Search is not available for this dataset
article
stringlengths 4.36k
149k
| summary
stringlengths 32
3.35k
| section_headings
listlengths 1
91
| keywords
listlengths 0
141
| year
stringclasses 13
values | title
stringlengths 20
281
|
|---|---|---|---|---|---|
Quality assessment is essential for the computational prediction and design of RNA tertiary structures . To date , several knowledge-based statistical potentials have been proposed and proved to be effective in identifying native and near-native RNA structures . All these potentials are based on the inverse Boltzmann formula , while differing in the choice of the geometrical descriptor , reference state , and training dataset . Via an approach that diverges completely from the conventional statistical potentials , our work explored the power of a 3D convolutional neural network ( CNN ) -based approach as a quality evaluator for RNA 3D structures , which used a 3D grid representation of the structure as input without extracting features manually . The RNA structures were evaluated by examining each nucleotide , so our method can also provide local quality assessment . Two sets of training samples were built . The first one included 1 million samples generated by high-temperature molecular dynamics ( MD ) simulations and the second one included 1 million samples generated by Monte Carlo ( MC ) structure prediction . Both MD and MC procedures were performed for a non-redundant set of 414 RNAs . For two training datasets ( one including only MD training samples and the other including both MD and MC training samples ) , we trained two neural networks , named RNA3DCNN_MD and RNA3DCNN_MDMC , respectively . The former is suitable for assessing near-native structures , while the latter is suitable for assessing structures covering large structural space . We tested the performance of our method and made comparisons with four other traditional scoring functions . On two of three test datasets , our method performed similarly to the state-of-the-art traditional scoring function , and on the third test dataset , our method was far superior to other scoring functions . Our method can be downloaded from https://github . com/lijunRNA/RNA3DCNN .
RNA molecules consist of unbranched chains of ribonucleotides , which have various essential roles in coding , decoding , regulation , expression of genes , and cancer-related networks via the maintenance of stable and specific 3D structures [1–5] . Therefore , their 3D structural information would help fully appreciate their functions . In this context , experiments such as X-ray crystallography , nuclear magnetic resonance ( NMR ) spectroscopy , and cryoelectron microscopy are the most reliable methods of determining RNA 3D structures , but they are costly , time-consuming , or technically challenging due to the physical and chemical nature of RNAs . As a result , many computational methods have been developed to predict RNA tertiary structures [6–32] . These methods usually have a generator producing a large set of structural candidates and a discriminator evaluating these generated candidates . A good generator should be able to produce structural candidates as close to native structures as possible , and a good discriminator should be able to recognize the best candidates . Moreover , a discriminator can direct generator searching structural space in heuristic prediction methods . For protein or RNA tertiary structure prediction , a discriminator generally refers to a free energy function , a knowledge-based statistical potential , or a scoring function . Several statistical potentials have been developed to evaluate RNA 3D structures , such as RASP [33] , RNA KB potentials [34] , 3dRNAscore [35] and the Rosetta energy function [9 , 16] . Generally , these potentials are proportional to the logarithm of the frequencies of occurrence of atom pairs , angles , or dihedral angles based on the inverse Boltzmann formula . The all-atom version of RASP defines 23 atom types , uses distance-dependent geometrical descriptions for atom pairs with a bin width of 1 Å , and is derived from a non-redundant set of 85 RNA structures . The all-atom version of RNA KB potential defines 85 atom types , also uses distance-dependent geometrical descriptions for atom pairs , and is derived from 77 selected representative RNA structures . Moreover , RNA KB potentials are fully differentiable and are likely useful for structure refinement and molecular dynamics simulations . 3dRNAscore also defines 85 atom types and uses distance-dependent geometrical descriptions for atom pairs with a bin width of 0 . 15 Å , and is derived from an elaborately compiled non-redundant dataset of 317 structures . In addition to distance-dependent geometrical descriptions , 3dRNAscore uses seven RNA dihedral angles to construct the statistical potentials with a bin width of 4 . 5° , and the final output potentials are equal to the sum of the two energy terms with an optimized weight . The Rosetta energy function has two versions: one for low resolution and the other for high resolution . The low-resolution knowledge-based energy function explicitly describing the base-pairing and base-stacking geometries guides the Monte Carlo sampling process in Rosetta , while the more detailed and precise high-resolution all-atom energy function can refine the sampled models and yield more realistic structures with cleaner hydrogen bonds and fewer clashes . As the paper on 3dRNAscore reported , 3dRNAscore is the best among these four scoring functions . Overall , the choices of the geometrical descriptors and the reference states in the scoring functions can affect their performance significantly , and the optimization of the parameters also influences this . Recently , we have witnessed astonishing advances in machine learning as a tool to detect , characterize , recognize , classify , or generate complex data and its rapid applications in a broad range of fields , from image classification , face detection , auto driving , financial analysis , disease diagnosis [36] , playing chess or games [37 , 38] , and solving biological problems [39–42] , to even quantum physics [43–45] . Even this list is incomplete , and has the potential to be extended further in the future . Therefore , we expect that machine learning methods will be able to help evaluate the structural candidates generated in the process of RNA tertiary structure prediction . Inspired by the successful application of 2D convolutional neural networks ( CNNs ) in image classification , we believe that 3D CNNs are a promising solution in that RNA molecules can be treated as a 3D image . Compared with other machine learning methods employing conventional hand-engineered features as input , 3D CNNs can directly use a 3D grid representation of the structure as input without extracting features manually . 3D CNNs have been applied to computational biology problems such as the scoring of protein–ligand poses [46 , 47] , prediction of ligand–binding protein pockets [48] , prediction of the effect of protein mutations [49] , quality assessment of protein folds [50] , and prediction of protein–ligand binding affinity [51] . Here , we report our work on developing two new scoring functions for RNA 3D structures based on 3D deep CNNs , which we name RNA3DCNN_MD and RNA3DCNN_MDMC , respectively . Our scoring functions enable both local and global quality assessments . To our knowledge , this is the first paper to describe the use of 3D deep CNNs to assess the quality of RNA 3D structures . We also tested the performance of our approaches and made comparisons with the four aforementioned energy functions .
The environment surrounding a nucleotide refers to its neighboring . To determine the neighboring atoms of a nucleotide , a local Cartesian coordinate system is specified first by its atoms C1’ , O5’ , C5’ , and N1 for pyrimidine or N9 for purine . Specifically , the origin of the local coordinate system is located at the position of atom C1’ . The x- , y- , and z-axes of the local coordinate system , denoted as x , y , and z , respectively , are decided according to Eqs 1–6 where rC1′ , rO5′ , rC5′ and rN stand for the vectors pointing from the origin in the global coordinate system to the atoms C1’ , O5’ , C5’ , and N1 or N9 , respectively . x = r N - r C 1 ′ ( 1 ) x = x ∥ x ∥ ( 2 ) y = r O 5 ′ + r C 5 ′ 2 - r C 1 ′ ( 3 ) z = x × y ( 4 ) z = z ∥ z ∥ ( 5 ) y = z × x ( 6 ) The environment surrounding a nucleotide consists of the atoms whose absolute values of x , y , and z coordinates are less than a certain threshold . Here , the threshold is set to 16 Å , which means that the environment surrounding a nucleotide contains the atoms within a cube of length 32 Å centered at this very nucleotide , as shown in Fig 1A . For a colorful 2D image , the input of a 2D CNN is an array of pixels of RGB channels . Similarly , in our work , the nucleotide and its surrounding environment are transformed into a 3D image consisting of an array of voxels . As shown in Fig 1A , the box of size 32 × 32 × 32 Å is partitioned into 32 × 32 × 32 grid boxes . Each grid box represents a voxel of three channels and its values are calculated by the accumulations of the occupation number , mass , or charge of the atoms in the grid box . The mass and charge information of each type of atoms is listed in S1 Table . After transformation , the input of the 3D CNN is a colorful 3D image of 32 × 32 × 32 voxels with three channels corresponding to RGB channels presented in Fig 1B . Practically , each channel is normalized to [0 , 1] by min-max scaling . The output of our CNN is the nucleotide unfitness score characterizing how poorly a nucleotide fits into its surroundings . For a nucleotide , its unfitness score is equal to the RMSD of its surroundings plus the RMSD of itself after optimal superposition between its conformations in the native structure and the assessed structure . The latter RMSD is generally very small , but the former varies in a large range . Nucleotides with smaller unfitness scores are in a conformation closer to the native conformation , and a score of 0 means that the nucleotide fits into its surrounding environment perfectly and is in its native conformation . Practically , the nucleotide unfitness score is normalized to [0 , 1] by min-max scaling . For the global quality assessment , the unfitness scores of all nucleotides are accumulated . Fig 1C exhibits the architecture of our CNN , a small VGG-like network [52] containing a stack of convolutional layers , a maxpooling layer , a fully connected layer , and 4 , 282 , 801 parameters in total . VGGNet is a famous image classification CNN . It is a very deep network and uses 19 weight layers , consisting of 16 convolutional layers stacked on each other and three fully-connected layers . The input image size 224 × 224 in VGGNet is much larger than our input size 32 × 32 × 32 in terms of the side length , and thus we used a smaller architecture . There are only four 3D convolutional layers in our neural network . The numbers of filters in each convolutional layer are 8 , 16 , 32 , and 64 , and the receptive fields of the filters in the first two convolutional layers and in the last two convolutional layers are 5 × 5 × 5 voxels and 3 × 3 × 3 voxels , respectively . The convolution stride is set to one voxel . No spatial padding is implemented in the convolutional layers . Moreover , a max-pooling layer of stride 2 is placed following the first two consecutive convolutional layers . Subsequently , one fully connected layer with 128 hidden units is stacked after the convolutional layers . The final output layer is a single number , namely , the unfitness score . All units in hidden layers are activated by the ReLU nonlinear function , while the output layer is linearly activated . The neural network was trained to reduce the mean squared error ( MSE ) between the true and predicted unfitness scores . A back-propagation-based mini-batch gradient descent optimization algorithm was used to optimize the parameters in the network . Batch size was set to 128 . The training was regularized by dropout regularization for the second , fourth convolutional layers , and the fully connected layer with a dropout ratio of 0 . 2 . The Glorot uniform initializer was used to initialize the network weights . The learning rate was initially set to 0 . 05 , and then decreased by half whenever the MSE of the validation dataset stopped improving for five epochs . The training process stopped when the learning rate decreased to 0 . 0015625 . Our 3D CNN was implemented using the python deep learning library Keras [53] , with Theano library as the backend . To construct the training dataset , first a list of 619 RNAs was downloaded with the search options “RNA Only” and “Non Redundant RNA Structures” from the NDB website http://ndbserver . rutgers . edu/ , which means that our training dataset includes RNA-only structures and the RNAs are non-redundant in both sequence and geometry . Second , the RNAs with an X-ray resolution >3 . 5 Å were removed from the list above . Finally , the RNAs in the test dataset were removed and the RNAs in the equivalence classes with the test dataset were also removed . “Structures that are provisionally redundant based on sequence similarity and also geometrical similarity are grouped into one equivalence class , ” as Leontis et al . defined [54] . Thus , 414 native RNAs were left to construct the training dataset . According to their length , the 414 RNAs were randomly divided into two groups , namely , 332 RNAs for training and 82 RNAs for validation in the CNN training process . Practically , the training samples were generated in two ways , namely , by MD and MC methods elaborated as follows . To evaluate our CNN-based scoring function and make comparisons with the traditional statistical potentials , three test datasets were collected from different sources . Test dataset I comes from the RASP paper [33] which is generated by the MODELLER computer program from the native structures of 85 non-redundant RNAs given a set of Gaussian restraints for dihedral angles and atom distances , and contains 500 structural decoys for each of the 85 RNAs . The RMSDs are in different ranges for these RNAs . The narrowest are from 0 to 3 . 5 Å , the broadest are from 0 to 13 Å , and the RMSDs of most decoys are less than 10 Å . This dataset can be downloaded from http://melolab . org/supmat/RNApot/Sup . _Data . html . Test dataset II comes from the KB paper [34] , which is generated by both position-restrained dynamics and REMD simulations for 5 RNAs and the normal-mode perturbation method for 15 RNAs . For the MD dataset , there are 3 , 500 decoys for each of four RNAs whose RMSDs range from 0 to >10 Å , and 2 , 600 decoys for one RNA ( PDB ID: 1msy ) whose RMSDs range from 0 to 8 Å . Meanwhile , for the normal-mode dataset , there are about 490 decoys for each of the 15 RNAs , whose RMSDs range only from 0 to 5 Å . This dataset can be downloaded from http://csb . stanford . edu/rna . One point that should be noted is that the downloaded pdb files name atom O2 in pyrimidine bases as “O . ” Test dataset III comes from RNA-Puzzles rounds I to III [55–57] , a collective and blind experiment in 3D RNA structure prediction . Given the nucleotide sequences , interested groups submit their predicted structures to the RNA-Puzzles website before the experimentally determined crystallographic or NMR structures of these target sequences are published . Therefore , the dataset is produced in a real RNA modeling scenario and can reveal the real performance of the existing scoring function . Marcin Magnus compiled the submitted structures from rounds I to III , and now the predicted models of 18 target RNAs can be downloaded from https://github . com/RNA-Puzzles/RNA-Puzzles-Normalized-submissions . There are only 12–70 predicted models for the 18 RNAs , some of whose RMSDs range from 2 to 4 Å , while some cover a wide range from 20 to 60 Å . Two neural networks were trained based on two sets of training samples . The first set included only MD training samples and the second set included both MD and MC training samples . And the two network models are named RNA3DCNN_MD and RNA3DCNN_MDMC , respectively . We tested test datasets I and II using RNA3DCNN_MD , and tested test dataset III using RNA3DCNN_MDMC . The reason why we trained two neural networks is that the three test datasets come from two kinds of methods . Test dataset I and II were produced by MD and normal-mode methods initiated from native structures , while test dataset III was produced by MC structure prediction methods , covering a broad structural space . After testing , for test datasets I and II , RNA3DCNN_MD performed better than RNA3DCNN_MDMC . But for test dataset III , RNA3DCNN_MDMC was superior . The results are reasonable . RNA3DCNN_MD is more accurate in the region close to native structures in that most of the MD training samples are not very far away from native structures or native topologies . However , when MC training samples were included , the neural network RNA3DCNN_MDMC became not as accurate as RNA3DCNN_MD for the structures around native ones and biased the non-native . On the contrary , RNA3DCNN_MD did not see the more random training structures far away from native states and thus it did not perform as well as RNA3DCNN_MDMC for test dataset III .
In general , a scoring function with good performance should be able to recognize the native structure from a pool of structural decoys and to rank near-native structures reasonably . Consequently , two metrics were used for a quantitative comparison with other scoring functions . One was the number of native RNAs with minimum scores in the test dataset , and the other was the Enrichment Score ( ES ) [34 , 35 , 58] , which characterizes the degree of overlap between the structures of the top 10% scores ( Etop10% ) and the best 10% RMSD values ( Rtop10% ) in the structural decoy dataset . The ES is defined as E S = | E t o p 10 % ∩ R t o p 10 % | 0 . 1 × 0 . 1 × N d e c o y s ( 7 ) where |Etop10% ∩ Rtop10%| is the number of structures in both the lowest 10% score range and the lowest 10% RMSD range , and Ndecoys is the total number of structures in the decoy dataset . If the score and RMSD are perfectly linearly correlated , ES is equal to 10 . If they are completely unrelated , ES is equal to 1 . If ES is less than 1 , the scoring function performs rather poorly with respect to that decoy dataset . We compared our CNN-based scoring function with four traditional statistical potentials for RNA , namely , 3dRNAscore , KB , RASP , and Rosetta . First , the number of native RNAs with minimum scores was counted as listed in Table 1 . As the 3dRNAscore paper reported , 3dRNAscore identified 84 of 85 native structures , KB 80 of 85 , RASP 79 of 85 , and Rosetta 53 of 85 . 3dRNAscore is thus clearly the best among the four statistical potentials . Our RNA3DCNN identified 62 of 85 native structures , and the unidentified native structures generally had the second or third lowest scores , almost the same as the lowest scores . Fig 2A shows an example in test dataset I in which the native structure was identified by our method , and Fig 2B shows an example in test dataset I in which the native structure had a slightly higher score calculated by our method than the structure of an RMSD of 0 . 9 Å . The RMSD-score plots of all 85 examples are provided in S1 Fig . The result that our method identified fewer native structures is reasonable . Specifically , the input and output of our neural network are geometry based , and thus similar structures have similar scores . The structures in the 0–1 Å range generally resemble each other and thus , for our scoring function , all the non-native structures with minimum scores have an RMSD ∼1 Å . Meanwhile , for the statistical potentials , atom steric clashes , angle , or dihedral angle deviations from the native form may quickly increase the potential values . Second , the ES was calculated . The mean ES values of the 85 RNAs calculated by 3dRNAscore , RASP , Rosetta , and our method RNA3DCNN were 8 . 69 , 8 . 69 , 6 . 7 , and 8 . 61 , respectively . The mean ES calculated by KB is not given in that we cannot open its original website and download its program , and the results of KB method shown in this paper come from the papers on KB and 3dRNAscore . The ES values of 3dRNAscore and our method are almost the same . The mean ES values of three methods are very large , suggesting that the RMSDs and scores calculated by the different methods are highly linearly correlated and that this test dataset is an easy benchmark to rank near-native decoys . For the MD decoys in test dataset II , 3dRNAscore and KB identified 5 of 5 native structures , RASP 1 of 5 , Rosetta 2 of 5 , and our method 4 of 5 , as listed in Table 1 . Our method gave the lowest score to the decoy of an RMSD of 0 . 97 Å for RNA 1f27 , as shown in Fig 3B . The ES values of the MD decoys using different scoring functions are listed in Table 2 . Fig 3A shows the relationship between RMSD and the score calculated by our method for the RNA 434d with the best ES . The RMSD-score plots of all five examples are provided in S2 Fig . From the table , we can see that our method performed better than 3dRNAscore for 2 of 5 RNAs , slightly worse for 1 of 5 RNAs , and worse for 2 of 5 RNAs , especially for the RNA 1f27 , in that the native structure had a slightly higher score than the decoys of RMSD around 1 Å . Moreover , our method performed better than KB , RASP , and Rosetta for 3 of 5 RNAs , comparably for 1 of 5 RNAs , and worse for the RNA 1f27 , as explained above . For the normal-mode decoys in this dataset , 3dRNAscore identified 12 of 15 native structures , RASP 11 of 15 , Rosetta 10 of 15 , KB and our method 15 of 15 , as listed in Table 1 . The ES values of the normal-mode decoys using different scoring functions are also listed in Table 2 . From the table , we can see that our method performed better than 3dRNAscore for 7 of 15 RNAs , equally for 4 of 15 RNAs , and worse for only 4 of 15 RNAs . Moreover , our method performed better than KB , RASP , and Rosetta for 12 , 11 , and 13 of 15 RNAs . The mean ES values of 3dRNAscore and our method were the same , and were greater than the other scoring functions . The RMSD-score plots of all 15 examples are provided in S2 Fig . The structures in test dataset III are derived from different groups by different RNA modeling methods . There are only dozens of predicted models for each target RNA and the RMSDs are almost always greater than 10 Å , and often even greater than 20 , or 30 Å . Consequently , we did not calculate the ES for this dataset and gave only the RMSDs of models with minimum scores in Table 3 . The results of method KB were not provided in that we could not open its website and get the program . From the table , we can see that our RNA3DCNN identified 13 of 18 native RNAs , 3dRNAscore 5 of 18 , RASP 1 of 18 , and Rosetta 4 of 18 . For puzzle 2 , though the native structures were not identified , our method gave the lowest RMSD among four methods . And for puzzle 3 , our method gave the RMSD as low as other two methods . Fig 4A shows an example in test dataset III in which the native structure was well identified by our method , and Fig 4B is the one not identified . The RMSD-score plots of all 18 examples are provided in S3 Fig . For test datasets I and II , all decoys are obtained from native structures , which means that they almost always stay around one local minimum in the energy landscape . But for test dataset III , in the real modeling scenario , the structures are far from native topologies and are located at different local minima in the energy landscape . For this reason , we trained two neural networks with two sets of training samples , that is , one set including only training samples from MD simulations initiated from native structures and another set including both MD training samples and MC training samples obtained in the broader and more complicated structural space . Our scoring function can evaluate each nucleotide , reveal the regions in need of further structural optimization , and guide the sampling direction in RNA tertiary structure modeling . Fig 5 portrays how our scoring function helps locate the unfit regions . In this figure , a decoy of RMSD 3 . 0 Å from test dataset II MD decoys and the native RNA 1nuj are superimposed , and thicker tubes show larger deviations from the native structure . The rainbow colors represent the calculated unfitness scores of each nucleotide , and the colors closer to red represent larger unfitness scores . We can see that the tubes in nucleotides 1 , 7 , 8 , 9 , and 14 are much thicker , and the colors of those regions are much closer to red , which means that our scoring function can rank the nucleotide quality correctly . Nucleotides 1 and 14 are the terminal nucleotides in two chains and are unpaired , so the deviations of these two are the largest . Nucleotides 7–9 are in the internal loop , so the deviations are larger than those of the remaining helical regions . The Pearson correlation coefficients between actual and predicted nucleotide unfitness scores were 0 . 69 and 0 . 34 for MD decoys and NM decoys in test dataset II , respectively , as shown in S4 Fig . The structures in NM decoys are all near native structures with RMSD ranging from 0 to 5 Å , thus making the correlation not strong . Saliency maps were used to visualize the trained network and help understand which input atoms are important in deciding the final output . In paper [59] , an image-specific class saliency map was first introduced to rank the pixels of an input 2D image based on their influence on the class score by computing the gradient of output class score with respect to the input image . The gradient can reveal how sensitive the class score is to a small change in input image pixels . Larger positive gradients mean that a slight decrease in the corresponding pixels can cause the true class score to drop markedly , and thus the corresponding pixels are more important in determining the right output class . Meanwhile , for our regression problem and a near-native conformation , the smaller output was better and the voxels of negative gradients were highlighted and important . Moreover , we mapped the gradients of each voxel back to the corresponding atoms . In Fig 6 , examples of saliency maps for the three input channels are presented . A , B , and C correspond to atomic occupation number , mass , and charge channels , respectively . The example is used to calculate the unfitness score of the 12th nucleotide in a helical region for the native RNA 1nuj . The nucleotide under assessment is drawn as spheres and sticks , its surrounding environment is drawn as sticks , while the atoms beyond its surrounding environment are shown as a black cartoon . The redder atoms represent smaller negative gradients , the bluer atoms represent larger positive gradients , and the nearly white atoms represent gradients close to 0 . The red regions are highlighted and more important in deciding the final output . In the atomic occupation number channel , atomic category differences disappear and only shapes count . From Fig 6A , we can see that the atoms in the nucleobases of the 10th–13th and 15th–19th nucleotides are highlighted and atom N3 in the 16th nucleotide is the most important , in accordance with the base-pairing and base-stacking interactions . In the atomic mass channel , the importance of atoms in the nucleobases described above declines somewhat , while atom P in the 12th nucleotide and atom N3 in the 16th nucleotide are the most important , in that atom P is much heavier than atoms C , N , and O and atom N3 is in the A12’s paired-base U16 . In the atomic charge channel , the seven most important atoms are N1 , P , N3 , and O3’ in the 12th nucleotide , atoms C4 and C2 in the 16th nucleobase , and atom N2 in the 17th nucleobase . Overall , from the analyses of the salient maps , it was found that the neural networks can learn the knowledge , such as the relevance of base pairing and stacking interactions to the score , from the training data automatically without any priori knowledge . It would be very interesting to see if neural networks can dig new knowledge out of data in the future work . We tested the computational time of 100 decoys of 91 nucleotides . The total time was 321 . 0 seconds . For a comparison , the C++ version of 3dRNAscore method took only 19 seconds . However , it was found that 99 . 6% of our computational time ( 319 . 7 seconds ) was used to prepare the input to CNN , and this time decreased to 2 seconds after we changed the code from Python to C++ . Therefore , the CNN-based approach is very efficient in terms of speed , and it is estimated that the overall computational time of our method will be approximately 3 seconds if we rewrite the entire code in C++ . However , the computational time of our method in Python version is acceptable for now , at least temporarily . We postpone the code rewriting work to the future when necessary . Moreover , our method can be downloaded from https://github . com/lijunRNA/RNA3DCNN . Recently , we have witnessed the astonishing power of machine learning methods in characterizing , classifying , and generating complex data in various fields . It is therefore interesting to explore the potential of machine learning in characterizing and classifying RNA structural data . In this study , we developed two 3D CNN-based scoring models , named RNA3DCNN_MD and RNA3DCNN_MDMC , for assessing structural candidates built by two kinds of methods . If the structural candidates are generated by MC methods such as fragment assembly , RNA3DCNN_MDMC is suggested . If the structural candidates are not very far away from the native structures , such as from MD simulations , the RNA3DCNN_MD model is better . We also compared our method with four other traditional scoring functions on three test datasets . The current 3D CNN-based approaches performed comparably with or better than the best statistical potential 3dRNAscore on different test datasets . For the first test dataset , the mean ES was almost the same as that of the best traditional scoring function , 3dRNAscore . The reason why the number of native structures identified by our method was much smaller than that by other scoring functions is that our method is structure-based and the scores of native structures and decoys of RMSD less than 1 . 0 Å are almost the same . This suggests that our method is robust if an RNA structure does not change much . For the second test dataset , our method generally performed similarly to 3dRNAscore and outperformed the three other scoring functions . For the MD decoys in the second test dataset , our method was slightly worse than 3dRNAscore . For the normal-mode decoys in the second test dataset , our method identified all the native structures , while 3dRNAscore identified only 12 of 15 native RNAs , and our method outperformed 3dRNAscore for 7 of 15 RNAs and underperformed it for only 4 of 15 RNAs . For the third test dataset from blind and real RNA modeling experiments , our method was far superior to the other scoring functions in identifying the native structures . Our method has some novel features . First , it is free of the choice of the reference state , which is a difficult problem in traditional statistical potentials . Second , it treats a cube of atoms as a unit like a many-body potential , while traditional statistical potentials divide them into atom pairs . Moreover , our method can evaluate each nucleotide , reveal the regions in need of further structural optimization , and guide the sampling direction in RNA tertiary structure prediction . Our method demonstrates the power of CNNs in quality assessments of RNA 3D structures and shows the potential to far outperform traditional statistical potentials . There remains great scope to improve the CNN models , such as by expanding them to include more input channels ( only three are considered currently ) , featuring more complex network architecture , and involving larger training datasets . Moreover , more RNA-related problems can be dealt with by 3D CNNs , such as protein–RNA binding affinity prediction and RNA–ligand docking and virtual screening .
|
RNA is an important and versatile macromolecule participating in various biological processes . In addition to experimental approaches , the computational prediction of RNA 3D structures is an alternative and important source of obtaining structural information and insights into their functions . An important part of these computational prediction approaches is structural quality assessment . For this purpose , we developed a 3D CNN-based approach named RNA3DCNN . This approach uses raw atom distributions in 3D space as the input of neural networks and the output is an RMSD-based nucleotide unfitness score for each nucleotide in an RNA molecule , thus making it possible to evaluate local structural quality . Here , we tested and made comparisons with four other traditional scoring functions on three test datasets from different sources .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"and",
"discussion"
] |
[
"molecular",
"dynamics",
"neural",
"networks",
"particle",
"physics",
"statistics",
"rna",
"structure",
"prediction",
"neuroscience",
"nucleotides",
"atoms",
"mathematics",
"forecasting",
"composite",
"particles",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"rna",
"structure",
"mathematical",
"and",
"statistical",
"techniques",
"chemistry",
"molecular",
"biology",
"physics",
"biochemistry",
"rna",
"molecular",
"structure",
"nucleic",
"acids",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"chemistry",
"chemical",
"physics",
"statistical",
"methods",
"macromolecular",
"structure",
"analysis"
] |
2018
|
RNA3DCNN: Local and global quality assessments of RNA 3D structures using 3D deep convolutional neural networks
|
Varicella-zoster virus ( VZV ) is a neurotropic human alphaherpesvirus that causes varicella upon primary infection , establishes latency in multiple ganglionic neurons , and can reactivate to cause zoster . Live attenuated VZV vaccines are available; however , they can also establish latent infections and reactivate . Studies of VZV latency have been limited to the analyses of human ganglia removed at autopsy , as the virus is strictly a human pathogen . Recently , terminally differentiated human neurons have received much attention as a means to study the interaction between VZV and human neurons; however , the short life-span of these cells in culture has limited their application . Herein , we describe the construction of a model of normal human neural progenitor cells ( NHNP ) in tissue-like assemblies ( TLAs ) , which can be successfully maintained for at least 180 days in three-dimensional ( 3D ) culture , and exhibit an expression profile similar to that of human trigeminal ganglia . Infection of NHNP TLAs with cell-free VZV resulted in a persistent infection that was maintained for three months , during which the virus genome remained stable . Immediate-early , early and late VZV genes were transcribed , and low-levels of infectious VZV were recurrently detected in the culture supernatant . Our data suggest that NHNP TLAs are an effective system to investigate long-term interactions of VZV with complex assemblies of human neuronal cells .
Varicella-zoster virus ( VZV ) is a ubiquitous human herpesvirus , as evidenced by a seroprevalence of more than 95% worldwide [1] . Among human herpesviruses , VZV has the smallest genome of approximately 125 kbp , which contains at least 70 open reading frames ( ORFs ) and consists of two unique regions , unique long ( UL ) and unique short ( US ) , each flanked by inverted repeat regions ( TRL , IRL , TRS , IRS ) [2] , [3] . Primary VZV infection typically causes childhood varicella ( chickenpox ) . During primary infection the virus gains access to and establishes latency in multiple cranial , dorsal root and autonomic ganglia . Varicella vaccination programs have successfully reduced the incidence of clinical disease in children in the USA by about 80% [4] , [5]; however , the attenuated vaccine , like wild-type VZV , is still able to establish latency in the peripheral nervous system [6] . Both wild-type and vaccine virus can reactivate from latency , particularly in elderly and immunocompromised individuals . Reactivation is associated with a declining VZV specific T-cell immunity and can result in herpes zoster ( shingles ) especially in the elderly , which is characterized by severe pain and often followed by postherpetic neuralgia [7] , [8] . Reactivation can also lead to progressive outer retinal necrosis and stroke by ischemic vasculopathy [9] , [10] . With respect to disease severity , duration and quality-of-life impairment , VZV reactivation in adulthood can be more serious than primary childhood infection . VZV pathogenesis , latency , and reactivation are difficult to study , as the virus exclusively infects humans and no animal model is currently available to investigate VZV latency . As a result , our knowledge of VZV latency is based on the analysis of human ganglia removed at autopsy . However , primary human ganglia have several disadvantages: ( i ) they can only be cultured for a few days in vitro , ( ii ) the availability of human ganglia is limited , ( iii ) the time between death and tissue collection cannot be predetermined , ( iv ) there are considerable human to human variations including preexisting virus burden and time since primary VZV infection , and ( v ) since most ganglia explants contain latent VZV , it is difficult to perform prospective studies using VZV with defined genetic mutations . Therefore , our understanding of VZV latency largely represents a snapshot of human ganglia that show a high degree of variability [11] taken shortly after death . Human fetal dorsal root ganglia ( DRG ) are typically free of latent VZV . Enzymatically dissociated human DRG show neurite outgrowth when maintained in culture and have facilitated the analysis of apoptosis inhibition mediated by VZV in neurons [12] , [13]; however , since the cultures can be maintained for only a few weeks , their application to study latent infection is limited . Human fetal DRG can be maintained for eight weeks as xenografts under the kidney capsule of SCID mice [14] , and the virus gene expression profile at eight weeks post-infection in DRG xenografts is comparable to latently infected adult trigeminal ganglia; however , difficulties in maintaining viable DGR xenografts along with ethical concerns regarding fetus-derived tissue has limited their widespread utilization . Human stem cells provide an alternative source of neurons . Stem cell-derived neurospheres can be maintained in tissue culture and in lateral ventricles of neonatal non-obese diabetic SCID mice . While human neurospheres maintained in mice can be infected with VZV and show no obvious virus-induced cytopathic effect , time course experiments using this system are technically challenging and costly [15] . In addition , differentiated human pluripotent stem cells can be infected with VZV and are useful to visualize retrograde virus axonal transport; however in these cultures , VZV initiates lytic replication that appears to be related to the neuronal purity [16] , [17] , [18] . Neuronal cultures lacking significant numbers of glia or astrocytes can be infected , show no virus-induced cytopathic effects ( CPE ) and do not produce infectious virus; but , it is challenging to obtain these short-lived cultures of pure neurons [19] . Taken together , the available neuronal models to study VZV interactions are hindered by their limited availability , elaborate maintenance requirements , limited lifespan , neuronal purity and significant ethical concerns . In order to overcome these problems , we developed a novel three-dimensional ( 3D ) culture system to maintain partially differentiated normal human neural progenitor ( NHNP ) cells as tissue-like assemblies ( TLAs ) that share some features with neurons found in human trigeminal ganglia . The 3D NHNP TLAs can be maintained for at least three months following VZV infection . During this time , VZV DNA was readily detected , VZV immediate-early , early and late genes were transcribed , no VZV-induced CPE was observed , but low amounts of infectious virus were released into the culture medium . Importantly , the VZV genome was stable throughout the three month infection period , suggesting that the 3D NHNP TLA culture system may be useful to investigate the long-term interplay between VZV and neural tissue .
NHNP cells were cultured in 3D on a scaffold of Cultispher beads in a rotating wall vessel ( RWV ) bioreactor and analyzed by multicolor fluorescence in situ hybridization ( mFISH ) , flow cytometry , confocal microscopy and RT-PCR to assess their genetic stability as compared with 2D NHNP cell cultures . Comparison of 2D and 3D cultured NHNP cells by mFISH respectively showed no chromosomal rearrangements or breaks , indicating genetic stability of NHNP TLAs for at least a six month period in culture ( Fig . 1A–B ) . Flow cytometry analysis confirmed that after 180 days in culture , NHNP TLAs expressed neuronal progenitor markers CXCR4 , CD133 , CD105-Endoglin , CD 90-Thy-1 and CD49f-α6 Integrin at levels comparable to the parental NHNP ( 2D ) cell population ( Table 1 ) . Markers of hematopoietic differentiation , CD38+ ( Table 1 ) and CD45+ ( data not shown ) were not detected in either culture . An example of the tissue-like complexity of the NHNP TLA constructs is shown by environmental scanning electron micrographs ( ESEM ) in Fig . 1C–F , which illustrate the relative size , density , and indistinguishable nature of the individual neural cells in the tissue assembly . In addition , confocal microscopy confirmed that NHNP TLAs expressed mature neuronal markers: neuron-specific nuclear protein ( NeuN ) , β-Tubulin-III , microtubule associated protein A&B ( MAP 2 A&B ) , and glial fibrillary acidic protein ( GFAP ) ( Fig . 2A ) . TLA microcarrier substrates alone did not show autofluorescence ( data not shown ) . Similarly , progenitor ( Nestin ) and mature ( β-Tubulin-III ) neuronal markers were detected by immunohistochemistry in human TG removed at autopsy ( Fig . 2B ) . Quantitative analysis of mRNA expression revealed higher levels of the early progenitor markers CXCR4 and CD133 in NHNP TLAs than in human TG ( Fig . 2C ) even though only few NHNP cells expressed the two proteins ( Table 1 ) . Later stage developmental markers , CD105 , CD90 and CD49f , were expressed to a lesser extent in NHNP TLAs than in human TG , suggesting that NHNP TLAs are less differentiated than primary TGs . Neuronal markers Nestin and β-Tubulin-III were expressed at comparable levels in TG and NHNP TLAs ( Fig . 2C ) . Lower expression levels in the NHNP TLAs than in human TG were observed for the neuronal marker neurofilament 200 ( NF-200 ) , again indicating NHNP TLAs are generally less differentiated than human TG . Taken together , our data indicated that NHNP TLAs , which had undergone 18–20 fold expansion over the three-month time period , share some similarities but are less differentiated when compared to mature human TGs removed at autopsy . Both copies of the diploid VZV gene ORF63/70 encoding the immediate-early protein 63 ( IE63 ) were mutated to identify expression originating from each gene . The fluorescent proteins eGFP and mRFP were inserted at the C-terminus of ORF63 and 70 , respectively , in the parental VZV Oka strain . For genetic manipulations , an infectious BAC clone ( pP-Oka ) was used and the insertion of the marker genes resulted in recombinant BAC p63G/70R ( Fig . 3A ) . Correct incorporation of eGFP and mRFP coding sequences was confirmed by PCR , DNA sequencing and multiple RFLP analyses ( Fig . 3B ) . The p63G/70R BAC was then transfected into MeWo cells , resulting in reconstitution of the recombinant virus ( v63G/70R ) . Upon virus reconstitution , ORF63-eGFP and ORF70-mRFP expression was readily detected in all plaques ( Fig . 3C , upper panel ) , confirming that both loci were expressed in infected MeWo cells . Upon subsequent passages of v63G/70R , plaques developed in which only either eGFP or mRFP were expressed ( Fig . 3C , middle and lower panel ) , suggesting that recombination between the IRS and TRS regions had occurred . PCR analysis of multiple isolated plaques positive for only eGFP or mRFP indicated homogeneity in the fluorescent tag , thus confirming that both ORF63 and ORF70 contained the same mutation ( data not shown ) , and confirming the suspected recombination . Growth kinetics showed no significant effect on virus replication due to insertion of the fluorescent proteins into ORF63/70 ( Fig . 4A ) . Taken together , our data indicated that recombination between the IRS and TRS region occurred frequently in cultured MeWo cells , which resulted in a stable replacement of IRS or TRS with their respective counterparts , TRS or IRS , respectively . Cell-free v63G/70R efficiently infected NHNP TLAs , as evidenced by an approximate 50-fold increase in VZV genome copies from 0 to 18 days post-infection ( dpi ) . After 18 dpi , VZV genome copy numbers remained constant ( Fig . 4B ) . To compare the overall health of VZV infected MeWo cells grown in 2D to VZV infected NHNP TLAs grown in 3D , glucose utilization pre- and post-VZV infection was monitored ( Fig . 4C ) . Each culture was maintained initially for 39 days to establish a baseline glucose consumption rate . Upon v63G/70R infection , glucose utilization rapidly declined in MeWo cells , most likely a consequence of lytic VZV replication as seen microscopically by extensive VZV-induced cytopathic effects ( data not shown ) . In contrast , glucose uptake in NHNP TLAs was not altered as a response to VZV infection ( Fig . 4C ) , suggesting that limited , if any , lytic VZV infection occurred . Quantitative RT-PCR was performed to determine if the virus genome was transcribed in infected NHNP TLAs . Transcript levels of the immediate-early gene ORF63 increased over time and correlated to the amount of VZV genome copies ( Fig . 5A ) . Further analysis showed that late VZV gene ORF 9 and ORF 40 transcripts also increased in abundance ( Fig . 5B ) . To determine if infectious virus was released from infected NHNP TLAs , cell-free supernatants from VZV infected NHNP TLAs were titrated using permissive MeWo cells and plaques stained using an anti-VZV IE63 antibody . Shortly after inoculation of the NHNP TLA culture with VZV , high levels of cell-free virus were detected ( Table 2 ) . Each day post infection , 80% of the culture medium was replaced , resulting in a rapid removal of cell-free virus from the reactor . However , low-level release of cell-free VZV was detected on 6 , 10 , 25 and 35 dpi , while no infectious virus was detected on 14 , 21 and 30 dpi , indicating intermittent de novo production and release of infectious virus ( Table 2 ) . It is noteworthy that the NHNP TLAs remained intact and healthy despite apparent virus production . Confocal analysis of NHNP TLAs infected with VZV containing eGFP fused to both ORF63 and ORF70 at 27 dpi revealed that the progenitor neuronal marker Nestin ( Fig . 6 A–C ) and the mature neuronal marker β-Tubulin-III colocalized with ORF63-eGFP ( Fig . 6D–F ) . Taken together , our data demonstrates that VZV infection of NHNP TLAs results in prolonged accumulation of virus DNA , mRNA and sporadic release of very small amounts of infectious cell-free virus . Recombination between inverted repeat regions in the genome of alphaherpesviruses during lytic replication is well documented [20] , [21] . To determine if VZV DNA recombination occurred in MeWo ( 2D ) or NHNP TLAs ( 3D ) cultures , we determined the recombination frequency between ORF63 and ORF70 by qPCR . During lytic replication in MeWo cells , recombination within the VZV genome that resulted in a replacement of ORF70-mRFP with ORF63-eGFP occurred frequently ( Fig . 7A ) , indicating a selective pressure against the ORF70-mRFP fusion protein . In contrast , the GFP/RFP ratios in v63G/70R infected NHNP TLAs remained unaltered for at least 69 days in culture , suggesting that the VZV genome is stably maintained in NHNP TLAs . In addition , confocal microscopy confirmed that both ORF63-eGFP and ORF70-mRFP are expressed in NHNP TLA cultures infected with v63G/70R ( Fig . 7B ) . Our data indicates a stable viral genome is preserved for an extended period in NHNP TLAs .
Establishment of latent infection and subsequent reactivation is integral to the alphaherpesvirus life cycle and ensures that virus is maintained in the population by intermittent virus production and transmission [22] . However , VZV appears to be unique among alphaherpesviruses with respect to establishment of and reactivation from the persistent/latent state [23]–[26] , emphasizing the need for accurate nomenclature . VZV-host interactions are classified as acute , “rapid production of infectious virions followed by rapid resolution and elimination of infection” , or persistent , “virus particles or products continue to be produced for long periods in which virions are continuously or intermittently produced” [27] . In most human cells in culture , VZV infection is acute and cells succumb to virus infection within 3–5 days most likely through apoptosis [28]–[31] . The low ratio ( 1∶40 , 000 ) of infectious to defective VZV particles indicates that production of complete virions is extremely inefficient [32] . Human neurons are known to be latently infected by VZV , but the lack of a suitable animal model has hindered investigations into this exceptional relationship between VZV and a host cell or organ . Our common understanding is that VZV infection of neurons results in latency . Latency is an extreme variant of persistent infection where , as exemplified by herpes simplex virus type 1 ( HSV-1 ) , “infectious virions can no longer be isolated” [27] . During latency , most HSV-1 genes are silenced through epigenetic modification of resident histones or by virus-specific miRNAs [33]–[35] . Therefore , HSV-1 gene transcription in infected neurons is restricted to the latency-associated transcript ( LAT ) at the cellular level , while no viral proteins and infectious virions are produced [36]–[38] . According to the above definitions and following the example provided by HSV-1 , the prototype alphaherpesvirus , VZV infection of NHNP TLAs maintained in 3D cultures is persistent and not latent . While VZV DNA is present and virus-induced CPE was not detected either microscopically or by monitoring total glucose uptake , VZV genes are transcribed , IE63 is translated , and small amounts of infectious virions are intermittently produced . The observation that the quantity of late VZV transcripts directly correlates to IE63 transcription and viral genome copies , suggests that many cells could express these transcripts and potentially produce infectious virus . Alternatively , late transcripts as well as infectious cell-free virions could be derived from very few cells that productively replicate VZV . While at least 12 VZV genes are transcribed in human ganglia removed at autopsy [39]–[41] and IE63 protein production is commonly detected [42] , it is known that the extent of VZV transcription increases with longer times between death and tissue analysis [43] . Hence , it has been argued that transcription of multiple virus genes detected in human ganglia most likely reflects reactivation of latent virus genomes [44] , [45] . Thus , while multiple viral genes are transcribed in infected NHNP TLAs and in human trigeminal ganglia containing VZV DNA at autopsy , the former is a long-term permissive infection and the later most likely reflects early stage virus reactivation . Our current work highlights the importance of NHNP TLAs in 3D culture for the study of VZV persistent infection , latency and reactivation . The ability to maintain VZV infection for extended times along with the ease of experimental manipulation allows us to investigate the VZV life cycle in an in vitro system for the first time . Our 3-month experiment was operator-terminated , and the healthy cultures still contained VZV DNA . As such , infected NHNP TLAs in 3D culture provide an excellent means to investigate chronic virus-neuron interactions . An additional attribute of the 3D cultures is that they allow introduction of autologous VZV-specific CD8 T-cells with the aim of immune surveillance that would remove cells actively producing virus . Previous studies on HSV-1 showed that CD8 T-cell surveillance within latently infected human and mouse trigeminal ganglia may help maintain latent infection [46]–[56] . This in turn could be also true for VZV and could allow the establishment of latently infected NHNP cultures by the addition of CD8 T-cells . The NHNP TLAs system developed here is based on growth on inert support microspheres that proliferate and remain viable in a 3D fluid environment using an optimized GTSF-2 medium . Various TLAs systems have been shown to exhibit morphologies , growth characteristics , and cytokine expression similar to human lung , liver , and small intestine [57]–[62] . Therefore , TLAs provide a promising platform to model viral infections of human cells in a 3D environment that is similar to the natural tissue . We have shown previously that human bronchio-tracheal tissue maintained as 3D-TLAs ( HLEMs ) can be used to investigate respiratory syncytial virus ( RSV ) , parainfluenza virus ( PIV3 ) and severe acute respiratory syndrome ( SARS ) [57] , [60] , [63] , [64] infections . HLEMs can be maintained for at least two months and can facilitate the discrimination between pathogenic and attenuated RSV and PIV3 strains based on viral replication efficiencies and inflammatory responses . In addition , SARS was shown to replicate for up to two weeks in HLEMs without TLA destruction [57] , [63] , [64] . The NHNP TLAs we constructed can also be used to investigate other neurotropic viruses ( West Nile , Dengue and Vesicular Stomatitis Virus [VSV] ) [65]–[66 , unpublished data] . As mentioned previously , VZV infection of most human cells results in acute cell death . This also occurs in neuronal cultures maintained in 2D , unless they are devoid of non-neuronal cells . Our 3D neuronal TLA cultures are unique in that VZV infection did not destroy the tissue-like assemblies even though low amounts of infectious virus were sporadically released and even when a mixed neural progenitor cell population was present . In this respect , neuronal progenitor cell cultures maintained in 3D are fundamentally different from the same cells maintained in 2D [17] , [19] . Therefore , this newly established NHNP TLA 3D system provides the unique opportunity to investigate neurotropic viruses in a neuronal setting over extended periods of time .
NHNP cells were obtained from Lonza ( Walkersville , MD , USA ) and propagated in GTSF-2 , a unique media containing glucose , galactose and fructose supplemented with 10% fetal bovine serum ( FBS ) , at 37°C under a 5% CO2 atmosphere [67]–[69] . NHNP cells were initially grown as monolayers in human fibronectin-coated flasks ( BD Biosciences , San Jose , CA ) and pooled from at least five donors , as described previously [70] . NHNP cell cultures were expanded , tested for viral contaminants as pre-certified by the manufacturer's production criteria ( Lonza ) , and cryopreserved in liquid nitrogen . Three-dimensional ( 3D ) NHNP TLAs were generated by seeding 3×105 NHNP cells/ml onto 3 mg/ml Cultispher beads ( Sigma-Aldrich , St . Louis , MO ) into a 55 ml rotating wall vessel bioreactor ( RWV; Synthecon , Houston , TX ) and grown at 37°C under a 5% CO2 atmosphere . Cells were allowed to attach to the beads for 48 h in the bioreactor before re-feeding with GTSF-2 containing 10% FBS . To maintain the TLA cultures within normal human physiological blood chemistry parameters ( pH 7 . 2 and a glucose concentration of 80–120 mg/dL ) , 20–90% of the media was replaced as required with fresh GTSF-2 media every 48 h , facilitating efficient TLA tissue growth and maturation prior to VZV infection . All metabolic determinations were made using an iStat hand held blood gas analyzer ( Abbott Laboratories , Abbott Park , IL ) . Human melanoma cells ( MeWo , American Type Culture Collection , ATCC , Manassas , VA . ) were propagated in Dulbecco's minimal essential medium supplemented with 10% fetal bovine serum , 100 U/ml penicillin and 0 . 1 mg/ml streptomycin ( Sigma-Aldrich , St . Louis , MO ) at 37°C under 5% CO2 [71] . Wild-type and recombinant viruses were passaged on MeWo cells by co-cultivation of infected with uninfected cells at a ratio of 1/5 [72] . MeWo cells for the infection of NHNP cultures were adapted to GTSF-2 medium over two passages prior to harvest of VZV . Cell-free VZV [73] was used for the NHNP TLA infections to avoid transfer of infected MeWo cells to the TLA culture . Briefly , infected cells were harvested at 96 h post-infection ( p . i . ) and resuspended in reticulocyte standard buffer ( 10 mM NaCl , 1 . 5 mM MgCl2 , 10 mM Tris-HCl , pH 7 . 4 ) . Cells were disrupted by Dounce ( type A ) homogenization ( Cole-Parmer , Vernon Hills , IL ) , debris removed by centrifugation at 900×g for 15 min and virus containing supernatant filtered using a 1 . 0 µm Millex filter unit ( Millipore Billerica , MA . ) . NHNP TLAs were infected in the RWV with cell-free VZV at a multiplicity of infection ( MOI ) of 0 . 1 by absorption at room temperature for 30 min in 20 ml GTSF-2 . Then , the RWVs were filled with fresh GTSF-2/10% FBS and transferred to a humidified incubator with a 5% CO2 atmosphere at 37°C . Every 24 h . p . i . , ∼80% of the culture media was replaced with fresh GTSF-2 containing 10% FBS . Samples were collected approximately every other day for ∼70 days to determine viral genome copies as described below . Supernatant of infected NHNP TLAs was removed , centrifuged to remove cells and debris and titrated in fresh MeWo cells to determine the amount of infectious VZV virions in the NHNP culture media . Multicolor mFISH for the analysis of chromosome integrity was conducted as described [74] , [75] . Briefly , NHNP cells and TLAs were cultured as previously stated , with the exception of being subcultured at a low density for 36 h . Cultures were incubated for 20 min with 50 nM calyculin-A ( Waco Chemicals , Japan ) , treated with 0 . 075 M KCl hypotonic solution at 37°C for 20 min , fixed in methanol/acetic acid ( 3∶1 vol/vol ) and stored at −20°C . Chromosome spreads were prepared as described [76] , [77] . Chromosome-containing slides were hybridized with Spectra Vysion Tm probes ( Vysis , Downers Grove , IL ) to detect specific chromosome pairs . Chromosomes were stained with DAPI and analyzed using a Zeiss Axioplan fluorescence microscope . Sample preparation , scanning and analysis of the TLAs was performed as described previously [78] , [79] . Images of NHNP TLAs were taken using a Philips XL 30 ESEM ( FEI Co . , Hillsboro , OR ) , at 150× , 650× and 2000× magnifications to illustrate the complexity and the tissue–like nature of the cultures . Neuronal markers were detected by indirect immunofluorescence using anti-human CXCR4 , CD133 , Nestin , CD105 ( Endoglin ) , CD90 ( Thy-1 ) CD49f ( ITGa6 ) and β-Tubulin-III at the following dilutions: CXCR4 50 µg/ml , CD133 25 µg/ml , Nestin 50 µg/ml , CD105 Endoglin 50 µg/ml , CD90 Thy-1 50 µg/ml , CD49f ( ITGa6 ) 25 µg/ml , β-Tubulin-III 25 µg/ml , CD34+ 25 µg/ml , and CD38 25 µg/ml . All reagents were obtained from R&D Systems ( Minneapolis , MN . ) with the exception of CXCR4 obtained through ABCAM ( Cambridge , MA . ) . Confocal and flow cytometry used the same reagents and confocal analyses were performed as described previously [57] , [60] , [80] . Flow cytometry followed the method of Morrison et al . [81] with minor modifications . Briefly , NHNP TLAs from RWV culture vessels were collected into a 50 mL conical tube and dissociated in 15 mL of 2% trypsin for <5 min , just long enough to dislodge cells . Trypsin was inactivated with 35 mL cold 10% FBS in PBS . Culture beads were allowed to sediment and free floating cells were transferred into a fresh conical tube , washed twice in PBS , centrifuged for 3 min at 600×g . The soft cell pellet was resuspended in GTSF-2 and permitted to recover for 1 h at 37°C . After recovery , viable cells ( trypan blue exclusion ) were counted and aliquots ( ∼1×105 ) were washed 2× in PBS and used for flow cytometry . Fluorochromes were added and incubated 45 min at 4°C in PBS , washed 2× in PBS at 4°C and brought to 400 µL with PBS . All analysis was done on freshly conjugated cells . Samples were either analyzed by confocal microscopy on a Leica TCS/SP2 3-laser confocal microscope or a Beckman Coulter XL2 Flow Cytometer ( Table 1 ) . Collection and use of human cadaver tissue was approved by the University of Colorado Internal Review Board and removed only after obtaining informed consent from next-of-kin ( Multi Institution Review Board Document No . B182 ) . Harvest of cadaver tissues is not considered human research as determined by the IRB . Trigeminal ganglia were fixed in 4% paraformaldehyde overnight , dehydrated through graded ethanol , and embedded in paraffin . Five micrometer sections were collected on Superfrost ( Cole-Parmer , Vernon Hills , IL ) slides and fixed at 72°C for 30 min . Samples were deparaffinized in xylene and rehydrated in graded ethanol . Antigen was retrieved by soaking the sections in 10 mM sodium citrate ( pH 6 . 0 ) in a steamer for 25 min . Slides were blocked in 5% normal goat serum for 1 h at room temperature and then incubated with mouse anti-Nestin monoclonal ( 1∶1 , 000 dilution; BD Transduction Labs , San Jose , CA ) or mouse anti-β-Tubulin-III monoclonal ( 1∶200 dilution; Cell Signaling , Danvers , MA ) antibody overnight at 4°C . Slides were washed in PBS and biotin-conjugated goat anti-mouse IgG ( 1∶1 , 000 dilution; Dako , Carpinteria , CA ) was applied for 1 h at room temperature . After secondary antibody application , slides were PBS washed and incubated with diluted alkaline phosphatase-conjugated streptavidin ( BD Biosciences , San Diego , CA ) for 1 h . The color reaction was developed for 2 min using the fresh fuchsin substrate system ( Dako , Carpinteria , CA ) in the presence of levamisole at a final concentration of 24 µg/ml . All images were acquired using Axiovision ( Zeiss ) digital imaging software with a Nikon Eclipse E800 microscope . DNA and RNA were extracted from cell pellets using DNeasy and RNeasy kits according to manufacturer's protocol ( Qiagen , Valencia , CA ) . Complementary DNA was synthesized from 100 ng of DNase-treated RNA using the Transcriptor System ( Roche Diagnostics , Mannheim , Germany ) as previously described [82] . Specific primers and probes for TaqMan and SYBR PCR were obtained from IDT ( Cedar Rapids , IW ) to quantify VZV ORF9 , ORF 40 , ORF 63/70 and the cellular GAPDH gene . To differentiate between VZV DNA and transcripts containing eGFP or mRFP fused to the 3′-end of ORF63/70 , we used specific primers and probes for each target ( Table 3 ) . PCR amplification using the 7500-Fast real-time PCR system ( Applied Biosystems , Foster City , CA ) was previously described [83] . All primers and probes are shown in Table 3 . VZV ORFs 63 and 70 were C-terminally labeled with the enhanced green fluorescent protein ( eGFP ) or the monomeric red fluorescent protein ( mRFP ) ( Fig . 5A ) using two-step Red-mediated en passant mutagenesis as described [84] . Briefly , the eGFP-I-SceI-aphAI and mRFP-I-SceI-aphAI cassette was amplified from pEP-eGFP-in and pEP-mRFP-in , respectively [84] , [85] with specific primers ( Table 3 ) . PCR products were introduced into pP-Oka , an infectious BAC clone of the P-Oka strain of VZV [86] , [87] by Red recombination performed in GS1783 E . coli cells ( a kind gift from Gregory A . Smith , Northwestern University , Chicago , IL ) . All clones were confirmed by DNA sequencing using primer sets specific for either ORF63 or ORF70 , as well as multiple restriction fragment length polymorphism analyses ( RFLP ) , to ensure the integrity of the genome ( Fig . 5B ) . Recombinant BAC DNA used for transfection was isolated using the plasmid Midi-prep kit according to the manufacturer's instruction ( Qiagen , Valencia , CA ) . MeWo cells were transfected with the Lipofectamine 2000 reagent ( Invitrogen , Carlsbad , CA ) as described [88] . Briefly , MeWo cells were transfected with 1 µg BAC DNA and 200 ng pCMV62 , a plasmid that contains the VZV IE gene ORF62 under the control of the human cytomegalovirus immediate-early promoter ( kindly provided by Dr . Paul Kinchington , University of Pittsburgh Medical School ) . Reconstituted viruses were propagated on MeWo cells as described above and analyzed for expression of the ORF63GFP/70RFP fusion proteins using a Zeiss Axiovert 25 fluorescence microscope system . In addition , VZV containing fluorescent tags at the C-terminus of both ORF63 and ORF70 upon recombination in MeWo cells was plaque purified and some viruses contained only either the eGFP or mRFP tag after plaque purification as shown by PCR and DNA sequencing analysis . MeWo cells ( 1×106 ) were inoculated with 100 plaque-forming units ( pfu ) of cell-associated virus . Infected cells were trypsinized at 24 , 48 , 72 or 96 h post infection ( h . p . i . ) , titrated in 10-fold dilutions , and added to MeWo cell monolayers seeded 24 h before . The number of plaques was determined by indirect immunofluorescence using an anti-VZV antibody exactly as described earlier [89] . Metabolic parameters of infected and uninfected NHNP TLAs were measured every 24 h over the course of the experiments to monitor a cellular development profile and to monitor the metabolic status of the tissues . Glucose consumption was determined using the iStat clinical blood gas analyzer using an EC8+ cartridge ( Abbott Laboratories , Abbott Park , IL ) according to the manufacturer's instructions [60] .
|
Varicella-zoster virus ( VZV ) , the alphaherpesvirus that typically causes childhood chickenpox and shingles in adults , becomes latent in neurons , thus remaining in the body for a lifetime . Unfortunately , few models are available to study the establishment of VZV latency since the virus infects only humans and establishes persistent infections and latency only in neurons , a slowly proliferating , short-lived cell in culture . We have successfully maintained normal human neural progenitor cells ( NHNP ) in tissue-like assemblies ( TLAs ) in 3-dimensional ( 3D ) cultures for up to 6 months . The 3D NHNP TLAs show some characteristics as those found in the human trigeminal ganglia , the site of VZV latency . NHNP TLAs infected with VZV remain viable for 3 months during which time VZV DNA replicates and remains genetically stable , virus genes are transcribed , and infectious VZV is sporadically released . The ability to maintain VZV infected NHNP cells in culture for extended times provides the unique opportunity to study the molecular interactions between this important human pathogen and neuronal tissue to an extent previously unattainable .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"infectious",
"diseases",
"chickenpox",
"shingles",
"viral",
"diseases"
] |
2013
|
Three-Dimensional Normal Human Neural Progenitor Tissue-Like Assemblies: A Model of Persistent Varicella-Zoster Virus Infection
|
We compared the neutralization sensitivity of early/transmitted HIV-1 variants from patients infected by subtype B viruses at 3 periods of the epidemic ( 1987–1991 , 1996–2000 , 2006–2010 ) . Infectious pseudotyped viruses expressing envelope glycoproteins representative of the viral quasi-species infecting each patient were tested for sensitivity to neutralization by pools of sera from HIV-1 chronically infected patients and by an updated panel of 13 human monoclonal neutralizing antibodies ( HuMoNAbs ) . A progressive significantly enhanced resistance to neutralization was observed over calendar time , by both human sera and most of the HuMoNAbs tested ( b12 , VRC01 , VRC03 , NIH45-46G54W , PG9 , PG16 , PGT121 , PGT128 , PGT145 ) . Despite this evolution , a combination of two HuMoNAbs ( NIH45-46G54W and PGT128 ) still would efficiently neutralize the most contemporary transmitted variants . In addition , we observed a significant reduction of the heterologous neutralizing activity of sera from individuals infected most recently ( 2003–2007 ) compared to patients infected earlier ( 1987–1991 ) , suggesting that the increasing resistance of the HIV species to neutralization over time coincided with a decreased immunogenicity . These data provide evidence for an ongoing adaptation of the HIV-1 species to the humoral immunity of the human population , which may add an additional obstacle to the design of an efficient HIV-1 vaccine .
Thirty years after the discovery of the human immunodeficiency virus ( HIV ) , the development of an effective vaccine remains an elusive goal . Experiments of passive immunization and vectored immunoprophylaxis in animal models have shown that human monoclonal ( HuMo ) broadly neutralizing antibodies ( NAbs ) can fully protect against HIV-1 infection [1]–[11] . However the design of an immunogen able to induce NAbs that would mediate potent cross-clade HIV-1 neutralization has not been reached so far . The identification of antibody specificities able to neutralize most of the currently circulating HIV-1 variants remains therefore a major focus of vaccine design . During natural HIV-1 infection , most of the patients develop autologous NAbs at the early stage of infection [12]–[17] . NAbs are directed against the gp120 and gp41 subunits of the viral envelope glycoprotein ( Env ) . The breadth of the autologous response is relatively narrow , as illustrated by its inability to neutralize heterologous isolates [12] , [18]–[20] and the absence or low level of protective activity against superinfection [21]–[23] . These antibodies do not seem to protect against disease progression but exert a selective pressure that drives the viral evolution and leads to the rapid selection of escape Env variants [12] , [13] , [24]–[26] . The molecular basis of HIV-1 escape to autologous neutralization involves multiple mechanisms , including single amino acids substitutions , insertions/deletions in the variable regions of the gp120 and an increased number and/or changing positions of potential N-linked glycosylation sites ( PNGS ) at its surface [13] , [20] , [24] , [27] , [28] . Nevertheless , it has become clear that a substantial number of HIV-1 infected individuals develop NAbs after 2 or 3 years of infection able to neutralize efficiently heterologous primary isolates of various subtypes [29]–[32] . This means that the relevant epitope ( s ) exist toward which a specific response can be mounted , at least in some individuals . Prior to 2009 , only four HuMo broadly Nabs , i . e . b12 , 2G12 , 2F5 and 4E10 , had been isolated from such individuals [33]–[37] . Recently , a “second generation” of HuMoNAbs ( particularly the PG , PGT and VRC series ) that are 10 to 100-fold more potent than the first generation HuMoNAbs were identified [38]–[41] . Several studies suggested that broad and potent neutralizing activity in most of the sera from patients with broadly NAbs arises through a limited number of specificities that correspond to the targets of these HuMoNAbs [42]–[45] . These targets are epitopes located within the surface glycoprotein gp120 . Some of them overlap the CD4 binding site [39] , [46] , [47] and others are more complex , of glycopeptidic nature , composed of conserved glycans and amino-acid residues of the V1 , V2 and V3 loops [48] , [49] . Two years ago , Bunnik et al suggested that HIV-1 might be evolving at the population level towards an enhanced resistance to antibody neutralization , subsequently to the selective pressure exerted by the individual NAbs responses [50] . Comparing HIV-1 variants isolated from patients of the Amsterdam Cohort Studies either early in the epidemic ( 1985–1989 ) or more recently ( 2003–2006 ) , they found an enhanced neutralization resistance of HIV-1 during the course of the epidemic , especially towards HuMoNAbs targeting epitopes of the CD4-binding site [50] , [51] . This finding may have major consequences for vaccine development . Therefore , we conducted the present study in order to both validate and extend the comprehension of the phenomenon . We compared the neutralization sensitivity of HIV-1 subtype B early/transmitted variants issued from French individuals at 3 periods of the epidemic , spanning more than 20 years ( 1987–1991/1996–2000/2006–2010 ) . Their neutralization sensitivity was tested using both polyclonal sera from HIV-1 infected patients and a large and updated panel of 13 HuMoNAbs , including the most efficient NAbs among those described to date . Our results confirm a clear continuous and progressive enhanced resistance to neutralization over time , providing evidence for an ongoing adaptation of the HIV-1 species to the humoral immunity of the human hosts over the course of the epidemic . However and despite this evolution , we found that one combination of two HuMoNAbs still should neutralize the most recently circulating HIV-1 variants , even at a relatively low concentration ( ≤1 µg/mL ) . Therefore , in addition to bring a basic knowledge on the interplay between HIV and the human species , our data provide a rationale for the selection of the HuMoNAbs that should be preferentially used for HIV immunoprophylaxis , especially for the emerging strategy of antibody gene transfer [11] .
The HIV-1 population that we studied was isolated from 40 patients enrolled at time of primary infection in the French ANRS PRIMO and SEROCO cohorts at three periods of the epidemic: between 1987 and 1991 ( Historical patients , HP ) , 1996 and 2000 ( Intermediate patients , IP ) and 2006 and 2010 ( Contemporary patients , CP ) . They were carefully selected to be comparable for each period in the following way: all patients were Caucasian men having sex with men ( MSM ) , infected by clade B viruses . They had similar distribution of viral loads ( median value: 5 . 0 , 5 . 1 and 5 . 2 log10 copies/mL for HP , IP and CP , respectively ) and similar distribution of CD4 T-cell counts ( median value: 507 , 619 and 571 cells/mm3 for HP , IP and CP , respectively ) ( Table S1 ) at time of sample collection . In order to limit both the viral diversity of the quasi-species infecting each patient and the influence of the development of an autologous humoral immune response on the neutralization sensitivity of viruses , blood samples were collected shortly after infection ( before 3 months post-infection except for a few cases ) ( Table S1 ) . The variants that we analyzed were therefore considered as early/transmitted viruses . Pseudotyped viruses expressing envelope glycoproteins ( Env ) variants representative of the viral quasi-species infecting each patient were generated from the entire env gene amplified by RT-PCR from plasma viral RNA . Infectious pseudoviruses were obtained for 11 to 15 patients infected at each of the 3 periods . They were compared for their sensitivity to neutralization by pools of sera from chronically infected patients infected early in the epidemic ( 1987–1991 ) or more recently ( 2003–2007 ) ( Table S2 ) , and by a panel of 13 HuMoNabs targeting major neutralizing epitopes . These included b12 , VRC01 , VRC03 and NIH45-46G54W , an engineered mutant derived from a clonal variant of VRC01 , which target the CD4-binding site of gp120 [39] , [40] , [46] , [47]; PG9 , PG16 and PGT145 which recognize a glycan-dependant quaternary epitope in the gp120 variable loops 1 and 2 ( V1/V2 ) [38] , [41] , [48] , [49]; PGT121 , PGT128 , PGT135 and 2G12 which target gp120 glycan-dependant epitopes within or near the variable loop 3 ( V3 ) of gp120 [35] , [41] , [49] , [52]; and 2F5 and 4E10 which target the membrane-proximal external region ( MPER ) of gp41 [37] , [38] , [53] , [54] . The sensitivity to neutralization was measured using a single round of infection in TZM-bl target cells . The 50% inhibitory concentrations ( IC50 ) were determined for each pool of sera ( reciprocal of dilution ) and for each HuMoNab ( antibody concentration ) . We observed a significantly progressive decrease over time in sensitivity to neutralization of the viruses by the two pools of sera ( 1987–1991 sera , median IC50: 138 . 3 , 54 . 0 and 37 . 8 for HP , IP and CP , respectively , P trend = 0 . 006; 2003–2007 sera , median IC50: 102 . 2 , 54 . 0 and 46 . 2 for HP , IP and CP , respectively , P trend = 0 . 02 ) ( Fig . 1 A and Table S3 ) . At high concentrations ( IC50≥20 ) , the neutralization coverage ( percentage of neutralized viruses ) by the pools of sera was high ( range: 73 . 3% to 100% ) ( Fig . 1 B ) . In contrast , when higher stringent conditions were considered ( IC50≥100 ) , a significant and progressive decrease in the percentage of neutralized viruses was observed over time for the 2003–2007 pool of sera ( P trend = 0 . 03 ) . Although not statistically significant , a similar trend was observed for the pool of sera from the 1987–1991 period ( Fig . 1 B ) . Taken together , these observations exclude the possibility that the lower sensitivity to neutralization of contemporary HIV variants is related to the calendar time from which the sera originated . A decrease in sensitivity to neutralization of the early/transmitted viruses from CP was also observed for VRC01 , VRC03 , NIH45-46G54W and b12 ( Fig . 2 A ) . The median IC50 of VRC01 and VRC03 increased progressively and significantly from 0 . 46 and 0 . 08 µg/mL for HP to 2 . 03 and 5 . 00 µg/mL for CP , respectively ( VRC01: P trend = 0 . 02 , VRC03: P trend = 0 . 002 ) . For NIH45-46G54W , the median IC50 increased from 0 . 08 for HP to 0 . 22 µg/mL for IP but remained stable for CP ( 0 . 18 µg/mL ) ( P trend = 0 . 03 ) . Inversely , the median IC50 of b12 did not differ between HP ( 4 . 58 ) and IP ( 3 . 73 ) but increased for CP ( 46 . 53 ) ( P trend = 0 . 02 ) . In addition , the neutralization breadth of VRC01 , VRC03 and NIH45-46G54W decreased progressively for viruses from HP to CP ( Fig . 2 E ) . This trend was more obvious at concentrations below 1 µg/mL for VRC01 ( P trend = 0 . 05 ) and less than 0 . 1 µg/mL for NIH45-46G54W ( P trend = 0 . 08 ) . The neutralization coverage of b12 was low , even for viruses from HP ( Fig . 2 E ) . HIV-1 variants from CP were also more resistant to PG9 ( median IC50: 2 . 18 , 9 . 30 and 10 . 00 µg/mL for HP , IP and CP respectively , P trend = 0 . 05 ) , PG16 ( median IC50: 0 . 38 , 0 . 95 and 10 . 00 µg/mL for HP , IP and CP respectively , P trend = 0 . 001 ) , and PGT145 ( median IC50: 0 . 02 , 4 . 31 and 10 . 00 µg/mL for HP , IP and CP respectively , P trend = 0 . 0001 ) ( Fig . 2 B ) . The neutralization breadth of these three HuMoNAbs decreased progressively from HP to CP , even at high concentrations ( Fig . 2 E ) . This evolution was particularly evident for PGT145 . The decrease in sensitivity to neutralization of viruses from CP was also observed for PGT121 and PGT128 ( Fig . 2 C ) . The median IC50 of PGT121 increased progressively from HP to CP ( median IC50: 0 . 01 , 0 . 05 and 0 . 32 µg/mL for HP , IP and CP , respectively ) . This trend was just above the limit of significance ( P trend = 0 . 07 ) . For PGT128 , the median IC50 increased from 0 . 02 µg/mL for HP to 0 . 11 µg/mL for IP but remained stable for CP ( 0 . 10 µg/mL ) ( P trend = 0 . 03 ) . The percentages of neutralized viruses by these two antibodies remained high for all subjects at high concentrations ( IC50≤10 µg/mL ) , but tended to decrease progressively from HP to CP when higher stringent conditions of neutralization were considered ( IC50≤1 µg/mL ) . The sensitivity to neutralization by 2G12 or PGT135 was low , even for viruses from HP ( Fig . 2 C and 2 E ) . The sensitivity to neutralization by 2F5 and 4E10 was much lower and no trend of an increased resistance to these HuMoNAbs was observed ( Fig . 2D and 2E ) . Altogether , these results showed that HIV-1 has become more resistant to antibody neutralization over the course of the epidemic . This effect was particularly marked for the pools of sera from HIV-1 infected patients and for the HuMoNAbs that target the CD4-binding site and the glycan-dependant epitopes in V1/V2 and V3 . Since the HIV-1 early/transmitted viruses from the contemporary patients were the most resistant to neutralization by HuMoNAbs , we tried to predict which antibodies combinations would be the most efficient . The IC50 heatmap analysis showed that the engineered antibody NIH45-46G54W was the broadest and most potent antibody ( Fig . 3A ) . It neutralized 12 of 14 viruses , with IC50<1 µg/mL for 11 viruses and between 1 and 10 µg/mL for the last one . An identical neutralization breadth was observed for its natural counterpart , VRC01 , but at a lower potency , only 6 viruses being neutralized with an IC50<1 µg/mL . Interestingly , the two viruses that were resistant to NIH45-46G54W and VRC01 ( issued from CP 590111 and 940218 ) were neutralized by PGT128 efficiently ( IC50<1 µg/mL ) , suggesting that combining NIH45-46G54W , or to a lesser extend VRC01 , with PGT128 would neutralize most or all HIV-1 variants from CP . To test this hypothesis , we investigated the neutralization sensitivity of viruses from CP against a 1∶1 combination of NIH45-46G54W and PGT128 , and a 1∶1 combination of VRC01 and PGT128 . The observed neutralization coverage by various concentrations of these two combinations was compared to the calculated theoretical coverage that would be obtained if the neutralizing activities were fully additive ( Fig . 3 B and C ) . The experimental data showed that the NIH45-46G54W–PGT128 combination was able to neutralize all the viruses with an IC50≤1 µg/mL . At this threshold , the VRC01-PGT128 combination neutralized 80% of the viruses . As shown in Fig . 3 B and C , the observed data reached approximatively the theoretical curves , indicating that these two categories of antibodies may counter HIV with an additive effect . Full-length env sequences of the early/transmitted viral population infecting each subject were obtained by direct sequencing from bulk env PCR products . These 40 sequences were compared to 160 env sequences isolated at the time of primary infection from clade B infected patients with documented year ( 1990–2009 ) and country of infection ( 111 from the United States , 25 from Europe , 6 from Australia , 15 from Trinidad and 1 from Africa ) . The phylogenetic analysis illustrated in the neighbor joining tree indicated that the env sequences of viruses from our study did not cluster in particular branches and were not particularly related to the country of origin or to the study period , historical , intermediate or contemporary ( Fig . 4 ) . It suggests that the biological properties that we describe may be representative of the entire clade B HIV-1 population . As it could be expected , the genetic diversity among the env sequences increased gradually from HP to CP ( mean genetic diversity: 8 . 8% , 12 . 4% and 14 . 8% for HP , IP and CP , respectively ) . It mirrors the global genetic evolution of HIV-1 over the course of the epidemic [55]–[57] , illustrated by the fact that our historical variants are located on shorter branches of the tree when compared to intermediate or contemporary variants ( Fig . 4 ) . It has been described regularly that an increase in both length and number of PNGS of gp120 , particularly in the variable regions , was associated with the evolution of HIV-1 towards resistance to neutralization at the individual level . We therefore compared these variables between variants issued from HP , IP and CP . No significant differences were observed between the 3 groups of subjects ( Fig . 5 ) . We just noticed a slight increase of the global length of gp120 sequences over time ( median aa numbers: 507 , 512 and 513 for HP , IP and CP , respectively ) . Because we observed a progressive increase over time in resistance to neutralization by most of the HuMoNAbs , we investigated if changes in key residues targeted by each of these antibodies might be associated with this evolution . Various amino acid substitutions have been identified at sites targeted by VRC01 , VRC03 , NIH45-46G54W and b12 [39] , [40] , [46] , [47] , [58] . We did not find any potential relationship between the resistance of viruses to neutralization by these antibodies and the presence or absence of specific residues at these key positions ( Fig . 6 A ) . In contrast , as previously reported [41] , [48] , [59] , a few amino acids substitutions were found to coincide with a neutralization resistance to PG9 , PG16 and PGT145 ( Fig . 6 B ) . These included the loss of a PNGS at position 160 either by the lack of N160 ( patients 770203 , 940139 , 750705 , 751002 and 1644 ) or the absence of the glycosylation sequon despite having N160 ( subject 330424 ) ; the presence of an Asn at position 166 ( patients 590111 , 751730 and 130206 ) ; the presence of a Thr at position 169 ( subject 1644 ) ; and the presence of a negatively charged residue ( Glu ) at position 171 ( patients 660118 and 440102 ) . There was a progressive increase in the number of viruses carrying at least one of these substitutions over the course of the epidemic: 1 of 11 ( 9% ) , 4 of 15 ( 27% ) and 6 of 14 ( 43% ) for HP , IP and CP ( P = 0 . 06 , Chi2 test for trend ) . We cannot exclude that other substitutions at these key residues also account for the neutralization resistance of some variants . For instance , the PG9/PG16/PGT145-resistant variant issued from patient 590110 carried three non-homologous substitutions , including the replacement of the positively charged Arg residue at position 166 by a negatively charged Asp residue . In contrast , a few variants were resistant to PG9 ( patients 562 and 920203 ) , PG16 ( patients 60101 and 920203 ) , or PGT145 ( patient 60101 ) , despite having conserved ( or similar ) residues at key positions required for neutralization sensitivity , suggesting that other molecular determinants are involved in resistance to these HuMoNAbs . Inspection of the V3 region targeted by PGT121 and 128 revealed a shift of the PNGS from N332 to N334 in seven neutralization ( or less sensitive ) -resistant variants ( Fig . 6 C ) , a modification already reported as involved in sensitivity to PGT121/128 [49] , [60] . However , neither the frequency of this change , nor other substitutions in V3 allowed to explain the increasing neutralization resistance to PGT121/128 of HIV-1 variants from CP . In order to check whether the evolution of the HIV-1 species towards a higher resistance to neutralization coincided with a poorer capability to induce NAbs , the neutralizing activity of sera from subtype B chronically-infected patients at the two extreme periods of the study , i . e . 1987–1991 ( n = 30 ) and 2003–2007 ( n = 30 ) ( Table S2 ) , was tested towards a panel of six heterologous subtype B isolates ( Table S4 ) . Sera were collected at least 36 months post-infection from untreated MSM infected by subtype B variants . Although , they were selected to be comparable between the two periods , the median time of collection differed by 4 . 5 months between the 1987–1991 sera and those of 2003–2007 ( 43 . 4 versus 38 . 9 months , respectively , P = 0 . 0006 , Wilcoxon test ) . However , the time ranges were very similar ( 34 . 7 to 51 . 6 versus 36 . 2 to 50 . 9 months post-infection for the 1987–1991 and 2003–2007 sera , respectively ) suggesting that the 4 . 5 months difference in median should not have any impact . These patients had similar viral loads ( median: 4 . 2 and 4 . 4 log10 copies/mL in historical and more recently infected patients , respectively ) and similar CD4 T-cell counts ( median: 468 and 464 cells/mm3 for historical and more recently infected patients , respectively ) at the time of sample collection . The panel of heterologous viruses included two clade B primary isolates ( BX08 and 92BR020 ) and four Env-pseudotyped viruses derived from subtype B env clones ( QH0692 . 42 , AC10 . 0 . 29 , RHPA4259 . 7 and REJO4541 . 67 ) , selected for their moderate ( tier 2 ) sensitivity to neutralization [30] , [61] , [62] . Highly neutralization-resistant ( tier 3 ) strains were discarded from our analysis due to the scarcity of sera showing any detectable neutralizing activity towards such viruses ( data not shown ) . For each strain , a lower frequency of detection of NAbs ( IC50 detectable at least at the first serum dilution , i . e . 1∶20 ) was observed among sera from recently infected patients ( 2003–2007 ) compared to those from the earliest period ( 1997–1991 ) ( Fig . 7 A , Table S4 ) . The differences were statistically significant when considering and adjusting for the strain the entire data set ( P = 0 . 001 ) . Similarly , NAbs titers were lower in the serum samples of the recent period when compared to the early period ( Fig . 7 B ) . The differences in Nab titers between the two periods were highly statistically significant when data were analyzed globally ( P = 0 . 0001 , 2way ANOVA test ) . We next compared the breadth of the neutralizing response between the two periods , by calculating the number of strains neutralized by each serum at least at the first serum dilution ( 1∶20 ) . A higher frequency of sera cross-neutralizing at least 4 strains was observed in the earliest period ( 12/30 versus 3/30 sera ) whereas most of the sera from the recently infected patients neutralized a limited number ( 1 to 3 ) of strains ( 19/30 versus 8/30 sera ) ( Fig . 7 C ) . The observed differences were just above the limit of significance ( P = 0 . 08 ) . To consider both the breadth and the intensity ( measured by the NAb titers ) of the neutralizing response , a neutralizing potency score was calculated for each serum ( see methods ) . Lower potency scores were observed for sera of the recent period ( median: 14 . 54 ) when compared to sera of the early period ( median: 29 . 76; P = 0 . 05 , Fig . 7 D ) . Altogether , these data suggest that patients who were infected more recently developed poorer NAb responses than did those who were infected in the earlier period of the epidemic .
There are strong evidences that HIV-1 evolves considerably faster within hosts than it does at the population level [63] , [64] . The preferential transmission of ancestral viruses , stored in long-lived memory CD4+ T cells , may be an important factor contributing to this difference of divergence [65] , [66] . However , several studies suggested that there is still an evolution of HIV-1 at the population level , which may have major implications in public health . Results of a recent meta-analysis are consistent with an increased virulence of HIV-1 over the course of the epidemic [67] . Complementary to these analyses and focusing on the sensitivity to NAbs , Bunnik et al suggested that HIV-1 was becoming more resistant to the humoral immune response of the host at the populational level [50] . This observation has importance for the fundamental knowledge on host-pathogen interactions , but also may have major consequences for prophylactic antibody-based interventions [68] . In the present study , we validated the reality of the phenomenon and bring detailed descriptive elements . This was made possible through the availability of samples collected at time of primary infection at three different calendar periods of the epidemic , spanning more than 20 years , and the recent availability of a large panel of extremely potent HuMoNAbs [38]–[41] . We found a continuous enhanced resistance of HIV-1 to antibody neutralization over time . This effect was observable for both polyclonal antibodies present in pools of sera from HIV-1 chronically infected patients and most of the HuMoNAbs targeting distinct epitopes of gp120 . It included VRC01 , VRC03 , NIH45-46G54W and b12 which target the CD4-binding site , PG9 , PG16 and PGT145 which recognize a glycan-dependant epitope in V1/V2 , and PGT121 and PGT128 which target a glycan-dependant epitope in V3 . In contrast , no increasing resistance was observed for the gp41-directed HuMoNAbs 2F5 and 4E10 . Therefore , our data strengthen the previous results of Bunnik et al [50] , [51] . Interestingly enough , consistent findings were obtained between these two studies despite the use of a different methodology , a peripheral blood mononuclear cells neutralization assay using primary viruses in the previous study versus a TZM-bl assay using pseudotyped viruses in the present study . Both studies focused on the sensitivity to neutralization of early/transmitted viruses , these variants being those selected during primary infection and therefore those that should be preferentially targeted in any prophylactic intervention . In order to limit the bias that could be introduced in such a study , we paid special attention to select samples which were collected as soon as possible after diagnosis of acute infection , or early seroconversion , from a similar transmission group of patients ( MSM ) , with comparable genetic background ( Caucasian ) , and before introduction of any antiretroviral treatment . In addition , our study that covers three periods allows to demonstrate convincingly the dynamic of the enhanced resistance of HIV-1 to the neutralizing response of the host species . Bunnick et al identified an enhanced resistance of HIV-1 to HuMoNAbs targeting the CD4 binding site , b12 and VRC01 , and a trend towards enhanced resistance to PG16 [50] , [51] . Our study revealed that the enhanced resistance of HIV-1 to neutralization at the population level is not restricted to the CD4-binding site but extends to the conformational glycan-dependant epitopes of the newly described potent HuMoNAbs . This was particularly evident for the HuMoNAbs PG9 , PG16 and PGT145 that target the V1/V2 loops , and to a lesser extend for the HuMoNAbs PGT121 and PGT128 that target the V3 loop . The entire set of data suggests that the neutralization escape ( or less sensitive ) variants persist over time either because this evolution does not affect significantly the viral fitness , or more likely because evolution is constrained through the acquisition of compensatory mutations [69] . However there is not yet any clear answer to that question since several recent reports suggested that HIV-1 was able to escape autologous broadly NAbs without any reduction in fitness [70] , [71] whereas others suggested that escape mutations resulted in a decreased ability of the virus to replicate in vivo [72] , [73] . We tried to explore the molecular mechanisms underlying the enhanced neutralization resistance of HIV-1 . We did not find any global changes in the viral envelope when we compared lengths and numbers of PNGS of either the entire gp120 sequences or each variable loop . This contrasted with previous observations made by Bunnik et al . who found that the increased neutralization resistance of HIV-1 coincided with longer V1 regions and more PNGS in this region [50] . The limited number of samples used in both studies may explain the lack of consensus . Alternatively , the difference might be attributed to the nature of the populations , maybe more restricted in the Amsterdam cohort than in our sample of patients enrolled in the entire country . When focusing on residues or regions targeted by each HuMoNAb , we did not identify any signature pattern in sequences from contemporary patients that could explain their increased resistance . Nevertheless , a few substitutions previously described as being associated to resistance were present in several of our resistant viruses . For instance , variants lacking the PNGS residue at position N160 were resistant to PG9 , PG16 and PGT145 , or variants with a shift of a PNGS from position N332 to N334 were resistant to PGT121 and PGT128 [41] , [48] , [49] , [59] , [60] . Although these mutations were present in some viruses , they were not sufficient to explain the observed difference in neutralization sensitivity between viruses from historical and contemporary patients . Thus , some HIV-1 variants were found to be resistant despite having conserved residues at key positions required for neutralization suggesting that other molecular determinants should be involved . Our study focused on subtype B HIV-1 variants issued from French patients . A phylogenetic analysis of the gp160 sequences of these variants with a large series of gp160 sequences issued from clade B variants isolated at the time of primary infection from patients of various geographic origins , suggested that our variants did not belong to a genetically-restricted subset of viruses , but could be considered as representative of the global evolution of clade B viruses worldwide over the three last decades . Taken together , the first study by Bunnik et al . and our independent work , performed in two different countries , suggest that the increased neutralization resistance of subtype B HIV-1 variants over the course of the epidemic is a global phenomenon . We cannot draw any conclusion about the evolution of the neutralization properties of other HIV-1 clades , even if a similar trend can be expected since it can be considered that they have been exposed to similar host selective pressures . Despite the increased neutralization resistance of HIV-1 over time , four HuMoNAbs , NIH45-46G54W , VRC01 , PGT121 and PGT128 , still displayed an important efficacy against the HIV-1 variants from contemporary patients . They neutralized from 78 . 6 to 85 . 7% of variants at concentrations ≤10 µg/mL and from 42 . 9 to 78 . 6% at concentrations ≤1 µg/mL , even if the breadth of VRC01 was approximately 2-fold lower than previously reported [39] , [59] , [60] . In contrast , a lower neutralization breadth was observed for PG9 , PG16 , PGT145 , PGT135 and VRC03 , which neutralized only from 21 . 4 to 57 . 1% of the contemporary HIV-1 variants at concentrations ≤10 µg/mL and from 7 . 1 to 42 . 9% at concentrations ≤1 µg/mL . The neutralization breadth of PG9 , PG16 and PGT145 was 4- to 5-fold lower than in previous reports which included viruses from various calendar periods [38] , [41] , [59] , [60] . In a vaccinal perspective , this emphasizes the importance of having an updated panel of circulating viruses to identify the epitope specificities that must be targeted preferentially . We next sought to determine which combination of HuMoNabs would provide the best neutralization coverage toward the contemporary HIV-1 variants with the best potency . An IC50 heatmap analysis showed that the engineered antibody NIH45-46G54W was the broadest and most potent antibody . It neutralized all HIV-1 variants except two . These two NIH45-46G54W-resistant variants were neutralized potently by PGT128 , suggesting that combining NIH45-46G54W with PGT128 would efficiently neutralize most variants . NIH45-46G54W is an engineered mutant derived from a clonal variant of VRC01 , whose substitution of a tryptophan for a glycine at position 54 was found to increase its potency to about tenfold [40] . As expected , VRC01 , the natural counterpart , showed the same neutralization breadth but a lower potency , suggesting that higher concentrations of a combination of VRC01 with PGT128 would be necessary to achieve similar neutralization . These assumptions were confirmed experimentally using mixtures of NIH45-46G54W and PGT128 and of VRC01 and PGT128 , showing the complementary effect of these antibodies . In addition , our experimental data clearly indicate that the two complementary categories of antibodies do not compete by steric hindrance and may be efficient in “real-life” conditions . These results are encouraging since a recent study showed that this combination neutralized 96% of a panel of 45 viruses of diverse subtypes [60] . However , most of the studied variants included in that study were issued from patients who had been infected more than 15 years ago . Recent advances in vector-mediated immunoprophylaxis suggest that this gene therapy-based approach could be a promising strategy to bypass the natural immune response and deliver broadly NAbs , thus conferring a sterilizing protection [8] , [11] , [74] . In this context , our data support that a NIH45-46G54W–PGT128 combination should be included in future human trials of immunoprophylaxis . In addition , to verify whether the evolution of HIV-1 towards enhanced resistance to neutralization has coincided with a poorer capability to induce NAbs , we compared the neutralizing activity of sera from 60 HIV-1 infected patients enrolled at the two extreme periods of the study ( 1987–1991; 2003–2007 ) . Again , we took a special attention to select samples from patients that would be as comparable as possible . Serum samples were collected at least 3 years ( extremes: 35–52 months ) after diagnosis of acute infection from patients of similar transmission group ( MSM ) , infected by a clade B virus , with comparable genetic background ( Caucasian ) , and before introduction of any antiretroviral treatment . This delay was selected based on the facts that NAbs , when present , are usually detected after 2 to 3 years , and that many patients were treated passed this delay . We observed a significant reduction of the neutralizing activity of sera from individuals infected later in the epidemic ( 2003–2007 ) ( lower frequency/titers of heterologous NAbs and reduced neutralization breadth ) when compared to that of sera from patients infected earlier in the epidemic ( 1987–1991 ) . Although the median time of collection differed by 4 . 5 months between the 1987–1991 sera and those of 2003–2007 it can be postulated that a so slight difference should not have any impact on the results . As similar observations were reported by Bunnik et al . [50] , the data suggest collectively that the increasing resistance of the HIV-1 viral species to antibody neutralization would be associated to a lowered immunogenicity . In conclusion , our results confirm a clear continuous and progressive enhanced resistance of HIV-1 to neutralization over time , providing evidence for an ongoing adaptation of the HIV-1 species to the humoral immunity of the human hosts over the course of the epidemic . However , despite this HIV-1 evolution , we found that one combination of two HuMoNAbs still should neutralize all the most recently circulating clade B variants , even at a relatively low concentration . These data provide a rationale for the selection of the HuMoNAbs that should be preferentially used for HIV immunoprophylaxis . They also suggest that a regular surveillance of the sensitivity to neutralization of the most recent transmitted variants will be necessary in the future , especially if this drift toward an enhanced resistance is also observed for other prevalent HIV-1 clades in other regions of the world .
Ethic national committees approvals were obtained for the two cohorts [SEROCO: Commission Nationale de l'Informatique et des Libertés ( CNIL ) ; PRIMO: Comité Consultatif de Protection des Personnes dans la Recherche Biomédicale ( CCPPRB ) Paris-Cochin and Comité de Protection des Personnes ( CPP ) Ile de France III] and all patients gave written informed consent to participate in the cohort . The evolution of the neutralization sensitivity of variants present at time of HIV-1 primary infection was analyzed on plasma samples collected from historical ( HP ) , intermediate ( IP ) and contemporary patients ( CP ) enrolled in two cohorts . HP were enrolled in the ANRS SEROCO CO2 cohort [75] whereas IP and CP were enrolled in the ANRS PRIMO CO6 cohort [76] . In order to be as comparable as possible , samples were selected from MSM , all Caucasian and infected by a subtype B virus . Viruses from 40 patients were studied: 11 infected between 1987 and 1991 ( HP ) , 15 infected between 1996 and 2000 ( IP ) and 14 between 2006 and 2010 ( CP ) ( Table S1 ) . The estimated date of infection was defined as the onset of symptoms minus 15 days for patients with symptomatic primary infection , or the date of incomplete western blot ( presence of antibodies to gp160 and P24 ) minus 1 month , or the midpoint between a negative and a positive ELISA result for asymptomatic patients [77]–[79] . The evolution of HIV immunogenicity was studied using serum samples from clade B-infected patients enrolled in the same two cohorts . They were infected either between 1987 and 1991 ( n = 30; early period ) or between 2003 and 2007 ( n = 30; late period ) . All patients were MSM , of Caucasian origin and none of them received an antiretroviral therapy before the date of serum samples used for this study . The selected serum samples were collected at least three years after infection ( Table S2 ) . This delay was chosen for two reasons . First , it is usually considered that 3 years are sufficient to detect NAbs . Second , many patients started to be treated after this delay . As shown in Table S2 , the delay since infection , the virus load and the CD4+ T-cell count at time of sample collection were similar in both groups of patients . HIV-1 RNA was extracted from plasma using the QIAamp Viral RNA Mini Kit ( Qiagen ) . Full length ( gp160 ) env genes were amplified by nested RT-PCR using group M env-specific degenerated primers . The outer primers pair was ExtM5 ( 5′-ATGGCAGGAAGAAGCGGARRC-3′ ) and ExtM3 ( 5′-CTTRTAAGTCATTGGTCTTAAA-GGYAG-3′ ) and the inner primers pair was IntM5XE ( 5′-AATTCTCGAGAATTCAGAAAGAGCAGAAGACAGTGGCAATG-3′ ) containing XhoI ( italicized ) and EcoRI ( underlined ) sites and IntM3MX ( 5′-GGCCACGCGTCTAGACTACTTTTTGACCACTTGCCMCCCAT-3′ ) containing MluI ( italicized ) and XbaI ( underlined ) sites . Reverse transcription ( RT ) was carried out using outer primer ExtM3 and the Superscript III First strand synthesis system ( Invitrogen ) . It was followed by the first round of PCR , using the Platinium PCR SuperMix High Fidelity ( Invitrogen ) with the following conditions: 2 min at 94°C , then 35 cycles of 15 s at 94°C , 30 s at 55°C and 3 min at 68°C , and a final extension step of 10 min at 68°C . A 5 µl aliquot of the products from the first round of PCR was then used as template for the second round of amplification under the same cycling conditions . The amplification products were libraries of env genes that represented the diversity of the viral env sequences present in the patient population . Each fragment was approximately 2 . 6 kb in length , spanning the entire open reading frame of the HIV-1 gp160 polyprotein . PCR amplification products were digested with XhoI and XbaI or XhoI and MluI restriction enzymes ( New England Biolabs ) , purified by agarose gel electrophoresis , and ligated into XhoI and XbaI or XhoI and MluI digested pCI mammalian expression vector ( Promega ) . The resulting pCI-env plasmids representing the amplified virus populations were propagated by transformation of Electromax DH5α electrocompetent Escherichia coli ( Invitrogen ) . Library of pCI-env plasmids were purified from transformed cultures using silica column chromatography ( Macherey-Nagel ) . Env-pseudotyped viruses were produced as previously described [23] by cotransfecting 3×106 293T cells with 4 µg of each patient-derived pCI-env library or of each pCDNA3 . 1-env clone ( reference env clones of clade B with moderate - tier 2 - sensitivity to neutralization: QH0692 . 42 , AC10 . 0 . 29 , RHPA4259 . 7 or REJO4541 . 67; NIH AIDS Reagent Program ) and 8 µg of pNL4 . 3 . LUC . R_E_[80] using FuGene-6 transfection reagent ( Promega ) . Virus stocks were harvested 72 h later , purified by filtration ( 0 . 45 µm filter ) and stored as aliquots at −80°C . Viral infectivity was monitored by infection of 1×104 TZM-bl cells , with serial fivefold dilutions of viral supernatants in quadruplicate , in the presence of 30 µg/ml DEAE-dextran . Infection levels were determined after 48 h by measuring the luciferase activity of cell lysates using the Bright-Glo luciferase assay ( Promega ) and a Centro LB 960 luminometer ( Berthold Technologies ) [81] . Wells producing relative luminescence units ( RLU ) >2 , 5 times the background were scored as positive . The TCID50 was calculated as described previously [82] . The primary isolates BX08 and 92BR020 were used as reference tier 2 strains [30] , [61] , [62] . Virus stocks were prepared by passaging isolates once or twice on phytohemagglutinin-stimulated peripheral blood mononuclear cells and stored as aliquots at −80°C . Viral infectivity was monitored as described above for Env-pseudotyped viruses . Sensitivity to heterologous sera and/or to HuMoNAbs of the pseudotyped viruses and of the primary isolates was assessed in duplicate in TZM-bl cells [83] , [84] . After titration , virus stocks were diluted to 400 TCID50/mL . Aliquots of 50 µL were then incubated for 1 h at 37°C with 50 µL of either 3-fold serial dilutions of heat inactivated serum samples ( 1∶20 to 1∶540 ) , b12 , 4E10 , 2G12 , 2F5 ( 50 µg/ml to 0 . 022 µg/ml , Polymun Scientific ) , or PGT121 , PGT128 , PGT135 , PGT145 , PG9 , PG16 , NIH45-46G54W , VRC01 and VRC03 ( 10 µg/ml to 0 . 0046 µg/ml , IAVI and NIH AIDS Reagent Program ) . The virus-antibody mixture was then used to infect 10 , 000 TZM-bl cells in the presence of 30 µg/ml DEAE-dextran . Infection levels were determined after 48 h by measuring the luciferase activities of cell lysates , as described above . Results were expressed as mean values . IC50 values were defined as the reciprocal of the serum dilution or the antibody concentration required to reduce RLUs by 50% . For immunogenicity studies , the potency score of each serum was calculated by summing the IC50 values obtained for each virus divided by the median IC50 value for each virus across all serum samples [85] . All env PCR products were sequenced according to the Dye Terminator cycle sequencing protocol ( Applied Biosystems , Foster City , Calif . ) . All sequences have been submitted to GenBank and assigned accession numbers KC699001 to KC699040 . Sequence alignments were performed using ClustalW in the software package of BioEdit 7 . 1 . 11 and edited manually . Amino acid positions were identified by the use of standard HxB2 numbering . Potential N-linked glycosylation sites ( PNGS ) were identified using N-Glycosite tool at the HIV database website ( http://www . hiv . lanl . gov/content/sequence/GLYCOSITE/glycosite . html ) . The env sequences corresponded to the population-based sequencing for each patient . Extensive studies have previously shown that the viral population present early after infection is highly homogeneous , in most of the cases represented by only one or a few variants [56] , [86] . Although the goal of our study was not to analyze in depth the diversity of the viral population in each patient , our careful examination of the sequencing profiles suggested that the population was highly homogeneous in each case . Phylogenetic analysis was inferred using the Minimum Evolution method [87] . The tree was drawn to scale , with branch lengths in the same units as those of the distances used to infer the phylogenetic tree . The distances were computed using the Maximum Composite Likelihood method [88] and are in the units of the number of base substitutions per site . Evolutionary analyses were conducted in MEGA5 [89] . A total of 200 nucleotide env sequences were included in the analysis . Forty sequences were derived from our study and 160 sequences were downloaded from the HIV database website ( http://www . hiv . lanl . gov/ ) . These sequences were selected based on the following criteria: they were issued from clade B infected patients at time of primary infection , with a documented year of isolation between 1990 and 2009 and a documented geographical origin ( 25 were from Europe , 113 from USA , 6 from Australia , 15 from Trinidad and 1 from Zambia ) . There was only one sequence per patient . Differences in IC50 values , length of envelope regions and numbers of PNGS between viruses from HP ( 1987–1991 ) , IP ( 1996–2000 ) and CP ( 2006–2010 ) were evaluated using a non-parametric Jonckheere-Tersptra test for trend . For calculations , viruses with IC50 >10 or <0 . 0045 ( for PGT121 , PGT128 , PGT135 , PGT145 , PG9 , PG16 , NIH45-46G54W , VRC01 and VRC03 ) , >50 or <0 . 02 ( for b12 , 4E10 , 2G12 , 2F5 ) , <20 or >540 ( for sera ) were assigned a value of 10 or 0 . 0015 , 50 or 0 . 0076 , and 6 . 66 or 540 , respectively . The evolution of the neutralization coverage of viruses from HP to CP by the pools of sera and the HuMoNAbs was evaluated using a Chi2 test for trend . The frequency of detection and the titers of NAbs among sera from patients infected in 1987–1991 compared to those from 2003–2007 were tested by Chi2 or ANOVA tests after adjustment for the strain ( n = 6 strains ) , respectively . Finally , the neutralization breadth and the potency score of the two groups of sera were compared using a Wilcoxon signed ranked test .
|
Most of the patients develop autologous neutralizing antibodies ( NAbs ) during HIV-1 infection . These NAbs drive the viral evolution and lead to the selection of escape variants at the individual level . The aim of our study was to check if , subsequently to the selective pressure exerted by the individual NAbs responses , the HIV-1 species has evolved at the population level towards an enhanced resistance to antibody neutralization . By comparing HIV-1 subtype B variants collected at three periods spanning more than 2 decades , we found a significantly progressive enhanced resistance to neutralization of the HIV-1 species over time . In addition , the enhanced resistance of the HIV species to neutralization coincided with a decreased capability of the virus to induce NAbs in infected patients . Despite this evolution , one combination of two human monoclonal broadly NAbs still were able to neutralize the most recent HIV-1 variants , suggesting that this combination should be preferentially included in future human immunoprophylaxis trials .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"humoral",
"immunity",
"medicine",
"infectious",
"diseases",
"viral",
"immune",
"evasion",
"immunity",
"hiv",
"virology",
"immunology",
"biology",
"microbiology",
"viral",
"diseases",
"viral",
"evolution",
"immunoglobulins"
] |
2013
|
Evidence for a Continuous Drift of the HIV-1 Species towards Higher Resistance to Neutralizing Antibodies over the Course of the Epidemic
|
Decision-making is usually accompanied by metacognition , through which a decision maker monitors uncertainty regarding a decision and may then consequently revise the decision . These metacognitive processes can occur prior to or in the absence of feedback . However , the neural mechanisms of metacognition remain controversial . One theory proposes an independent neural system for metacognition in the prefrontal cortex ( PFC ) ; the other , that metacognitive processes coincide and overlap with the systems used for the decision-making process per se . In this study , we devised a novel “decision–redecision” paradigm to investigate the neural metacognitive processes involved in redecision as compared to the initial decision-making process . The participants underwent a perceptual decision-making task and a rule-based decision-making task during functional magnetic resonance imaging ( fMRI ) . We found that the anterior PFC , including the dorsal anterior cingulate cortex ( dACC ) and lateral frontopolar cortex ( lFPC ) , were more extensively activated after the initial decision . The dACC activity in redecision positively scaled with decision uncertainty and correlated with individual metacognitive uncertainty monitoring abilities—commonly occurring in both tasks—indicating that the dACC was specifically involved in decision uncertainty monitoring . In contrast , the lFPC activity seen in redecision processing was scaled with decision uncertainty reduction and correlated with individual accuracy changes—positively in the rule-based decision-making task and negatively in the perceptual decision-making task . Our results show that the lFPC was specifically involved in metacognitive control of decision adjustment and was subject to different control demands of the tasks . Therefore , our findings support that a separate neural system in the PFC is essentially involved in metacognition and further , that functions of the PFC in metacognition are dissociable .
Decision-making is a process of evidence accumulation . That evidence may come from sensory signals of external stimuli or from mental representations of internal cognitive operations . Variations in evidence can create uncertainty in the person rendering a decision . The decision maker is normally explicitly or implicitly aware of uncertainties about a decision and consequently confirms or revises a decision even prior to , or in the absence of , external feedback . In the framework of cognitive control , the processes of decision uncertainty monitoring—and consequent decision adjustments—are termed metacognition , that is , ‘cognition about cognition’ [1–4] . Although metacognition generally accompanies decision-making with uncertainty , the underlying neural system of the metacognitive processes in decision uncertainty monitoring and consequent decision adjustments remains less clear than that of the decision-making process per se [5 , 6] . Much of the work on the neural bases of metacognition in humans has focused on metacognitive monitoring of internal states ( i . e . , confidence or uncertainty ) with regard to the cognitive processes such as episodic memory [7 , 8] and sensory perception [9 , 10] . Behaviorally , confidence ratings , which reflect subjective accuracy beliefs regarding decisions , have often been found to deviate from the accuracy of an actual decision [11–13] . These observations have suggested the existence of a separate neural processing system ( meta-level ) in the generation of decision confidence or uncertainty , independent of the decision-making process per se ( object-level ) . We hereafter refer to this description of metacognition as separable from decision-making as “Theory 1” [11–18] . The prefrontal cortex ( PFC ) has been proposed to play a critical role in metacognition [14] , and it has been demonstrated that interference with or lesions in PFC regions may impair metacognitive monitoring of perceptual decisions , but not decisions per se [15–18 , but see also 19] . A contrary theory , which we will refer to as “Theory 2 , ” suggests that metacognition may be merely dependent on the decision-making process and therefore exclusively reliant on accumulated evidence [20–24] . Specifically , this theory , based on bounded accumulation models , has interpreted divergence between decision accuracy and confidence reports as being caused by the accumulation of postdecisional evidence during the interval between decision-making and confidence reporting [20–24] . Furthermore , it implies that decision adjustment naturally occurs as a part of this continuous postdecisional evidence accumulation and therefore is an integrated part of the initial decision-making process [21 , 24] . Some proponents of this theory have argued that a separate neural system for metacognition to monitor and control decision-making should not be necessary because the processes are interdependent [24]; however , not all work supporting this theory insists on this notion [20] . Thus , one of the crucial issues in the debate between the two theories is whether a separate neural system for metacognition exists . Single-decision paradigms ( depicted in Fig 1A and 1C ) are not sufficient to determine the existence or nonexistence of separable systems because the decision-making process and the metacognitive process are inevitably coupled in such tasks . The purpose of retrospective metacognition is to confirm or revise foregone decisions . Given an opportunity to make a decision on the same situation again ( i . e . , make a “redecision” ) ( as depicted in tasks shown in Fig 1B and 1D ) , a decision maker may revise an initial decision as well as confidence in the decision once s/he detects uncertainty regarding the initial decision [25] . Thus , if a separate neural system for metacognition exists as proposed by Theory 1 , the metacognitive processes—in particular metacognitive control—should be more extensively involved in redecision , especially if the initial decision must be made quickly . On the contrary , the neural system involved in redecision should be the same as those involved in an initial decision if there is not a separate neural system for metacognition ( as proposed by Theory 2 ) . If a separate neural system for metacognition exists , the activity of this system should be manifest after an initial decision is reached , whereas Theory 2 suggests that they share the same underlying neural systems and that neural activity following either a single decision or a redecision should be the same . Therefore , comparing behavioral and neural differences between the two phases of initial decision and redecision may allow us to test which theory better accounts for the neural processing of metacognition . A specific perspective of metacognition derived from Theory 1 implies that decision uncertainty , rather than decision confidence , should be the key signal for metacognition . If there is no uncertainty regarding a decision , it should not evoke the processes of metacognitive monitoring and control . Therefore , the critical aim of this study was to elicit and analyze neural activity positively correlated with decision uncertainty , rather than that positively correlated with decision confidence . In the present study , we employed a novel “decision–redecision” experimental paradigm to investigate neural activity associated with metacognition . The participants were asked to make two consecutive decisions on the same situation using a perceptual decision-making task and a rule-based decision-making task ( Fig 2A ) . We combined this novel paradigm with the functional magnetic resonance imaging ( fMRI ) technique to formally test the two theories and systematically investigate the underlying neural substrates of metacognitive processes accompanying decision-making . Based on our previous study [25] , we expected that the frontoparietal control network would be associated with metacognitive processing . In the current study , we focused on specific functions of the regions in the network believed to be involved in metacognition . We found that dorsal anterior cingulate cortex ( dACC ) activity significantly correlated with metacognitive monitoring of decision uncertainty and that lateral frontopolar cortex ( lFPC ) activation correlated instead with metacognitive control . These findings provide evidence for distinct neural processes involved in metacognition and decision-making .
We developed a novel decision–redecision paradigm for this study ( Fig 2A ) . The participants made an initial decision ( decision phase ) , immediately followed by another decision on the same situation ( redecision phase ) . This allowed the participant the opportunity to revise the initial decision and update their confidence rating , even without feedback . The internal states of uncertainty regarding initial and final decisions were separately evaluated by confidence ratings . Confidence was rated on a scale of 1 to 4 , immediately following the corresponding decisions . Decision uncertainty was then the inverse of the confidence rating ( i . e . , a confidence rating of 4 corresponded to an uncertainty rating of 1 ) . Critically , our task differed from previous paradigms used to analyze ‘change of mind’ [21 , 24] . Previous task paradigms were only able to analyze the small portion of trials in which a participant happened to change their mind , while our paradigm allowed analysis of each trial . We used two different types of decision-making tasks in the present study: one was a rule-based decision-making task ( Sudoku ) , the other a perceptual decision-making task ( random-dot motion [RDM] ) , which has commonly been used to investigate the neural process of decision-making [5] and more recently , metacognition [21 , 22 , 24 , 26] . The decisions in the Sudoku task rely on internal informational operations , but decisions in the RDM task should be more dependent on accumulation of external information . It is possible to continue accumulating evidence from external stimuli that may affect decision-making in the RDM task , but that is less likely in the Sudoku task because it is rule based . For this reason , the two tasks should result in differential processing in metacognitive control to adjust initial decisions . The sequences of both tasks were identical ( Fig 2A , illustrated for the main fMRI experiment [fMRI1] ) . After a Sudoku problem or RDM stimulus was presented for 2 s , the participant made a choice from 4 possible solutions within 2 s and then reported their confidence rating on that decision within 2 s . A critical feature of our paradigm was that the same Sudoku problem or the same RDM stimulus was immediately repeated for 4 s , and the participant again made a choice within 2 s and again reported their confidence rating within 2 s . To better distinguish the metacognitive process from the decision-making process , we intentionally set a short initial decision phase ( 2 s ) , to minimize metacognition during the initial decision-making phase , but set a longer duration in the redecision phase ( 4 s ) to allow enough time for metacognitive processing in redecision . There was no explicit feedback or cue to indicate whether the decision was correct after either the initial decision or the redecision . For both tasks , the task difficulty of each trial ( Fig 2B ) was adaptively adjusted by a staircase procedure [9 , 27] so that the average accuracy for the initial decision was converged to approximately 50% ( chance level was 25% ) . For the control condition , the participant was shown a digital number in the target grid in the Sudoku task , and for the RDM task , s/he was shown an RDM stimulus with 100% coherence . For the former , the participant only needed to press the button matching the number , and for the latter , the participant indicated the unambiguous RDM direction . Prior to the experimental testing , the participant was trained to attain high-level proficiency in Sudoku problem-solving . The current study was composed of 4 fMRI experiments: Twenty-one participants took part in fMRI1 ( see Materials and methods ) . In both the Sudoku and RDM tasks , decision uncertainty levels were largely consistent with the percentage of incorrect initial decisions ( Fig 2C; Pearson’s r = 0 . 76 ± 0 . 12 [mean ± SD] , one-tailed t test , t21 = 7 . 3 , P = 1 . 7 × 10−7 in the Sudoku task; r = 0 . 71 ± 0 . 14 , t21 = 6 . 8 , P = 5 . 0 × 10−7 in the RDM task ) . To examine the trial-by-trial consistency between objective erroneous decisions and subjective decision uncertainty levels in individual participants , a nonparametric approach was employed to construct the receiver operating characteristic ( ROC ) curve by using the decision uncertainty levels as thresholds to characterize the likelihood of erroneous decisions . The area under curve ( AROC ) was then calculated to represent the individual uncertainty sensitivity , indicating how sensitive the participant was to the decision uncertainty [9] . As observed in the previous studies , the uncertainty sensitivity of individual participants markedly deviated from decision accuracy in both tasks , which were controlled around 50% ( Fig 2D ) . The response times ( RTs ) of option choices in the initial decision were positively correlated with the decision uncertainty levels ( Fig 2E; t21 = 6 . 9 , P = 4 . 0 × 10−7 in the Sudoku task; t21 = 4 . 3 , P = 1 . 6 × 10−4 in the RDM task ) . The correlation coefficient between RT of option choices and the decision uncertainty level ( rRT-uncertainty ) in the initial decision was highly correlated with the uncertainty sensitivity ( AROC1 ) across the participants ( Pearson’s r = 0 . 61 , z test , z = 3 . 4 , P = 4 . 0 × 10−4 in the Sudoku task; r = 0 . 48 , z = 2 . 4 , P = 0 . 0085 in the RDM task ) . Thus , the RT–uncertainty correlation also reflected individual uncertainty sensitivity . The level of decision uncertainty was reduced by redecision . The extent of decision uncertainty reduction via redecision was highly correlated with the decision uncertainty level in the initial decision phase ( Fig 2F; Goodman and Kruskal’s γ = 0 . 82 ± 0 . 11 , t21 = 8 . 8 , P = 2 . 1 × 10−8 in the Sudoku task; γ = 0 . 78 ± 0 . 14 , t21 = 7 . 7 , P = 8 . 2 × 10−8 in the RDM task ) . Accordingly , accuracy also improved along with uncertainty reduction ( Fig 2G; Pearson’s r = 0 . 54 ± 0 . 13 , t21 = 4 . 2 , P = 2 . 3 × 10−4 in the Sudoku task; r = 0 . 39 ± 0 . 14 , t21 = 2 . 8 , P = 5 . 6 × 10−3 in the RDM task ) . One could suspect that the improvement of uncertainty reduction and the change in accuracy in the redecision phase were caused by regression towards mean in the two separate decisions: higher uncertainty at the first measurement by chance would increase improvement at the second measurement . However , the decision accuracy and decision uncertainty levels for the final decision-making phase remained significantly differential ( Pearson’s r = 0 . 35 ± 0 . 15 , t21 = 2 . 1 , P = 0 . 032 in the Sudoku task; r = 0 . 36 ± 0 . 14 , t21 = 2 . 6 , P = 8 . 9 × 10−3 in the RDM task in Fig 2C; Pearson’s r = 0 . 32 ± 0 . 14 , t21 = 2 . 0 , P = 0 . 042 in the Sudoku task; r = 0 . 32 ± 0 . 15 , t21 = 2 . 2 , P = 0 . 028 in the RDM task in Fig 2G ) , indicating that the participants’ performance in redecision reflected metacognitive processing ability rather than chance . Despite the fact that both decision accuracy and decision uncertainty levels were improved in the redecision phase , the divergence between uncertainty sensitivity and decision accuracy remained significant ( Fig 2H ) . Indeed , neither the individual uncertainty sensitivities nor those of individual differences were altered by redecision ( Fig 2I; t21 = 0 . 82 , P = 0 . 21 in the Sudoku task; t21 = 1 . 0 , P = 0 . 15 in the RDM task ) . Similarly , neither the individual RT–uncertainty correlation coefficients nor those of individual differences were altered by redecision ( Fig 2E; t21 = −0 . 77 , P = 0 . 22 in the Sudoku task; t21 = 0 . 35 , P = 0 . 36 in the RDM task ) . These results show that individual uncertainty sensitivity was stable , was intrinsic to individual metacognitive ability , and was independent of the accumulated evidence and the type of decision-making required . Commonly across both tasks , brain activation during the initial decision phase was mainly restricted to brain areas posterior to the PFC , in particular the posterior portion of the PFC , the inferior frontal junction ( IFJ ) ( S1A Fig; Fig 3A ) . In the redecision phase , a frontoparietal control network—consisting of the lFPC , dACC , anterior insular cortex ( AIC ) , middle dorsolateral PFC ( mDLPFC ) , and anterior inferior parietal lobule ( aIPL ) —was more extensively recruited ( Fig 3B; S1B and S2 Figs; S1 Table ) . In contrast , the lFPC and mDLPFC regions of the frontoparietal control network were not activated when a new Sudoku problem or a new RDM stimulus was presented for the first time during the redecision phase , preceded by the control stimuli in the initial phase ( fMRI2 , n = 17; S1C and S3 Figs ) , while the dACC activity during the same phase became much weaker , and its response onset was much delayed from the onset of the stimulus presentation ( delay offset >3 s; S3 Fig ) . Thus , the frontoparietal control network , in particular the regions of the lFPC , mDLPFC , and dACC in the anterior FPC , were more extensively involved in redecision than in the initial decision phase . Trial-by-trial activity in the regions of the frontoparietal control network in redecision was positively correlated with decision uncertainty level for the initial decision ( Fig 3C and S2 Table ) . Critically , these correlations remained significant even for the correct trials ( S1E Fig ) , indicating that these regions were encoding the decision uncertainty signal rather than the error signal . Furthermore , task difficulty or RT could not explain their association with the decision uncertainty in these regions . The residual fMRI signal changes after the components associated with the task difficulty and RT were regressed out remained highly correlated with decision uncertainty level , but residual fMRI signal changes after the components of the decision uncertainty level were regressed out were not further correlated with the task difficulty and RT . Although the dACC and AIC regions were also partially activated during the initial decision phase ( S1A Fig and Fig 3A ) , this activity—as well as in other regions activated during the same phase—was neither positively nor negatively correlated with the decision uncertainty level ( S1D Fig ) . Activity in the ventromedial PFC ( VMPFC ) and posterior cingulate cortex ( PCC ) regions of the default-mode network in redecision were negatively correlated with decision uncertainty level or positively correlated with its inverse—decision confidence ( S1F Fig ) . Thus , the regional activity seen in the frontoparietal control network involved processes intrinsic to redecision but not the activity involved in decision-making for the initial phase . In the third fMRI experiment ( fMRI3 , n = 25 ) , we confirmed that the strength of activity in the frontoparietal control network depended critically on whether redecision was required after the initial decision phase . When decision uncertainty levels for initial decisions were matched in the two conditions ( two-tailed paired t test , t25 = 0 . 62 , P = 0 . 27 ) , activity in the frontoparietal control network was much stronger when redecision was required ( ‘redecision condition’ ) , in comparison with those when redecision was not required ( “non-redecision condition” ) ( Fig 3D ) , despite the fact that activation of the frontoparietal control network in the ‘non-redecision condition’ was also significant ( S1G Fig ) and was correlated with decision uncertainty level as well [25] . Thus , the frontoparietal control network , more strongly activated in redecision , should not only be involved in metacognitive monitoring of decision uncertainty of the initial decision but also in metacognitive control in redecision ( Fig 1D ) . We then putatively defined this frontoparietal control network as the metacognition network . Because the duration of the redecision phase in fMRI1 was longer ( 4 s ) than that of the initial decision phase ( 2 s ) , it raised the question of whether the fMRI activity predominately observed during the redecision phase was induced by the longer exposure , specifically in the trials with more difficult decisions . To address this , we scanned an independent group of participants ( fMRI4 , n = 20 ) while they underwent the same RDM task as fMRI1 except that the duration of redecision was set to 2 s . The same behavioral and neural results were replicated as in fMRI1 ( S4 Fig ) . Just as the extent of uncertainty reduction by redecision was found to be highly correlated with the decision uncertainty level of the initial decision ( Fig 2F ) , activity in the regions of the metacognition network were also found to be positively correlated with the extent of uncertainty reduction ( S1H Fig ) . However , the strength of the correlations decreased somewhat after the components associated with the decision uncertainty level were regressed out ( S1I Fig ) . Conversely , correlations with decision uncertainty level in the metacognition network remained significant after the components associated with the extent of uncertainty reduction were regressed out ( S1J Fig ) . These partial correlations complementarily confirmed that the metacognition network in redecision was involved in both metacognitive monitoring and metacognitive control , indicating that the two processes interacted in redecision processing . The two processes , although interactive , can be dissociated . In the region involved in uncertainty monitoring , activity strength should dynamically represent decision uncertainty level . As decision uncertainty levels were reduced by redecision , the strength of its activity should accordingly be reduced . Therefore , the neural activity change should be negatively correlated with the extent of decision uncertainty reduction . Alternatively , in the region that was critically involved in metacognitive control , its activity should become positively correlated with the extent of decision uncertainty reduction , representing the outcome or the extent of metacognitive control . We found that the activity in the dACC and AIC regions at the late phase of redecision did in fact negatively correlate with the extent of decision uncertainty reduction after the components associated with the decision uncertainty level of the initial decision were regressed out ( Fig 4A , S1K and S1L Fig ) . Conversely , the lFPC activity in the Sudoku task was positively correlated with the extent of decision uncertainty reduction after components associated with the decision uncertainty level were regressed out ( Fig 4B ) , but negatively in the RDM task ( Fig 4B and S1I Fig ) . In addition , VMPFC activity was also positively correlated with the extent of decision uncertainty reduction in both tasks ( S1I Fig ) . The regional activity in the default-mode network appeared intrinsically anticorrelated with the regional activity in the metacognition network ( further detail regarding activity in the default-mode network associated with metacognition will be discussed in another study ) . Thus , the dACC and AIC regions were specifically involved in metacognitive monitoring . In contrast , the lFPC was specifically involved in metacognitive control in redecision , particularly in the Sudoku task . Therefore , their functional roles in metacognition appear to dissociate in redecision processing . In the Sudoku task , whether the problem would be better solved should be conditioned to individual intrinsic motivation to engage metacognitive control because metacognitive control was effortful . The ventral striatum ( VS ) activity during the redecision phase was positively correlated with the extent of decision uncertainty reduction in the Sudoku task , but not in the RDM task ( Fig 4C ) . VS might encode the intrinsic motivation or the internal reward on reduction in uncertainty during the redecision phase in the Sudoku task . Critically , the lFPC activity was significantly coupled with the interaction between the VS activity and the decision uncertainty level of the initial decision ( Fig 4D; see psycho–physiological interaction [PPI] analysis in Materials and methods ) . Furthermore , the accuracy change of each participant by redecision was positively correlated with the coupling strength in the Sudoku task ( Fig 4E ) . These results implied that the efficiency of lFPC involvement in metacognitive control in rule-based decision-making tasks ( i . e . , Sudoku ) might be facilitated by the VS activity . The abilities of metacognitive monitoring and control are behaviorally embodied in two components: uncertainty sensitivity and accuracy change , respectively . Throughout all sessions , including fMRI1 and the other repeated behavioral experiments , the individual uncertainty sensitivity was highly consistent across different sessions of the Sudoku task ( Cronbach’s α = 0 . 91; Fig 5A , left column , upper panel ) and the RDM task ( α = 0 . 89 , Fig 5A , left column , middle panel ) , as well as across the two tasks ( α = 0 . 85; Fig 5A , left column , lower panel ) . In contrast , the individual accuracy change in redecision was not consistent across the two tasks ( α = 0 . 03; Fig 5A , right column , lower panel ) , although it was consistent between different sessions of the Sudoku task ( α = 0 . 80; Fig 5A , right column , upper panel ) and the RDM task ( α = 0 . 76; Fig 5A , right column , middle panel ) . Thus , individual metacognitive abilities of uncertainty monitoring were reliably consistent , but individual metacognitive control was dissociable in the two tasks . Accordingly , the individual uncertainty sensitivity ( AROC ) was positively correlated with the uncertainty-level regression β value of the fMRI signal changes ( i . e . , neural uncertainty sensitivity ) , primarily in the dACC and AIC regions ( Fig 5B , P < 0 . 001 , cluster size = 20; and Fig 5C upper , one-tailed t test , Pearson’s r = 0 . 79 , t19 = 5 . 6 , P = 6 . 0 × 10−6 in the Sudoku task; r = 0 . 55 , t19 = 2 . 9 , P = 0 . 0049 in the RDM task; S3 Table ) , but not in the lFPC region ( Fig 5B and 5C bottom; Pearson’s r = 0 . 17 , t19 = 0 . 8 , P = 0 . 22 in the Sudoku task; r = 0 . 21 , t19 = 1 . 0 , P = 0 . 17 in the RDM task ) , commonly in both tasks . The differences of correlations were significant between the two regions ( t19 = 3 . 8 , P = 5 . 6 × 10−4 in the Sudoku task; t19 = 2 . 3 , P = 0 . 016 in the RDM task ) . In contrast , the individual accuracy change was significantly correlated with the mean activity in the lFPC region ( Fig 5D , P < 0 . 001 , cluster size = 20; and Fig 5E bottom; Pearson’s r = 0 . 69 , t19 = 4 . 2 , P = 2 . 2 × 10−4 in the Sudoku task; r = −0 . 39 , t19 = 1 . 9 , P = 0 . 041 in the RDM task ) , but not in the dACC region ( Fig 5D and 5E upper; Pearson’s r = 0 . 18 , t19 = 0 . 8 , P = 0 . 21 in the Sudoku task; r = −0 . 02 , t19 = 0 . 09 , P = 0 . 47 in the RDM task ) . When the lFPC activity was stronger , the accuracy change was more in the Sudoku task but became less in the RDM task ( Fig 5E ) . The differences of correlations were significant between the two regions ( t19 = 2 . 7 , P = 0 . 007 in the Sudoku task; t19 = 1 . 8 , P = 0 . 045 in the RDM task ) . Thus , the dACC activity ( AIC as well ) commonly represented individual metacognitive abilities in monitoring of decision uncertainty , whereas the lFPC differentially modulated individual metacognitive abilities in control of decision adjustment—in both the Sudoku and RDM tasks—consistent with their dissociated functional roles in metacognitive monitoring and metacognitive control , respectively . The regions of the metacognition network were also activated in the trials of both tasks with confidence level 4 in comparison with their respective control conditions ( Fig 6B and S2 Fig ) . These activity differences might be partially caused by differentially subjective uncertain states of the two conditions that were not reflected by the four-scale confidence ratings ( i . e . , the ceiling effect ) . The averaged accuracy was about 80% in the certain trials of the tasks ( Fig 2C ) , but it was about 95% in the control conditions . Nevertheless , the task baseline activity in the certain trials of the tasks could also predict the individual uncertainty monitoring bias and potential abilities of efficient metacognitive control of decision adjustment . Individual uncertainty monitoring bias—as estimated by averaging the decision uncertainty levels of all trials in each session of the tasks , representing the individual’s overconfident or underconfident tendency—was consistent between different sessions in the Sudoku task ( α = 0 . 95; Fig 6A , left panel ) and in the RDM task ( α = 0 . 94 , Fig 6A , middle panel ) , as well as across the two tasks ( α = 0 . 91; Fig 6A , right panel ) . Accordingly , individual uncertainty monitoring bias was positively correlated with the mean task baseline activity in the dACC region ( Fig 6C P < 0 . 001 , cluster size = 20; and Fig 6F left , Pearson’s r = 0 . 50 , t19 = 2 . 5 , P = 0 . 0096 in the Sudoku task; r = 0 . 44 , t19 = 2 . 1 , P = 0 . 022 in the RDM task ) but not in the lFPC region ( Fig 6C and 6F right; Pearson’s r = 0 . 18 , t19 = 0 . 80 , P = 0 . 22 in the Sudoku task; r = −0 . 04 , t19 = 0 . 17 , P = 0 . 43 in the RDM task ) , commonly in both tasks . The differences of correlations were significant between the two regions ( t19 = 2 . 1 , P = 0 . 026 in the Sudoku task; t19 = 1 . 8 , P = 0 . 042 in the RDM task ) . Meanwhile , the individual accuracy change in the Sudoku task was positively correlated with the mean task baseline activity in the lFPC region ( Fig 6D and 6G right; Pearson’s r = 0 . 45 , t19 = 2 . 2 , P = 0 . 020 ) but not with that in the dACC region ( Fig 6G left; one tailed t test , r = 0 . 14 , t19 = 0 . 62 , P = 0 . 27 ) . In contrast , the individual accuracy change in the RDM task was negatively correlated with the mean task baseline activity in the lFPC region ( Fig 6E and 6G right; Pearson’s r = −0 . 40 , t19 = 1 . 9 , P = 0 . 035 ) but not with that in the dACC region ( Fig 6G left; r = −0 . 13 , t19 = 0 . 57 , P = 0 . 29 ) . The differences of correlations were significant between the two regions ( t19 = 1 . 9 , P = 0 . 039 in the Sudoku task; t19 = 1 . 8 , P = 0 . 046 in the RDM task ) . Furthermore , the differences of correlations with the individual uncertainty monitoring bias and the individual accuracy change in the dACC ( t19 = 2 . 2 , P = 0 . 020 in the Sudoku task; t19 = 2 . 8 , P = 0 . 0055 in the RDM task ) , as well as in the lFPC ( t19 = 1 . 8 , P = 0 . 046 in the Sudoku task; t19 = 2 . 0 , P = 0 . 030 in the RDM task ) , were significant . Thus , the task baseline activity in the dACC region commonly reflected the individual uncertainty monitoring bias in both tasks , whereas that in the lFPC region could predict the individually differential potential abilities of metacognitive control for decision adjustment in both tasks . Thus far , we have shown that the neural system of metacognition could be dissociated into at least two subsystems: the dACC and AIC regions involved in metacognitive monitoring of decision uncertainty , and the lFPC region involved in metacognitive control of decision adjustment . To further elaborate the subsystems of the metacognition network , we performed analyses of interregional functional connectivity in the metacognition network . By regressing out the mean activity and the modulations by the decision uncertainty level , the RT and the extent of uncertainty reduction , as well as their interactions , we calculated trial-by-trial correlations between each pair of regions in the metacognition network ( see Materials and methods ) . The interregional functional connectivity patterns in both the task condition ( Fig 7A ) and the control condition ( Fig 7B ) were almost identical across the two types of tasks and were also similar to those at the resting state ( Fig 7C ) . The interregional functional connectivity patterns consistently showed that the metacognition network might be divided into three subsystems: the lFPC region; the dACC and AIC regions; and the DLPFC and aIPL regions . The interregional functional connectivity within each of the subsystems was systematically stronger than that across the subsystems ( paired t test , P < 0 . 05 in all comparisons ) . So far , the functional roles of the subsystem consisting of the DLPFC and aIPL regions in metacognition remain unclear . It is worth noting that the functional connectivity between the dACC and the regions of the other two subsystems in the task conditions was numerically stronger than the corresponding one at the resting state but was not statistically significant .
In the present study , we utilized a novel decision–redecision paradigm to examine the behavioral and neural associations of metacognitive processing in redecision , as compared to the processing in an initial decision . The robust findings from our study showed that individual uncertainty sensitivity ( both AROC and rRT-uncertainty ) remained markedly stable over two consecutive decisions on the same situational task , between different sessions of the same tasks , and across the different tasks . This indicates that individual uncertainty sensitivity was independent of evidence accumulation or the form of the decision-making process . These findings provide evidence to contradict the theoretical prediction of Theory 2 . If the processes of metacognitive processing and decision-making were integrated in one network , it should follow that , as more evidence is accumulated after redecision , the uncertainty sensitivity ( i . e . , AROC ) should be also improved [28] . Our study did not support that but rather led us to suggest the existence of an additional neural process in the brain to nonuniformly transform evidence accumulated in the decision-making processes to neural signals encoding decision confidence/uncertainty . The decision confidence/uncertainty should be much constrained by this neural process , as proposed by Theory 1 . Using fMRI , we identified patterns of neural activity in the frontoparietal control network that were more extensively involved in redecision than with initial decisions . Furthermore , activity in the regions of this network was positively scaled with decision uncertainty and became stronger in the condition requiring redecision than that in the condition not requiring redecision after the initial decision . These findings suggest that this network is involved both in metacognitive monitoring of decision uncertainty and metacognitive control of decision adjustment . Taken together , the evidence supports the theoretical proposal that metacognition utilizes a separate neural system to monitor and control decision-making ( i . e . , Theory 1 ) . We have putatively referred to the network revealed by our experiments as the metacognition network . We further propose that this network could be segregated into three subsystems ( as shown in Fig 7 ) . The subsystem consisting of the dACC and AIC regions was involved in metacognitive monitoring of decision uncertainty , common in the two tasks . The neural uncertainty sensitivity ( the uncertainty-level regression β value ) in the two regions was highly correlated with the behavioral uncertainty sensitivity . Furthermore , their task baseline activity could predict the individual uncertainty bias . Thus , the decision uncertainty signal could be finally represented by the dACC and AIC activity , which might be the outcome of transforming the uncertainty information from the decision-making process [29] . We thus inferred that uncertainty monitoring might indeed consist of two-order processes . We suggest the possibility that the first-order process coincides with the decision-making process that simultaneously generates the uncertainty information , implicitly associated with decision uncertainty , as proposed by Theory 2 , and that the second-order process then transforms this uncertainty information from different decision-making processes into common decision uncertainty scales , which are encoded in the dACC and AIC regions , as activity observed in this study would support . This hypothesis then integrates the two theories together and consistently accounts for observed evidence from both sides . It is worth noting that our results differed from previous neuroanatomical studies showing that the lFPC region was associated with individual behavioral uncertainty sensitivity [9 , 18] . The dACC and AIC regions have been well recognized for their involvement in conflict and error monitoring of the preceding cognitive processes to signal the need for more control [30–32] . Our results suggest that it is decision uncertainty , rather than decision error or conflict information , that serves as the primary signal to evoke monitoring [25] . Our findings are also profoundly different from previously reported accounts of dACC function in performance monitoring . First , there was no explicit feedback or cue to indicate whether the decision was correct or incorrect . The participants evaluated decision uncertainty via individual internal signals rather than external cues . Secondly , the task difficulty and RT does not explain the dACC and AIC activity in association with decision uncertainty . The dACC and AIC regions have been shown to broadly monitor subjective feelings such as pain , emotion , and others [33] . Critically , the salient information that elicits conscious monitoring in these regions is not necessarily from the somatosensory stimulation [34] . Similarly , the prospective monitoring of uncertainty in judgments of learning ( JOL ) and feeling-of-knowing ( FOK ) has also been shown to activate these regions , prior to execution of the decision-making tasks [35] . Therefore , decision uncertainty monitoring in the dACC and AIC regions should be domain general , independent of the sources of uncertainty information and the forms of decision-making tasks . In short , the individual uncertainty sensitivity is a unique and core trait of each individual decision maker , presumably dependent on the circuit of the dACC and AIC regions [33] . Decision uncertainty monitoring could be a bottom-up process . It occurred automatically without any explicit requirement of redecision ( fMRI3 , S1G Fig ) [25] . However , the subsequent metacognitive control of decision adjustment should require top-down cognitive control . In the Sudoku task , lFPC activity was positively correlated with the extent of decision uncertainty reduction within individual participants and the accuracy change by redecision across the participants , suggesting critical involvement of the lFPC region in metacognitive control . Uncertainty-driven exploration could be a critical process in metacognitive control [25 , 36–39] . Revising foregone decisions usually requires an exploration of alternative-solution approaches because the previously used solution approach would likely lead to the same unsatisfactory solution . Through exploration of alternative solutions , a more satisfactory option could be found by which the decision uncertainty would therefore be reduced . During this process of exploration , strategy management could be a key function of lFPC involvement in metacognitive control . This top-down strategic signal might regulate the activity in other frontal cortical areas and posterior parietal cortex , to execute the processes of altering the previous uncertain choice [25 , 36 , 39] or to explore a non-default option [37 , 38] . This would lead to the expectation that the lFPC would not be involved in metacognitive control in the RDM task because revising the preceding perceptual decision might simply require more attention to the stimulus in redecision to continue evidence accumulation , not necessarily exploration . However , lFPC activity remained activated as well and was negatively correlated with the extent of decision uncertainty reduction and accuracy change . It is possible that the process of exploration in the lFPC might be competitive with the simultaneous process of exploitation in the posterior brain areas when these two-level systems are not well coordinated [40 , 41] . Indeed , an FPC lesion in nonhuman primates enhanced the animals’ performance of a well-learned decision-making task [42] . However , it remains enigmatic why the lFPC was kept activated when it was not necessary and would not facilitate the engaging task . Presumably , the signal for increased control derived from the dACC region that is sensitive to decision uncertainty might nonselectively activate the lFPC because lFPC activity was also conditioned by decision uncertainty . The automaticity of eliciting lFPC involvement in metacognitive control may facilitate uncertainty resolution in the majority of difficult real-world situations , to relieve effort for engagement in metacognitive control , but failure of disengagement could impair the performance adjustment in simple tasks . Instead , intrinsic motivation might boost metacognitive control of decision adjustment in demanding tasks through VS activity . Metacognitive control is a form of cognitive control; however , not all forms of cognitive control are metacognitive . Although the Sudoku task and the RDM task appeared very different , to our surprise , the fMRI activation patterns associated with the decision-making process in the initial decisions were quite similar between the two tasks . Critically , the IFJ at the posterior PFC was commonly activated . IFJ is ubiquitously engaged in online task execution , in involving cognitive control [43 , 44] , and attention [45] . Thus , IFJ might play a critical role in object-level cognitive control , generally in decision-making tasks [25 , 46 , 47] . The segregation of the meta-level cognitive control in the anterior PFC and the object-level cognitive control in the posterior PFC is aligned with the hypothesis of the rostrocaudal functional division of the PFC in cognitive control [25 , 48 , 49] . However , the PFC functional division proposed in the current paper is subject to the strategy of task implementation [25] rather than to level of task complexity [48 , 49] . The initial decision-making merely recruits the posterior PFC to implement the default strategy of exploiting routine processes , whereas metacognition is evoked when the initial decisions are uncertain , recruiting the frontoparietal control network , including the anterior PFC , to control exploration of alternative processes [50] . Therefore , the metacognition process in redecision is not incorporating prior information acquired in the initial decision , but rather it is prone to altering the initial decision . There were some potential pitfalls for the fMRI data analyses in the current study . Because the metacognition process should automatically accompany the decision-making process with uncertainty , it excludes the possibility of inserting time jitters between the initial decision phase and the redecision phase , as conventionally used in fMRI paradigms . Thus , the two events of the decision-making process and the metacognition process in the general linear models ( GLMs ) could be collinear and result in inflation of standard errors of the estimated parameters for the regions involved in both processes . Fortunately , the activation of the regions of interest ( ROIs ) predominately involved in metacognition appeared in the redecision phase . Of note , the variance inflation factor ( VIF ) was approximately 2 . 4 , which suggests that the collinearity of the GLMs used in the current study was not severe . In summary , we have constructed and proposed the extent and generality of the functional architecture of the metacognition neural system , which is separate from the decision-making neural system ( Fig 8 ) . The metacognition neural system is composed of the metacognitive monitoring system and the metacognitive control system . The metacognitive monitoring system , consisting of the dACC and AIC regions , is domain general . It reads out the uncertainty information from the decision-making process and quantitatively encodes the decision uncertainty states . The metacognitive control system of the lFPC region implements high-level cognitive control ( e . g . , strategy ) , dominant in rule-based and abstract inference tasks ( e . g . , the Sudoku task ) , and may compete with low-level cognitive control ( e . g . , attention ) , dominant in perceptual tasks ( e . g . , the RDM task ) . The high-level cognitive control by the lFPC region could be modulated by intrinsic motivational signals from the VS region . These two subsystems separately monitor and control the decision-making system , in which the IFJ region is critically involved . Thus , the decision-making neural system and the metacognition neural system form a closed-loop system to control and adapt our behavior towards desired goals .
All participants were university students , who were recruited through a campus bulletin board system ( BBS ) . Informed consent was obtained from each individual participant in accordance with a protocol approved by Beijing Normal University Research Ethics Committee . Twenty-one participants ( 19–33 y old , 12 female ) took part in the main fMRI experiment ( fMRI1 ) and the resting fMRI experiment . Out of them , 16 participants ( 19–33 y old , 9 female ) took part in all sessions of the repeated behavioral experiments . In addition , 17 participants ( 19–25 y old , 10 female ) took part in the second fMRI experiment ( fMRI2 ) , 25 participants ( 19–27 y old , 14 female ) took part in the third fMRI experiment ( fMRI3 ) , and 20 participants ( 19–26 y old , 11 female ) took part in the fourth fMRI experiment ( fMRI4 ) . In an aperture with a radius of 3 degrees ( visual angle ) , about 300 white dots ( radius: 0 . 08 degrees; density: 2 . 0% ) on a black background that were moving in different directions at a speed of 8 . 0 degrees/s were displayed . The time span of the movement of each dot lasted for 3 frames . A portion of dots was moving in the same direction ( one of the four directions: left , right , up , and down ) , while the others were moving in different , random directions . The participant was required to discriminate the net motion direction . According to the proportion of coherently moving dots , discrimination difficulty was classified into 10 levels ( Fig 2B ) , for which the movement coherence varied from 1 . 6% to 51 . 2%; coherence of moving dots in the control condition was 100% . To fill a 4 × 4 grid matrix , each digital number from 1 to 4 should be filled in once and only once in each column , each row , and each corner with 4 grids . The task used in the present study required participants to fill in a target grid with a digital number from 1 to 4 in a partially completed Sudoku problem . Each problem had a unique solution . A Sudoku generator ( custom codes ) was used to create thousands of different Sudoku problems . Problem difficulty was classified into 10 levels according to the minimum number of logic operation steps required to arrive at the solution; this classification scheme largely matched with participants’ reported subjective difficulty level ( Fig 2B ) . In the control condition , the presented problem was made up of symbols ( “#” ) replacing the digital numbers except for the space in the target grid where the digital number was illustrated . Thus , the participant only needed to press the corresponding button . Participants were trained in the cognitive skills used to solve the 4 × 4 Sudoku problems under experimenters’ guidance for at least two h per d over a continuous span of 4 d . The participants practiced solving problems with no time constraints first in two to four runs of 40 problems at a set difficulty level . Once the average accuracy of a session crossed 90% , s/he then practiced solving problems at the same level within 2 s . Once the average accuracy of the run in a 2-s time frame reached 70% , the participant repeated these steps at the next level of difficulty . After the 4-d intensive training program , each participant had attained high-level proficiency in solving 4 × 4 Sudoku problems in 2 s , as the mean task difficulty finally approached the fifth level . The sequences of both the Sudoku and RDM tasks were identical . In fMRI1 , each trial started with a green-cross cue to indicate that the task stimulus would be presented 1 s later . The stimulus was presented for 2 s , then four options were presented , and the participant made a choice within 2 s . After a choice was made , four confidence level ratings from 1 ( lowest ) to 4 ( highest ) were presented , and the participant reported their confidence rating within 2 s . The same stimulus was immediately presented again for 4 s , and the participant again selected a choice and again reported their confidence rating . Each trial lasted for 15 s . The control trials were intermingled with task trials . The sequence of the control trials was identical to that of the task trials . In each task , there were 4 runs , and each run consisted of 30 task trials and 10 control trials . The task difficulty of each trial was adjusted by a staircase procedure through which one level was upgraded after two consecutive correct trials , was downgraded by one level after two consecutive erroneous trials , and was kept at the same level otherwise , so that the mean accuracy converged at about 50% . Prior to each experiment , two runs were carried out to allow each participant to practice and stabilize performance . The Sudoku problems used in the learning and practice sessions were different from those used in the fMRI and behavioral experiments . In addition , a 10-min resting fMRI experiment was conducted when the participant was in a resting state with eyes opened . The second fMRI experiment ( fMRI2 , Fig 4 and S1C Fig ) was carried out to examine whether the metacognition network would also be essentially involved in the decision-making process in the initial decision if a new Sudoku problem or new RDM stimulus was presented during the redecision phase , following a control stimulus presentation in the initial decision phase , as used in fMRI1 . In fMRI2 , a randomized selection of a control stimulus , a new Sudoku problem , or a new RDM stimulus was presented in the redecision phase . The appearance of a new stimulus in the redecision phase occurred in half of the trials . The Sudoku problems or RDM stimuli used in this experiment were selected from those at the middle level of task difficulty . This design was used to reduce the participant’s decision uncertainty in this experiment by controlling the difficulty of the task . In all other ways , the task sequence was the same as that used in fMRI1 . In total , there were 120 trials across two runs in each task . The third fMRI experiment ( fMRI3 , Fig 3D and S1E Fig ) was carried out to compare brain activity in a redecision-task condition ( like the task in fMRI1 ) to activity in a task that did not require redecision . In non-redecision trials , a control stimulus was presented in the redecision phase , therefore no redecision was required . The task sequence was the same as the design used in fMRI1 with the exception that the presentation time of the stimulus was 3 s during the redecision phase . In each task , the ‘redecision’ condition and the ‘non-redecision’ condition were randomized , and each consisted of 60 trials across 3 runs . The fourth fMRI experiment ( fMRI4 , S4 Fig ) was carried out to confirm that the engagement of metacognition was independent of the duration of redecision . The task sequence was exactly the same as was used in fMRI1 , with the exception that the redecision phase was 2 s instead of 4 s . Only the RDM task was used in this experiment . In total , there were 120 task trials and 40 control trials across 4 runs . In the fMRI experiments , the participants viewed images of the stimuli on a rear-projection screen through a mirror ( resolution , 1 , 024 × 768 pixels; refresh rate , 60 Hz ) . Normal or corrected-to-normal vision was achieved for each participant . All images were restricted to 3 degrees surrounding the fixation cross . All fMRI experiments were conducted using a 3 T Siemens Trio MRI system with a 12-channel head coil ( Siemens , Germany ) after the 4-d Sudoku training . Functional images were acquired with a single-shot gradient echo T2* echo-planar imaging ( EPI ) sequence with volume repetition time ( TR ) of 2 s , echo time ( TE ) of 30 ms , slice thickness of 3 . 0 mm , and in-plane resolution of 3 . 0 × 3 . 0 mm2 ( field of view [FOV]: 19 . 2 × 19 . 2 cm2; flip angle [FA]: 90 degrees ) . Thirty-eight axial slices were taken , with interleaved acquisition , parallel to the anterior commissure–posterior commissure ( AC–PC ) line . To test the reliability of the participants’ metacognitive abilities , behavioral experiments were carried out using the same paradigms as the Sudoku and RDM tasks . Each of the participants completed 6 sessions of behavioral experiments on consecutive days . Each session was composed of 4 runs of the Sudoku task and 4 runs of the RDM task , the same as was used in fMRI1 . A nonparametric approach was employed to assess each participant’s uncertainty sensitivity . The ROC curve was constructed by characterizing the incorrect probabilities with different uncertainty levels for initial decisions as thresholds . The area under curve ( AUC ) was calculated to represent how well the participant was at detecting and rating their decision uncertainty [9] . The individual uncertainty bias was estimated by the mean uncertainty level of each session , regressed out the factor of Aroc . The accuracy change was the change in mean accuracy from the first decision to the second decision . The individual uncertainty sensitivity and uncertainty bias , as well as accuracy change , were calculated for each session of the fMRI and behavioral experiments . The analysis was conducted with FMRIB’s Software Library ( FSL ) [51] . To correct for the rigid head motion , all EPI images were realigned to the first volume of the first scan . Data sets in which the translation motions were larger than 2 . 0 mm or the rotation motions were larger than 1 . 0 degree were discarded . It turned out that no data had to be discarded in the fMRI experiments . The EPI images were first aligned to individual high-resolution structural images and were then transformed to the Montreal Neurological Institute space by using affine registration with 6 degrees of freedom and resampling the data with a resolution of 2 × 2 × 2 mm3 . A spatial smoothing with a 4-mm Gaussian kernel ( full-width at half-maximum ) and a high-pass temporal filtering with a cutoff of 0 . 005 Hz were applied to all fMRI data . Each trial in fMRI1 was modeled with three regressors . The first regressor represented the decision-making process in the initial decision , which was time-locked to the onset of the first stimuli presentation , with summation of the presentation time ( 2 s ) and the differential RT from the mean RT of control trials as the event duration . The second regressor represented the neural process following the initial decision , including the metacognition process and the decision-making process in the redecision , and was time-locked to the onset of the first confidence judgment—with summation of the confidence report , the second presentation time ( 4 s ) of the stimuli , and the differential RT from the mean RT of control trials as the event duration . The third regressor represented the baseline during the intertrial intervals ( ITIs ) , time-locked to the onset of ITI , with the ITI duration as the event duration . The uncertainty level of the initial decision , the RT , and the level of uncertainty reduction ( differences in the uncertainty level between the final decision and the initial decision ) were implemented as modulators of the second regressor by demeaning the variances of the uncertainty level ( Fig 3C ) and consequently orthogonalizing the RT and the level of uncertainty reduction with each other ( Fig 3A–3C and S1I Fig ) , or reversing the orthogonalization order ( S1J Fig ) . It should be noted that the orthogonalization processes were equal to stepwise regression analyses on these covariates . The same analyses were applied to the fMRI data of fMRI2 and fMRI3 . For group-level analysis , we used FMRIB’s local analysis of mixed effects ( FLAME ) , which model both “fixed effects” of within-participant variance and ‘random effects’ of between-participant variance using Gaussian random-field theory . Statistical parametric maps were generated by a threshold with P < 0 . 05 with false discovery rate ( FDR ) correction , unless noted otherwise . The regressions of the individual uncertainty sensitivity ( AROC ) , the individual RT–uncertainty correlation coefficient , the individual mean uncertainty level , and the individual accuracy change with the β weights of uncertainty levels ( Figs 5B , 5D and 6C ) —or with the task baseline activity ( Fig 6C–6E ) —were calculated at the third level of group analyses . For these analyses , statistical parametric maps were generated by a threshold of P < 0 . 001 with the cluster-size threshold as 20 ( family-wise error correction ) . The ROIs of the metacognition network were defined by the voxels that were significantly activated during the redecision phase in the task trials compared to those during the same phase in the control trials across both tasks using conjunction analysis ( P < 0 . 001 , cluster-wise correction; green areas in statistical parametric maps ) . ROI analyses were obtained from both hemispheres of the same region . The VS ROI was anatomically defined by the striatum atlas of FSL templates [52] . The time courses were derived from the ROIs , calculating a mean time course within an ROI in each participant individually . We then averaged the time courses of the same condition across the participants ( S2 and S3 Figs ) ; or , we oversampled the time course by 10 and created epochs from the beginning of an event onward , then applied the corresponding GLM to every pseudo-sampled time point separately . By averaging the β weights across participants , we created the time courses shown in Fig 4 . SEMs were calculated between participants . The PPI analysis ( Fig 4D ) was conducted with the demeaned VS time courses after removing the mean activity and the component correlated with the uncertainty level as the physiological factor , and the uncertainty level convolved with the canonical hemodynamic response function ( HRF ) during the redecision phase as the psychological factor . The two factors per se , and the interaction between the two factors as confound regressors , were put together into a new GLM analysis across the whole brain . Functional connectivity analyses were independently conducted for the task and resting fMRI data . For the task fMRI data in each ROI , the mean activity and the components associated with the uncertainty level , RT , level of uncertainty reduction , and their interactions were regressed out; the residual time courses were then averaged across the voxels of the region and segmented into the individual trials of the task and control conditions in the Sudoku and RDM task , respectively . The segmented data of each trial were then modeled using a single regressor during the redecision phase convolved with the canonical HRF , and then a regression value was obtained for each trial . The correlation coefficient of the regression values between each pair of the ROIs in the metacognition network was calculated across the trials of the task or control condition in each participant . Finally , the averaged correlation coefficients were shown ( Fig 7A and 7B ) . For the resting fMRI data , the standard processing was carried out [53] , and the averaged correlation coefficients were shown ( Fig 7C ) .
|
Decision-making is often accompanied by a sense of uncertainty regarding the outcome . In many situations , there is no explicit feedback or cue to indicate whether the decision is correct or not . Fortunately , our brain can evaluate decision uncertainty using the internal signals and subsequently make appropriate adjustments to initial decisions . The process of considering the outcome of a decision and whether a decision should be adjusted is called metacognition , and it tends to be automatically induced . Thus , decision-making is usually accompanied by metacognition , and the two processes are inevitably coupled . However , the neural systems supporting metacognitive processing remain unclear and have often been misattributed to the neural system of the decision-making process per se . Here , we have analyzed this process in several volunteers by imaging the brain activity in specific regions while they performed Sudoku and random-dot motion ( RDM ) tasks . Our results suggest the existence of a neural system located in the prefrontal cortex ( PFC ) mainly involved in metacognition and independent from the neural system of decision-making .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"cingulate",
"cortex",
"control",
"theory",
"medicine",
"and",
"health",
"sciences",
"decision",
"theory",
"diagnostic",
"radiology",
"functional",
"magnetic",
"resonance",
"imaging",
"decision",
"making",
"engineering",
"and",
"technology",
"prefrontal",
"cortex",
"applied",
"mathematics",
"brain",
"social",
"sciences",
"neuroscience",
"magnetic",
"resonance",
"imaging",
"control",
"engineering",
"cognitive",
"psychology",
"mathematics",
"systems",
"science",
"statistics",
"(mathematics)",
"cognition",
"brain",
"mapping",
"neuroimaging",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"imaging",
"techniques",
"metacognition",
"psychology",
"radiology",
"and",
"imaging",
"diagnostic",
"medicine",
"anatomy",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"cognitive",
"science"
] |
2018
|
The neural system of metacognition accompanying decision-making in the prefrontal cortex
|
The unfolded protein response ( UPR ) regulates cell fate following exposure of cells to endoplasmic reticulum stresses . PERK , a UPR protein kinase , regulates protein synthesis and while linked with cell survival , exhibits activities associated with both tumor progression and tumor suppression . For example , while cells lacking PERK are sensitive to UPR-dependent cell death , acute activation of PERK triggers both apoptosis and cell cycle arrest , which would be expected to contribute tumor suppressive activity . We have evaluated these activities in the BRAF-dependent melanoma and provide evidence revealing a complex role for PERK in melanoma where a 50% reduction is permissive for BrafV600E-dependent transformation , while complete inhibition is tumor suppressive . Consistently , PERK mutants identified in human melanoma are hypomorphic with dominant inhibitory function . Strikingly , we demonstrate that small molecule PERK inhibitors exhibit single agent efficacy against BrafV600E-dependent tumors highlighting the clinical value of targeting PERK .
Folding and maturation of secreted proteins occurs in the endoplasmic reticulum ( ER ) . Cellular stresses that generate mis-folded proteins trigger a stress response termed the unfolded protein response pathway ( UPR ) [1–5] . Activation of the UPR is characterized by increased transcription of genes encoding ER molecular chaperones such as BiP/GRP78 and GRP94 , protein disulfide isomerase , and CHOP ( C/EBP homologous protein ) [6–10] . Mammalian cells contain three ER transmembrane effectors of the UPR . Ire1 is composed of a luminal domain that senses stress , a single transmembrane domain , and a cytosolic tail that contains both a protein kinase domain and an Rnase domain [11 , 12] . Ire1 regulates expression of numerous ER chaperones through activation of the X-box binding protein 1 ( Xbp1 ) transcription factor [13] . Accumulation of Xbp1 is mediated by Ire1-dependent splicing that generates a shorter Xbp1 mRNA that is more efficiently translated [14 , 15] . PERK , also an ER transmembrane protein kinase , is activated in a manner analogous to the Ire1 [16] and catalyzes serine 51 phosphorylation of eIF2α resulting in reduced protein synthesis [17–19] . The third signaling components are the transmembrane transcription factors ATF6α/β . While normally tethered to the ER , upon stress , ATF6 migrates to the trans-Golgi , where it is processed by S1P and S2P proteases to release the N-terminal DNA-binding transcription factor domain [20–22] . Physiologically , the UPR is an adaptive pathway . Through increased synthesis of chaperones , reduced protein synthesis and cell cycle arrest , cells have a window of opportunity to restore ER homeostasis prior to committing to apoptosis . Consistently , knockout of individual UPR signaling molecules , such as PERK or Ire1 , severely compromises cell survival following stress [23–26] . When a cell is unable to alleviate the burden of mis-folded proteins , such as under conditions of chronic stress , the UPR triggers apoptosis [27–31] . Among the various pathways engaged , Perk-dependent activation of the pro-apoptotic CHOP transcription factor is the most heavily investigated [28–34] . The balance of pro-survival and pro-apoptotic signals following stress ultimately determines cell fate . Although perturbations in protein folding in the ER can be achieved through the use of pharmacological agents that disrupt protein glycosylation ( tunicamycin ) or perturb calcium homeostasis ( thapsigargin ) [35–38] , the rapid expansion of tumor cells results in a microenvironment wherein critical metabolic nutrients such as glucose , oxygen and growth factors become limiting resulting in UPR activation . Acute expression of oncogenes is also associated with UPR engagement [39–42] . Normal cells respond to chronic UPR activation via growth arrest and/or apoptosis thereby preventing cell expansion , while tumor cells typically bypass the anti-proliferative impact of UPR activation and instead depend upon the pro-adaptive signaling suggesting a potential point of therapeutic intervention . Indeed deletion of PERK can reduce tumor progression [42 , 43] . Likewise , deletion of Xbp1 , a transcription factor whose accumulation is dependent upon Ire1 activity , also reduces tumorigenesis [44] . Such results have stimulated attempts to generate small molecules that inhibit PERK or Ire1 . Consequently , highly specific and potent inhibitors of the PERK enzyme have been developed [45–48] . While the UPR is considered important for tumor progression , there is potential for tumor suppressive activity given it antagonizes cyclin D1 . With the advent of PERK specific inhibitors and an eye towards therapeutic utility , we have addressed the role of PERK in BrafV600E driven melanoma and provide evidence for a dose-dependent function of PERK in melanoma genesis .
Perk harbors anti-proliferative activity [49] in addition to cell survival activities , suggesting a potential for tumor suppressive properties . We ascertained the impact of deletion of one versus two alleles of Perk in melanocytes harboring activated BrafV600E . We utilized a conditional allele of Perk to circumvent issues of pancreatic atrophy that occurs in a global Perk knockout [50–52] . Previous work with the mice wherein BrafV600E expression alone is induced in melanocytes revealed induction of cellular senescence rather than tumor development [53] . Bypass of BrafV600E-dependent senescence has only been observed in mice wherein a second tumor suppressor such as p16Ink4A [54–57] , PTEN [53] , or Fbxo4 has been deleted [58] . Remarkably , BrafV600ECA/+;Perk+/- mice developed melanoma with high penetrance within 4–6 weeks which rapidly disseminated to peripheral tissue ( Fig 1A–1C ) . Immunohistochemistry ( IHC ) for S100 confirmed melanocytic origin melanocytes ( Fig 2 ) . To address underlying mechanisms , we analyzed “pre-malignant skin” from TyrCre; BrafCA/+;Perk +/+ or Perk+/- or Perk -/- mice . IHC revealed reduced accumulation of , p-Akt and pS6 in Perk+/- skin relative to wt , but higher than Perk-/- consistent with dosage dependence ( Fig 2 ) . Likewise , consistent with the presence of one wild type Perk allele , modestly elevated p-eIF2α and CHOP was observed relative to Perk-/- tissue . BrafV600ECA/+;Perk+/- tissue exhibited the highest level of staining CD31 staining , with BrafV600ECA/+;Perk+/+ having intermediate levels and BrafV600ECA/+;Perk-/- exhibiting the lowest level consistent with previous work [10 , 59] . These results demonstrate that deletion of one allele of Perk reduces p-eIF2α , pro-apoptotic CHOP yet maintains or even increases vascularity , as determined by CD31 staining . The observation that PERK+/- melanocytes are permissive for BrafV600E-dependent transformation implies that acute activation of BrafV600E triggers PERK activity and PERK tumor suppression . To assess this hypothesis , primary human melanocytes were infected with retrovirus encoding BrafV600E . Expression of mutant Braf triggered increased p-eIF2α and elevated CHOP ( Fig 3A and 3B ) . Conversely chaperone expression was not increased suggesting the absence or weak activation of ATF6 ( Fig 3A ) . BrafV600E expression was associated with increased SAβ-galactosidase consistent oncogene induced senescence ( Fig 3B ) . Armed with evidence for BrafV600E-dependent activation of Perk in vitro , we assessed BrafV600E-dependent activation of Perk in vivo . Following activation of BrafV600E expression specifically in melanocytes with topical application of 4-OHT , we noted increased expression of Chop; however , no increase in chaperone expression was observed and Xbp1 splicing was reduced suggesting that BrafV600E selectively induces Perk in vivo ( Fig 3C ) . To assess oncogene induced senescence , we measure SA-βGal in the skin of mice harboring BrafV600E in Perk+/+ , +/- and -/- backgrounds . Here we noted reduced SA-βGal staining specifically in Perk+/- relative to +/+ tissue demonstrating that deletion of one Perk allele permitted bypass of BrafV600E-induced senescence ( Fig 3D; quantification , S1 Fig ) . We also noted significant overexpression of cyclin D1 in BrafV600ECA/+/Perk+/- relative to BrafV600ECA/+/Perk+/+ ( Fig 3E and 3F ) . Consistent with Perk functioning as an antagonist of cyclin D1 protein synthesis [60] . The susceptibility of Perk+/- but not Perk-/- mice to BrafV600E-dependent melanoma genesis , along with the retention of 50% Perk protein expression in all tumors examined ( Fig 1 ) suggested the intriguing possibility that the remaining Perk allele was necessary for malignant progression . We addressed whether the remaining Perk allele was necessary for tumor progression by treating tumor-bearing mice with LY-4 . LY-4 is a PERK specific inhibitor with a 2nM IC50 and little activity towards other eIF2α kinases ( S1 Table ) . Kinome and Treespot analysis demonstrates the selectivity of LY-4 for PERK relative to > 400 additional kinases ( S2A and S2B Fig; S2 Table ) . After confirming that LY-4 inhibits PERK activity in cultured melanoma cells ( Fig 4A ) , mice were exposed to tamoxifen to induce BRAFV600E and delete a single Perk allele; LY-4 treatment was initiated when tumors were 2-3mm3 . LY-4 treatment reduced tumor growth by nearly 90% ( Fig 4B and 4C ) . Treatment reduced p-eIF2α ( Fig 4D ) , p-Akt , [61]; LY-4 treatment also elevated p62 and reduced LC3BII ( Fig 4D–4F ) consistent with reduced autophagy . LY-4 also reduced phospho-H3 , Ki67 and CD31 , while increasing TUNEL positivity demonstrating reduced proliferation and increased apoptosis ( S3A Fig ) as mechanisms contributing to LY-4-dependent tumor inhibition . No pancreatic toxicity was noted in LY-4 treated animals ( S3B and S3D Fig ) . Finally , LY-4 did not inhibit MAPK activation ( Fig 5F; S3C Fig ) demonstrating its impact on tumor growth does not reflect inhibition of downstream BrafV600E targets . The observation that Perk+/- was permissive while Perk-/- was resistant to BrafV600E melanoma , suggested a model where tumors remained dependent upon Perk for its pro-survival activity , but that reduced Perk dosage permitted senescence bypass , through either lack of apoptosis ( e . g . reduced CHOP induction ) or cyclin D1 induction reflecting reduced inhibition of translation under Perk deficiency . To test this , we 1 ) reduced expression of pro-apoptotic Chop or 2 ) reduced cyclin D1 levels . If reduced Chop was a key progression , Chop+/- or -/- mice should be susceptible to BrafV600E-melanoma . However , deletion of Chop in BrafV600ECA/+ mice did not permit melanoma development ( Fig 4G ) . To evaluate the role of cyclin D1 overexpression , in BrafV600E/Perk+/- tissue , we deleted one allele of CCND1 and deletion completely abrogated melanoma genesis ( Fig 4H ) . Overexpression of cyclin D1 drives development of lymphomas by triggering DNA damage , which in turn activates p53 [62 , 63]; as such , tumor progression selects either apoptosis or p53 loss [62 , 64–66] . Consistent with cyclin D1 overexpression contributing to melanoma in the Perk+/- background , p53 was overexpressed suggesting stabilizing mutations ( Fig 4F ) . DNA sequencing revealed p53 mutations in 7 of 7 tumors with mutations apparent throughout the DNA-binding domain ( S3 Table ) . The results presented above reveal that Perk+/- melanocytes are permissive for BrafV600E-melanomagenesis while , Perk-/- are not . The capacity of LY-4 to inhibit progression of BrafV600E;Perk+/- melanomas , implies an “addiction” to the remaining Perk allele suggesting a potential therapeutic threshold for Perk inhibition . To address Perk function in a mouse model of metastatic melanoma [58] , we generated Tyr-Cre/BrafV600ECA/+/Fbxo4mt/Perkf/f or Tyr-Cre/BrafV600ECA/+/Fbxo4mt/Perkf/+ permitting inducible activation of BrafV600E and deletion of one or two alleles of Perk in melanocytes upon application of 4-OHT . We have previously demonstrated that inactivation of Fbxo4 in the BrafV600ECA/+ background triggers cyclin D1-dependent , metastatic melanoma [58] . Deletion of both Perk alleles effectively inhibited BrafV600E-dependent melanoma in the setting of Fbxo4-deficiency , while deletion of one allele of Perk ( Perk+/- ) was not sufficient to either inhibit or accelerate melanoma genesis ( Fig 5A; S4A Fig ) . The absence of tumor inhibition is consistent with data in Fig 1 demonstrating the permissiveness of Perk+/- melanocytes to BrafV600E . The fact that we do not observe decreased latency in the Fbxo4mt/Perk+/- background is consistent with both Fbxo4 and Perk signaling converging on the inhibition of cyclin D1 . To address mechanism , we assessed Perk activity and downstream readouts in premalignant skin . Perk deletion reduced eIF2α phosphorylation and Chop expression in premalignant skin as detected by western analysis ( Fig 5B ) . The variability in signal likely reflects the fact that Perk is only deleted in melanocytes and we are analyzing whole skin . We next assessed various markers by IHC . We first stained sections with S100 to identify melanocytes and subsequent sections with the antibodies indicated . Deletion of Perk was associated with increased cyclin D1 specifically in pre-malignant melanocytes ( Fig 5C ) consistent with Perk-antagonizing cyclin D1 translation [49 , 67 , 68] . IHC also revealed decreased p-Akt and CD31 consistent with Perk-dependent regulation of both Akt signaling [61 , 69–71] and angiogenesis [72–74] ( Fig 5C ) . We next utilized BrafV600E;Pten-/- mice , an independent melanoma model , to determine whether Perk was required for tumor progression [75 , 76] . Mice were exposed to 4-OHT to induce BRAFV600E and delete Pten; LY-4 treatment was initiated when tumors were 2-3mm3 . LY-4 inhibited melanoma progression ( Fig 5D , S4E Fig ) and this outcome was accompanied by reduced eIF2α phosphorylation and CHOP accumulation suggesting on-target effects of this drug ( Fig 5E ) . LY-4 treatment also led to accumulation of p62 suggesting reduced autophagy , and elevated cleaved caspase 3 indicative of increased rate of apoptosis ( Fig 5E ) . IHC confirmed that LY-4 treatment resulted in decreased p-eIF2α , Chop , and p-H3 ( S4B Fig ) . Reduced CD31 staining was also noted ( S4B Fig ) , consistent with previous work linking Perk signaling with tumor angiogenesis [72–74] . IHC also revealed increased TUNEL positive tumor cells ( S4B Fig ) . We also monitored blood glucose levels in LY-4 treated mice , given previous evidence that Perk inhibition caused pancreatic toxicity ( 45 ) . Importantly , blood glucose levels remained stable ( blood glucose level was lower than 200 mg/dL ) , with no evidence of pancreatic damage during the course of treatment ( S4C and S4D Fig ) . If melanoma progression-depends upon the retention of functional PERK , it stands to reason that human melanoma-derived cell lines will maintain and depend upon PERK activity . To address the contribution of PERK to melanoma , we determined whether PERK was functional in melanoma cell lines . We utilized melanoma cells lines lacking detectable mutations in PERK , but expressing the following Braf alleles: BrafWT or BrafV600E/D ( 1205 Lu , 451LU , WM983B , WM35 , WM3918 , WM239A , WM3211 , WM1791C , C8161 ( http://www . wistar . org/lab/meenhard-herlyn-dvm-dsc/page/melanoma-cell-lines-0 ) . PERK expression was detected and was functional in all cell lines assessed ( Fig 6A ) . To assess PERK contributions to melanoma cell survival following ER stress , we established two independent melanoma cell lines ( WM3918 , WM239A ) expressing a previously validated , tetracycline-inducible shRNA directed against human PERK [43 , 77] . PERK expression was undetectable 3-days post-doxycycline in shRNA-harboring cells ( Fig 6B ) . To assess whether PERK function is important for melanoma cell growth and survival , we took advantage of previous work revealing a role for PERK in promoting survival following cell detachment from solid matrix [78 , 79] . Consistently , PERK knockdown in either WM3918 or WM239A reduced colony formation in soft agar ( Fig 6C; left graphs ) . Addition of thapsigargin further reduced anchorage-independent growth ( right graphs ) . PERK function also increased cell survival in clonogenic survival assays ( Fig 6D ) . In contrast , PERK knockdown cells grew well on plastic under normal growth conditions ( Fig 6D top; Fig 6E ) . As an independent method for assessing PERK contribution to melanoma cell survival thereby ensuring that the impact of shRNA was PERK-dependent , we utilized a previously characterized small molecule inhibitor of PERK , GSK2656157 [45–47] or LY-4 . GSK2656157 or LY-4 treatment inhibited PERK activity ( as judged by reduced auto-phosphorylation ) and suppressed melanoma cell survival under ER-stress ( S5A and S5B Fig ) . These data demonstrate that retention of functional PERK is critical for melanoma cell survival . The susceptibility of the PERK+/- genotype to melanoma genesis suggests a potential for inactivation of PERK in human melanoma . We searched the human cancer genome atlas and identified mutations throughout PERK coding exons at a frequency of ~7% ( Fig 7A ) . To assess the functional consequence of these mutations to PERK function , we generated analogous alleles in murine Perk ( A418V , T424A , H432Y , Y470C , P479Q , P991R , Δ910-analogous to human 911fs which deletes AA910-1116 ) and reconstituted Perk-/- MEFs by retroviral transduction . Importantly , all mutants remained localized to ER structures analogous to wild type Perk ( S6B Fig ) . Nevertheless , all mutants exhibited reduced activities with PerkΔ910 exhibiting the least activity as determined by p-eIF2α , Chop induction and cyclin D1 repression ( Fig 7B and 7D ) . Consistent with all mutations compromising Perk function , Perk-/- MEFs reconstituted with melanoma-derived Perk mutants exhibited increased sensitivity to ER stress as determined by clonogenic survival assay ( Fig 7E ) . Importantly , co-administration of LY-4 with tunicamycin further reduced cell survival demonstrating that mutant Perk allele activity still contributed to survival following ER stress and that all alleles remained LY-4 sensitive ( Fig 7E ) . Given that reduced Perk function cooperated with BRAF in vivo , we expected that hypomorphic Perk might increase spontaneous cell transformation . Consistently , cells expressing mutant Perk formed foci and grew in soft agar , both surrogates of cell transformation ( Fig 7F; S6A Fig ) . Finally , because Perk mutations will occur in the context of one wild type Perk allele , we considered the potential of these tumor-derived Perk mutants to exhibit dominant negative activity relative to endogenous Perk . To address this , we transduced NIH3T3 cells , which retain wild type Perk , with retrovirus encoding selected mutant Perk alleles ( Fig 7G ) . Following transduction , cells were exposed to thapsigargin and we subsequently measured Chop expression as a read out of Perk function . Consistent with dominant negative activity , we noted reduced Chop expression in cells expressing most Perk mutants . This data reveals that hypomorphic alleles of Perk exhibit dominant inhibitor activity with respect to endogenous Perk and suggests physiological relevance of Perk mutants , in melanoma development .
While PERK has pro-survival and thus pro-tumorigenic activities , it also triggers pro-apoptotic signals and opposes cell division via inhibition of cyclin D1 translation . This latter PERK attribute begs the question of whether under certain conditions , or in specific tissues , PERK might function as a tumor suppressor . Supporting potential tumor suppressive function for the UPR , HRAS-dependent transformation of primary human melanocytes was potentiated by genetic inhibition of PERK , Ire1 and ATF4 [40] . To directly assess the contribution of Perk to melanoma genesis in vivo , Perk was deleted in a mouse melanocytes coordinately with activation of BrafV600E [58] . Deletion of single Perk allele resulted in melanocytes permissive to overt transformation through expression of BrafV600E alone . Typically , BrafV600E-dependent melanoma can be achieved by deletion or inactivation of known tumor suppressors , such as PTEN [53 , 80] , p16Ink4A [81–83] , Fbxo4 [58] . In the absence of inactivating mutations in one of these genes , BrafV600E expression is associated with permanent growth arrest or senescence . Mono-allelic deletion of Perk thus represents a previously unappreciated mechanism for bypass of BrafV600E-dependent senescence . In addition , it is the only example wherein deletion of one versus two alleles results in diametrically opposing results with regard to tumor suppression versus tumor progression . The basis of Perk haploinsufficiency likely reflects dose-dependent signaling duration and/or intensity . Accordingly , loss of one allele reduces pro-apoptotic signals ( CHOP expression reduced ) , increases expression of a pro-oncogenic protein ( cyclin D1 ) and maintains sufficient pro-survival signals through the remaining allele . Additionally , we noted that excision of one Perk allele suppressed BrafV600E-induced senescence . This likely reflects dysregulation of cyclin D1 , given previous work that associates cyclin D1/CDK4 function with senescence [84 , 85] . Importantly , BrafV600E/Perk+/- melanomas are dependent upon the remaining Perk allele . There have been sporadic reports implicating potential tumor suppressor like functions for Perk . Acute ablation of Perk in mammary epithelium increased tumor formation due to the accumulation of genomic instability [43] . Anti-proliferative activity of Perk was also attributed to differential impacts on mammary tumorigenesis [79] . In contrast to mono-allelic deletion , excision of both Perk alleles did not cooperate with BrafV600E demonstrating that retention of one functional allele of Perk is required for tumor progression . The demonstration that BrafV600E/Perk+/- melanomas are dependent upon the remaining Perk allele supports this conclusion . These results reveal an unanticipated role for Perk in melanoma initiation , given previous work arguing that Perk is not required for tumor initiation [42 , 43] . Bi-allelic Perk deletion did not impact oncogene-induced senescence or Erk1/2 phosphorylation demonstrating Perk is not required for BrafV600E signaling . This data reveals that Perk is required for the establishment of BrafV600E melanoma , but does not address whether PERK is a therapeutic target . Strikingly , treatment of mice with established BrafV600E/PTEN-dependent tumors triggered significant inhibition of tumor growth providing support for Perk as a bona fide therapeutic target . LY-4 treatment was associated with increased apoptosis , reduced markers of angiogenesis and decreased proliferation . Importantly , LY-4 treatment inhibited eIF2α phosphorylation and CHOP induction demonstrating on target effects of this agent . While a reduction in CHOP expression may also facilitate tumor progression by limiting cell death , it is not the primary driver susceptibility , as a Chop deletion in the BrafV600E background does not permit melanoma genesis . In contrast , cyclin D1 overexpression is a key tumorigenesis-driving event . Dysregulation of the cyclin D1/CDK4 pathway occurs in a majority of melanomas and increased cyclin D1/CDK4 activity ( e . g . loss of p16Ink4A or Fbxo4 ) cooperates with BrafV600E to drive melanoma . Second , reducing cyclin D1 expression through by deletion of a single allele inhibits melanoma genesis . Third , we have previously shown that cyclin D1-driven tumors are specifically opposed by p53 [63] and p53 is subject to mutations within its DNA binding domain in Perk+/- tumors while p16Ink4a is still expressed . If Perk heterozygosity were relevant to BrafV600E-dependent human melanoma , one would predict the occurrence of inactivating mutations in PERK in human melanoma . Consistently PERK mutations occur at a frequency of approximately 7% . Initial characterization of these mutants revealed reduced PERK activity . Cells expressing these mutants fail to effectively repress cyclin D1 , exhibit reduced CHOP expression and are prone to spontaneous transformation , analogous to cyclin D1 overexpressing cell lines . Collectively , our data reveals a complex role for Perk in melanoma genesis . While Perk pro-survival activity is necessary for melanoma genesis and melanoma progression , its ability to antagonize cell division through cyclin D1 supports a model wherein Perk regulation of cell fate is a delicate balance wherein less is not necessarily better . Importantly , LY-4 exhibited clear therapeutic potential for BrafV600E-dependent melanoma . Since Perk mutant tumors are dependent upon the remaining Perk allele , Perk status is unlikely to feature into patient response . An open question remains as to whether Perk tumor suppressive function is tissue specific . While previous work revealed no evidence for such in mammary tissue ( 43 ) , additional analysis is required to address this issue .
All animal use and experiments were approved by The Medical University of South Carolina Office of Compliance and Institutional Animal Care and Use Committee ( IACUC ) ( approved animal use protocol #AR3340 ) . All animals and experiments were carried out in compliance with Institutional Animal Care and Use Committee guideline of Medical University of South Carolina . Animals were obtained from the followings: TyrCreER , BrafCA/+ , Fbxo4+/- [58]; Perkl/l [51]CyclinD1+/- ( from Jackson Laboratory ) ; Chop-/- ( from Jackson Laboratory , stock #: 005530 ) . Appropriate intercrosses were established to generate TyrCre;BrafV600ECA/+;Fbxo4+/- or -/-;Perk+/+ , TyrCre;BrafCA/+;Fbxo4+/-or-/-; Perkl/+ , TyrCre;BrafCA/+;Perkl/+ , TyrCre;BrafV600ECA/+;Perkl/+;D1+/- , TyrCre;BrafV600ECA/+;Chop+/-or-/- mice and controls . For LY-4 treatment studies , tumor volume was measured twice per week and calculated with the following formula: volume = ( length × width × width ) /2 . LY-4 treatment was initiated when a tumor reached ~3mm3 ( formulated in 20% Captisol in 25mM of NaH2PO4 buffer , pH 2 . 0 ) by oral gavage twice per day . Blood glucose was monitored every five days using a Freestyle meter ( TheraSense , Inc . ) during LY-4 treatment . Animal experiments were conducted in accordance with IACUC protocols and University Laboratory Animal Research ( ULAR ) guidelines . 4-Hydroxytamoxifen ( 4-OHT ) was freshly prepared in dimethyl sulfoxide ( DMSO ) ( 5mM ) and applied topically for three consecutive days to postnatal day 2 pups . The mouse skin tissue and human primary melanocytes were homogenized in Trizol ( Invitrogen ) and total RNA was extracted with chloroform . The cDNA was synthesized by using MMLV reverse transcriptase III and random primers ( Invitrogen ) following the manufacturer’s protocol . QPCR assay was prepared by using SyBr PCR mix ( Applied Biosystems ) and amplified using ABI Prism 7000 Sequence Detection System ( Applied Biosystems ) with the following primers: Chop ( F: 5’-CCAACAGAGGTCACACGCAC-3’; R: 5’-TGACTGGAATCTGGAGAGCGA-3’ ) , Uxbp1 ( F: 5’-CACCTTCTTGCCTGCTGGAC-3’; R: 5’-GGGAGCCCTCATATCCACAGT-3’ ) , Sxbp1 ( F: 5’-GAGTCCGCAGCAGGTG-3’; R: 5’-GTGTCAGAGTCCATGGGA-3’ ) , Bip ( F: 5’-ACCCTTACTCGGGCCAAATT-3’; R: 5’-AGAGCGGAACAGGTCCATGT-3’ ) , Gapdh ( F: 5’-GGAGCGAGACCCCACTAACA-3’; R: 5’-ACATACTCAGCACCGGCCTC-3’ ) . Infected skin melanocytes were lysed directly with TRIzol reagent ( Sigma ) followed by the isolation of total RNA according the user’s instructions . One μg total RNA was reverse transcribed using a Maxima First Strand cDNA Synthesis Kit for qRT-PCR ( Thermo Fisher ) . Fast SYBR . Green Master Mix ( life Technologies ) was used with cDNA template and primers to evaluate the expression of target genes and GAPDH . Primers used were purchased from Integrated DNA Technologies . Amplifications were performed using an Applied Biosystems . 7500 Real-Time PCR System ( Life Technologies ) . All experiments were performed in triplicate . Expression ratios of controls were normalized to 1 . The melanoma tumor tissues and cultured cells were harvested in Tween 20 buffer containing 50mM HEPES ( pH 8 . 0 ) , 150mM NaCl , 2 . 5mM EGTA , 1mM EDTA , 0 . 1% Tween 20 , and protease/phosphatase inhibitors ( 1mM phenylmethylsulphonyl fluoride , 20 U of aprotinin/ml , 5mg of leupeptin/ml , 1mM DTT , 0 . 4mM NaF , and 10mM β-glycerophosphate ) . Lysates were sonicated prior to clearing by centrifugation at 4°C for 30 min . Proteins were resolved by SDS-PAGE , transferred to membrane , and subjected to immunoblot . Antibodies utilized include PERK ( Rockland ) , p-eIF2α S51 ( Cell Signaling ) , BiP ( Cell Signaling ) , total eIF2α ( Cell Signaling ) , Cyclin D1 ( mouse monoclonal D1-72-13G ) , Cul4a ( Bethyl , A300-739A ) , p-AktS473 ( Cell Signaling ) , total Akt ( Cell Signaling ) , GAPDH ( Cell Signaling ) and β-actin ( Sigma Aldrich ) . Tumor tissue samples were harvested from mice and snap frozen in tissue-Tek O . C . T . 6-micron thick frozen tissue sections were prepared according to standard procedures ( Sigma , Senescence-Galactosidase Staining Kit #9860 ) . 10% buffered formalin was used to fix tissues ( overnight ) , followed by dehydration with ethanol , paraffin embedding , and sectioning . 5- to 8-μm sections were used for immunohistochemistry ( IHC ) , sections were dewaxed and rehydrated in gradient ethanol followed by melanin depigmentation . Sections were immersed in 10% hydrogen peroxide and boiled for 20 min at 65°C . After microscopic inspection , the sections were rinsed with tap water for 5 min . Standard protocols were utilized for hematoxylin and eosin ( H&E ) staining . Antibodies utilized for IHC include: cyclin D1 ( mouse monoclonal D1-72-13G ) , S-100 ( Dako ) , CHOP ( Cell Signaling ) , p-eIF2α ( Cell Signaling ) , pAkt ( Cell Signaling ) , pS6 ( Cell Signaling ) , γH2Ax ( Cell Signaling ) , pATM ( Cell Signaling ) , p21 ( Santa Cruz Biotechnology ) , H4R3me2 ( Epigenetic ) , pRb ( Santa Cruz Biotechnology ) , CD31 ( Cell Signaling ) , pERK ( Cell Signaling ) , pH3 ( Cell Signaling ) , Carbonic anhydrase IX/CA9 ( Novus Biologicals ) . Antigens were retrieved with Antigen Retrieval Citra Plus ( Biogenic ) by boiling for 15 min , and antibodies were visualized with a Vectastain ABC Elite kit ( Vector Laboratories ) and a peroxidase substrate kit DAB ( Vector Laboratories ) . Sections were also tested for apoptosis by using TdT In Situ Apoptosis Detection Kit—DAB ( R&D Systems ) . For quantification of IHC , the average percentage of positively stained cell evaluated form each section were scored according to staining index ( SI ) scale ( - = no stain; + = < 33%; ++ = 33%-66%; +++ = >66% ) . Representative fields from each sections were chosen are presented in figures . Soft agar assays , were performed in 6-well plates ( 2500 cells seeded ) containing 0 . 4% low melting point agarose ( Lonza ) lower layer , and on top of 0 . 8% agarose-top layer . Cells were growth in 37°C , 5%CO2 for 21–26 days and colonies were quantified . GraphPad Prism software was utilized to analyze Kaplan-Meier tumor-free survival graphs . A two-tailed Student t test was utilized for other statistical analyses ( P values of <0 . 05 indicating statistical significance ) . Error bars in the figures represent .
|
PERK is critical for progression of specific cancers and has provided stimulus for the generation of small molecule PERK inhibitors . Paradoxically , the anti-proliferative and pro-death functions of PERK have potential tumor suppressive qualities . We demonstrate that PERK can function as either a tumor suppressor or a pro-adaptive tumor promoter and the nature of its function is determined by gene dose . Preclinical studies suggest a therapeutic threshold exists for PERK inhibitors .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"cell",
"death",
"medicine",
"and",
"health",
"sciences",
"cell",
"cycle",
"and",
"cell",
"division",
"cancer",
"treatment",
"cell",
"processes",
"biological",
"cultures",
"cancers",
"and",
"neoplasms",
"epithelial",
"cells",
"oncology",
"cell",
"cultures",
"chromatophores",
"melanoma",
"cells",
"immunologic",
"techniques",
"research",
"and",
"analysis",
"methods",
"animal",
"cells",
"melanocytes",
"melanomas",
"biological",
"tissue",
"signal",
"transduction",
"immunohistochemistry",
"techniques",
"cell",
"biology",
"anatomy",
"apoptosis",
"epithelium",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"cultured",
"tumor",
"cells",
"cell",
"signaling",
"histochemistry",
"and",
"cytochemistry",
"techniques",
"cyclins"
] |
2016
|
PERK Is a Haploinsufficient Tumor Suppressor: Gene Dose Determines Tumor-Suppressive Versus Tumor Promoting Properties of PERK in Melanoma
|
Genetic variants in intron 1 of the fat mass– and obesity-associated ( FTO ) gene have been consistently associated with body mass index ( BMI ) in Europeans . However , follow-up studies in African Americans ( AA ) have shown no support for some of the most consistently BMI–associated FTO index single nucleotide polymorphisms ( SNPs ) . This is most likely explained by different race-specific linkage disequilibrium ( LD ) patterns and lower correlation overall in AA , which provides the opportunity to fine-map this region and narrow in on the functional variant . To comprehensively explore the 16q12 . 2/FTO locus and to search for second independent signals in the broader region , we fine-mapped a 646–kb region , encompassing the large FTO gene and the flanking gene RPGRIP1L by investigating a total of 3 , 756 variants ( 1 , 529 genotyped and 2 , 227 imputed variants ) in 20 , 488 AAs across five studies . We observed associations between BMI and variants in the known FTO intron 1 locus: the SNP with the most significant p-value , rs56137030 ( 8 . 3×10−6 ) had not been highlighted in previous studies . While rs56137030was correlated at r2>0 . 5 with 103 SNPs in Europeans ( including the GWAS index SNPs ) , this number was reduced to 28 SNPs in AA . Among rs56137030 and the 28 correlated SNPs , six were located within candidate intronic regulatory elements , including rs1421085 , for which we predicted allele-specific binding affinity for the transcription factor CUX1 , which has recently been implicated in the regulation of FTO . We did not find strong evidence for a second independent signal in the broader region . In summary , this large fine-mapping study in AA has substantially reduced the number of common alleles that are likely to be functional candidates of the known FTO locus . Importantly our study demonstrated that comprehensive fine-mapping in AA provides a powerful approach to narrow in on the functional candidate ( s ) underlying the initial GWAS findings in European populations .
The association between variants in the fat mass and obesity associated ( FTO ) gene on chromosome 16q12 . 2 and body mass index ( BMI ) is well-established in populations of European descent . Genome-wide association studies ( GWAS ) and subsequent replication studies have identified several strongly correlated single nucleotide polymorphisms ( SNPs ) located in intron 1 of FTO associated with increased BMI and increased risk of obesity [1]–[13] . With an observed effect size of 0 . 35 kg/m2 ( 0 . 1 z-score units of BMI ) per risk allele , the FTO locus has a substantially stronger effect on BMI than any other identified common locus [14] . While this impact on BMI may seem small , it has a potential public health bearing , as even a 1 unit increase in BMI results in an estimated 8% increase in coronary heart disease [15] , and excess weight in midlife is associated with increased mortality [16] . Thus , a seemingly small increase in BMI can have a marked impact , particularly in countries with an increasing burden of excess weight , such as the US , where an estimated 68% of adults were overweight or obese in 2008 [17] . Studies in non-European populations have had varied success in replicating the findings at the FTO locus . While several studies showed an association between FTO SNPs and obesity-related phenotypes in Hispanic [5] , [6] , [18] , [19] and Asian populations [20]–[26] , studies of African or African American ( AA ) subjects showed limited support for some of the most consistent FTO GWAS findings initially identified in subjects of European descent [2] , [5] , [6] , [14] , [18] , [19] , [27]–[32] . This lack of generalization in AA may be attributable to the lower levels of linkage disequilibrium ( LD ) with the underlying functional variant ( s ) at the 16q12 . 2/FTO locus , as compared with European Americans ( EAs ) . SNPs discovered in GWAS ( i . e . “index SNPs” ) are often not the functional variants; however , they do tag genomic regions harboring strongly correlated variants , one or more of which are the potentially functional variant ( s ) . Because different ancestral populations differ in their LD patterns , an index SNP discovered in one ancestral group ( e . g . EA ) may or may not be strongly correlated with the functional variant ( s ) in a different ancestral group ( e . g . AA ) . Thus , the index SNP may not show evidence for replication in AAs; however , other SNPs in the region may be in high LD with the functional variant ( s ) , and , hence , measuring these SNPs may further characterize associations within the genomic region . Therefore , a full exploration of potential replication/generalizability of GWAS findings in other ancestral groups requires investigating not only at the index SNP ( s ) , but also examining , if possible , all variants of the region tagged by the index SNP ( s ) . African populations are particularly suited for these studies because the LD pattern between SNPs tends to be substantially weaker than in other ancestral groups , as has been demonstrated for the FTO gene [30] . This process can reduce the number of potential functional variants for follow-up molecular investigation [33] . Given that functional studies can be labor- and cost-intensive , narrowing the associated region is an important step toward elucidating the underlying molecular mechanism . While molecular evaluation of the 16q12 . 2/FTO locus provides some promising leads [34] , the putative functional variant ( s ) in this locus remain under investigation , and fine-mapping studies have been limited with respect to the number of tested variants , sample sizes and inclusion of non-EA populations . Fine-mapping studies , particularly when conducted within a broader region , may also identify additional independent signal ( s ) implicating multiple functional variants in the region . To better understand the relationship of genetic variation at this locus and BMI in AAs , we comprehensively assessed the association of BMI with variants in the 16q12 . 2/FTO region and flanking gene RPGRIP1L in over 20 , 000 AA samples from the Population Architecture using Genetics and Epidemiology ( PAGE ) consortium . We used the Metabochip as genotyping platform [35] , which included all suitable SNPs discovered in the 1000 Genomes Project Pilot 1 for the 16q12 . 2/FTO region and led to successfully genotyping of over 1 , 500 SNPs . This together with imputation into the updated 1000 Genomes Project allowed us to densely fine-mapping this region . In addition , we performed a detailed bioinformatic analysis to propose candidate polymorphisms for follow-up functional evaluation .
The age of the 20 , 488 participants ranged from 20 to 85 years with an average age of 58 . 5 years across the cohorts ( Table 1 ) . The fraction of men included in each cohort varied from 0% to 43% . In all studies that included both genders , men had on average a lower BMI than the women . Participants in the Hypertensive Genetic Epidemiology Network ( HyperGEN ) had the highest BMI while participants in the Multiethnic Cohort ( MEC ) had the lowest BMI . The obesity rate ( defined as BMI≥30 kg/m2 ) across the studies was 46% ranging from 31% to 58% . The targeted 16q12 . 2/FTO fine-mapping region spans 646 kb from 53 , 539 , 509 to 54 , 185 , 773 ( build 37 ) on the long arm of chromosome 16 ( 16q12 . 2 ) , including the large FTO gene ( 411 kb ) as well as 198 kb downstream of FTO , which includes the RPGRIP1L gene and 37 kb upstream of FTO ( Figure 1 ) . On average , we successfully genotyped or imputed one SNP per 172 bp ( = 646 , 264 bp/3 , 756 SNPs ) . The allele frequency distribution of the 3 , 756 SNPs in this region is shown in Table 2 . In contrast to GWAS platforms , a large fraction of SNPs have allele frequencies <5% ( 52 . 5% ) , including 16 . 7% with allele frequency <1% . Twenty-one SNPs are within the exons of FTO and RPGRIP1L , of which 7 are synonymous and 14 are missense ( 2 synonymous and 4 missense are located in FTO ) . The most significant SNPs in the 16q12 . 2/FTO fine-mapping region was rs56137030 ( p-value 8 . 3×10−6 ) , which showed no evidence for heterogeneity ( p = 0 . 13; Table 3 ) . Each A allele of this variant ( allele frequency = 0 . 12 ) increased BMI by 1 . 35% ( 95% confidence interval ( 0 . 76%–1 . 95% ) . Table 3 also displays results for the three next most significant SNPs showing similar significant associations ( p-values 1 . 4×10−5 to 1 . 1×10−5 ) and Table S1 shows results separately for each study . All three SNPs were correlated with rs56137030 ( r2≥0 . 73 in AA and r2≥0 . 91 in EA; Table S2 ) . rs56137030 and the three next most significant SNPs are located in intron 1 of the FTO gene , approximately in the middle of the region ( Figure 1 ) . The nine GWAS index SNPs previous highlighted in EA studies are also located in the same intron 1 FTO region and results for these nine GWAS index SNPs are provided in Table 3 showing p-values<0 . 05 for five out of the nine index SNPs ( 0 . 03 to 3 . 0×10−4 ) . The most significant variant , rs56137030 , was correlated at r2>0 . 5 with 103 SNPs in Europeans ( Figure 1a ) , including eight of the nine GWAS index SNPs ( Table S2 ) . In contrast in AAs rs56137030 was correlated with only 28 SNPs at r2>0 . 5 ( Table S2 , Figure 1b ) . All 28 variants correlated with rs56137030 in AA showed some evidence of association with BMI ( p-values 0 . 0057 to 1 . 1×10−5 ) and no or limited evidence of heterogeneity ( all p for heterogeneity >0 . 04 ) . To investigate if any of these correlated SNPs were associated with BMI independently from rs56137030 we adjusted each SNP for rs56137030 ( including rs56137030 and a second SNP simultaneously in one model ) . None of these variants remained significant at p<0 . 05 ( Table S2 ) . As expected , the p-values of rs56137030 were also less significant in these conditional analyses , particularly for SNPs highly correlated in AA , demonstrating that these findings are not independent . All 28 SNPs correlated with rs56137030 ( r2>0 . 5 ) in AA are located between 53 , 800 , 954 and 53 , 845 , 487 , spanning a 44 . 5 kb region about 104 . 8 kb downstream of the exon 1 boundary , and ending about 1 . 4 Kb after exon 2 ends [no variant in exon 2 was genotyped or imputed and , to our knowledge , only 3 variants ( rs116753298 , rs149393601 , and chr16:53844100 ) all with allele frequency <0 . 1% have been reported in FTO exon 2 so far [36] , [37]] . To predict the molecular mechanisms underlying the genetic association signals , and to identify candidate variants for functional follow-up , rs56137030 and variants in LD ( r2>0 . 5; n = 28 ) were assessed for overlap with eleven different genome-wide functional annotation datasets ( Table S3 , Material and Methods ) . Among these 29 variants , six ( rs11642015 , rs17817497 , rs3751812 , rs17817964 , rs62033408 , and rs1421085 ) were located within candidate intronic regulatory elements , including two ( rs3751812 and rs1421085 ) that were within highly sequence-conserved elements among vertebrates , and two ( rs11642015 and rs1421085 ) that were predicted to have allele-specific binding affinities for different transcription factors . Specifically , we predicted that only the T allele at rs11642015 binds Paired box protein 5 ( PAX5 ) and that the C allele at rs1421085 has a substantially reduced binding affinity for Cut-like homeobox 1 ( CUX1; Figure S1 ) . Outside the known FTO intron 1 region , we observed no strong evidence for a second independent signal: When we adjusted each SNP for rs56137030 the most significant SNPs outside of the FTO intron 1 region was a SNP located at position 53710931 ( no rs number reported ) intronic of the neighbor gene RPGRIP1L ( conditional p-value 7 . 7×10−4 ) followed by rs8051873 in intron 8 of FTO ( conditional p-value 0 . 0011 ) . Table S4 shows results for variants highlighted in previous AA studies . Of these nine SNPs only two SNPs had a p-values<0 . 05 ( rs3751812 , p value 0 . 0012 ) .
In this large study of over 20 , 000 AAs we densely fine mapped the entire FTO gene and adjacent RPGRIP1L gene , spanning a total region of almost 650 kb . We observed significant associations for variants in the known locus in intron 1 of FTO . Due to reduced correlation in AA compared with EA with the most significant SNPs , we were able to substantially reduce the number of functional candidates . Six SNPs were located within candidate intronic regulatory elements , including rs1421085 , for which we predicted allele-specific binding affinity for the transcription factor CUX1 . Because we did not focus solely on the known FTO intron 1 region , we were able to comprehensively investigate the region; however , this approach revealed no evidence for a second independent signal in AA . This is one of the first studies of the Metabochip in an ancestral group that is particularly well suited for fine-mapping GWAS loci , due to its distinct linkage disequilibrium ( LD ) patterns and lower LD overall . Our example clearly shows the powerful approach of studying a large AA population , substantially reducing the number of possible functional variants compared with European descent populations . While very large number of Europeans and EAs are genotyped on the Metabochip , the number of Minority populations genotyped on this chip is substantially smaller; however , this is the focus of the PAGE Study . As many papers using the Metabochip in European populations will be published over the next years , our results show the important contribution that Minority populations and , in particularly AA , will have for systematic mapping of GWAS loci . The importance of the 16q12 . 2/FTO locus for obesity-related traits was identified in genome-wide scans of Europeans . These scans highlighted several variants within the FTO intron 1 , all of which , except for rs6499640 , are in high LD with each other in EA [1]–[7] . Consistently these studies showed an increase in BMI ( ∼1 . 1% to 1 . 3% per risk allele; [7] ) . However , AA studies showed very limited or no evidence for an association with rs9939609 [2] , [5] , [6] , [14] , [18] , [28]–[31] , rs1121980 [18] , [19] , [28] , [30] , [31] , rs17817449 [18] , [19] , [28]–[31] , or the previously reported functional variant rs8050136 [2] , [6] , [19] , [27]–[29] , [31] , [32] . We observed nominal evidence for association with BMI for rs17817449 and rs8050136 , but results were not among the most significant associations . rs1421085 was significantly associated in our study ( p = 3×10−4;Table S2 ) and was also found to be associated in the study from Nock et al . [29] ( n = 469 , p-value = 7×10−4 ) , Hassanein et al . [31] ( n = 4 , 217 , p-value = 3×10−4 ) , and Hester et al . [32] ( n = 4 , 992 , p-value = 0 . 07 ) but not with four smaller AA studies ( ≤1000 AA subjects ) [5] , [18] , [28] , [38] . Consistent with our finding Hassanein et al . [31] also observed a significant association with rs9941349 and rs1558902 ( n = 9 , 881 , p-value = 4×10−6 and n = 4 , 217 , p-value = 2×10−5 , respectively ) , these SNPs have not been genotyped in other AA studies . However , to put these findings into context with other variants in this region a comprehensive evaluation of all variants is needed . To conduct a more comprehensive evaluation of the FTO locus , some AA studies extended the SNPs list from the EA index SNPs described above ( Table S4 ) . Grant et al . [27] analyzed eleven FTO SNPs genotyped as part of their GWAS in about 2 , 000 AA children and only rs3751812 showed a marginally significant association with BMI ( p-value = 0 . 02 ) . Wing et al . [18] , [19] genotyped up to 27 SNPs in the intron 1 of FTO in a cohort study and family study including 288 and 604 AA , respectively , and observed an association of BMI with rs1108102 ( p-value = 5×10−4; [18] ) ; however , this finding was not confirmed in their cohort study or in our larger meta-analysis ( Table S4 ) . A fine mapping study of 47 tagging SNPs in 497 AA children [28] identified an association of rs8057044 ( p-value = 5×10−4 ) with BMI , which was not replicated by the current study . In two fine-mapping studies [5] , [14] , no FTO-BMI associations were noted , possibly because the majority of subjects were lean ( ∼75% had a BMI between 18–25 ) in one study [14] , or because the sample size was small in both studies ( about 1 , 100 ) . Hassanein et al . [31] genotyped 34 tagging SNPs in the FTO intron 1 region in 4 , 217 AA and followed up findings from two variants ( rs3751812 , p-value = 4×10−4 and rs9941349 , p-value = 6×10−5 ) in four additional studies ( n = 5 , 664 ) , adding some support ( p-values ranging from 0 . 016 to 0 . 64 ) which resulted in overall p-values of 2 . 6×10−6 and 3 . 6×10−6 , respectively . The authors concluded that this finding reduced the potential functional variants to those correlated with these two variants . Both variants ( rs3751812 and rs9941349 ) were also associated in our meta-analysis ( p-value 0 . 001 and 0 . 005 , respectively; Table S4 ) , although they were not among the most significant findings . In summary , except for Hassanein et al . [31] , studies in AA were relatively small and showed mixed results that were mainly not replicated in our study . As the number of variants genotyped in any AA study was limited ( 10 to 50 SNPs ) , we were able to investigate if genotyping or imputing additional variants in this region ( in total 3 , 756 ) may even further reduce the list of possible functional variants and search for second independent signal ( s ) . In our analysis , rs56137030 was most significantly associated with BMI . In EA , rs56137030 has a similarly high allele frequecy as the previously reported GWAS index SNPs ( A allele frequency = 0 . 42 ) and was highly correlated with the GWAS index SNPs ( r2≥0 . 87 ) , except for rs6499640 ( r2 = 0 . 12 ) , which is also not strongly correlated with any of the other GWAS index SNPs in EA or AA . In AA , the correlation between rs56137030 and GWAS index SNPs varied substantially ( r2 = 0 . 001 to 0 . 73 ) , demonstrating that studying AA can substantially reduce the bin of correlated SNPs defined by the index signal identified in EA . Specifically , the number of SNPs correlated with rs56137030 at r2>0 . 5 was 103 in EA , but only 28 in AA . Including these 28 SNPs together with rs56137030 in conditional analyses showed that the significance of each of the SNPs , as well as rs56137030 , was substantially reduced , supporting the idea that any of the highly correlated SNPs is a potential functional candidate . rs56137030 and all 28 highly correlated SNPs are non-coding , suggesting that the functional variant is likely to have a cis-regulatory effect . Among these variants , six are located within the candidate intronic regulatory elements , two of which are highly conserved , and two that are predicted to confer allele-specific binding affinities for transcription factors . The variant rs1421085 is within a highly conserved element and may be particularly interesting , because the C allele has a substantially reduced binding affinity for CUX1 , which has been previously implicated in the transcriptional regulation of FTO [34] . Accordingly , this variant is a compelling candidate for follow-up functional evaluation , though outside the scope of the present study . We did not observe any evidence for a second independent signal within the broader 16q12 . 2/FTO region . This finding is consistent with the only other AA study that extended the fine-mapping approach beyond intron 1 to the entire FTO gene including 262 tagSNPs in 1 , 485 subjects [30] . However , even within our substantially larger study of over 20 , 000 AAs , power to identify second independent signals may still have been limited , particularly for less frequent variants or variants with weaker effects , given the increased burden of multiple comparisons that needs to be adjusted for when testing all SNPs across the entire region . Several limitations warrant consideration to inform fine-mapping and functional characterization studies . To comprehensively evaluate the region we not only included directly genotyped SNPs , which included all SNPs known at the time of the chip development and suitable for genotyping SNPs but we also imputed to the most recent version of the 1000 Genomes Project . While this approach provides a rather complete list of variants imputed SNPs ted to be called with varying accuracy . To account for the imputation accuracy we used the dosage , which we showed results in unbiased estimates [39] . However , we also note that the overall p-value of a SNP is impacted by the imputation accuracy ( lower imputation accuracy results in higher p-values ) . Accordingly , it is important for the interpretation of the results that not only the most significant SNPs will be considered as functional variants but also those correlated with the most significant SNPs as done in this paper . Second , for a part of the WHI samples directly genotyped SNPs were only available from a smaller subset of SNPs as genotyping in this subset was based on a GWAS platform and not the dense Metabochip . However , the imputation Rsq as a measurement of the accuracy was very high ( Table S1 ) . Third , despite the relative large samples size of over 20 , 000 AA the statistical significance of the finding is relative weak compared to previous studies in European descent populations for the FTO region . We note that the relative weak power of our study is not due to differences in the observed effect size , e . g . we observed a 1 . 35% change in BMI per allele for the most significant SNP while the replication stage of Willer et al . [7] observed a 1 . 25% change in BMI per allele of their most significant FTO SNP ( rs9939609 ) . However , the substantially lower allele frequency in AA compared with EA ( 12% vs . 41% ) and the larger variance of BMI in AA populations ( e . g . the standard deviation in our study was 6 . 4 kg/m2 compared with 4 . 2 kg/m2 in European populations [7] explains the reduce power . Fourth , our functional characterization is based on in silico analyses and requires experimental validation . Finally , the majority of study participants were female and it is unclear how a predominantly female population may have influenced the results . To our knowledge , this is the largest and most comprehensive fine-mapping study conducted to date in AA . Our findings likely rules out that several of the EA index SNPs in intron 1 of FTO such as rs9939609 , as well as a large fraction of SNPs correlated in EA but not in AA are the underlying functional variants . With rs56137030 and its correlated SNPs , our finding points us closer to the functional variant ( s ) . Among these , rs1421085 is the most compelling candidate for follow-up functional evaluation . Importantly , our study demonstrated that comprehensive fine-mapping in AA provides a powerful approach to narrow in on the functional candidate ( s ) underlying the initial GWAS findings in EA .
All studies were approved by Institutional Review Boards at their respective sites , and all participants provided informed consent . PAGE involves several studies , described briefly below and in more detail in Text S1 as well as at the PAGE website ( https://www . pagestudy . org ) [40] . In brief , participants were recruited from Atherosclerosis Risk in Communities Study ( ARIC ) , GenNet , Hypertension Genetic Epidemiology Network ( HyperGEN ) , Multiethnic Cohort ( MEC ) , and Women's Health Study ( WHI ) . ARIC randomly selected and recruited 15 , 792 participants aged 45–64 at four U . S . communities [41] . GenNet and HyperGEN are two family-based studies designed to investigate the genetics of hypertension and related conditions [42] . The Multiethnic Cohort ( MEC ) is a population-based prospective cohort study of over 215 , 000 men and women in Hawaii and California aged 45–75 at baseline ( 1993–1996 ) and primarily of five ancestries [43] . The WHI encompasses four randomized clinical trials as well as a prospective cohort study of 161 , 808 post-menopausal women aged 50–79 , recruited ( 1993–1998 ) and followed up at 40 centers across the US [44] . All studies collected self-identified racial/ethnic group via questionnaire . We selected all AA participants from ARIC , HyperGEN , and GenNet for genotyping . In MEC , a subset of AAs was selected based on availability of biomarker or as controls for nested case-control studies . WHI included all AAs who provided consent for DNA analysis . We excluded underweight ( BMI<18 . 5 kg/m2 ) and extremely overweight ( BMI>70 kg/m2 ) individuals with the assumption that these extremes could be attributable to data coding errors , an underlying illness or possibly to a familial syndrome and hence , a rare mutation . We also limited analysis to adults defined as age >20 years . For ARIC HyperGEN , GenNet and WHI , BMI was calculated from height and weight measured at time of study enrollment . In MEC , self-reported height and weight were used to calculate baseline BMI . A validation study within MEC has shown high validity of self-reported height and weight . Specifically this study showed that BMI was under-estimated based on self-reported compared to measured weight , but the difference was small ( <1 BMI unit ) and comparable to the findings from national surveys [45] . SNPs were included as part of the Metabochip , a 200 k Illumina customized iSelect array developed through the collaborative efforts of several consortia working on metabolic syndrome related diseases . Details on the design can be found elsewhere ( http://www . sph . umich . edu/csg/kang/MetaboChip/ ) . In brief , SNPs within the 16q12 . 2/FTO region were selected based on 1000 Genomes Pilot 1 and HapMap phase 2 . The boundaries around each GWAS index SNP were determined by identifying all SNPs with r2≥0 . 5 with the index SNP , and then expanding the initial boundaries by 0 . 02 cM in either direction using the HapMap-based genetic map [46] . The total interval size of the 16q12 . 2/FTO region was 646 kb . All 1000 Genomes Pilot 1 SNPs obtained from Sanger Institute ( August 12 , 2009 ) and the Broad Institute ( August 11 , 2009 ) were considered as potential fine mapping SNPs , unless SNP allele frequency was <0 . 01 in all three HapMap samples ( CEU , YRI and HBC/JPT ) . SNPs were excluded if ( a ) the Illumina design score was <0 . 5 or ( b ) there were SNPs within 15 bp in both directions of the SNP of interest with allele frequency of >0 . 02 among Europeans ( CEU ) . SNPs annotated as nonsynonymous , essential splice site , or stop codon were included regardless of allele frequency , design score , or nearby SNPs in the primer . Samples were genotyped on the Metabochip at the Human Genetics Center of the University of Texas-Houston ( ARIC , GenNet and HyperGEN ) , the University of Southern California Epigenome Center ( MEC ) , and the Translational Genomics Research Institute ( WHI ) . Each center also genotyped 90 HapMap YRI ( Yoruba in Ibadan , Nigeria ) samples to facilitate cross-study quality control ( QC ) , as well as 2–3% study-specific blinded replicates to assess genotyping reproducibility . Genotypes were called separately for each study using GenomeStudio with the GenCall 2 . 0 algorithm . Samples were called using study-specific cluster definitions ( based on samples with call rate >95% , ARIC , MEC , WHI ) or cluster definitions provided by Illumina ( GenNet , HyperGEN ) and kept in the analysis if call rate was >95% . We excluded SNPs with GenTrain score <0 . 6 ( ARIC , MEC , WHI ) or <0 . 7 ( GenNet , HyperGEN ) , cluster separation score <0 . 4 , call rate <0 . 95 , and Hardy-Weinberg Equilibrium p<1×10−6 . We also excluded SNPs based on Mendelian errors in 30 YRI trios >1 , replication errors >2 with discordant calls ( when comparing across studies ) in 90 YRI samples >3 , and discordant calls for 90 YRI genotyped in PAGE versus HapMap database >3 . In total , we successfully genotyped 1 , 694 out of 1 , 818 variants in the 16q12 . 2/FTO region . After excluding 165 SNPs that were monomorphic or had very low allele frequency ( <0 . 01% ) we included 1 , 529 variants in the analysis . For ARIC , MEC and WHI we identified related persons using PLINK by estimating identical-by-descent ( IBD ) statistics for all pairs . When apparent first-degree relative pairs were identified , we excluded from each pair the member with the lower call rate . We excluded from further analysis samples with an inbreeding coefficient ( F ) above 0 . 15 ( ARIC , MEC , WHI ) [47] . We determined principal components of ancestry in each study separately using EIGENSOFT [48] , [49] and excluded apparent ancestral outliers from further analysis as described elsewhere [50] . In total 240 subjects failed genotyping ( ARIC = 27 , GenNet = 9 , HyperGEN = 26 , MEC = 140 , and WHI = 27 ) . After further excluding subjects based on age and BMI ( see above ) , a total of 14 , 162 subjects with Metabochip data were included ( 3 , 297 from ARIC , 517 from GenNet , 1 , 171 from HyperGEN , 3 , 865 from MEC , and 5 , 312 from WHI ) . In addition we included 6 , 326 WHI participants genotyped as part of the SNP Health Association Resource ( SHARe ) on the Affymetrix 6 . 0 platform . Details can be found elsewhere [51] . To impute to the 1000 Genomes Project we used as the reference panel the haplotypes of the 1092 samples ( all populations ) from release version 2 of the 1000 Genomes Project Phase I ( ftp://ftp-trace . ncbi . nih . gov/1000genomes/ftp/release/20110521 ) [52] . Combining reference data from all populations has been found to improve imputation accuracy of the low-frequency variants [53] , and , hence , is recommended . We built the target panel by combining all genotype data in the FTO region from all studies . We used genotyped data from the Metabochip for all studies , except for WHI SNP Health Association Resource ( SHARe , n = 6 , 326 samples ) where we used genotype data from the Affymetrix 6 . 0 platform . The target panel was phased using Beagle [54] . We then performed a haplotype-to-haplotype imputation to estimate genotypes ( as allele dosages ) at 1000 Genomes Project variants . The phased target panel was imputed to the interval 53 . 5–54 . 2 Mb on chromosome 16 of the 1000 genomes reference panel using minimac [55] . To evaluate the quality of each imputed SNP we calculated Rsq . We excluded imputed SNPs with Rsq<0 . 9 for SNPs with allele frequency <0 . 5% , Rsq<0 . 8 for SNPs with allele frequency >0 . 5–1% , Rsq<0 . 7 for SNPs with allele frequency >1–3% , Rsq<0 . 6 for SNPs with allele frequency >3–5% , and Rsq<0 . 5 for SNPs with allele frequency >5% . Given the large reference panel this resulted in high imputation quality [51] . To calculate pairwise correlation between variants we used the 1000 Genomes Project data , specifically we used 61 African Americans from the South-west ( ASW ) and 65 European Americans ( Utah residents with Northern and Western European ancestry from the CEPH collection , CEU ) . The association between each SNP and natural log-transformed BMI ( lnBMI ) was estimated using linear regression . SNP genotypes were coded assuming an additive genetic model ( i . e . , 0 , 1 , or 2 copies of the coded allele ) . All analyses were adjusted for age ( continuous ) , sex , and study site ( as applicable ) . All models ( except WHI ) included sex*age interaction terms to account for possible effect modification by sex . In addition , we adjusted for the top two principal components of ancestry . Family data from GenNet and HyperGEN was analyzed using mixed models ( variance component models ) to account for relatedness . Results ( β and SEs for lnBMI ) were combined with fixed-effects meta-analysis weighting the effect size estimates ( β-coefficients ) by their estimated standard errors , using METAL [56] . We evaluated Q-statistic and I2 as a measure of heterogeneity [57] , [58] , to describe the presence or absence of excess variation between the PAGE cohorts . For ease of interpretation , we calculated the % change in BMI per copy of the effect allele based on the beta for the lnBMI . To graphically display the results , we used LocusZoom [59] . We tested for independence of findings by including the most significant variant and each of the other variant into the same model ( i . e . we included 2 variants simultaneously in one model ) . We evaluated if SNPs are independent by investigating the p-value . For an independent SNP the p-value would remain low/similar after adjusted for the most significant SNP . We conducted a bioinformatic characterization for the most significant SNP and all SNPs correlated with the most sigificant SNP ( r2>0 . 5 ) . We implemented in-house Perl scripts to query bioinformatic databases , and assigned each of the 16 SNPs to one or more of the functional annotation datasets listed in Table S3 . These datasets are not mutually exclusive . For example , a SNP can be located within both a candidate regulatory element ( dataset #7 ) and a CTCF binding site ( dataset #10 ) . Because FTO is expressed and may have functional relevance in a wide array of tissues , we defined candidate cis-regulatory elements ( dataset #7 ) as DNaseI hypersensitive sites ( open chromatin loci ) that are present in at least one human cell type . For SNPs that occur within predicted transcription factor binding sites ( datasets #3 and #8 ) , we computed transcription factor binding affinity for each SNP allele using the PWM-scan algorithm [60] , as described previously [61] .
|
Genetic variants within the fat mass– and obesity-associated ( FTO ) gene are associated with increased risk of obesity . To better understand which specific genetic variant ( s ) in this genetic region is associated with obesity risk , we attempt to genotype or impute all known genetic variants in the region and test for association with body mass index as a measurement of obesity in over 20 , 000 African Americans . We identified 29 potential candidate variants , of which one variant ( rs1421085 ) is a particularly interesting candidate for future functional follow-up studies . Our example shows the powerful approach of studying a large African American population , substantially reducing the number of possible functional variants compared with European descent populations .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genome-wide",
"association",
"studies",
"medicine",
"public",
"health",
"and",
"epidemiology",
"epidemiology",
"genetics",
"population",
"genetics",
"biology",
"human",
"genetics",
"public",
"health",
"genetics",
"and",
"genomics"
] |
2013
|
A Systematic Mapping Approach of 16q12.2/FTO and BMI in More Than 20,000 African Americans Narrows in on the Underlying Functional Variation: Results from the Population Architecture using Genomics and Epidemiology (PAGE) Study
|
APOBEC3G ( A3G ) /APOBEC3F ( A3F ) are two members of APOBEC3 cytidine deaminase subfamily . Although they potently inhibit the replication of vif-deficient HIV-1 , this mechanism is still poorly understood . Initially , A3G/A3F were thought to catalyze C-to-U transitions on the minus-strand viral cDNAs during reverse transcription to disrupt the viral life cycle . Recently , it was found more likely that A3G/A3F directly interrupts viral reverse transcription or integration . In addition , A3G/A3F are both found in the high-molecular-mass complex in immortalized cell lines , where they interact with a number of different cellular proteins . However , there has been no evidence to prove that these interactions are required for A3G/A3F function . Here , we studied A3G/A3F-restricted HIV-1 replication in six different human T cell lines by infecting them with wild-type or vif-deficient HIV-1 . Interestingly , in a CEM-derived cell line CEM-T4 , which expresses high levels of A3G/A3F proteins , the vif-deficient virus replicated as equally well as the wild-type virus , suggesting that these endogenous antiretroviral genes lost anti-HIV activities . It was confirmed that these A3G/A3F genes do not contain any mutation and are functionally normal . Consistently , overexpression of exogenous A3G/A3F in CEM-T4 cells still failed to restore their anti-HIV activities . However , this activity could be restored if CEM-T4 cells were fused to 293T cells to form heterokaryons . These results demonstrate that CEM-T4 cells lack a cellular cofactor , which is critical for A3G/A3F anti-HIV activity . We propose that a further study of this novel factor will provide another strategy for a complete understanding of the A3G/A3F antiretroviral mechanism .
Cytidine deaminases are RNA-editing enzymes that target cytosines for conversion to uracils ( C-to-U ) . They belong to the apolipoprotein B mRNA-editing enzyme catalytic polypeptide-like ( APOBEC ) family , which includes activation-induced deaminase ( AID ) , APOBEC1 ( A1 ) , APOBEC2 ( A2 ) , a group of APOBEC3 ( A3 ) , and APOBEC4 ( A4 ) in humans [1] . A1 is the original member of this family and remains the best characterized . It has the capability to introduce a premature termination codon on apolipoprotein B100 ( apoB ) mRNA by C-to-U editing to produce a truncated form of this protein [2] . AID is the second member identified . It edits specific “hotspots” on immunoglobulin gene loci in activated B cells to direct somatic hypermutation and isotype class switching to generate different antibodies [3] . The human A3 subgroup contains seven members: A3A , A3B , A3C , A3DE , A3F , A3G , and A3H . All these proteins have antiretroviral activities against different targets , including exogenous retroviruses and endogenous retroelements [1] . The replication of human immunodeficiency virus type 1 ( HIV-1 ) is inhibited by A3B , A3DE , A3F , A3G , and A3H [4]–[11] , and A3G shows the most powerful anti-HIV-1 activity [9] . A3G/A3F also blocks various retroelements , including LTR retrotransposons and non-LTR retrotransposons [12]–[18] . Nevertheless , HIV-1 is able to elude this defense mechanism and cause disease in humans for two reasons . First , A3B and A3H are poorly expressed in vivo [5] , [7] , [19] , [20] . Second , HIV-1 produces a viral infectivity factor ( Vif ) that binds to and mediates the destruction of A3DE , A3F , and A3G in 26S proteasomes via recruitment of the Cullin5 ubiquitin E3 ligase [6] , [10] , [11] , [21]–[25] . Recently , a protein degradation–independent mechanism was also reported [26] . The antiretroviral mechanism of these A3 proteins has been extensively studied . Initially , it was found that A3G proteins deaminate deoxycytidines ( dCs ) to form deoxyuridines ( dUs ) on viral minus-strand cDNAs during viral reverse transcription [27]–[30] . These C-to-U mutations could either cause the degradation of the viral minus-strand cDNAs or result in G-to-A hypermutations in the plus-strand viral cDNAs , which create havoc in viral transcripts and produce noninfectious virions . Although this model has been favored for a while , recent investigations suggest that the cytidine deamination may not be absolutely required for A3 antiretroviral activity . First , it was found that anti-HIV activity of A3G/A3F does not correlate with hypermutations , but correlates with the reduction of viral reverse transcripts [31] . In addition , A3F and A3H were shown to inhibit HIV-1 replication in the absence of hypermutations [5] , [32] . Second , A3G/A3F and other A3 proteins were also shown to inhibit the replication of some other retroviruses or retrotransposons in the absence of hypermutation [14] , [15] , [18] , [33]–[37] . Further investigations demonstrated that A3G/A3F could interrupt viral reverse transcription by reducing the efficiency of tRNAlys3 priming to the viral RNA template , elongation , and DNA strand transfer [38]–[43] . Moreover , they could also block viral integration [42] , [44] . However , how these viral enzymatic reactions are inhibited by A3G/A3F is still not clear . To understand the mechanism of A3G/A3F antiretroviral activity , efforts have made in another direction by isolating their cellular binding proteins , although there is no evidence to prove that they are functionally required . Both A3G and A3F were found in a ∼700 kDa high-molecular-mass ( HMM ) complex in immortalized cell lines [45] , [46] . Unlike A3F , the A3G HMM complexes are RNase-sensitive; when treated with RNase A , they fall apart into 100 kDa low-molecular-mass ( LMM ) complexes . Biochemical isolations of A3G/A3F-binding proteins from these HMM complexes have generated a list that contains almost 100 different proteins , some of which are shared by A3G and A3F [47]–[50] . However , it is still unclear whether these interactions are essential for A3G/A3F antiviral activities . In this report , we studied the anti-HIV-1 activity of A3G/A3F in six different human T cell lines . We found that A3G/A3F lost anti–HIV-1 activity in the CEM-derived cell line CEM-T4 . Further investigation demonstrated that CEM-T4 cells lacked a cellular cofactor . These results lead to a new direction of investigation on the mechanism of how A3G/A3F inhibits retroviral replication .
To study A3 protein antiretroviral activity , six human T cell lines , including HUT 78 , HUT 78–derived H9 and PM1 , and CEM-derived CEM-SS , CEM-T4 , and A3 . 01 , were selected for HIV-1 infection . HUT 78 and H9 cells are nonpermissive cells because they express A3G and restrict HIV-1ΔVif virus replication; A3 . 01 cells are semipermissive cells because the HIV-1ΔVif virus is not completely restricted although it expresses A3G; and CEM-SS cells are permissive cells because they do not express A3G and no longer restrict HIV-1ΔVif replication [51] . CEM-T4 is a natural subclone of CEM and isolated by its relatively high CD4 expression ( Paul Clapham , personal communication ) . The permissiveness of PM1 and CEM-T4 for the HIV-1ΔVif virus has not been determined . To understand whether there is a significant difference in CD4 levels in these cells , the surface expression of CD4 and CXCR4 was determined by flow cytometry . No significant difference in CD4 and CXCR4 expression was observed , although CEM-T4 cells did show a slight high CD4 expression ( Figure 1A ) . Next , these cells were infected with wild-type or vif-defective HIV-1 , and viral replication curves were determined for 8 d . Although a robust replication of the wild-type virus was observed in all six cell lines , a significant variation in vif-defective virus replication was found ( Figure 1B ) . The replication of HIV-1ΔVif was completely restricted in HUT 78 , H9 , and PM1 cells; less severely restricted in A3 . 01 cells; and not restricted at all in CEM-SS and CEM-T4 cells . This result indicated that CEM-T4 should belong to the permissive cell type with no A3G/A3F expression . To confirm this , A3G and A3F expressions were determined by Western blotting . High levels of A3G were detected in CEM , A3 . 01 , H9 , HUT 78 , and PM1 cells , and no A3G expression was detected in CEM-SS cells ( Figure 1C ) , which is consistent with previous observations . Strikingly , a high-level of A3G expression was also detected in CEM-T4 cells . In addition , A3F expression was also detected in CEMT4 as well as H9 , A3 . 01 , and HUT 78 cells , but not in CEM-SS cells ( Figure 1C ) . These unexpected results caused us to conclude that although CEM-T4 cells express A3G/A3F , they are unable to block HIV-1 replication . Because A3G has much more potent anti–HIV-1 activity than A3F , we decided to further characterize the A3G protein from CEM-T4 cells to understand why A3G could not inhibit HIV-1 replication . First , we cloned and sequenced the A3G gene . No mutation was found in this gene from this cell line ( unpublished data ) . Second , we determined whether a defect at a post-translational level was present that could disrupt A3G interaction with other cellular partners . The A3G protein complex was isolated from CEM-T4 cells by sedimentation of cell lysates in sucrose gradients as described previously [46] . A3G was found in the HMM complexes in CEM-T4 cells as it was in 293T and A3 . 01 cells and , importantly , they were sensitive to conversion to LMM complexes upon RNase A treatment ( Figure 2A ) . This result indicated an intact ability of A3G to interact with cellular RNAs and proteins . Third , we determined A3G cytidine deaminase activity by a scintillation proximity-based assay using an A3G-specific template . As presented in Figure 2B , cells such as 293T and CEM-SS that do not express A3G had marginal deaminase activity; CEM-T4 and A3 . 01 cells that express A3G had higher levels of activity , which were significantly increased after RNase A treatment . These results confirmed that A3G is present in HMM complexes in these cells ( Figure 2A ) and further indicated that A3G in CEM-T4 cells is enzymatically active . Thus , no apparent defect on A3G was detected from CEM-T4 cells . Since the antiretroviral activity of A3G is associated with its presence in virions , we decided to further determine whether A3G could be packaged into HIV-1 virions from CEM-T4 cells . To produce sufficient amounts of virions , cell lines stably producing HIV-1 virions were generated . HIV-1 viruses carrying a neomycin-resistant gene were initially produced from 293T cells after transfection with pNL-Neo or pNL-NeoΔVif . A3 . 01 , H9 , CEM-SS , and CEM-T4 cells were then infected with these viruses , and stably infected cell lines were created by G418 selection . Virions were then purified by ultra-centrifugation , and viral proteins were determined by Western blotting . All the cell lines were able to produce viruses as evidenced by the detection of p24Gag; Vif was only detectable in samples of the wild-type virus but not the vif-deficient virus ( Figure 2C ) . A3G was consistently detected in virions from H9 , A3 . 01 , and CEM-T4 cells , but not CEM-SS cells , and more A3G proteins were found in the vif-defective than the wild-type virions . The level of A3G encapsidation from CEM-T4 cells was lower than that from H9 , but was at least comparable to those from A3 . 01 cells ( Figure 2C; compare lanes 2 , 4 , 6 ) . Thus , A3G was effectively encapsidated by HIV-1 from CEM-T4 cells . We further compared the infectivity of viruses from CEM-SS , A3 . 01 , and CEM-T4 . We found that HIV-1ΔVif infectivity was significantly reduced only in A3 . 01 cells , not CEM-SS and CEM-T4 cells ( Figure 2D ) . This result indicated that the vif-deficient virions produced from CEM-T4 cells were still infectious , which explained why CEM-T4 cells were permissive for HIV-1ΔVif replication . We developed two hypotheses to explain why CEM-T4 cells are permissive for HIV-1ΔVif replication: 1 ) CEM-T4 cells lack a cofactor essential for A3G/A3F anti-HIV activity; and 2 ) CEM-T4 cells express a dominant inhibitor that blocks A3G/A3F activity . A previously described trans-complementation assay was used to test these hypotheses [52] , [53] . In this assay , CEM-T4 cells were fused with 293T cells to form heterokaryons . Because 293T can support A3G/A3F antiviral activity , if A3G/A3F antiviral activity is restored in heterokaryons , it would indicate that a cofactor is missing in CEM-T4 cells; otherwise , CEM-T4 cells should express an inhibitor . To ensure that infectious virions are produced exclusively from heterokaryons , we expressed env-deficient viral particles ( HIV-1ΔEnv ) in CEM-T4 cells and HIV-1 Env protein in 293T cells . No infectious particle could be produced from these two cell lines unless they formed heterokaryons by HIV-1 Env and CD4/CXCR4-mediated cell fusion and trans-complemented for the missing viral components ( Figure 3A ) . The infectious viral particles were then detected by infecting the HIV-reporter cell line TZM-bI , which contains an integrated firefly luciferase gene under the control of the HIV LTR . Initial control experiments were performed with T cells that support A3G/A3F anti–HIV-1 activity . A2 . 01 , a CEM-derived human T cell line that does not express CD4 , was used as a negative control , whereas A3 . 01 , HUT 78 , and H9 were used as positive controls . Since A2 . 01 cells should not fuse to 293T cells , no infectious particles should be recovered . Indeed , very low luciferase activity was detected from TZM-bI cells inoculated with culture supernatant from A2 . 01 and 293T coculture ( Figure 3B; lane 1 ) . In sharp contrast , when A2 . 01 was replaced by A3 . 01 , HUT 78 , or H9 , 10- to 30-fold higher luciferase activities were detected from TZM-bI , indicating a high efficiency of heterokaryon formation and release of infectious particles from these heterokaryons ( Figure 3B; lanes 2–4 ) . Since CEM-SS cells do not express A3G/A3F , they were further used to test the sensitivity of this system to A3G/A3F and Vif activities . When A3G or A3F was not expressed in 293T cells , infectious particles were recovered from heterokaryons regardless of whether Vif was expressed ( Figure 3B; lanes 5 and 8 ) . However , when A3G or A3F was expressed , infectious particles were only recovered from heterokaryons in the presence of Vif ( Figure 3B; lanes 6 and 7 ) , not in the absence of Vif ( Figure 3B; lanes 9 and 10 ) . These results not only demonstrated the efficiency and accuracy of this trans-complementation assay , but also confirmed the expression of corresponding proteins from different constructs . Finally , we fused CEM-T4 cells with 293T cells . When Vif was expressed in heterokaryons , infectious particles were recovered regardless of whether A3G or A3F was expressed ( Figure 3B; lanes 11–13 ) . In sharp contrast , when Vif was not expressed and A3G or A3F was expressed either from CEM-T4 or 293T , no infectious particles were recovered ( Figure 3B; lanes 14–16 ) . These results indicated that A3G/A3F in the heterokaryon between CEM-T4 and 293T could block HIV-1 replication . Thus , we concluded that CEM-T4 cells lack a cellular cofactor required for A3G/A3F anti-HIV activities . Although we found that the endogenously expressed A3G/A3F in CEM-T4 cells lost anti-HIV activity , we wanted to know whether this defect is still present when these proteins are overly expressed . We therefore attempted to express A3G or A3F genes transiently in an HIV-based vector . A3F , A3G , or the noncatalytic A3G ( A3GE259Q ) gene with a 3′ HA tag was inserted into the Nef open reading frame in either pNL4-3 or pNL4-3ΔVif vector so that these A3 proteins could be expressed during HIV-1 replication ( Figure 4A; top panel ) . In total , six different HIV-1 proviral constructs were generated: pNLA3F , pNLA3FΔVif , pNLA3G , pNLA3GΔVif , pNLA3GE259Q , and pNLA3GE259QΔVif . To test their activities , recombinant viruses were first produced by transfection of 293T cells and viral infectivity was determined . To further confirm A3G/A3F gene function in these vectors , we cotransfected the vif-defective version of these vectors with pNL-A1 . pNL-A1 is a Vif expression vector created from pNL4-3 with most of the viral genes deleted , including the viral RNA packaging signal , and can only provide Vif expression in trans for a single round . High levels of A3G or A3GE259Q expression were detected in transfected 293T cells in the absence of Vif by Western blotting ( Figure 4A; middle panels; lanes 5 and 8 ) , and their expressions were decreased by Vif expressed either in cis or in trans ( Figure 4A; lanes 4 , 6 , 7 , 9 ) . The expression of A3F was relatively low ( Figure 4A; lane 2 ) , and Vif further decreased this expression ( Figure 4A; lanes 1 and 3 ) . Next , recombinant viruses were collected to infect TZM-bI cells for a single-round replication . Overall , these viruses had a very similar infectivity in the presence of Vif ( Figure 4A; bottom panel; lanes 1 , 3 , 4 , 6 , 7 , 9 ) . However , A3 proteins decreased viral infectivity when Vif was absent . The wild-type A3G had the most powerful anti-HIV activity , which reduced viral infectivity by around 10-fold ( Figure 4A; lane 5 ) . Both A3F and A3GE259Q mutant reduced viral infectivity by around 2-fold ( Figure 4A; lanes 2 and 8 ) . The low anti-HIV activity of A3F could be due to its low expression , and the low activity of A3GE259Q is consistent with previous reports [16] , [54]–[56] . Nevertheless , this result confirmed that these constructs expressed functional A3G or A3F proteins . Next , we infected CEM-SS and CEM-T4 cells with these recombinant HIV viruses . In this experiment , pNLA3 indicates HIV-1 expressing both Vif and A3 proteins , pNLA3ΔVif indicates viruses expressing only A3 protein , and pNLA3ΔVif+Vif indicates viruses expressing Vif provided in trans by pNL-A1 and A3 proteins . As presented in Figure 4B , the replication of both pNLA3FΔVif and pNLA3FΔVif+Vif were slightly delayed in CEM-SS cells when compared with pNLA3F , and this result was reversed in CEM-T4 cells; the replication of both pNLA3GE259QΔVif and pNLA3GE259QΔVif+Vif were also slightly delayed in CEM-SS cells during the first 3 d of infection , and all three A3GE259Q-expressing viruses replicated equally well in CEM-T4 cells; the replication of pNLA3GΔVif and pNLA3GΔVif+Vif but not pNLA3G viruses was strictly restricted in CEM-SS cells , and all 3 viruses replicated almost equally well in CEM-T4 cells . The growth curve of these viruses in CEM-SS cells was consistent with their infectivity data in Figure 4A , confirming that A3F only weakly inhibited viral replication due to low expression , A3G potently inhibited viral replication , and A3GE259Q had very low antiviral activity . Nonetheless , since all vif-deficient viruses expressing A3G or A3F replicated very well in CEM-T4 cells , these results indicated that A3G/A3F lost their antiviral activity , which further supported that CEM-T4 cells lack a cofactor . The slight delay of pNLA3GΔVif virus replication in CEM-T4 cells could be due to an incorporation of this cofactor from 293T cells that compensated A3G activity during the first round of infection , which provides another piece of evidence that CEM-T4 cells do not express this cofactor . To further confirm these observations , we stably transduced CEM-T4 or CEM-SS cells with an A3G , A3GE259Q , A3F , or GFP gene by the murine leukemia virus ( MuLV ) -based vector pMSCVneo . These genes , containing a 3′ HA-tag , were inserted into pMSCVneo , and recombinant MuLV viruses were created by transfection of Phoenix-AMPHO cell line . Viruses were then used to infect CEM-T4 or CEM-SS cells , and 8 stable cell lines were created by G418 selection . All transduced genes were expressed although the expression of A3F and GFP was lower than that of A3G and A3GE259Q ( Figure 5A ) . It is known that human A3G inhibits MuLV replication , but one group reported that this inhibition might not depend on cytidine deamination [57] . Nevertheless , we wanted to make sure that the tranduced gene did not contain any mutation . The exogenous A3G and A3F genes in CEM-T4 cells were cloned and sequenced , and no mutation was found ( unpublished data ) . Thus , these cell lines should express functional exogenous A3G/A3F proteins . Another possibility that A3G/A3F lost anti-HIV activity is that they are mislocalized in CEM-T4 cells . To exclude this possibility , we compared A3G subcellular localization in CEM-T4 and CEM-SS cells by confocal microscopy . A3G and A3F are both known as cytoplasmic proteins , and can be found in the mRNA processing ( P ) bodies [58] , [59] . A recent report showed that A3F is colocalized with cellular protein MOV10 [48] , which is also a P-body protein [60] . When stable CEM-T4 and CEM-SS cells expressing A3G were stained with anti-A3G and anti-MOV10 antibodies , A3G protein was found in the cytoplasm of both cell lines and was colocalized with MOV10 ( Figure 5B ) . Thus , A3G/A3F should not be mislocalized in CEM-T4 cells . Finally , we determined HIV-1 replication in these cell lines . As presented in Figure 5C , both wild-type and vif-deficient HIV-1 replicated equally well in CEM-T4 cells expressing A3G , A3GE259Q , A3F , or GFP , suggesting that the vif-deficient virus is not restricted by any of these genes . However , in CEM-SS cells , although the replication of vif-deficient virus was not restricted by GFP or A3GE259Q , it was severely restricted by A3G or A3F . The anti-HIV activity of A3F in CEM-SS cells was relatively lower than that of A3G , which could be due to its low expression as shown in Figure 5A . In addition , the poor activity of A3GE259Q further confirmed that this noncatalytic A3G mutant has poor antiviral activity . Thus , we concluded that A3G/A3F failed to inhibit HIV-1 replication in CEM-T4 cells even though they were overexpressed .
In this report , we studied A3G/A3F anti–HIV-1 activity in 6 different human T cell lines . We found that in one cell line , CEM-T4 , A3G/A3F lost anti–HIV-1 activity due to the absence of a cellular factor , which is very critical for A3G/A3F anti-HIV activity . Although A3G/A3F potently inhibits HIV-1 replication , this mechanism is still poorly defined . A3G/A3F has two conserved zinc-binding domains . In the process of blocking HIV-1 replication , these two domains have different functions . The N-terminal domain has a high affinity for RNAs , which normally serves as a virion-packaging signal , and the C-terminal domain has cytidine deamination activity , which is the real catalytic domain [61]–[63] . Initially , it was found that the noncatalytic A3G mutant E259Q still retained intact anti-HIV activity , suggesting that the cytidine deaminase activity is not required for antiviral activity [64] . However , this result could not be reproduced by the other investigators [16] , [54]–[56] , and we also found that the E259Q mutant had very marginal anti-HIV activity ( Figures 4B and 5C ) . Nevertheless , it is clear that A3G could inhibit the replication of hepatitis B virus and human T cell leukemia virus type 1 in the absence of cytidine deamination [36] , [37] , and many similar cases have been found in blocking HIV-1 , adeno-associated virus , and retrotransposon replications by different A3 proteins [5] , [14] , [15] , [18] , [32]–[35] . Thus , although cytidine deamination is not required , the cytidine deaminase activity is absolutely required for A3 antiretroviral activities . As introduced before , A3G/A3F can reduce the accumulation of HIV-1 cDNAs , which correlates well with their anti-HIV activity . Previously , two groups reported that uracil DNA glycosylases-2 ( UNG ) , a host DNA repair enzyme , is involved in the degradation of viral cDNAs containing uracils [65] , [66] . However , another two groups obtained conflicting results and dismissed the role of this enzyme in A3G antiviral activity [16] , [67] . In addition , A3G/A3F were shown to inhibit tRNAlys3 priming , elongation , or DNA strand transfer during reverse transcription and viral integration [38]–[44] . How A3G/A3G can virtually disrupt these critical reactions in the viral life cycle needs to be understood . Notably , the presence of an A3G/A3F antiretroviral cofactor may help decipher this poorly defined mechanism . One possibility is that this cofactor has nuclease activity that directly degrades viral reverse transcripts . Alternatively , it may increase the affinity of A3G/A3F to viral RNAs or cDNAs so that they can compete with viral reverse transcriptase and integrase for their substrates . Although it was expected that A3G/A3F would specifically interact with viral RNAs or cDNAs to block HIV-1 replication , in vitro study with recombinant A3G proteins failed to demonstrate such specificity [68] , [69] . In the case of another deaminase , A1 , it has been shown that its editing activity is a highly sequence-specific process dependent upon the primary , secondary , and perhaps even tertiary structure of the RNA substrate . Further investigations have identified two host factors that are required to complement A1 for apoB mRNA editing: ACF ( APOBEC1 complementation factor ) and its splice variant ASP ( APOBEC1 stimulating protein ) [70] , [71] . ACF is very homologous to the RNA-binding protein GRY-RBP and binds to the U-rich “mooring” sequence of apoB mRNA [71] . Thus , similar cofactors may also be required for A3G/A3F . Two groups have reported that human A3G/A3F could inhibit yeast LTR retrotransposon Ty1 in Saccharomyces cerevisiae [12] , [17] . Whether a similar cofactor is required for A3G/A3F antiretroviral activity in yeast therefore becomes an interesting question . If it is required , it would imply that a highly conserved antiviral ortholog gene exists in different organisms . Otherwise , it may indicate that this cofactor is very specific for HIV-1 . The latter possibility could also suggest that A3G/A3F uses different antiviral mechanisms to target different retroviruses . Nevertheless , knowledge of the cofactor involved in the process of blocking HIV-1 replication by A3G/A3F is critical to our understanding of HIV pathogenesis . Further characterization of this cofactor will lead to a complete understanding of A3G/A3F anti-HIV activity .
HIV-1 proviral constructs pNL4-3 and pNL4-3ΔVif and human A3G or A3F expression pcDNA3 . 1-V5-6XHis vectors have been described previously [6] , [11] . The noncatalytic A3G mutant ( A3GE259Q ) was created by site-directed mutagenesis in pcDNA3 . 1 vector . pNL-Neo , pNL-NeoΔVif , pNLA3G , pNLA3GΔVif , pNLA3F , pNLA3FΔVif , pNLA3GE259Q , and pNLA3GE259QΔVif were created by replacing the firefly luciferase gene in pNL-Luc and pNL-LucΔVif with a neomycin-resistant gene or an A3G , A3F , or A3GE259Q gene containing a 3′ HA-tag by NotI/XhoI digestion , respectively . In addition , an A3F , A3G , A3GE259Q , or GFP gene with a 3′ HA tag was inserted into the pMSCVneo vector by EcoRI/XhoI digestion . pNL4-3ΔGag and pNL4-3ΔEnv were created by SphI/AgeI double digestion or NheI single digestion of pNL4-3 , followed by large Klenow fragment treatment before T4 ligation . The NheI site is still active in pNL4-3ΔEnv although the env gene was inactivated by frame-shift . To create pNL4-3ΔGagΔVif and pNL4-3ΔEnvΔVif , the parental plasmids were digested with PfiMI and filled in with a linker from annealing two DNA oligonucleotides ( 5′-CTAGCTAGCTAGCCGGCAGA-3′ , 5′-GCCGGCTAGCTAGCTAGTCT-3′ ) . The HIV indicator cell line TZM-bI and human T cell lines HUT 78 , H9 , PM1 , CEM-SS , CEM-T4 , A3 . 01 , and A2 . 01 were from the National Institutes of Health ( NIH ) AIDS Research and Reference Reagent Program . The Phoenix-AMPHO cell line was from Dr . G . Nolan ( Stanford University ) . T cell lines were cultured in RPMI 1640 with 10% fetal bovine serum ( HyClone ) . Phoenix-AMPHO , 293T , and TZM-bI were cultured in DMEM with 10% bovine calf serum ( HyClone ) . HIV-1 or MuLV viruses were produced from 293T or Phoenix-AMPHO cells by the standard calcium phosphate transfection . A total of 1×105 cells were incubated with 100 ng wild-type or Vif-defective HIV viruses at 37°C for 3 h . After removal of the inocula , followed by 3 extensive washings , cells were cultured in 24-well plates for 8 d . Culture supernatants were then collected daily for measurement of p24Gag by ELISA . A scintillation proximity-based assay was used as described previously [46] , [72] . The rabbit anti-human A3G polyclonal antibody was from the NIH AIDS Research and Reference Reagent Program . The mouse anti-human A3F polyclonal antibody was from Abnova , Taiwan . Actin was detected by a polyclonal antibody ( C-11; Santa Cruz Biotechnology ) . HIV-1 p24Gag and Vif were detected by antibodies ( nos . 3537 and 6459 ) from the NIH AIDS Research and Reference Reagent Program . HRP-conjugated anti-rabbit or mouse IgG secondary antibodies were from PIERCE . Detection of the HRP-conjugated antibody was performed using Supersignal Wetpico Chemiluminescence Substrate kit ( PIERCE ) . A previously established protocol was adopted [52] , [53] . Briefly , 293T cells were seeded in 6-well plates at 8×105/well in 2 ml medium . After 12 h , cells were transfected with 6 μg of HIV Env expression vector pNL4-3ΔGag or pNL4-3ΔGagΔVif in the presence or absence of A3G or A3F expression vector and washed with PBS 4 h later . Simultaneously , 8×105 T cells were infected with 500 ng of VSV-pseudotyped Env-defective HIV-1 from pNL4-3ΔEnv– or pNL4-3ΔEnvΔVif–transfected 293T cells at 37°C for 3 h . After removal of the inocula and extensive washing , infected T cells were added to the Env-expressing 293T cell culture . After 48 h , supernatants from these cocultures were collected to infect TZM-bI cells . Viral infectivity was finally determined by measuring cellular luciferase activities after another 48 h . CEM-T4 cells stably expressing exogenous A3G from the pMSCVneo vector were fixed in a buffer ( 5% formaldehyde+2% sucrose in PBS ) . Fixed samples were permeabilized for 30 min at room temperature in a permeabilization buffer ( 1% Triton X-100 , 10% sucrose in PBS ) prior to incubation with antibodies . Cells were then stained with a mouse anti-A3G monoclonal antibody at 1∶100 ( ImmunoDiagnostics , obtained from the NIH AIDS Research and Reference Reagent Program ) and a rabbit anti-MOV10 polyclonal antibody at 1∶100 ( Proteintech Group ) . Cover slips were washed once in PBS ( 5 min at room temperature ) and incubated with secondary antibodies , including goat anti-mouse IgG Alexa Fluor 488 and goat anti-rabbit IgG Alexa Fluor 594 ( Invitrogen ) . Cells were further stained with 1 μg/ml Hoechst 33342 ( Sigma-Aldrich , St . Louis , Missouri , United States of America ) . Cover slips were then washed twice with PBS and mounted onto microscope slides with glycerol gelatin ( Sigma-Aldrich ) and were stored at 4°C in the dark until analyzed by a confocal microscope Olympus Fluoview 1000 . The GenBank accession numbers for human APOBEC3G and APOBEC3F are BC024268 and BC038808 .
|
Cytidine deaminases are host enzymes that remove the amino group from the cytidine base on single-stranded DNA or RNA , resulting in a replacement of the cytidine with a uracil . Such replacement may alter the amino acid–coding sequence of the gene and change protein function . It has been well documented that APOBEC1 and AID play very important roles in protein metabolism and immune response via this mechanism . Interestingly , recent advances in retroviral researches have discovered that the seven cytidine deaminases ( APOBEC3A to 3H ) on human Chromosome 22 can restrict retrovirus replication . In particular , APOBEC3G and APOBEC3F have the most powerful anti–HIV-1 activity and also inhibit other retroviruses , including retrotransposons . They could inhibit viral replication in either a cytidine deamination-dependent or -independent manner , but the precise mechanism remains to be defined . In this report , we found that in a particular human T cell line , APOBEC3G and APOBEC3F failed to block HIV-1 replication . Further analyses indicated that this cell line lacks a cellular factor , which is very critical for APOBEC3G and APOBEC3F antiviral activity . Thus , APOBEC3G and APOBEC3F require a cofactor to inhibit viral replication , and identification of this cofactor will provide an important strategy to decipher this poorly defined antiretroviral mechanism .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"virology/immunodeficiency",
"viruses",
"microbiology/medical",
"microbiology",
"microbiology/innate",
"immunity"
] |
2008
|
APOBEC3G and APOBEC3F Require an Endogenous Cofactor to Block HIV-1 Replication
|
There has been much interest in studying evolutionary games in structured populations , often modeled as graphs . However , most analytical results so far have only been obtained for two-player or linear games , while the study of more complex multiplayer games has been usually tackled by computer simulations . Here we investigate evolutionary multiplayer games on graphs updated with a Moran death-Birth process . For cycles , we obtain an exact analytical condition for cooperation to be favored by natural selection , given in terms of the payoffs of the game and a set of structure coefficients . For regular graphs of degree three and larger , we estimate this condition using a combination of pair approximation and diffusion approximation . For a large class of cooperation games , our approximations suggest that graph-structured populations are stronger promoters of cooperation than populations lacking spatial structure . Computer simulations validate our analytical approximations for random regular graphs and cycles , but show systematic differences for graphs with many loops such as lattices . In particular , our simulation results show that these kinds of graphs can even lead to more stringent conditions for the evolution of cooperation than well-mixed populations . Overall , we provide evidence suggesting that the complexity arising from many-player interactions and spatial structure can be captured by pair approximation in the case of random graphs , but that it need to be handled with care for graphs with high clustering .
Graphs are a natural starting point to assess the role of population structure in the evolution of cooperation . Vertices of the graph represent individuals , while links ( edges ) define interaction and dispersal neighborhoods . Classical models of population structure , such as island models [1 , 2] and lattices [3 , 4] , often developed before the current interest in complex networks [5 , 6] , can all be understood as particular instances of graphs [7 , 8] . More recently , the popularity of network theory has fueled a renewed interest in evolutionary dynamics on graphs , especially in the context of social behaviors such as cooperation and altruism [7–21] . When selection is weak on two competing strategies , such that fitness differences represent only a small perturbation of a neutral evolutionary process , a surprisingly simple condition for one strategy to dominate the other , known as the “sigma rule” , holds for a large variety of graphs and other models of spatially structured populations [22] . Such a condition depends not only on the payoffs of the game describing the social interactions , but also on a number of “structure coefficients” . These coefficients are functions of demographic parameters of the spatial model and of its associated update protocol , but are independent of the payoffs . In the case of two-player games , the sigma rule depends on a single structure coefficient σ . The larger this σ , the greater the ability of spatial structure to promote the evolution of cooperation or to choose efficient equilibria in coordination games [22] . Partly for this reason , the calculation of structure coefficients for different models of population structure has attracted significant interest during the last years [8 , 21–27] . Despite the theoretical and empirical importance of two-player games , many social interactions involve the collective action of more than two individuals . Examples range from bacteria producing extracellular compounds [28–31] to human social dilemmas [32–36] . In these situations , the evolution of cooperation is better modeled as a multiplayer game where individuals obtain their payoffs from interactions with more than two players [37–43] . An example of such multiplayer games is the volunteer’s dilemma , where individuals in a group must decide whether to volunteer ( at a personal cost ) or to ignore , knowing that volunteering from at least one individual is required for a public good to be provided [44–46] . Importantly , such a multiplayer interaction cannot be represented as a collection of pairwise games , because changes in payoff are nonlinear in the number of co-players choosing a particular action . Multiplayer games such as the volunteer’s dilemma can also be embedded in graphs , assuming , for instance , that nodes represent both individuals playing games and games played by individuals [47–49] . Most previous studies on the effects of graph structure on multiplayer game dynamics have relied on computer simulations [49] . However , similar to the two-player case , some analytical progress can be made if selection is assumed to be weak . In the multiplayer case , the sigma rule depends no longer on one , but on up to d − 1 structure coefficients , where d is the number of players [50] . Although exact formulas for structure coefficients of multiplayer games can be obtained for relatively simple models such as cycles [51] , analysis has proved elusive in more complex population structures , including regular graphs of arbitrary degree . Indeed , extending analytical results on evolutionary two-player games on graphs to more general multiplayer games is an open problem in evolutionary graph theory [52] . Here , we contribute to this body of work by deriving approximate analytical expressions for the structure coefficients of regular graphs updated with a Moran death-Birth model , and hence for the condition of one strategy to dominate another according to the sigma rule . The expressions we find for the structure coefficients suggest that regular graphs updated with a Moran death-Birth model lead to less stringent conditions for the evolution of cooperation than those characteristic of well-mixed populations . Computer simulations suggest that our approximations are good for random regular graphs , but that they systematically overestimate the condition for the evolution of cooperation in graphs with more loops and higher clustering such as rings and lattices . In these cases , cooperation can be no longer promoted , but even be hindered , with respect to the baseline case of a population lacking spatial structure .
We consider stochastic evolutionary dynamics on a graph-structured population of size N . Each individual is located at the vertex of a regular graph of degree k . Individuals obtain a payoff by interacting with their k neighbors in a single d-person symmetric game ( i . e . , d = k+1 ) . If j co-players play A , a focal A-player obtains aj whereas a focal B-player obtains bj , as indicated in Table 1 . We model the stochastic evolutionary dynamics as a Markov process on a finite space state . More specifically , we consider a Moran death-Birth process [12 , 14 , 53] according to which , at each time step: ( i ) a random individual is chosen to die , and ( ii ) its neighbors compete to place a copy of themselves in the new empty site with probability proportional to 1 − w + w × payoff , where the parameter w measures the intensity of selection . Without mutation , such a Markov process has two absorbing states: one where all vertices are occupied by A-players and one where all vertices are occupied by B-players . Let us denote by ρA the fixation probability of a single A-player in a population of B-players , and by ρB the fixation probability of a single B-player in a population of A-players . We take the comparison of fixation probabilities , i . e . ρ A > ρ B , ( 1 ) as a measure of evolutionary success [54] and say that A is favored over B if condition ( 1 ) holds . Under weak selection ( i . e . , w ≪ 1 ) the condition for A to be favored over B holds if the sigma rule for multiplayer games [50] is satisfied , i . e . , if ∑ j = 0 d - 1 σ j f j > 0 , ( 2 ) where σ0 , … , σd−1 are the d structure coefficients ( constants that depend on the population structure and on the update dynamics ) , and f j = a j - b d - 1 - j , j = 0 , 1 , … , d - 1 , ( 3 ) are differences between payoffs , which we will refer to in the following as the “gains from flipping” . The gains from flipping capture the change in payoff experienced by a focal individual playing B in a group where j co-players play A when all players simultaneously switch strategies ( so that A-players become B-players and B-players become A-players ) . It turns out that the payoffs of the game only enter into condition ( 1 ) via the gains from flipping Eq ( 3 ) , as the structure coefficients are themselves independent of aj and bj . Structure coefficients are uniquely determined up to a constant factor . Setting one of these coefficients to one thus gives a single structure coefficient for d = 2 [22] . For d > 2 , and in the usual case where structure coefficients are nonnegative , we can impose ∑ j = 0 d - 1 σ j = 1 without affecting the selection condition ( 2 ) . For our purposes , this normalization turns out to be more useful than setting one coefficient to one , as it allows us to rewrite the sigma rule Eq ( 2 ) as ∑ j = 0 d - 1 ς j f j = E f ( J ) > 0 , ( 4 ) where f ( j ) ≡ fj , and J is the random variable with probability distribution prescribed by the “normalized structure coefficients” ς j = σ j / ∑ i = 0 d - 1 σ i . In light of condition ( 4 ) , the sigma rule can be interpreted as stating that strategy A is favored over B if the expected gains from flipping are greater than zero when the number of co-players J is distributed according to the normalized structure coefficients . From this perspective , different models of population structure lead to different normalized structured coefficients and hence to different expected gains from flipping , which in turn imply different conditions for strategy A to be favored over B in a given multiplayer game [51] . For instance , a well-mixed population with random group formation updated with either a Moran or a Wright-Fisher process leads to normalized structure coefficients given by [39 , 40]: ς j W = N d ( N - 1 ) if 0 ≤ j ≤ d - 2 N - d d ( N - 1 ) if j = d - 1 . ( 5 ) A normalized sigma rule such as the one given by Eq ( 4 ) holds for many spatial models and associated updating protocols [50 , 51] . Here , we focus on the case of regular graphs updated with a Moran death-Birth process . We provide exact expressions for the case of cycles for which k = 2 . For k ≥ 3 , we bypass the difficulties of an exact calculation by using a combination of pair approximation [55 , 56] and diffusion approximation [14] . Our approach implicitly assumes that graphs are equivalent to Bethe lattices ( or Cayley trees ) with a very large number of vertices ( N ≫ k ) . In addition , weak selection intensities ( wk ≪ 1 ) are also required for an implicit argument of separation of timescales to hold . In order to assess the validity of our analytical approximations , we implemented a computational model of a Moran death-Birth process in three different types of regular graphs ( rings , random graphs , and lattices ) with different degrees and estimated numerically the fixation probabilities ρA and ρB as the proportion of realizations where the mutant succeeded in invading the wild-type .
Going beyond the complete graph representing a well-mixed population , the simplest case of a regular graph is the cycle , for which k = 2 ( and consequently d = 3 ) . In this case , we find the following exact expressions for the structure coefficients ( S1 Text , Section 1 ) : ς 0 G = 1 2 ( N - 2 ) , ς 1 G = 1 2 , ς 2 G = N - 3 2 ( N - 2 ) . ( 6 ) For large N , the structure coefficients reduce to ς 0 G = 0 , ς 1 G = ς 2 G = 1 / 2 and the sigma rule Eq ( 4 ) simplifies to a 1 + a 2 > b 1 + b 0 . ( 7 ) This is also the condition for the boundary between a cluster of A-players and a cluster of B-players to move in favor of A-players for weak selection [57] ( Fig 1 ) . Condition ( 7 ) implies that A can be favored over B even if A is strictly dominated by B ( i . e . , aj < bj for all j ) as long as the payoff for mutual cooperation a2 is large enough so that a2 > b0+ ( b1 − a1 ) ; a necessary condition for this inequality to hold is that A strictly Pareto dominates B ( i . e . , a2 > b0 ) . Such a result is impossible in well-mixed populations , where the structure coefficients Eq ( 5 ) prevent strictly dominated strategies from being favored by selection . Condition ( 7 ) provides a simple example of how spatial structure can affect evolutionary game dynamics and ultimately favor the evolution of cooperation and altruism . For regular graphs of degree k ≥ 3 , we find that the structure coefficients can be approximated by ( S1 Text , Section 2 ) ς j G = ( k - 2 ) k - 1 - j ( k + 2 ) ( k + 1 ) k 2 ∑ ℓ = 0 k - 1 ( k - ℓ ) k 2 - ( k - 2 ) ℓ υ ℓ , j , k + 2 k + ( k - 2 ) ℓ τ ℓ , j , k , ( 8 ) where υ l , j , k = ( k − 1 − l k − 1 − j ) 1 ( k − 1 ) k − 1 − l + ( l k − j ) k − 2 ( k − 1 ) l , ( 9 ) and τ l , j , k = ( k − 1 − l k − j ) k − 2 ( k − 1 ) k − 1 − l + ( l k − 1 − j ) 1 ( k − 1 ) l . ( 10 ) These expressions are nontrivial functions of the degree of the graph k and thus difficult to interpret . For instance , for k = 3 , we obtain ς G = ( 7 144 , 31 144 , 61 144 , 45 144 ) . The previous results hold for any symmetric multiplayer game with two strategies . To investigate the evolution of multiplayer cooperation , let us label strategy A as “cooperate” , strategy B as “defect” , and assume that , irrespective of the focal player’s strategy , the payoff of a focal player increases with the number of co-players playing A , i . e . , a j + 1 ≥ a j and b j + 1 ≥ b j for all j . ( 11 ) This restriction on the payoffs is characteristic of “cooperation games” [51] in which playing A is beneficial to the group but might be costly to the individual . Well-known multiplayer games belonging to this large class of games include different instances of volunteer’s dilemmas [44 , 46] , snowdrift games [58] , stag hunts [59] , and many other instances of public , club , and charity goods games [43] . We are interested in establishing whether graph-structured populations systematically lead to structure coefficients that make it easier to satisfy the normalized sigma rule Eq ( 4 ) than well-mixed populations ( the baseline case scenario of a population with no spatial structure ) for any cooperation game satisfying condition ( 11 ) . In other words , we ask whether a graph is a stronger promoter of cooperation than a well-mixed population . Technically , this is equivalent to asking whether the set of games for which cooperation is favored under a graph contains the set of games for which cooperation is favored under a well-mixed population , i . e . , whether a graph is greater than a well-mixed population in the “containment order” [51] . A simple sufficient condition for this is that the difference in normalized structure coefficients , ςG − ςW , has exactly one sign change from − to + [51] . This can be verified for any N > 3 in the case of cycles ( k = 2 ) by inspection of eqs ( 5 ) and ( 6 ) . For large regular graphs of degree k ≥ 3 and hence multiplayer games with d ≥ 4 players , we checked the condition numerically by comparing eqs ( 5 ) and ( 8 ) for k = 3 , … , 100 . We find that ςG − ςW always has a single sign change from − to + and hence that , in the limit of validity of our approximations , regular graphs promote more cooperation than well-mixed populations for all games fulfilling Eq ( 11 ) ( Fig 2 ) . In the following , we explore in more detail the sigma rule for particular examples of multiplayer games . To assess the validity of our approximations , we compare our analytical results with explicit simulations of evolutionary dynamics on graphs ( Fig 3 , N = 100; S1 Fig , N = 500 ) . We implemented three different kinds of regular graphs: ( i ) random regular graphs , ( ii ) rings ( generalized cycles in which each node is connected to k/2 nodes to the left and k/2 nodes to the right ) , and ( iii ) lattices ( a square lattice with von Neumann neighborhood with k = 4 , a hexagonal lattice with k = 6 , and a square lattice with Moore neighborhood and k = 8 ) . Analytical predictions are in good agreement with simulation results in the case of cycles ( i . e . , rings with k = 2 , for which our expressions are exact ) and for all random regular graphs that we explored . Contrastingly , for rings with k ≥ 4 and lattices , our approximations tend to underestimate the critical benefit-to-cost ratio beyond which the fixation probability of cooperators is greater than that of defectors . In other words , our analytical results seem to provide necessary but not sufficient conditions for cooperation to be favored . Such discrepancies stem from the fact that our analysis assumes graphs with no loops such as Cayley trees; the error induced by our approximations is more evident when looking at the actual fixation probabilities ( S2 Fig , N = 100 , S3 Fig , N = 500 ) and not just at their difference . As all graphs with k > 2 we considered do contain loops , such mismatch is expected—in particular for rings and lattices , which are characterized by high clustering . Perhaps more importantly , our simulations suggest that the critical benefit-to-cost ratio for the volunteer’s dilemma without cost sharing in rings and lattices with k ≥ 6 is greater than the corresponding values for random graphs and well-mixed populations . This illustrates a case in which a graph-structured population updated with a death-Birth process leads to less favorable conditions for the evolution of cooperation than a well-mixed population .
We studied evolutionary multiplayer game dynamics on graphs , focusing on the case of a Moran death-Birth process on regular structures . First , we used a combination of pair approximation and diffusion approximation to provide analytical formulas for the structure coefficients of a regular graph , which together with the payoffs from the game determine when a strategy is more abundant than another in the limits of weak selection and weak mutation . Such a condition is valid for any symmetric multiplayer game , including the volunteer’s dilemma [44–46] and other multiplayer social dilemmas discussed in the recent literature [38 , 41 , 58 , 59 , 61] . The condition can be used to determine the specific conditions ( in terms of the degree of the graph and the parameters of the game , such as payoff costs and benefits ) under which cooperation will thrive . The structure coefficients also provide a way of comparing the graph with other population structures , such as the well-mixed population . In particular , and to the extent that our approximations are valid , graphs updated with a death-Birth process are more conducive to the evolution of cooperation than well-mixed populations for a large class of games ( see condition ( 11 ) ) . Second , we used numerical simulations to estimate the fixation probabilities and the difference in fixation probabilities of different strategies for particular examples of games ( volunteer’s dilemma with and without cost sharing ) and graphs ( random regular graphs , rings , and lattices ) . Although simulations agree very well with the analytical approximations in the case of random regular graphs , discrepancies are evident in the case of rings and lattices , which are characterized by higher clustering and for which pair approximation is not sufficiently accurate . In these cases , the analytical approximations systematically overestimate the ability of a graph to promote the evolution of cooperation . Importantly , in the case of the volunteer’s dilemma without cost sharing and for rings or lattices of relatively large degree , the critical benefit-to-cost ratio above which cooperation is favored is greater , not smaller , than the corresponding value for a well-mixed population . Even though detrimental effects of spatial structure on cooperation have been previously noted in similar studies [62] , our results are counterintuitive given the updating protocol and the intensity of selection we explored . Indeed , a death-Birth Moran process under weak selection would always favor cooperation ( with respect to a well-mixed population of the same size ) for any linear cooperation game , including any collection of two-player cooperation games . Our simulations show that this might not be the case when social dilemmas are instead modelled as nonlinear games such as the volunteer’s dilemma . We used pair approximation and diffusion approximation to find approximate values for the structure coefficients , but other approaches can be used to obtain better estimates of them . In particular , coalescent theory [63] allows us to write the sigma rule in terms of selection coefficients ( dependent on the payoffs of the game and the demographic parameters of the model ) and expected coalescence times under neutrality [64 , 65]; however , such expected coalescence times can be difficult to obtain exactly . Alternatively , for small graphs , the sigma rule and hence the structure coefficients can be explicitly calculated from the transition matrix of the evolutionary process ( cf . Appendix C of Ref . [26] ) . Finally , we note that even in cases for which the structure coefficients are difficult to obtain by purely analytical means , they can be estimated numerically , either indirectly ( by estimating the expected times to coalescence ) or directly ( by computing and comparing fixation probabilities ) . For simplicity , we assumed that a focal player obtains its payoff from a single multiplayer game with its k immediate neighbors . Such assumption allowed us to consider multiplayer interactions on graphs in a straightforward way . However , this is in contrast with a common assumption of many studies of multiplayer spatial and network games in which a focal player’s total payoff is the sum of payoffs obtained in k+1 different games , one “centered” on the focal player itself and the other k centered on its neighbors [47–49] . As a result , focal players interact not only with first-order but also with second-order neighbors , which would lead to more intricate structure coefficients . For example , in this case the structure coefficients of a cycle are given by [51 , 66] ς 0 G * = N + 1 3 ( 2 N - 3 ) , ς 1 G * = 2 N - 1 3 ( 2 N - 3 ) , ς 2 G * = N - 3 2 N - 3 . ( 22 ) These values are different from those we calculated under the assumption that individuals play a single game with first-order neighbors , given by Eq ( 6 ) . For N > 4 , the structure coefficients fulfill ς G ≥ con ς G * , meaning that our assumption of payoffs from a single game leads to less restrictive conditions for cooperation to be favored by selection . This observation is in line with previous results for pairwise games on graphs suggesting that the condition for the evolution of cooperation is optimized when interaction and replacement neighborhoods coincide [67] , which corresponds to our assumption of individuals playing a single game . Future work should consider the calculation of structure coefficients for the cases where the payoff to a player also depends on games centered on neighbors and how the condition for the promotion of cooperation differs from the one resulting from our simplifying assumption . We modelled social interactions as multiplayer matrix games with two discrete strategies ( A and B ) and obtained our results by assuming that selection is weak ( w is small ) . Alternatively , one could model the same multiplayer game but assume instead that players can choose between two similar mixed strategies z and z + δ , where z and z + δ refer to the probability of playing A for each strategy , and δ is small [43 , 68 , 69] . In such a “δ-weak selection” scenario , and for any number of players , only a single structure coefficient is needed to identify conditions under which a higher probability of playing A is favored by natural selection . For transitive graphs of size N and degree k , this structure coefficient is given by [7 , 25] σ = ( k + 1 ) N - 4 k ( k - 1 ) N . ( 23 ) Exchanging the structure coefficient σ for the “scaled relatedness coefficient” κ of inclusive fitness theory via the identity κ = ( σ − 1 ) / ( σ+1 ) [65] , we obtain [16] κ = N - 2 k k ( N - 2 ) . ( 24 ) With such a value , recent results on multiplayer discrete games in structured populations under δ-weak selection [43] can be readily applied to show that , for all cooperation games as we defined them and for a death-Birth protocol , A is favored over B more easily for a graph-structured population than for a well-mixed population , as long as N > k+1 . Such prediction qualitatively coincides with the one obtained from our analytical approximations , but does not capture our numerical results for the volunteer’s dilemma in rings and lattices . To sum up , we have shown that even for multiplayer games on graphs , which are routinely analyzed by simulation only , some analytical insight can be generated . However , fully accounting for the complexity of evolutionary multiplayer games in graphs with high clustering remains a challenging open problem .
|
Cooperation can be defined as the act of providing fitness benefits to other individuals , often at a personal cost . When interactions occur mainly with neighbors , assortment of strategies can favor cooperation but local competition can undermine it . Previous research has shown that a single coefficient can capture this trade-off when cooperative interactions take place between two players . More complicated , but also more realistic , models of cooperative interactions involving multiple players instead require several such coefficients , making it difficult to assess the effects of population structure . Here , we obtain analytical approximations for the coefficients of multiplayer games in graph-structured populations . Computer simulations show that , for particular instances of multiplayer games , these approximate coefficients predict the condition for cooperation to be promoted in random graphs well , but fail to do so in graphs with more structure , such as lattices . Our work extends and generalizes established results on the evolution of cooperation on graphs , but also highlights the importance of explicitly taking into account higher-order statistical associations in order to assess the evolutionary dynamics of cooperation in spatially structured populations .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"recreation",
"infographics",
"taxonomy",
"applied",
"mathematics",
"population",
"genetics",
"social",
"sciences",
"data",
"management",
"mathematics",
"population",
"biology",
"computer",
"and",
"information",
"sciences",
"games",
"behavior",
"social",
"psychology",
"evolutionary",
"systematics",
"approximation",
"methods",
"game",
"theory",
"psychology",
"computer",
"modeling",
"data",
"visualization",
"natural",
"selection",
"graphs",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"evolutionary",
"biology",
"evolutionary",
"processes"
] |
2016
|
Evolutionary Games of Multiplayer Cooperation on Graphs
|
The evolutionary success of primate lentiviruses reflects their high capacity to mutate and adapt to new host species , immune responses within individual hosts , and , in recent years , antiviral drugs . APOBEC3G ( A3G ) and APOBEC3F ( A3F ) are host cell DNA-editing enzymes that induce extensive HIV-1 mutation that severely attenuates viral replication . The HIV-1 virion infectivity factor ( Vif ) , expressed in vivo , counteracts the antiviral activity of A3G and A3F by inducing their degradation . Other APOBECs may contribute more to viral diversity by inducing less extensive mutations allowing viral replication to persist . Here we show that in APOBEC3C ( A3C ) -expressing cells infected with the patient-derived HIV-1 molecular clones 210WW , 210WM , 210MW , and 210MM , and the lab-adapted molecular clone LAI , viral G-to-A mutations were detected in the presence of Vif expression . Mutations occurred primarily in the GA context and were relatively infrequent , thereby allowing for spreading infection . The mutations were absent in cells lacking A3C but were induced after transient expression of A3C in the infected target cell . Inhibiting endogenous A3C by RNA interference in Magi cells prevented the viral mutations . Thus , A3C is necessary and sufficient for G-to-A mutations in some HIV-1 strains . A3C-induced mutations occur at levels that allow replication to persist and may therefore contribute to viral diversity . Developing drugs that inhibit A3C may be a novel strategy for delaying viral escape from immune or antiretroviral inhibition .
The evolutionary success of primate lentiviruses is evident from their prevalence in Old-World primates and their capacity to spread to new host species , frequently leading to the emergence of zoonotic disease [1 , 2] . Establishing persistent infection in individual hosts requires high mutation rates and rapid and extensive viral adaptation , which allows the virus to escape from humoral and cell-mediated immune responses [3 , 4] . Rapid viral adaptation also produces drug resistance that limits the effectiveness of therapy in many patients . Thus , understanding lentiviral genetic variation is crucial for HIV therapy . An important mechanism of HIV genetic variation is G-to-A mutation during reverse transcription [5 , 6] . Such mutations can be mediated by a family of DNA-editing enzymes with a strong preference for specific dinucleotide contexts [7–10] . For instance , APOBEC3G ( A3G ) induces high frequency of GG-to-AG mutations [11–15] whereas APOBEC3B ( A3B ) and APOBEC3F ( A3F ) cause GA-to-AA mutations [7 , 11 , 12] . In contrast , APOBEC3C ( A3C ) acts on both GA and GG dinucleotides , with a preference for GA over GG [7 , 12 , 16] . G-to-A mutations have been detected in at least 43% of HIV-1-infected patients , indicating that such mutations occur in a setting of persistent replication [17] . A3G- and A3F-induced mutations are suppressed by HIV-1 virion infectivity factor ( Vif ) , which is typically expressed in vivo , thus limiting their contribution to the adaptation of viral populations [15 , 18–22] . In contrast , A3B and A3C are relatively resistant to the effects of Vif [7 , 16 , 23] , which suggests they may play a role in HIV diversity . However , unlike A3C [24] , A3B is not expressed in the lymphoid cells that serve as targets for HIV-1 infection , which limits its potential role in the evolution of wild-type HIV-1 [9 , 23] . We hypothesized that G-to-A mutation in vif-expressing viruses is caused by an APOBEC that does not have a strong antiviral activity , is relatively resistant to Vif action , and is expressed in HIV-1 target cells . To contribute to the adaptation of viral populations , the APOBEC activity should be weak/moderate , such that mutations are sufficiently infrequent to allow some mutant progeny to survive . APOBEC3C ( A3C ) is a candidate because it is expressed in cells targeted by HIV-1 , including PBMC , macrophages , and thymocytes [24] . A3C can cause mutations in the GA and GG contexts [7 , 12] . A3C has diverged by at least 40% from the A3G , A3F , and A3B , which are known to have antiviral activity against HIV-1 [7 , 9 , 24 , 25] . Furthermore , a recent report showed that APOBEC3C exerts potent antiviral activity against simian immuno-deficiency virus , but not to a lab-adapted clone of HIV-1 [16] . Low-level activity of DNA-editing enzymes would allow some HIV-1 progeny to be viable enough to be selected by immune responses or antiviral drugs . In this study , we assessed the influence of A3C on G-to-A mutation and HIV-1 replication using multiple infectious viruses differing in Gag and/or Pro . The first group of isogenic vif-expressing NL4–3 molecular clones ( 210WW , 210WM , 210MM , 210MW ) contain combinations of pre- ( W ) and post-therapy ( M ) gag and pro genes from an HIV-1-infected patient who rapidly developed resistance to a protease inhibitor–containing regimen [26] . A second group of drug-resistant viruses includes three NL4–3-based molecular clones with point mutations introduced in the protease and reverse transcriptase ( RT ) genes . NL4–3 and LAI were also included as controls .
To determine whether A3C is responsible for inducing G-to-A mutation in HIV-1 , we inhibited A3C expression in HIV-1-infected Magi cells and examined the viral sequences . Magi cells , which normally express A3C ( Figure 1 ) , were transfected with siRNA 1 targeting A3C RNA at position 167–185 relative to the start codon . A FITC-conjugated RNA oligo was cotransfected to mark the transfected cells . FITC-positive cells were sorted 48 h after transfection . We found that no A3C mRNA was detectable in FITC-positive cells transfected with A3C-specific siRNA ( Figure 2A , lane 3 ) after 48 h in culture . By contrast , cells transfected with a control , scrambled RNA , had normal A3C mRNA levels . The cells were then infected in a single-round replication assay with NL4–3 and 210WW viruses produced by 293T transfection ( both virus stocks were p24-normalized , delta env and VSV-G-pseudotyped ) . After an additional 24 h in culture , cellular DNA and RNA were extracted . A3C mRNA was undetectable by RT-PCR ( unpublished data ) . The presence of G-to-A mutations in NL4–3 and 210WW was then examined using a sensitive mutation assay adapted from Janini et al . , 2000 [17] and designed for detecting G-to-A viral mutants over a more abundant wild-type background . This assay sensitively detects G-to-A mutants in the viral population and measures the relative frequency of mutation in the GA versus GG contexts ( see Materials and Methods for a detailed description ) . We found no G-to-A mutation in virus grown in Magi cells transfected with the A3C siRNA 1 ( Figure 2B ) . By contrast , G-to-A mutations were detected in a parallel experiment where Magi cells were transfected with scrambled RNA ( Figure 2B ) , indicating that A3C can induce G-to-A mutations in 210WW . The dinucleotide context of G-to-A mutation also correlates with the activity of A3C since both GA and GG contexts were modified with GA relatively more preferred than GG ( 17 GA over five GG , Figure 2B ) . Surprisingly , no G-to-A mutations were detected in NL4–3 infections of Magi cells transfected with the siRNA 1 or with the scrambled RNA ( unpublished data ) . To confirm that the siRNA 1 was specific to A3C , we examined the expression profile of other APOBEC genes . We found that the Magi cell line did not express mRNA for A3G or A3F , either before or after siRNA treatment ( Figure 2A , lanes 5–8 ) . Magi cells do express A3B [27] , but A3B mRNA levels were not altered by the A3C siRNA treatment ( Figure 2A , lanes 1 and 2 ) as expected because there were nine nucleotide mismatches between the A3C siRNA 1 and A3B mRNA sequences . To ensure that the suppression of G-to-A mutations in siRNA 1–transfected cells is not due to an off-target effect of the siRNA , we repeated these experiments with a second siRNA ( siRNA 2 ) , which targets residues 107–125 , and causes a partial knockdown of A3C mRNA expression ( Figure S3 ) . When transfected into 293T cells , endogenous A3C mRNA levels were reduced by 74% and the frequency of G-to-A mutations was correspondingly reduced by 70% compared to the control scrambled RNA . siRNA 2 was specific for A3C mRNA; no reduction in A3B , A3F , and A3G expression was observed compared to the scrambled control . Thus , specific inhibition of A3C expression by siRNA eliminated 210WW G-to-A mutations after infection of Magi ( Figure 2B ) and 293T cells . This result suggests that A3C expression is required for the induction of G-to-A mutations in 210WW . To further confirm that A3C in the target cells was responsible for the G-to-A mutations , the A3C gene was transfected into SupT1 cells , which do not express A3C mRNA ( Figure 1 ) . A3C cDNA was cloned into an expression plasmid that expresses the bicistronic A3C-IRES-GFP under the R PGK promoter and carries the influenza hemaggluttinin ( HA ) epitope tag ( see Materials and Methods ) . A3C expression in 293T-transfected cells was confirmed by western blotting with a polyclonal antibody specific for HA , displaying the predicted electrophoretic mobility of 23 kDa ( Figure 3A ) . The A3C expression plasmid was then transfected into human SupT1 cells and the level of protein expression was assessed by quantifying GFP by FACS ( Figure 3B ) . After 48 h , the cells were infected with p24 normalized delta env VSV-G pseudotyped NL4–3 and 210WW viruses produced by 293T cells . After an additional 24 h , DNA was extracted , amplified using the sensitive mutation assay for detecting G-to-A mutation in the viral population [17] , and sequenced . In SupT1 cells transfected with A3C , G-to-A mutation was predominately in the GA context with 210WW showing significantly higher levels of mutation in comparison to NL4–3 ( Figure 3C ) . No mutation was detected following infection with 210WW or NL4–3 in SupT1 cells transfected with the empty vector ( Figure 3C ) . Therefore , A3C is sufficient to induce mutations in the patient-derived infectious molecular clone of HIV-1 . Because 210WW and NL4–3 have different mutation frequencies in the presence of A3C in a setting of single-round infection , we wanted to assess the influence of A3C-induced mutation on the viral replication kinetics in a spreading infection . 293T-derived viruses were normalized by p24 and used to inoculate PBMC , CEMSS , and SupT1 cells , three cell types that have different APOBEC expression profiles . Consistent with previous reports , we found that PBMC expressed A3C , A3G , and A3F [9 , 16] , and that SupT1 was negative for all APOBEC RNA analyzed [27] ( Figure 1 ) . In contrast to a previous study [7] , we found that CEMSS expressed A3C only ( Figure 1 ) . Viral replication kinetics were monitored by measuring HIV-1 p24 concentration in the culture supernatant . We found that 210WW and NL4–3 had comparable growth kinetics in all three cell types ( Figure 4A ) . The presence and sequence context of G-to-A mutations in NL4–3 and 210WW was next examined in the three cell types at day 5 post infection using the same sensitive mutation assay as above . Frequent G-to-A mutations were found in 210WW sequences extracted from PBMC and CEMSS cells ( A3C-positive , Figure 4B ) . Consistent with A3C activity , they were detected both at GA and GG contexts with GA preferred to GG [7 , 12] . In contrast , no G-to-A mutation was detected in SupT1 cells ( Figure 4B ) ; although the viruses had achieved comparable levels of replication as in the other cells ( Figure 4A ) . NL4–3 sequences showed little to no mutation in any of the three cell types ( Figure 4B ) . Therefore , A3C does not seem to reduce the replication kinetics during spreading infection of 210WW in comparison to NL4–3 despite their difference in A3C susceptibility . To understand why A3C-induced mutations did not reduce the replication kinetics of 210WW we assessed the number of stop codons present in a region of RT using a quantitative assay based on blue and white β-galactosidase complementation ( see Materials and Methods ) . In this assay , the wild-type RT sequence gives rise to a blue colony whereas a mutated sequence containing a stop codon will give rise to a white colony . At day 5 post infection by 210WW , cellular DNA from PBMC , CEMSS , and SupT1 were amplified using primers that cannot distinguish between mutated and non-mutated sequences . The PCR fragments , containing six tryptophan codons susceptible to become stop codons if G-to-A mutations occur , were inserted in frame with the β-galactosidase gene . After validating the assay ( see Materials and Methods ) , we counted the number of white and blue colonies obtained from the different infections . In SupT1 cells infected with 210WW , four out of 1 , 219 clones were white ( 0 . 3% ) . This rate is equivalent to the background of the assay . Upon sequencing , the few white clones that were found contained insertion-deletion mutations leading to frame-shifts , rather than G-to-A mutations leading to stop codons , consistent with the absence of A3C in SupT1 cells . In contrast , 78 out of 937 clones from PBMC were white ( 8 . 3% ) , and 61 out of 1 , 290 clones from CEMSS were white ( 4 . 7% ) . The low frequencies of lethal mutations ( 4%– 8% ) induced in A3C-expressing cells could explain the comparable kinetics observed between 210WW , which is susceptible to A3C , and NL4–3 , which is relatively resistant . The low frequency of lethal mutation suggests that the overall rate of G-to-A mutation within the viral population should also be low in cells expressing A3C . In order to quantify the overall mutation rate , a clonal analysis of a 230-nt region of pro in proviral DNA of 210WW and NL4–3 infection from PBMC was performed . Neutral primers that do not distinguish between mutated and non-mutated sequences were used to amplify 210WW and NL4–3 DNA from PBMC that had been infected 5 d earlier . PBMC infection with NL4–3 Δvif virus served as a positive control for G-to-A mutation . Of 45 210WW clones , 33 were wild type and 12 were mutated ( 27% ) , with only one to two G-to-A mutations each in the region analyzed . Of 47 NL4–3 clones , 43 were wild type and four were mutated ( 8% ) . Of 41 clones of Δvif virus , four were wild type and 37 were mutated ( 90% ) . We then determined the frequency of mutations leading to lethal premature stop codons: four of 45 ( 9% ) contained premature stop codons consistent with the result of the blue and white β-galactosidase complementation assay and the non-attenuated replication kinetics of 210WW ( Figure 4A ) . Two of 47 ( 4% ) NL4–3 clones had lethal stop codons , and 28 of 41 ( 68% ) Δvif clones had lethal stop codons in the pro gene . To investigate whether G-to-A mutation could enhance the capacity of patient-derived viruses to adapt to antiviral drugs , we screened mutated clones for drug-resistance mutations . We looked for pro D30N , a mutation in the protease gene that occurs during nelfinavir treatment [28] and is caused by G-to-A mutation in the GA context . Such a modification was found in two of 45 ( 4% ) 210WW clones in the absence of drug selection but in none of the NL4–3 or Δvif clones . These results suggest that limited G-to-A mutation could be deleterious for some viral progeny , but surviving progeny may have greater genetic diversity that could allow adaptation to antiviral drugs . To determine if A3C-induced G-to-A mutation occurs in viruses other than 210WW , and to correlate the susceptibility for G-to-A mutation to Gag and/or Pro , we examined G-to-A mutation in a number of viruses . The first group is the 210 virus family ( 210WW , 210MW , 210WM , and 210MM ) that consists of reconstructed HIV-1 molecular clones derived from combinations of pre- and post-therapy gag and pro genes from a patient who had developed resistance to ritonavir within 4 wk of treatment and that were cloned into a NL4–3 genetic background [26] . These molecular clones demonstrate substantially different phenotypes with respect to drug susceptibility , replication capacity , and Gag cleavage characteristics [26 , 29 , 30] . The difference in replication capacity between 210WW and 210WM is 5-fold . The chimeric viruses ( MW and WM ) have intermediate phenotypes representing a 2 . 5-fold difference in replication capacity with respect to 210WW [26] . There are 11 amino acid differences between the wild-type Gag in 210WW and the mutant Gag in 210MM as well as two amino differences between wild type and mutant protease ( Figure S2 ) . The second group consists of RT- and protease inhibitor–resistant viruses constructed by site-directed mutagenesis of NL4–3 ( Protease Mutant 1 , 2 , and JF4A ) . NL4–3 and LAI were also used . We repeated the experiments in Figure 4B using an expanded number of viruses from 293T transfections and substituting H9 cells for PBMC ( also A3C- , A3G- , and A3F-positive , Figure 1 ) . Population sequencing followed by the dinucleotide context analysis of G-to-A mutations was performed on cellular DNA at day 7 post infection using the G-to-A sensitive detection assay . Preferential ( approximately 15-fold ) G-to-A mutation in the GA as opposed to the GG context was seen in 210WW , 210WM , 210MW , and 210MM ( Figure 4C ) . Susceptibility to G-to-A mutation was not restricted to the 210 viruses: LAI showed an intermediate to low level of mutation in H9 and CEMSS . G-to-A mutations in LAI also occurred predominantly in the GA context . The 210 viruses showed significantly higher levels of mutation in comparison to the other NL4–3-based and LAI viruses tested in H9 and CEMSS cells ( Figure 4C ) . No mutations were detected in SupT1 cell infections with any of the viruses tested . To determine whether A3C from the producer cell , target cell , or both induces G-to-A mutations , we measured mutations in A3C-positive and A3C-negative target cells from viruses derived from A3C-positive 293T cells directly , or after propagation for 7 d in A3C-negative cells ( SupT1 ) . Viral infection titers were normalized by supernatant Gag p24 . Infections were carried out and G-to-A mutations were determined as described in Figure 4B . We found no G-to-A mutations in non-expressing A3C target cells infected with 210WW regardless of A3C expression in the producer cells ( Table 1 ) . However , G-to-A mutations were induced in target cells expressing A3C . The result suggests that only A3C in the target cell is active .
In this study , we asked whether A3C has a role in HIV-1 G-to-A mutations . We found that transient A3C expression was sufficient to cause mutation of a patient-derived infectious HIV-1 molecular clone in SupT1 cells . Reciprocally , specific abrogation of A3C expression prevented G-to-A mutation in Magi and 293T cells , demonstrating that A3C is required for mutation in these cell lines . Furthermore , G-to-A mutation is detected in CEMSS cells , which we found express A3C but not A3B , A3F , or A3G . A role for A3C in PBMC is suggested by the activity of the enzyme that can act on both GA and GG contexts: Both mutational patterns are widely observed in vivo with G-to-A mutation in the GA context more predominant than in the GG context [5 , 6 , 17] . The preferential activity of A3C on GA context reported by previous studies [7 , 12 , 16] is consistent with our findings . Indeed , in CEMSS cells that endogenously express only A3C , G-to-A mutation was predominately detected in the GA context ( Figure 4B ) . These data are consistent with the notion that A3C is necessary and sufficient to induce G-to-A mutation in HIV-1 . G-to-A mutations occurring at low rates in a high background of wild-type viruses were detected in this study using a sensitive G-to-A detection assay adapted from Janini et al , 2000 [17] . Although this approach is useful to sensitively detect G-to-A mutations in the viral population , and to assess the relative frequency of mutation in the GA versus GG context , it is not quantitative . To confirm and quantify the rate of mutation detected in A3C-expressing cells we used the blue and white complementation assay and clonal analysis of sequences . The blue and white complementation assay and clonal analysis of gene segments amplified using non-selective primers indicated that the majority of viral sequences was not altered after exposure to A3C: 27% of the 210WW clones harbor only one to two G-to-A mutations . We also found that mutated sequences containing premature stop codons induced by G-to-A mutation reflect less than 9% of the 210WW clones . A low rate of G-to-A mutation likely explains the negligible antiviral effect of A3C on the 210 and LAI viruses . Other studies reported that A3C did not exhibit strong antiviral activity against HIV-1 NL4–3 replication , which is consistent with our own observations [7 , 16] . The minimal antiviral effect of A3C reflects the low frequencies of mutation documented in blue and white complementation assay and clonal analysis of virus populations . Limited G-to-A mutation associated with A3C may be deleterious for some viral progeny , but surviving progeny would have greater genetic diversity . Protease D30N mutations , which confer clinically important resistance to the protease inhibitor nelfinavir [28] , appeared in 5% of 210WW clones from PBMC infections . While other proteins expressed in PBMCs may have contributed to the D30N mutations observed , A3C is a likely inducer for the following reasons: A3C is expressed in PBMCs; D30N mutations were observed exclusively in infections with 210WW , which is susceptible to A3C , but not in infections with NL4–3 or NL4–3 Δvif , which are susceptible to other deaminases; and all three viruses have the same reverse transcriptase , suggesting that reverse transcriptase errors could not account for the differential rates of mutation observed in the different infections . The action of A3C was observed primarily in infectious molecular clones of HIV-1 , derived from a patient who rapidly developed drug resistance and virological drug failure during therapy with a protease inhibitor [26] . G-to-A mutation associated with A3C expression was also observed in the molecular clone LAI , which may suggest that A3C action is not restricted to the 210 viruses . Susceptibility to A3C-mediated mutation was higher in the 210 family of viruses compared to the NL4–3 family of viruses . The mechanism for this difference in susceptibility to A3C is not known , although the 210 family of viruses differs from NL4–3 in Gag and Pro ( the DNA alignment as well as the amino acid alignment of 210WW , LAI , and NL4–3 is provided in Figures S1 and S2 ) . The patient-derived gag segment in 210 is 6% divergent from NL4–3 overall . Differences in Gag may alter the stability of the viral core , or the pre-integration complex , which would affect how long nascent single-stranded cDNA is available to bind with A3C in the target cell . More work is needed to identify the viral determinants of A3C susceptibility , and to seek clinical correlations of A3C susceptibility . The action of A3C differs from that of A3G , A3F , and A3B in several ways . A3G appears to act both in the virion and the target cell: target cell action appears to involve impaired reverse transcription rather than cytidine deamination [31] . A3G , A3F , and A3B cause G-to-A hypermutation when incorporated in the budding virion , producing lethal hypermutation early in the synthesis of viral cDNA that frequently occurs in the virion . In contrast , A3C seems to act only in the target cell based on the following observations: G-to-A mutation was not observed in SupT1 cells ( using the sensitive and the blue and white assays ) that were infected with viruses produced in 293T cells , which express A3C ( Figure 1 ) , while viruses propagated in SupT1 acquire mutation only when they infect A3C-expressing cells ( Table 1 ) . Target cell action was further demonstrated by transient expression or transient knockdown of A3C in target cells , which determined the appearance of G-to-A mutation in single-cycle infections . The exposure of the viral cDNA to A3C in the target cell could be limited , which may decrease the frequency of viral mutation to sublethal levels , allowing surviving progeny to have greater genetic diversity to adapt and escape from immune responses or antiviral drugs . HIV diversity is the major obstacle for effective treatment of HIV patients . It has been suggested that natural genetic variations in A3G/A3F and/or vif may result in an incomplete neutralization of those cytidine deaminases allowing the remaining enzymes to exert a low level of mutation on the HIV-1 genome and potentially inducing viral diversity [32 , 33 , 34] . Although A3G mutants have been recovered from different individuals , there is no functional evidence for a defect in the antiviral activity of those mutants and in their ability to induce massive hypermutation resulting in defective viruses [33 , 34] . Similarly , vif variants that partially fail to neutralize A3G have been identified; however , the recovered proviral sequences were highly hypermutated and replication-defective [32] . We find evidence of a low level of cytosine deaminase activity in vif-expressing viruses , although some Vif-A3C interactions have been observed [7 , 9 , 16] . A3C activity may be relatively resistant to Vif [7 , 19 , 35 , 36] because A3C has a tyrosine instead of the aspartic acid at position 128 , which is critical for the interaction with and sensitivity to Vif . In conclusion , we demonstrate here that A3C is necessary and sufficient for G-to-A mutations in some HIV-1 strains . A3C-induced mutations occur at levels that allow replication to persist and may therefore play a role in driving viral diversity .
The cell lines H9 , CEMSS , SupT1 , Magi , and 293T were directly obtained from the National Institutes of Health AIDS Research and Reference Reagent Program . Cell line identity was confirmed using DNA fingerprinting by PCR amplification with allele-specific primers ( Research Genetics ) . PBMCs from leukocyte-enriched fractions of whole blood from HIV-1-seronegative donors were isolated by density-gradient centrifugation ( Histopaque-1077 , Sigma ) . Cells were stimulated with PHA-P ( 5 μg/ml , Sigma ) for 24 h , washed , and maintained at 2 × 106/ml in RPMI-1640 ( Irvine Scientific ) supplemented with L-glutamine ( 2 mM ) , 20% FBS ( Gemini Bioproducts ) , and IL-2 ( 10 U/ml purified human , Roche Diagnostics ) before and after infection . pNL4–3 , obtained from the NIH AIDS Research and Reference Reagent Program ( Catalog Number: 114 ) , is a laboratory wild-type HIV-1 molecular clone . pM46I/L63P/V82T/I84V ( Catalog Number: 4595 ) , and pL10R/M46I/L63P/V82T/I84V ( Catalog Number: 4596 ) are NL4–3-based infectious molecular clones with protease mutations introduced in vitro . pJF4A ( Catalog Number: 1412 ) is a full-length infectious molecular clone that is identical to pNL4–3 except for two nucleotide substitutions introduced by site-directed mutagenesis at nt 2754 ( A-G ) and 2755 ( C-A ) that render it resistant to ddC . pLAI . 2 ( Catalog Number: 2532 ) is a laboratory wild-type molecular clone . The 210 viruses are isogenic infectious molecular clones having the NL4–3 background and combinations of pre-therapy and post-therapy gag and pro gene segments from a patient who had developed resistance mutations to ritonavir 4 wk after treatment [26] . The 210 constructs differ from NL4–3 in a portion of the 5′ untranslated region upstream of gag , gag , and pro ( nt 702-2563 relative to NL4–3; see Figure S1 ) . 210WW has pre-therapy gag and pro , while 210MM has drug-adapted gag and pro , which contains pro mutations I54V and V82A . 210WM has the pre-therapy gag with drug-adapted pro , and 210MW has drug-adapted gag with pre-therapy pro . NL4–3 Δenv and 210WW Δenv viruses were generated by digesting each plasmid with Nhe1 restriction enzyme ( Invitrogen ) . The DNA polymerase I Klenow fragment ( New England Biolabs ) was used to fill in the ends and the plasmids were religated using the T4 DNA ligase ( Promega ) . Pseudotyped VSV-G ΔEnv NL4–3 and 210WW viruses were produced by cotransfecting 293T with the pMD . G plasmid expressing the VSV-G envelope protein [37] and the plasmid expressing ΔEnv HIV-1 viruses , respectively . NL43 Δvif is a NL4–3 variant with three stop codons introduced into vif by directed mutagenesis ( C . de Noronha , unpublished data ) . Virus stocks were prepared by Lipofectamine-mediated transfection of 293T cells . The clarified viral supernatants were quantified and normalized by HIV-1 Gag p24 ELISA ( Perkin Elmer ) . DNase treatment of viral supernatant was carried out at 37 °C for 30 min with the equivalent of 10 units DNase ( Sigma ) /1 ml of the clarified crude virus . Spreading infections were monitored over time by accumulation of p24 Gag in the culture supernatants . Single-cycle infectivity was performed by challenging the cell lines ( 5 × 105 cells ) with VSV-G Δenv pseudotyped viruses for 24 h . PolyA+ RNA was isolated from cells with the QuickPrep Micro mRNA Purification Kit ( Pharmacia ) . The open reading frames of A3B , A3C , A3F , and A3G were amplified by RT-PCR with the following primers: 3B forward 5′-ATGAATCCACAGATCAGAAATCC-3′ and reverse 5′-TCAGTTTCCCTGATTCTGG-3′; 3C: forward 5′-ATGAATCCACAGATCAGAAACC-3′ and reverse 5′-TCACTGGAGACTCTCCCGTA-3′; 3F: forward 5′-ATGAAGCCTCACTTCAGAAAC-3′ and reverse 5′-TCACTCGAGAATCTCCTGC-3′; 3G: forward 5′-ATGAAGCCTCACTTCAGAAACACAG-3′ and reverse 5′-TCAGTTTTCCTGATTCTGGAGAATGG-3′ . The amplicons were cloned into pGEM-T easy vectors ( Promega ) and confirmed by sequencing . The A3C gene with HA tag sequence at its 3′ terminus was amplified from mRNA of H9 cells by RT-PCR , and the identity of the product was confirmed by sequencing . It was then cloned into pTT-IRES-GFP ( a gift from David Fenard , Gladstone Institute of Virology and Immunology ) for protein expression . The 293T or Magi cells were transfected with Lipofectamine 2000 transfection reagent ( Invitrogen ) , according to the manufacturer's recommendations . The SupT1 cells were transfected by Amaxa electroporation technology with the program T29 , according to the manufacturer's recommendations . DNA was extracted from infected cells with the DNA Easy kit ( Invitrogen ) and quantified by spectrometry . PolyA + RNA was isolated with QuickPrep Micro mRNA Purification kit ( Pharmacia ) . Reverse transcription reaction was performed with random hexamers using SuperScript III reverse transcriptase ( Invitrogen ) . An RT minus control was always performed in parallel to check for DNA contamination . PCR was performed with Easy-A High Fidelity PCR Cloning Enzyme ( Stratagene ) . APOBEC reading frames ( A3B , A3C , A3F , and A3G ) were amplified with primers specific for a single gene and listed above in the APOBEC plasmid construction section . The specificity of the primers was verified by sequencing the PCR products . The PCR products were sequenced by dye terminator cycle sequencing with Big Dye v3 . 1 ( Applied Biosystems ) or ET Terminators ( Amersham Biosciences ) on an ABI 3100 microcapillary sequencer . Raw sequence data was analyzed and aligned using Sequencher ( Gene Codes ) . This assay is designed to detect G-to-A viral mutated sequences present in a high wild-type background . A nested PCR procedure specific for a 230-nt region in the protease gene amplifies both G ( wt ) and A ( mut ) species in the GA and GG context using the HIV-1 pro primers Hypa 10 and Hypa 11 in the first PCR and DP 16 and DP 17 in the second round PCR as described [17] . The first round PCR uses primers specific for the mutated sequences ( Hypa 10 and Hypa 11 ) . The goal is to enrich the mutated sequences that are usually present in low concentration . The second round PCR uses primers that can amplify both mutated ( if present from the first round PCR ) and the input wild-type DNA ( carried over from the first PCR ) [17] . Under these amplification conditions , no PCR-induced mutations were detected using control plasmids as templates . Protease sequences were determined as population sequences . A mixture of G and A peaks was typically seen at the mutation site . A change from G-to-A was considered a true mutation only if A represented at least 20% of the mixture . The clonal analysis was performed using the neutral primers DP16–2 and DP17–2 . The sequence of DP16–2 is 5′-TATCCTTTAGCTTCCCTCA-3′ . The sequence of DP17–2 is 5′-TAATGGGAAAATTTAAAGTGCAG-3′ . PCR products were cloned in pCR 4-TOPO ( Invitrogen ) . The plasmids were then sequenced and analyzed for their G-to-A content . The genetic assay was carried out with a 111-nucleotide viral DNA containing a HIV-1 pol fragment that includes six tryptophan codons susceptible to become stop codons if G-to-A mutations occur . Viral DNA was amplified using the primers Nsi-Trp6 Fw: 5′-CCA ATGCATACTCCTAAAT TTAAATTACC CATACAAAA-3′ and Trp6 Rev Spe: 5′-GACTAGTTAAGGGAGGG GTATTGA-3′ . The amplified DNA was phenol extracted , precipitated with ethanol , subjected to treatment with Nsi1 and SpeI , and then purified from a 1% agarose gel . This fragment encoding six Trp codons was cloned in frame into the lacZ fragment of the pGEM-T vector . For this purpose pGEM-T was digested first with Nsi1 and then with SpeI to obtain the linearized vector , which was purified from a 1% agarose gel and treated with alkaline phosphatase before ligation to the PCR products . Escherichia coli XL-1 Blue was transformed and plated onto minimal agar plates containing 8% X-Gal ( 5-bromo-4-chloro-3-indolyl-d-galactopyranoside ) and 20% IPTG ( isopropyl-d-thiogalactopyranoside ) . The plates were incubated at 37 °C for approximately 15 h . We validated the assay by first mixing 0%–5%–25%–50%–75%–100% of G-to-A mutated clones with wild-type clones . We then amplified each mix , cloned the PCR product , and counted the number of blue and white colonies . We confirmed that the number of blue and whites clones obtained after every specific mixture with the level of mutated sequences included in the mix . We also determined that the background of this assay was around 0 . 3% , which corresponded to the percentage of white colonies detected using 100% wild-type plasmids . The sequences of the white clones obtained in this specific mix revealed insertions or deletions interrupting the reading frame , and not stop codons . We also sequenced 20 white and 20 blue clones obtained from the other mix to verify their mutation content and the presence of stop codons . 20/20 blue clones were wild-type sequences and 20/20 white clones were mutated sequences . Blue and white clones were counted using the EagleSight ( Stratagene ) instrument and confirmed by manual counting . A3C protein expression in 293T transfected cells was detected by western blot with an HA-probe ( Y-11 ) and HRP polyclonal antibody specific for the HA tag ( Santa Cruz Biotechnology ) . siRNAs targeting A3C from positions 167–185 ( siRNA 1 ) and 107–125 ( siRNA 2 ) bases relative to the start codon were transcribed in vitro ( RiboMax kit , Promega ) from an oligo DNA template containing the T7 promoter synthesized by Invitrogen . The sequence of the scrambled RNA for both siRNAs was generated with the Promega siRNA target designer . For the siRNA transfection , 293T and Magi cells were seeded into six-well plates at a density of 104 cells/ml . After about 2 h , purified A3C siRNAs along with RNA oligo conjugated to FITC ( Block It kit , Invitrogen ) were transfected into the cells with Lipofectamine 2000 , according to the manufacturer's protocols . A scrambled siRNA was transfected separately as a control . The efficiency of transfection was monitored by fluorescent microscopy at 24 , 48 , and 72 h after siRNA transfection . For flow cytometry analysis , the cells were analyzed with a Becton Dickinson FACS cytometer . Samples were counted and analyzed with FlowJo software ( Tree Star ) . Non-transfected cells were used as a control . FITC-positive cells were sorted ( DIVA instrument , Becton Dickinson ) for further analysis and infection .
The GenBank ( http://www . ncbi . nlm . nih . gov/Genbank/index . html ) accession numbers for the genes and proteins mentioned in the text are: A3B ( NM_004900 ) , A3C ( NM_014508 ) , A3F ( NM_001006666 ) , A3G ( NM_021822 ) , LAI gag-pro ( K02013 ) , NL4–3 gag-pro ( AF324493 ) , 210MM gag-pro ( EU100418 ) , and 210WW gag-pro ( EU100417 ) .
|
HIV has shown a chameleon-like nature , always changing to adapt to its environment . Defining the factors that drive and regulate genetic changes in HIV over time is key to understanding how HIV causes disease and escapes from the body's immune responses and drug treatment . The diversity of HIV has implications for the development of effective drugs and for fostering better immune responses . In this study , we showed that a human protein , called APOBEC3C ( A3C ) , could induce certain mutations in some HIV-1 strains , including those derived from a patient who had developed rapid drug resistance . Eliminating the expression of this protein prevented the mutations in two different types of cells . Furthermore , short-term A3C expression was sufficient to cause mutation . We conclude that A3C is necessary and sufficient to induce signature mutations in HIV-1 . A3C-induced mutations may provide potential benefit for the virus if the mutation rate is low enough such that the majority of viruses are able to replicate , while accumulating a limited number of novel mutations that may allow the virus to survive in the face of antiviral drugs or immune responses .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] |
[
"viruses",
"infectious",
"diseases",
"virology",
"evolutionary",
"biology",
"molecular",
"biology"
] |
2007
|
Target Cell APOBEC3C Can Induce Limited G-to-A Mutation in HIV-1
|
Small cell lung cancer ( SCLC ) is an aggressive disease with poor survival . A few sequencing studies performed on limited number of samples have revealed potential disease-driving genes in SCLC , however , much still remains unknown , particularly in the Asian patient population . Here we conducted whole exome sequencing ( WES ) and transcriptomic sequencing of primary tumors from 99 Chinese SCLC patients . Dysregulation of tumor suppressor genes TP53 and RB1 was observed in 82% and 62% of SCLC patients , respectively , and more than half of the SCLC patients ( 62% ) harbored TP53 and RB1 mutation and/or copy number loss . Additionally , Serine/Arginine Splicing Factor 1 ( SRSF1 ) DNA copy number gain and mRNA over-expression was strongly associated with poor survival using both discovery and validation patient cohorts . Functional studies in vitro and in vivo demonstrate that SRSF1 is important for tumorigenicity of SCLC and may play a key role in DNA repair and chemo-sensitivity . These results strongly support SRSF1 as a prognostic biomarker in SCLC and provide a rationale for personalized therapy in SCLC .
Small cell lung cancer ( SCLC ) represents 13% of all newly diagnosed cases of lung cancer worldwide with more than 180 , 000 cases per year [1] . It is an aggressive neuroendocrine malignancy with a unique natural history of a short doubling time , high growth fraction , and early development of widespread metastases [2] . Most patients are very sensitive to thoracic radiotherapy and platinum drugs such as cisplatin and carboplatin , but suffer disease recurrence or progression in a very short period of time following initial treatment [1] . Currently , for recurrent or progressive SCLC , the only drug approved in the United States and Europe is topotecan , a topoisomerase 1 ( Top1 ) inhibitor which provides some benefit , though the five year survival rate of SCLC has remained unchanged at~5% for the last four decades [2] . To improve patient outcomes in SCLC , it is critical to understand the key genetic alterations that contribute to the specific disease phenotypes and their utility for potential therapeutic targets . However , systematic genetics and genomics analyses of large cohorts of SCLC patients remains a challenge , primarily because SCLC usually presents as extensive disease upon diagnosis and hence is rarely treated surgically , thus causing a lack of suitable tumor specimens for comprehensive analysis . To date , these types of extensive genome-wide molecular analyses have been performed on relatively small patient cohorts , which provide utility restricted to the disease population sampled [3 , 4 , 5] . Within these studies , among genes recurrently affected by genomic alterations in SCLC , TP53 , RB1 , as well as the amplification of MYC family members and SOX2 have been identified . However , the molecular factors related to chemo-sensitivity or resistance remain unknown . Additionally , clinical outcome such as survival in relation to genetic alterations remains unreported , particularly in the SCLC Chinese patient population . Here , we conducted the first comprehensive genetic landscape survey of Chinese SCLC patients with whole exome sequencing ( WES ) and transcriptomic sequencing of primary tumors from 99 SCLC patients with detailed clinical history and survival data . Our study not only identified novel recurrent genetic alterations such as CDH10 and DNA repair pathways which may influence outcomes in SCLC patients , but also revealed SRSF1 , an RNA-splicing factor which can form complexes with TP53 and Top1 , and plays a critical role in SCLC patient survival .
WES of 25 normal [normal adjacent tissue ( NAT ) or blood] and matched tumor pairs , and 74 tumors only ( no normal tissue ) from Chinese SCLC patients revealed 32 , 566 somatic non-silent single nucleotide variants ( SNVs ) or insertion/deletions ( indels ) , an average of 329 per patient and non-silent/silent ratio of 2 . 11 . The patient summary is described in Table 1 and S1 Table . The most frequent transition and transversion changes were G>A and G>T , respectively , consistent with a previous report in SCLC [2] . Genes harboring the most recurrent somatic SNVs or indels were TP53 ( 82% ) , RB1 ( 47% ) , CSMD3 ( 47% ) , NOTCH1 ( 18% ) and NOTCH3 ( 15% ) ( S2 Table ) . TP53 and RB1 have been reported previously as the most recurrent genes harboring nonsilent somatic SNVs in SCLC [2 , 3 , 4] . Oncogenic gain-of-function mutations in NOTCH1 commonly occur in human T-cell acute lymphocytic leukemia ( T-ALL ) and B-cell chronic lymphocytic leukemia [6 , 7 , 8] . Loss-of-function mutations in Notch receptors have been recently reported to likely play a tumor suppressor role in lung squamous cell carcinoma and SCLC patients [9 , 10] . Additionally , the concordance between the top 100 genes harboring the most recurrent nonsilent somatic SNVs or indels in this study and a recent WES study of Asian SCLC patients ( Japanese; n = 51 ) was 62% ( S2 Table ) , with strong consistency of recurrence prevalence in TP53 ( 82% vs . 80% ) , RB1 ( 47% vs . 39% ) , and CSMD3 ( 47% vs . 37% ) , among other genes , between the two studies [5] . To further narrow down the most disease-relevant mutated genes , we first generated a list of genes harboring the most recurrent and significant nonsilent somatic mutations ( identified with two independent algorithms ) . Then this list was intersected with two independent lists of significantly mutated genes in SCLC generated by both Peifer et al [4] and Umemura et al [5] studies . Aside from TP53 and RB1 , neural cell transmembrane genes TMEM132D , NCAM2 , and CDH10 were shared in all three independent studies ( S3 Table ) . The mutation rates of TMEM132D , NCAM2 , CDH10 in our Chinese patient cohort were 14% , 13% and 12% , respectively . To evaluate the impact of these mutations in these three genes on patient outcomes , we used a Cox proportion hazard ( PH ) regression model to correlate the mutation status with survival . The patients were split into two groups: those harboring at least one nonsilent somatic mutation and those without . Among these three genes , patients with mutations in CDH10 , a cadherin which is predominantly expressed in brain [11] , displayed a significant association with poor survival , after adjusting for age , gender , tumor stage , and chemotherapy status ( p = 0 . 0127 ) . Twelve of 99 patient harbored CDH10 mutations , mostly located in the cadherin domain with high confidence protein affecting predictions ( i . e . SIFT ) ( Fig 1 ) . To better understand the genetic basis of chemo sensitivity and resistance in SCLC , we systematically surveyed SNVs and indels in all known DNA repair genes [12] . Eighty-seven percent ( 87% ) of patients harbored ≥1 nonsilent somatic SNV in a DNA repair gene besides TP53 ( S4 Table ) ; similarly , within a Japanese SCLC study cohort in a previous study , 69% of patients were identified by the same criterion [5] . The patient prevalence of nonsilent somatic SNVs in genes classified as mismatch repair ( MMR ) , nucleotide excision repair ( NER ) , homologous recombination , or DNA polymerase were 22% , 30% , 26% and 35% , respectively . Twelve percent of patients harbored nonsilent somatic SNVs in DNA polymerase genes that are involved in DNA replication in NER and MMR ( POLD1 and POLE , [13] ) . POD1 , POLG and POLQ were most recurrently mutated among the 15 DNA polymerase genes . These somatic SNVs cause protein truncations and amino acid changes in the polymerase , exonuclease , and helicase domains ( Fig 2A–2C ) . Fanconi anemia pathway genes were most recurrent with prevalence of 36% . Within this specific pathway , multiple genes involved in DNA inter-strand crosslink repair such as FANCM ( 7% ) and BRIP1/FANCJ ( 7% ) were among the most mutated ( Fig 2D ) . Finally , 29% of patients harbored nonsilent somatic SNVs in genes that affect sensitivity of mammalian cells to topoisomerase inhibitors , in addition to TP53 [14] . Somatic copy number variants ( CNVs ) were identified from exome-sequencing data . Our results confirmed key oncogenic genes with recurrent CN gains/amplifications that were previously reported in SCLC [3 , 5 , 15 , 16 , 17] , including MYC ( 8% ) , KIT ( 16% ) , and SOX2 ( 67% ) . Significant copy number gains or amplifications were observed across a cluster on chromosome 3q26-29 [5] ( S5 Table ) . Genes with CN losses previously reported in SCLC [2 , 4 , 5] include RB1 ( 34% ) , RASSF1 ( 57% ) , FHIT ( 54% ) , KIF2A ( 16% ) , and PTEN ( 13% ) . A long segment along chromosome 3p22 was also detected to have significant CN loss . Recurrence rates of these genes affected by CNVs were comparable to those reported previously [3 , 5] . In addition , we found recurrent gains of SRSF1 ( 50% ) as well as concordant over-expression of mRNA for those patients with gains ( p = 0 . 005; two-tailed two-sided Welch’s t-test; Fig 3A ) . Among these 96 Chinese patients , 28% had both CN gain and mRNA over-expression of SRSF1; in an independent cohort of 25 Caucasian SCLC patients ( commercially purchased specimens–see Methods ) , we identified 32% with the same result . Further , SRSF1 CN gain was determined to be 30% ( 8/27 SCLC patients ) in a re-analysis of the available WES data published from a previous Caucasian SCLC patient cohort–a result very similar between both Caucasian SCLC cohorts [3] . CN gains/amplifications or losses and somatic SNVs for relevant genes are summarized in S1 Fig . SRSF1 CN status was evaluated by FISH assay ( N = 34 ) . Using a FISH criterion described in the Methods for deviations from disomy [18] , the sensitivity and specificity were 47% and 71% respectively ( positive and negative predictive values of 57% and 62% , respectively ) . This is comparable to a previous study’s concordance reported between FISH and sequencing using much greater sequencing depth ( 843X ) detecting an EML4-ALK fusion in lung cancer [19] . Further , a clinical study detecting ALK fusions in lung cancer reported a positive predictive value between sequencing and FISH as 68% ( 19/28 ) among diagnostic characterized patients , and only 46% ( 6/13 ) when reduced to those patients with clinical outcomes ( 11/13 were sequencing positive and partial responders to crizotinib ) [20] . These studies support both the lack of sensitivity in FISH assays compared to sequencing for detecting variants and comparability in concordance between these two assays in this study and two previous studies , both of which were detecting a much larger genetic variant ( S6 Table; S2 Fig ) . For patients with both survival and WES data ( N = 96 ) , genes within CN gain or loss regions were correlated with survival . The cohorts were separated into a discovery set ( patients with tumors/matched normal; N = 22 ) and a validation set ( patients with tumors only; N = 74 ) . Kaplan-Meier analyses were conducted between patients with or without CN gains in the discovery cohort first ( see Methods ) . Then this gene list was reduced to those with log-rank p<0 . 05 in the validation cohort . For the remaining genes , patients with both RNASeq and survival data were interrogated ( N = 48 ) and SRSF1 was the only gene that correlated between both CN gain and mRNA over-expression at a p<0 . 05 ( log-rank p = 0 . 008; Fig 3B ) as well as between over-expression and survival using a Cox proportion hazard ( PH ) regression model adjusting for age , gender , tumor stage , and chemotherapy status ( p = 0 . 047; HR = 2 . 7; Fig 3C ) . Patients with SRSF1 mRNA over-expression or CN gain demonstrated significantly worse survival . The discovery ( log-rank test p = 0 . 062 ) , validation ( log-rank test p = 0 . 03 ) , and combined patient cohort ( Cox PH p = 0 . 012; HR = 2 . 1; log-rank test p = 0 . 005 ) analyses are provided in Fig 3D–3F and S7 Table CN gains in SRSF1 from The Cancer Genome Atlas ( TCGA ) were interrogated for correlation with survival ( S3 Fig to evaluate the specificity of SRSF1 CN gains associating with survival in other cancer indications . We used a threshold of at least 3 patients for a particular cancer indication harboring a CN gain in SRSF1 to minimize biases in sample groups for survival analysis . Among cancer indications in TCGA with ≥3 patients harboring CN gains in SRSF1 ( BRCA , KIRP , SARC , SKCM , and UCEC ) , uterine corpus endometrial carcinoma ( UCEC ) was the only indication with a correlation between SRSF1 CN gain and poor survival ( log-rank test p = 0 . 003 ) , though the patient number with a CN gain group was highly unbalanced compared to those without ( n = 8 vs . n = 437 , respectively ) , likely driving the low p-value . This result demonstrates how this CN gain in SRSF1 is specific to SCLC . We next evaluated SRSF1 as a potential tumor driver in SCLC . We first screened SRSF1 DNA CNs in 13 SCLC cell lines using TaqMan assays . Five of thirteen had SRSF1 CN> = 3: Four including NCI-H82 had 3 copies , and DMS114 had 4 copies . These cell lines also expressed high levels of SRSF1 protein ( S4 Fig ) . SRSF1 siRNA was transfected into DMS114 , and the growth effect of SRSF1 ablation in two dimensional cell culture either alone or in conjunction with a sub-lethal dose of cisplatin or topotecan ( two of the most common standard of care treatments in SCLC ) , was evaluated ( Fig 4A ) . SRSF1 knockdown alone caused a 35% decrease in the proliferation rate . Treatment with a low dose of cisplatin or topotecan only induced a modest decrease of cell growth . However , combination with SRSF1 siRNA significantly enhanced the overall growth inhibition effect . SRSF1 has also been shown to regulate the BCL2 pathway by alternative splicing of BIM , which results in a protein lacking pro-apoptotic activity [21 , 22] . In this study , we see that SRSF1 gene expression is positively correlated with BIM ( r = 0 . 58 , p<0 . 0001 ) and SRSF1 CN gain or amplification also shows concordantly high expression of BIM ( S5 Fig ) . Furthermore , we performed caspase-3/7 assays on similarly treated cells ( Fig 4B ) to evaluate the synergistic effect between SRSF1 knockdown and standard chemotherapy . SRSF1 siRNA alone induced modest but statistically significant caspase-3 activation , similar to cisplatin treatment alone . The combination of the two produced a substantially higher caspase induction . A similar trend was revealed with topotecan . Comparable results were also obtained in other SCLC models ( S4 Fig ) . The effect of SRSF1 knockdown on SCLC cells when grown as 3D spheroids was evaluated next . Cells transfected with non-targeting siRNA produced large and well-organized spheroids; in contrast , cells transfected with SRSF1 siRNA did not form well-organized structures but mainly existed as single cells with poor viability ( Fig 4C and S6B Fig ) . Results were confirmed by colony formation assays ( S6D Fig ) . The effect of SRSF1 siRNA is mediated by specific target loss as demonstrated by a reconstitution study with a siRNA-resistant Flag-tagged expression construct which efficiently rescued the spheroid growth in the presence of the SRSF1 siRNA ( Fig 4D ) . A similar rescue effect was also achieved in NCI-H82 cells ( S6C Fig ) . A tumor formation study was conducted using siRNA-transfected DMS114 and SHP-77 cells . Equal numbers of viable transfected cells were injected in immunocompromised mice and tumor growth was monitored for up to three weeks . SRSF1 knockdown completely suppressed the tumor growth in both SCLC models ( Fig 4E and S7A Fig ) . DNA-damage induction as a potential effect of SRSF1 knockdown based on our DNA-repair analysis was assessed . Inductions of p-H2AX and Chk2 , established markers of DNA-strand breaks and DNA-repair response [23 , 24] , were consistently observed upon SRSF1 abrogation in DMS114 and SHP-77 ( Fig 5A and S7B Fig ) , and increased phosphorylations were observed when we combined SRSF1 siRNA transfection and treatment with cisplatin or topotecan . To better understand the role of SRSF1 CN gain on downstream pathways in SCLC , we performed differential gene expression analysis between SRSF1 CN gain and SRSF1 CN neutral patients . A total of 861 genes were identified to be significantly expressed between these patient cohorts . Pathway analysis revealed that PIK3CA and MAPK3 were two of the top activated master regulators , which suggests that SRSF1 CN gain regulates PI3K/Akt and MAPK pathway activity with certain causality ( S12 Table ) . Therefore , we investigated the impact of SRSF1 loss on both PI3K/Akt and Ras/Raf MAPK kinase signaling pathways in SCLC cells through phospho-kinase array profiling ( Fig 5B ) . Control siRNA-transfected DMS114 displayed strong phospho-AKT and ERK signals , which were abrogated by SRSF1 siRNA . Western blot confirmed this in both DMS114 and NCI-H1048 cells ( Fig 5C ) . This demonstrated that SRSF1 promotes SCLC growth and survival by sustaining PI3K/AKT and MEK/ERK pathways , two of the most well-established oncogenic pathways .
Our study represents the first comprehensive genetic landscape survey of Chinese SCLC patients with detailed clinical history , revealing key recurrent genetic alterations associated with patients’ outcomes . Mutations identified in previous SCLC genomic studies shared little consensus for significantly mutated genes other than TP53 and RB1 . However , by leveraging our data with these previous SCLC studies , we were able to identify three additional common significantly mutated genes ( TMEM132D , NCAM2 , and CDH10 ) with over 10% prevalence in SCLC . Interestingly , all three genes encode transmembrane proteins involved in neural cell adhesion . This finding will need to be further evaluated for the impact on neuroendocrine association in SCLC . Cadherins ( CDHs ) are important in maintenance of cell adhesion and polarity , alterations of which contribute to tumorigenesis . Recurrent mutations in CDH10 have recently been reported in EGFR/KRAS/ALK mutation-negative lung adenocarcinoma in never-smokers [25] and as a prognostic mutation signature in colorectal cancer [26] . Our study indicated that CDH10 is not only the most commonly and significantly mutated gene in SCLC but also associated with poor survival in SCLC . CDH1/E-cadherin , the founding member of the CDH/cadherin family , undergoes loss-of-function mutations across multiple tumor types such as breast , gastric , colorectal and ovarian cancer . Its functional inactivation contributes to cancer progression by increasing cell invasion , migration , metastasis and proliferation and EMT process [27] . We speculate that the recurrent CDH10 mutations we detected in SCLC may perform similar roles as CDH1 mutations in other cancers to promote SCLC aggressiveness , leading to poor patient survival . We are currently conducting experiments to test this hypothesis . Our study suggests that genetic alteration of DNA repair pathways influence chemotherapy outcomes in SCLC patients . The Fanconi anemia ( FA ) pathway is essential for the repair of DNA inter-strand cross-linking agents , such as cisplatin , which has been used as first-line treatment in SCLC . It was demonstrated several decades ago that the FA patient-derived cells which contain genetic defects in FA genes display hypersensitivity to DNA cross-linking agents [28] . Our data strongly suggest that high prevalence mutations in FA pathway genes may contribute to initial hypersensitivity of SCLC to platinum-based treatment such as cisplatin . Multiple reports with experimental evidence show that the efficacy of various chemotherapeutic agents , including cisplatin , requires a functional TP53 protein for efficient induction of apoptosis and that loss of TP53 function enhances resistance to cytotoxic agents used in cancer therapy [29 , 30 , 31] . Further , a combination of TP53 inactivation and MMR deficiency has also been observed to confer cisplatin resistance [32] . Our data suggest that high frequency mutations in TP53 combined with other DNA repair mutations such as mismatch repair , nucleotide excision repair , homologous recombination , and key DNA polymerases may confer early sensitivity and latent resistance to cisplatin in SCLC . Of particular importance is our discovery of the prevalence of SRSF1 CN gain and mRNA over-expression , and its role as a prognostic marker for poor patient survival—reported for the first time in SCLC . SRSF1 occurs in the same protein complex with topoisomerase 1 ( Top1 ) [33] . Topotecan is a Top1 inhibitor and the only agent with regulatory approval for the treatment of relapsed SCLC [34] . In normal cells , Top1 cooperates with SRSF1 to prevent the formation of DNA-RNA hybrids ( R-loops ) , unscheduled replication fork arrest , and genomic instability . In Top1 deficient cells , R-loops are formed and lead to replication fork stalling , phosphorylation of H2AX , and genomic instability . Treatment of Top1+ cells with diospyrin , to inhibit Top1phosphorylation of SRSF1 or with a siRNA targeting SRSF1 mimics a Top1-deficient phenotype [35] . Although significant correlation between SRSF1 and Top1 gene expression is not observed in our data , our experiment clearly demonstrates that SRSF1 loss induces phosphorylated H2AX signal in SCLC cell lines , which suggests that SRSF1 may help maintain the genomic integrity of SCLC to safeguard against DNA-damage and cell death . With these factors in mind , we propose that SRSF1 may also rely on modulating H2AX signal to sustain the tumorigenicity in some SCLC tumor patients . In the absence of specific limited stage ( LS ) or extensive stage ( ES ) disease determination in this study and a recent comprehensive SCLC study [9] , a simplified approach was used to classify SCLC patients into early and late stage disease activity . Based on known TNM information , early stage ( TNM stage I/II ) patients are M0 , who are usually designated as LS patients , while late stage ( TNM stage III/IV ) patients are M1a or M1b , and usually classified into ES patients . We then evaluated SRSF1 expression between early ( TNM stage I/II ) and late stage ( TNM stage III/IV ) SCLC patients . Results indicated that SRSF1 gene expression does not significantly differ between these patient groups in both this study and the George et al study ( p = 0 . 81 and p = 0 . 91 , respectively; S8 Fig ) . This may suggest that SRSF1 is not the key driver of cancer metastasis in SCLC . SRSF1 is one of the critical downstream transcriptional targets of Myc [36] . Myc family genes ( MYC and MYCN ) were shown to have significant CN gain or amplification events in our Chinese SCLC patients ( 14% ) . SRSF1 gene over-expression in both Myc and N-Myc amplified SCLC cell lines and Myc amplified SCLC tumor patients , however , was not observed ( p = 0 . 29 and p = 0 . 33 , respectively ) , though the number of amplified cell lines or patient tumors with available gene expression data was sparse for each comparison ( S9 Fig ) . SRSF1 is a key cancer driver , as demonstrated by the profound tumor-suppressive effect of specific SRSF1 knockdown in SRSF1-amplified or overexpressed SCLC models . Previous reports demonstrate that overexpression of SRSF1 results in oncogenic transformation of immortalized rodent fibroblasts [37] , human mammary epithelial cells [38] and mouse hepatocytes [39] . In these models , SRSF1 overexpression promoted cell proliferation , resistance to apoptosis , and formed tumors in orthotopic mouse models . It is likely that this transformation is a cumulative result of SRSF1’s many different functions , including a combination of several alternatively spliced oncogenic variants in response to an increase in SRSF1 levels . A number of such variants have been identified , but these probably represent only a small fraction of potential effectors [40] . Das et al , previously summarized various spliced products of SRSF1 and isoform mechanisms driving oncogenic phenotypes [40] , though these were not detected with reliability using RNASeq here–a challenge with this technology that currently persists in splice variant detection , especially in FFEE specimens . Furthermore , we demonstrate here that SRSF1 mediates the activation of both PI3K/AKT and MEK/ERK pathways as evidenced by both gene expression pathway analyses and the suppression of these pathways through SRSF1 knockdown . It is interesting to note that several SRSF1-regulated targets involved in regulating cell proliferation are downstream of these two pathways , including RPS6KB1 , MKNK2 , and CCND1 genes [37 , 41] . RPS6KB1 encodes the protein S6 kinase 1 , a downstream effector in the PI3K/AKT/mTOR signaling pathway and has been shown to be involved in mediating SRSF1-induced transformation [37 , 42] . MKNK2 is an effector in the MAPK/ERK pathway [43] . Splicing functionality has been shown to be critical for some , but not all oncogenic activities of SRSF1 . An SRSF1 variant that is confined to the nucleus has been shown to be critical for its oncogenic role in mammary epithelial cells [38] . However , this variant was not able to promote tumor formation in hepatocellular xenografts [39] . In this particular model , SRSF1-mediated oncogenesis was attributed to activation of Raf-MEK-ERK pathway [39] . This demonstrates that SRSF1 can be oncogenic via both nuclear and cytosolic activities through either canonical ( splicing-related ) or non-canonical ( AKT/ERK-related ) pathways under various cellular contexts . It may be of future interest to explore and pinpoint which effector pathway of SRSF1 drives its oncogenic roles in SCLCs . In conclusion , our discovery firmly establishes SRSF1 as a compelling therapeutic target for SCLC , especially for the population with poor outcome , as predicted by SRSF1 over expression .
The study protocol and informed consent from all studies in this study were approved by the Ethics Committee of Shanghai Chest Hospital and Nanjing Medical University . Informed consent in writing was obtained from each patient and the study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki as reflected in a priori approval by the Ethics Committee of Shanghai Chest Hospital and Nanjing Medical University . Ninety-nine Chinese patients who were diagnosed with primary SCLC were recruited prospectively into an ongoing study at the Jiangsu Cancer Hospital or Shanghai Chest Hospital from July 2004 to July 2013 . The diagnosis of SCLC was made by pathologists in the above hospitals by hematoxylin and eosin ( H&E ) staining according to histology plus the immunohistochemistry for chromogranin A and synaptophysin . Patients were followed up prospectively via routine hospital visits or telephone calls . The phone calls were conducted by trained medical staff to patients or their family contacts once every three month until death or last time of follow-up . All patients were treated with at least one cycle of chemotherapy after surgery . The clinical features of the patients are summarized in Table 1 and S1 Table . Of the 99 patients , 25 had matched normal adjacent tissue or blood , while 74 patients only had tumor specimens . All tissues samples were FFPE archived samples collected from surgery ( not biopsy ) . Eighty-six tumor samples were treatment naïve and 13 of 99 patients were treated with standard chemotherapy before surgery . Tumor contents in each tumor and normal adjacent tissue ( NAT ) was assessed by H&E stain and the tumor and NAT were subjected to macro-dissection and tumor purity was >70%; the tumor content in each NAT was< 3% . The Caucasian SCLC patient cohort consisted of 25 FFPE lung tumor tissue specimens with matched normal adjacent tissue pairs , which were purchased from Conversant Biologics , Inc ( Huntsville , AL ) ( S11 Table ) . The diagnosis of SCLC was confirmed by two independent pathologists in Medimmune by H&E staining . All samples were treatment naïve surgical samples . All patients were Caucasian with 24 males and 1 female . The average age of the patients was 63 . 3 years ( range of 40–76 years ) . The tumor stages ranged from stage I to IV . The tumor and NAT were macro-dissected and tumor purity was >70%; the tumor content in each NAT was< 3% . DNA whole exome sequence ( WES ) and RNA sequencing data ( RNASeq ) data was generated using the Illumina standard library preparation and sequencing protocols as described in [44] The SureSelect Human All Exon V5 capture kit was used to capture coding regions of genes included in the major genomic databases . Paired end FASTQ files of 90mer sequence reads for both sequence data types were provided to MedImmune . RNASeq data has been deposited into GEO under accession GSE60052 while WES data was deposited into dBGaP under accession 12059 . All sequence data was QCd for read counts , quality values , kmer usage , GC-content , and all other relevant parameters with FastQC ( v0 . 10 . 1 ) . The DNA read sequences were aligned to the human genome ( UCSC hg19; Feb 2009 release; Genome Reference Consortium GRCh37 ) using GATK ( v2 . 3 . 4; [45] ) and both insertion/deletion ( indel ) realignment and PCR duplicate removal was conducted using GATK ( v2 . 3 . 4; [45] ) and Picard ( v1 . 85; [46] ) respectively . Both coverage and depth statistics for all 99 tumor specimens are provided in S10 Table . For the 25 tumor/normal matched Chinese and 25 tumor/normal matched Caucasian ( commercially purchased ) specimens , both Mutect ( v1 . 1 . 4; [47] ) and SAMtools ( v0 . 1 . 18; [48] ) were used to make somatic variant calls . SAMtools mpileup arguments: Qphred>30 and mapping quality>30 with minimum coverage >20; MuTect arguments: default settings . GATK SomaticIndelDetector with default settings and SAMtools mpileup were used to identify small indels . The SNVs and indels which were in common between GATK and Samtools were retained . SNVs and indels were further filtered by 1000 genomes and NHLBI-ESP project with 6500 exomes minor allele frequency ( MAF ) in all races of <1% or unknown MAF . The retained SNVs/indels were further filtered by dbSNP129 and dbSNP135 , following known issues between the two dbSNP versions . Finally , genes were removed from the SNV/indel list that had been identified from a previous study as potential artifact genes , to further minimize false positive variant calls [49] All dbSNPs which were retained in dbSNP135 and had Cosmic IDs were noted for further study . For the 74 DNA tumor specimens without a matched normal specimen , Samtools mpileup was used to call SNVs and indels relative to the human reference genome ( UCSC hg19; Feb 2009 release; Genome Reference Consortium GRCh37 ) . Germline polymorphisms were removed by retaining only mutations with MAF in all races of <1% or unknown MAF within the 1000 genomes and NHLBI-ESP project with 6500 exomes database . The retained SNVs/indels were further filtered by dbSNP129 and dbSNP135 similar to previously described . The most recurrent SNVs/indels between the matched and unmatched patient cohorts are provided in S2 Table , along with patient recurrence summaries from a previous Japanese SCLC cohort of 51 patients , to highlight comparability in results and a validation of the SNV/indel calling strategy [5] . A similar strategy for calling and filtering somatic SNVs in the absence of a matched germline control specimen was conducted in a previous prostate cancer whole exome study [50] . All patient-level somatic SNV or indel calls with associated read depth and annotation parameters are provided in S8 Table . SNV and indel annotation was conducted with ANNOVAR [51] To verify the identity and matching between the tumor and normal paired WES samples , a selection of 300 heterozygous single nucleotide polymorphisms ( SNPs ) with MAFs>0 . 3 and <0 . 7 were selected from the 1000 genomes database . All DNA samples were clustered to observe any major discrepancies in subject or specimen labeling ( S10 Fig ) . All somatic mutations in the coding regions ( plus splicing mutations ) were selected for driver gene prediction analysis to identify those genes with the most recurrent nonsilent mutations . MutsigCV [39] and the method described by Youn et al [52] were implemented independently and Q value<0 . 05 ( MutsigCV ) and Q value = 0 . 00 ( Youn’s method ) were used as thresholds to detect significantly recurrently mutated genes . Genes predicted by both methods were selected as high confidence driver genes ( S3 Table ) . Amino acid change mutations were mapped onto corresponding structures using mutagenesis wizard implemented in PyMOL ( Schrodinger , LLC ) . For POLG coordinates of human mitochondrial DNA polymerase holoenzyme from Protein Data Bank ( PDB , [53] ) entry 3IKM [54] were used . The Q52E mutation could not be mapped since that part of the protein was absent in the structure . For DNA polymerase delta subunit the PDB entry 3IAY of yeast that shares 48/65% sequence identity/similarity over 908 amino acids was used . For RNASeq data , the average read count per mate was 50 million . RNA reads were mapped to the human genome ( UCSC hg19; Feb 2009 release; Genome Reference Consortium GRCh37 ) using TopHat2 ( v2 . 0 . 9; [55 , 56] ) and the human reference gtf annotation file ( GRCh37 . 68 ) . Transcript counts were calculated and normalized using htseq-count and DESeq ( v1 . 12 . 1; [57] ) . The DESeq negative binomial distribution was used to calculate the p-value and fold changes between 48 lung tumor and 6 normal adjacent lung samples using adjusted p<0 . 05 and |fold change|>2 as a threshold . The full transcriptome summary table is provided ( S9 Table ) . Due to the low fidelity and lack of reproducibility in splice variant detection using RNASeq , analysis was not conducted to examine spliced products of SRSF1 . For CNV analysis , the R package ExomeCNV [58] was used . This method makes CNV calls not by defining a mandatory cut-off to detect gains or losses , rather the specificity and sensitivity ( power ) of detecting CNV based on depth of coverage and log ratio of all exons is calculated , and a CN call is made when sufficient specificity and sensitivity are achieved . We used default parameters setting of ExomeCNV ( sensitivity and specificity = 99 . 9% ) . For the 22 tumor/normal matched Chinese as well as the 25 tumor/normal Caucasian ( commercially purchased ) specimens , the standard ExomeCNV pipeline was employed , in which a tumor and its adjacent normal pair were used to make the call . For the 74 tumor specimens without matched normal tissue , 1 normal FFPE lung tissue specimen ( N08-4579A ) was used as baseline with each of the 74 tumor specimens using ExomeCNV . This method was also conducted with 6 normal FFPE lung tissue specimens and results were very similar between the use of a single normal or average of 6 normals . The overview of the most prevalent CNV calls ( ≥20% patients harboring gains or losses , to limit the table size ) for matched Chinese patient tumor/normal or Chinese patient tumor only results are provided in S5 Table . All cancer indications in TCGA were assessed for correlation with survival using OncoLand ( OmicSoft Corp; Cary , NC ) . To avoid issues of unbalanced comparisons , only indications where at least 3 patients harboring a CN gain in SRSF1 were analyzed . These included: breast invasive carcinoma ( BRCA ) , kidney renal papillary cell carcinoma ( KIRP ) , sarcoma ( SARC ) , skin cutaneous melanoma ( SKCM ) , and uterine corpus endometrioid carcinoma ( UCEC ) . UCEC was the only indication with a correlation between patients harboring CN gain of SRSF1 and poor survival ( log-rank test p = 0 . 003 ) , though the number of patients harboring a CN gain was highly unbalanced compared to those without ( n = 8 vs . n = 437 , respectively; S3 Fig ) . Time-to-event analyses were used to correlate both the CN gain status of SRSF1 and SRSF1 gene expression with overall survival of Chinese SCLC patients . First , a Kaplan-Meier ( KM ) analysis was used to evaluate the difference of survival curves for SRSF1 CN gain group and no CN gain group . Those genes with a trend of significance ( log-rank p<0 . 1 ) in the Chinese patient discovery cohort ( n = 22; SRSF1 in Fig 3D ) and with 10% CNV calls among the cohort were evaluated in the Chinese patient validation cohort ( n = 74; 1 , 707 genes; SRSF1 in Fig 3E ) . Since the discovery cohort was approximately 1/3 the size of the validation cohort and thus less powered , a modest log-rank test threshold was used . Among those 1 , 707 genes , 215 had p-values<0 . 05 from the log-rank test and CNV calls in more than 10% of the patients in the cohort . Among these 215 genes , SRSF1 was the only gene that correlated with DNA CN gain status using a Welch’s modified t-test ( p<0 . 01; Fig 3B ) . Next , both the Chinese patient discovery and validation cohorts were combined ( n = 96 ) and both a KM and multivariate Cox proportion hazard ( PH ) regression analysis was conducted to compare the SRSF1 CN gain and no CN gain patient groups . Differences were assessed with p-values for the grouping difference ( log-rank ) and the hazard ratio with adjustment for age , gender , tumor stage and chemotherapy treatment status before sampling ( Cox PH model; Fig 3F ) . Then , the gene expression of SRSF1 in the 48 Chinese SCLC patients with RNASeq and clinical data were divided into two groups according to SRSF1 gene expression level ( >75% percentile of overall expression and < = 75% percentile of overall expression ) . Similar KM analysis as well as a Cox PH regression analysis was performed to compare the survival curves of SRSF1 over- expressed versus not over-expressed groups with the same covariate adjustments in the Cox PH model as conducted previously with WES data ( Fig 3C ) . The R package survival was used to perform these analyses and model summaries are provided in both Fig 3 and S7A and S7B Table . A similar time-to-event analysis adjusting for age , gender , tumor stage and chemotherapy treatment status was conducted using the nonsilent mutation status to split patients into two groups . SRSF1 gene copy number change was conducted via a dual-probe FISH test . The SRSF1 FISH probe was a SpectrumRed ( Cat #02N34-050 , Enzo Life Sciences , Inc . , New York , USA ) labeled fluorescent DNA probe , generated in-house from a bacterial artificial clone CTD-2061E5 ( Invitrogen , Carlsbad , USA ) . CEP17 probe ( Vysis , Cat #06J37-017 ) was a SpectrumGreen labeled fluorescent DNA probe specific for the alpha satellite DNA sequence at the centromeric region of chromosome 17 . FISH assays were performed as reported previously . In brief , assays were run on 4 micron dewaxed and dehydrated FFPE samples from 34 small cell lung cancer patients . The SpotLight Tissue pretreatment Kit ( Cat #00–8401 , Invitrogen , Carlsbad , USA ) was used for pretreatment according to the manufacturer’s instructions . Sections and probes were codenaturated at 79oC for 6 minutes and then hybridized at 37oC for 48 hours . After a quick post wash off process ( 0 . 3%NP40/2xSSC at 75 . 5 oC for 2 minutes , twice in 2×SSC at room temperature for 2 minutes ) , sections were finally mounted with 0 . 3μg/ml DAPI ( Cat #H-1200 , Vector Laboratories , Inc . , Burlingame , USA ) . CN gains were scored using the criteria outlined by Cappuzzo et al ( 18 ) where disomy was scored by ≤2 copies in ≥90% of cells , low trisomy was scored by ≤2 copies in ≥40% of cells and ≥3 copies in 10–40% of the cells , high trisomy was scored by ≤2 copies in ≥40% of the cells and ≥3 copies in ≥40% of the cells , and polysomy was scored by ≤2 copies in <40% of the cells . High trisomy and polysomy were called CN gain positive . ( S6 Table ) . Genomic DNA ( gDNA ) from cultured cells was prepared using QIAamp DNA Micro Kit . Copy number assay of SRSF1 ( Hs00944074_cn ) and reference assay RNase P ( VIC ) were ordered from ABI/Life Technologies . Assays were set up based on ABI reference with four replicates for each sample . The assays were run on ABI 7900HT ( SDS v2 . X ) and the data files were analyzed using the CopyCaller Software . Reference probe RNAse-P was used to determine the SRSF1 copy number gain status: copy number > 2 was considered a gain status . All SCLC cell lines were grown in RPMI1640 medium supplemented with 10% fetal bovine serum . SRSF1 ( SF2/ASF ) antibody ( 96 ) was supplied by Santa Cruz Biotechnology . Phospho-Histone H2A . X ( Ser139 ) ( 20E3 ) and Phospho-Chk2 ( Thr68 ) ( C13C1 ) were supplied by Cell Signaling Technology . Cell proliferation was determined by CellTiter-Glo Luminescent Cell Viability Assay ( Promega ) . Caspase-Glo 3/7 Assay Systems ( Promega ) were used to analyze cell apoptosis . SiRNA reverse transfections were carried out using Lipofectamine RNAiMAX ( Life Technologies ) . siRNAs targeting SRSF1 were ordered as “HP custom siRNA” from Qiagen . The sequences is and CCAACAAGATAGAGTATAA ( SRSF1 siRNA ) . AllStars Neg . Control siRNA ( Qiagen ) was used as negative control for transfection . Both control siRNA and SRSF1 siRNAs were transfected at a final concentration of 100nM . Culture medium were was replaced with fresh medium at 48 hour after transfection , and cell lysates were prepared at 72 hour for Western blotting . For clonogenic assay , SCLC cell lines were transfected with SRSF1 siRNAs for 48 hrs and then seeded in a 1% methylcellulose H4100 medium ( StemCell Technologies ) consisting of RPMI1640 medium with 10% FBS at 2 , 000 cells/mL . After 5 days , colonies with more than 40 cells per colony were counted . SCLC cell lines were transfected with SRSF1 siRNAs for 48 hrs and then seeded in ultralow attachment plates ( Corning ) in sphere forming media: DMEM/F12 with 0 . 4% BSA , 10ng/mL bFGF , 20ng/mL EGF , 5ug/mL insulin , 1% KnockOut Serum Replacement ( Life Technologies ) . Cells were treated with Cisplatin ( 0 . 001 ug– 10 ug/ml ) for 4 days , after which viability of spheres was quantitated by CellTiter-Glo Assay ( Promega ) . Images were taken with EVOS FL Auto Cell Imaging System . SCLC cell lines were cotransfected with 800 ng myc/flag-tagged SRSF1 vector ( Origene ) encoding the open reading frame of either the wildtype gene ( NM_006924 . 4 with 25 nM of either non-targeting siRNA or SRSF1 siRNA-2 using Lipofectamine RNAiMAX ( Life Technologies ) . SRSF1 siRNA targets the 3’UTR of SRSF1 , and therefore does not affect expression of the SRSF1 ORF vector . After 48 hr , cells were harvested and then seeded in ultralow attachment plates ( Corning ) in sphere forming media: DMEM/F12 with 0 . 4% BSA , 10ng/mL bFGF , 20ng/mL EGF , 5ug/mL insulin , 1% KnockOut Serum Replacement ( Life Technologies ) . Cells were also harvested and lysed with Novex Tris-Glycine SDS Sample Buffer ( Life Technologies ) for Western blotting . Viability of spheres was quantitated after 4 days by CellTiter-Glo Assay ( Promega ) . Images were taken with EVOS FL Auto Cell Imaging System All animal procedures were conducted in accordance with all appropriate regulatory standards under protocols approved by the Medimmune Institutional Animal Care and Use Committee . Since the SRSF1 siRNA had shown good knockdown efficacy of SRSF1 protein at day7 after transient transfection ( by western blot of sphere assays ) , and prolonged effects on colony formation ( about 2 weeks after transfection ) , we used transient siRNA knockdown in the mice xenograft study . Immunocompromised athymic nude ( nu/nu ) female mice were purchased from Harlon Laboratories at 3–4 week of age . SHP-77 and DMS-114 cells were transfected with either control siRNA or SRSF1 siRNA at a final concentration of 100nM . Two days after transfection , ten million viable cells in 50% matrigel were inoculated subcutaneously ( SC ) into right flank of each mouse . The length and width of each tumor was measured with an electronic cliper 2 times per week . Tumor growth curves of DMS114 and SHP77 parental cell lines are displayed in S11 Fig . Tumor volume ( mm3 ) was calculated based on the following formula: [length ( mm ) x width ( mm ) 2] ÷ 2 .
|
SCLC patients are initially highly chemo-sensitive with response rates of greater than 80% in both limited and extensive diseases , but suffer uniform disease recurrence or progression in a very short period of time . In the absence of well-defined genomic biomarkers and insights into the resistance mechanism , many targeted treatments have yielded negative results in the last decade Using integrated next generation sequencing ( NGS ) technology in combination with a high quality surgical sample set with comprehensive clinical annotation , our study not only identified novel recurrent genetic alterations in genes such as CDH10 and DNA repair pathways which may influence outcomes in SCLC patients , but also discovered the expression of SRSF1 , an RNA-splicing factor which can both regulate key oncogenic and survival pathways such as BCL2 , and play a critical role in patient survival .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"gene",
"regulation",
"cancer",
"treatment",
"cancers",
"and",
"neoplasms",
"ethnicities",
"oncology",
"dna",
"small",
"interfering",
"rnas",
"chinese",
"people",
"lung",
"and",
"intrathoracic",
"tumors",
"gene",
"expression",
"people",
"and",
"places",
"biochemistry",
"rna",
"carcinogenesis",
"small",
"cell",
"lung",
"cancer",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"dna",
"repair",
"human",
"genetics",
"non-coding",
"rna"
] |
2016
|
Genomic Landscape Survey Identifies SRSF1 as a Key Oncodriver in Small Cell Lung Cancer
|
Identifying optimal dosing of antibiotics has proven challenging—some antibiotics are most effective when they are administered periodically at high doses , while others work best when minimizing concentration fluctuations . Mechanistic explanations for why antibiotics differ in their optimal dosing are lacking , limiting our ability to predict optimal therapy and leading to long and costly experiments . We use mathematical models that describe both bacterial growth and intracellular antibiotic-target binding to investigate the effects of fluctuating antibiotic concentrations on individual bacterial cells and bacterial populations . We show that physicochemical parameters , e . g . the rate of drug transmembrane diffusion and the antibiotic-target complex half-life are sufficient to explain which treatment strategy is most effective . If the drug-target complex dissociates rapidly , the antibiotic must be kept constantly at a concentration that prevents bacterial replication . If antibiotics cross bacterial cell envelopes slowly to reach their target , there is a delay in the onset of action that may be reduced by increasing initial antibiotic concentration . Finally , slow drug-target dissociation and slow diffusion out of cells act to prolong antibiotic effects , thereby allowing for less frequent dosing . Our model can be used as a tool in the rational design of treatment for bacterial infections . It is easily adaptable to other biological systems , e . g . HIV , malaria and cancer , where the effects of physiological fluctuations of drug concentration are also poorly understood .
The rise of antibiotic resistance underlines the need for employing existing antibiotics prudently . Although antibiotic dosing regimens have been investigated for more than half a century [1] , we do not yet have a sufficient understanding of the link between drug dosing and bacterial killing to design rational treatment strategies [2 , 3] . Even for antibiotic regimens that have been standard of care , substantial improvements in dosing levels [4] , treatment frequency [5] and treatment duration[6–8] have been made decades after their introduction . Most experimental and some clinical studies investigate antibiotic concentrations at a constant or at an average concentration . However , drug concentration at target tissues can fluctuate substantially over time . These fluctuations can influence the effectiveness of treatment , with the importance of such fluctuations differing substantially between classes of antibiotics [9] . Three alternative descriptions of effective antibiotic concentration are commonly used ( so-called pharmacokinetic drivers ) : i ) the total concentration integrated over a given time interval ( area under the curve , AUC ) , ii ) the peak concentration ( Cmax ) or iii ) the time during which the concentration exceeds a specific threshold ( time above MIC , TC>MIC , Fig 1A ) . For some drugs Cmax correlates best with bacterial clearance [10] , for example in clinical trials with isoniazid [11] . Even once-weekly dosing was slightly superior to daily dosing for the novel TB drug bedaquiline when holding total drug administration constant [12] . For rifampicin [13] and quinolones , the total amount of drug [14] appears to be the best predictor of treatment success . For beta-lactams , the time above the minimal inhibitory concentration ( MIC ) correlates best with bacteriological response [2] . For some antibiotics , such as tetracycline , antibacterial action depends on both TC>MIC and AUC [10] . Each of these three measures of exposure ( AUC , Cmax and TC>MIC ) would be optimized by employing different dosing strategies , for example by using large intermittent doses to increase Cmax or by employing extended release formulations to increase TC>MIC [15] . A clear mechanistic understanding of antibiotic pharmacodynamics has not yet been achieved , and this lack of knowledge is a major obstacle for the design of rational treatment regimens . Treatment strategies for bacterial infections ( e . g . dose levels , dosing frequency , and duration of therapy ) are usually developed based on pharmacodynamic and pharmacokinetic data collected through expensive in vitro and in vivo studies [9 , 16–18] . Specifically , the question of which pharmacokinetic driver governs antibiotic efficacy has to be determined experimentally with hollow-fiber systems or animal models [11 , 19 , 20] . This experimental information in turn can be incorporated into mathematical models [21] , but to our knowledge there is no mathematical model that can guide these experiments . Thus , the development of models that can inform optimal dosing strategies from data collected in early phases of antibiotic development could speed the drug development process and help to identify promising compounds that should be prioritized [22] . Here , we extend a modeling framework [23] that integrates bacterial population biology with the intracellular reaction kinetics of antibiotic-target binding to investigate how the kinetics of drug-target binding affect bacterial response to fluctuating antibiotic concentrations . We find that the physicochemical characteristics of drug action predict differences in antibiotic pharmacodynamics at fluctuating concentrations and correlate well with observed data .
Using three models that incorporate complexity and realism in a stepwise fashion ( Fig 1 ) , we consider how reaction kinetics govern the expected bacterial responses to antibiotics . First , we use a simple model that considers only drug-target binding to explore the general principles of antibiotic-target reaction kinetics . Then , we use more complex models to simulate the action of two specific antibiotics , ampicillin and tetracycline , under a range of different dosing strategies . We assess which physicochemical characteristics of these two drugs explain their distinct pharmacodynamic behavior and evaluate how an understanding of these physicochemical characteristics can inform more effective dosing regimens .
Recommended antibiotic dosage varies widely depending on the employed antibiotic and the targeted pathogen . It is therefore difficult to compare antibiotic action in terms of absolute concentrations . Typically , all measures of antibiotic efficacy are defined relative to the MIC of the specific bacteria/drug pair ( Cmax/MIC , AUC/MIC and TC>MIC ) to circumvent this problem . To be able to use a modeling framework based on physicochemical characteristics of drug action , it is therefore useful to define the MIC based on physicochemical properties [23] . In the simplest case , when we assume a constant antibiotic concentration in this framework , the MIC depends on two parameters: the drug target affinity ( KD ) and the threshold of bound target ( fc ) at which the net growth of a bacterial population is zero ( Eq ( 3 ) ) . Fig 2A illustrates the expected MIC according to Eq ( 3 ) which depends on drug target affinity KD and the critical threshold fc . The absolute concentration of antibiotic at MIC rises with the threshold occupied target required for bacterial suppression ( fc ) . Given any threshold level of target occupancy , drugs with a higher binding affinity ( lower KD ) will require smaller concentrations to prevent bacterial growth ( lower MIC ) . Classical models of antibiotic pharmacodynamics typically assume that the antibiotic concentration at any time point determines the net bacterial growth rate at that same time point . This assumes both that the antibiotic acts instantaneously and that previous antibiotic exposure has no continuing influence on bacterial growth . In reality , however , there is typically a delay between initial exposure and antibiotic effect and there may also be post-antibiotic period in which bacterial growth remains suppressed even after the antibiotic is removed from the extracellular space . Here we use our modeling framework to understand how the onset and end of antibiotic action are affected by the physicochemical properties of drug-target binding . We use the reaction kinetics of drug target binding ( Eq ( 4 ) ) to show the dynamics when the antibiotic is applied at a concentration slightly above MIC ( 1 . 01 x MIC ) . We use this concentration for illustration purposes because at this concentration , the critical fraction of binding fc is reached in finite time , but never substantially exceeds this threshold . ( The influence of higher concentrations is explored in Fig 3 . ) The different scenarios in Fig 2B illustrate the time course of drug-target binding at the same concentration relative to the MIC , but different absolute antibiotic concentrations . From the limited number of studies in which antibiotic-target dissociation rates have been directly measured , we assume that these rates range between 10−3/s and 10−4/s [24 , 25] . Slower turnover of drug-target binding ( i . e . a longer half-life of the drug-target complex ) is associated with a delayed onset of action ( compare Fig 2B dotted to solid lines ) . Surprisingly , we find that the system approaches equilibrium more quickly when fc is higher . This effect can be explained as follows: The absolute antibiotic concentration at MIC rises sharply with the threshold fc , and can map to very different absolute drug concentrations ( see Fig 2A ) . Initially , only the forward reaction is relevant when a negligible amount of target is bound and proceeds with the rate kf[A][T0] ( i . e . as the product of forward reaction rate , antibiotic concentration and target molecule concentration ) . Under conditions where the antibiotic concentration is held constant , the equilibrium fraction of bound target [A] is given by [AT]eq=[A][A]+KD and asymptotically approaches 1 . Therefore , the velocity of the reaction increases more quickly than the fraction of bound target at equilibrium . This produces a paradoxical finding: when dosing antibiotics at the same levels relative to their respective MIC , those that require a high threshold of bound target to be effective are expected to act more quickly ( Fig 2B ) . The delay until an antibiotic is effective depends on many physiological and biochemical factors . Since this model focuses on the reaction kinetics alone ( ignoring diffusion barriers and concentration gradients ) , Model 1 provides a lower bound for the expected delay until onset of antibiotic action . Even here , for reasonable parameter settings , we find that even this delay can extend for several hours . One potential approach for speeding antibiotic-target binding and reducing delay to onset of action is to increase antibiotic exposure through higher dosing . Lower thresholds and slower turnover are associated with delays until antibiotic action; these effects can be overcome by increasing the drug concentration ( Fig 3 ) . The light blue solid and dotted vertical lines in Fig 2B indicate when the fast and slow reactions reach the fc , i . e . the time of onset of the antibiotic action ( tonset ) , and can be compared to the blue and green lines in Fig 3B at 1 . 01MIC . When the antibiotic-target reaction equilibrates slowly , a high dose of antibiotic is especially beneficial and minimizes the opportunity for additional bacterial replication events prior to onset of antibiotic action . Bacterial growth often remains suppressed after the antibiotic concentration drops below the MIC ( i . e . the post-antibiotic effect ) . This effect occurs because drug-target complex dissociation is not instantaneous . Therefore , high drug concentrations that saturate the target beyond the threshold required for antibiotic action fc may have additional benefits if they extend bacterial suppression beyond the time that the antibiotic concentration exceeds the MIC . We use our model to identify the conditions in which high antibiotic concentrations are expected to prolong antibiotic action . For simplicity , for these simulations we assume that at the time of antibiotic withdrawal , 99 . 9% of the target is bound and that the antibiotic concentration both inside and outside of the bacterial cell immediately drops to zero . Under these assumptions , Eq ( 1 ) can be simplified and the unbinding of the antibiotic corresponds to a simple exponential decay . Fig 4 illustrates the expected dissociation of the drug-target complex for antibiotics with different half-lives tbound . When the threshold required for antibiotic action fc is very high , the antibiotic stops working very rapidly and the length of the post-antibiotic effect is brief and relatively insensitive to the half-life of the drug-target complex . Conversely , when there is both a low threshold and a slow turnover time of drug-target binding , the post-antibiotic period may last for several hours .
Beta-lactams acetylate penicillin-binding proteins ( PBPs , the target molecules ) , and thereby inhibit cell wall synthesis . The acetylation of PBPs consumes beta-lactams , and therefore the drug-target reaction is not reversible . However , PBPs are constantly de-acetylated and the effects of the antibiotic are therefore reversible . The kinetics of PBP acetylation and de-acetylation as well as target occupancy at MIC have been determined experimentally ( Table 1 ) . In single cell experiments , ampicillin has no detectable sub-MIC activity ( S1 Fig ) so we assume that antibiotic is effective only while the fraction of bound antibiotic exceeds fc . To explore whether TC>MIC , AUC or Cmax are the best predictors of antibiotic efficacy , we model three simplified dosing strategies: i ) an idealized bolus injection where the drug concentration immediately reaches its peak and then declines exponentially , ii ) a hypothetical pharmacokinetic curve where the antibiotic concentration is maintained just above the MIC ( 1 . 01x MIC ) for the same length of time >MIC as in i ) and then falls instantaneously to 0 , and iii ) a curve of similar shape to ii ) that retains the same area under the curve as i ) ( see Fig 5A ) . In ii ) , the time above MIC is identical to i ) but we eliminate the excess binding that occurs because of the initial high peak in i ) . In iii ) , the AUC is the same as in i ) but instead of the high peak concentration , there is a significantly prolonged TC>MIC . In other words , all graphs in the middle column of Fig 5 have the same time > MIC as those in the left column , and all graphs in the right column have the same AUC as those in the left column . First , we investigate whether our modeling framework can reproduce the time-dependent action of beta-lactams based on the known physicochemical characteristics of the drug and its target . Fig 5A shows numerical simulations of Eq ( 6 ) using experimentally determined parameters ( see Table 1 ) . Note that this equation includes diffusion across the bacterial cell envelope . However , penicillin-binding proteins are either located in the cell envelope or in case of gram-negatives in the periplasm , and we therefore assume here that the diffusion barrier to the target is negligible . We adapted Eq ( 6 ) to describe the consumption of beta-lactams during target acetylation by dropping the backward reaction term kr[AT] for the differential equation describing the intercellular antibiotic [A]i . For all three dosing strategies , antibiotic action starts immediately after the drug concentration rises above MIC and stops as soon as the drug concentration falls below MIC , i . e . increasing the AUC alone without increasing TC>MIC does not change antibiotic action substantially ( compare Fig 5A left and right panel ) . This is in accordance with the observation that the efficacy of beta-lactams strongly depends on the time above MIC . Taken together , we can reproduce time-dependent action of beta-lactams solely based on reaction kinetics . The high threshold required for activity as well as the extracellular location of the target lead to a fast onset of drug action as the antibiotic concentration rises above the MIC and a nearly immediate end of antibiotic action as the antibiotic concentration drops below the MIC . To illustrate the different dynamics of bolus injections and constant dosing , we visualize time course of ampicillin action for a bolus injection in S1 Movie ( Fig 5A , left panel ) and for a constant concentration in S2 Movie ( Fig 5A , middle panel ) . Beta-lactams are the antibiotic class for which time-dependent action is most widely accepted , and we can reproduce their time-dependent action with our model . This suggests that physicochemical characteristics may be responsible for this behavior . For most other antibiotic classes , antibacterial efficacy is better correlated with AUC or Cmax [13 , 14] . We hypothesized that alteration in specific physicochemical parameters could generate AUC and Cmax-dependent action . To investigate this hypothesis , we modified the parameters for ampicillin one at a time to determine whether , through such parameter modification , we could reproduce AUC and Cmax-dependent action . Because antibiotic treatments are usually given over several days and the time between individual doses is typically in the range of hours , we first investigated parameter changes that produce an equilibration time of several hours . For example , if the drug must diffuse across a cell envelope with a diffusion rate of p = 10−4 /s , this leads to a half-life of free intracellular drug of 1h 55min . Fig 5B shows a comparison of the same dosing strategies as used for the upper panel ( Fig 5A ) with this additional diffusion barrier . With such a strong diffusion barrier the antibiotic concentration inside the cell also remains above MIC after a bolus injection of 50x MIC for several hours because antibiotic molecules are retained within the cell . Consequently , the activity after such a bolus administration can be extended by several hours ( Fig 5B ) . However , the diffusion barrier also delays the onset of antibiotic action . This delay is dependent on the antibiotic concentration , the left panel of Fig 5B shows a delay of 14 minutes while the right panel shows a delay of 13h . This is because the equilibration of intra- and extracellular concentration is slower when there are smaller differences between the concentrations outside and inside the bacterial cell . If the antibiotic dose is only slightly above MIC , >10h are required to reach the threshold for inhibition fc ( Fig 5B , right panel ) . Thus , a dosing strategy with an equivalent time >MIC as the 50x MIC bolus administration will never achieve bacterial suppression , while a dose with an equivalent area under the curve is approximately 5 . 4 times more effective than the bolus injection ( bolus injection: antibacterial activity from 14min to 11h9min , constant concentration with same MIC: activity from 13h to 71 . 5h ) . S2 Fig shows the dynamics when keeping Cmax constant but varying the drug half-life . Again , we would expect beta-lactam action to start immediately and end immediately when the drug falls below MIC; however , for an antibiotic with a substantial diffusion barrier , we would expect delays until the onset and cessation of drug action . This behavior is not limited to slow equilibration rates due to diffusion barriers . Earlier , we identified two parameters that affect the onset and the end of antibiotic action: target occupancy at MIC ( fc ) and the half-life of the drug-target complex ( tbound ) . A long exposure to ampicillin at MIC is expected to result in a target saturation that just reaches the critical threshold fc = 95 . 4% [38] at equilibrium . After withdrawal of the drug , the target saturation would immediately fall below this critical threshold and the antibiotic would no longer be active . The duration of antibiotic action might be extended with higher drug concentrations since this would produce higher target saturation and lead to a longer delay until the fraction bound target falls below the critical threshold fc . However , increasing the target saturation at the beginning from 95 . 4% to 99 . 9% is expected to extend the action of ampicillin only by about 9 minutes ( 0 . 999ln ( fc ) −kr ) . On the other hand , if the critical threshold fc was 10% instead of 95 . 4% , achieving a target saturation of 99 . 9% would extend the expected time of antibiotic action by over 6h . Similarly , if the rate of de-acetylation is decreased 10-fold ( kr = 10−5 ) , the expected duration of antibiotic action after achieving a target saturation of 99 . 9% is extended by over 1 . 5h . Changing both parameters to values that result in equilibration rates in the range of hours leads to the same qualitative behavior as when the equilibration rate is in the range of hours because of a diffusion barrier ( S3 Fig ) . Thus , our model predicts that changing a single physiochemical parameter ( equilibration times to the range of hours ) has major impact on pharmacodynamics: instead of TC>MIC alone , the AUC becomes another predictor of antibiotic efficacy and both are needed to predict antibiotic action . S1 Movie and S2 Movie illustrate the time course of the action of a hypothetical antibiotic that has the same binding and de-acetylation rates as ampicillin , but where the antibiotic must cross a diffusion barrier with p = 10−4 /s and a threshold of fc = 10% . As in S1 Movie and S2 Movie , we compare a bolus injection ( S3 Movie ) and a concentration with the same time above MIC ( S4 Movie ) . We now examine the consequences of slow drug equilibration rates ( i . e . in the range of days ) on predicted antibiotic pharmacodynamics . We slow the diffusion rate across the bacterial cell envelope to p = 10−5 , which corresponds to a half-life of 19h 15min . In this case , the antibiotic concentration inside the cell remains above MIC after a bolus injection of 50x MIC for a day . Exposure to the antibiotic at a concentration only slightly above MIC ( 1 . 01 x MIC ) is insufficient to achieve the required amount of bound target , even when maintained for several days ( Fig 5C ) . Thus , in situations where antibiotics are expected to equilibrate slowly , a high peak concentration is necessary to achieve antibiotic action and the Cmax is expected to be the best predictor of antibiotic action . We further tested this finding by investigating the reaction kinetics of a drug with a very different mechanism of action: the antitubercular pro-drug isoniazid ( INH ) . In this case , target binding occurs after drug activation to the adduct INH-NAD which depends on NAD content and oxygen saturation . Importantly , the majority of the active drug INH-NAD remains in the mycobacterial cell and is not able to cross the cell envelope [49 , 50] . Because INH-NAD remains in the cell , the expected amount of bound target does not decline even when the external concentration of INH declines . We therefore interpret treatment success here as the required time to reach fc , i . e . the expected time after which an average bacterium is killed . Again , we can reproduce experimental and clinical findings that INH treatment efficacy is significantly correlated with both Cmax and AUC in univariate regressions ( Fig 6 ) . S1 Table gives an overview of all parameters combinations used in the simulations . In a multivariate regression , only Cmax is significantly correlated with the time to reach the required threshold to kill bacteria ( S4 Fig ) , although the best model according to the Akaike Information Criterion includes all three pharmacokinetic indices . Due to the prodrug-activation , it takes several hours to reach the threshold required for killing , and this delay can be reduced with high peak concentrations . Since the active drug , INH-NAD , is trapped inside the cell , the intracellular drug concentration does not decrease when the external concentration decreases . Therefore , it is not necessary to keep the external pro-drug concentration above MIC for the drug to be active .
The time above the MIC is expected to be a reasonable predictor of antibiotic action in situations where antibiotic concentrations below the MIC have little effect on bacteria . For example , cells exposed to 80% MIC ampicillin show no measurable defect in either growth or elongation rates , and all cells remain intact ( S1 Fig ) . In contrast , translation inhibitors such as chloramphenicol and tetracycline do affect bacterial growth below MIC , and it has previously been shown a nearly complete suppression of growth at 80% MIC [23] . When fitting Eq ( 10 ) to data from single cells exposed to constant sub-MIC concentrations of antibiotics [23] , it has previously been estimated that a high threshold of bound ribosomes must be met to interrupt all bacterial growth ( fc = 98% ) and that there is a low diffusion barrier ( p = 1 . 2x 10−2/s ) . Experimental values from the literature suggest a short half-life of the drug-target complex ( Table 1 ) . Based on the values of these parameters , we expect that the TC>MIC should be the best predictor of tetracycline effects . However , experimental and clinical evidence suggests that both AUC and TC>MIC determine the efficacy for tetracycline [23] . Accordingly , we used Model 3 ( Eq ( 10 ) populated with parameters for tetracycline ) , to investigate how sub-MIC activity affects antibiotic pharmacodynamics under different dosing strategies . Fig 7A shows simplified pharmacokinetics of a tetracycline bolus injection with initial concentrations ranging from 0 . 1–5 x MIC . Fig 7B shows the effects of dosages above MIC on the bacterial growth rate . Given the low diffusion barrier , bacterial growth is completely suppressed as long as the antibiotic concentration is retained above MIC . As soon as the antibiotic concentration falls below MIC , bacterial growth immediately resumes and continues to increase in rate as the antibiotic is cleared . Thus , when considering pharmacokinetic measures that correlate with the complete suppression of bacterial growth , the time above MIC is the best predictor of antibiotic action . However , the model also suggests that sub-MIC concentrations may substantially affect the total expected bacterial load over 24h ( Fig 7C ) . Accordingly , our model predicts that TC>MIC may be an imperfect predictor of antibiotic action ( at least as measured by its effect on the total bacterial burden over 24h ) since sub-MIC exposure can impact expected bacterial burden ( first 5 data points in Fig 8A , highlighted in blue ) . In contrast , for antibiotic dosages above MIC , TC>MIC correlates well with efficacy ( last 5 data points in Fig 8A , highlighted in red ) , however , it should be noted that the overall effect is very small . Nevertheless , in a treated patient with low remaining bacterial burden the additional killing of few bacteria can make the difference between cure ( i . e . extinction of bacterial population ) or relapse . Our model is deterministic and therefore cannot capture extinction , however , depending on the initial population size a very small frequency of survivors that translates to less than one bacterium effectively means extinction . Over a wider range of antibiotic dosages in bolus injection , the area under the curve correlates more strongly with antibiotic effects because it is a measure that also reflects actions that occur below MIC ( Fig 8B ) .
It is well known that certain pharmacokinetic measures ( i . e . AUC , Cmax or TC>MIC ) are better predictors of the pharmacodynamics of some antibiotics than of others , but we currently have limited quantitative understanding of the mechanisms that drive this phenomenon . In this paper , we extend a model that links chemical reaction kinetics to bacterial population biology [23] and suggest a potential mechanistic explanation for this phenomenon . Based on this model , we suggest how physicochemical and biochemical characteristics of drug-target interaction may shape antibiotic dose response curves . Differences in characteristics between antibiotics offer a compelling explanation for the observation that different measures of drug exposure correlate best with antibacterial activity . Specifically , we identified four factors that govern patterns of drug effects: i ) the half-life of the antibiotic-target complex , ii ) the diffusion barrier between extracellular antibiotic and its target , iii ) the threshold of bound target required to suppress bacterial growth ( i . e . target molecule occupancy at MIC ) and iv ) drug effects when the antibiotic is present only at sub-MIC levels . The first three factors , the half-life of drug-target complex , the diffusion barrier and the threshold required for bacterial suppression , all influence the time until the antibiotic starts and stops acting ( i . e . the equilibration rate of the reaction ) . When the onset of action of an antibiotic is rapid , we expect that achieving drug concentrations just above MIC should be sufficient to trigger the antibacterial effect . If an antibiotic stops acting quickly , antibiotic effects should cease as soon as the concentration falls below MIC . In these circumstances , we expect that the time above MIC would be a good measure for antibiotic efficacy . We demonstrated that our model , when parameterized with relevant drug-target binding data from the literature , can reproduce such time-dependent pharmacodynamics of ampicillin . Beta-lactams are somewhat unique in that their targets are located outside the cytosol [54] . Therefore , there is negligible diffusion barrier between the antibiotic molecules surrounding a bacterial cell and their targets . Our model predicts that this leads to a fast onset and end of antibiotic action . Also , almost all target molecules are occupied at MIC [38] , and we demonstrate here that this also should lead to a rapid onset and cessation of antibiotic activity . Time-dependent efficacy of beta-lactams is well established both experimentally and clinically . For example , it is recommended that beta-lactams are given as continuous infusion rather than bolus injections [27] . For most other antibiotic classes , antibacterial efficacy is correlated with AUC or Cmax [10 , 14] . Many antibiotics have targets that are located in the cytosol ( e , g . ribosomal-targeting antibiotics such as streptomycin or gyrase-targeting antibiotics such as ciprofloxacin ) . Also , unlike beta-lactams , many antibiotics will have effects before the majority of target molecules are bound . We therefore investigated whether our model can also reproduce concentration-dependent patterns of antibiotic action , in which antibiotic efficacy is best described by either Cmax or AUC . Indeed , our model predicts that the TC>MIC is not highly correlated with treatment efficacy when the time until an antibiotic starts and stops being active ( i . e . the equilibration time ) is in the range of hours or longer . The delay until an antibiotic is effective depends on many physiological and biochemical factors . Here , we focus on the reaction kinetics alone , which provide a lower bound for the expected time to onset of antibiotic action . We note that even these lower-bound estimates may be as long as a few hours , potentially permitting several additional generations of bacterial replication . We would therefore suggest high doses , at least initially , for antibiotics that: 1 ) act at low thresholds of bound target; 2 ) diffuse only slowly through the cell envelope; or 3 ) have a slow turnover rate ( i . e . a long half-life of drug-target binding ) . A similar argument can be made for the anti-tuberculosis drug isoniazid , which is a prodrug that is activated by bacterial cells . The activation rate of the drug alone is sufficient to explain the slow onset of action of the drug [23] , and this delay can likely be reduced with higher antibiotic doses . Indeed , the efficacy of isoniazid has been linked to high peak doses [55] , a finding we were able to reproduce here . Additionally , when equilibration rates are slow , higher dose of antibiotics can extend the action of the antibiotic beyond the time the antibiotic concentration outside the bacteria exceeds the MIC . Thus , high doses have the additional benefit of prolonging the post-antibiotic period for antibiotic-target pairs that equilibrate slowly . In isoniazid , this extension of drug action predicted by our model is especially pronounced , because the drug is trapped in the cell such that declining external drug concentrations have little effect . In principle , these delays in onset and end of action are a similar phenomenon to the concept of a “biophase lag” [56] although the underlying mechanisms are not the same . To examine the conditions in which each of these pharmacokinetic metrics provides the best measure of drug effect , we compared a dosing strategy with a high peak concentration that facilitates rapid onset of an antibacterial effect with a dosing strategy that has an equivalent AUC , but a lower peak concentration and a substantially longer exposure time ( Fig 5 ) . If an antibiotic equilibrates slowly , the onset of antibiotic action at low doses is so delayed that the required fraction of bound target cannot be reached before the antibiotic falls below MIC in the low dose/long exposure strategy ( Fig 5 ) . Obviously , the exact parameter ranges in which this is the case depend on the definition of “long” ( in our case , days ) . If equilibration is too slow compared to the relevant timeframe ( for example due to the accumulation of activated isoniazid in the cell ) , we would expect that Cmax is a better predictor of antibacterial efficacy than the AUC . Whether the peak concentration ( Cmax ) or the total exposure ( AUC ) is the best predictor of antibiotic efficacy thus depends on both the observed timeframe and the equilibration rate . In addition to the onset and end of antibiotic action , we found that the biological activity of the antibiotic at sub-MIC concentrations also determines which pharmacokinetic measure best predicts treatment efficacy . A similar argument has been made for rifampicin therapy in tuberculosis[57] . Some antibiotics such as ampicillin ( S1 Fig ) have very little effect below MIC . In contrast , some antibiotics like tetracycline have some sub-MIC activity . Clearly , the time above MIC alone cannot predict treatment success when sub-MIC concentrations partially suppress bacterial growth . Indeed , our model predicts that treatment efficacy with tetracycline depends both on TC>MIC and AUC which is in concordance with clinical and experimental studies [10] . Taken together , our mechanistic model can reproduce the pharmacodynamic characteristics of both ampicillin and tetracycline . It offers an intuitive explanation for differences in optimal dosing strategies between antibiotic classes . However , the parameters needed to inform even such a simple model have not yet been measured for many antibiotic/bacterial pairs . We note that most of the kinetic measurements for antibiotic-target binding were published decades ago [24 , 34 , 35] . To our knowledge , beta-lactams are the only antibiotic class for which target occupancy at MIC has been experimentally determined . Furthermore , the number of target molecules per cell and especially the concentration of free antibiotic at the target site are rarely known , despite being a focus of active research in tuberculosis [58–61] . We suggest that experiments to address these knowledge gaps should be prioritized as the results of these studies could inform new approaches for the rational dosing of antibiotics . Identifying optimal antibiotic dosing strategies is challenging and in this paper we have addressed only a subset of the considerations that must be accounted for when determining treatment recommendations . For example , antibacterial efficacy and toxicity must be balanced and the frequency of dosing may affect adherence; these are important factors that should doubtless affect treatment recommendations . In addition , our simple models do not consider host immune responses to infection , which may further modify our expectations regarding treatment success [57 , 62 , 63] . Nevertheless , given the urgent need to preserve the efficacy of existing antibiotics and the need to develop new agents [64] , we see a promising role for mechanistic models that can suggest the most promising dosing strategies based on the physicochemical and biochemical characteristics of drug-target interactions . Such novel pharmacodynamics models can also be integrated into more complex frameworks that include host responses and more sophisticated pharmacokinetics [57 , 62 , 63] . Our model is general and we believe it could be usefully adapted to improve dosing strategies for treatment of other diseases . For example , we note that the effects of the physiological fluctuations of drug concentration are also poorly understood in the treatment of cancer [65] , HIV [66] and malaria [67] and similar questions arise regarding the effects of exposure to harmful substances in toxicology [68] .
Previously , we have shown that models that consider drug-target binding kinetics can explain complex patterns of antibiotic action such as post-antibiotic effects , inoculum effects , and persistence [23] . The central assumption of these models is that bacterial replication decreases and/or bacterial killing increases with the fraction of bound target molecules . Here , we extend this approach using three different mathematical models that incorporate additional complexity and biological realism in a stepwise fashion ( Fig 1B ) . In all these models we follow the entire bacterial biomass rather than single cells . For our purposes here and in contrast to previous work [23] , we can simplify the model by assuming that there is negligible heterogeneity between single cells . Table 2 lists all parameters and variables of these models .
To build our understanding of the drug-target reaction kinetics as antibiotic concentrations fluctuate within a host , Model 1 focuses only on the drug-target binding that occurs after exposure and withdrawal of an antibiotic . For Model 1 we make the following simplifying assumptions ( which are subsequently relaxed in Models 2 and 3 ) : The chemical reaction of antibiotics with their targets is described by the following equation: A+T ⇌ AT . The intracellular antibiotic molecules A react with target molecules T with a rate kf and form an antibiotic-target molecule complex . If the reaction is reversible , the complex dissociates with a rate kr , leading to a dynamic equilibrium . The dynamics of this system are governed by the concentrations [A] , [T] , [AT] rather than the absolute number of molecules . We assume that the total concentration of target/cell [T0] is constant . In this case , the concentration of free target can be described as [T] = [T0] − [AT] . Assuming that cells are treatment-naïve , i . e . there are no bound target molecules at the beginning , the kinetics of antibiotic-target reaction can then be described by a single differential equation , which can be simplified if we assume the intracellular antibiotic concentration [A] is constant: d[AT]dt=kf[A] ( [T0]−[AT] ) −kr[AT] ( 1 ) and solved as: [AT] ( t ) =kf[A][T0] ( 1−e− ( kr+kf[A] ) t ) kr+kf[A] ( 2 ) At a certain point , the fraction of bound target reaches a critical threshold at which the net growth of the bacterial population is zero . In this framework , the MIC is characterized as the minimal antibiotic concentration at which this critical percentage of bound target , fc , is reached . Thus , the MIC is the antibiotic concentration at which the equilibrium fraction of bound antibiotic is exactly fc: i . e . [AT]MIC[T0]=fc . After simplifying , this yields: MIC=KDfc1−fc ( 3 ) with the affinity constant KD=krkf . Expressing all antibiotic concentrations as fold-MIC ( xMIC ) and thereby replacing [A] with MICKDfc1−fc , Eq ( 2 ) can then be transformed: [AT] ( t ) =fcT0xMIC1−fc ( 1−xMIC ) ( 1−e−kr ( 1−fc ( 1−xMIC ) ) t1−fc ) ( 4 ) The time to the onset of antibiotic action , i . e . the delay until the fraction of bound target first exceeds fc after antibiotic administration , can be expressed as: tonset=fc−1kr ( 1+fc ( xMIC−1 ) ) log ( − ( xMIC−1 ) ( fc−1 ) xMIC ) ( 5 )
Model 2 includes the following compartments: Ae , the number of extracellular antibiotic molecules , Ai , the number of intracellular antibiotic molecules , T , the number of free target molecules , and AT , the number of drug-target complexes . For bolus injections , the model is described by the following set of equations: dAedt=−ln ( 2 ) tclAe−p ( AeViVe−Ai ) dAidt=p ( AeViVe−Ai ) −kfnAViAiT+krATdTdt=−kfnAViAiT+krATdATdt=kfnAViAiT−krAT ( 6 ) To model an alternative drug administration approach in which the antibiotic concentration is maintained at a constant level c and after a specified time ( tend ) is assumed to fall instantaneously to 0 ( i . e . intravenous dosing ) , the extracellular antibiotic concentration is given by: Ae={cfort<tend0fort≥tend ( 7 ) We express the antibiotic concentration as fold-MIC ( xMIC ) using Eq ( 3 ) . The terms describing the chemical kinetics of drug-target reaction are equivalent to Eq ( 1 ) . In addition , we describe the diffusion through the cell envelope with a permeability coefficient p depending on the concentration difference inside and outside of the bacterial cells and the clearance of the extracellular antibiotic; its half-life is tcl . In our simulations , drug binding and diffusion from extra- to intracellular space changes the dynamics of external drug concentrations negligibly , even though this may change at very high bacterial loads with a high number of targets per cell [69] . Here , we use the same equations and parameters as in Figure 7 in [23] , extended by diffusion across the cell envelope and a decay term that describes the elimination of the drug from the blood after a bolus injection with t1/2 . In the case of the prodrug isoniazid ( INH ) , target binding occurs after drug activation to the adduct INH•NAD ( equivalent to A before ) which depends on NAD content and oxygen saturation . Here , we focus on INH binding to the enoyl reductase InhA , which is then present in its inactive form InhAi . Assuming NAD and target molecule concentration as well as oxygen saturation remain constant , the number of molecules in each compartment is described by the following set of equations: dINHedt=−ln ( 2 ) tclINHe−p ( INHeViVe−INHi ) dINHidt=p ( INHeViVe−INHi ) −kNAD , O2INHidINH∙NADdt=kNAD , O2INHi−kfnAViINH∙NADInhA+krInhAidInhAidt=kfnAViINH∙NADInhA−krInhAi ( 8 ) This set of equations is based on ( 6 ) and we additionally model prodrug activation .
Finally , Model 3 expands on Model 2 by allowing the reproduction of target molecules that would occur as a result of bacterial replication and also allows for unspecific binding . ( This extension relaxes assumption 3 and partially relaxes assumption 4 in the list provided above . ) This model describes antibiotics that only suppress bacterial growth but do not increase bacterial killing ( i . e . bacteriostatic agents ) . For bacteriostatic translation inhibitors such as tetracycline , the bacterial replication rate depends linearly on the fraction of free ribosomes [70] . We therefore assume that the bacterial growth rate r is proportional to ffree=[T][T]+[AT] above ff = 1- fc and that there is no growth when the fraction of free ribosomes falls below this critical threshold: r ( ffree ) ={0forffree<ffrnodrug11−ff ( ffree-ff ) forffree>ff ( 9 ) Here , we track bacterial cells B ( scaled in number of cells per liter ) that can reproduce until they reach a maximal carrying capacity K , the extracellular and intracellular number of antibiotics Ae and Ai , and the intracellular concentration of drug-target complexes AT and unspecifically bound antibiotic AU . The rates kf and kr describe specific binding and dissociation , the rates ku , f and ku , r describe the rates for unspecific binding and dissociation . Data indicate that the total number of ribosomes increases linearly with cell volume; this means that the intracellular concentration within a single cell between the time of its “birth” and the split into two daughter cells remains relatively constant [31] . We can therefore write the number of free target molecules as T = BT0 − AT with T0 describing the fixed number of total target molecules per cell . The growth of bacteria exposed to sub-MIC concentrations of a translation inhibitor can then be described by the following set of differential equations ( note that we are again following molecules , not molar concentrations ) : dBdt=r ( B , T0 , [AT] , fc ) B ( 1−BK ) dAedt=−ln ( 2 ) tclAe−p ( AeViVe−Ai ) dAidt=p ( AeViVe−Ai ) −kfnAViAiT+krAT−ku , fAiB+ku , rAUdATdt=−kfnAViAiT+krATdAUdt=ku , fAiB−ku , rAU ( 10 ) Again , this equation is based on ( 6 ) , in addition , we model bacterial population biology by following the total amount of bacteria B .
|
In this era of rising concerns about antibiotic resistance , the rational design of optimal antibiotic treatment regimens remains an important unrealized goal . At this time , the characteristics of antibiotic treatment regimens ( e . g . dosing levels , treatment duration , route of administration ) are determined largely based on costly in vivo experiments . The sheer number of possible dosing strategies that must be tested contributes to the delay and cost of the development of new drugs and may limit the feasibility of finding optimal regimen characteristics . Here , we demonstrate how modeling the chemical kinetics of drug-target binding can identify the best time-concentration profile of antibiotics . Using both analytical approaches and numerical simulations , we find that the physicochemical characteristics of drug-target binding are sufficient to explain the pharmacodynamics of commonly used antibiotics such as ampicillin , isoniazid and tetracycline . In practical terms , our models can be used as a tool in the rational design of treatment for bacterial infections . Because of the generality of drug-target binding kinetics , these approaches may also be adapted to other diseases where the effects of physiological fluctuations of drug concentration are also poorly understood , such as HIV , malaria and cancer .
|
[
"Abstract",
"Introduction",
"Results",
"Model",
"1:",
"General",
"Principles",
"of",
"Antibiotic-Target",
"Reaction",
"Kinetics",
"Model",
"2:",
"Onset",
"and",
"End",
"of",
"Antibiotic",
"Action",
"Determine",
"Pharmacodynamic",
"Properties",
"of",
"Ampicillin",
"and",
"Isoniazid",
"Model",
"3:",
"Antibiotic",
"Action",
"below",
"MIC",
"Modifies",
"Pharmacodynamic",
"Properties",
"of",
"Tetracycline",
"Discussion",
"Methods",
"Model",
"1:",
"Drug-Target",
"Binding",
"Only",
"Model",
"2:",
"Drug-Target",
"Binding",
"with",
"Diffusion",
"Barrier",
"and",
"Bolus",
"Injection",
"Model",
"3:",
"Including",
"Unspecific",
"Binding",
"and",
"Bacterial",
"Replication"
] |
[
"antimicrobials",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"intracellular",
"pathogens",
"pathogens",
"drugs",
"microbiology",
"developmental",
"biology",
"antibiotic",
"resistance",
"tetracyclines",
"antibiotics",
"pharmacodynamics",
"pharmacology",
"microbial",
"growth",
"and",
"development",
"antimicrobial",
"resistance",
"microbial",
"physiology",
"isoniazid",
"bacterial",
"growth",
"pharmacokinetics",
"microbial",
"control",
"biology",
"and",
"life",
"sciences"
] |
2017
|
Using Chemical Reaction Kinetics to Predict Optimal Antibiotic Treatment Strategies
|
Host defense against the intracellular protozoan parasite Trypanosoma cruzi depends on Toll-like receptor ( TLR ) -dependent innate immune responses . Recent studies also suggest the presence of TLR-independent responses to several microorganisms , such as viruses , bacteria , and fungi . However , the TLR-independent responses to protozoa remain unclear . Here , we demonstrate a novel TLR-independent innate response pathway to T . cruzi . Myd88−/−Trif−/− mice lacking TLR signaling showed normal T . cruzi-induced Th1 responses and maturation of dendritic cells ( DCs ) , despite high sensitivity to the infection . IFN-γ was normally induced in T . cruzi-infected Myd88−/−Trif−/− innate immune cells , and further was responsible for the TLR-independent Th1 responses and DC maturation after T . cruzi infection . T . cruzi infection induced elevation of the intracellular Ca2+ level . Furthermore , T . cruzi-induced IFN-γ expression was blocked by inhibition of Ca2+ signaling . NFATc1 , which plays a pivotal role in Ca2+ signaling in lymphocytes , was activated in T . cruzi-infected Myd88−/−Trif−/− innate immune cells . T . cruzi-infected Nfatc1−/− fetal liver DCs were impaired in IFN-γ production and DC maturation . These results demonstrate that NFATc1 mediates TLR-independent innate immune responses in T . cruzi infection .
The host defense against invasion of intracellular pathogens relies on Th1 cell-derived IFN-γ that activates macrophages to kill the engulfed pathogens [1] . Toll-like receptor ( TLR ) -mediated recognition of pathogens has been established to induce activation of innate immune cells such as dendritic cells ( DCs ) and subsequent development of Th1 cells [2] , [3] . However , recent evidence also indicates the presence of TLR-independent mechanisms for the recognition of microorganisms such as bacteria , viruses , and fungi [4] , [5] . Accordingly , TLR-independent mechanisms for Th1 development have been demonstrated in several infectious models such as fungal and bacterial infections [6] , [7] . However , TLR-independent recognition of protozoa remains unknown . Trypanosoma cruzi is an intracellular protozoan parasite that causes Chagas' disease , a chronic disorder characterized by cardiomyopathy and malformation of the intestine [8] . Several components of T . cruzi have been shown to induce TLR-dependent activation of innate immunity and subsequent development of Th1 cells [9]–[14] . The absence of TLR-dependent activation of innate immunity results in high susceptibility to T . cruzi infection [15] , [16] due to defective type I interferon ( IFN ) -mediated induction of the GTPase IFN-inducible p47 ( IRG47 ) [17] . Invasion of infective metacyclic trypomastigotes of T . cruzi into host cells induces a close interaction between the parasites and the host , because T . cruzi utilize several host-derived factors in order to establish the infection . These include activation of Ca2+ signaling pathways and phosphatidylinositol-3 kinases [18]–[20] . However , it remains unclear how T . cruzi-mediated activation of host cytoplasmic signaling pathways is regulated and whether it is TLR-dependent or -independent . In T cells , the nuclear factor of activated T cells ( NFAT ) family of transcription factors has been shown to mediate production of cytokines including IFN-γ [21] , [22] . The NFAT family of proteins comprises four closely related members ( NFATc1 , NFATc2 , NFATc3 , and NFATc4 ) that are activated by Ca2+ signaling , and NFAT5 that is regulated by osmotic stress . The role of NFAT proteins in T cells has been well characterized [21] , [22] . However , little is known about the role of NFAT proteins in innate immune responses , although some of the NFAT members are highly expressed and can modulate gene induction in macrophages ( Mφ ) [23] , [24] . Here , we analyzed the mechanisms of TLR-independent activation of innate immunity during T . cruzi infection using T . cruzi-infected Myd88−/−Trif−/− mice . Our results demonstrate that NFATc1 mediates TLR-independent induction of IFN-γ in innate immune cells , leading to effective Th1 responses during T . cruzi infection .
Previously , we demonstrated that mice lacking both MyD88 and TRIF , in which TLR-dependent activation was abolished , are highly sensitive to infection with T . cruzi [17] . Because TLRs have been shown to control development of Th1 cells , we analyzed Th1 responses in T . cruzi-infected mice . Mice were intraperitoneally ( i . p . ) infected with T . cruzi trypomastigotes , and at 6 days of infection CD4+ T cells were isolated from the spleen and stimulated with anti-CD3 antibody ( Ab ) ( Figure 1A ) . In T . cruzi-infected wild-type mice , there was considerable production of IFN-γ compared with that in non-infected control mice , indicating induction of potent Th1 responses . In Myd88−/− and Myd88−/−Trif−/− mice , IFN-γ production was similar to that in wild-type mice following T . cruzi infection . Next , we analyzed the antigen-specific Th1 response at 0 , 4 , 6 , and 10 days after T . cruzi infection by stimulating CD4+ T cells with freeze-thawed T . cruzi in the presence of antigen presenting cells ( APC ) ( Figure 1B ) . This stimulation induced marked production of IFN-γ at 6 and 10 days of the infection in wild-type mice . Even in CD4+ T cells derived from T . cruzi-infected Myd88−/− and Myd88−/−Trif−/− mice , antigen-specific production of IFN-γ was induced to levels similar to that of wild-type mice . Thus , the antigen-specific Th1 response was not impaired in Myd88−/− and Myd88−/−Trif−/− mice . We also analyzed IFN-γ production from CD4+ T cells by intracellular staining ( Figure 1C , Figure S1A ) . The number of IFN-γ-producing CD4+ T cells was almost equally elevated in wild-type , Myd88−/− and Myd88−/−Trif−/− mice at 6 days ( Figure 1C ) as well as at 10 days ( Figure S1A ) after infection . Consistent with previous studies [25] , [26] , the number of IFN-γ producing CD8+ T cells and NK1 . 1+ natural killer cells was not increased at 10 days after T . cruzi infection ( Figure S1B ) . Development of Th1 cells is critically controlled by DCs [27] , [28] . In addition , stimulation of TLRs induces maturation of DCs [3] . Therefore , we analyzed expression of MHC class II and co-stimulatory molecules on T . cruzi-infected DCs . Bone marrow-derived DCs ( BMDCs ) were infected with T . cruzi trypomastigotes for 6 h , then were washed , cultured for 48 h , and analyzed for expression of MHC class II , CD40 , and CD86 by flow cytometry ( Figure 1D ) . T . cruzi infection resulted in enhanced expression of these molecules in wild-type BMDCs . Expression was also increased in BMDCs derived from Myd88−/−Trif−/− mice after T . cruzi infection , indicating normal maturation of T . cruzi-infected DCs of Myd88−/−Trif−/− mice . Thus , Th1 cell development and DC maturation were induced during T . cruzi infection even in the absence of TLR-dependent activation of innate immunity . T . cruzi infection induced maturation of DCs in the absence of TLR signaling . Therefore , we screened genes that were normally induced in T . cruzi-infected DCs of Myd88−/−Trif−/− mice . BMDCs from wild-type , Myd88−/− and Myd88−/−Trif−/− mice were infected with T . cruzi trypomastigotes for 6 h , then mRNA was extracted and used for DNA microarray analysis . Approximately 80% of genes that were induced in T . cruzi-infected wild-type DCs ( about 4000 genes ) were MyD88-dependent , as the T . cruzi-mediated induction was reduced in Myd88−/− DCs ( Figure S2 ) . Some of the genes that were normally induced in Myd88−/− DCs , but not induced in Myd88−/−Trif−/− DCs ( MyD88/TRIF-dependent genes; 14% of genes that were induced in wild-type DCs ) are known to be induced by type I IFNs . In addition , a majority of the small number of genes that were induced even in Myd88−/−Trif−/− DCs ( 6% ) were IFN-γ-inducible genes ( Figure S3 ) . In order to corroborate that IFN-γ-inducible genes are normally induced in T . cruzi-infected Myd88−/−Trif−/− DCs , we analyzed mRNA expression of Ifng and IFN-γ-inducible genes , including Stat1 , and Irgm , by real-time RT-PCR ( Figure 2A ) . T . cruzi infection resulted in robust induction of Ifng , Stat1 , and Irgm in wild-type , Myd88−/− , Trif−/− , and Myd88−/−Trif−/− DCs . T . cruzi-induced expression of Stat1 and Irgm in wild-type DCs was inhibited by addition of a de novo protein synthesis inhibitor , cycloheximide ( CHX ) ( Figure 2B ) . In contrast , CHX did not inhibit T . cruzi-induced Ifng expression ( Figure 2C ) . Next , in order to analyze whether the expression of the IFN-γ-inducible genes was secondary to induction of Ifng , we used BMDCs derived from Ifngr1−/− mice in which the IFN-γ-mediated response was abolished ( Figure 2D , E ) . In Ifngr1−/− BMDCs , T . cruzi-mediated induction of Stat1 and Irgm was reduced , whereas induction of Ifng was unimpaired . These data indicate that Ifng was induced primarily in response to T . cruzi infection , and Stat1 and Irgm were induced secondary to Ifng induction . In peritoneal Mφ , similar patterns of T . cruzi-mediated gene expression were observed ( Figure S4 ) . Recently , the CD11clowB220+NK1 . 1+ subset of cells was identified as a natural killer ( NK ) cell subset with a high capacity for IFN-γ production in response to IL-12 or a TLR9 ligand [29]–[31] . To exclude the possibility of contamination of these cells in preparation of BMDCs or Mφ , we purified CD11chighB220−NK1 . 1− population from the spleen , and analyzed for IFN-γ expression ( Figure S5A ) . CD11chighB220−NK1 . 1− cells showed very low levels of IFN-γ expression in response to IL-12/IL-18 stimulation compared with the NK cell subset ( Figure S5B ) . However , these cells from wild-type and Myd88−/−Trif−/− mice expressed IFN-γ in response to T . cruzi infection ( Figure 2F ) . Flow cytometric analysis further demonstrated that T . cruzi-infected CD11chigh splenic DCs expressed IFN-γ protein ( Figure 2G ) . These findings indicate that T . cruzi infection induces IFN-γ production in DCs . Next , we analyzed whether IFN-γ was involved in TLR-independent DC maturation and Th1 responses during T . cruzi infection . In BMDCs derived from Ifngr1−/− mice , T . cruzi-induced enhancement of CD40 , CD86 , and MHC class II was partially reduced ( Figure 3A ) . Furthermore , expression of these molecules was completely abolished in T . cruzi-infected DCs of Myd88−/−Trif−/−Ifngr1−/− mice . Enhanced expression of these molecules in response to exogenous IFN-γ was not observed in Ifngr1−/− BMDCs ( Figure S6A ) . These findings indicate that IFN-γ produced from T . cruzi-infected DCs mediated DC maturation . We also analyzed IFN-γ production from splenic CD4+ T cells of T . cruzi-infected mice ( Figure 3B ) . In both Ifngr1−/− and Myd88−/−Trif−/−Ifngr1−/− mice , T . cruzi antigen-dependent production of IFN-γ was severely reduced . Importance of IFN-γ production was further underscored by the finding that Ifngr1−/− mice were more sensitive to T . cruzi infection than Myd88−/−Trif−/− mice ( Figure S6B ) . These results demonstrate that IFN-γ mediates TLR-independent DC maturation and Th1 development during T . cruzi infection . Importance of IL-12 in Th1 cell development has been established [32] . Indeed , IL-12p40-deficient mice were highly susceptible to T . cruzi infection with severely reduced Th1 responses ( [33] , [34] and Figure S7A ) . In addition , IL-12p40 concentration in the serum was decreased in T . cruzi-infected Ifngr1−/− mice ( Figure S7B ) . In T . cruzi-infected Myd88−/−Trif−/− mice , IL-12p40 production was severely reduced , but still induced [17] , suggesting that IL-12 is produced via TLR-dependent and -independent pathways . Thus , IFN-γ , which is produced via the TLR-independent pathways , might induce IL-12p40 to activate T cells to fully differentiate into Th1 cells . Next , we analyzed the molecular mechanisms for TLR-independent induction of IFN-γ after T . cruzi infection . In Myd88−/−Trif−/− DCs , T . cruzi-induced phosphorylation of MAP kinases such as ERK , p38 , and JNK , as well as degradation of IκBα was not observed at all ( Figure S8A ) . In addition , T . cruzi infection did not induce DNA binding activity of NF-κB in Myd88−/−Trif−/− DCs ( Figure S8B ) . Thus , T . cruzi-mediated activation of NF-κB and MAP kinases was not induced in the absence of TLR signaling . Next , we stimulated DCs with T . cruzi trypomastigotes killed by repeated freeze-thaw steps . Live T . cruzi , but not killed parasites , induced Ifng expression ( Figure 4A ) . Because many studies have demonstrated that T . cruzi utilize the host Ca2+ signaling to establish the infection [35] , we assessed the intracellular Ca2+ concentration in T . cruzi-infected BMMφ using a fluorescent Ca2+ indicator Fluo-4 AM ( Figure 4B , Figure S9A , B ) . T . cruzi infection led to rapid increase in intracellular Ca2+ level in both wild-type and Myd88−/−Trif−/− Mφ , which returned to the basal level after 18 min of the infection . Epimastigotes , which are not able to invade the host cells , did not induce the elevation of Ca2+ concentration in Mφ ( Figure S9C ) . These results prompted us to examine whether Ca2+ mobilization induced by intracellular invasion of T . cruzi contributed to the TLR-independent Ifng induction . Accordingly , we treated wild-type and Myd88−/−Trif−/− BMDCs with an intracellular Ca2+ chelator , bis- ( o-aminophenoxy ) -ethane-N , N , N′ , N′-tetraacetic acid tetra ( acetoxymethyl ) ester ( BAPTA-AM ) , and infected with T . cruzi . In BAPTA-AM pre-treated DCs , T . cruzi-induced Ifng expression was severely reduced , although lipopolysaccharide ( LPS ) -induced response was not impaired ( Figure 4C ) . In this condition , T . cruzi-induced elevation of intracellular Ca2+ concentration was severely reduced ( Figure S10 ) . In addition , stimulation with both phorbol myristate acetate ( PMA ) /Ca2+ ionophore or Ca2+ ionophore alone , which mimics Ca2+ signaling , induced expression of Ifng in both wild-type and Myd88−/−Trif−/− Mφ ( Figure 4D ) . Taken together , these findings indicate that T . cruzi-dependent intracellular Ca2+ mobilization mediates TLR-independent Ifng induction . In the host cells , especially in T lymphocytes , Ca2+ mobilization induces activation of cytokine genes via calmodulin/calcineurin-dependent activation of the transcription factor NFAT . Therefore , we treated Myd88−/−Trif−/− Mφ with FK506 to block calcineurin activation , and infected with T . cruzi . Treatment of FK506 resulted in a marked decrease in T . cruzi-induced expression of Ifng , despite normal LPS-induced response ( Figure 5A ) . Among NFAT members , Nfatc1 , Nfatc3 , and Nfat5 mRNA were abundantly expressed in BMDCs ( Figure S11 ) . A previous study has shown that NFATc1 increased anti-CD3/anti-CD28-induced IFN-γ promoter activity in T cells [36] . Furthermore , it has been demonstrated that IFN-γ production was normal in NFATc3-deficient splenocytes [37] . In addition , NFAT5 has been shown to be activated by osmotic stress , but not by Ca2+ signaling [38] . Thus , we focused on NFATc1 . In wild-type and Myd88−/−Trif−/− BMMφ , T . cruzi trypomastigotes infection induced nuclear translocation of NFATc1 ( Figure 5B ) . T . cruzi-induced nuclear translocation of NFATc1 was blocked by the pre-treatment with BAPTA-AM in wild-type and Myd88−/−Trif−/− BMMφ ( Figure S12A , B ) . These results indicate that NFATc1 is activated in response to T . cruzi infection in a TLR-independent manner . Next , we analyzed whether NFATc1 was involved in the T . cruzi-induced IFN-γ production . We obtained RAW264 . 7 macrophage clones expressing different levels of NFATc1 ( Figure 5C ) . In NFATc1 expressing RAW264 . 7 cells , T . cruzi-induced expression of Ifng , Stat1 , and Irgm was enhanced , and the extent of fold-induction correlated with the NFATc1 expression level ( Figure 5D ) . T . cruzi-induced Ifng expression was severely reduced in the presence of FK506 ( Figure 5E ) . These findings indicate the possible involvement of NFATc1 in mediating IFN-γ production in T . cruzi-infected innate immune cells . The NFAT family of transcription factors has been shown to interact with different transcription factors to effectively induce gene activation [22] . In the case of activation of the human IFNG promoter , T-bet ( encoded by Tbx21 ) and NFAT have been shown to act synergistically on the promoter [39] . T-bet is a transcription factor that is essential for IFN-γ production in T cells and DCs [40] . Tbx21 has also been shown to be induced by IFN-γ in monocytes and DCs [41] . In accordance with those studies , Tbx21 was normally induced in T . cruzi-infected Myd88−/−Trif−/− Mφ ( Figure 6A ) . Expression of NFATc1 alone weakly activated the Ifng promoter in RAW264 . 7 Mφ , but introduction of both NFATc1 and T-bet synergistically activated the Ifng promoter in RAW264 . 7 Mφ ( Figure 6B ) . These findings indicate that NFATc1 mediates activation of the Ifng promoter together with T-bet in innate immune cells , like the case in T cells . To determine the role of NFATc1 in T . cruzi-infected DCs , we investigated IFN-γ production and maturation in T . cruzi-infected Nfatc1−/− DCs . Because mice lacking Nfatc1 are lethal before day14 . 5 of gestation [42] , we obtained fetal liver-derived DCs ( FLDCs ) . Fetal liver cells of both wild-type and Nfatc1−/− embryos at day 12 . 5 of gestation differentiated into DCs expressing similar levels of CD11c in the presence of GM-CSF , Flt3 ligand and SCF ( Figure S13 ) . Moreover , LPS-induced maturation , as determined by enhanced surface expression of CD40 , CD86 , and MHC class II , was not impaired in Nfatc1−/− mice-derived cells ( Figure S14A ) , indicating that development and LPS-induced maturation of Nfatc1−/− FLDCs was not compromised . T . cruzi-infected wild-type FLDCs expressed increased amounts of Ifng . In contrast , in FLDCs from Nfatc1−/− embryos the T . cruzi-induced Ifng expression was impaired ( Figure 7A ) . On the other hand , T . cruzi-induced Il6 and Tnf expression was observed normally in Nfatc1−/− FLDCs ( Figure 7B ) . In T . cruzi-infected Nfatc1−/− FLDCs , expression of Il12b ( encoding IL-12p40 ) was partially decreased ( Figure S14B ) . Next , we analyzed T . cruzi-induced expression of MHC class II and co-stimulatory molecules on wild-type and Nfatc1−/− FLDCs . In Nfatc1−/− FLDCs , T . cruzi-mediated enhancement of CD40 , CD86 , and MHC class II was dramatically reduced ( Figure 7C ) . The impaired surface expression of these molecules in Nfatc1−/− FLDCs was rescued by addition of exogenous IFN-γ ( Figure 7C ) . These results indicate that NFATc1 mediates T . cruzi-induced IFN-γ production and maturation of DCs .
In the present study , we analyzed TLR-independent innate immune responses against the intracellular protozoan parasite T . cruzi . T . cruzi-infected Myd88−/−Trif−/− mice displayed normal Th1 responses and normal DC maturation . A comprehensive analysis of gene expression profiles of T . cruzi-infected DCs identified IFN-γ as a TLR-independent gene which mediated DC maturation and Th1 responses even in the absence of TLR signaling . T . cruzi infection induced an increase in intracellular Ca2+ level in DCs and macrophages , which led to NFATc1 activation and IFN-γ induction in a TLR-independent manner . In Nfatc1−/− DCs , T . cruzi-induced IFN-γ production and DC maturation was impaired . These findings demonstrate that NFATc1 is responsible for TLR-independent innate immune responses during T . cruzi infection . The family of TLRs has been established to be critical for the innate recognition of T . cruzi [14] . TLR signaling pathways consist of two major components mediated by MyD88 and TRIF [43] . Myd88−/− mice show a high susceptibility to T . cruzi infection [15] , [16] , while mice deficient in both MyD88 and TRIF are even more susceptible to T . cruzi infection [17] . These findings indicate that TLR-dependent recognition of T . cruzi is crucial to the host defense against the parasite . In this regard , TLR-dependent induction of IFN-β might be responsible for high susceptibility to T . cruzi infection in Myd88−/−Trif−/− mice in spite of the normal Th1 responses [17] . A previous study showed the MyD88-dependent IFN-γ production in T . cruzi-infected mice [16] . However , surprisingly , we found that Myd88−/−Trif−/− mice exhibited normal Th1-dependent IFN-γ production . Discrepancy between both studies might be due to distinct experimental protocols . IFN-γ is known to facilitate IL-12 production . Indeed , IL-12p40-deficient mice were highly susceptible to T . cruzi infection with severely reduced Th1 responses [33] , [34] . In T . cruzi-infected Myd88−/−Trif−/− mice , IL-12p40 production was severely reduced , but still induced [17] , suggesting that IL-12 is produced via TLR-dependent and -independent pathways . Considering that T . cruzi-infected Ifngr1−/− mice showed decreased level of serum IL-12p40 and that T . cruzi-infected Nfatc1−/− FLDCs exhibited reduced expression of IL-12p40 , NFATc1-dependent IFN-γ production may facilitate IL-12p40 production . Alternatively , the direct involvement of NFATc1 in activation of IL-12p40 gene has been also shown [23] . Collectively , T . cruzi infection might cause not only the TLR-dependent IL-12p40 production , but also the NFATc1-mediated ( TLR-independent ) production of IFN-γ and IL-12p40 , coordinating the host Th1 response . IFN-γ was identified as a gene induced in T . cruzi-infected Myd88−/−Trif−/− DCs . IFN-γ production by DCs was first demonstrated in IL-12-stimulated CD8α+ lymphoid DCs [44] . Subsequently , CD11clowB220+NK1 . 1+ cells were shown to produce high amounts of IFN-γ in response to IL-12 or a TLR9 ligand [45] , [46] . These CD11clowB220+NK1 . 1+ cells have been shown to be a subset of NK cells [29]–[31] . Thus , IFN-γ production from DCs as well as Mφ is controversial [47] , [48] . In order to exclude the possibility that our DC or Mφ preparations were contaminated by NK cell subsets , we isolated CD11chighB220−NK1 . 1− cells and analyzed IFN-γ production . These cells showed a severely reduced level of IL-12/IL-18-induced IFN-γ production compared with NK cells . In addition , IL-12/IL-18-induced IFN-γ expression was less than T . cruzi-induced expression in these cells ( our unpublished data ) . In Nfatc1−/− FLDCs , IL-12/IL-18-induced IFN-γ expression was not impaired ( our unpublished data ) . These results indicate that T . cruzi-induced IFN-γ production in DCs is mediated by a pathway distinct from the IL-12 ( or the TLR9 ligand ) -induced one in NK subsets . Protozoan parasites including T . cruzi require Ca2+ for their survival within the host cells [19] . In addition , T . cruzi evokes elevation of intracellular Ca2+ concentration in the host cells to establish the invasion [49] , [50] . Our findings indicate that T . cruzi-induced activation of host Ca2+ signaling mediates IFN-γ production . In the host cells , the family of NFAT transcription factors , which is activated by calmodulin/calcineurin , is known to bridge Ca2+ to promote gene expression [51] . The role of NFAT proteins has been well characterized in T lymphocytes , and can induce activation of the Ifng gene [22] , [52] . However , the role of NFAT proteins in innate immune cells remains unclear . Several reports indicate that NFAT proteins are activated in macrophages [24] , [53] . In addition , cyclosporin A , which blocks calcineurin-dependent NFAT activation , has been shown to inhibit DC functions [54] , [55] . In accordance with these reports , in the present study NFATc1 was activated in Myd88−/−Trif−/− Mφ in response to T . cruzi infection . Furthermore , analysis using FLDCs derived from Nfatc1−/− embryos demonstrated that NFATc1 mediates T . cruzi-induced IFN-γ production and DC maturation . These findings establish a new signaling pathway mediating an innate immune response during T . cruzi infection . It is well established that the TLR-dependent pathway initiates innate immune responses against pathogens . In addition , in the case of invasion of protozoan parasites triggering activation of intracellular Ca2+ signaling , NFATc1 mediates the TLR-independent innate immune responses through induction of IFN-γ . Bradykinin B2 receptor has been shown to mediate T . cruzi-dependent generation of inositol 1 , 4 , 5-trisphosphate , leading to elevated level of intracellular Ca2+ [56] . However , several other mechanisms that induce intracellular Ca2+ influx have been proposed in T . cruzi infection [18] . Identification of critical molecules that lead to NFATc1 activation during T . cruzi infection would be a future issue to be addressed . It would be also interesting in the future to analyze whether the NFAT pathway is involved in innate immune responses against other protozoan parasites such as Toxoplasma and Leishmania species . In this study , we focused on DCs and Mφ , which initiate Th1 responses . However , in an in vivo condition , T . cruzi are expected to invade into several other types of cells than DCs and Mφ . Therefore , it is possible that the invasion of T . cruzi into cells of non-innate immune cell populations indirectly influences Th1 polarization in vivo . In the future , we should analyze whether the NFAT family of transcription factors is involved in these processes . In summary , in the present study we revealed a new TLR-independent mechanism for the interaction between protozoan parasites and host innate immunity . Ca2+ is critical for both living organisms , and therefore the parasite utilizes host Ca2+ for its benefit . On the host side , Ca2+ signaling leads to activation of NFATc1 to eliminate the parasite . It would be interesting in the future to analyze the precise role of NFATc1 in protozoan parasite infection using innate immune cell-specific NFATc1-deficient mice .
All animal experiments were conducted in accordance with guidelines of the Animal Care and Use Committee of Osaka University and Kyushu University . Myd88−/− , Trif−/− , Ifngr1−/− , and Nfatc1−/− mice were generated as previously described [17] , [42] . Each mouse strain was backcrossed to C57BL/6 for at least five generations , and then used to generate double or triple-mutant mice . PE-conjugated anti-CD11c , PE-conjugated anti IFN-γ , APC-conjugated anti-CD11c , APC-conjugated anti-CD40 , FITC-conjugated anti-NK1 . 1 , FITC-conjugated anti-CD40 , FITC-conjugated anti-CD86 , FITC-conjugated anti-I-Ab and Pacific Blue-conjugated anti-B220 antibodies were purchased from BD Pharmingen . Anti-NFATc1 and anti-β actin antibodies were purchased from Santa Cruz . Ca2+ ionophore A23187 ( C7522 ) , PMA ( P1585 ) , and FK506 ( F4679 ) were purchased from Sigma . Fluo-4 AM was purchased from Invitrogen . BAPTA-AM was from Calbiochem . Cycloheximide was from Nacalai tesque . To isolate peritoneal Mφ , mice were i . p . injected with 2 ml of 4% thioglycolate medium ( Sigma ) , and peritoneal exudate cells were isolated from the peritoneal cavity at three days post injection . The cells were incubated for 2 h , then washed three times with HBSS . The remaining adherent cells were used as peritoneal Mφ in experiments . To prepare bone marrow-derived DCs or Mφ , bone marrow cells were prepared from the femur and tibia , and cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum ( FBS ) , 100 µM 2-mercaptoethanol ( 2ME ) , and 10 ng/ml GM-CSF ( Pepro Tech ) or 30% L cell culture supernatant , respectively . After six days , the cells were used as bone marrow DCs or bone marrow Mφ in experiments . To prepare fetal liver-derived DCs ( FLDCs ) , fetal liver ( FL ) were obtained from 12 . 5 days post-coitum murine embryos , and FL cells were dissociated by pipetting , passed through a nylon mesh , and then cultured in RPMI 1640 medium supplemented with 10% FBS , 100 µM 2ME , 20 ng/ml GM-CSF , 10 ng/ml Flt3 ligand ( Pepro Tech ) , and 10 ng/ml SCF ( Pepro Tech ) as described [57] . After eight days , the floating cells were used as FLDCs in experiments . RAW 264 . 7 cells were transfected with the NFATc1 ( pcDNA3 ) expression plasmid . The cells expressing NFATc1 were selected in the presence of 0 . 4 mg/ml G418 and cloned . Splenocytes from wild-type mice were irradiated ( 30 Gy ) and used as APC . The trypomastigote stage of T . cruzi Tulahuen strain was maintained in vivo in Ifngr1−/− mice by passages every other week or in vitro in LLC-MK2 cells by passages every four days . For in vitro experiments , 5×105 Mφ or DCs were infected with 1 . 5×106 trypomastigotes . For in vivo experiments , mice were i . p . injected with 6×101 trypomastigotes or PBS . Epimastigotes of Tulahuen strain were grown at 26°C in liver infusion tryptose liquid medium , supplemented with 2 . 5% hemoglobin and 10% fetal calf serum . Total RNA was isolated with TRIzol reagent ( Invitrogen ) , and 1–2 µg of RNA was reverse transcribed using M-MLV reverse transcriptase ( Promega ) and random primers ( Toyobo ) after treatment with RQ1 DNase I ( Promega ) . Real-time RT-PCR was performed on an ABI 7300 ( Applied Biosystems ) using the TaqMan Universal PCR Master Mix ( Applied Biosystems ) . All data were normalized to the corresponding gene Eef1a1 encoding elongation factor-1α ( EF-1α ) expression , and the fold difference relative to the EF-1α was shown . Amplification conditions were: 50°C ( 2 min ) , 95°C ( 10 min ) , 40 cycles of 95°C ( 15 s ) , and 60°C ( 60 s ) . Primers of Tbx21 , Stat1 , Irgm , and Tnf were purchased from Assay on Demand ( Applied Biosystems ) . Sequences for EF-1α , Il12b , Il6 , and Ifng are as follows: EF-1α probe 5′-gcacctgagcagtgaagccagctgct-3′ . forward primer 5′-gcaaaaacgacccaccaatg-3′ . reverse primer 5′-ggcctggatggttcaggata-3′; Il6 probe 5′-ccttcttgggactgatgctggtgaca-3′ . forward primer 5′-ctgcaagagacttccatccagtt-3′ . reverse primer 5′-aagtagggaaggccgtggtt-3′; and Ifng probe 5′-gtcaccatccttttgccagttcctccag-3′ . forward primer 5′-tcaagtggcatagatgtggaagaa-3′ . reverse primer 5′-tggctctgcaggattttcatg-3′ . Splenic cells were isolated from T . cruzi-infected mice at the indicated time point and stimulated with PMA and ionomycin for 4 h in the presence of 10 µg/ml brefeldin A . In experiments to detect IFN-γ production from CD11c+ cells , splenic cells were infected with T . cruzi ( 1∶1 ) for 12 h , and further cultured for 6 h in the presence of 10 µg/ml brefeldin A . After staining of surface CD11c , CD4 , CD8 or NK1 . 1 , the cells were fixed with CytopermCytofix ( BD Biosciences ) for 20 min and incubated with PE-conjugated anti-IFN-γ Ab . Flow cytometric analysis was performed on FACSCantoII ( BD Biosciences ) . For in vivo experiments , mice were i . p injected with T . cruzi , and CD4+ T cells were isolated from the spleen at the indicated days after the infection . 2 . 5×105 CD4+ T cells were stimulated with anti-CD3 Ab or freeze-thawed T . cruzi in the presence of 2 . 5×105 APC for 24 h . The culture supernatants were collected and diluted at 1∶5 . ELISA was performed with anti-mouse IFN-γ Ab , avidin-HRP , and TMB solution purchased from eBioscience . Optical densities were determined at 450 nm wavelengths with reference at 570 nm . Levels of IFN- γ were calculated from the standard curve by using purified mouse IFN- γ purchased from eBioscience . Bone marrow DCs or FLDCs were infected with T . cruzi for 6 h , washed and then cultured for 24 or 48 h . The T . cruzi-infected cells were stained with the combination of PE-conjugated anti-CD11c and the indicated antibodies at 4°C for 20 min , and washed . Flow cytometric analysis was performed on FACSCalibur or FACSCant II flow cytometer ( BD Biosciences ) and using FlowJo software ( Tree Star ) . CD11chigh cells and NK cells were sorted using FACS Aria ( BD Biosciences ) . The instrumental compensation was set in each experiment using single color , 2-color or 4-color stained samples . This assay was performed as described [58] . In brief , bone marrow Mφ plated on glass-bottom dishes were incubated in serum-free RPMI 1640 supplemented with 2 µM Fluo-4 AM , the increase in fluorescent intensity of which indicates increased Ca2+ level , at 37°C for 30 min . The cells were then washed to remove the free extracellular dye , and were maintained in culture medium during the whole experiment . The analysis of changes of basal intracellular calcium concentrations in response to T . cruzi infection was performed using an IX71 fluorescence microscope ( Olympus ) . Bone marrow Mφ were transfected with pcDNA3-NFATc1 by nucleofection ( mouse macrophage nucleofector kit; Amaxa ) . After 24 h , the cells were infected with T . cruzi for 30 min , washed with Tris-buffered saline ( TBS ) , and then fixed with 3 . 7% formaldehyde in TBS for 15 min at room temperature . After permeabilization with 0 . 2% Triton X-100 , cells were washed with TBS , incubated with anti-NFATc1 antibody in TBS containing 1% bovine serum albumin , then incubated with Alexa Fluor 594-conjugated goat anti-mouse immunoglobulin G ( Molecular Probes ) . To stain the nucleus , cells were cultured with 0 . 5 mg/ml 4 , 6-diamidino-2-phenylindole ( DAPI; Wako ) . Stained cells were analyzed using an LSM510 confocal microscope ( CarlZeiss ) . RAW 264 . 7 cells were transfected with the indicated expression plasmids together with the reporter plasmid IFN-γ-Luc and the internal control plasmid phRG-TK by Nucleofection ( Nucleofector Kit V; Amaxa ) . After 18 h , the cells were infected with T . cruzi for 18 h , and the luciferase activities of whole cell lysates were measured using the Dual-luciferase reporter assay system ( Promega ) and Lumat LD 9507 ( Berthold ) . Differences between control and experimental groups were evaluated by the Student's t-test .
|
Trypanosoma cruzi is an intracellular protozoan parasite that causes Chagas diseases in humans . Invasion of T . cruzi into the host is sensed by Toll-like receptors ( TLRs ) , which recognize microbial components that are present in microbes but not in the host . TLRs are essential for the initiation of immune responses against pathogens . Recent evidence indicates the presence of TLR-independent mechanisms for the recognition of microbes , such as bacteria , viruses , and fungi . However , TLR-independent recognition of protozoa remains unknown . We found that immune responses against T . cruzi were induced even in the absence of TLR signaling . The TLR-independent responses were found to be mediated by IFN-γ production in innate immune cells . Furthermore , the TLR-independent IFN-γ production was revealed to be mediated by Ca2+-dependent activation of NFATc1 , which has been shown to play a pivotal role in cytokine production in T lymphocytes . Our study provides a novel mechanism for the TLR-independent innate immune response against protozoan parasites . It is also worth noting that the host defense mechanism utilizes a factor ( Ca2+ ) that is a prerequisite for the survival of intracellular protozoan parasites .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"infectious",
"diseases/protozoal",
"infections",
"immunology/innate",
"immunity"
] |
2009
|
NFATc1 Mediates Toll-Like Receptor-Independent Innate Immune Responses during Trypanosoma cruzi Infection
|
Animal circadian clocks are based on multiple oscillators whose interactions allow the daily control of complex behaviors . The Drosophila brain contains a circadian clock that controls rest–activity rhythms and relies upon different groups of PERIOD ( PER ) –expressing neurons . Two distinct oscillators have been functionally characterized under light-dark cycles . Lateral neurons ( LNs ) that express the pigment-dispersing factor ( PDF ) drive morning activity , whereas PDF-negative LNs are required for the evening activity . In constant darkness , several lines of evidence indicate that the LN morning oscillator ( LN-MO ) drives the activity rhythms , whereas the LN evening oscillator ( LN-EO ) does not . Since mutants devoid of functional CRYPTOCHROME ( CRY ) , as opposed to wild-type flies , are rhythmic in constant light , we analyzed transgenic flies expressing PER or CRY in the LN-MO or LN-EO . We show that , under constant light conditions and reduced CRY function , the LN evening oscillator drives robust activity rhythms , whereas the LN morning oscillator does not . Remarkably , light acts by inhibiting the LN-MO behavioral output and activating the LN-EO behavioral output . Finally , we show that PDF signaling is not required for robust activity rhythms in constant light as opposed to its requirement in constant darkness , further supporting the minor contribution of the morning cells to the behavior in the presence of light . We therefore propose that day–night cycles alternatively activate behavioral outputs of the Drosophila evening and morning lateral neurons .
Circadian rhythms are controlled by endogenous clocks that tick with an approximately 24-h period fitted to the rotation of the earth . They are synchronized to day–light cycles by environmental cues , the strongest of which is light . Since activity must occur at the most favorable time of the day , the rest–activity rhythm is one of the most tightly clock-controlled behaviors . In natural conditions , many animal species display bimodal rest–activity profiles with activity peaks that anticipate dawn and dusk , and adjust to seasonal changes in day length [1 , 2] . A similar activity pattern is observed in laboratory light–dark ( LD ) conditions . A lengthening of the light episode induces a morning-peak advance and an evening-peak delay in mice , suggesting the existence of morning and evening oscillators in the mammalian brain that contribute to seasonal adaptation [3] . The cellular basis of such a dual oscillator has not been characterized in mammals , but has been recently described in Drosophila . The Drosophila behavioral clock rests upon approximately 150 neurons that express the PERIOD ( PER ) protein , divided into three lateral and three dorsal groups , as well as a recently described lateral-posterior group [4–6] . The lateral neurons ( LNs ) can be divided into cells that express the pigment-dispersing factor ( PDF ) neuropeptide , and PDF-negative cells . The PDF-expressing cells are four to five large ventral lateral neurons ( l-LNvs ) and four small ventral lateral neurons ( s-LNvs ) , whereas the PDF-negative cells are a single s-LNv ( the fifth s-LNv ) and six dorsal lateral neurons ( LNds ) . In LD cycles , PER expression in the four PDF-expressing s-LNvs is sufficient to drive activity that anticipates lights-ON , and hence these cells contain a morning oscillator ( MO ) , whereas the addition of four PDF-negative LNs ( fifth s-LNv plus three LNds ) is sufficient to drive lights-OFF anticipation; hence , the latter cells contain an evening oscillator ( EO ) [7] . Another group reported similar results [8] . They additionally indicated that dorsal neurons ( DNs ) could contribute to both the MO and EO . We will therefore specifically refer to the morning oscillator residing in PDF-positive LNs as the LN-MO and to the evening oscillator residing in the PDF-negative LNs as the LN-EO . We have previously shown that , in constant darkness ( dark–dark; DD ) , clock function restricted to the LN-MO is sufficient to generate robust 24-h activity rhythms , whereas clock function in the LN-EO is not [7] . This suggested that , in the absence of light , the LN-MO is the driving oscillator of the circadian network . Indeed , it has been shown that at least part of the LN-EO behaves in DD as a driven oscillator , reset by the LN-MO in each circadian cycle [9] . Circadian clocks are very sensitive to light and respond to it in different ways . First , light is the main clock synchronizer , and LD cycles entrain the Drosophila brain clock through two separate light-input pathways . The blue-light–sensitive protein cryptochrome ( CRY ) is present in most clock neurons [10 , 11] . Light-activated CRY binds to the TIMELESS ( TIM ) protein and induces its degradation , which is likely to reset the molecular oscillator [12–15] . cryb mutants do not respond to short light pulses and fail to quickly resynchronize to a shift of the LD cycle [10 , 11 , 16–18] . The cryb mutation is located in the flavin-binding domain and certainly abolishes CRY photoreceptive function [17] . Although the CRYb protein is barely detectable by anti-CRY antibodies [17] , very low amounts may still be present and play some non-photoreceptive function in the mutants . The visual system , which includes the compound eye and the extra-retinal Hofbauer-Büchner eyelet , provides additional rhodopsin-dependent light inputs to the brain clock [19–21] . They are not sufficient for clock responses to short light pulses , but allow entrainment by LD cycles ( although less efficiently than CRY ) . Only flies depleted for both functional CRY and the visual system appear circadianly blind [11 , 18] . Besides entrainment , light affects other parameters of the circadian system , including its internal synchrony as well as the robustness of the rhythm and its period [22] . In constant light ( light–light; LL ) , wild-type flies become arrhythmic , whereas cryb mutants retain robust rhythmicity with a 24–26-h period [11 , 14 , 15 , 18 , 23 , 24] , presumably because the absence of functional CRY prevents the light-induced disappearance of TIM in the mutants . Indeed , mutations affecting the CRY-dependent degradation of the TIM protein also produce robust 24–26-h activity rhythms in LL [24 , 25] . Two studies reported that cryb mutants display split rhythms in LL , with a major long-period ( ∼25 h ) component and a minor short-period ( ∼22 . 5 h ) one [26 , 27] . These slow and fast components appear to correlate with molecular oscillations in some of the PDF-negative LNs and in the PDF-positive s-LNvs , respectively [27] , suggesting that they may originate from these subsets . Genetic background , light specifications , and behavioral setup are likely to influence splitting occurrence , but the main reason why split rhythms have only been observed in these two studies is likely related to their longer activity recordings , since split components usually appear after several days in LL [26 , 27] . The present work is aimed at understanding how the previously defined LN-MO and LN-EO control rhythmic behavior in the presence or the absence of light . We have generated flies that were mosaic with respect either to CRY signaling or to the presence of a functional clock . In particular , we altered functional CRY levels separately in either PDF-positive or PDF-negative neurons . We similarly restored PER expression in per0;; cryb double mutants only in precisely targeted neurons . The results indicate that light has opposite effects on the LN-MO and LN-EO , activating the rhythmic behavioral output induced by the evening cells and inhibiting the rhythmic behavioral output induced by the morning cells . Surprisingly , we found that light acts downstream from the molecular clock , since the behavior , but not the molecular oscillations , is light-dependent . We also show that cryb pdf0 double mutants are rhythmic in LL , further supporting the light-induced preeminence of PDF-negative cells .
We first analyzed PER oscillations in cryb mutants in LL , under conditions in which split behavioral rhythms do not occur ( see Materials and Methods ) . As previously described ( see above ) , the mutants displayed a slightly lengthened period ( Tables 1 and S1 ) . In cryb brains dissected on the third day in LL , the PDF-positive s-LNvs and some PDF-negative LNs showed PER cycling , whereas the l-LNvs and three subsets of DNs did not ( unpublished data ) . This is very similar to the molecular oscillations described by Rieger et al . [27] for cryb mutants in LL , before splitting would eventually occur . Since PER cycling in LL appeared to be restricted to the PDF-positive and PDF-negative LNs , we decided to focus our study on these groups of clock neurons . In addition , we decided to center the study upon the effect of light on the rhythmicity of the two LN oscillators , and we voluntarily put aside the role of cryptochrome and the visual system in their entrainment pathways . To first clarify the heterogeneity of the LNds group ( see also [27] ) , we examined the Mai179-Gal4–driven green fluorescent protein ( GFP ) expression profile , which includes the previously characterized LN-EO [7] ( see Figure S1 ) . In LL , PER cycling was detected in all four LN-EO neurons ( Figures 1A and 2A ) . The fifth s-LNv and the previously described [27] cycling LNd ( called here LNdM* ) displayed the strongest oscillations , but the two other Mai179-Gal4–positive LNds ( LNdMs ) also showed robust , although slightly delayed , oscillations ( trough at circadian time [CT]68 instead of CT64 ) . Conversely , constant PER levels were observed in the three Mai179-Gal4–negative LNds ( LNdOs; Figures 1A and 2A ) . Interestingly , CRY immunoreactivity was detected in the three Mai179-Gal4–positive LNds ( one LNdM* + two LNdMs ) , but not in the Mai179-Gal4–negative LNdOs in DD ( Figure 1B and 1C ) . These data strongly support the existence of two LNd subgroups: three Mai179-Gal4–expressing CRY-positive cells constituting the LN-EO with the fifth s-LNv , and three Mai179-Gal4- and CRY-negative cells , whose function is unknown . We then checked whether the presence of functional CRY affects PER expression in an oscillator-autonomous manner in LL , using the two most strongly cycling Mai179-Gal4–expressing PDF-negative LNs as reporters for the LN-EO . The main additional Gal4 lines we used here were pdf-Gal4 [28] to drive expression in the PDF-positive cells only , and tim-Gal4 [29] to drive expression in all clock cells . The pdf-Gal80 transgene was used to inhibit GAL4 activity in the PDF-positive cells , and thus “subtract” their contribution from any wider GAL4-expressing cell ensemble [8] . As in wild-type flies , PER levels remained low or undetectable in all cells that contained functional CRY , and PER oscillations were observed exclusively in some of the cells expressing either strongly reduced CRY levels ( through cry RNA interference [RNAi] ) or the mutated CRYb protein ( Figures 2B–2E and S2 ) . The four PDF-positive s-LNvs and the two selected PDF-negative LNs displayed oscillations whenever they were made CRY deficient ( Figures 2B and S2 ) . We conclude that , when functional CRY is reduced or absent , PER oscillations in LL persist in the previously characterized LN-MO and LN-EO . We then analyzed the behavior of flies with PER oscillations in either the PDF-expressing or the PDF-negative neurons in LL . Genotypes with CRY only ( and consequently no PER ) in PDF-expressing cells were almost as rhythmic as cryb mutants ( Figure 3 and Table 1; see also Table S1 ) , with a consistently long period . Contrary to DD , the PDF-negative cells can therefore drive behavioral rhythms autonomously in LL , in the absence of any PER oscillations in the PDF-positive LNvs . Conversely , flies with CRY only ( and consequently no PER ) in PDF-negative cells are mostly arrhythmic ( Table 1 ) , despite robust PER oscillations in their PDF-positive s-LNvs ( Figure 2C and 2D ) . This demonstrates that the four LN-MO neurons cannot drive robust behavioral rhythms autonomously in LL , as opposed to their ability to do so in DD . PER oscillations persist in such flies at least up to the fifth day in LL ( Figure 2G ) , whereas their behavior becomes arrhythmic within the very first days ( Figure 3 ) . We thus conclude that in cryb flies , constant light appears to inhibit the behavioral output of the LN-MO , but not the molecular oscillator itself . To understand whether light inputs coming from the visual system participate to the LL behavioral rhythms of cryb flies , we induced its genetic ablation by expressing the apoptotic gene head involution defective ( hid ) under the control of photoreceptor-specific regulatory sequences . The GMR-hid strain [30] was previously shown to completely lack all visual glass gene-dependent structures , but to retain the glass-dependent set of DN1s that express PER in the adult brain [11] . GMR-hid–induced ablation of the visual system restored the behavioral function of the PDF-expressing neurons in LL , now driving rhythms with a 24 . 4-h period ( Figure 3 and Table 1 ) , although there was no detectable change in PER oscillations ( compare Figure 2D and 2F ) . This indicates that the inhibition of the LN-MO behavioral output by light depends on the visual system . What is the neuronal basis of the long-period , LN-MO–independent rhythmicity of cryb flies in LL ? Restoring CRY in only the LN-EO ( three LNds plus the PDF-negative fifth s-LNv ) ( Figures 2E and S2 ) rendered the flies as arrhythmic as the wild type ( Figure 3 and Table 1 ) . Thus the LN-EO is necessary for that long-period rhythmicity . Indeed , per0 ;; cryb double mutants with Mai179-Gal4–driven PER expression restricted to the LN-EO displayed robust activity rhythms , with a long 25 . 7-h period ( Figure 3 and Table 1 ) . In these LN-EO–only flies , PER levels robustly cycled in all four neurons , displaying a trough after 65 h rather than 60 h in LL , consistent with a period close to 25 . 5 h rather than 24 h ( Figures 4A and S3 ) . A similar behavior was obtained with the cry-Gal4–19 driver ( Table 1 ) , which gives a PER expression pattern very close to Mai179-Gal4 ( Figure S1 ) . The LN-EO is thus not only necessary , but also sufficient to drive rhythmic behavior in LL , whereas it is not sufficient in DD ( [7] and Table 1 ) . However , genotypes with PER cycling in the LN-MO neurons in addition to the LN-EO neurons displayed a slightly shorter period than flies with PER in the LN-EO neurons only ( Table 1 ) , suggesting that the LN-MO somehow influences the period of the LN-EO and therefore participates in the LN-EO–driven rhythmic behavior . Interestingly , long-period PER oscillations in the LN-EO neurons were observed in DD ( trough after 64 h in Figure 4B or between 64 and 72 h in Figure 4C , to be compared with 60 h in Figure4D; see also Figure S3 ) , similarly to LL , although such LN-EO–only flies were behaviorally arrhythmic , contrary to LN-MO–only flies ( Table 1 ) . We conclude that in the absence of light , the LN-EO is running at the molecular level , but that its behavioral output is inhibited since it cannot drive activity rhythms . Since PDF is required for robust behavioral rhythmicity in DD [28] , we asked whether rhythmicity in LL would also depend on PDF signaling . We therefore constructed cryb pdf0 double mutants and tested them in LL . Such flies indeed displayed strong rhythmicity ( Figure 5 and Table 2 ) , similar in robustness to that of cryb mutants ( see high power values in Tables 1 and 2 ) , but with a short 22 . 8-h period . We then analyzed PER oscillations in the double mutants in LL . PER cycling in the EO neurons was in good agreement with the short-period behavior , whereas PER cycling in the MO neurons was not ( Figure S4 ) . These data strongly suggest that the EO neurons drive LL activity rhythms in the cryb pdf0 flies , whereas the robustly cycling MO neurons do not contribute significantly to the PDF-independent LL behavior . We conclude that the LN-EO does not require PDF to generate behavioral rhythms in LL , although PDF strongly influences its period . Conversely , the double mutants were mostly arrhythmic in DD ( Table 2 ) , with a fraction of the flies displaying a weak short-period rhythmicity as reported for pdf0 mutants in DD [11 , 28 , 31] . The rhythmicity of pdf0 mutants was not improved in LL , showing that the strong rhythmic behavior of the double mutants in LL results from the cryb mutation .
The PDF-expressing LNs and the PDF-negative LNs were previously characterized as morning and evening cells , respectively , in LD conditions [7 , 8] . Furthermore , the morning LNs were able to drive robust 24-h rhythms in DD , whereas evening LNs were not [7] . We show in this study that in LL , the evening LNs drive robust rhythms when cryptochrome signaling is absent or reduced , whereas the morning cells are not able to do so . Surprisingly , the molecular oscillations of both groups can be uncoupled from behavioral rhythmicity , depending on light conditions . In DD , the two LN groups show autonomous molecular cycling , but there is no behavioral output when the LN-EO is cycling alone . In LL ( and reduced CRY signaling ) , both groups still show autonomous cycling , but there is no behavioral output when the LN-MO is cycling alone . We therefore conclude that light has opposite effects on the behavioral output of the two LN oscillators , activating it from the evening LNs and inhibiting it from the morning LNs . The opposite effects of light on the behavioral outputs do not appear to be related to entrainment , since PER oscillations in both the PDF-positive and PDF-negative LNs are synchronized to the LD cycles even in the absence of CRY signaling . The inhibiting effect of light on the LN-MO behavioral output is abolished when the visual system is genetically ablated . This suggests that the projections of the visual system photoreceptors convey , not only input information to the PDF cells ( light entrainment ) , but also signals to control their behavioral output ( light inhibition ) . It is tempting to speculate that light exerts both effects through a direct connection of the PDF cells with the visual system . The Hofbauer-Büchner eyelet photoreceptors that project directly to the LN-MO neurons and participate in the entrainment [19 , 20] provide a possible pathway . It was recently reported that the overexpression of PER [32] or of the SHAGGY ( SGG ) kinase [33] in the PDF-negative clock neurons induced rhythmic behavior in LL . The rhythmicity was associated with the cycling of PER subcellular localization in some of the DNs , whereas the PDF-expressing cells were molecularly arrhythmic . These studies therefore concluded that some DN subsets are able to drive behavioral rhythms in LL . Different groups of PDF-negative cells may thus be able to drive behavioral rhythms in constant light , depending on whether and how the molecular clock is manipulated . Such manipulation could also directly affect molecular oscillations , making them less easy to detect . Since CRY does not appear to have a core clock function in the brain , our data are largely based on situations in which the clock mechanism is little if at all altered . The data support a major contribution of the LN-EO to the robust rhythms of cryb mutants in LL . The strong rhythmicity of the cryb pdf0 double mutants in LL contrasts with their weak rhythmic behavior in DD . Altogether , our results strongly suggest that this robust rhythm is generated by the LN-EO , which would therefore behave as a PDF-independent autonomous oscillator . However , the period of the oscillator is clearly influenced by PDF signaling , and thus by the LN-MO , going from 24–25 h in cryb to 22–23 h in cryb pdf0 flies . An attractive possibility is that the strong short-period rhythm observed in the cryb pdf0 double mutant in LL has the same neuronal origin as the weak short-period rhythm described for pdf0 mutants in DD [28] . The cellular basis of this PDF-independent oscillator in DD remains unclear [11 , 31 , 34] , although the presence of similar rhythms in flies genetically ablated for the PDF-expressing neurons [28 , 35] suggests that it originates from other clock cells . Different results were obtained for the recently described DN-based LL oscillators . When transferred to a pdf0 background , all SGG-overexpressing flies were found to be arrhythmic [33] , whereas about 60% of the PER-overexpressing flies displayed long-period rhythms [32] . This suggests that different types of DNs with different sensitivity to PDF may have been analyzed in these two studies . Although some DNs may contribute to the PDF-independent rhythms , our data suggest a strong contribution of PDF-negative LNs to the rhythmic behavior that persists in pdf0 mutants . The weakness of the short-period rhythm of pdf0 flies in DD may reflect the inhibition of the LN-EO output in the absence of light . Our results indicate that whereas the LN-MO autonomously drives rhythmic behavior in constant darkness , the LN-EO plays this role in constant light , if CRY signaling is abolished or reduced . We thus suggest that in natural LD conditions , Drosophila behavior could be driven by the LN-MO during the night , and by the LN-EO during the day , when cryptochrome is quickly degraded by light . This supports a model of a light-induced switch between the circadian oscillators of the LNs ( Figure 6 ) that would allow a better separation of the dawn and dusk activity peaks in day–night conditions . It has been shown that PDF-expressing LNs drive the clock neuronal network in short days , whereas PDF-negative DN subsets take the lead in long days [33] . Our results suggest that the PDF-negative cells of the LN-EO could also be a major player during the long days . Surprisingly , we find that light does not seem to act on the molecular oscillations , but inhibits the LN-MO behavioral output and promotes the LN-EO behavioral output , which may provide an efficient fine tuning of the contributions of the two oscillators . It therefore appears that the visual system controls both the input ( entrainment ) and the behavioral output of the LN oscillators in the Drosophila brain clock . In species such the honeybee or the flour beetle , which appear to lack a light-sensitive CRY protein [36 , 37] , this role of the visual system may be particularly important .
The cryRNAi construct produces a double-stranded RNA ( dsRNA ) that corresponds to the 300–799 region of the cry-RA mRNA ( see http://flybase . bio . indiana . edu/reports/FBgn0025680 . html ) . The primers used for PCR were: 5′ primer: AAGGCCTACATGGCCGGACCGATGTGGGTTACAATCGGATGC 3′ primer: AATCTAGAGGTACCGAAGCCCATGTTGTCTCCATA . The 500-bp DNA fragment was inserted into the pUAST-R57 vector as described here: http://www . shigen . nig . ac . jp/fly/nigfly/about/aboutRnai . jsp . Two UAS-cryRNAi insertions were generated , and the line with the strongest expression ( R3 ) was used in this study . When combined with UAS-cry and the pdf-Gal4 driver , this UAS-cryRNAi insertion reduced CRY levels by at least 80% ( unpublished data ) , as judged by immunocytofluorescence with anti-CRY . The UAS-cry [10] and UAS-per16 [35] insertions have been described previously . tim-Gal4 is expressed in all clock neurons in addition to several non-clock neuronal groups [29] , and pdf-Gal4 is specifically expressed in the PDF-positive LNvs [28] . The cry-Gal4–19 insertion was generated by jumping out the P element of the original cry-Gal4 insertion [10] . It has a more restricted expression pattern than the previously described cry-Gal4–39 insertion [11] . The expression patterns of cry-Gal4–19 and Mai179-Gal4 ( see also [7] ) are described in Figure S1 . We used the pdf-Gal80 line 96A , which contains two insertions and completely abolishes pdf-Gal4-driven expression in the PDF-positive LNvs [8] . Experiments were carried out with 1–7-d-old flies at 20 °C in Drosophila activity monitors ( TriKinetics ) as previously described [38] . Light was provided by standard , white-fluorescent low-energy bulbs . Light intensity at fly level was in the range of 300–1 , 000 μW/cm2 , depending on the position of the monitor in the incubator . For LL and DD analysis , flies were first entrained in 12 h:12 h LD cycles during at least 4 d , and activity data were analyzed for 6 d , starting from the second day in DD or in LL . Under these LL conditions , cryb mutants displayed robust activity rhythms , and no split rhythms could be observed . Data analysis was done with the FaasX 0 . 9 . 8 software , which is derived from the Brandeis Rhythm Package . FaasX runs on Apple Macintosh OSX and is freely available upon request . Rhythmic flies were defined by χ2 periodogram analysis with the following criteria ( filter ON ) : power ≥ 20 , width ≥ 2 h , with selection of 24 h ± 6 h upon period value . Power and width are the height and width of the periodogram peak , respectively , and give the significance of the calculated period . Actograms represent absolute activity levels for each 0 . 5-h interval , averaged over groups of flies of a given genotype . The hash density of the actogram ( number of activity events per hash ) varies from 15 to 35 , according to the activity level of the genotype . This allows the comparison of activity profiles between genotypes that display very different activity levels . Mean daily activity ( number of events per 0 . 5 h ± standard error of the mean [s . e . m . ] ) is calculated over the whole period of DD or LL , and is reported in Tables 1 , 2 , and S1 for all genotypes . All behavioral experiments were reproduced two or three times with very similar results . All experiments were done on whole-mounted adult brains . GFP reporter expression , anti-PER , anti-CRY , and anti-PDF labeling was done as previously described [11 , 20] . Fluorescence signals were analyzed with a Zeiss Axioplan2 epifluorescence microscope equipped with a SPOT2 ( Diagnostic Instruments ) digital camera . Fluorescence intensity was quantified from digital images with the ImageJ software . We applied the formula: I = 100 × ( S − B ) /B , that gives the fluorescence percentage above background ( where S is the fluorescence intensity , and B is the average intensity of the region adjacent to the positive cell ) . Confocal imaging was performed on a Leica SP2 microscope . Stacks of approximately 20 images were obtained , which spanned the breadth of the brain between the LNvs ( posterior ) and the DN1s ( anterior ) . Maximum intensity projections were generated from such stacks .
|
Living organisms have evolved circadian clocks that anticipate daily changes in their environment . Their clockwork is fully endogenous , but can be reset by external cues . ( Light is the most efficient cue . ) The circadian neuronal network of the fruit fly ( Drosophila ) brain perceives light through the visual system and a dedicated photoreceptor molecule , cryptochrome . Flies exhibit a bimodal locomotor activity pattern that peaks at dawn and dusk in light–dark conditions . These morning and evening activity bouts are controlled by two distinct neuronal clocks in the fly brain . By using flies with a deficient cryptochrome pathway , we have uncovered an unexpected role for light in the circadian system . In addition to synchronizing the two oscillators to solar time , light also controls their behavioral output . The morning oscillator can periodically rouse the fly when in constant darkness , but not in constant light , whereas the evening oscillator can do the same in constant light , but not in constant darkness . This suggests the existence of a light-dependent switch between oscillators that appears to require the visual system . Such a mechanism likely contributes to better separate the active periods of the fly at dawn and dusk , and may help the animal to adapt to seasonal changes in day length .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"neuroscience",
"genetics",
"and",
"genomics"
] |
2007
|
Light Activates Output from Evening Neurons and Inhibits Output from Morning Neurons in the Drosophila Circadian Clock
|
The search for genes that regulate stem cell self-renewal and differentiation has been hindered by a paucity of markers that uniquely label stem cells and early progenitors . To circumvent this difficulty we have developed a method that identifies cell-state regulators without requiring any markers of differentiation , termed Perturbation-Expression Analysis of Cell States ( PEACS ) . We have applied this marker-free approach to screen for transcription factors that regulate mammary stem cell differentiation in a 3D model of tissue morphogenesis and identified RUNX1 as a stem cell regulator . Inhibition of RUNX1 expanded bipotent stem cells and blocked their differentiation into ductal and lobular tissue rudiments . Reactivation of RUNX1 allowed exit from the bipotent state and subsequent differentiation and mammary morphogenesis . Collectively , our findings show that RUNX1 is required for mammary stem cells to exit a bipotent state , and provide a new method for discovering cell-state regulators when markers are not available .
Adult stem cells are functionally defined based on their ability to regenerate tissues . This unique regenerative ability can be recapitulated in culture models , where single stem cells , but not differentiated cells , form tissue rudiments in three-dimensional extracellular matrices . These tissue rudiments , or organoids , exhibit many of the topological , functional and phenotypic traits of the corresponding tissue . For example , mammary stem cells form ducts and lobules in collagen matrices that resemble structures present in the breast [1–3] , while colon stem cells form mini-crypts in Matrigel that resemble analogous structures in the small intestine [4] . Given their potential for regenerative medicine , there is significant interest in identifying genes that regulate self-renewal or differentiation of stem cells . In systems with well-defined markers of stem , progenitor and differentiated states , this can be accomplished by inhibiting candidate genes and assessing the resulting effects on cell state proportions [5] . However , for many tissues markers of stem cells and early progenitors are not available , and even in cases where such markers are available they often only enrich for states of interest . This lack of defining markers has complicated efforts to screen for cell-state regulators , because changes in the number of cells expressing an enriching marker may not quantitatively reflect changes in the stem or progenitor cell types of interest . We have addressed this difficulty by developing a new approach that identifies cell state regulators without requiring defining markers of cell state , termed Perturbation-Expression Analysis of Cell States ( PEACS ) . Application of PEACS to mammary stem cells led to the discovery of a novel role for RUNX1 in exit from the bipotent state . We anticipate that PEACS will be useful in the many contexts where defining markers are not available , and have implemented the algorithm as a software tool available to the scientific community .
The analysis underlying PEACS is based on several observations . First , populations of stem cells propagated in culture are heterogeneous , and invariably include early progenitors and other more differentiated cell types . While typically considered a drawback of maintaining stem cells in culture , this heterogeneity is essential for the computational analysis underlying PEACS . Second , experimental conditions that perturb transitions between stem and progenitor states will also perturb the relative proportions of stem and progenitor cells in a heterogeneous population of cells . For example inhibiting a gene required for stem cell self-renewal will reduce the proportion of stem cells in a heterogeneous population , with a concomitant relative increase in progenitors or other more differentiated cell types . The computational challenge then is to use the population expression vectors—one for each perturbation—to infer which perturbations modulate cell-state proportions . However , without knowing either the cell state proportions or the gene-expression vectors of the individual states , it may appear that there is insufficient information to make such an inference . The solution lies in a third key observation: the gene-expression profiles ( vectors ) of heterogeneous populations of cells are weighted linear combinations of the expression profiles ( vectors ) of the component states within the population , with the weights in this linear combination corresponding to cell-state proportions . In other words , the gene-expression signal of the population is a linear mixture of component signals , the latter of which are unknown . The key is to deconvolute this signal ( Fig 1 ) . Several computational algorithms have been designed precisely for this purpose—to infer the constituent components of mixed signals—under the assumption that the mixed signal is a weighted linear combination of constituent components . The most commonly used algorithm to infer linear components , SVD/PCA , iteratively minimizes the reconstruction error of a mixed signal , under the constraint that the component newly identified in a given iteration be orthogonal to all of the previously identified components . Given the immense success of SVD/PCA in solving many problems across diverse fields , we decided to assess its effectiveness for our problem . A second algorithm , NMF , reconstructs mixed signals by identifying components which have only non-negative loadings . Some researchers have found this non-negative constraint to be appealing , since negative loadings of genes can be difficult to interpret biologically; for this reason we also included this method for comparison . A third algorithm , ICA , does not require that the constituent components be orthogonal to one another—and instead identifies components by maximizing their independence in a statistical sense . ICA has proven useful for deconstructing mixed signals ( e . g . , audio ) into their constituent parts . Although our goal in developing PEACS was to apply it in settings where neither the state expression vectors nor cell-state proportions are known , to assess the effectiveness of the algorithms described above ( SVD , NMF , ICA ) we needed an idealized context in which cell-state proportions could be experimentally defined . Experimentally defining cell-state proportions would make it possible to assess , for each algorithm , how well it identified changes in cell-state proportions across experimental conditions . To generate such idealized experimental conditions we mixed three different breast cancer cell lines ( T47D , SUM159 , MDA-MB-231 ) in defined proportions—for example 1:1:1 , 1:2:2 , 1:1:0—with 10 mixtures in total . In this idealized experiment the three cancer lines represented different “cell states” that were mixed in defined proportions to create heterogeneous populations ( Fig 2A; T47D = State A , MDA-MB-231 = State B , SUM159 = State C ) . We isolated total mRNA from these heterogeneous populations and profiled the expression of 17 differentiation-related genes and GAPDH , thereby generating a gene-expression profile for each heterogeneous population ( S1 Table ) . Lastly , we applied SVD , NMF and ICA to the gene expression matrix to assess the relative performance of these algorithms in identifying changes in cell-state proportions . The results of the SVD , NMF and ICA analyses are presented in Fig 2B–2D . SVD/PCA successfully identified components that closely correlated with the proportions of the cell states in our idealized experiment: the first component exhibited a strong negative correlation with the fraction of cells within the population in State A ( r2 = 0 . 92 ) , while the second component correlated with the fraction of cells in State B ( r2 = 0 . 47 ) . Additionally , the replicates for each perturbation clustered closely together in the space spanned by these first two components identified by the SVD/PCA algorithm ( Fig 2B right ) . Moreover , the first two SVD components together explained ~90% of the variation in the gene-expression data ( as can be seen by the Scree plot in S1A Fig ) , which is consistent with the two degrees of freedom inherent in the design of this idealized experiment . In contrast to SVD/PCA , the two components identified by NMF both correlated strongly with the fraction of cells in State A ( r2 = 0 . 92 and 0 . 92 respectively ) —with component 1 correlating negatively with the proportion of cells in State A , and component 2 correlating positively with the proportion of cells in State A ( Fig 2C , S1E Fig ) ; neither NMF component 1 nor 2 was correlated with states B or C ( all r2 < 0 . 43; S1E Fig ) . For this analysis the NMF factorization was performed with parameter k = 2 , because the two components together explained over 95% of the variance in the gene-expression data ( S1B Fig ) . As was the case for the SVD/PCA algorithm , the replicates for each perturbation clustered closely in the space spanned by the two components identified by NMF; this strongly suggested that the components identified by the algorithm reflected biological signal rather than experimental noise . Unlike the SVD/PCA and NMF algorithms , the first two ICA components did not correlate with the fraction of cells in any of states A , B or C ( all pairwise r2 < 0 . 13 , Fig 2D , S1F Fig ) . Moreover , in almost all cases the various replicates for a given perturbation did not cluster together in the space spanned by the first two components identified by ICA ( Fig 2D right ) . Collectively these observations indicated that both the SVD/PCA and NMF algorithms effectively identified components that correlated strongly with cell-state proportions , while ICA failed to do so . Moreover , these observations showed that only the SVD/PCA components spanned the 2 degrees of freedom inherent this idealized experiment , which , by design , involved cellular populations that were mixtures of exactly 3 cell states . One potential explanation for why the SVD and NMF components tracked cell-state proportions is that the components were identifying genes differentially expressed between cell states . We could directly compare gene loadings in the various components with gene expression in the various states because the gene-expression profiles of the pure states were known in our idealized experimental conditions ( Fig 2E ) . This comparison revealed that genes with the highest loadings in SVD component 1 were uniquely expressed or repressed in state A; this was consistent with the observation that this component tracked with the fraction of cells in state A . Similarly , NMF components 1 and 2—both of which also tracked with state A—identified a very similar set of genes uniquely expressed by state A ( Fig 2E ) . A key difference , however , was that unlike SVD component 1 , which included positive and negative loadings corresponding respectively to genes down or up in state A , both of the NMF components had only positive gene loadings—with NMF component 1 having positive gene loadings for the genes down in state A , and NMF component 2 having positive loadings for the genes up in state A ( Fig 2E ) . In contrast , SVD component 2 identified the only two genes that were strongly differentially expressed between states B and C ( HOXA5 , FOXO1; Fig 2E ) ; these two genes , HOXA5 and FOXO1 , were respectively down and up in state B relative to state C , and were expressed near median levels in state A . Thus , the highest loadings of SVD1 in this idealized experiment marked genes differentially expressed between luminal and basal cells , including the established luminal markers GATA3 and STAT5A . More generally , these findings suggested that the highest loadings in the SVD component vectors may serve to identify markers of specific cell states in contexts where such markers are not known . Since our goal in developing PEACS was to identify perturbations that affect cell state proportions , we needed a method for reducing the SVD component weights to a single score that quantifies the extent of change in cell-state proportions . For this purpose the Euclidian metric , which corresponds to the natural notion of ‘distance’ in 1 , 2 and 3-dimensional space , was attractive for several reasons . First , we expect distances in SVD space to scale linearly with the extent of the change in cell state proportions . Consistent with this , analysis of the SVD1 v SVD2 replicate plot for the idealized experiment ( Fig 2B right panel ) revealed that small perturbations in cell state proportions ( e . g . 1:1:1 to 1:2:2 ) resulted in small distances in component space , whereas large changes in cell state proportions ( e . g . 1:1:1 to 0:1:1 ) resulted in large distances in SVD component space . Second , the Euclidian metric makes it straightforward to quantify how noise in the various dimensions impacts the reliability of multidimensional distance estimates . We therefore used the Euclidean metric to compare distances between samples in the space spanned by the first k SVD components , where k was chosen using the standard approach of looking for an ‘elbow’ in the corresponding Scree plot . To account for biological variability across replicates ( or different shRNAs targeting the same gene ) , we defined the PEACS score as the Euclidean distance divided by the standard error about the mean for each set of replicates ( Fig 3 ) . Intuitively , this PEACS score—Euclidean distance divided by standard error—can be thought of as a ‘signal-to-noise’ ratio , which scales the magnitude of a change by the error in the distance estimate . Empirical p-values for PEACS scores were determined by Monte Carlo sampling: for a given perturbation with n replicates , a null distribution was obtained by randomly sampling n expression profiles from the experimental data , calculating a PEACS score , and iterating this process 10 , 000 times to generate a PEACS score null distribution . The empirical p-value was then determined by ranking the PEACS score for the given perturbation relative to the PEACS scores generated by this Monte Carlo procedure . We next applied PEACS to the MCF10A human stem cell model of mammary morphogenesis [6] . When seeded into a three-dimensional collagen matrix , MCF10A cells form ductal , lobular , and ductal-lobular tissue rudiments ( Fig 4A–4C ) . These tissue rudiments are monoclonal , indicating that they arise from single stem cells , and are morphologically similar to structures present in the human mammary gland ( Fig 4C; S2 and S3 Fig; S1 Movie ) . As a first step , we used gene-expression profiling to identify 39 developmentally implicated transcription factors ( TFs ) expressed in MCF10A cells ( S2 Table ) . We next inhibited these factors with 3–5 shRNAs targeting each TF , with two biological replicates per shRNA , resulting in a total of 240 genetically perturbed lines . For each genetically perturbed line , we then profiled the expression of all 39 factors and housekeeping genes using high-throughput qRT-PCR . These experiments generated a large data matrix with rows corresponding to gene expression values , and columns corresponding to shRNA perturbations . From this data matrix , we eliminated genes that were not inhibited by at least 3 distinct shRNAs . Application of the PEACS algorithm to this filtered data matrix produced a score that quantified the extent to which TF inhibition affected cell-state proportions . Based on this PEACS score , most genetic perturbations had small effects on cell state proportions , which were comparable to the effects of hairpins that did not successfully knockdown their targeted genes ( Fig 5A , S3 Table ) . When inhibited , several genes caused large , reproducible changes in cell state proportions , which could be seen when the perturbations were plotted in 3D SVD component space or as PEACS scores ( Fig 5A and 5B ) . We used the first three SVD components for this analysis because the elbow of the Scree plot occurred at three dimensions ( S1C Fig ) . The top three factors identified by this analysis were NR3C1 , RUNX1 and TCF3 ( Fig 5C , S3 Table ) . Identification of the glucocorticoid receptor ( NR3C1 ) , the highest-scoring factor , was significant because of its established role in regulating mammary ductal differentiation and lactation [7] . TCF3 , the third-highest scoring factor , was recently reported to be a mammary stem cell regulator [8] . RUNX1 , which was the second-highest scoring factor , is mutated in a subset of breast cancers but has not been previously implicated as a regulator of mammary stem cell biology [9–11] . Since the other hits identified by PEACS were established regulators of mammary stem cells or differentiation , we suspected that RUNX1 might also play a role in one or both of these processes , and therefore decided to further explore its function . In this dataset , RUNX1 primarily affected the expression of SVD component 1 . We therefore investigated the loadings of SVD component 1 to identify the genes that have the highest contribution to this component ( Fig 5D ) . The highest loadings of SVD component 1 were ETS1 , HIF1A , HOXA5 , NFYA , RUNX1 , YY1 , and RB1 . As expected , these genes were significantly decreased in the RUNX1 knockdown condition compared to perturbation conditions that did not change SVD component 1 ( Fig 5D ) . While we do not know what the state corresponding to SVD component 1 is , these markers may be useful for future studies investigating mammary lineages . To evaluate the functional role of RUNX1 we inhibited its expression with shRNAs ( Fig 6B ) and assessed the ability of MCF10A cells to form tissue rudiments in polymerized collagen . RUNX1-inhibited cells formed spheres that did not hollow ( Fig 6D ) , indicating that they were not mature lobules , and rarely formed ducts or ductal-lobular rudiments ( 71% reduction relative to control ) ; the rare ducts that did form were shorter in length ( 25% reduction ) and did not exhibit the branched morphology seen in wild type structures ( Fig 6A , 6C ) . As a control , cells that were either mock-infected or expressed a control shRNA were not affected in their ability to form tissue rudiments . These results indicated that RUNX1 is required for mammary cells to differentiate into ducts and mature lobules . To assess if the phenotype caused by RUNX1 inhibition was reversible , we generated an MCF10A line in which RUNX1 could be reversibly inhibited by a doxycyline ( dox ) -inducible shRNA ( Fig 7B ) . When cultured in collagen in the presence of dox , these MCF10A cells formed solid spheres and few ducts , recapitulating the phenotype observed above when RUNX1 was constitutively inhibited by shRNAs . When RUNX1 was re-expressed by withdrawing dox , the spheres rapidly sprouted ducts and began to hollow—often within 12–24 hours ( Fig 7A ) . This finding indicated that the RUNX1-inhibited spheres were still capable of forming both ducts and lobules upon RUNX1 re-expression , raising the possibility that these spheres might consist of bipotent cells reversibly arrested in their differentiation . To directly examine this possibility we assessed whether single cells from RUNX1-inhibited spheres could form tissue rudiments when seeded into collagen . Parental MCF10A cells largely lose this ability upon differentiating in collagen ( Fig 7C ) . We seeded cells with dox to form RUNX1-inhibited spheres , harvested and dissociated the spheres by treatment with collagenase and trypsin , and then reseeded single cells into collagen with or without dox . Cells reseeded in dox again gave rise to solid spheres . However , those reseeded without dox formed lobules and ducts that matured into complex ductal-lobular structures ( Fig 7C ) , doing so with efficiency comparable to that of parental MCF10A cells maintained in 2D culture . These observations strongly suggested that parental MCF10A cells dissociated from tissue rudiments lost the ability to reseed tissue rudiments because they had differentiated and lost stem and progenitor activity; in contrast , cells within RUNX1-inhibited MCF10A spheres maintained their ability to reseed tissue rudiments because they did not differentiate in collagen and remained bipotent . We next examined if RUNX1 also affected the differentiation of primary human breast stem and progenitor cells . To this end we isolated primary human breast epithelial cells from reduction mammoplasty tissue samples , modulated RUNX1 expression , and assessed stem and progenitor cells using colony forming assays ( Fig 8A ) [12–14] . In these assays the majority of stem and progenitor cells form colonies containing differentiated luminal or basal cells . However a fraction of bipotent stem cells proliferate but do not differentiate; these form micro-colonies of 2–16 cells that remain uncommitted and co-express both luminal and basal markers . Inhibiting RUNX1 expression caused a 2-fold increase in the number of stem cell micro-colonies , suggesting that this transcription factor was required for primary human breast stem cells to differentiate in culture ( Fig 8B ) . Consistent with this interpretation , inhibiting RUNX1 expression reduced the number of differentiated colonies by nearly 90% , while its over-expression led to a 300% increase in differentiated colonies . We next examined whether transiently inhibiting RUNX1 would expand the population of functional stem cells in culture . For this experiment we first infected primary cells with the dox-inducible shRUNX1 lentivirus , and plated cells with dox to assay for colony-forming ability . After micro-colonies of stem cells had formed ( 7 days after plating ) , we removed the dox so that RUNX1 would be re-expressed . We found that re-expressing RUNX1 caused the stem cell micro-colonies to differentiate within 48–96 hours , and resulted in the formation of heterovalent colonies that included both bipotent stem cells and lineage-committed basal and luminal cells ( Fig 8C ) . These heterovalent colonies were never observed in colony-forming assays with control primary cells , or in assays with primary cells in which RUNX1 had been stably inhibited . Collectively , these findings indicate that RUNX1 inhibition enables primary breast stem cells to expand in an uncommitted state while retaining the functional ability to differentiate in culture .
We have shown that PEACS identifies perturbations that affect cell-state transitions , taking as input the gene-expression profiles of perturbed cellular populations . We validated PEACS by applying it to a mammary stem cell model with shRNAs as a source of perturbations . In this context , the method identified several established regulators ( e . g . , NR3C1 and TCF3 ) of mammary stem cell biology , as well as a novel gene , RUNX1 , which had not previously been implicated as a mammary stem cell regulator . Follow-up studies revealed that inhibiting RUNX1 prevented mammary stem cells from differentiating , indicating that this gene is required for stem cells to exit a bipotent state . Although our study focused on shRNA perturbations , there is every reason to believe that PEACS would be equally effective for gene over-expression or chemical perturbations . Several computational methods for analyzing gene-expression profiles have been previously reported [15–19] . PEACS differs from these in three important ways . First , the goal of PEACS is to specifically identify perturbations that influence how cells transition between differentiation states; we are not aware of other methods that do this . Second , the method does not require any markers of stem , progenitor or differentiated states . Third , our method analyzes bulk populations of cells to identify changes in cell state ratios , rather than analyzing large numbers of single cells . We anticipate that this marker-free approach will be particularly useful in the many contexts where stem , progenitor , and differentiated cells have been identified functionally , but where markers that distinguish these states are not yet available . It is worth emphasizing that , although markers that enrich for stem and progenitor states have been identified in many systems , few systems offer markers that sort stem or progenitor cells to purity; this latter ability is essential if these markers are to be used to identify genes that regulate state transitions . In cases where such markers are in fact available—or when they are used to define states de facto without consideration of the underlying biology—we have previously shown that a Markov model can be used to quantify the rates of transition between states , and predict the equilibrium proportions of cell states [20] . Targeting RUNX1 may offer unique possibilities for therapeutic applications . Stem cells have a strong tendency to differentiate when propagated in culture , even under conditions that are intended to maintain them in an undifferentiated state . This problem has been observed with human ES cells , HSCs , and many other stem cell types . We have shown that primary human mammary stem cells can be expanded in a bipotent state by transiently inhibiting RUNX1; moreover these cells spontaneously differentiate once RUNX1 expression is re-established . A chemical compound that inhibits RUNX1 could therefore be used to propagate mammary stem cells in culture . It will be of interest to examine if inhibiting RUNX1 can also prevent other types of stem cells from exiting a bi- or multipotent state . In support of this possibility , a dominant-negative RUNX1 translocation has been found in a subset of leukemias , and expression of this protein blocks the differentiation of leukemic cells and promotes the self-renewal of hematopoietic stem cells [21 , 22] . Additionally , a RUNX1 ortholog , Runt , has been shown in planaria to be required for neoblast stem cells to differentiate at wound sites [23] . Taken together , these observations suggest the intriguing possibility that this function of RUNX1/Runt is conserved across species and cell types .
Primary tissues were obtained with consent in compliance with laws and institutional guidelines , as approved by the Institutional Review Board of Maine Medical Center . Exemption status for human research was obtained from the Committee on the use of Humans as Experimental Subjects ( COUHES ) at MIT , based on de-identification of the samples . All patient samples are de-identified prior to distribution for research use . The data collected and stored is limited to basic demographic data , specimen handling information ( ex: related to chain of custody ) , specimen quality data , and histopathologic data . At no time is any patient identifier provided to any researcher . MCF10A cells were obtained from ATCC and cultured in MEGM with 100 ng/ml cholera toxin , GlutaMax , Penicillin and Streptomycin ( Lonza CC-3150 ) . HEK293T cells were maintained in DMEM supplemented with 10% FBS , GlutaMax , Penicillin and Streptomycin . SUM159 ( Asterland ) , MDA-MB-231 ( ATCC ) , and T47D ( ATCC ) cells were cultured in DMEM with 10% FBS , GlutaMax , Penicillin and Streptomycin . Human organoids were isolated from breast tissues from patients undergoing elective reduction mammoplasty . Primary tissues were obtained with consent in compliance with laws and institutional guidelines , as approved by the Institutional Review Board of Maine Medical Center . Organoids were aliquoted in 1:1 DMEM/Hams-F12 media supplemented with 5% calf serum , 10 ng/mL insulin , 10 μg/mL epidermal growth factor , 10 μg/mL hydrocortisone , and 10% DMSO and stored in liquid nitrogen . Doxycycline ( dox ) , where applicable , was used at a concentration of 4μg/ml . Lentivirus production , target cell infection , and selection were performed as previously described [24] . Constitutive shRNA plasmids in a pLKO . 1 vector were obtained from the Broad Institute RNAi consortium ( https://www . broadinstitute . org/rnai/trc3 ) , and inducible hairpins ( dox “ON”; pTRIPZ vector ) were obtained from Thermoscientific . Overexpression constructs were obtained through gateway cloning of the appropriate ORF into the pLenti6 . 2-ccdB-3xFLAG-V5 construct . MCF10A cells were seeded onto a 96 well plate at a density of 7500 cells per well and infected the next day with hairpin lentivirus targeting an expressed developmental transcription factor . One day after infection , cells were selected with 5ug/ml puromycin containing media . Two days later RNA was collected with the Qiagen RNeasy 96 Biorobot 8000 kit and cDNA synthesized with the iScript cDNA synthesis kit ( BioRad 170–8890 ) . Microfluidic qPCR was carried out according to the manufacturer’s Protocol ( Protocol 37: Fast Gene Expression Analysis Using EvaGreen on the BioMark or BioMark HD System ) . The 39 TFs profiled were selected by profiling gene-expression in MFC10As and selecting all TFs implicated in differentiation that were confirmed to be expressed by qPCR . The cDNA was preamplified for 14 cycles with a mix of 41 primer sets ( 39 TFs , BTub , and GAPDH ) and mastermix , then treated with ExoI . Prior to analysis with PEACS , the data matrix with the Fluidigm CT values was normalized to GAPDH and median normalized by gene such that the median CT value for each gene was 0 . For the idealized experiment , gene expression was profiled using standard qPCR and the 17 genes profiled were randomly selected transcription factors expressed by MCF10A cells and implicated in differentiation . Let M be a data matrix of perturbation-expression values with rows corresponding to perturbations and columns corresponding to the genes whose expression was profiled . We used the reduced singular value decomposition to transform M , so that M=[ ↑u1↓…↑un↓ ]︷mxn[ σ1⋱σn ]︷nxn[ ←v1t→⋮←vnt→ ]︷nxn=∑i=1nσiuivit ( 1 ) where u1 , … , un and v1 , … , vn are respectively the left and right eigenvectors corresponding to the singular values σi of the reduced singular value decomposition of M . We have assumed here that n < m , i . e . , the number of perturbations exceeds the number of genes whose expression is profiled . Because the dimensions of ui and vi are ( mx1 ) and ( nx1 ) , respectively , and the σi are scalars , M can be viewed as a weighted sum of the rank-one matrices uivit . We then use the first k singular values and vectors to reconstruct a low-rank approximation of M: M~[ ↑u1↓…↑uk↓ ]︷mxk[ σ1⋱σk ]︷kxk[ ←v1t→⋮←vkt→ ]︷kxn ( 2 ) The value k is chosen using a Scree plot , as described in the main text . In the case of k = 3: The gene-expression vector for the perturbation p can therefore be approximated by the following weighted sum of the first 3 SVD eigenvectors: Thus the gene expression data for each perturbation p is mapped into the space spanned by linear combinations of the first k gene-expression SVD eigenvectors v1 , … , vk . Again for k = 3: These coordinates in SVD-space are plotted as ‘Component scores’ in Figs 2B , 3 , and 5A . Finally , to determine the PEACS score , we first calculate the Euclidean distance between the u1p , u2p , u3p for a given perturbation p ( averaged across all its replicates ) , and the median centroid vector ( u¯1 , u¯2 , u¯3 ) taken across all m perturbations: Finally , the PEACS score is calculated by dividing the distance in ( Eq 6 ) by the standard error across replicates for a given perturbation . To calculate a p-value , a Monte Carlo sampling algorithm was implemented . For each set of n perturbations , a null distribution of PEACS scores was obtained by sampling n random perturbations 10 , 000 times without regards for perturbation labels . The p-value was defined as the rank of the real PEACS score in the null distribution divided by 10 , 000 . The PEACS code for MATLAB is available as a supplemental file ( S2 Text ) and on our lab website at: http://guptalab . wi . mit . edu/ . 7 . 5x103 MCF10A cells were resuspended in 0 . 2ml of collagen solution ( 1 . 25mg/ml rat tail collagen I in PBS , brought to pH 7 . 3 with 0 . 1N NaOH ) and plated on a single chamber of a 4-chamber slide . Collagen was polymerized for 2 hours at 37°C , after which they were detached and cultured in 1ml of MCF10A medium . MCF10A cells were grown in collagen matrix through day 7 , at which time the collagen pads were collected and incubated in 100 ug/ml collagenase in PBS at 37°C for 10 minutes . The structures were collected by centrifugation ( 500 RPM , 5 min ) , resuspended in 0 . 25% trypsin , and incubated for 20–25 minutes at 37°C . Cells were counted in trypan blue , spun down ( 500RPM , 5 min ) , and resuspended in MCF10A media; 7500 living cells were reseeded into a new collagen pad . Samples were fixed with 4% paraformaldehyde for 15 minutes at room temperature . Pads were permeabilized using 0 . 1% TritonX-100 and incubated with blocking solution ( PBST with 10% goat serum and 3% BSA ) for 1 hr at room temperature and stained with the appropriate primary antibody in blocking buffer for 1–2 hours at room temperature or overnight at 4°C . The samples were washed with PBS , and incubated with an Alexa Fluor-labeled secondary antibody . Samples were washed , stained with 1ug/ml DAPI . Images of phalloidin-AF594 and DAPI- stained collagen structures were analyzed by image segmentation software ( CellProfiler; [25] ) , with an analysis pipeline that differentially detected lobules and ducts based on size , area and form factor adjustments . Primary human organoids were thawed and plated on a 10cm dish in 10ml of RMFC ( DMEM + 10% Calf Serum ) media for 1–2 hours . The non-adherent fraction , fibroblast reduced organoids , was collected , spun 10 minutes at 233 gravity , resuspended in cold PBS and passed 10 times through an 18-gauge needle . The organoids were once again pelleted 5 minutes at 335 gravity , resuspended in 2ml of 0 . 05% trypsin , and incubated 10 minutes at 37°C . We then added 8ml of RMFC media and 0 . 5mg of DNaseI ( Roche 10104159001 ) . The cell suspension was passed through a 40 um filter and the cells counted . Thirty thousand cells were plated per well of a 6-well plate in MEGM , and assayed for cytokeratin expression after 7–11 days , using CK8/18 antibody ( Vector VP-C407 ) and CK14 antibody ( Thermo 9020-P ) . Some plates were visualized using IHC , while others were visualized using IF with AF488/AF555 conjugated secondary antibodies . Colonies grown on 6-well plates were fixed in 100% methanol for 5 minutes , washed with PBS , permeabilized with 0 . 1% triton X-100 followed by serial blocking in 3% hydrogen peroxide and 1% BSA + 2% horse serum . The plates were incubated overnight at 4°C with 1:750 CK8/18 antibody . The plates were incubated with 1:200 αMouse-IgG-HRP ( Vector BA-2000 ) for 30 minutes , and stained with DAB according to the manufacturer’s protocol ( Vector ABC elite PK-6100; Vector ImmPACT DAB SK-4105 ) . Excess avidin/biotin was blocked with the Vector Avidin/Biotin blocking kit SP-2001 . Plates were re-blocked for 1 hour in PBS + 1% BSA and 2% goat serum , then incubated for 1 hour at room temperature with 1:750 CK14 antibody in PBS + 1% BSA then incubated at room temperature with αRabbit-IgG-HRP ( Vector BA-1000 ) for one hour . The plates were then stained with VIP according to the manufacturer’s protocol ( Vector ABC elite PK-6100; Vector ImmPACT VIP SK-4605 ) , washed with water and stored dry . Western blots were performed with standard procedures . RUNX1 was blotted with 1:1000 Ab23980 ( AbCAM ) .
|
The discovery of stem cell regulators is a major goal of biological research , but progress is often limited by a lack of definitive markers capable of distinguishing stem cells from early progenitors . Even in cases where markers have been identified , they often only enrich for certain cell states and do not uniquely identify states . While useful in some contexts , such enriching markers are ineffective tools for discovering genes that regulate the transition of cells between states . We present a method for identifying these cell state regulatory genes without the need for pre-determined markers , termed Perturbation-Expression Analysis of Cell States ( PEACS ) . PEACS uses a novel computational approach to analyze gene expression data from perturbed cellular populations , and can be applied broadly to identify regulators of stem and progenitor cell self-renewal or differentiation . Application of PEACS to mammary stem cells resulted in the identification of RUNX1 as a key regulator of exit from the bipotent state .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2015
|
Perturbation-Expression Analysis Identifies RUNX1 as a Regulator of Human Mammary Stem Cell Differentiation
|
Genomic rearrangements ( gross chromosomal rearrangements , GCRs ) threatens genome integrity and cause cell death or tumor formation . At the terminus of linear chromosomes , a telomere-binding protein complex , called shelterin , ensures chromosome stability by preventing chromosome end-to-end fusions and regulating telomere length homeostasis . As such , shelterin-mediated telomere functions play a pivotal role in suppressing GCR formation . However , it remains unclear whether the shelterin proteins play any direct role in inhibiting GCR at non-telomeric regions . Here , we have established a GCR assay for the first time in fission yeast and measured GCR rates in various mutants . We found that fission yeast cells lacking shelterin components Taz1 or Rap1 ( mammalian TRF1/2 or RAP1 homologues , respectively ) showed higher GCR rates compared to wild-type , accumulating large chromosome deletions . Genetic dissection of Rap1 revealed that Rap1 contributes to inhibiting GCRs via two independent pathways . The N-terminal BRCT-domain promotes faithful DSB repair , as determined by I-SceI-mediated DSB-induction experiments; moreover , association with Poz1 mediated by the central Poz1-binding domain regulates telomerase accessibility to DSBs , leading to suppression of de novo telomere additions . Our data highlight unappreciated functions of the shelterin components Taz1 and Rap1 in maintaining genome stability , specifically by preventing non-telomeric GCRs .
The integrity of chromosomal DNA can be compromised by mutations that vary in size , ranging from small perturbations , such as point mutations and short insertions/deletions , to large changes , such as deletions , duplications , inversions , and translocations of long chromosome segments . The latter are collectively called genomic rearrangements or gross chromosomal rearrangements ( GCRs ) , which have profound implications in cancers as well as genetic diseases . Recent advances in DNA sequencing technology have enabled us to trace the history of GCRs in cancer cells , and it is now well-known that cancer development is accompanied by the frequent occurrence of GCRs [1] . Thus , elucidation of the molecular mechanism underlying GCR control is of critical importance in understanding the progression of cancer malignancy . Previous studies have pointed to the requirement of chromosome maintenance mechanisms for suppression of GCRs , including DNA repair and telomere protection pathways [2 , 3] . The telomere is a huge DNA-protein complex that is located at the termini of linear chromosomes . In humans , telomeric DNA comprises hexanucleotide TTAGGG repeats and consists of a double-stranded ( ds ) region and a single-stranded ( ss ) overhang . The telomeric dsDNA recruits TRF1-TRF2-Rap1 , whereas the ss telomeric DNA recruits POT1-TPP1 , and these two subcomplexes are bridged by TIN2 to form a complex known as shelterin ( reviewed in [4] ) . This shelterin complex helps cells distinguish telomeres from DNA double-strand breaks ( DSBs ) that must be repaired . For instance , TRF2 depletion brings about the frequent occurrence of chromosome end-to-end fusions , which is due to deregulation of the non-homologous end joining ( NHEJ ) repair pathway at telomeres . Resultant dicentric chromosomes are unstable , leading to another round of chromosomal rearrangements ( reviewed in [5] ) . It is thus evident that telomere protection by the shelterin complex is vital for repressing GCRs . While the shelterin complex primarily serves to protect telomeric DNA , the telomere-associated DNA polymerase named telomerase is implicated in GCRs [6 , 7] . On the one hand , telomerase is able to elongate the telomere repeat sequence using its RNA subunit as a template , thereby counteracting gradual telomere shortening at each round of DNA replication . At the same time , however , telomerase poses a potential threat to genome stability . In budding yeast , telomerase promotes GCRs through de novo addition of telomere repeats to DSB sites , resulting in terminal deletion of chromosomal DNA [7] . It has been reported that de novo telomere addition is suppressed through two mechanisms: activation of Pif1 helicase , which was proposed to remove telomerase from DSBs; and inhibition of Cdc13 accumulation by DNA damage signaling [8–10] . However , a previous study showed that fission yeast Pif1 is not a negative regulator of telomerase [11] . In human cells , recruitment of telomerase to telomeres and the activity of telomerase are regulated by the shelterin complex ( reviewed in [12] ) . However , it is still unclear whether shelterin is also involved in the regulation of de novo telomere addition at non-telomeric sites . Fission yeast , Schizosaccharomyces pombe , serves as a useful model to dissect the functions of shelterin , because this unicellular organism shares most of the shelterin components with humans . Fission yeast shelterin is composed of six proteins: Taz1 , Rap1 , Poz1 , Tpz1 , Pot1 , and Ccq1 . Among these , Taz1 , Tpz1 and Pot1 are orthologs of human TRF1/2 , TPP1 , and POT1 , respectively . Fission yeast Rap1 and human Rap1 are also homologous to each other , sharing several domains including a single BRCT domain at their N termini . Taz1 and Rap1 form a subcomplex that binds to telomeric ds DNA , while Poz1 , Tpz1 , Ccq1 , and Pot1 form another subcomplex at the telomeric ss DNA . Similar to human shelterin , these two subcomplexes at telomeric ds and ss DNA are bridged by the physical interaction between Rap1 and Poz1 [13 , 14] . To date , the shelterin components in fission yeast have been extensively investigated . Taz1 , Rap1 , and Poz1 negatively regulate telomerase activity and promote telomere heterochromatin formation [13 , 15–17] . On the other hand , Tpz1 and Pot1 are essential for telomere protection , and thus telomere DNA is aggressively degraded after deletion of the tpz1+ or pot1+ gene [13 , 18] . Ccq1 recruits telomerase to telomeres through direct binding to telomerase [19 , 20] . Taz1 and Rap1 prevent telomere end fusions that would otherwise be caused by aberrant activation of the NHEJ repair pathway at telomeres [21 , 22] . Taz1 and Rap1 also tether telomeres to the nuclear periphery via inner nuclear membrane ( INM ) protein Bqt4 in vegetative cell growth [23] . As such , the shelterin components perform distinct functions , even though they form a complex . It is known that disruption of shelterin can trigger frequent GCRs through breakage of dicentric chromosomes formed by chromosome end-to-end fusion [5] . However , it is unclear whether the shelterin complex has an additional GCR-suppressive function apart from preventing such chromosome end-to-end fusions; this uncertainty can be ascribed to technical limitations in precisely measuring the occurrence rate of GCRs in mammalian cells . In budding yeast , an assay has been developed to measure the GCR rates , aptly termed the “GCR assay” [24 , 25] . GCR rates are deduced from loss of two tandem counter-selective markers inserted in a non-essential chromosomal region . In this study , we adopted the GCR assay to fission yeast and examined whether the individual shelterin components as well as other telomere-binding proteins suppress GCRs in non-telomeric regions . We found that a fraction of the shelterin components , including Taz1 and Rap1 , are required for GCR suppression . Deletion of DNA ligase IV , which is essential for NHEJ , did not rescue the increased GCR rates in taz1Δ and rap1Δ mutant cells , suggesting that Taz1 and Rap1 do not prevent GCR via suppressing NHEJ , unlike the Taz1- and Rap1-dependent protection of telomeres from fusion . Instead , derepression of telomerase is responsible for the increased GCR rates in taz1Δ and rap1Δ strains . Dissection of the Rap1 protein identified the N-terminal BRCT domain as an important domain for the GCR suppression . Moreover , when DSBs are site-specifically induced at a non-telomeric locus by I-SceI endonuclease , Taz1 and Rap1 are required for cellular survival and for inhibiting erroneous repair . We propose that Taz1 and Rap1 prevent GCRs by regulating telomerase activity and DSB repair , even in non-telomeric regions .
To measure GCR rates in fission yeast , we applied the assay system that was previously developed for budding yeast ( Fig 1A ) [24] . We constructed a DNA cassette containing two neighboring marker genes , ura4+ and TK ( the latter encodes herpes virus thymidine kinase ) in tandem . Cells expressing ura4+ and TK are sensitive to 5-fluoroorotic acid ( 5-FOA ) and 5-fluoro-2’-deoxyuridine ( FUdR ) , respectively . As expected , fission yeast cells with this marker cassette integrated at approximately 150 kb from the right telomere of chromosome I ( the precise location is described in Materials and Methods ) showed sensitivity to both of the drugs ( 5-FOA/FUdR ) ( S1A Fig ) . This strain is expected to become resistant to both drugs when the ura4+ and TK genes undergo simultaneous deletions and/or loss-of-function point mutations . However , such simultaneous point mutations seem highly unlikely because the probability of simultaneous point mutations occurring in two specific genes is thought to be quite low ( ~10−14/cell division , given that the spontaneous incidence of loss-of-function mutations for each gene , independently , is ~10−7 , see Methods ) [26] . Thus , as in the budding yeast GCR assay system , the vast majority of drug-resistant survivors in our assay should be derived from GCRs that result in simultaneous loss of the two marker genes . Because an essential gene closest to the marker cassette is sec16+ , which is located about 16 . 8 kb centromeric from the cassette , and there is no essential gene telomeric to the cassette , our system can detect GCRs that take place within this ~16 . 8-kb region ( Fig 1A ) . Hereafter , we will refer to this GCR target region as the “breakpoint region” . Because it lacks any sequence that shares apparent homology with other chromosome regions , our GCR assay is expected to detect GCRs that are mediated by no or little homology . In the GCR assay , we counted the number of colonies on a plate with or without 5-FOA/FUdR and estimated GCR rates per cell division using fluctuation analysis [24] . In the case of wild-type cells , a GCR rate determined in our system was 2 . 6 × 10−9 per cell division ( Fig 1B ) . This rate is actually far greater than the expected probability of dual independent point mutant survivors ( 10−14 per cell division ) , confirming that our system primarily detects GCRs . We then isolated 5-FOA/FUdR-resistant clones and performed DNA sequencing at the breakpoint region ( See Materials and Methods ) . Based on the sequencing data , GCRs were classified into deletion and translocation types ( Fig 1C , wild-type ) . In the deletion type , DSBs led to deletion of the chromosomal terminus containing the drug selection cassette . The sequencing analysis detected ectopic telomeric DNA repeats at the breakpoints , suggesting that de novo telomere addition healed the DSB ( Fig 1D ) . Twelve out of fifteen wild-type-derived GCR clones examined here belonged to this type . In two other clones , breakpoints were fused with unique sequences from the left arm of chromosome I ( opposite the right arm where the original marker cassette had been located prior to the rearrangement ) in a head-to-tail orientation ( same direction towards telomeres ) . Such fusions could be derived from either break-induced DNA replication or DNA recombination ( Fig 1D , Translocation as diagramed in S1C Fig ) . Indeed , the breakpoint junctions consisted of 7 or 8 bp of microhomology in these survivors ( Fig 1D ) . In both types of GCRs , the locations of the breakpoints seemed to be uniformly distributed in the breakpoint region , rather than clustering at a particular hotspot ( Fig 1E ) . In the last of the 15 survivors , we established the loss of the marker cassette but could not determine the precise change in the breakpoint region sequence . As we expected , we did not obtain any clones with simultaneous point mutations in both ura4+ and TK , validating the usefulness of our assay system to specifically evaluate GCR rates . Of note , the GCR rate measured in our fission yeast system is comparable to that in the original budding yeast system: 2 . 6 × 10−9 /cell division in a 16 . 8 kb-long breakpoint region in wild-type fission yeast ( this study ) vs . 2 . 27 × 10−9 in a 19 . 2 kb breakpoint region in wild-type budding yeast [27] . If we assume that GCRs occur randomly throughout the genome , then GCR rates would be expected to be proportional to the DNA length of the breakpoint region for the GCR assay . The GCR rates normalized to unit length are 1 . 5 × 10−10/cell division/kb in fission yeast and 1 . 2 × 10−10/cell division/kb in budding yeast . Thus , wild-type cells of the two yeast species showed comparable normalized GCR rates . In addition , GCRs in wild-type fission yeast cells are mostly associated with terminal deletions , whereas translocations are relatively rare , just as in budding yeast [24] . We thus asked whether the GCR suppression mechanism identified by the budding yeast GCR assay system also functions in fission yeast . Budding yeast strains lacking nuclease FEN-1 or Mre11 show a 914- and 628-fold increases in GCR rates , respectively [20] . We therefore tested the effects of loss of FEN-1 and Mre11 in rad2Δ and mre11Δ S . pombe cells and observed a ~100-fold increase in GCR rates ( Fig 1B ) . In budding yeast , Pif1 helicase suppresses telomerase-mediated telomere elongation at native telomeres and DSBs through destabilizing annealing of telomerase RNA template and single-stranded telomere DNA substrates [28] . We examined the impact of inactivating pfh1 , the fission yeast homologue of PIF1 on GCRs . pfh1-mt* is a mutant that lacks nuclear functions but retains the essential mitochondrial functions [29] . pfh1-mt* showed 32-fold higher GCR rates than wild type , similar to the results reported for the budding yeast corresponding mutant , pif1-m2 ( S1D Fig ) [7] . From these similarities observed in the two distinct yeast species , we surmise that the regulatory mechanism suppressing GCRs is evolutionarily conserved , underscoring the significance of studying the GCR mechanism in fission yeast . Since most of the GCR survivors that we isolated were derived from de novo telomere addition , we investigated a possible involvement of the telomere proteins in GCR regulation . In fission yeast , Taz1 directly binds to both Rap1 and ds telomeric DNA , thereby recruiting Rap1 to telomeres ( Fig 2A ) . We found that taz1Δ and rap1Δ cells showed greatly increased GCR rates , 1 . 1 × 10−7 and 0 . 85 × 10−7 /cell division ( 42-fold and 33-fold higher than wild-type cells ) , respectively ( Fig 2B ) . A Rap1-I655R mutant , in which the recruitment of Rap1 to telomeres is diminished due to a compromised Taz1-Rap1 interaction [30] , showed an increased GCR rate that was comparable to taz1Δ or rap1Δ ( Fig 2B , 2 . 6 × 10−7 /cell division , a 97-fold increase over wild-type ) , suggesting that Taz1 represses GCRs primarily through the physical interaction with Rap1 . Consistent with this notion , taz1+ and rap1+ were found to be epistatic: a taz1Δ rap1Δ double mutant ( 1 . 2 × 10−7 /cell division ) showed similarly increased GCR rates compared to each single mutant taz1Δ or rap1Δ ( Fig 2B ) . We determined the sequences of GCR breakpoints in taz1Δ and rap1Δ survivors in GCR assay . We found only deletion type GCRs in taz1Δ and rap1Δ survivors , although the fractions of deletion in these mutants are not significantly higher than in wild type ( Fig 2C & 2D ) . These results imply that the two shelterin components Taz1 and Rap1 function in the same pathway to prevent GCRs . We also measured GCR rates of wild-type , taz1Δ , and rap1Δ strains at 20°C . It is known that taz1Δ , but not wild-type or rap1Δ delay the cell cycle progression and lose viability due to chromosomal entanglement at this temperature [31] . We found that taz1Δ , but not wild-type and rap1Δ , showed more than one order of magnitude higher GCR rates at 20°C than at 32°C ( S2 Fig ) . Given the correlation between the increased GCR rate and cold-sensitivity among the three strains , it is possible that the chromosome entanglement contributes to the high GCR rate with taz1Δ at 20°C . Future study is necessary for concluding the molecular link between these phenotypes . In sharp contrast to taz1+ and rap1+ deletion , deletion of the poz1+ gene , which encodes another Rap1-interacting shelterin component , did not affect the GCR rate ( 3 . 9 × 10−9 /cell division , Fig 2B ) . Interestingly , taz1Δ poz1Δ and rap1Δ poz1Δ double mutants showed lower GCR rates than the taz1Δ and rap1Δ single mutants , demonstrating that Poz1 is required for the derepression of GCRs in taz1Δ and rap1Δ cells ( Fig 2E ) . Consistently , the abrogation of Poz1-Tpz1 binding by a I501A/R505E mutation in Tpz1 [32] , another shelterin component that directly binds to Poz1 , similarly suppressed the increased GCR rates in taz1Δ and rap1Δ mutant backgrounds ( Fig 2E ) . This result suggests that Poz1 recruitment promotes GCRs in taz1Δ and rap1Δ , given that Poz1-Tpz1 binding is essential for telomere localization and function of Poz1 [14 , 33] . We noticed that poz1+ deletion and Tpz1-I501A/R505E mutation individually caused strong reduction in GCR rates in rap1Δ but not much in taz1Δ cells . These results suggest that the increased GCR rates in taz1Δ and rap1Δ are caused by different mechanisms . It was reported that the formation of Rap1-Poz1-Tpz1 trimer is a hierarchical process in vitro [34] . First , Poz1 and Tpz1 form a dimer . The Rap1-binding domain of Poz1 undergoes allosteric changes upon the Poz1-Tpz1 dimer formation , which greatly increases the affinity with Rap1 , and induces the Rap1-Poz1-Tpz1 trimer formation . Therefore , it is possible that the Tpz-Poz1 dimer stably exists in rap1Δ but not in taz1Δ . Such unusual shelterin subcomplexes may contribute to the differential effects of poz1+ deletion in taz1Δ and rap1Δ , as revealed in Fig 2E . These results suggest that genetic interaction of Taz1 , Rap1 , and Poz1 regarding GCR suppression is complex , similarly to that as for telomere length regulation and cold sensitivity [20 , 22] . We also examined Stn1 , a non-shelterin protein that binds to telomeric ssDNA . Because Stn1 is essential for telomere protection , we investigated a temperature sensitive mutant stn1-1 , which has slightly elongated telomeres at semi-permissive temperature 25°C [35] . We found a moderately increased GCR rate at that temperature ( Fig 2B ) , 1 . 6 × 10−8 /cell division , a 6-fold increase over wild-type . Because both Taz1 and Rap1 are multi-functional ( see below ) , we set out to dissect which specific function ( s ) is related to the GCR inhibition . It is known that , in taz1Δ and rap1Δ cells , but not in poz1Δ cells , telomeres are prone to fuse to each other by NHEJ when cells are arrested at G1 phase [21 , 36] . It is thus possible that a failure to suppress NHEJ in taz1Δ and rap1Δ could lead to formation of dicentric chromosomes , which would trigger DSBs which could result in the observed chromosome terminal deletions . In order to examine whether NHEJ is responsible for frequent GCRs in taz1Δ and rap1Δ mutant cells , we deleted the DNA ligase IV-encoding lig4+gene , which is essential for NHEJ in fission yeast . It was reported that a lack of lig4+ suppresses the frequent telomere fusions in taz1Δ and rap1Δ [21 , 22] . We found that disruption of lig4+ in taz1Δ and rap1Δ did not significantly suppress the increased GCR rates observed with taz1Δ and rap1Δ ( Fig 3A ) , suggesting that NHEJ is dispensable for the high incidence of GCRs in taz1Δ and rap1Δ cells . Because fission yeast in exponentially growing phase shows very short G1 phase , and NHEJ is active only in G1 but not in S and G2 phase , it was possible that NHEJ was dispensable for GCRs due to a small fraction of cells staying in G1 phase . We therefore arrested cells in G1 phase through nitrogen starvation , and measured the GCR frequency . Briefly , cells exponentially growing in YES media were divided into two groups , which were incubated in EMM media with ( N+ ) or without ( N- ) ammonium sulfate for 24 hr , respectively , and then transferred to YES media for growth overnight . Then equal numbers of N+ and N- cells were subjected to the GCR assay . We found that taz1Δ ( N- ) cells showed an approximately two-fold increase in GCR frequencies compared to taz1Δ ( N+ ) cells ( S3A Fig ) . Because taz1Δ ( N- ) cells were G1-arrested and/or lost viability [21] while taz1Δ ( N+ ) cells actively proliferated in EMM media ( with or without supplementing nitrogen ) , the total number of cell divisions was greater in taz1Δ ( N+ ) cells than in taz1Δ ( N- ) cells . Therefore , the two-fold increase of GCR frequencies is most likely an underestimate of a larger GCR rate ( which is normalized per cell division ) in taz1Δ ( N- ) compared to that in taz1Δ ( N+ ) . Interestingly , GCR frequencies in taz1Δ ( N- ) were partially suppressed by ligase IV deletion , while those in taz1Δ ( N+ ) were not . Taken together , NHEJ also contributes to the increase of GCR frequencies of taz1Δ cells in G1 phase . Because G1 cells are rare in cells with unperturbed cell cycles , this effect is negligible in exponentially growing cell populations . These results suggest that the break-fusion-bridge cycle via formation of telomere-fusion-mediated dicentric chromosomes plays a minor role , if any , in the increased GCR rate in cycling taz1Δ cells . Both Taz1 and Rap1 are essential for heterochromatin formation at telomeres and their adjacent regions , subtelomeres [37] . To examine whether telomere heterochromatin structure is important for suppression of GCRs , we deleted the clr4+ and swi6+ genes , both of which encode essential factors for heterochromatin formation ( S3B Fig ) . Deletion of swi6+ in the wild-type background led to a small increase in the GCR rate , suggesting a potential contribution of heterochromatin to the suppression of GCRs . We examined poz1-W209A mutation . The shelterin component Poz1 is required for telomere silencing , and the poz1-W209A mutation is known to specifically disrupt the telomere heterochromatin regulatory function , among others [38] . No increase in the GCR rate was observed in poz1-W209A strain ( S3C Fig ) . In contrast , deletion of clr4+ in rap1Δ background showed small but significant decrease GCR rates , although the underlying mechanism is unclear . From these results , we conclude that heterochromatin does not play a significant role in suppressing GCR except in rap1Δ . In fission yeast , Taz1 and Rap1 , but not Poz1 , tether telomeres to the INM via binding of Rap1 to INM protein Bqt4 in vegetative cell growth [23 , 39] . It is thus possible that the telomere tethering to the INM contributes to GCR suppression through regulation of chromosome positioning within the nucleus . We found that bqt4Δ cells showed moderately increased GCR rates ( 4 . 3 × 10−8 /cell division , Fig 3B ) . The GCR rate was also significantly increased by deletion of bqt3+ ( 3 . 0 × 10−8 /cell division ) , whose protein product Bqt3 stabilizes Bqt4 [23] . It was reported that the Ku70/80 complex and two INM proteins Lem2 and Man1 also promote tethering of telomeres to the nuclear envelope , although Man1 plays a minor role [40 , 41] . Deletion of pku70+ or lem2+ , but not man1+ , led to moderately higher GCR rates ( 2 . 9 × 10−8 and 4 . 2 × 10−8 /cell division , respectively ) than the wild-type strain ( Fig 3B ) . These results imply that tethering of telomeres to the nuclear envelope facilitates GCR suppression . Bqt4 localizes to the INM through its C-terminus transmembrane domain , and its N-terminal half is necessary and sufficient for binding Rap1 [23] . While telomeres are dissociated from the nuclear envelope in bqt4Δ , expression of an artificial fusion protein between Rap1 and an N terminus-truncated Bqt4 ( Rap1-GFP-Bqt4ΔN ) in bqt4Δ resumed telomere clustering at the nuclear envelope [23] . With our GCR assay , we found that bqt4Δ cells expressing the Rap1-GFP-Bqt4ΔN fusion protein from bqt4 promoter showed only a slightly lower GCR rate than bqt4Δ cells expressing GFP-Bqt4ΔN , in which telomeres are not tethered to the INM ( Fig 3B , 2-fold difference ) . Moreover , the rap1-5E mutant ( consisting of S213E , T378E , S422E , S456E , S513E mutations ) , in which the interaction between Rap1 and Bqt4 is impaired [39] , displayed a comparable GCR rate to wild-type cells ( Fig 3B ) . We also found that simultaneous deletion of bqt4+ significantly increased GCR rates in taz1Δ and rap1Δ cells ( Fig 3C ) . These results suggest that Rap1-Bqt4 binding plays a minor role in suppressing GCRs , and that Bqt4 regulates GCRs at least in part by a Taz1- and Rap1-independent mechanism . By the same token , this result suggests that Rap1 utilizes Bqt4-independent mechanisms for suppressing GCRs . Taz1 and Rap1 suppress telomerase-mediated telomere DNA elongation [42 , 43] . Given that all GCRs examined in taz1Δ and rap1Δ were terminal deletions involving de novo telomere additions at breakpoints , it was likely that deregulated telomerase reactions facilitated GCRs through enhanced de novo telomere addition in taz1Δ and rap1Δ cells . Inactivation of trt1+ , the gene encoding the catalytic subunit of telomerase , leads to chromosome self-circularization [42] , making the GCR assay results difficult to compare with other cases . We therefore explored the effect of a Pof8 disruption on GCR rates . Pof8 is involved in maturation of telomerase RNA , and deletion of pof8+ leads to telomere shortening without extensive chromosome circularization , in contrast to a trt1+ deletion [44–47] . We found that taz1Δ pof8Δ and rap1Δ pof8Δ cells , which also do not show chromosome circularization , showed GCR rates which were lower than taz1Δ and rap1Δ cells , and similar to wild type cells ( Fig 4A ) . These results suggest that telomerase activity is essential for the high GCR rates in taz1Δ and rap1Δ . In contrast , the GCR rate of a rad2Δ pof8Δ strain was in between that of rad2Δ alone and wild type cells , suggesting that Taz1 and Rap1 specifically suppress telomerase-dependent GCRs . We considered two possibilities for how telomerase activity affects GCRs in taz1Δ and rap1Δ: ( 1 ) increased telomerase accessibility directly facilitates de novo telomere addition at breakpoints , or ( 2 ) abnormally elongated native telomeres indirectly affect non-telomeric GCRs . To determine which is the case , we examined GCR rates using cells with circular chromosomes in the presence or absence of Trt1 [4] . It is known that circular chromosomes in trt1Δ do not contain telomere DNA sequences [48] . For this purpose , we introduced a Trt1-expressing plasmid into trt1Δ cells with circularized chromosomes . In this setting , the majority of the trt1Δ cells harboring the Trt1 plasmid maintained circular chromosomes I and II ( S4 Fig ) . As for trt1Δ taz1Δ , chromosomal configuration depends on the order of gene deletions during the strain preparation . When trt1+ is deleted first , followed by taz1+ deletion , the strain contains circular chromosomes . In contrast , linear chromosomes are maintained when taz1+ is deleted first , followed by trt1+ deletion [42] . Below , we will describe experiments using trt1Δ taz1Δ maintaining circular chromosomes , except otherwise noted . We also confirmed that trt1Δ taz1Δ expressing ectopic Trt1 retains circular chromosomes ( S4 Fig ) . When we subjected circular chromosome-containing cells to the GCR assay , it was expected that circular chromosomes needed to undergo complicated changes , such as two independent DSBs at the both sides of the selection cassette , and healing of the two DSBs by telomere addition to produce linear chromosomes . Consistently , all of the various strains maintaining circular chromosomes ( except Trt1-overproducing trt1Δ taz1Δ ) showed GCR rates below the detection sensitivity of the assay ( Fig 4B ) . When trt1Δ , trt1Δ taz1Δ and trt1Δ rap1Δ ( all containing circular chromosomes ) were transformed with Trt1-expressing plasmids , trt1Δ taz1Δ showed a significant increase in GCR frequency , while trt1Δ and trt1Δ rap1Δ did not ( Fig 4B ) . These results suggest two points: first , Taz1 prevents GCR formation independent of its specific DNA binding to telomere DNAs , since circular chromosomes lack all telomere DNAs [48]; second , Taz1 has additional roles , which are not shared by Rap1 , in preventing GCR formation from circular chromosomes . When we deleted poz1+ in trt1Δ taz1Δ cells , followed by over-expression of Trt1 , GCR rates were decreased , suggesting that Poz1 promotes GCRs in taz1Δ cells in the absence of telomere DNA ( Fig 4B , compare lanes 4 and 8 ) . rap1Δ trt1Δ cells showed similar GCR rates to wild type even after Trt1 re-expression . We confirmed that both trt1Δ taz1Δ poz1Δ cells and rap1Δ trt1Δ cells maintained circularization of chromosomes I and II before and after Trt1 re-expression ( S4 Fig ) . In contrast to circular chromosomes-containing trt1Δ taz1Δ , linear chromosome-maintaining trt1Δ taz1Δ ( see above ) , showed significantly increased GCR rates compared to linear-chromosome-containing wild-type cells ( Fig 4B ) . Ectopic Trt1-over-expression further increased the GCR rates to the level of taz1Δ cells . To further dissect the precise mechanism of GCR repression by Rap1 , we exploited previously reported sequential N-terminal Rap1 truncations , Rap1-A to G [31] ( Fig 5A ) . Among these , we found that only the Rap1-G mutant showed an increased GCR rate . Because the Rap1-A to F mutant strains all retain the Poz1-binding domain ( Rap1 457–512 amino acids ) but Rap1-G does not , the results raised the possibility that Rap1-Poz1 binding is required for the GCR suppression . However , deletion of the Poz1-binding domain alone did not increase GCR rates ( Rap1ΔP , Fig 5A and 5B ) . Further dissection of Rap1 revealed that simultaneous deletion of the BRCT domain at the N terminus as well as the Poz1-binding domain led to an increase in GCR rates that was comparable to that in rap1Δ cells ( Rap1-AΔP , Fig 5A and 5B , and S4 Fig ) . The phenotype of the rap1-AΔP strain was similar to rap1Δ regarding GCR suppression; GCRs from the rap1-AΔP strain primarily showed terminal deletion ( Fig 5C ) , and deletion of pof8+ canceled the increased GCR rates . These results indicate that the BRCT domain and the Poz1-binding domain redundantly suppress GCRs . Since the Rap1-A mutant , which lacks the BRCT domain , maintains normal telomere length , we suggest that the BRCT domain does not regulate telomerase action at native telomeres , while the Poz1-binding domain suppresses GCRs through inhibition of telomerase at both telomeres and non-telomeric DSBs . The Rap1 BRCT domain may be involved in a general DNA repair pathway , failure of which causes various consequences including erroneous telomere addition by telomerase at non-telomeric regions . Given that GCRs are thought to arise from aberrant DSB repair , the telomerase-independent GCR repression mechanism could potentially include DSB processing . In order to examine this possibility , we constructed a conditional , site-specific DSB induction system ( Fig 6A ) . The DNA sequence-specific endonuclease I-SceI was expressed under the control of a tetracycline-inducible promoter , and a single I-SceI cut site ( I-SceIcs ) was integrated at approximately 150 kb centromeric from the right telomere of chromosome I , exactly at the same locus as the marker cassette that was inserted in our GCR assay strains . Addition of anhydrotetracycline ( ahTET ) to the culture media leads to a DSB at the I-SceIcs . Indeed , two hours after ahTET addition , quantitative PCR amplification of genomic DNA using primers flanking the I-SceIcs decreased to 40–50% of control levels in wild-type , taz1Δ , rap1Δ , and poz1Δ backgrounds , demonstrating that DSBs were induced in these strains with similar efficiencies ( Fig 6B ) . With this system , we examined how efficiently the wild-type and mutant cells could repair the DSB . We transiently induced DSB formation at I-SceIcs by culturing cells in liquid media containing ahTET for two hours . After that , ahTET was washed out and the cells were spread onto ahTET-free plate media . Cells that were unsuccessful in repairing the I-SceI DSB ( or healing it , e . g . by de novo telomere addition ) would not form colonies . We examined genomic DNA extracted from 10 colonies each from wild type , taz1Δ , rap1Δ strains and confirmed that none of them contained mutation in I-SceIcs , indicating that GCRs were not involved in generating survivors . Subsequent to the transient DSB induction in wild-type cells , the frequency of colony formation decreased to 62% of control ( uncut ) levels ( Fig 6C ) . Strikingly , taz1Δ and rap1Δ cells showed further lower viabilities ( 37% and 33% , respectively ) . This result suggests that Taz1 and Rap1 promote DSB repair . taz1Δ pof8Δ cells showed similar survival with taz1Δ , consistent with the idea that the survivors occur not through telomerase-mediated GCRs , but through DSB repair , and suggesting that Taz1 promotes DSB repair independently of telomerase regulation . Interestingly , the BRCT domain-lacking rap1-A mutant showed significantly lower survival ( 42% ) than wild-type , indicating that the Rap1 BRCT domain plays a significant role in DSB repair . We note , however , that the survival rate of the rap1-A strain was still slightly but significantly higher than rap1Δ , suggesting that Rap1 is involved in two pathways that promote DSB repair: one BRCT-domain-dependent and the other independent ( Fig 6D ) . We examined whether Taz1 and Rap1 physically bind to the DSBs by chromatin immunoprecipitation ( ChIP ) . No significant ChIP signal was detected for both Taz1 and Rap1 at the sites 1 . 5 and 5 kb apart from I-SceIcs at 2 and 4 hrs after DSB induction , while they localized at telomeres ( S6B Fig ) . It is possible that Taz1 and Rap1 were only transiently recruited to DSBs , which made the ChIP detection difficult . Alternatively , Taz1 and Rap1 are indirectly involved in the DSB repair ( see Discussion ) . To examine if the impaired DSB repair in taz1Δ and rap1Δ cells leads to GCRs , we measured GCR frequencies caused by the I-SceI-induced DSB . We allowed the I-SceI endonuclease to be continuously active by culturing cells on plate media containing ahTET . Under this condition , faithful DSB repair would be detrimental to cell viability because it would regenerate I-SceIcs , leading to incessant cut and repair cycles . In contrast , I-SceIcs would become resistant to I-SceI cleavage when the I-SceIcs was lost via mutagenic DSB repair , including GCRs , indels , and point mutations . In the wild-type background , only 0 . 58% of cells survived and formed colonies , suggesting that GCRs and erroneous DNA repair is rare ( Fig 6F ) . The survival was likely caused by de novo telomere addition , because trt1Δ showed significantly lower survival rates than wild type ( Fig 6E ) . ccq1Δ also decreased the survival rate to a similar level shown by trt1Δ , suggesting that Ccq1 is required for de novo telomere addition at DSBs , as in the case of telomerase-mediated telomere elongation at native telomeres . In contrast , a higher fraction of cells survived in the taz1Δ and rap1Δ backgrounds . In all wild-type , taz1Δ , and rap1Δ survivors , in which the breakpoints were identified ( n = 9 , 9 , and 7 , respectively ) , the I-SceIcs was eliminated by terminal deletion associated with de novo telomere addition , suggesting that the observed viabilities reflect frequencies of terminal deletions in wild-type , taz1Δ and rap1Δ strains . Therefore , Taz1 and Rap1 not only promotes faithful DSB repair ( Fig 6C ) , but also prevents erroneous DSB repair by suppressing terminal deletion coupled with de novo telomere addition . In wild-type cells , breakpoints of the deletions were close to ( 1~10 bp ) I-SceIcs in all 9 clones in which the breakpoints were identified ( Fig 6G ) . In contrast , breakpoints of 4 out of 9 clones in the taz1Δ survivors and 5 out of 7 clones in rap1Δ survivors were within this range , but the breakpoints in the other 5 taz1Δ survivor clones ( one was ~300-bp centromeric , and four were 9~13-kb centromeric to the I-SceIcs ) and two rap1Δ clones were located far ( >10 bp ) from the I-SceIcs . Therefore , we monitored DNA resection around DSBs indirectly through ChIP experiments of RPA . Localization of RPA subunit Rad11 was increased 2 hours after ahTET addition at 1 . 5 kb distant from I-SceIcs , while it was increased 4 hours at 5 and 13 kb both in wild type and taz1Δ strains ( S6C Fig ) . These results suggest that Taz1 and Rap1 repress de novo telomere addition associated with DNA resection . The uncontrolled telomere addition may have contributed to the higher frequency of chromosome deletions found in taz1Δ and rap1Δ compared to wild-type clones ( Fig 6E and 6F ) . As rap1-A and poz1Δ strains did not show increased survival , disruption of either Rap1 BRCT domain or Rap1-Poz1 pathway is not sufficient for inducing mutagenic DNA repair , and they function redundantly for suppressing telomere addition . We also examined distribution of breakpoints of telomere addition in rap1-A survivors . Since the breakpoints often located far from I-SceIcs ( >10 bp , 3/6 ) similarly with taz1Δ and rap1Δ , the broader distribution of breakpoints shown in the shelterin mutants would be caused consistently by a loss of the Rap1 BRCT domain-dependent DSB repair pathway . Collectively , these results suggest that Taz1 represses GCRs by facilitating proper DSB repair and suppressing de novo telomere addition . In a previous research , Taz1 was implicated in DSB repair because taz1Δ cells were sensitive to DNA damaging agents , such as methyl methane sulfonate ( MMS ) and bleomycin [31] . The sensitivity was augmented by simultaneous deletion of cds1+ [31] . We tested if a loss of the impaired DSB repair was responsible for the higher GCR rates in taz1Δ and rap1Δ . We found that the increased GCR rates and sensitivity to DNA-damaging reagents were not always correlated: rap1Δ was not sensitive to bleomycin ( S6D Fig ) , consistent with a study showing that rap1Δ is not sensitive to MMS [36] . Moreover , taz1Δ cds1Δ and rap1Δ cds1Δ double-mutant strains showed decreased GCR rates relative to taz1Δ and rap1Δ ( S6E Fig ) . These results indicate that the high GCR rates in taz1Δ and rap1Δ is not simply due to defective DSB repair .
How do shelterin components Taz1 and Rap1 repress GCRs at the breakpoint region ? Although it has been reported that Taz1 can be recruited to telomere-like sequences outside telomeres , the breakpoint region of our GCR assay system does not have any telomeric DNA motif and a previous genome-wide chromatin immunoprecipitation analysis failed to detect Taz1 in the breakpoint region [49] . In addition , other genome-wide studies have indicated that telomeres are unlikely to reside stably in close proximity to the breakpoint region by examining 3D chromosome positioning in the nucleus [50 , 51] . One possibility is that the higher GCR rates observed in the taz1Δ and rap1Δ strains were caused indirectly from the abnormal telomeres in these mutants . For example , aberrantly elongated telomeres ( taz1Δ and rap1Δ ) or gapped telomeres ( taz1Δ ) sequester substantial amounts of DSB repair factors [52] , thereby compromising DSB repair efficiency outside telomeres . However , we showed that Taz1 still significantly suppressed GCRs in cells with no telomeric DNA ( circular chromosomes ) ( Fig 4B ) . This result favors another hypothesis that DSBs directly recruit Taz1 and Rap1 in a telomeric DNA sequence-independent manner , rather than the indirect model . However , we have not detected any localization of Taz1 and Rap1 at DSBs in ChIP experiments ( S6B Fig ) . It is possible that although Taz1-Rap1 associates with DSBs , the association is very limited either temporally or stoichiometrically , which made detection by the ChIP experiment difficult . In humans , it was reported that human TRF1 and TRF2 are recruited to DNA damage sites to promote homologous recombination-directed DSB repair ( HDR ) . It is known that the association happens only transiently immediately after DSB induction [53–55] . In budding yeast , inner nuclear envelope protein Mps3 binds unrepaired DSBs , thereby spatially recruiting them close to the nuclear envelope [56] . We showed that Taz1 and Rap1 promote survival after a transient site-specific DSB induction , suggesting that they are involved in DSB repair . Notably , the experiment with constitutive DSB induction showed that GCR breakpoints in taz1Δ were in some cases far ( > ~10 kb ) from the original break site . These large deletions in taz1Δ can be explained by excessive resection of DSB ends or defective HDR . Firstly , Taz1 might suppress DNA end resection at non-telomeric DSB sites . This is not surprising given that Taz1 and Rap1 suppress extensive resection at telomeres [57] . Excessively resected DNA may provoke loss of the opposite strand , because it is known that 3’-end strands are degraded several hours after resection of the 5’ strand at a DSB in budding yeast [58] . It is possible that de novo telomere addition occurred at such new DSB sites distant from the original DSB site , which may account for the de novo telomere additions that are far from the original DSB site in taz1Δ cells . Alternatively , since the Taz1 homolog TRF1 promotes HDR at non-telomeric DSBs [52 , 53] , it is possible that Taz1 suppresses large deletions by promoting HDR . According to this scenario , in the taz1Δ strain , cells that engage in grossly defective HDR to repair DSBs are inviable , but an increased accessibility of telomerase to DSBs would promote formation of de novo telomere addition to the DSBs . In addition , the longer reaction time required by the inefficient HDR in taz1Δ , compared to taz1+ , would also lead to extensive resection , resulting in large deletions . Similar to our results , budding yeast pif1 mutants , in which de novo telomere addition is highly promoted , showed large deletions concomitant with spontaneous and HO endonuclease-induced de novo telomere addition [52] . This also can be explained by inefficient HDR , because it was recently shown that Pif1 promotes HDR [58] . Previous studies with budding yeast and mammalian cells have proposed that telomere-binding proteins prevent GCRs solely by suppressing fatal inter-chromosomal fusions or de novo telomere additions . Our results , however , raise the possibility that components of the shelterin complex have a previously unappreciated mechanism for suppressing genome instability . Remarkably , our Rap1 truncation analysis demonstrated that the previously uncharacterized BRCT domain together with the Poz1-binding domain suppresses GCRs . Although we do not know at this moment the detailed molecular mechanism of GCR suppression by the BRCT domain , it does not depend on NHEJ because DNA ligase IV is dispensable for the increased GCR rate in rap1Δ . The BRCT domain-dependent mechanism would also be independent of telomerase regulation , given that the Rap1-A strain lacking the BRCT domain has normal telomere lengths [27] . Rather , the BRCT domain would be involved in DSB repair , because the Rap1-A strain showed reduced survival after transient DSB induction . Although the N-terminal BRCT domain is a conserved feature of Rap1 among other species , including budding yeast and humans , its function has been unclear , except that a mutation in the BRCT domain of budding yeast Rap1 affects its regulatory activity related to transcription [59] . It would be interesting to examine whether the BRCT domain of mammalian Rap1 is involved in suppression of genome instability or promotion of DSB repair . Further analysis of fission yeast shelterin components and their counterparts in mammals will reveal the detailed mechanism of the GCR suppression .
All of the experiments were performed using the S . pombe strains listed in S1 Table . Growth media , basic genetics , and construction of strains carrying deletion alleles or epitope-tagged proteins were described previously [60] . Point mutations were introduced using a QuikChange Multi Site-Directed Mutagenesis Kit ( Agilent , 200514 ) , or by manually using Dpn I and PCR . All plasmids constructed in this study were sequence-verified . TK was cloned from FY2317 [61] and was flanked by a cytomegalovirus promoter and S . cerevisiae LEU2 terminator . ura4+ and TK were inserted between SPAC29B12 . 14c and SPAC1039 . 01 on chromosome I ( at nt 5442736 of the chromosome I sequence described in Pombase: http://www . pombase . org/ ) For immunoblot of Rap1 and Cdc2 , polyclonal anti-Rap1 antibody [29] and Cdc2 p34 [PSTAIRE] antibody ( sc-53 , Santa Cruz Biotechnology , Inc . ) were used . For measurements of GCR rates , a previously described method using fluctuation analysis in budding yeast was applied to fission yeast , with some modifications [62] . Cells were streaked on YES agar plates and incubated at 32°C to form single colonies . Whole colonies were picked up by excising colonies with a sterile scalpel , suspended in YES liquid media , and incubated at 32°C until saturation . A portion of the saturated cells ( at most 500 μl per plate ) was plated on a YES agar plate supplemented with 1 mg/ml 5-FOA and 20 μg/ml FUdR . At the same time , 100 μl of a 105-fold dilution of the cell suspension was plated on another YES agar plate . Both plates were placed at 32°C , and colony numbers of selective and non-selective media were counted after 5- and 2-day incubations , respectively . The total number of GCRs in each liquid culture was estimated from the colony numbers using a following equation [63]: m= ( r/z‐0 . 693 ) /ln ( r/z+0 . 367 ) where m represents a number of GCRs formed de novo , not formed by duplication of pre-formed GCRs , in each total liquid culture , r represents the number of colonies on the selective media , and z represents the fraction of cells plated on the selective media . This procedure was performed with at least 7 independent colonies , and median values of [ ( number of GCRs: m ) / ( total cell number ) ] were shown as the GCR rate . When the median value was zero , a tentative median value was calculated by assuming m = 1 in all colonies and shown in figures as an upper bound , as described previously [62] . Confidence intervals for the GCR rates were calculated as previously described [62] , and shown in graphs as error bars . In order to determine statistical significance of differences in GCR rates , a Mann-Whitney test was performed . To determine breakpoint sites in GCR survivors , breakpoint sequences were mapped to ~400 bp resolution by sequential PCR analysis , as previously described [25] ( S7A Fig ) . For survivors with no loss of amplification by any primer sets , PCR products spanning ura4+ and TK ( in GCR assay ) or I-SceIcs ( in the site-specific DSB assay ) were sequenced to check whether they contained point mutations . To assess whether de novo telomere addition had occurred , PCR analysis was performed with a primer designed to anneal to a sequence centromeric from breakpoint and a primer including telomere sequence ( M13R-T1: 5’-caggaaacagctatgacctgtaaccgtgtaaccgtaac-3’ or M13R-19: 5’-caggaaacagctatgaccctgtaaccccctgtaacc-3’; the underlined sequence was added to increase specificity of amplification after the 2nd cycle ) using iProof DNA polymerase ( Bio-Rad ) ( S7B Fig ) [64] . Reaction mixtures were incubated at 98°C for 30 s , then cycled 30 times at 98°C for 10 s , 68°C for 20 s , and 72°C for 1 min . For survivors without de novo telomere addition , the breakpoint sequence was determined by ligation-mediated PCR , as described previously [65] ( S7C Fig ) . To examine how G1 arrest affects GCR frequency , we modified the protocol for measurement of GCR frequencies after treatment with DNA damaging agents [66] . Cells exponentially grown in YES liquid medium were divided equally , washed twice with EMM medium with/without ammonium chloride ( nitrogen source ) , and incubated with the same medium for 24 hours . The incubated cells were washed once with YES medium and incubated with YES liquid medium overnight until saturation . Defined numbers of cells were plated on YES plates containing/lacking 5-FOA and FUdR . Resulting colony numbers were counted , and the ratio of surviving cells in 5-FOA and FUdR was calculated to obtain the respective GCR frequencies . Site-specific DSB induction was performed essentially as described previously [67] , with the following modifications . The tetracycline-inducible I-SceI integration plasmid containing LEU2 was integrated at the leu1-32 locus . A plasmid containing the I-SceI cleavage site as well as ura4+ and hygromycin B selectable markers was integrated at the same site as the ura4+-TK cassette in our GCR assay strain . DSB was induced by addition of ahTET ( Sigma , 3 μM final ) . qPCR was performed using a StepOnePlus real-time PCR system ( Applied Biosystems ) . Sequences of PCR primer sets are listed in S2 Table . ChIP assays were performed essentially as described previously [13] , with the following modifications . The cell concentration was adjusted to 1 . 0 × 107 cells/mL just before addition of ahTET and a defined volume of the culture was collected at indicated time points for fixation . Immunoprecipitation was performed with anti-myc antibody ( 9B11 , Cell Signaling ) using Dynabeads M‐280 Sheep anti‐Mouse IgG ( Invitrogen ) . DNA was purified and extracted from washed beads and input samples with Chelex 100 Resin ( BioRad ) as described previously [68] , and analyzed by qPCR . According to a previous report [26] , fission yeast acquires 5-FOA resistance by spontaneous inactivating point mutation of either ura4+ or ura5+ , and the rate of spontaneous mutation which confers 5-FOA resistance is 1 . 3 × 10−7 ( /cell division ) . Among these mutations , ratio of mutation in ura5/ura4 is 1 . 85 , so inactivating mutation rate of ura4+ can be calculated as 1 . 3 × 10−7 / ( 1 . 85+1 ) = 4 . 6 × 10−8 . Given that the length of ORF of ura4+ and TK are ~800 and ~1100 bp , respectively , we estimated mutation rate of TK as 4 . 6 × 10−8 × ( 1100/800 ) = 6 . 3 × 10−8 . If this estimate is true , the rates of point mutations that confer 5-FOA and FUDR resistance are both nearly or less than 10−7 per cell division .
|
Tips of chromosomes , telomeres , are bound and protected by a telomere-binding protein complex called shelterin . Most previous studies focused on shelterin’s telomere-specific role , and its general role in genome maintenance has not been explored extensively . In this study , we first set up an assay measuring the spontaneous formation rate per cell division of gross chromosomal rearrangements ( GCRs ) in fission yeast . We found that the rate of GCRs is elevated in mutants defective for shelterin components Taz1 or Rap1 . Detailed genetic experiments revealed unexpectedly that Taz1 and Rap1 have a novel role in repairing DNA double-strand breaks ( DSBs ) and suppressing GCRs at non-telomeric regions . Given that shelterin components are conserved between fission yeast and humans , future studies are warranted to test whether shelterin dysfunction leads to genome-wide GCRs , which are frequently observed in cancers .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"chromosome",
"structure",
"and",
"function",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"cloning",
"telomeres",
"mutation",
"fungi",
"model",
"organisms",
"non-homologous",
"end",
"joining",
"experimental",
"organism",
"systems",
"dna",
"molecular",
"biology",
"techniques",
"schizosaccharomyces",
"research",
"and",
"analysis",
"methods",
"saccharomyces",
"chromosome",
"biology",
"animal",
"studies",
"chromosomal",
"aberrations",
"schizosaccharomyces",
"pombe",
"molecular",
"biology",
"yeast",
"biochemistry",
"chromosomal",
"deletions",
"point",
"mutation",
"eukaryota",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"yeast",
"and",
"fungal",
"models",
"saccharomyces",
"cerevisiae",
"dna",
"repair",
"organisms",
"chromosomes"
] |
2019
|
Telomere-binding proteins Taz1 and Rap1 regulate DSB repair and suppress gross chromosomal rearrangements in fission yeast
|
Complete metamorphosis ( Holometaboly ) is a key innovation that underlies the spectacular success of holometabolous insects . Phylogenetic analyses indicate that Holometabola form a monophyletic group that evolved from ancestors exhibiting hemimetabolous development ( Hemimetaboly ) . However , the nature of the changes underlying this crucial transition , including the occurrence of the holometabolan-specific pupal stage , is poorly understood . Using the holometabolous beetle Tribolium castaneum as a model insect , here we show that the transient up-regulation of the anti-metamorphic Krüppel-homolog 1 ( TcKr-h1 ) gene at the end of the last larval instar is critical in the formation of the pupa . We find that depletion of this specific TcKr-h1 peak leads to the precocious up-regulation of the adult-specifier factor TcE93 and , hence , to a direct transformation of the larva into the adult form , bypassing the pupal stage . Moreover , we also find that the TcKr-h1-dependent repression of TcE93 is critical to allow the strong up-regulation of Broad-complex ( TcBr-C ) , a key transcription factor that regulates the correct formation of the pupa in holometabolous insects . Notably , we show that the genetic interaction between Kr-h1 and E93 is also present in the penultimate nymphal instar of the hemimetabolous insect Blattella germanica , suggesting that the evolution of the pupa has been facilitated by the co-option of regulatory mechanisms present in hemimetabolan metamorphosis . Our findings , therefore , contribute to the molecular understanding of insect metamorphosis , and indicate the evolutionary conservation of the genetic circuitry that controls hemimetabolan and holometabolan metamorphosis , thereby shedding light on the evolution of complete metamorphosis .
Insects are , by far , the most successful and diversified animal group , with more than two million species described ( approximately half of all animal species reported ) . One of the reasons of this taxonomic richness lies in the appearance of specific novel phenotypic characters known as “key innovations” that has allowed the adaptive radiation of insect species . Several lines of evidence suggest that wings and complete metamorphosis are the two key innovations that have had the most relevant effect on insect diversity through evolution [1 , 2] . However , whereas the origin , evolution and development of wings have been investigated intensively [3–5] , less data is available on the origin and evolution of complete metamorphosis [6–9] . Since their origination from arthropod ancestors , approximately 479 million years ago ( Ma ) [10] , insects have undergone extreme evolution in their postembryonic development , emerging different types of metamorphosis: ametaboly , hemimetaboly and holometaboly [6 , 11] . The most primitive type is ametaboly , in which immature individuals are miniature versions of the wingless adult form and sexual maturity is achieved through successive molts . In hemimetabolous insects , juvenile ( nymphs ) and adult forms are very similar and the metamorphosis of the adult-specific structures , the wings and the genitalia , occur during a single stage , the last nymphal instar . Finally , in holometabolous insects the immature larva undergoes a complete morphological transformation to form the adult . The body reorganization is so radical that a two-stage metamorphic process bridged by the holometabolous-specific intermediate pupal stage is required to transform the larva into a winged adult . Despite the relevance of complete metamorphosis in the taxonomic success of Holometabola , the nature of the changes underlying the appearance of the holometabolan pupa remains a puzzling problem in evolutionary and developmental biology . From an endocrine perspective , the genetic switch between juvenile and adult programs in hemimetabolous and holometabolous insects relies on the same hormone: the sesquiterpenoid juvenile hormone ( JH ) synthesized by the corpora allata glands [12–17] . While JH prevents metamorphosis during the pre-ultimate immature stages , its disappearance in the final juvenile stage allows metamorphosis to occur . The anti-metamorphic effect of JH is mediated by the induction of the C2H2 zinc-finger type transcription factor-encoding gene krüppel-homolog 1 ( Kr-h1 ) [13] . RNAi-mediated knockdown of Kr-h1 triggers premature adult development in hemimetabolous insects and induces precocious pupation in pre-ultimate instar larvae of holometabolous insects [18–21] . A second important metamorphic gene is Broad-complex ( Br-C ) , which encodes a member of the bric-a-brac-tramtrack-broad family of transcription factors [22 , 23] . In contrast to the conserved role of Kr-h1 , the functions of Br-C have critically changed from hemimetabolous to holometabolous insects . RNAi analysis in the hemimetabolous insects Oncopeltus fasciatus , Pyrrochoris apterus and Blattella germanica revealed that Br-C is specifically required for regulation of wing development , in particular size , shape and vein formation [18 , 24 , 25] , a function that is conserved in the holometabolous Tribolium castaneum , Drosophila melanogaster and Bombyx mori [23 , 26–30] . In contrast , Br-C functions in holometabolous insects have expanded to the metamorphic control of pupal commitment , pupal morphogenesis and the inhibition of adult differentiation [26–29 , 31 , 32] . In addition to Kr-h1 and Br-C , we have recently described E93 as the conserved master factor that promotes adult metamorphosis in hemimetabolous and holometabolous insects [33] . RNAi-mediated depletion of E93 prevents adult metamorphosis and induces endless repetitions of nymphal molts in hemimetabolous insects and the repetition of the pupal program in holometabolous insects [33] . In addition , E93 is also required to repress the expression of Kr-h1 and Br-C during the last immature stages of hemimetabolous and holometabolous insects , thus ensuring the transition to the adult forms [33] . Given the functional relevance of Kr-h1 , E93 and Br-C genes ( hereafter referred to as the metamorphic genetic network ) , we hypothesized that the appearance of the holometabolan-specific pupal stage was facilitated by changes in the timing of expression and/or regulation of the metamorphic network genes from hemimetabolous to holometabolous insects . In this study , we use the holometabolous insects Tribolium castaneum and Drosophila melanogaster to test this hypothesis . Here , we show that a transient peak of Kr-h1 at the end of the final larval stage , a particular event specific of holometabolous insects , has been critical for the occurrence of the pupa . This late pulse of Kr-h1 prevents the precocious up-regulation of E93 during this stage , thus pausing the implementation of the adult differentiation program initiated at the prepupal stage , and allowing the strong up-regulation of Br-C , which is critical for the correct formation of the pupa . In addition , we use the hemimetabolous insect Blattella germanica to demonstrate that the functional relation between Kr-h1 and E93 is evolutionary conserved , thus suggesting that the occurrence of the pupal stage has been facilitated by the co-option of regulatory mechanisms already present in hemimetabolous insects .
Three events are critical to promote metamorphosis in last instar nymphs of hemimetabolous insects: ( i ) the drop in the JH titer , ( ii ) the down-regulation of the anti-metamorphic factor Kr-h1 , and ( iii ) the up-regulation of E93 , which induces adult differentiation [33] . Likewise , these three events also occur during the onset of the last larval instar of the holometabolous T . castaneum ( S1 Fig ) [20 , 33] . However , a distinguishing event in T . castaneum is the up-regulation of TcKr-h1 expression during the prepupal stage at the end of the final ( L7 ) larval instar ( S1 Fig ) , suggesting that this Kr-h1 prepupal elevation might be a key event in the evolution of complete metamorphosis . Although the role of TcKr-h1 in pre-ultimate larval stages of T . castaneum has been analyzed previously [20] , its specific function during the prepupal stage is unknown . Thus , to examine the function of the late larval peak of TcKr-h1 expression , we depleted TcKr-h1 by RNAi in vivo by injecting TcKr-h1 dsRNA into newly emerged last L7 instar larvae ( TcKr-h1i animals ) . Specimens injected with dsMock were used as negative controls ( Control animals ) . Whereas Control animals pupated normally , the majority of the TcKr-h1i larvae arrested development after six days of injection ( S1 Table ) . Remarkably , removal of the apolysed larval cuticle from the arrested TcKr-h1i larvae revealed a strong precocious adult development , especially in the head and thorax ( Fig 1A–1D ) . For example , in the head several rows of ommatidia were clearly developed in the compound eyes of TcKr-h1i animals ( Fig 1Q–1S ) . The antennae presented adult sensillae and the shape and segmentation of the funicle and club were clearly adult-like ( Fig 1I–1K ) . The maxillae showed well-defined and segmented palps , lacinia and galea ( Fig 1M–1O ) . In the thorax , the legs presented the double claws typical for the adult legs and the different segments , including the tarsal ones , were clearly defined ( Fig 1E–1G ) . The elytra were highly sclerotized with the cuticle showing the typical adult microsculpture ( Fig 1U–1W ) . The adult-specific microsculpture was also detected in all body appendages ( Fig 2 ) . Furthermore , the dark brown pigmentation of the tanned cuticle in the head and thorax , including the appendages and part of the abdomen , resembled that of the adult ( Fig 1H , 1L , 1P , 1T and 1X ) . On the other hand , the abdomen of TcKr-h1i animals showed less pronounced adult differentiation and several pupal structures , such as the gin traps and urogomphi , were also visible ( S2 Fig ) . Confirming the premature activation of the adult genetic program in TcKr-h1i animals , high levels of the adult-specific cuticle gene TcCPR27 , normally occurring at the end of the pupal stage [34] , were precociously detected at the end of the larval stage ( Fig 3A ) . Overall , our data demonstrate that the specific up-regulation of TcKr-h1 at the end of larval development prevents the larva to metamorphose directly into the adult bypassing the pupal stage . Scale bars represent 0 . 5 mm in ( B ) and ( D ) ; 300 μm in ( E ) and ( F ) ; 200 μm in ( I ) , ( J ) , ( N ) and ( T ) ; 100 μm in ( G ) , ( H ) , ( K ) , ( M ) , ( O ) , ( R ) and ( X ) ; 50 μm in ( L ) , ( P ) , ( Q ) , ( S ) , ( V ) and ( W ) ; 30 μm in ( U ) . In order to understand how the prepupal pulse of TcKr-h1 prevents direct adult differentiation , we next analyzed the expression of TcBr-C and TcE93 in TcKr-h1i animals , as these factors are required for the correct formation of the pupal and adult forms , respectively [26–28 , 33] . In wild type T . castaneum , TcE93 mRNA levels start to increase moderately in the last larval instar to reach maximum levels during the pupal stage , while TcBr-C expression is restricted to a strong pulse during the prepupal stage ( S1 Fig ) [26–28 , 33] . Consistent with the precocious differentiation of adult features in TcKr-h1i animals , we found that TcE93 mRNA levels were prematurely up-regulated in these animals and that the prepupal peak of TcBr-C mRNA was strongly suppressed ( Fig 3B and 3C ) . These results suggest that the direct transition from larva to adult in TcKr-h1i animals stems from the premature up-regulation of TcE93 at the end of L7 and the concomitant repression of TcBr-C . In light of our previous study showing that TcE93 represses TcBr-C expression in the pupal stage [33] , our findings also suggest that the repression of TcBr-C in the TcKr-h1i animals may depend on the untimely increase of TcE93 . If both suggestions were correct , then depleting TcKr-h1 and TcE93 simultaneously in L7 would be sufficient to impair premature adult differentiation and to allow the normal induction of TcBr-C , thus redirecting the molt again to a normal pupal stage . Consequently , we injected TcKr-h1 and TcE93 dsRNAs into newly emerged last L7 instar larvae ( TcKr-h1i+TcE93i animals ) . Most of the TcKr-h1i+TcE93i animals pupated properly 6 days after the injection ( Fig 4A and 4B , S2 Table ) . Remarkably , all the appendages of these animals showed normal pupal-like morphology with no signs of precocious adult differentiation ( Fig 4A and 4B ) . In addition , some of the double knockdown animals arrested development at the prepupal-pupal transition ( Fig 4C and S2 Table ) . When the larval cuticle of these animals was removed , the morphology of the appendages was similar to those that had successfully pupated , including well-developed gin traps in the abdomen ( Fig 4C–4K ) . Consistent with the observed phenotype , TcKr-h1i+TcE93i prepupae presented a normal peak of TcBr-C mRNA and the premature up-regulation of the adult-specific TcCPR27 gene was completely prevented ( Fig 4L ) . Altogether , our results show that the transient pulse of TcKr-h1 at the end of the larval development prevents the direct transformation of larval tissues to adult ones by maintaining low levels of TcE93 during this period . In doing so , TcKr-h1 allows the strong up-regulation of TcBr-C and the occurrence of a new developmental stage , the pupa . Next , we asked to what extent the adult features observed in TcKr-h1i animals , consequence of the premature TcE93 up-regulation , were due to the TcE93-dependent repression of TcBr-C . To this aim , we injected TcBr-C dsRNA in newly molted L7 larvae ( TcBr-Ci animals ) to mimic the absence of the prepupal TcBr-C peak observed in TcKr-h1i animals . Consistent with previous reports [26–28] , the majority of the TcBr-Ci larvae arrested development at the end of the prepupal stage or just after the pupal molt showing a mix of larval , pupal and adult characters ( Fig 5A–5C and S3 Table ) . Detailed analyses of the TcBr-Ci specimens that undergo pupation revealed that , in addition to short and blister wings , abnormal urogomphi and absence of gin traps , their appendages , including antennae , maxillae , mandibles , and legs , presented an adult-like segmentation although with larval-like pigmentation ( S3 Fig; [26–28] ) . However , the extent of adult differentiation of TcBr-Ci animals was clearly weaker than that observed after TcKr-h1 removal ( Fig 1 ) , suggesting that the premature adult differentiation observed in TcKr-h1i specimens was not exclusively channeled through TcBr-C repression . We , then , measured the expression levels of TcE93 and TcKr-h1 in TcBr-Ci animals during the prepupal stage and found that the mRNA levels of both genes were similar to Control animals ( Fig 5D and 5E ) , indicating that phenotypic alterations observed in TcBr-Ci animals are not due to variations of TcE93 or TcKr-h1 levels . Taken together , our data suggest that high levels of TcBr-C , together with low levels of TcE93 , which depend of the TcKr-h1 peak , are required during the prepupal stage of T . castaneum to prevent premature adult differentiation and allow the proper formation of the pupa . Next , we investigated whether the metamorphic genetic circuitry is conserved in more derived holometabolous insects and turned to the dipteran D . melanogaster . To this end , we used the Gal4/UAS system [35] to modify the expression of the metamorphic genes . First , we knocked down DmKr-h1 expression in the whole animal using the UAS-DmKr-h1RNAi transgene driven by the ubiquitous ActinGal4 ( ActGal4 ) driver , and measured the levels of DmE93 and DmBr-C in white prepupa . Similar to T . castaneum , we found that depletion of DmKr-h1 lead to a significant increase of the mRNA levels of the two DmE93 isoforms , DmE93A and DmE93B , and to the concomitant decrease of DmBr-C mRNA levels ( Fig 6A ) . To confirm this result specifically in a metamorphic tissue , we overexpressed UAS-DmKr-h1RNAi specifically in the pouch region of the wing disc using the rotundGal4 ( rnGAL4 ) driver . As expected , depletion of DmKr-h1 led to a remarkable increase of DmE93A and DmE93B in the wing pouch during the prepupal stage ( Fig 6B ) , and to the disappearance of DmBr-C protein ( Fig 6C ) . However , unlike T . castaneum where the viability of the wings was not affected ( S4 Fig ) , DmKr-h1-depleted wings showed clear signs of necrosis at the pupal stage ( Fig 6D ) , probably due to a deficient wing evertion . The wing phenotype is consistent with previous reports where imaginal discs failed to elongate properly in D . melanogaster Br-C mutants [23 , 30 , 36] . To further confirm the repression of DmBr-C , at the protein level , UAS-DmKr-h1RNAi was specifically expressed in the anterior compartment of the wing disc by using the cubitus interruptusGAL4 ( CiGal4UASGFP ) driver , and its effect there was compared with the control posterior compartment . As Fig 7A shows , depletion of DmKr-h1 with CiGal4UASGFP drastically reduced DmBr-C protein levels in the anterior compartment . The disappearance of DmBr-C was not due to reduced viability of the DmKr-h1-depleted cells as they showed normal protein levels of Spalt , a protein whose expression is independent of either DmKr-h1 or DmE93 ( Fig 7B ) . The same result was obtained when DmKr-h1RNAi clones were generated in the wing disc ( Fig 7C ) , confirming again the cell autonomous repression of DmBr-C . Next , we analyzed whether the disappearance of DmBr-C was due to the precocious upregulation of DmE93 . To do this , we depleted both DmKr-h1 and DmE93 simultaneously in the anterior compartment of the wing pouch using the CiGal4UASGFP driver . Under these conditions , the levels of DmBr-C protein returned to normal ( Fig 7D ) , allowing the normal evertion of the wing disc . Likewise , when DmKr-h1 and DmE93 were depleted in the wing pouch under the control of the rnGal4 driver , the wings did not degenerate and everted properly , and the adult flies emerged with small and undifferentiated wings , a phenotype similar to that observed in DmE93-depleted adult wings ( Fig 7E ) . Altogether , our results show that the regulatory interactions between the metamorphic toolkit genes are conserved in more derived holometabolous insects . Given that Holometaboly evolved from hemimetabolous ancestors some 345 Ma [10] , we finally sought to determine whether the effect of Kr-h1 in preventing precocious adult metamorphosis through E93 is present also in hemimetabolous insects . To this end , we turned to the German cockroach Blattella germanica as a model of hemimetabolous development . B . germanica undergoes six nymphal instars ( N1-N6 ) before molting into the adult . In contrast to holometabolous insects , B . germanica metamorphosis occurs during a single period , the last ( N6 ) nymphal instar , and is restricted to the transformation of the wing primordia into functional wings , to the acquisition of functional genitalia and to marked changes in cuticle pigmentation [33 , 37] . In agreement with previous data [19] , RNAi-mediated depletion of BgKr-h1 in penultimate ( N5 ) instar nymphs ( BgKr-h1i animals ) caused precocious differentiation of adult features after the ensuing molt ( Fig 8A and 8B , S4 Table ) . Precocious BgKr-h1i adults were smaller but had all the external characteristics of a normal adult: functional hind- and forewings , adult cerci , and adult-specific pigmentation of the cuticle ( Fig 8B ) . Consistent with the phenotype observed , BgKr-h1i N5 nymphs presented a significant precocious up-regulation of BgE93 when compared to Control nymphs ( Fig 8D ) . This result was consistent with a previous report [38] , although in their experiments , BgE93 levels were measured specifically in the tergal gland of male nymphs . Interestingly , as happened in holometabolous insects , we also found that BgBr-C mRNA levels were strongly reduced in BgKr-h1i N5 nymphs ( Fig 8E ) . To confirm that the premature activation of the adult genetic program in BgKr-h1i animals depends on the precocious up-regulation of BgE93 , as in T . castaneum TcKr-h1i prepupae , we next depleted BgKr-h1 and BgE93 simultaneously in penultimate N5 nymphs . As expected , double RNAi for BgKr-h1 and BgE93 ( BgKr-h1i+BgE93i animals ) in newly emerged N5 nymphs resulted in normal N6 nymphs after the following molt instead of undergoing precocious metamorphosis ( Fig 8C and S4 Table ) . BgKr-h1i+BgE93i N6 nymphs presented all the morphological characteristics of a nymph: black cuticle , two thick stripes of black melanin in the pronotum , nymphal cerci and external wing pads ( Fig 8C ) . BgKr-h1i+BgE93i N6 nymphs kept molting into supernumerary nymphal stages until reaching N10 when they arrested development due to problems in shedding the exuvia . The inability to undergo metamorphosis of BgKr-h1i+BgE93i nymphs is consistent with our recent observation [33] , where RNAi-mediated depletion of BgE93 in nymphs of B . germanica prevented the nymphal-adult transition and caused endless reiterations of nymphal molts . Notably , BgBr-C mRNA levels in BgKr-h1i+BgE93i N6 nymphs were similar to Control nymphs ( Fig 8E ) , confirming that the high levels of BgE93 in BgKr-h1i N5 nymphs are responsible for the down-regulation of BgBr-C expression . Finally , we aimed to determine whether the BgKr-h1-dependent repression of BgE93 observed in N5 nymphs is also present in younger nymphal instars . To do this , we injected newly eclosed N4 nymphs with BgKr-h1 dsRNA and assessed the expression of BgE93 in these animals . As shown in Fig 9C , BgE93 mRNA levels were not up-regulated in BgKr-h1i N4 nymphs . Consequently , all BgKr-h1i N4 nymphs molted properly into normal N5 nymphs ( Fig 9A and 9B , S5 Table ) and became premature adults in the ensuing molt instead of molting into N6 nymphs as the Control animals , which is quite consistent with previous observations [19] . Interestingly , BgBr-C mRNA levels in BgKr-h1i N4 nymphs were significantly reduced ( Fig 9D ) despite BgE93 not being up-regulated in these animals , indicating that BgKr-h1 is necessary to maintain BgBr-C expression during the antepenultimate N4 stage , which correlates with previous studies that show that JH enhances Br-C expression in hemimetabolous insects [18 , 24 , 25] . Overall , these data show that 1 ) the anti-metamorphic effect of BgKr-h1 during the penultimate N5 instar is channeled through the repression of BgE93 , suggesting that the metamorphic landscape of hemimetabolous insects has offered the substrate for the evolution of complete metamorphosis and the occurrence of the holometabolan pupa; and 2 ) the capacity of BgKr-h1 to prevent BgE93 up-regulation and , hence , premature adult differentiation in B . germanica is restricted to the penultimate N5 nymphal stage .
How complete metamorphosis is controlled at the molecular level is a critical question towards understanding how Holometaboly has evolved from hemimetabolan ancestors . Previous studies have revealed that metamorphosis in both types of insects requires the down-regulation of the anti-metamorphic Kr-h1 and the up-regulation of the adult specifier E93 transcription factor genes [18–20 , 33] . However , a difference arises in holometabolous insects with the transient pulse of Kr-h1 at the end of the final larval stage , absent in hemimetabolous nymphs . Based on our data , we propose a model through which this late peak of Kr-h1 might allow the appearance of the new holometabolan-specific pupal stage ( Fig 10A ) . This model is based on the TcKr-h1-dependent repression of TcE93 expression once metamorphosis has been initiated in the prepupal period ( Fig 10A ) . We propose that the low levels of TcE93 at this particular stage of development are essential for two reasons; first , it prevents the direct transformation of the larva into the adult , as it happens in the last nymphal instar of hemimetabolous insects . Second , it allows the stage-specific burst of TcBr-C expression in the prepupal stage , as TcE93 is a potent repressor of TcBr-C expression ( Figs 3 and 7; [33] ) . This pulse of TcBr-C is crucial to coordinate the morphogenesis of the different body parts that would give rise to the normal pupal morphology , as evidenced by the occurrence of larva-pupa-adult mosaics in TcBr-C-depleted animals [26–28] . Thus , as a result of the crosstalk between TcKr-h1 , TcE93 and TcBr-C in the prepupal stage , the larva molts into a new metamorphic stage , the pupa , instead of undergoing the terminal adult molt . Finally , our previous results showed that after the pupal molt , high levels of TcE93 repress TcKr-h1 and TcBr-C expression , ensuring the completion of the metamorphic process ( Fig 10A ) [33] . From an evolutionary perspective , our results strongly suggest that the metamorphic genetic landscape of hemimetabolous insects has served as the substrate for the evolution of complete metamorphosis . First , RNAi analysis in the hemimetabolous insect B . germanica ( Figs 8 and 9 ) , coupled with previous results [33] , reveals the fundamental conservation of the functional interactions between the metamorphic network genes Kr-h1 , E93 and Br-C in hemimetabolous and holometabolous insects ( Fig 10B ) . Second , in the absence of the prepupal pulse of TcKr-h1 ( TcKr-h1i animals ) , the expression dynamics of the metamorphic network genes during the last larval instar of T . castaneum closely resemble those in the last nymphal instar of hemimetabolous insects ( Fig 10C ) . Therefore , it seems likely that the transient re-induction of Kr-h1 midway through the metamorphic process in holometabolous ancestors , with the consequent redeployment of the metamorphic toolkit circuit , occurred in the origin of the pupal stage , transforming the single-stage metamorphic period of hemimetabolous insects into the two-stages process of holometabolous insects . Consistent with this scenario , the expression profile of Kr-h1 during the post-embryonic development in thrips ( Thysanoptera ) , one of the closest hemimetabolous relatives of holometabolous insects that present quiescent and non-feeding stages called propupa and pupa ( Neometaboly ) , are comparable to those in holometabolous insects [39] . Based on our results , we thus propose that the two metamorphic periods of holometabolous insects , the last larval instar and the pupal period , are ontogenetically homologous to the last nymphal instar of hemimetabolous insects . Although the regulatory architecture between the metamorphic toolkit genes is mostly conserved in winged insects , as are the metamorphic functions of Kr-h1 and E93 , the specific role of Br-C in relation to metamorphosis has dramatically changed during the evolution of holometaboly . While Br-C does not exert any metamorphic role in hemimetabolous insects , and its function is mainly limited to the control of wing development , particularly in relation to size , form and vein formation [18 , 24 , 25] , in holometabolous insects Br-C has been specifically recruited for new stage-specific metamorphic functions [26–29 , 31 , 32] . Therefore , Br-C acts as the pupal coordinator that ensures proper pupal morphogenesis and prevents premature adult morphogenesis in the holometabolan context ( Fig 5 ) . The acquisition by Br-C of new metamorphic functions has been favored , in part , by changes in its expression , from being constantly expressed during embryogenesis and throughout nymphal development in Hemimetaboly to be confined to the strong prepupal-specific pulse characteristic of holometabolous insects [20 , 26 , 27 , 40–42] . Two events have probably facilitated this change . First , the conserved repressive activity of E93 upon Br-C expression , already present in hemimetabolous insects , that ensured the repression of Br-C during the pupal stage [33] . Second , a shift in the JH regulatory activity on Br-C expression , from being inductive in hemimetabolous insects [18 , 24 , 25] to repressive in holometabolous insects [40 , 43 , 44] . The JH-dependent repression of Br-C has restricted the expression of this factor in young larvae until the onset of the last larval instar , when the decline of JH and the temporal disappearance of Kr-h1 allow the induction of Br-C by the ecdysteroid hormone 20-hydroxyecdysone ( 20E ) [40 , 41 , 45] . Recently , it has been shown in B . mori that the early larval repression of Br-C depends on the direct binding of Kr-h1 to the Br-C gene [46] . Paradoxically , the repressive activity of Kr-h1 on Br-C expression does not occur during the prepupal stage of holometabolous insects as the strong prepupal pulse of Br-C parallels that of Kr-h1 . Consistent with this observation , the overexpression of BmKr-h1 during the prepupal stage in transgenic B . mori cannot prevent the normal appearance of the BmBr-C pulse [47] . According to that , our data demonstrate that the prepupal surge of Kr-h1 is required to allow the normal expression of Br-C through the repression of E93 , which suggests that the regulatory activity of Kr-h1 upon Br-C expression is stage-specific , from directly inhibiting its expression in young larvae to allowing its induction in the prepupal stage through the repression of E93 . Further studies are needed to reveal the molecular mechanisms through which Br-C is not repressed by Kr-h1 in the prepupal stage . Given the functional relevance of the reappearance of Kr-h1 during the prepupal period , it remains to be established what is the precise signal that controls it . As Kr-h1 expression is induced by JH in hemimetabolous and holometabolous species [18–20 , 48–51] , it is plausible that a prepupal pulse of circulating JH controls TcKr-h1 upregulation . Consistent with this possibility , allatectomy ( the surgical elimination of the gland that synthesizes JH , the corpora allata ) in the final larval instar of the lepidopterans Manduca sexta and Hyalophora cecropia caused partial premature adult development [52 , 53] . However , the double knockdown of TcJHAMT and TcCYP15A1 , the two enzymes that catalyze the final two steps of the JH biosynthetic pathway in T . castaneum , does not cause precocious differentiation of adult structures [54] . Likewise , allatectomized Bombyx mori larvae developed into normal pupa , which indicates that the prepupal BmKr-h1 pulse is corpora allata-independent [47] . As BmKr-h1 expression in vitro is up-regulated by JH , it is possible that BmKr-h1 could be induced by JH synthesized from other tissues than the corpora allata [47] . Unlike lepidopteran species , Kr-h1 expression in D . melanogaster is not only induced by JH but also by 20E [51 , 55] . Overall , these data suggest that the critical prepupal up-regulation of Kr-h1 in holometabolous insects is controlled by a combination of factors , including JH and 20E , still to be clearly identified . On the other hand , it is of great interest to know how E93 expression is regulated . While the repressive activity of Kr-h1 on E93 expression is a common trait of hemimetabolous and holometabolous insects ( Figs 3 and 8; [38] ) , the signals that induce E93 expression have only been characterized in two holometabolous insects , D . melanogaster and B . mori . In the fly , DmE93 expression is induced by 20E [56 , 57] . Likewise , in B . mori , it has been shown that BmE93 is also induced by 20E and repressed by JH [58] . Given the relevance of such regulation , future studies should investigate the molecular basis underlying the regulation of E93 expression in hemimetabolous and holometabolous insects . In conclusion , we have established the critical stage-specific interactions between the metamorphic toolkit genes that underlie the formation of the pupa in holometabolous insects . Although the full details of the origin of the holometabolan pupa still remain to be determined , our results provide a molecular framework to explain how complete metamorphosis is regulated , thus shedding light into the evolution of complete metamorphosis .
Wild-type T . castaneum strain and the enhancer-trap line pu11 ( obtained from Y . Tomoyasu , Miami University , Oxford , Ohio ) were reared on organic wheat flour containing 5% nutritional yeast , and maintained at 29°C in constant darkness . Flies were raised on standard D . melanogaster medium at 25°C , unless otherwise required . Oregon R flies ( OR-R , used as a wild type control ) , ActGAL4 , rnGal4 , CiGAL4 , UASGFP and UASdicer ( used to enhance RNAi effectiveness ) were obtained from the Bloomington Stock Center ( BDSC ) . UAS-DE93RNAi ( KK108140; GD4449 ) , and UAS-DKr-h1RNAi ( KK107935; GD51282 ) are from the Vienna Drosophila RNAi Center ( VDCR ) . For clonal analysis , hsflp;Tub>y>Gal4;UASGFP females were crossed with males carrying UASKr-h1RNAi . Embryos were kept at 25°C until late L2 , incubated 1 hour at 37°C and transferred to 25°C until late L3 . B . germanica specimens were reared in the dark at 30 ± 1°C and 60–70% relative humidity . Total RNA was isolated with the GenElute Mammalian Total RNA kit ( Sigma ) , DNAse treated ( Promega ) and reverse transcribed with Superscript II reverse transcriptase ( Invitrogen ) and random hexamers ( Promega ) . To obtain cDNA from the wing pouch of rnGal4 and rnGal4>UASDmKr-h1RNAi animals , the wing pouch was specifically separated from the rest of the wing . Relative transcripts levels were determined by real-time PCR ( qPCR ) , using Power SYBR Green PCR Mastermix ( Applied Biosystems ) . To standardize the qPCR inputs , a master mix that contained Power SYBR Green PCR Mastermix and forward and reverse primers was prepared ( final concentration: 100nM/qPCR ) . The qPCR experiments were conducted with the same quantity of tissue equivalent input for all treatments and each sample was run in duplicate using 2 μl of cDNA per reaction . All the samples were analyzed on the iCycler iQ Real Time PCR Detection System ( Bio-Rad ) . For each standard curve , one reference DNA sample was diluted serially . Primers sequences for qPCR analyses were: T . castaneum: TcE93: TcE93-F: 5’-CTCTCGAAAACTCGGTTCTAAACA-3’ TcE93-R: 5’-TTTGGGTTTGGGTGCTGCCGAATT-3’ TcBr-C: TcBr-C-F: 5’-TCGTTTCTCAAGACGGCTGAAGTG-3’ TcBr-C-R: 5’-CTCCACTAACTTCTCGGTGAAGCT-3’ TcKr-h1: TcKr-h1-F: 5’-AAGAAGAGCATGGAAGCACACATT-3’ TcKr-h1-R: 5’-GAATCGTAGCTAAGAGGGTCTTGA-3’ TcCPR27: TcCPR27-F: 5’-AGGTTACGGCCATCATCACTTGGA-3’ TcCPR27-R: 5’-ATTGGTGGTGGAAGTCATGGGTGT-3’ TcRpL32: TcRpL32-F: 5’-CAGGCACCAGTCTGACCGTTATG-3’ TcRpL32-R: 5’-CATGTGCTTCGTTTTGGCATTGGA-3’ D . melanogaster: DmE93A: DmE93A-F: 5’-CACATCAGCAGCTATGAAATA-3' mDE93A-R: 5’- AACCGGCTATTGCTATGGGCTGTT-3' DmE93B: DmE93B-F: 5’-TCCACAGATATGCTGCATATTGTG-3' DmE93A-R: 5’- AACCGGCTATTGCTATGGGCTGTT-3' DmBr-C: DmBr-C-F: 5’-CATCTGGCTCAGATACAGAACCT-3’ DmBr-C-R: 5’-CTTCAGCAGCTGGTTGTTGATGT-3’ DmRpL32: DmRpL32-F: 5’-CAAGAAGTTCCTGGTGCACAA-3’ DmRpL32-R: 5’-AAACGCGGTTCTGCATGAG-3’ B . germanica: BgE93: BgE93-F: 5’-CAAGCGGGGCAAATATCGCAATTA-3’ BgE93-R: 5’-TGACCTTGTACTCGAGTGTGG-3’ BgBr-C: BgBr-C-F: 5’-CTTAAAGCTCATAGAGTGGTGTTG-3’ BgBr-C-R: 5’-CACTTCACCATGGTATATGAATTC-3’ T . castaneum—RNAi in vivo was performed as previously described [33 , 59] . Control dsRNA consisted of a non-coding sequence from the pSTBlue-1 vector ( dsControl ) . For the in vivo treatment , dsRNAs , concentrated up to 4 µg/µl , were injected into the abdomen of last instar larvae ( L7 ) of the pu11 line . B . germanica–RNAi in vivo was performed as previously described [60 , 61] . A dose of 1 µl ( 5–8 µg/µl ) of the dsRNA solution was injected into the abdomen of newly ecdysed penultimate ( N5 ) or antepenultimate ( N4 ) instar nymphs , and left until analysed . In case of coinjection of two dsRNAs in T . castaneum or B . germanica , the same volume of each dsRNA solution was mixed and applied in a single injection . To maintain the RNAi effect during the successive nymphal instars , the same dose of dsRNAs was reapplied to all treated animals after molting into new nymphal stages . The primers used to generate templates via PCR for transcription of the dsRNAs were: T . castaneum: dsTcKr-h1: dsTcKr-h1-F: 5’‐ AATCCTCCTGCTCATCCAGCACTA-3’ dsTcKr-h1-R: 5’‐ CAGGATTCGAACTAGGAGGTGTTA-3’ dsTcE93: dsTcE93-F: 5’-AAATAACGGTGATACAGTGTCAAG-3’ dsTcE93-R: 5’-TTGTAGTCCATCTCGGAGATGGAA-3’ dsTcBr-C: dsTcBr-C-F: 5’-CAATTACCAAAGCAGCATCACATC-3’ dsTcBr-C-R: 5’-GGCTTTGTACTTGCGCCAACTGTT-3’ B . germanica: dsBgKr-h1: dsBgKr-h1-F: 5’- GAATCTCAGTGTGCATAGGCG-3’ dsBgKr-h1-R: 5’- CCTTGCCACAAATGACACAA-3’ dsBgE93: dsBgE93-F: 5’-GAAACAGAACCTCCTTTCAAAAGG-3’ dsBgE93-R: 5’-AAAGTGTGAACCTGCCCGATGAA-3’ T . castaneum dissections were carried out in Ringer’s saline and the different appendages were mounted directly in Glycerol 70% . For D . melanogaster immunohistochemistry , wings from 0 h after puparium formation animals were collected and stained as described [33] . Antibodies: mouse anti-Broad-Complex ( 1:100; Developmental Studies Hybridoma Bank ( DSHB ) ) ; rat anti-Spalt ( Sal ) ( 1:200; a gift from R . Barrio ) ; rabbit anti-Caspase 3 ( 1:250; Cell Signaling Technologies ) ; Alexa Fluor 555-conjugated secondary antibodies ( 1:200; Molecular Probes ) . All samples were examined with AxioImager . Z1 ( ApoTome 213 System , Zeiss ) microscope , and images were subsequently processed using Adobe photoshop . Control and TcE93i animals of T . castaneum were carefully taken out of the larval cuticle with forceps when necessary . Then , they were fixed in 80% ethanol , and dehydrated with a series of graded ethanol solutions ( 90% , 95% and 100% ) for 15 min in each solution , critical-point dried using CO2 , sputter-coated with gold-palladium , and observed under a Hitachi S-3500N scanning electron microscope .
|
Complete metamorphosis is an evolutionary innovation that has been critical for the success of insects . Phylogenetic relationships reveal that holometabolous insects evolved from ancestors displaying hemimetabolous development . Yet , little is known about the molecular nature of the changes required for such transition , including the evolution of the holometabolan-specific pupal stage . Here , by using Tribolium castaneum , we report that the crosstalk between Krüppel-homolog 1 ( Kr-h1 ) , E93 and Broad-Complex genes at the end of the larval development has been a key event underlying the formation of the pupa . Interestingly , we show that the interaction between Kr-h1 and E93 is also present in hemimetabolous insects , suggesting that the pupal stage has evolved by the co-option of regulatory mechanisms already present in hemimetabolous insects .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"invertebrates",
"insect",
"metamorphosis",
"animals",
"developmental",
"biology",
"animal",
"anatomy",
"nymphs",
"pupae",
"hemimetabolism",
"zoology",
"wings",
"holometabolism",
"insects",
"arthropoda",
"entomology",
"metamorphosis",
"biology",
"and",
"life",
"sciences",
"larvae",
"organisms"
] |
2016
|
The Occurrence of the Holometabolous Pupal Stage Requires the Interaction between E93, Krüppel-Homolog 1 and Broad-Complex
|
Standard approaches to data analysis in genome-wide association studies ( GWAS ) ignore any potential functional relationships between gene variants . In contrast gene pathways analysis uses prior information on functional structure within the genome to identify pathways associated with a trait of interest . In a second step , important single nucleotide polymorphisms ( SNPs ) or genes may be identified within associated pathways . The pathways approach is motivated by the fact that genes do not act alone , but instead have effects that are likely to be mediated through their interaction in gene pathways . Where this is the case , pathways approaches may reveal aspects of a trait's genetic architecture that would otherwise be missed when considering SNPs in isolation . Most pathways methods begin by testing SNPs one at a time , and so fail to capitalise on the potential advantages inherent in a multi-SNP , joint modelling approach . Here , we describe a dual-level , sparse regression model for the simultaneous identification of pathways and genes associated with a quantitative trait . Our method takes account of various factors specific to the joint modelling of pathways with genome-wide data , including widespread correlation between genetic predictors , and the fact that variants may overlap multiple pathways . We use a resampling strategy that exploits finite sample variability to provide robust rankings for pathways and genes . We test our method through simulation , and use it to perform pathways-driven gene selection in a search for pathways and genes associated with variation in serum high-density lipoprotein cholesterol levels in two separate GWAS cohorts of Asian adults . By comparing results from both cohorts we identify a number of candidate pathways including those associated with cardiomyopathy , and T cell receptor and PPAR signalling . Highlighted genes include those associated with the L-type calcium channel , adenylate cyclase , integrin , laminin , MAPK signalling and immune function .
Much attention continues to be focused on the problem of identifying SNPs and genes influencing a quantitative or dichotomous trait in genome wide scans [1] . Despite this , in many instances gene variants identified in GWAS have so far uncovered only a relatively small part of the known heritability of most common diseases [2] . Possible explanations include the presence of multiple SNPs with small effects , or of rare variants , which may be hard to detect using conventional approaches [2]–[4] . One potentially powerful approach to uncovering the genetic etiology of disease is motivated by the observation that in many cases disease states are likely to be driven by multiple genetic variants of small to moderate effect , mediated through their interaction in molecular networks or pathways , rather than by the effects of a few , highly penetrant mutations [5] . Where this assumption holds , the hope is that by considering the joint effects of variants acting in concert , pathways GWAS methods will reveal aspects of a disease's genetic architecture that would otherwise be missed when considering variants individually [6] , [7] . In this paper we describe a sparse regression method utilising prior information on gene pathways to identify putative causal pathways , along with the constituent variants that may be driving pathways association . Sparse modelling approaches are becoming increasingly popular for the analysis of genome wide datasets [8]–[11] . Sparse regression models enable the joint modelling of large numbers of SNP predictors , and perform ‘model selection’ by highlighting small numbers of variants influencing the trait of interest . These models work by penalising or constraining the size of estimated regression coefficients . An interesting feature of these methods is that different sparsity patterns , that is different sets of genetic predictors having specified properties , can be obtained by varying the nature of this constraint . For example , the lasso [12] selects a subset of variants whose main effects best predict the response . Where predictors are highly correlated , the lasso tends to select one of a group of correlated predictors at random . In contrast , the elastic net [13] selects groups of correlated variables . Model selection may also be driven by external information , unrelated to any statistical properties of the data being analysed . For example , the fused lasso [14] , [15] uses ordering information , such as the position of genomic features along a chromosome to select ‘adjacent’ features together . Prior information on functional relationships between genetic predictors can also be used to drive the selection of groups of variables . In the present context , information mapping genes and SNPs to functional gene pathways has recently been used in sparse regression models for pathway selection . Chen et al . [16] describe a method that uses a combination of lasso and ridge regression to assess the significance of association between a candidate pathway and a dichotomous ( case-control ) phenotype , and apply this method in a study of colon cancer etiology . In contrast , Silver et al . [17] use group lasso penalised regression to select pathways associated with a multivariate , quantitative phenotype characteristic of structural change in the brains of patients with Alzheimer's disease . In identifying pathways associated with a trait of interest , a natural follow-up question is to ask which SNPs and/or genes are driving pathway selection ? We might further ask a related question: can the use of prior information on putative gene interactions within pathways increase power to identify causal SNPs or genes , compared to alternative methods that disregard such information ? One way to answer these questions is by conducting a two-stage analysis , in which we first identify important pathways , and then in a second step search for SNPs or genes within selected pathways [18] , [19] . There are however a number of problems with this approach . Firstly , highlighted variants are then not necessarily those that were driving pathway selection in the first step of the analysis . Secondly , the implicit ( and reasonable ) assumption is that only a small number of SNPs in a pathway are driving pathway selection , so that ideally we would prefer a model that has this assumption built in . The above considerations point to the use of a ‘dual-level’ sparse regression model that imposes sparsity at both the pathway and SNP level . Such a model would perform simultaneous pathway and SNP selection , with the additional benefit of being simpler to implement . A suitable sparse regression model enforcing the required dual-level sparsity is the sparse group lasso ( SGL ) [20] . SGL is a comparatively recent development in sparse modelling , and in simulations has been shown to accurately recover dual-level sparsity , in comparison to both the group lasso and lasso [20] , [21] . SGL has been used for the identification of rare variants in a case-control study by grouping SNPs into genes [22]; for the identification of genomic regions whose copy number variations have an impact on RNA expression levels [23]; and to model geographical factors driving climate change [24] . SGL can be seen as fitting into a wider class of structured-sparsity inducing models that use prior information on relationships between predictors to enforce different sparsity patterns [25]–[27] . Hierarchical and mixed effect modelling approaches have also been suggested as a means of leveraging pathways information for the simultaneous identification of SNPs or genes within associated pathways . Brenner et al . [28] propose such a method for identifying SNPs in a priori selected candidate pathways by comparing results from multiple studies in a meta-analysis . This approach is similar in motivation to the two-stage methods described above . The method proposed by Wang et al . [29] is closer in spirit to our own , in that it provides measures of pathway significance , and also ranks genes within pathways . Both of these methods however use results from univariate tests of association at each gene variant as input to the models , in contrast to our joint-modelling approach . Here we describe a method for sparse , pathways-driven SNP selection that extends earlier work using group lasso penalised regression for pathway selection . This latter method was previously shown to offer improved power and specificity for identifying associated pathways , compared with a widely-used alternative [30] . In following sections we describe our method in detail , and demonstrate through simulation that the incorporation of prior information mapping SNPs to gene pathways can boost the power to detect SNPs and genes associated with a quantitative trait . We further describe an application study in which we investigate pathways and genes associated with serum high-density lipoprotein cholesterol ( HDLC ) levels in two separate cohorts of Asian adults . HDLC refers to the cholesterol carried by small lipoprotein molecules , so called high density lipoproteins ( HDLs ) . HDLs help remove the cholesterol aggregating in arteries , and are therefore protective against cardiovascular diseases [31] . Serum HDLC levels are genetically heritable [32] . GWAS studies have now uncovered more than 100 HDLC associated loci ( see www . genome . gov/gwastudies , Hindorff et al . [33] ) . However , considering serum lipids as a whole , variants so far identified account for only 25–30% of the genetic variance , highlighting the limited power of current methodologies to detect hidden genetic factors [34] .
We arrange the observed values for a univariate quantitative trait or phenotype , measured for N unrelated individuals , in an response vector . We assume minor allele counts for P SNPs are recorded for all individuals , and denote by the minor allele count for SNP j on individual i . These are arranged in an genotype design matrix . Phenotype and genotype vectors are mean centred , and SNP genotypes are standardised to unit variance , so that , for . We assume that all P SNPs may be mapped to L groups or pathways , , , and begin by considering the case where pathways are disjoint or non-overlapping , so that for any . We denote the vector of SNP regression coefficients by , and additionally denote the matrix containing all SNPs mapped to pathway by , where , is the column vector of observed SNP minor allele counts for SNP j , and is the number of SNPs in . We denote the corresponding vector of SNP coefficients by . In general , where P is large , we expect only a small proportion of SNPs to be ‘causal’ , in the sense that they exhibit phenotypic effects . A key assumption in pathways analysis is that these causal SNPs will tend to be enriched within a small set , , of causal pathways , with , where denotes the size ( cardinality ) of . We denote the set of causal SNPs mapping to pathway by , and make the further assumption that most SNPs in a causal pathway are non-causal , so that , where denotes the size ( cardinality ) of . A suitable sparse regression model imposing the required , dual-level sparsity pattern is the sparse group lasso ( SGL ) . We illustrate the resulting causal SNP sparsity pattern in Figure 1 , and compare it to that generated by the group lasso ( GL ) , a group-sparse model that we used previously in a sparse regression method to identify gene pathways [17] , [30] . With the SGL [20] , sparse estimates for the SNP coefficient vector , are given by ( 1 ) where and are parameters controlling sparsity , and is a pathway weighting parameter that may vary across pathways . ( 1 ) corresponds to an ordinary least squares ( OLS ) optimisation , but with two additional constraints on the coefficient vector , , that tend to shrink the size of , relative to OLS estimates . One constraint imposes a group lasso-type penalty on the size of . Depending on the values of and , this penalty has the effect of setting multiple pathway SNP coefficient vectors , , thereby enforcing sparsity at the pathway level . Pathways with non-zero coefficient vectors form the set of ‘selected’ pathways , so thatA second constraint imposes a lasso-type penalty on the size of . Depending on the values of and , for a selected pathway , this penalty has the effect of setting multiple SNP coefficient vectors , , thereby enforcing sparsity at the SNP level within selected pathways . SNPs with non-zero coefficient vectors then form the set of selected SNPs in pathway l , so thatThe set of all selected SNPs is given byThe sparsity parameter controls the degree of sparsity in , such that the number of pathways and SNPs selected by the model increases as is reduced from a maximal value , above which . The parameter controls how the sparsity constraint is distributed between the two penalties . When , ( 1 ) reduces to the group lasso , so that sparsity is imposed only at the pathway level , and all SNPs within a selected pathway have non-zero coefficients . When , solutions exhibit dual-level sparsity , such that as approaches 0 from above , greater sparsity at the group level is encouraged over sparsity at the SNP level . When , ( 1 ) reverts to the lasso , so that pathway information is ignored . For the estimation of we proceed by noting that the optimisation ( 1 ) is convex , and ( in the case of non-overlapping groups ) that the penalty is block-separable , so that we can obtain a solution using block , or group-wise coordinate gradient descent ( BCGD ) [35] . A detailed derivation of the estimation algorithm is given in the accompanying Supplementary Information S1 , Section 3 . From ( S . 9 ) and ( S . 10 ) , the criterion for selecting a pathway l is given by ( 2 ) and the criterion for selecting SNP j in selected pathway l by ( 3 ) where and are respectively the pathway and SNP partial residuals , obtained by regressing out the current estimated effects of all other pathways and SNPs respectively . The complete algorithm for SGL estimation using BCGD is presented in Box 1 . We test the hypothesis that where causal SNPs are enriched in a given pathway , pathway-driven SNP selection using SGL will outperform simple lasso selection that disregards pathway information in a simple simulation study . We simulate genetic markers for individuals . Marker frequencies for each SNP are sampled independently from a multinomial distribution following a Hardy Weinberg equilibrium frequency distribution . SNP minor allele frequencies are sampled from a uniform distribution . SNPs are distributed equally between 50 non-overlapping pathways , each containing 50 SNPs . We then test each competing method over 500 Monte Carlo ( MC ) simulations . At each simulation , a baseline univariate phenotype is sampled from . To generate genetic effects , we randomly select 5 SNPs from a single , randomly selected pathway , to form the set of causal SNPs . Genetic effects are then generated as described in Supplementary Information S1 , Section S3 . To enable a fair comparison between the two methods ( SGL and lasso ) , we ensure that both methods select the same number of SNPs at each simulation . We do this by first obtaining the SGL solution , , with and , which ensures sparsity at both the pathway and SNP level . We use a uniform pathway weighting vector . We then compute the lasso solution using coordinate descent over a range of values for the lasso regularisation penalty , , and choose the setwhere is the number of SNPs previously selected by SGL , and is the number of SNPs selected by the lasso with . We measure performance as the mean power to detect all 5 causal SNPs over 500 MC simulations , and test a range of genetic effect sizes ( see Supplementary Information S1 , Section S3 ) . In a follow up study , we compare the performance of the two methods in a scenario in which pathways information is uninformative . For this we repeat the previous simulations , but with 5 causal SNPs drawn at random from all 2500 SNPs , irrespective of pathway membership . Results are presented in Figure 2 . Referring to Figure 2 , we see that where causal SNPs are concentrated in a single causal pathway ( Figure 2 - left ) , SGL demonstrates greater power ( and equivalently specificity , since the total number of selected SNPs is constant ) , compared with the lasso , above a particular effect size threshold ( here ) . Where pathway information is not important , that is causal SNPs are not enriched in any particular pathway ( Figure 2 - right ) , SGL performs poorly . To gain a deeper understanding of what is happening here , we also consider the power distributions across all 500 MC simulations corresponding to each point in the plots of Figure 2 . These are illustrated in Figure 3 . The top row of plots illustrates the case where causal SNPs are drawn from a single causal pathway . Here we see that there is a marked difference between the two distributions ( SGL vs lasso ) . The lasso shows a smooth distribution in power , with mean power increasing with effect size . In contrast , with SGL the distribution is almost bimodal , with power typically either 0 or 1 , depending on whether or not the correct causal pathway is selected . This serves as an illustration of the advantage of pathway-driven SNP selection for the detection of causal SNPs in the case that pathways are important . As previously found by Zhou et al . [6] in the context of rare variants and gene selection , the joint modelling of SNPs within groups gives rise to a relaxation of the penalty on individual SNPs within selected groups , relative to the lasso . This can enable the detection of SNPs with small effect size or low MAF that are missed by the lasso , which disregards pathways information and treats all SNPs equally . Where causal SNPs are not enriched in a causal pathway ( bottom row of Figure 3 ) , as expected SGL performs poorly . In this case SGL will only select a SNP where the combined effects of constituent SNPs in a pathway are large enough to drive pathway selection . Finally , with many pathways methods an adjustment to pathway test statistics is made to account for biases due to variations in pathway size , that is the number of SNPs in a pathway [6] . We explore potential biases using SGL for pathway selection using the simulation framework described above , but this time allowing for varying pathway sizes , ranging from 10 to 200 SNPs . We find no evidence of a pathway size bias ( see Supplementary Information S1 , Section 5 for further details ) . We discuss the issue of accounting for pathway size and other potential biases in pathway and SNP selection when using real data in a later section . The assumption that pathways are disjoint does not hold in practice , since genes and SNPs may map to multiple pathways ( see ‘Pathway mapping’ section below ) . This means that typically for some . In the context of pathways-driven SNP selection using SGL , this has two important implications . Firstly , the optimisation ( 1 ) is no longer separable into groups ( pathways ) , so that convergence using coordinate descent is no longer guaranteed [35] . Secondly , we wish to be able to select pathways independently , and the SGL model as previously described does not allow this . For example consider the case of an overlapping gene , that is a gene that maps to more than one pathway . If a SNP mapping to this gene is selected in one pathway , then it must be selected in each and every pathway containing the mapped gene , so that all pathways mapping to the gene are selected . We instead want to admit the possibility that the joint SNP effects in one pathway may be sufficient to allow pathway selection , while the joint effects in another pathway containing some of the same SNPs do not pass the threshold for pathway selection . A solution to both these problems is obtained by duplicating SNP predictors in , so that SNPs belonging to more than one pathway can enter the model separately [30] , [36] . The process works as follows . An expanded design matrix is formed from the column-wise concatenation of the sub-matrices , , to form the expanded design matrix of size , where . The corresponding parameter vector , , is formed by joining the pathway parameter vectors , , so that . Pathway mappings with SNP indices in the expanded variable space are reflected in updated groups . The SGL estimator ( 1 ) , adapted to account for overlapping groups , is then given by ( 4 ) With this overlap expansion , the model is then able to perform pathway and SNP selection in the way that we require , and the corresponding optimisation problem is amenable to solution using the BCGD estimation algorithm described in Box 1 . However , for the purpose of pathways-driven SNP selection , the application of this algorithm presents a problem . This arises from the replication of overlapping SNP predictors in each group , , that they occur . Consider for example the simple situation where there are two pathways , , containing sets of causal SNPs and respectively . Here the indicates that SNP indices refer to the expanded variable space . We begin by assuming that and contain the same SNPs , so that in the unexpanded variable space , . We then proceed with BCGD by first estimating . We assume that the correct SNPs are selected , so that , and otherwise . For the estimation of , the estimated effect , of these overlapping causal SNPs is removed from the regression , through its incorporation in the block residual . Since no other causal SNPs exist in pathway , so that the criterion for pathway selection , ( 2 ) is not met . That is is not selected . Now consider the case where additional , non-overlapping causal SNPs , possibly with smaller effects , occur in , so that in the unexpanded variable space , . In other words , causal SNPs are partially overlapping ( see Figure 4 ) . This is the situation for example where multiple causal genes overlap both pathways , but one or more additional causal genes occur in . During BCGD pathway is then less likely to be selected by the model , than would be the case if there were no overlapping SNPs , since once again the effects of overlapping causal SNPs , , are removed . For pathways-driven SNP selection , we will argue that we instead require that SNPs are selected in each and every pathway whose joint SNP effects pass a revised pathway selection threshold , irrespective of overlaps between pathways . This is equivalent to the previous pathway selection criterion ( 2 ) , but with the additional assumption that pathways are independent , in the sense that they do not compete in the model estimation process . We describe a revised estimation algorithm under the assumption of pathway independence below . We justify the strong assumption of pathway independence with the following argument . In reality , we expect that multiple pathways may simultaneously influence the phenotype , and we also expect that many such pathways will overlap , for example through their containing one or more ‘hub’ genes , that overlap multiple pathways [37] , [38] . By considering each pathway independently , we aim to maximise the sensitivity of our method to detect these variants and pathways . In contrast , without the independence assumption , a competitive estimation algorithm will tend to pick out one from each set of similar , overlapping pathways , and miss potentially causal pathways and variants as a consequence . We illustrate this idea in the simulation study in the following section . One potential concern is that by not allowing pathways to compete against each other , specificity may be reduced , since too many pathways and SNPs may be selected . We discuss the issue of specificity further in the context of results from the simulation study . A detailed derivation of the SGL model estimation algorithm under the independence assumption is given in Supplementary Information S1 , Section 2 . The main results are that the pathway ( 2 ) and SNP ( 3 ) selection criteria become ( 5 ) respectively . The key difference is that partial derivatives and are replaced by , that is each pathway is regressed against the phenotype vector . This means that there is no block coordinate descent stage in the estimation , so that the revised algorithm utilises only coordinate gradient descent within each selected pathway . For this reason we use the acronym SGL-CGD for the revised algorithm , and SGL-BCGD for the previous algorithm using block coordinate gradient descent . The new algorithm is described in Box 2 . Finally , we note that for SNP selection we are interested only in the set of selected SNPs in the unexpanded variable space , and not the set . Since , under the independence assumption , the estimation of each does not depend on the other estimates , , we do not need to record separate coefficient estimates for each pathway in which a SNP is selected . Instead we need only record the set of SNPs selected in each selected pathway . This has a useful practical implication , since we can avoid the need for an expansion of or , and simply form the complete set of selected SNPs as We now explore some of the issues raised in the preceding section , specifically the potential impact on pathway and SNP selection power and specificity of treating the pathways as independent in the SGL estimation algorithm . We do this in a simulation study in which we simulate overlapping pathways . The simulation scheme is specifically designed to highlight differences in pathway and SNP selection with the independence assumption ( using the SGL-CGD estimation algorithm in Box 2 ) and without it ( using the standard SGL estimation algorithm in Box 1 ) . SNPs with variable MAF are simulated using the same procedure described in the previous simulation study , but this time SNPs are mapped to 50 overlapping pathways , each containing 30 SNPs . Each pathway overlaps any adjacent ( by pathway index ) pathway by 10 SNPs . This overlap scheme is illustrated in Figure 5 ( top ) . As before we consider a range of overall genetic effect sizes , . A total of 2000 MC simulations are conducted for each effect size . At MC simulation , we randomly select two adjacent pathways , where . From these two pathways we randomly select 10 SNPs according to the scheme illustrated in Figure 5 ( bottom ) . This ensures that causal SNPs overlap a minimum of 1 , and a maximum of 2 pathways , with . The true set of causal pathways , , is then given by , or ( although simulations where will be extremely rare ) . Genetic effects on the phenotype are generated as described previously ( Supplementary Information S1 , Section S3 ) . SNP coefficients are estimated for each algorithm , SGL-BCGD and SGL-CGD , using the same regularisation with and for both . The average number of pathways and SNPs selected by SGL-BCGD and SGL-CGD across all 2000 MC simulations is reported in Table 1 . As expected , for both models , the number of selected variables ( pathways or SNPs ) increases with decreasing effect size , as the number of pathways close to the selection threshold set by increases . For each model , at MC simulation we record the pathway and SNP selection power , and respectively . Since the number of selected variables can vary slightly between the two models , we also record false positive rates ( FPR ) for pathway and SNP selection as and respectively . The large possible variation in causal SNP distributions , causal SNP MAFs etc . makes a comparison of mean power and FPR between the two methods somewhat unsatisfactory . For example , depending on effect size , a large number of simulations can have either very high , or very low pathway and SNP selection power , masking subtle differences in performance between the two methods . Since we are specifically interested in establishing the relative performance of the two methods , we instead illustrate the number of simulations at which one method outperforms the other across all 2000 MC simulations , and show this in Figure 6 . In this figure , the number of simulations in which SGL-CGD outperforms SGL , i . e . where SGL-CGD power>SGL-BCGD power , or SGL-CGD FPR<SGL-BCGD FPR , are shown in green . Conversely , the number of simulations where SGL-BCGD outperforms SGL-CGD are shown in red . We first consider pathway selection performance ( top row of Figure 6 ) . For both methods , the same number of pathways are selected on average , across all effect sizes ( Table 1 ) . At low effect sizes , there is no difference in performance between the two methods for the large majority of MC simulations , and where there is a difference , the two methods are evenly balanced . As with SGL Simulation Study 1 , this is the region ( with ) where pathway selection fairs no better than chance . With , SGL-CGD consistently outperforms SGL , both in terms of pathway selection sensitivity and control of false positives ( measured by FPR ) . To understand why , we turn to SNP selection performance ( bottom row of Figure 6 ) . At small effect sizes , in the small minority of simulations where the correct pathways are identified , SGL-BCGD tends to demonstrate greater power than SGL-CGD ( Figure 6 bottom left ) . However , this is at the expense of lower specificity ( Figure 6 bottom right ) . These difference are due to the slightly larger number of SNPs selected by SGL-BCGD ( see Table 1 ) , which in turn is due to the ‘screening out’ of previously selected SNPs from the adjacent causal pathway during BCGD , as described previously . This results in the selection of a larger number of SNPs when any two overlapping pathways are selected by the model . In the case where two causal pathways are selected , SNP selection power is then likely to be higher , although at the expense of a greater number of false positives . When pathway effects are just on the margin of detectability , SGL-CGD is more often able to select both causal pathways , although this doesn't translate into increased SNP selection power . This is most likely because at this effect size neither model can detect SNPs with low MAF , so that SGL-CGD is detecting the same ( overlapping ) SNPs in both causal pathways . Note that once again SGL-BCGD typically has a higher FPR than SGL-CGD , since more SNPs are selected from non-causal pathways . As the effect size increases , the number of simulations in which SGL-CGD outperforms SGL-BCGD for SNP selection power grows , paralleling the former method's enhanced pathway selection power . This is again a demonstration of the screening effect with SGL-BCGD described previously . This means that SGL-CGD is more often able to select both causal pathways , and to select additional causal SNPs that are missed by SGL . These additional SNPs are likely to be those with lower MAF , for example , that are harder to detect with SGL , once the effect of overlapping SNPs are screened out during estimation using BCGD . Interestingly , as before SGL-CGD continues to exhibit lower false positive rates than SGL . This suggests that , with the simulated data considered here , the independence assumption offers better control of false positives by enabling the selection of causal SNPs in each and every pathway to which they are mapped . In contrast , where causal SNPs are successively screened out during the estimation using BCGD , too many SNPs with spurious effects are selected . The relative advantage of SGL-CGD over SGL-BCGD on all performance measures starts to decrease around , as SGL-BCGD becomes better able to detect all causal pathways and SNPs , irrespective of the screening effect . One issue that must be addressed is the problem of selection bias , by which we mean the tendency of SGL to favour the selection of particular pathways or SNPs under the null , where no SNPs influence the phenotype . Possible biasing factors include variations in pathway size or varying patterns of SNP-SNP correlations and gene sizes . Common strategies for bias reduction include the use of dimensionality reduction techniques and permutation methods [39]–[42] . In earlier work we described an adaptive weight-tuning strategy , designed to reduce selection bias in a group lasso-based pathway selection method [30] . This works by tuning the pathway weight vector , , so as to ensure that pathways are selected with equal probability under the null . This strategy can be readily extended to the case of dual-level sparsity with the SGL . Our procedure rests on the observation that for pathway selection to be unbiased , each pathway must have an equal chance of being selected . For a given , and with tuned to ensure that a single pathway is selected , pathway selection probabilities are then described by a uniform distribution , , for . We proceed by calculating an empirical pathway selection frequency distribution , , by determining which pathway will first be selected by the model as is reduced from its maximal value , , over multiple permutations of the response , . This process is described in detail in Supplementary Information S1 , Section 4 . We note that alternative methods for the construction of ‘null’ distributions , for example by permuting genotype labels , have been used in existing pathways analysis methods [6] . In the present context we choose to permute phenotype labels in order to preserve LD structure , since we expect this to be a significant source of bias with our data . Our iterative weight tuning procedure then works by applying successive adjustments to the pathway weight vector , , so as to reduce the difference , , between the unbiased and empirical ( biased ) distributions for each pathway . At iteration , we compute the empirical pathway selection probability distribution , determine for each pathway , and then apply the following weight adjustmentThe parameter controls the maximum amount by which each can be reduced in a single iteration , in the case that pathway l is selected with zero frequency . The square in the weight adjustment factor ensures that large values of result in relatively large adjustments to . Iterations continue until convergence , where . Note that when multiple pathways are selected by the model , the expected pathway selection frequency distribution under the null will not be uniform . This is because pathways overlap , so that selection frequencies will reflect the complex distribution of overlapping genes , as indeed will unbiased empirical selection frequencies . We have shown previously that this adaptive weight-tuning procedure gives rise to substantial gains in sensitivity and specificity with regard to pathway selection [30] . With most variable selection methods , a choice for the regularisation parameter , , must be made , since this determines the number of variables selected by the model . Common strategies include the use of cross validation to choose a value that minimises the prediction error between training and test datasets [43] . One drawback of this approach is that it focuses on optimising the size of the set , , of selected pathways ( more generally , selected variables ) that minimises the cross validated prediction error . Since the variables in will vary across each fold of the cross validation , this procedure is not in general a good means of establishing the importance of a unique set of variables , and can give rise to the selection of too many variables [44] , [45] . For the lasso , alternative approaches , based on data subsampling or bootstrapping have been shown to improve model consistency , in the sense that the correct model is selected with a high probability [45]–[47] . These methods work by recording selected variables across multiple subsamples of the data , and forming the final set of selected variables either as the intersection of variables selected at each model fit , or by assessing variable selection frequencies . Examples of the use of such approaches can be found in a number of recent gene mapping studies involving model selection using either the lasso or elastic net [9] , [19] , [44] , [48] . Motivated by these ideas , we adopt a resampling strategy in which we calculate pathway , gene and SNP selection frequencies by repeatedly fitting the model over B subsamples of the data , at fixed values for and . Each random subsample of size is drawn without replacement . Our motivation here is to exploit knowledge of finite sample variability obtained by subsampling , to achieve better estimates of a variable's importance . With this approach , which in some respects resembles the ‘pointwise stability selection’ strategy of Meinshasen and Bühlmann [45] , selection frequencies provide a direct measure of confidence in the selected variables in a finite sample . This resampling strategy also allows us to rank pathways , genes and SNPs in order of their strength of association with the phenotype , so that we expect the true set of causal variables to achieve a high ranking , whereas non-causal variables will be ranked low . There have however been suggestions that the use of lasso-type penalties in combination with a subsampling approach can be problematic when applied to GWAS data , where there is widespread correlation between SNPs [49] . This is due to the lasso's tendency to single out different SNPs within an LD block from subsample to subsample , depressing variable selection frequencies for groups of SNPs with high LD . Possible remedies include the use of grouping or sliding-window type strategies , so that neighbouring SNPs in high LD are added to the set of selected SNPs at each subsample . We test the relative performance of these different strategies in a final simulation study described in the next section . For pathway ranking , we denote the set of selected pathways at subsample b bywhere is the estimated SNP coefficient vector for pathway l at subsample b . The selection probability for pathway l measured across all B subsamples is thenwhere the indicator function , if , and 0 otherwise . Pathways are ranked in order of their selection probabilities , . For SNP ranking , we denote the set of SNPs selected at subsample b ( in the unexpanded variable space ) by , and further denote the set of all SNPs within a specified LD threshold , r of SNPs in by ( including SNPs in ) . We use an correlation coefficient for this threshold . Using the same procedure as for pathway ranking , we then obtain two possible expressions for the selection probability of SNP j across B subsamples aswhere the indicator functions , if , and 0 otherwise; and if , and 0 otherwise . Finally , for gene ranking we denote the set of selected genes to which the SNPs in are mapped by , where is the set of gene indices corresponding to all G mapped genes . An expression for the selection probability for gene g is thenwhere the indicator function if , and 0 otherwise . SNPs and genes are ranked in order of their respective selection frequencies . Software implementing the methods described here , together with sample data is available at http://www2 . imperial . ac . uk/~gmontana/psrrr . htm . We evaluate the performance of the above strategies for ranking pathways , SNPs and genes in a final simulation study . For this study we use real genotype and pathways data so that we can gauge variable selection performance in the presence of LD , and variations in the distribution of gene and pathway sizes and of overlaps . For these simulations we use genome-wide SNP data from the ‘SP2’ dataset and map SNPs to pathways from the KEGG pathways database ( see following sections for further details ) . This dataset comprises 1 , 040 individuals , each genotyped at 542 , 297 SNPs , of which 75 , 389 SNPs can be mapped to 4 , 734 genes and 185 pathways with a mean pathway size of 1 , 080 SNPs . We test a number of different scenarios in which we vary the numbers of causal SNPs and SNP effect sizes . For each scenario we perform 400 MC simulations . For each MC simulation we select k causal SNPs at random from a single randomly selected causal pathway . Note however that because pathways can overlap , different numbers of causal SNPs ( up to a maximum number k ) may overlap more than one pathway . We then generate a quantitative phenotype in which we control the per-locus effects size , , where is the proportionate change in phenotype per causal allele , and m is the locus minor allele frequency . GV is then the total proportion of trait variance attributable to each causal locus under an additive model , and under Hardy-Weinberg equilibrium [50] . We also report the total variance , TV , which is the proportion of trait variance attributable to all causal loci . Using contemporaneous GWAS data , Park et al . [50] , report values for GV ranging from 0 . 0004 to 0 . 02 for three complex traits ( height , Crohns disease and breast , prostate and colorectal ( BPC ) cancers ) , although clearly only the largest studies will have sufficient power to identify the smallest genetic effects . They additionally produce estimates ranging from 67 to 201 for the total number of susceptibility loci using these effect sizes , with corresponding values for TV ranging from 0 . 1 to 0 . 36 ( 95% CI ) . It is interesting to note that for certain diseases there is also evidence for polygenic modes of inheritance involving many thousands of SNPs with small effects [51] . While it is currently impossible to translate findings from these and other GWAS into an understanding of how causal SNPs might be distributed within putative causal pathways , we are guided in part by these reported values in constructing our six simulation test scenarios , which are listed in Table 2 . These are designed to cover cases where the number of causal SNPs is relatively small , or large relative to pathway size , and to test cases where the proportion of trait variance explained by causal SNPs spans a realistic range . For simplicity , we set the regularisation parameter to be very close to , to ensure that a single pathway is selected at each of the subsamples generated for each simulation . We set and characterise the resulting SNP sparsity in the final two columns of Table 2 . At each MC simulation , all causal SNPs used to generate the phenotype are removed from the genotype data prior to model fitting . In Figure 7 ( g ) we present the proportion of subsamples ( across all MC simulations ) in which the correct causal pathway is selected , for each of the scenarios described in Table 2 . Since pathways overlap , a causal pathway is here defined as any pathway containing one or more causal SNPs . Since only one pathway is selected at each subsample , true positive rates for each scenario represent the mean number of subsamples in which a causal pathway is selected , across all MC simulations . In Figure 7 ( a ) – ( f ) we present results for SNP and gene ranking performance using SGL-CGD in combination with our resampling-based ranking strategy , using the three different selection frequency measures , and , described in the previous section . For SNP rankings , since actual causal SNPs used to generate phenotypes are removed , true positives are defined as selected SNPs that tag at least one causal SNP with an coefficient . False positives are selected SNPs which do not tag any causal SNP . For gene rankings , causal genes are defined as those that map to a true causal SNP . True positives are then selected causal genes , and false positives are selected non-causal genes . Since the number of ranked variables varies across simulations , mean true positive rates across all simulations are plotted against the number of selected false positives for each scenario . Thus , for a particular simulation , if the highest ranking false positive is at rank z , then the number of true positives is , and the true positive rate for a single false positive is the proportion of true causal variables ( SNPs or genes ) that are tagged by these selected variables . SNP and gene rankings using a univariate , regression-based quantitative trait test ( QTT ) for association are also presented for comparison . For SNP rankings , variables are ranked by their QTT p-value . For gene rankings , SNPs are first mapped to genes , and genes are then ranked by their smallest associated SNP p-value . SNP to gene mappings for all methods are determined in the same way as for mapping SNPs to pathways , that is SNPs are mapped to genes within 10 kbp upstream or downstream of the SNP in question ( see ‘Pathway mapping’ section below ) . It is immediately apparent that the best performance , both in terms of power and control of false positives , is obtained by grouping selected SNPs into genes , that is when ranking by gene selection frequency , . As described elsewhere [49] , simple ranking by SNP selection frequency gives poor results , even if we extend SNP selection to include nearby SNPs in strong LD with selected variants . A notable feature of our method is highlighted by comparing scenarios ( c ) and ( e ) . In scenario ( c ) , the genetic variance explained by each causal locus is relatively high , and gene ranking performance for both QTT and SGL is very good . For scenario ( e ) , the proportion of total phenotypic variance explained by causal loci is the same as that in ( c ) , but in the former relatively small genetic effects are distributed across a larger number of causal loci vs . . Pathway selection power is maintained by SGL for both scenarios , and SGL is also able to maintain superior gene ranking performance with relatively high power and good control of false positives compared to QTT where performance is poor . Also of interest is the fact that SGL gene ranking performance is able to outperform QTT SNP and gene ranking , even at the smallest per-locus effect sizes ( measured by GV - scenarios ( a ) and ( d ) ) , where pathway selection performance is relatively low . Note that in some cases ( most notably in scenario ( a ) ) , SGL SNP and gene ranking power can exceed pathway selection power . This is because true positive SNPs or genes may be ranked higher than false positives , even in the case that a causal pathway is selected in relatively few subsamples . Indeed this ability to distinguish true from false positives in variable rankings at low signal to noise thresholds is one of the attractive features of our subsampling approach . We conclude from this simulation study that SGL in combination with gene ranking using our proposed subsampling approach is able to demonstrate good power and specificity over a range of scenarios using real genotype and pathways data . We next use this approach in an application study which we describe in the remainder of this article . Our application study using pathways-driven SNP selection to search for pathways and genes associated with variation in serum high-density lipoprotein cholesterol levels is carried out using data from two separate cohorts of Asian adults . These datasets have previously been used to search for novel variants associated with type 2 diabetes mellitus ( T2D ) in Asian populations . The first ( discovery ) cohort is from the Singapore Prospective Study Program , hereafter referred to as ‘SP2’ , and the second ( replication ) dataset is from the Singapore Malay Eye Study or ‘SiMES’ . Detailed information on both datasets can be found in [52] , but we briefly outline some salient features here . Both datasets comprise whole genome data for T2D cases and controls , genotyped on the Illumina HumanHap 610 Quad array . For the present study we use controls only , since variation in lipid levels between cases and controls can be greater than the variation within controls alone . The use of both cases and controls in our analysis might then lead to a confounded analysis , where any associations could be linked to T2D status or some other spurious factor . A full investigation of population stratification for the SP2 dataset was carried out for the original GWAS study using PCA with 4 panels from the International Hapmap Project and the Singapore Genome Variation Project , to ensure that this dataset contained only ethnic Chinese [52]–[54] . The SiMES dataset comprises ethnic Malays , and shows some evidence of cryptic relatedness between samples . For this reason , the first two principal components of a PCA for population structure are used as covariates in our analysis of this dataset . Again full details of the stratification analysis can be found in [52] and associated Supplementary Information . A summary of information pertaining to genotypes for each dataset , both before and after imputation and pathway mapping , is given in Table 3 , along with a list of phenotypes and covariates . After the initial round of quality control , genotypes for both datasets have a maximum SNP missingness of 5% . Since our method cannot handle missing values , we perform ‘missing holes’ SNP imputation , so that all missing SNP calls are estimated against a reference panel of known haplotypes . SNP imputation proceeds in two stages . First , imputation requires accurate estimation of haplotypes from diploid genotypes ( phasing ) . This is performed using SHAPEIT v1 ( http://www . shapeit . fr ) . This uses a hidden Markov model to infer haplotypes from sample genotypes using a map of known recombination rates across the genome [55] . The recombination map must correspond to genotype coordinates in the dataset to be imputed , so we use recombination data from HapMap phase II , corresponding to genome build NCBI b36 ( http://hapmap . ncbi . nlm . nih . gov/downloads/recombination/2008-03_rel22_B36/ ) . Following the primary phasing stage , SNP imputation is performed using IMPUTE v2 . 2 . 2 ( http://mathgen . stats . ox . ac . uk/impute/impute_v2 . html ) . IMPUTE uses a reference panel of known haplotypes to infer unobserved genotypes , given a set of observed sample haplotypes [56] . The latest version ( IMPUTE 2 ) uses an updated , efficient algorithm , so that a custom reference panel can be used for each study haplotype , and for each region of the genome , enabling the full range of reference information provided by HapMap3 [57] to be used . Following IMPUTE 2 guidelines , we use HapMap3 reference data corresponding to NCBI b36 ( http://mathgen . stats . ox . ac . uk/impute/data_download_hapmap3_r2 . html ) which includes haplotype data for 1 , 011 individuals from Africa , Asia , Europe and the Americas . SNPs are imputed in 5MB chunks , using an effective population size ( Ne ) of 15 , 000 , and a buffer of 250 kb to avoid edge effects , again as recommended for IMPUTE 2 . Pathways GWAS methods rely on prior information mapping SNPs to functional networks or pathways . Since pathways are typically defined as groups of interacting genes , SNP to pathway mapping is a two-part process , requiring the mapping of genes to pathways , and of SNPs to genes . A consistent strategy for this mapping process has however yet to be established , a situation compounded by a lack of agreement on what constitutes a pathway in the first place [58] . The number and size of databases devoted to classifying genes into pathways is growing rapidly , as is the range and diversity of gene interactions considered ( see for example http://www . pathguide . org/ ) . Databases such as those provided by KEGG ( http://www . genome . jp/kegg/pathway . html ) , Reactome ( http://www . reactome . org/ ) and Biocarta ( http://www . biocarta . com/ ) classify pathways across a number of functional domains , for example apoptosis , cell adhesion or lipid metabolism; or crystallise current knowledge on specific disease-related molecular reaction networks . Strategies for pathways database assembly range from a fully-automated text-mining approach , to that of careful curation by experts . Inevitably therefore , there is considerable variation between databases , in terms of both gene coverage and consistency [59] , so that the choice of database ( s ) will itself influence results in pathways GWAS . The mapping of SNPs to genes adds a further layer of complexity , since although many SNPs may occur within gene boundaries , on a typical GWAS array the vast majority of SNPs will reside in inter-genic regions . In an attempt to include variants potentially residing in functionally significant regions lying outside gene boundaries , SNPs may be mapped to nearby genes using various distance thresholds . Various values for SNP to gene mapping distances , measured in thousands of nucleotide base pairs ( kb ) , have been suggested in the literature , ranging from mapping SNPs to genes only if they fall within a specific gene , to the attempt to encompass upstream promoters and enhancers by extending the range to 10 , 20 or even 500 kb and beyond [18] , [39] , [58] . This process is illustrated schematically in Figure 8 . Notable features of the SNP to pathway mapping process include the fact that genes ( and therefore SNPs ) may map to more than one pathway , and also that many SNPs and genes do not currently map to any known pathway [7] . Following imputation , SNPs for both datasets in the present study are mapped to KEGG canonical pathways from the MSigDB database ( http://www . broadinstitute . org/gsea/msigdb/index . jsp ) . SNPs are mapped to all genes , upstream or downstream of the SNP in question . We exclude the largest KEGG pathway ( by number of mapped SNPs ) , ‘Pathways in Cancer’ , since it is highly redundant in that it contains multiple other pathways as subsets . Details of the pathway mapping process are given in Figures 9 and 10 . Note that there is a difference in the number of SNPs available for the pathway mapping between the two datasets , and this results in a small discrepancy in the total number of mapped genes ( SP2: 4 , 734 mapped genes; SiMES: 4 , 751 ) . However , both datasets map to all 185 KEGG pathways , and a large majority of mapped genes and SNPs overlap both datasets . Detailed information on the pathway mapping process for the two datasets is presented in Table 4 . An ethics statement covering the SP2 and SiMES datasets used in this study can be found in [52] .
For the SP2 dataset we consider two separate scenarios for the regularisation parameters and . For the two scenarios we set the sparsity parameter , , but consider two values for , namely . We test each scenario over 1000 subsamples . We also compare the resulting pathway and SNP selection frequency distributions with null distributions , again over 1000 subsamples , but with phenotype labels permuted , so that no SNPs can influence the phenotype . The parameter controls how the regularisation penalty is distributed between the ( pathway ) and ( SNP ) norms of the coefficient vector . Each scenario therefore entails different numbers of selected pathways and SNPs , and this information is presented in Table 5 . Comparisons of empirical and null pathway selection frequency distributions for each scenario are presented in Figure 11 . The same comparisons for SNP selection frequencies are presented in Figure 12 . In these plots , null distributions ( coloured blue ) are ordered along the x-axis according to their corresponding ranked empirical selection frequencies ( marked in red ) . This is to help visualise any potential biases that may be influencing variable selection . To interpret these results , we begin by noting from Table 5 that many more SNPs are selected with , resulting in higher SNP selection frequencies , compared to those obtained with ( see Figure 12 ) . This is as expected , since a lower value for implies a reduced penalty on the SNP coefficient vector , resulting in more SNPs being selected . Perhaps surprisingly , given that the group penalty is increased , the number of selected pathways is also greater . This must reflect the reduced penalty , which allows a greater number of SNPs to contribute to a putative selected pathway's coefficient vector . This in turn increases the number of pathways that pass the threshold for selection . This raises the question of what might be considered to be an optimal choice for the regularisation-distributional parameter , since different assumptions about the number of SNPs potentially influencing the phenotype may affect the resulting pathway and SNP rankings . To answer this , we turn our attention to the pathway and SNP selection frequency distributions for each value in Figures 11 and 12 . At the lower value of ( top plots in Figures 11 and 12 ) , empirical pathway and SNP selection frequency distributions appear to be biased , in the sense that there is a suggestion that pathways and SNPs with the highest empirical selection frequencies also tend to be selected with a higher frequency under the null , where there is no association between genotype and phenotype . This relationship appears to be diminished with , when fewer SNPs are selected by the model . We investigate this further by plotting empirical vs . null selection frequencies as a sequence of scatter plots in Figure 13 , and we report Pearson correlation coefficients and p-values for these in Table 6 . These provide further evidence of increased correlation between empirical and null selection frequency distributions at the lower value for both pathways and SNPs , again suggesting increased bias in the empirical results , in the sense that certain pathways and SNPs tend to be selected with a higher frequency , irrespective of whether or not a true signal may be present . Further qualitative evidence of reduced bias with is suggested by the clearer separation of empirical and null distributions at the higher value in Figures 11 and 12 . For example , the maximum empirical pathway selection frequency is reduced by a factor of 0 . 29 ( 0 . 35 to 0 . 25 ) as is increased from 0 . 85 to 0 . 95 , whereas the maximum pathway selection frequency under the null is reduced by a factor of 0 . 81 ( 0 . 29 to 0 . 054 ) . Similarly for SNPs , the maximum empirical SNP selection frequency is reduced by a factor of 0 . 37 ( 0 . 52 to 0 . 33 ) , whereas the maximum SNP selection frequency under the null is reduced by a factor of 0 . 9 ( 0 . 11 to 0 . 011 ) . The increased bias with is most likely due to the selection of too many SNPs , in the sense that many selected SNPs do not exhibit real phenotypic effects . These extra SNPs effectively add noise to the model , in the form of multiple weak , spurious signals . This in turn will add bias to the resulting selection frequency distributions , tending to favour , for example , SNPs that overlap multiple pathways , and the pathways that contain them . As is increased , we would expect this biasing effect to be reduced , until a point where too few SNPs are selected , when there is then a risk that some of the true signal may be lost . Note that the reduced but still significant correlations between empirical and null selection frequency distributions at in Table 6 are not unexpected . These may reflect the complex overlap structure between pathways , meaning that pathways ( and associated SNPs ) with a relatively high degree of overlap with other pathways , due for example to the presence of so called ‘hub genes’ , are more likely to harbour true signals , as well as spurious ones [38] , [60] , [61] . Another potential source of correlations between empirical and null distributions is the effect of LD depressing SNP selection frequencies , highlighted earlier . Taking all the above into consideration , we choose to report results with , where there is less evidence of bias due to the selection of too many SNPs . The top 30 pathways , ranked by their selection frequency , are presented in Table 7 , and the top 30 ranked genes , ranked by are presented in the left hand part of Table 8 . Versions of these tables extending to lower ranks are provided in Tables S1 and S2 . For the replication SiMES dataset , we repeat the above analysis design , but consider only the ‘low bias’ scenario where and . Once again we test each scenario over 1000 subsamples , and compare the resulting pathway and SNP selection frequency distributions with null distributions generated over 1000 subsamples with phenotype labels permuted . Pathway and SNP selection frequency distributions are presented in Figure 14 . An investigation of pathway and SNP selection bias is presented in the form of scatter plots illustrating potential correlation between empirical and null selection frequencies in Figure 15 , with corresponding Pearson correlation coefficients and p-values presented in Table 9 . The top 30 ranked pathways and genes are presented in Tables 10 and 8 ( right hand part ) respectively , and extended rankings are provided in Tables S3 and S4 . We now consider the problem of comparing the pathway and gene rankings obtained for each dataset . To do this we require some measure of distance between each pair of ranked lists . Ideally this measure should place more emphasis on differences between highly-ranked variables , since we expect the association signal , and hence agreement between the ranked lists , to be strongest there . By the same reasoning , we expect there to be little or no agreement between variables at lower rankings , where selection frequencies are low . Indeed a consideration of empirical and null selection frequency distributions ( Figures 11 ( bottom ) , 12 ( bottom ) and 14 ) suggests that only the very top ranked variables are likely to reflect any true signal , so that we would additionally like our distance metric to be able to accommodate consideration of the top-k variables only , with , where p is the total number of variables ranked in either dataset . One complication with top-k lists is that they are partial , in the sense that unlike complete lists , a variable may occur in one list , but not the other . In order to consider this problem , we introduce the following notation . We denote the complete set of ranked predictors by , and begin by assuming that all variables are ranked in both datasets . We denote the rank of each variable in list 1 by , so that if variable 5 is ranked first and so on . The corresponding ranks for list 2 are denoted by . A suitable metric describing the distance between two top-k rankings is the Canberra distance [62] , ( 6 ) This has the properties that we require , in that the denominator ensures more emphasis is placed on differences in the ranks of highly ranked variables in either dataset . Furthermore , this distance measure allows comparisons between partial , top-k lists , since a variable occurring in one top-k list but not the other is assigned a ranking of in the list from which it is missing . Note also that a variable i that is not in either of the top-k ranks , that is , makes no contribution to . In order to gauge the extent to which the distance measure ( 6 ) differs from that expected between two random lists , we require a value for the expected Canberra distance between two random lists , which we denote . Jurman et al . [62] derive an expression for this quantity , and we use this to compute the normalised Canberra distance , ( 7 ) Note that this has a lower bound of 0 , corresponding to exact agreement between the lists . For two random lists , the upper bound will generally be close to 1 , although it can exceed 1 , particularly for small k , since the expected value for random lists is not necessarily the highest value . Finally , we compare gene rankings for each cohort obtained using our method with those from a standard GWAS in which SNPs are tested separately for their association with HDLC . Results from the latter study form part of an ongoing multi-cohort study and so are reported in summary form only . Further details are presented in Supplementary Information S1 , Section 6 . By considering only SNPs that map to pathways in each cohort , we find that the top 50 ranked genes using our method are highly enriched amongst genes mapping to highly-ranked SNPs in their respective GWAS ( by permutation ) . Furthermore 4 out of the top 10 ranked genes in the SP2 dataset using our method are also in the top 10 of 4 , 734 genes ranked in the SP2 GWAS . The corresponding figure for the SiMES cohort is 2 out of 10 . As with our gene ranking results ( Table 8 ) , we find little concordance between high ranking genes in both GWAS , with for example no gene occurring amongst the top 10 gene ranks in both cohorts . Note that none of the subset of SNPs in either GWAS that map to pathways in our study achieves genome-wide significance after correcting for multiple testing ( SP2 cohort , 75 , 389 SNPs , minimum SNP p-value = ; SiMES cohort , 78 , 933 SNPs , minimum SNP p-value = ) .
|
Genes do not act in isolation , but interact in complex networks or pathways . By accounting for such interactions , pathways analysis methods hope to identify aspects of a disease or trait's genetic architecture that might be missed using more conventional approaches . Most existing pathways methods take a univariate approach , in which each variant within a pathway is separately tested for association with the phenotype of interest . These statistics are then combined to assess pathway significance . As a second step , further analysis can reveal important genetic variants within significant pathways . We have previously shown that a joint-modelling approach using a sparse regression model can increase the power to detect pathways influencing a quantitative trait . Here we extend this approach , and describe a method that is able to simultaneously identify pathways and genes that may be driving pathway selection . We test our method using simulations , and apply it to a study searching for pathways and genes associated with high-density lipoprotein cholesterol in two separate East Asian cohorts .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results"
] |
[] |
2013
|
Pathways-Driven Sparse Regression Identifies Pathways and Genes Associated with High-Density Lipoprotein Cholesterol in Two Asian Cohorts
|
Viruses have two modes spread in a host body , one is to release infectious particles from infected cells ( global infection ) and the other is to infect directly from an infected cell to an adjacent cell ( local infection ) . Since the mode of spread affects the evolution of life history traits , such as virulence , it is important to reveal what level of global and local infection is selected . Previous studies of the evolution of global and local infection have paid little attention to its dependency on the measures of spatial configuration . Here we show the evolutionarily stable proportion of global and local infection , and how it depends on the distribution of target cells . Using an epidemic model on a regular lattice , we consider the infection dynamics by pair approximation and check the evolutionarily stable strategy . We also conduct the Monte-Carlo simulation to observe evolutionary dynamics . We show that a higher local infection is selected as target cells become clustered . Surprisingly , the selected strategy depends not only on the degree of clustering but also the abundance of target cells per se .
For the first step , we checked whether a strain with a certain G value can be endemic or not by using the stability analysis of the disease free equilibrium . A similar analysis is done by Hiebeler [18] but the endemic condition was calculated for extreme cases ( G = 0 or G = 1 ) in that study . Here we showed that the endemic condition is also obtained for a virus strain with intermediate G value . In addition to the stability analysis , we also obtained the next generation matrix [20] from the linearized dynamics and calculated basic reproductive number ( R0 ) , as , R0=12α ( gxC+lqC/C+ ( gxC+lqC/C ) 2+4gψpCC ) , ( 6 ) where g = βGG , l = βL ( 1−G ) ( 1−θ ) , and ψ = βL ( 1−G ) θ . The derivation is shown in Appendix A ( S1 Text ) . When G = 1 , R0 is βGxC/α ≡ ρ1 , which is consistent with the result from the SIS model without spatial structure . In this case , the infection becomes endemic when ρ1 > 1 that means an infected cell infects more than one susceptible cell . When G = 0 , R0 is βL ( 1−θ ) qC/C/α ≡ ρ0 . ρ0 > 1 is consistent with the "dyad heuristic" of Levin and Durrett [21] , that is , a pair of infected cells will reproduce more than one pair of infected cells . For general values of G , the endemic condition is obtained by stability analysis of the disease free equilibrium , α2−α ( gxC+lqC/C ) −gψpCC<0 . ( 7 ) By solving ( 7 ) with respect to α , the result becomes consistent with R0 > 1 . For the intermediate value of G , the condition ( 7 ) is rewritten by using ρ1 and ρ0 , [1−1ρ1G][1−1ρ0 ( 1−G ) ]<11−θ . Fig 1 shows the region of G that satisfies endemic condition ( 7 ) with changing the recovery rate α . The difference among Fig 1D–1F is the degree of spatial correlation , pCC/xC2 ( but it is not exactly same as the spatial correlation ( pCC−xC2 ) / ( xC−xC2 ) ) . Of course , viruses cannot be endemic when the recovery rate α is too high . With increasing α , highly locally infecting strains drop out first when host cells distribute like CSR or when cells distribute more uniform than CSR . On the other hand , strains with intermediate G value are more resistant to the increase of α than other strains when host cells distribute with a positive correlation . Results of the simulation ( Fig 1G–1I ) indicate that the probability of survival in several trials show similar dependence as predicted by condition ( 7 ) . Using the invasibility analysis , we drew a pairwise invasibility plot ( PIP ) ; Fig 2A is an example of a single parameter set in which intermediate value of G is evolutionarily stable strategy ( ESS ) . In all parameter region examined , there is a unique ESS and ESS strategy is any of completely global ( G = 1 ) , a mixture of local and global infections ( an intermediate G ) , or completely local ( G = 0 ) . In the present model , there are no other patterns like the evolutionary branching , or more than two evolutionary singular points in the present model . The dependence of ESS on the parameters is shown in Fig 2B–2D . When target cells are relatively less clustered like CSR , ESS G is independent of α ( red line in Fig 2B ) . In contrast , when target cells are relatively clustered , increase of α makes the ESS proportion of global infection higher ( green and blue lines in Fig 2B ) . This is because high recovery rate increases the density of disease-free clusters , which makes global infection beneficial in accessing the isolated clusters . This dependence is also observed with βG < βL ( S1 Fig ) . In Fig 2C and 2D , the dependence of ESS on the degree of clustering for target cells , pCC/xC2 , is shown . In general , the higher the degree of target cell clustering becomes , the more local infection is optimal . When the rates of global and local infection are equal ( βG = βL ) , the threshold below which the completely global infection becomes ESS is CSR , pCC/xC2=1 ( Fig 2C ) . In addition to the dependence on pCC/xC2 , the ESS proportion of global infection also depends on the fraction of cells xC alone ( Fig 2C ) . If xC becomes higher with fixing pCC/xC2 , the ESS level of global infection becomes lower . The reason is that in this alteration , the conditional probability that a randomly chosen target cell has a target cell at its nearest neighbor , qC/C = pCC/xC , becomes higher . Thus , local infection becomes more efficient in finding susceptible cells than global infection . The threshold point at which the completely global strain cannot be ESS does not change by altering xC , which is analytically shown in the next section . When the two infection rates differ , the ESS G value tends to prefer the infection mode of better efficiency ( Fig 2D ) . It should be noted that the ESS proportion of global infection is not always a R0 maximizing strategy ( Fig 2E ) . When pCC/xC2 is very high , the ESS proportion is much lower than the G value that maximizes R0 . In most epidemiological or infection dynamics without structure , the ESS trait is to maximize R0 [22 , 23] . When cells are spatially clustered , however , the ESS is not always maximizing R0 , which may be due to a “self-shading” problem as discussed in a previous study [8] . Here we consider the special case in which a mutant strain with a certain G′ ( < 1 ) invades a completely global resident strain ( G = 1 ) . The endemic equilibrium of the completely global strain is obtained from Eq ( 4 ) ( for the calculation , see equations ( A9 ) in S1 Text ) , x^I=xC−αβG , p^SI= ( 1−αβGxC ) αβGxCpCC , p^IO= ( 1−αβGxC ) ( xC−pCC ) , where x^I , p^SI and p^IO are pair densities at the equilibrium . In this case , we can analytically obtain the condition for a mutant strain to increase its density around the resident's endemic equilibrium ( for derivation , see Appendix B in S1 Text ) , βGβL ( 1−θ+G′θ ) <pCCxC2 . ( 8 ) The right-hand side of ( 8 ) denotes the degree of spatial correlation , and this is the reason why we choose pCC/xC2 as the horizontal axis of Fig 2C and 2D . The left-hand side of ( 8 ) represents an increasing function of the mutant’s proportion of global infection G’ . Therefore if βG/βL>pCC/xC2 holds , the completely global strain prevents any kind of mutant from invading and it is an ESS . Especially when βG = βL , the threshold of the spatial configuration at which the completely global strain can be the ESS corresponds to complete spatial randomness ( CSR ) . It means that when cells distribute with a negative correlation , the completely global strain is the ESS , but when cells distribute with a positive correlation , some degree of contact infection can be beneficial . As predicted in this section , the threshold point moves to βG/βL when βG ≠ βL ( Fig 2D ) . The mean G in the population quickly converges to a certain level , and fluctuates around that level ( Fig 3A ) . As long as we use the same parameters , the evolutionary outcomes are similar to each other regardless of the initial condition and the number of trials . We found that evolutionary branching never occurs and the distribution of strains is always unimodal ( see S2 Fig ) . Fig 3B and 3C shows evolutionary outcomes with changing the degree of spatial correlation pCC/xC2 like Fig 2C and 2D , where 20 trials are conducted for each parameter set . In general , the results are similar to the pair approximation , notably , 1 ) high pCC/xC2 prefers local infection , 2 ) evolutionary outcome depends on both pCC/xC2 and xC , and 3 ) increasing xC promotes the local infection . The difference between the results from pair approximation ( Fig 2C and 2D ) and those from simulations ( Fig 3B and 3C ) is that the mean value of G in the simulations does not converge to extreme values ( the completely global or the completely local ) . This may be because the population is always polymorphic as a result of mutations . In addition , there are two other reasons for the region in which the completely local is the ESS according to the pair approximation . The first reason is the effect of finite population size . Especially , since the completely local strain spreads only in a contiguous cluster , the number of available hosts is smaller than for other strains . Therefore , diffusing to other clusters becomes adaptive . The second one is the limitation of pair approximation . In this approximation , we approximate qσ/σ′σ″ the conditional probability that a randomly chosen nearest of a σ'σ'' pair has a σ site by qσ/σ′ the conditional probability that a randomly chosen nearest neighbor of σ' site is a σ site . When viruses are too biased toward local infection , infected cells tend to form a large cluster and the approximation does not work well . For these reasons , there is a discrepancy between pair approximation and simulation . To check whether these results are specific to our method of generating a spatial structure , we also conducted the same evolutionary simulations on the several different deterministically generated structures each having the same global and pair densities ( xC and pCC ) as those in randomly generated structure . Fig 4 shows the comparison of results between randomly and deterministically generated structures . These results suggest the robustness of our results on the evolution of local and global infections based on the randomly generated spatial configurations for given singlet and doublet densities , xC and pCC . In the above sections , we assumed a linear trade-off , that is , the proportion of local infection decreases at the same amount as the proportion of global infection G is increased . However , this should not be always the case in reality . When the local infection decreases in proportion to G0 . 5 , the result is quite different ( Fig 5 ) . In this case , the PIP shows bistability in which the evolutionary outcome depends on an initial state ( Fig 5B ) . This prediction by pair approximation is also confirmed by simulation ( Fig 5C ) .
From an ecological point of view , our model can be applied for considering the evolution of long and short dispersal . Target cells correspond to habitats for animals or plants , with S sites and I sites corresponding to unoccupied and occupied sites , and non-target cells representing unsuitable sites to settle . According to the endemic condition ( 6 ) , the persistence of a species with some local colonization rate depends on the spatial structure . Our results , if applied to a conservation biological setting , suggest that , even if we conserve the abundance of habitats for an endangered species , extinction might occur only due to the change in spatial arrangement of habitats . In terms of the evolution of dispersal distance , previous studies add other settings such as a trade-off between survivability and dispersal range [30] , the population dynamics in local patches [31] , kin selection [32] , the existence of pests [33] , or the disturbance structure [34 , 35] . However , the effect of spatial structure per se has not been clarified yet . We therefore checked how spatial structure affects the evolution of shot and long dispersal . Our results suggest that the ESS proportion of short dispersal depends not only on the degree of clustering but also on the density of habitats per se . Short dispersal is selected when habitats are clustered , and this tendency is strong when the abundance of habitats is high . Our model is also regarded as a model of dispersal rate evolution that considers how much an offspring should disperse beyond its natal patch if we arrange habitats like patches . In this interpretation , local infection denotes staying in a natal patch and global infection represents outgoing from a natal patch . The infection rates βG and βL correspond to survival rate of outgoing and staying individuals , respectively . The conventional result in this situation suggests that even if the survival rate of an outgoing individual ( βG ) is much lower than that of a staying individual ( βL ) , at least half of all offsprings should go out [36] . Pair approximation in the present study does not predict this result but this is clearly due to the limit of pair approximation . Actually , Monte-Carlo simulation shows that the completely local strategy is never selected . In addition to the previous result , we suggest that even if the survival rate of outgoing individual is higher than that of staying one ( βG > βL ) , there is a parameter region in which some offspring should be left due to high pCC/xC2 . Although a similar result has been suggested [37] , the reasoning differs to our study . The previous result is due to the variation of patch quality , which means that an individual born in a good patch should leave its offspring in a natal patch . In our model , high clustering of habitats prefers leaving offsprings in a natal patch without assuming differences in patch quality . When the habitats are highly clustered ( pCC/xC2>1 ) , the probability of finding new habitat is higher for local dispersal ( qC/C = pCC/xC ) than for global dispersal ( xC ) . Therefore , the advantage of finding a new habitat can outweigh the disadvantage of low survival rate . Our model predicts the evolutionarily stable dispersal strategy only but some previous models suggest evolutional branching of dispersal rate [38 , 39] or evolutionary bistability in which the evolutionary outcome differs in initial state [40] . In general , branching or bistability may occur when the fitness of a phenotype depends on the frequencies of other existing phenotypes and possible phenotypes have a proper trade-off [41 , 42] . In our model , there is a trade-off between global and local infection . The reason why we only observe ESS is that the linear trade-off is not suitable for causing evolutionary branching or bistability . In fact , if we assume nonlinear trade-off between global and local infection is assumed , we can observe the evolutionary bistability ( Fig 5 ) . Our model suggests that spatial structure has an important role , while it is not commonly considered in the field of virology . In in vitro experimental cases , two-dimensional cell culture is commonly used and this condition is similar to our model . Hence , there is a possibility that the evolution of global and local infection can occur . In an example of culture of measles viruses ( MVs ) , a higher level of local infection was selected for in continuing passages [43] , indicating the emergence of mutant viruses with a high ability to induce membrane fusion in vitro . When the evolution of global or local infection occur , other viral traits like virulence will evolve as shown in Boots and Sasaki [8] . They analytically showed that a lower virulence is predicted as infection becomes more local . The importance of local infection in the evolution of influenza viruses is shown experimentally [3 , 4]: cell-to-cell transmission promotes a faster expansion of the diversity of virus quasispecies and may facilitate viral evolution and adaptation when influenza viruses’ neuraminidases are inhibited and virus release from infected cells is suppressed [3 , 4] . Therefore , we emphasize the relationship between the spatial distribution of target cells and the evolution of viral infection mode when studying in vitro infectious dynamics . If we manipulate cell density and the efficiency of viral transmission by antibodies , viruses that have favorable level of global and local infection may be obtained . In a host body , it is rare that cells distribute like two-dimensional cell culture systems except for epithelia . Epithelial cells form a continuous sheet and they may be different in susceptibility because of surface molecule expression , response to interferons and other immune cell activities . Therefore , epithelia can be a place where evolution of cell-to-cell infection can occur , and in fact , some viruses are known to have an ability for cell-to-cell infection in epithelia like MVs [44] ( for other examples , see Table 1 in [5] ) . In contrast , influenza viruses can also infect epithelial cells but the evolution of local infection is not known . The reason may be the short length of infection period; influenza infection period can be as short as a week but some viruses like HIV survive in the host body for a long time . Such persistent viruses may have sufficient time for the within-host evolution and the adaptive dynamics framework can be applied . Since MVs also have an ability to establish persistent infection , these viruses may also have a chance to evolve efficient cell-to-cell infection in the host body . It has been shown that cell-to-cell viral transmission through virological synapse occurs in retroviruses such as human T-lymphotropic virus type 1 ( HTLV-1 ) [45] and HIV [46] . This process is thought to have important role because the contribution of cell-to-cell infection on HIV spread is estimated to be equal to or more that of cell-free infection by comparing static and shaking culture conditions [47 , 48] . Since infected cells can move in the lymphoid tissue and find a connection to susceptible cells , the spatial viscosity of the infected target cells in the lymph nodes should be weakened in these viruses . However , there would remain some non-random correlation of uninfected target cells ( pCC/xC2>1 ) because of the locality of T cells in the lymph nodes , and this could favor cell-to-cell transmission over cell-free transmission in retroviruses too . Therefore , we suggest new conditions that evolutionarily promote cell-to-cell infection in those viruses: highly localized distribution of target cells in the lymphoid tissues ( pCC/xC2>1 ) . These points have not been suggested in the previous theoretical study [16] . Our model may explain the emergence of mutant MVs that are isolated from patients of subacute sclerosing panencephalitis ( SSPE ) . These viruses have mutations that provide high ability of cell fusion ( i . e . high level of local infection ) [49] and can infect central nervous system cells , while wild type MVs cannot [43 , 49] . Since MVs can spread in a cell-to-cell manner between epithelial cells , epithelia is a candidate place in which the evolution of local infection occurs . Another possibility is lymph nodes because the main target cell of MVs is SLAM ( signaling lymphocytic activation molecule , also known as CD150 ) expressing immune cells such as T and B cells etc . [50 , 51] . As discussed in the case of HIV , concentrating target cells in the lymph nodes satisfies the condition under which local infection is selected . Consequently , MVs are prone to evolve local infection in a host body and to gain the ability to infect cells of central nervous system but how MVs reach the central nervous system remains unknown . In conclusion , our results suggest that the mode of viral spread , global or local infection , may undergo adaptive evolutionary change in vitro and in vivo . In the future , we can consider more realistic situation such as evolutionary dynamics in three-dimensional space or the repulsion of superinfecting virions that attenuates self-shading effect [29] . In order to examine the emergence of mutant virus or the evolution of virulence , we need to take into account the fact that the mode of infection itself is subject to selection .
|
Viruses such as human immunodeficiency virus and measles virus can spread through physical contact between infected and susceptible cells ( cell-to-cell infection ) , as well as normal cell-free infection through virions . Some experimental evidences support the possibility that high ability of cell-to-cell infection is selected in the host . Since the mode of spread affects the evolution of life history traits , it is important to reveal what condition favors high ability of cell-to-cell infection . Here we address what level of cell-to-cell infection is selected in different target cell distributions . Analysis of ordinary differential equations that keep track of dynamics for spatial configuration of infected cells and the Monte-Carlo simulations show that higher proportion of local infection is selected as target cells become clustered . The selected strategy depends not only on the degree of clustering but also the abundance of target cells per se . Our results suggest viruses have more chances to evolve the ability of local infection in a host body than previously thought . In particular , this may explain the emergence of measles virus strains that gained the ability to infect the central nervous system .
|
[
"Abstract",
"Results",
"Discussion"
] |
[
"organismal",
"evolution",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"applied",
"mathematics",
"microbiology",
"viral",
"structure",
"simulation",
"and",
"modeling",
"viruses",
"algorithms",
"evolutionary",
"computation",
"mathematics",
"rna",
"viruses",
"statistics",
"(mathematics)",
"microbial",
"evolution",
"measles",
"virus",
"research",
"and",
"analysis",
"methods",
"paramyxoviruses",
"medical",
"microbiology",
"mathematical",
"and",
"statistical",
"techniques",
"microbial",
"pathogens",
"monte",
"carlo",
"method",
"evolutionary",
"immunology",
"virions",
"viral",
"evolution",
"evolutionary",
"algorithms",
"viral",
"persistence",
"and",
"latency",
"computational",
"techniques",
"virology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"evolutionary",
"biology",
"statistical",
"methods",
"organisms"
] |
2018
|
The role of spatial heterogeneity in the evolution of local and global infections of viruses
|
Trypanosoma brucei is the causative agent of African sleeping sickness . The polyamine biosynthetic pathway has the distinction of being the target of the only clinically proven anti-trypanosomal drug with a known mechanism of action . Polyamines are essential for cell growth , and their metabolism is extensively regulated . However , trypanosomatids appear to lack the regulatory control mechanisms described in other eukaryotic cells . In T . brucei , S-adenosylmethionine decarboxylase ( AdoMetDC ) and ornithine decarboxylase ( ODC ) are required for the synthesis of polyamines and also for the unique redox-cofactor trypanothione . Further , trypanosomatid AdoMetDC is activated by heterodimer formation with a catalytically dead homolog termed prozyme , found only in these species . To study polyamine regulation in T . brucei , we generated inducible AdoMetDC RNAi and prozyme conditional knockouts in the mammalian blood form stage . Depletion of either protein led to a reduction in spermidine and trypanothione and to parasite death , demonstrating that prozyme activation of AdoMetDC is essential . Under typical growth conditions , prozyme concentration is limiting in comparison to AdoMetDC . However , both prozyme and ODC protein levels were significantly increased relative to stable transcript levels by knockdown of AdoMetDC or its chemical inhibition . Changes in protein stability do not appear to account for the increased steady-state protein levels , as both enzymes are stable in the presence of cycloheximide . These observations suggest that prozyme and ODC are translationally regulated in response to perturbations in the pathway . In conclusion , we describe the first evidence for regulation of polyamine biosynthesis in T . brucei and we demonstrate that the unique regulatory subunit of AdoMetDC is a key component of this regulation . The data support ODC and AdoMetDC as the key control points in the pathway and the likely rate-limiting steps in polyamine biosynthesis .
Human African trypanosomiasis is a neglected disease of sub-Saharan Africa caused by the protozoan parasite Trypanosoma brucei . Current estimates are that more than 50 million people are at risk for infection [1] . Without treatment the disease is always fatal and available drug therapy is limited by toxicity , difficult dosing regimes , and emerging resistance [2] . Eflornithine ( D , L-α-difluoromethylornithine ) , a suicide inhibitor of ODC , is one of only two drugs available for the treatment of the late-stage disease , and its effectiveness has focused attention on the importance of the polyamine biosynthetic pathway for parasite growth . Polyamines are essential organic cations found in all species , and the metabolic pathway has been extensively studied as a potential target for the development of drugs to treat infectious and proliferative diseases [3] , [4] . In most eukaryotes the diamine putrescine is synthesized from L-ornithine by ODC , and it serves as the precursor for the formation of the longer chain amine spermidine ( Figure 1 ) . AdoMetDC catalyzes the formation of decarboxylated S-adenosylmethionine required as the aminopropyl group donor in the formation of the longer chain polyamines . Unique to the trypanosomatids , spermidine is conjugated to glutathione ( GSH ) to produce trypanothione , which is required in cellular redox reactions and necessary for nucleotide synthesis [5] . Gene knockout studies in both T . brucei and Leishmania have been reported for several of the polyamine and trypanothione biosynthetic enzymes demonstrating that they are essential for growth [6]–[13] . Genetic studies have not been reported for T . brucei AdoMetDC , however several promising in vivo trials have shown that AdoMetDC inhibitors cure T . brucei infections in mice , providing chemical evidence that AdoMetDC is an important drug target against this pathogen [14] , [15] . Polyamine homeostasis is essential for normal cellular function . An excess of polyamines leads to hyperproliferation or tumorigenesis , while polyamine deficiency is associated with cell growth arrest leading to death [3] , [4] . As a consequence , polyamine pools in eukaryotic cells are tightly regulated , and cells carefully orchestrate a balance of polyamine biosynthesis , degradation and transport into and out of the cell . In mammalian cells , ODC and AdoMetDC are controlled by transcriptional , translational and post-translational mechanisms [3] , [16] , [17] . Polyamine levels are also regulated by back converting enzymes and polyamine transport . ODC and AdoMetDC both have rapid intracellular turnover rates in mammalian cells . Polyamines accelerate ODC degradation above this basal rate through the action of a protein inhibitor , antizyme [18] . T . brucei does not appear to encode the genes for antizyme nor for the back conversion of polyamines , and it lacks the general transcriptional control mechanisms found in other eukaryotes [19] , leaving open the question of how polyamines are regulated in the parasite . Recently , we discovered that T . brucei AdoMetDC is activated 1 , 200-fold ( on kcat ) by dimerization with a catalytically dead paralog we termed prozyme [20] . This mechanism for controlling AdoMetDC activity is unique to the trypanosomatid parasites , and the finding raised the possibility that regulation of prozyme expression could provide a parasite-specific mechanism to control polyamine homeostasis in trypanosomatids . In order to study the potential for AdoMetDC or prozyme to function as regulators in polyamine biosynthesis we utilized RNA interference ( RNAi ) or regulated knockout approaches in blood form T . brucei parasites to deplete the cells of these proteins . Loss of AdoMetDC or prozyme leads to decreases in spermidine and trypanothione and to cell death . A large compensatory induction in the expression levels of prozyme and ODC was observed after either genetic depletion or chemical inhibition of AdoMetDC . Our data support a translational control mechanism for the regulation of these proteins and they provide the first demonstration that polyamine biosynthesis is regulated in T . brucei . Thus AdoMetDC and prozyme appear to play a central role in controlling polyamine homeostasis in the parasite through a mechanism that is not found in other eukaryotic cells . These data suggest that prozyme arose as a mechanism to regulate the polyamine metabolic flux in trypanosomes and they illustrate the multiplicity of regulatory control mechanisms that have evolved in this essential metabolic pathway .
To test the effects of reduced AdoMetDC expression on blood form T . brucei parasites , we generated a stable cell line with an inducible AdoMetDC targeted RNAi . This line contains a tetracycline ( Tet ) inducible stem-loop vector with 620 bp fragments of AdoMetDC in opposite orientations integrated into the rRNA gene locus ( Figure S1 ) . Addition of Tet leads to production of a double stranded stem-loop RNA targeting AdoMetDC mRNA for degradation . Uninduced AdoMetDC RNAi cells grew at the same rate as the parent 90-13 cells ( data not shown ) . Induction of the AdoMetDC RNAi ( +Tet ) leads to a reduction in AdoMetDC protein that was maintained until the cells die ( Figure 2A and 2B ) . Cell growth arrest was observed within 4 days of induction , followed by cell death ( day 11 ) . Exogenous spermidine ( 0 . 1 mM ) restored normal growth to the induced cells , demonstrating that the AdoMetDC RNAi specifically targeted spermidine biosynthesis ( Figure 2A ) . The polyamine and trypanothione metabolite profile was assessed in the AdoMetDC RNAi cells . Intracellular putrescine levels increased 6–7-fold within 2 days of RNAi induction , while spermidine , glutathionyl-spermidine ( GSH-Spd ) and trypanothione levels gradually declined , with a near complete depletion of GSH-Spd and trypanothione being observed by day 6 after induction ( Figure 2C and 2D , and Table S1 ) . The loss of these metabolites correlated well with the point of cell growth arrest . Spermidine was decreased to just below 40% of uninduced controls . Addition of exogenous spermidine to the medium restored the GSH-Spd and trypanothione pools to 70–80% of wild-type levels , while spermidine levels were returned to 60–70% of controls . Under these conditions the cells grow normally . Cell growth could be rescued by spermidine up to four days post-induction , however by day six it was no longer effective ( Figure S2 ) . This correlates to the time frame when complete depletion of GSH-Spd/trypanothione is observed ( Figure 2D ) suggesting that the cells undergo an irreversible event in the absence of reduced GSH-Spd/trypanothione that commits them to death . To determine if AdoMetDC knockdown is coupled to changes in the expression of other enzymes in the pathway , Western analysis of prozyme , ODC , spermidine synthase ( SpdSyn ) , trypanothione synthase ( TrypSyn ) and trypanothione reductase ( TrypRed ) was undertaken over the time course of the AdoMetDC RNAi induction ( Figure 2B ) . Prozyme protein levels were markedly increased relative to the tubulin control by day 2 after RNAi induction ( Figure 2B ) . ODC was also consistently higher upon depletion of AdoMetDC . This induction of the prozyme and ODC proteins was observed in the presence and absence of exogenous spermidine , demonstrating that the increased expression is not a result of cell growth effects . No significant effects on the expression of the other pathway enzymes were observed . Control experiments on 90-13 cells ( ±Tet ) demonstrated that Tet had no affect on the expression patterns of any of the tested enzymes ( Figure S3 ) . The importance of prozyme to growth of blood form T . brucei was evaluated by the generation of a prozyme conditional knock out ( cKO ) cell line . T . brucei is a diploid organism , thus to generate the KO line the first prozyme allele was replaced with T7 polymerase and a G418 selectable marker , a Tet responsive FLAG-tagged prozyme gene was integrated into the rRNA locus , and finally the second prozyme allele was replaced by the Tet repressor gene and a hygromycin selectable marker ( Figure S1B ) . Southern blotting confirmed the correct integration of the three vectors ( Figure S4 ) . Prozyme cKO cells maintained in the presence of Tet expressed the FLAG-tagged prozyme protein and had similar growth rates to wild-type 427 cells ( data not shown ) . Upon removal of Tet , prozyme expression was reduced to undetectable levels , leading to a rapid arrest of cell growth ( day 2 ) followed by cell death ( day 6 ) ( Figure 3A and 3B ) . Unlike the AdoMetDC RNAi line , the addition of exogenous spermidine did not restore cell growth . These data demonstrate that prozyme is essential for the growth of blood form parasites , and they show that the low activity of homodimeric AdoMetDC ( <0 . 1% of AdoMetDC/prozyme heterodimer activity ) is insufficient to promote cell growth . In addition the data confirm that the AdoMetDC/prozyme heterodimer is the functional configuration of AdoMetDC in the cell . Loss of prozyme in cKO cells resulted in a large increase in putrescine levels ( about 10 fold increase ) , similar to what is seen during AdoMetDC knockdown ( Figure 3C and Table S2 ) . However , in these cells a more substantial reduction in spermidine was observed and it was decreased to 5–7% of the levels detected in the uninduced control cells . Similarly to the AdoMetDC RNAi line , GSH-Spd and trypanothione declined to 5–7% of control levels , while GSH pools were 30–50% of controls ( Figure 3D ) . The decline in the GSH pools suggests that part of this pool has been oxidized as the cell attempts to compensate for the loss of reduced trypanothione . Exogenous spermidine partially restored the intracellular spermidine pools ( up to 26% of control cell levels ) ; however , the GSH-Spd and trypanothione levels remained below 10% , explaining why the cell growth defect was not rescued . To look for potential regulation of the polyamine biosynthetic enzymes , we monitored the protein levels of other pathway enzymes by Western blotting in prozyme cKO cells ( Figure 3B ) . Changes in the protein levels of AdoMetDC , TrypSyn , SpdSyn , or TrypRed were not observed in response to the shut down of prozyme expression . However , ODC protein levels increased similarly to that observed upon knockdown of AdoMetDC expression . In order to determine if AdoMetDC and prozyme are present at similar concentrations , or if one is in excess over the other we measured AdoMetDC activity in cell lysates in the presence and absence of exogenous recombinant prozyme ( 1 µM ) . AdoMetDC activity in 427 parental cells was compared to the AdoMetDC RNAi cells ( ±Tet ) and to the prozyme cKO cells ( ±Tet ) ( Figure 4 ) . In the control cells ( 427 cells or AdoMetDC RNAi ( −Tet ) ) the addition of prozyme stimulates AdoMetDC by 5–8-fold , demonstrating that under normal growth conditions prozyme is present in limiting concentration in comparison to AdoMetDC . Knockdown of AdoMetDC ( RNAi +Tet ) or of prozyme ( cKO–Tet ) , reduced AdoMetDC activity by 60% and 80% , respectively , showing that while neither knockdown approach completely eliminates the protein targets , the prozyme cKO is more efficient at depleting AdoMetDC activity than the induction of AdoMetDC RNAi . The addition of exogenous prozyme has a minimal effect on the activity of the Tet induced AdoMetDC RNAi cells , however activity is increased significantly for the prozyme cKO cells ( −Tet ) , showing that added prozyme can restore activity to the lysate that lacks endogenous prozyme . These data demonstrate that prozyme is present in limiting concentration under typical growth conditions of the wild-type 427 cells , positioning the cell to increase pathway flux by upregulating prozyme levels . We undertook quantitative analysis of the effects on protein and mRNA levels to gain mechanistic insight into the observed induction of prozyme and ODC upon knockdown of AdoMetDC or prozyme . Protein amounts were determined by quantitative Western blotting with fluorescent secondary antibodies , and Northern blots were developed by phosphorimaging; the density of these signals was then quantitated using imaging software and the effects determined relative to tubulin controls ( Figure 5A and 5B and Table S3 show a representative data set ) . AdoMetDC protein and mRNA levels were reduced by 70–80% compared to controls by induction of RNAi , consistent with the activity data . In response , as observed in Figures 2 and 3 , both prozyme and ODC protein levels are induced . The induction of prozyme is consistently more robust and occurs earlier in the time course than for ODC . Prozyme protein increased by an average of 25-fold ( range 12–40-fold , n = 3 ) , while ODC increased 7-fold ( range 5–10-fold , n = 3 ) ( day 4 +Tet; Figure 5C and Table S4 ) . In the presence of spermidine , knockdown of AdoMetDC also led to induction of prozyme and ODC protein , with observed average increases of 10 and 5-fold , respectively ( day 4 +Tet , +Spd ) . Finally , quantitative analysis of the prozyme cKO line showed that the loss of prozyme expression also led to induction of ODC , with ODC protein levels increasing by 4- and 5-fold ( day 2 and 3 without Tet , respectively ) ( Figure S5 and Table S4 ) . In order to confirm that the magnitude of the ODC protein induction was accurately measured by our quantitative analysis we also performed activity assays on the same sample set . ODC activity increased between 8–13-fold ( days 4 and 6 no Spd ) and by 4–5 fold ( days 4 and 6 plus Spd ) after induction of AdoMetDC RNAi , and by 9-fold after the expression of prozyme was turned off in the cKO line ( day 3 −Tet ) . These data show that the observed induction of ODC protein correlates with a similar increase in ODC activity . For both prozyme and ODC the mRNA levels remained relatively constant , and were within 2-fold of the levels detected for control samples ( Figure 5B and Table S3 ) , thus the observed changes in protein expression arise predominately from increased protein levels . These data suggest that the expression of prozyme and ODC is controlled either at the level of translation , or through a post-translational mechanism . In contrast , the expression levels of the remaining pathway enzymes are not regulated under the experimental conditions and the protein and mRNA levels remain within 2-fold of the uninduced controls ( Figure 5 and Table S3 ) . In order to determine if the induction of prozyme and ODC in the AdoMetDC RNAi line was due to the loss of AdoMetDC protein or AdoMetDC activity , we monitored protein levels by Western blotting in the presence of MDL 73811 ( 5′- ( [ ( Z ) -4-amino-2-butenyl]methylamino ) -5′-deoxyadenosine ) , a suicide inhibitor of AdoMetDC [15] . The growth effects of MDL 73811 were first determined for blood form 427 parasites and for the AdoMetDC RNAi line in the absence of Tet ( EC50 = 25–50 nM ) . Cells were then cultured in MDL 73811 concentrations near the EC50 ( 25 and 75 nM MDL 73811 ) and AdoMetDC , prozyme , ODC , TrypSyn and TrypRed were followed by Western over a 24–48 h time course ( Figures 5 , 6A , and 6B , and Table S4 ) . The AdoMetDC RNAi cell line in the absence of MDL 73811 was followed as a control ( ±Tet ) . Knockdown of AdoMetDC by RNAi led to induction of prozyme and ODC protein within 12 h of Tet addition . Prozyme and ODC protein levels were induced by MDL 73811 as early as 3 h after its addition to the culture , with protein levels increasing by 10 and 6-fold , respectively after 24 h ( Figures 5 and 6 ) . AdoMetDC levels declined by 40% in the presence of MDL 73811 , while the levels of TrypSyn , SpdSyn and TrpRed were unchanged . The mRNA levels of the pathway enzymes were within 2-fold of the levels for the control cells after MDL 73811 treatment . These data demonstrate that prozyme and ODC protein levels are rapidly upregulated by either genetic knockdown of AdoMetDC or by its chemical inhibition . The protein synthesis inhibitor cycloheximide was added to the cells to determine if changes in protein turnover rates account for the induction of prozyme and ODC protein upon loss of AdoMetDC activity . Cycloheximide prevented the induction of both prozyme and ODC by MDL 73811 , demonstrating that new protein synthesis is required to observe the upregulation of both proteins . AdoMetDC , prozyme and ODC protein levels in the control cells were stable over the course of the 6 h experiment ( Figure 6C and 6D ) , suggesting that changes in protein turnover rates do not account for the rapid induction of protein levels observed either in the presence of MDL 73811 or during genetic depletion of AdoMetDC . Therefore , changes in translational efficiency are implicated as the mechanism for induction of prozyme and ODC .
Polyamines are essential growth modulators that are highly regulated in eukaryotic cells . The studies herein demonstrate that AdoMetDC and prozyme are both essential for the growth of blood form T . brucei , and they provide the first evidence that the polyamine biosynthetic pathway is also regulated in T . brucei . Our data implicate both prozyme and trypanothione in regulation of polyamine homeostasis in T . brucei , suggesting that the evolution of these trypanosomatid-specific factors may have been linked to the need to acquire regulatory control mechanisms to modulate the growth effects of polyamines on the cell . Few targets in the T . brucei parasite are validated both genetically and chemically , and the data presented here contribute to a compelling case that AdoMetDC is a promising drug target for the development of new treatments for African sleeping sickness . We have shown that prozyme is an essential activator of AdoMetDC activity and that the low activity of the AdoMetDC homodimer on its own is insufficient to maintain cell growth . Thus inhibitors that block heterodimerization , or lock the AdoMetDC structure into an inactive confirmation would provide a parasite-specific mechanism to inhibit this essential enzyme . Successful strategies for these approaches have recently been described , e . g . for Abl kinase [21] and the Bcl-X/BAK interaction [22] . The AdoMetDC knockdown by RNAi led to an 70–80% reduction in AdoMetDC protein that was sufficient to cause cell death , suggesting that a small molecule inhibitor of AdoMetDC would not need to fully inhibit AdoMetDC to be effective . A postulated mechanism for selective toxicity of ODC inhibitors is differential rates of intracellular protein turnover between the rapidly degraded mammalian ODC and the stable trypanosome enzyme [23] . We find that T . brucei AdoMetDC is also a stable protein , while like ODC mammalian AdoMetDC has a short intracellular half-life [24] . Thus differential enzyme turnover may also contribute to selective toxicity of AdoMetDC inhibitors in T . brucei . Finally , the AdoMetDC suicide inhibitor MDL 73811 is a potent anti-trypanosomal agent [14] , [15] . Our finding that prozyme levels are induced by MDL 73811 provides strong evidence supporting AdoMetDC inhibition as the mechanism of action of MDL 73811 in T . brucei , strengthening the chemical validation of the target . It appears that T . brucei regulates the polyamine pathway flux to maintain spermidine homeostasis while depleting the conjugated thiol pools . Indeed spermidine levels do not fall below 40% of the control levels after induction of AdoMetDC RNAi , and they are only fully depleted in the prozyme cKO , which generated a more complete knockdown of AdoMetDC activity . Addition of exogenous spermidine to the prozyme cKO line partially restored the spermidine pools , but not the GSH-Spd and trypanothione pools , suggesting that the cell does not build up the GSH-Spd and trypanothione pools until spermidine levels reach an appropriate steady-state set point . In support , eflornithine treatment has been reported to cause only a partial depletion of spermidine [25] , [26] . These data further provide evidence that the loss of trypanothione pools is a central factor in cell death for both the AdoMetDC knockdown cells and during eflornithine treatment . TrypSyn contains both a synthetic domain catalyzing the formation of GSH-Spd and trypanothione , as well as a catabolic domain that is able to degrade both conjugates back to free spermidine and GSH [27] . The function of the catabolic domain has not been demonstrated in vivo , however our data supports the idea that trypanothione serves as a cellular reservoir for spermidine and that its catabolism contributes to spermidine homeostasis in the cell . This mechanism would provide an analogous situation to that observed in mammalian cells where catabolism of spermine and spermidine play crucial roles in regulating pathway flux [28] . The finding that spermidine was able to rescue the cell growth defect caused by the AdoMetDC RNAi knockdown , but not the prozyme cKO is not entirely understood . Underlying this observation is the finding that exogenous spermidine restored the spermidine and trypanothione pools to the AdoMetDC RNAi cells but not the prozyme cKO cells . We did not observe any difference in spermidine uptake between these lines ( unpublished observation ) but did find as has been previously published that spermidine is poorly taken up by T . brucei blood form parasites [10] . The prozyme cKO led to a 30% greater reduction in AdoMetDC activity than the AdoMetDC RNAi knockdown , and this correlates with a more rapid and complete depletion of spermidine and trypanothione and to more rapid cell growth arrest . This data , in combination with the spermidine pool analysis , suggests that spermidine incorporation is only sufficient to restore GSH-Spd and trypanothione pools after AdoMetDC RNAi , where more residual AdoMetDC remains to help bolster the spermidine pools than is observed in the prozyme cKO ( Figure 4 ) . A further caveat is the observation that spermidine is able to rescue the AdoMetDC RNAi lines when added up to day 4 after Tet induction , while it can not rescue these lines if added as late as day 6 . This corresponds to the point in the growth curve where the GSH-Spd and trypanothione pools have become completely depleted , suggesting that the cells have undergone an irreversible event in the absence of these pools that commits them to death . For the prozyme cKO line , even at the earliest time point measured ( day 2 ) , the GSH-Spd and trypanothione pools were completely depleted , so it may be that the prozyme cKO cells undergo the irreversible transition too quickly for spermidine rescue to be effective . T . brucei responds to loss of spermidine by upregulating two key proteins required for spermidine production , ODC and prozyme . Either the genetic knockdown of AdoMetDC or its chemical inhibition by MDL 73811 led to a rapid and robust increase in prozyme and ODC protein levels . Spermidine rescue of the AdoMetDC RNAi line fully restored normal cell growth , yet the induction of the prozyme and ODC proteins still occurred , providing strong evidence that the induction was not linked to non-specific cell growth effects . The fact that either genetic or chemical loss of AdoMetDC activity leads to increased protein levels of ODC and prozyme provides cross-validation for the observation and demonstrates that the effects do not result from some peculiarity of the genetic system . In wild-type cells prozyme is present in limiting concentration in comparison to AdoMetDC . Thus prozyme is positioned to function not only as an essential activator of AdoMetDC , but by dynamically altering its levels the cell has a mechanism to modulate spermidine production depending on its metabolic state . The pyruvoyl-dependant mechanism of AdoMet decarboxylation is inherently prone to enzyme inactivation through abortive deamination , which occurs at a significant frequency during the catalytic cycle [29] . Mammalian cells may utilize rapid turnover of AdoMetDC as a mechanism to replace the damaged enzyme . Perhaps in the trypanosomatids , where AdoMetDC is a stable protein , the ability to rapidly upregulate the allosteric activator , prozyme , provides an alternative mechanism to evade substrate-mediated inactivation of AdoMetDC by allowing the AdoMetDC pools to remain inactive until needed . The expression of prozyme and ODC both appear to be under translational control . First the protein levels of prozyme and ODC increased substantially upon loss of AdoMetDC activity , while the mRNA levels were not significantly affected . Thus the increased protein accumulation results from either increased translational or from post-translational changes in protein stability . Secondly , prozyme , ODC and AdoMetDC were stable over the 6 h incubation with cycloheximide demonstrating that changes in protein turnover are unlikely to contribute to the increase in steady-state protein levels . The control of gene expression in T . brucei is unusual in comparison to other eukaryotic cells . In T . brucei mRNAs are transcribed as polycistronic units and polymerase II gene transcription is not regulated at the level of transcription initiation [19] . Instead gene expression is regulated post-transcriptionally through changes in message stability , protein stability or through translational regulation . Examples of translational control in T . brucei are limited but include developmental regulation of procyclin [30]–[32] , cytochrome oxidase isoforms [33] and of Nrk protein kinase [34] . Cytochrome c has been shown to be regulated post-translationally by differential protein stability [35] . The finding that polyamine biosynthesis in T . brucei is likely to be translationally regulated shows that translational control in T . brucei is not limited to developmental regulation but also functions as a mechanism to regulate house-keeping enzymes involved in primary metabolism during the cell cycle . An open question is what triggers the de-repression of ODC and prozyme protein expression in response to loss of AdoMetDC . Protein levels are increased by either knockdown or inhibition of AdoMetDC , thus loss of AdoMetDC activity appears to be the common element in triggering the response . The precedent in other eukaryotic cells is that changes in spermidine levels affect the ribosome and thus influence translation . Expression of the ODC inhibitor antizyme is controlled by ribosomal frame-shifting which is stimulated by spermidine , or by putrescine at higher concentrations [36] , [37] . Translation of AdoMetDC in mammalian cells and plants is regulated by a ribosome-stalling peptide that traps the ribosome so that AdoMetDC mRNA is translated efficiently only at low spermidine concentration [38] , [39] . In T . brucei there are no apparent upstream open reading frames in prozyme or ODC , and changes in spermidine concentration appear unlikely to trigger the observed translational changes , since protein levels were induced even in the spermidine rescued AdoMetDC RNAi knockdown cells . The elevated putrescine levels ( 6–10-fold ) observed in the metabolite profiles for cells that express higher levels of prozyme and ODC is a potentially significant observation , but it seems an unlikely trigger for an increase in ODC expression . Human and malaria dihydrofolate reductase have been shown to bind their mRNA to regulate translation of the message [40] , [41] . For the human enzyme binding to mRNA is prevented by the enzyme inhibitor methotrexate . This suggests an intriguing hypothesis that is consistent with our data . AdoMetDC may bind to both the prozyme and ODC mRNA and thereby inhibit their translation . The regulated expression of prozyme is a unique mechanism for controlling polyamine pathway flux in T . brucei not found in other eukaryotic cells . However , despite differences in the protein components the metabolic control points of the pathway in T . brucei are similar to those found in mammalian cells , with regulation being focused on control of ODC and AdoMetDC . Thus the data suggests that similarly to mammalian cells , ODC and AdoMetDC catalyze the rate-limiting steps in polyamine biosynthesis in T . brucei . While the mechanisms differ , both utilize translational regulation of regulatory proteins to control the pathway flux . Mammalian cells translationally regulate an ODC inhibitor protein , antizyme , while T . brucei regulates an AdoMetDC activator protein , prozyme . Thus T . brucei has evolved a unique but parallel regulatory mechanism to control polyamine metabolism to that found in mammalian cells . These data provide evidence for convergent evolution of translational control mechanisms centered on regulatory binding proteins in the polyamine metabolic pathway .
Bloodstream form trypanosomes ( 90-13 or 427 ) were cultured in HMI-9 media supplemented with 10% serum at 37°C , 5% CO2 as described [42] . Chicken serum was used in place of fetal bovine serum , allowing for the addition of spermidine ( 0 . 1 mM ) without encountering polyamine oxidase-driven toxicity [7] . Cells were grown with the appropriate antibiotics ( G418 , 2 . 5 µg/ml; hygromycin 5 µg/ml; phleomycin 2 . 5 µg/ml , blasticidin 2 . 5 µg/ml ) and were split every 24–48 hours to maintain cultures in log phase ( 105 to 106 cells/ml ) . Cell densities were determined by counting on a hemocytometer ( Brightline , Fisher ) . Growth curves are represented as total cell number ( product of cell density and total dilution ) and all data were collected in biological triplicate . To determine the effects of MDL 73811 , AdoMetDC RNAi or 427 cells were cultured in the in the presence of a range of MDL 73811 concentrations ( 10–150 nM MDL 73811 ) and cells were counted after 3 d to determine the EC50 . For Western analysis cultures of blood form T . brucei wild-type cells were split into fresh media 3 h before dosing with inhibitors . Drug ( 25 or 75 nM ) , cycloheximide ( 50 µg/ml ) or both were then added and samples were collected at the indicated time points . The ability of this concentration of cycloheximide to inhibit protein synthesis in T . brucei has previously been established [23] , [33] , [43] . The techniques for manipulating gene expression in T . brucei by RNAi have been well established [44] . The pLEW100 and pJM326 vectors [45] were used to generate the AdoMetDC RNAi plasmid . A 620 base pair portion of the T . brucei AdoMetDC gene ( starting at the 78th coding nucleotide ) was amplified by PCR ( primers in Table S5 ) from genomic DNA isolated from T . brucei 427 cells and cloned into the pJM326 vector in the forward direction and into the pLew100 vector in the reverse direction . The AdoMetDC fragment fused to the stuffer region was excised from pJM326 with HindIII and XbaI and inserted into the modified pLew100 vector . The resulting plasmid contains two copies of the AdoMetDC gene fragment in opposite orientation separated by the stuffer DNA sequence ( Figure S1A ) . The production of stemloop RNA ( with double stranded RNA targeting the AdoMetDC message ) is driven from the Tet inducible procyclin promoter . Log phase T . brucei 90-13 bloodstream form cells were transfected with the linearized AdoMetDC stemloop RNAi vector ( 80 µg ) , and phleomycin resistant cells containing the construct integrated into the rRNA locus were selected as previously described [11] . Clonal lines were generated by limiting dilution . Synthesis of double stranded RNA ( RNAi ) targeting the AdoMetDC message in a stemloop structure was induced by Tet ( 1 µg/ml ) , which was added fresh every 24 h . The prozyme conditional knockout line was generated using previously described methods [46] , [47] . Identical 300 bp segments of the prozyme 5′ and 3′ UTRs ( corresponding to nucleotide −321 to −1 and 1023 to 1348 , respectively where base 1 represents the ATG start and base 978 the TGA stop codons in the protein coding region ) were cloned into the pLEW13 and pLEW90 vectors . The pLEW100 vector was altered by liberating the phleomycin resistance cassette with DraIII and BssHII and replacing it with the blasticidin resistance gene from pcDNA6V5His ( Invitrogen ) . The resulting small open reading frame prior to the blasticidin coding region was removed by site directed mutagenesis . The resulting vector ( pLEW300 ) was used to generate the Tet inducible expression plasmid containing Flag-tagged prozyme , using the tagged gene isolated from the previously described pT7-FLAG1-prozyme vector [20] . The primers used to generate the constructs are provided in Table S5 , and the construct diagrams are in Figure S1B . Log phase T . brucei 427 bloodsteam form cells were transfected with the Amaxa nucleofector as described [48] to introduce in successive steps: 1 ) the pLew13-prozyme SKO-N vector ( which contains coding sequence for T7 polymerase and the G418 resistance cassette inside the prozyme 5′ and 3′ UTRs ) integrated into the prozyme locus; 2 ) the pLew300-Flag prozyme construct ( which contains coding sequence for a FLAG tagged prozyme and the blasticidin resistance cassette ) targeted to the rDNA spacer region; and , 3 ) pLew90-prozyme SKO-H vector ( which contains coding sequence for the tet repressor and the hygromycin resistance cassette inside the prozyme 5′ and 3′ UTRs ) integrated into the second prozyme locus . Clonal lines were generated by limiting dilution . Prozyme expression was maintained in these lines by the addition of Tet ( 1 µg/ml ) . Southern blot analysis was used to confirm the genotypes of cKO cell lines . Genomic DNA ( 10 µg ) was isolated from cells using standard methods , digested with BanI , separated on a 1% agarose gel and transferred to a positively charged membrane ( Ambion Bright Star ) by vacuum in 10× SSC . The membrane was crosslinked by UV , prehybridized in Ambion UltraHyb solution and then hybridized with probe overnight . Blots were washed ( 0 . 1× SSC and 0 . 1% SDS ) and visualized with a FLA 5000 phosphoimager . Radiolabeled [32P]dATP probes were prepared using the Strip-EZ PCR kit ( Ambion ) from genomic DNA . The primers used for generation of the probe by PCR are provided in Table S5 . The probe for the endogenous locus was a 600 bp region starting at 891 bp upstream of the prozyme ATG start site . The probe for the coding region included nucleotides 78 to 678 of the prozyme gene . Cells ( ∼108 ) were harvested by centrifugation ( 3 , 000 rpm ) , washed twice in cold phosphate-buffered saline ( PBS , pH = 7 . 4 ) , resuspended in lysis buffer ( 50 mM HEPES pH 8 , 100 mM NaCl , 5 mM 2-mercaptoethanol , 2 mM phenylmethylsulfonyl fluoride , 1 µg/ml leupeptin , 2 µg/ml antipain , 10 µg/ml benzamidine , 1 µg/ml pepstatin , 1 µg/ml chymostatin ) , and lysed by three successive freeze/thaw cycles . The lysate was clarified by centrifugation ( 13 , 200 rpm for 5 minutes ) and protein concentration was determined colormetrically [49] . Protein ( 20 µg/lane ) was separated by 12% SDS/PAGE and transferred to a polyvinylidene difluoride ( PVDF ) membrane ( Hybond-P , Amersham ) . Membranes were blocked in 5% non-fat milk in Tris-buffered saline ( TBS: 20 mM Tris·HCl , 137 mM NaCl , pH 7 . 6 ) . Primary and secondary antibody incubations were carried out in 5% milk TBS-T ( TBS plus 0 . 1% ( v/v ) Tween-20 ) . Rabbit polyclonal antibody against T . brucei prozyme was generated ( by Proteintech Group , Inc , Chicago , IL ) to recombinant His6-prozyme purified as described [20] and was used at 4 , 000-fold dilution . All other antibodies have been previously described [8] , [11] , [12] , [20] and were used at the following dilutions: T . brucei AdoMetDC ( 2 , 500-fold dilution ) , ODC ( 10 , 000-fold dilution ) , L . donovani SpdSyn ( 1000-fold dilution ) , T . brucei TrypSyn ( 1 , 000-fold dilution ) and T . cruzi TrypRed ( 1000-fold dilution ) . Mouse anti-FLAG M2 antibody ( Sigma ) ( 1 , 000-fold dilution ) , horseradish peroxidase- ( HRP ) linked donkey anti-rabbit or anti-mouse IgG secondary antibodies ( Amersham Biosciences ) were used at 10 , 000 fold dilution . Antigen recognition was visualized using the ECL chemiluminescent HRP substrate reagents ( Amersham ) , followed by detection on film . For the AdoMetDC RNAi cells ( Figure 2 ) , separate gels were prepared for each antibody and the full molecular weight range was transferred to membrane and probed . This strategy allowed any potential modifications of the proteins to be detected , and none were observed . Thus only the appropriate molecular weight range for the probed protein is displayed in Figure 1 . For the prozyme cKO cells ( Figure 3 ) parallel gels were loaded and transferred , but each membrane was sectioned into three strips by molecular weight to allow for probing of three antibodies from one gel . Western analysis was repeated in biological triplicates and representative experimental sets are displayed ( Figures 2B and 3B ) . Samples were prepared as described above , parallel gels were loaded , and each gel was cut into three sections to allow for probing of multiple antibodies . Each gel was probed for tubulin as a control . mRNA from these samples was also collected and used for quantitative northern blot analysis ( see below ) . Primary antibody raised against AdoMetDC was used at a 500-fold dilution , tubulin was used at 50 , 000-fold dilution , and all other antibodies were used at a 1000-fold dilution in 5% milk-TBS-T . Alexa-Fluor 647 goat anti-rabbit IgG , Alexa-Fluor 647 goat anti-mouse IgG or Alexa-Fluor 546 goat anti-mouse IgG ( Invitrogen ) secondary antibodies were used at a 500-fold ( for AdoMetDC blot ) or 1 , 000-fold ( all other antibodies ) in 5% BSA ( bovine serum albumin ) TBS-T . Fluorescent signals were measured on a Typhoon 8610 ( Molecular Dynamics , with λex = 633 and λem = 670 or λex = 532 and λem = 580 ) . The bands were quantified by ImageQuant 5 . 2 software ( Molecular Dynamics ) . Density from each antibody was normalized by tubulin density , and each experimental condition was normalized to control samples . Intracellular polyamine content was analyzed by conjugation to the fluorescent AccQ-tag reagent ( 6-aminoquinolyl-n-hydroxysuccinimidyl in acetonitrile , Waters ) followed by separation on a Waters AccQtag ( 3 . 9×150 mm ) column using a Beckman System Gold HPLC with a Ranin Dynamax Fluorescence detector . HPLC buffers and gradients have been previously described [50] . The reduced intracellular thiols were quantitated as described [11] , [26] by labeling with monobromobimane and separation on a Phenomonex Nucleosil C18 column ( 30×4 . 6 mm ) . AdoMetDC and ODC activity in cell lysates were determined as previously described [20] , [50] . Lysates ( 40 µg ) were incubated with 14C-AdoMet ( Amersham; 40 µM ) ± recombinant prozyme ( 1 µM ) for assay of AdoMetDC or with 1-14C-ornithine ( Amersham; 80 µM ) for assay of ODC . Reactions were allowed to proceed for 30 minutes at 37°C before quenching with HCl . Liberated 14CO2 was trapped by barium hydroxide soaked filter paper and measured by scintillation counting . Northern blot analysis was facilitated by the NorthernMax kit ( Ambion ) . Briefly , mRNA was isolated from trypanosome cells ( at least 1×108 ) using the micro polyA purist kit ( Ambion ) and separated by denaturing 1% agarose gel electrophoresis ( 1 µg/lane ) . The mRNA was transferred to a positively charged nylon membrane ( BrightStar-Plus , Ambion ) and cross-linked . Radiolabeled [32P]dATP probes were prepared using the Strip-EZ PCR kit ( Ambion ) with genomic or plasmid DNA serving as the template . Data displayed represents a single membrane that was stripped and reprobed for each of the displayed genes . Signals were either developed on film , or with a phosphorimager screen . For quantitation , the signal from the phosphorimager screen ( Molecular Dynamics ) was measured on the Typhoon 8610 scanner ( Molecular Dynamics , using the 633 nm laser ) , and quantified my ImageQuant 5 . 2 software ( Molecular Dynamics ) . Gene accession numbers have been taken from http://www . genedb . org/ and are as follows: AdoMetDC , Tb927 . 6 . 4460/Tb927 . 6 . 4410; prozyme , Tb927 . 6 . 4470; ODC , Tb11 . 01 . 5300; Spd Syn , Tb09 . v1 . 0380; Tryp Syn , Tb927 . 2 . 4370 and Tryp Red , Tb10 . 406 . 0520 .
|
Human African trypanosomiasis ( HAT ) is an important vector-borne pathogen . The World Health Organization estimates that more than 50 million people are at risk for the disease , which occurs focally , in remote regions , and periodically reaches epidemic levels . Untreated HAT is always fatal , and the available drugs compromise toxicity and emerging resistance . The only safe treatment for late-stage disease is an inhibitor of an essential metabolic pathway that is involved in the synthesis of small organic cations termed polyamines . In this paper , we use genetic approaches to demonstrate how the parasite regulates this essential metabolic pathway . By modulating the protein levels of a trypanosome-specific activator of polyamine biosynthesis , the parasite has developed a mechanism to regulate pathway output . We also demonstrate that this pathway activator is essential to parasite growth . Our data strengthen the genetic and chemical validation of a key enzyme in this pathway as a drug target in the parasite , and they provide new insight into parasite-specific approaches that could be used to design novel drugs against this deadly disease .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry/chemical",
"biology",
"of",
"the",
"cell",
"biochemistry/transcription",
"and",
"translation",
"biochemistry/biocatalysis",
"infectious",
"diseases/protozoal",
"infections"
] |
2008
|
Regulated Expression of an Essential Allosteric Activator of Polyamine Biosynthesis in African Trypanosomes
|
Bacteria sense and respond to many environmental cues , rewiring their regulatory network to facilitate adaptation to new conditions/niches . Global transcription factors that co-regulate multiple pathways simultaneously are essential to this regulatory rewiring . CodY is one such global regulator , controlling expression of both metabolic and virulence genes in Gram-positive bacteria . Branch chained amino acids ( BCAAs ) serve as a ligand for CodY and modulate its activity . Classically , CodY was considered to function primarily as a repressor under rich growth conditions . However , our previous studies of the bacterial pathogen Listeria monocytogenes revealed that CodY is active also when the bacteria are starved for BCAAs . Under these conditions , CodY loses the ability to repress genes ( e . g . , metabolic genes ) and functions as a direct activator of the master virulence regulator gene , prfA . This observation raised the possibility that CodY possesses multiple functions that allow it to coordinate gene expression across a wide spectrum of metabolic growth conditions , and thus better adapt bacteria to the mammalian niche . To gain a deeper understanding of CodY’s regulatory repertoire and identify direct target genes , we performed a genome wide analysis of the CodY regulon and DNA binding under both rich and minimal growth conditions , using RNA-Seq and ChIP-Seq techniques . We demonstrate here that CodY is indeed active ( i . e . , binds DNA ) under both conditions , serving as a repressor and activator of different genes . Further , we identified new genes and pathways that are directly regulated by CodY ( e . g . , sigB , arg , his , actA , glpF , gadG , gdhA , poxB , glnR and fla genes ) , integrating metabolism , stress responses , motility and virulence in L . monocytogenes . This study establishes CodY as a multifaceted factor regulating L . monocytogenes physiology in a highly versatile manner .
Listeria monocytogenes is a Gram-positive facultative intracellular pathogen transmitted by ingesting contaminated foods . L . monocytogenes causes a disease termed listeriosis associated with a mortality rate of up to 30% . Listeriosis typically manifests as a mild gastroenteritis in healthy people , however it can lead to meningitis in elderly and immunocompromised people and cause stillbirth in pregnant women [1] . The replicative niche of L . monocytogenes inside the host is within the cell cytosol [2] . The bacteria invade non-phagocytic host cells by expressing specialized proteins termed internalins that induce active internalization; in the case of phagocytic cells the bacteria are simply phagocytosed [3 , 4] . Subsequently , L . monocytogenes rapidly escapes from the endosome/phagosome vacuole using primarily the listeriolysin O toxin ( LLO ) , two phospholipases ( PlcA and PlcB ) , and components of the competence system [5–8] . Having gained entry to the host cell cytosol L . monocytogenes replicates rapidly ( at a growth rate similar to that exhibited in rich laboratory medium ) , and spreads from cell to cell using actin based motility , which is mediated by the virulence factor , ActA [9 , 10] . Remarkably , all the above-mentioned virulence factors ( and other factors ) are positively regulated by PrfA , a Crp/Fnr like transcription regulator that is considered the master virulence activator of L . monocytogenes [11 , 12] . The transition from saprophyte to deadly pathogen relies largely on L . monocytogenes sensing multiple host-specific signals that are transduced to trigger PrfA expression and activity [13] . For example , sensing of temperature , certain carbon sources ( e . g . , glucose-1-phosphate ) , the availability of amino acids ( e . g . , isoleucine ) , iron and glutathione , were all shown to trigger PrfA via different mechanisms , and thus activate downstream virulence genes [14–21] . These examples highlight that multiple mechanisms have evolved to sense various host-specific cues and metabolic signals that conjointly inform L . monocytogenes of its intracellular location and the need to switch to the virulent state . To better understand the metabolic environment within the host cell and the signals that activate L . monocytogenes virulence genes during intracellular growth , we previously performed a genome scale integrative study that combined transcriptome analysis and metabolic modeling in silico [18] . Various bacterial metabolic pathways were identified to be highly active during L . monocytogenes infection and contribute to bacterial intracellular growth in macrophage cells . Notably , the biosynthesis of branch-chained amino acids ( BCAAs ) ( i . e . , isoleucine , leucine and valine ( ILV ) ) was highly induced in intracellularly grown bacteria , a pathway encoded by the ilv operon , suggesting that BCAAs may be limiting in macrophage cells . In light of this finding , we reasoned that sensing of metabolite availability within the host cell might alert the bacteria of their intracellular location and the need to activate the virulence state . Accordingly , we found that growing L . monocytogenes in minimal defined medium with limiting amounts of BCAAs indeed leads to robust activation of the virulence genes [18] . Under low concentrations of BCAAs , in particular of isoleucine , prfA and some of its downstream-regulated genes were highly expressed concomitantly with the BCAA biosynthesis pathway . These observations identified BCAAs as an important metabolic signal for L . monocytogenes within the mammalian niche . Subsequently , we demonstrated that CodY , a global regulator and sensor of BCAAs , is responsible for the upregulation of prfA and the virulence genes under low BCAAs concentrations [17 , 18] . CodY was found to bind directly within the coding sequence of the prfA gene ( 15 nucleotides down-stream the ATG start codon ) and activate transcription , in turn leading to upregulation of virulence genes . These findings were surprising since CodY was thought to function primarily as a repressor and to bind DNA primarily under rich media conditions ( i . e . under high BCAA concentrations ) [17] . CodY is a Gram-positive specific global regulator that was discovered two decades ago in Bacillus subtilis as a general repressor of stationary phase genes , though now it is known to regulate many cellular processes [22–24] . CodY responds to cellular levels of BCAAs by directly binding these amino acids , an interaction that influences its structural conformation and activity [25–27] . In some Gram-positive bacteria CodY also binds GTP , however the effect of this interaction is not completely understood [28] . Initial studies indicated that CodY acts as a general repressor ( when bound to GTP or BCAAs ) of many metabolic genes including the BCAAs biosynthesis pathway . However , later studies in B . subtilis demonstrated that CodY also functions as an activator , in the presence of its ligand ( i . e . , under high concentrations of BCAAs ) [29 , 30] . In L . monocytogenes , CodY was shown to repress genes involved in amino acid metabolism , nitrogen assimilation and sugar uptake under conditions of high BCAA concentrations [31] . A role for CodY in regulation of virulence was demonstrated in Gram-positive pathogens other than L . monocytogenes , including Clostridium perfringens , Bacillus anthracis and Streptococcus pyogenes where CodY was shown to activate indirectly the expression of certain virulence genes [32–35] . As mentioned above , before the present study , CodY activity was primarily documented under rich growth conditions , with the ability of CodY to bind DNA demonstrated in the presence of its regulatory ligand [25 , 36] . However , our previous observation that CodY possesses a regulatory activity and DNA binding capacity also under minimal growth conditions , exhibiting limited amounts of BCAAs ( as shown for the prfA gene ) , raised the possibility that CodY may bind DNA and function also when BCAAs are at low concentrations , maybe even in its unliganded form [17] . Support for this premise came from the observation that a CodY protein mutated within its BCAA-binding site ( i . e . , harboring a R61A substitution within its GAF domain ) loses the ability to repress metabolic genes under high BCAAs concentrations but retains the ability to activate prfA under low BCAAs concentrations [17] . To further delineate if indeed CodY possesses diverse activities under rich and minimal growth conditions and thus regulates different target genes , we took a system-level approach employing L . monocytogenes bacteria . The CodY regulon , in particular its direct target genes , were analyzed in bacteria grown in rich and minimal media ( the latter containing low concentrations of BCAAs ) using RNA-Seq and ChIP-Seq techniques . Notably , the data reveal that CodY retains multiple regulatory activities under both conditions , orchestrating the expression of metabolic , stress and virulence genes in a highly versatile manner .
Before the present study , CodY’s regulon and direct target genes were assessed under rich growth conditions ( or that are rich in BCAAs ) , and thus a role for CodY was documented in the presence of its ligand isoleucine [33 , 37–42] . To better decipher CodY’s activity in relation to the availability of its regulatory ligand/s , we analyzed the CodY regulon under both rich and minimal growth conditions ( the latter containing low levels of BCAAs ) using the RNA-Seq technique . Wild-type ( WT ) and ΔcodY L . monocytogenes bacteria were grown in rich brain heart infusion ( BHI , containing excess amounts of BCAAs > 800 μM ) and low BCAAs minimal defined media ( LBMM; containing ~80 μM of BCAAs ) . Total bacterial RNA was extracted at mid-logarithmic growth and subjected to deep sequence analysis using Illumina HiSeq 2500 . Of note , these experimental conditions were chosen in concordance with our previous findings demonstrating that during growth in BHI , CodY effectively represses metabolic genes ( e . g . , the ilv operon , the hallmark of CodY regulation ) , whereas during growth in LBMM this repression is relieved concomitantly with CodY activation of the prfA gene [17 , 18] . As reported previously , WT and ΔcodY bacteria exhibit similar growth in LBMM medium , whereas ΔcodY bacteria grown in BHI displayed a slightly reduced growth as compared to WT bacteria ( S1 Fig ) [17] . RNA-Seq analysis was performed in triplicate and the reproducibility of the biological repeats was high ( a mean R2–0 . 956 ) ( S2 Fig ) . A total of 368 genes ( ~14% of L . monocytogenes genome ) were found to be affected by CodY ( directly and indirectly ) under both conditions . Among these , 237 genes were upregulated and 131 genes were down regulated in the ΔcodY mutant in comparison to WT bacteria . In BHI medium , 334 genes were differentially regulated by CodY ( directly and indirectly ) , in agreement with previous reports [37–43] . Under this condition , 111 genes were down regulated and 223 genes were upregulated in the ΔcodY mutant in comparison to WT bacteria ( Fig 1A and 1C ) . Notably in LBMM , 181 genes were differentially regulated by CodY , among them 55 genes were down regulated and 126 were upregulated in the ΔcodY mutant , demonstrating for the first time a global regulatory role for CodY under low BCAAs conditions ( Fig 1B and 1C ) . Notably , among all of the upregulated genes ( 237 genes ) , 112 ( 45% ) were upregulated under both rich and minimal conditions , whereas among the down regulated genes ( 131 genes ) , 36 ( ~27% ) were down regulated under both conditions ( Fig 1B ) . Overall these findings indicate that CodY may serve as both a repressor and as an activator of genes under high and low BCAAs levels . Moreover , they demonstrate that another mode of CodY regulation might exist that is independent of BCAAs . The latter may be mediated by additional regulatory factors such as GTP . Applying manually curated criteria clustering and hierarchical clustering on all of the differentially regulated genes yielded 6 distinct gene clusters representing all modes of CodY regulation ( Fig 1B and 1C and S1 Table ) . Specifically , 111 genes were identified to be repressed ( cluster I ) and 76 genes to be activated ( cluster II ) by CodY ( directly and indirectly ) exclusively under rich growth conditions , 14 genes were identified to be repressed ( cluster III ) and 19 genes to be activated ( cluster IV ) by CodY exclusively under minimal growth conditions , while 112 genes were identified to be repressed ( cluster V ) and 36 genes to be activated ( cluster VI ) by CodY under both rich and minimal growth conditions ( clusters are listed in S1 Table ) . Taken together , the data highlight CodY’s plasticity and ability to regulate genes in diverse spectrum of growth conditions . Pathways and responses were identified manually and by functional enrichment analysis using the MIPS server ( Fig 2 and S2 Table ) [44] . Genes repressed by CodY under rich growth conditions are involved in amino acid metabolism and transport ( e . g . , BCAAs and histidine biosynthesis pathways ) , peptide and sugar transport systems ( e . g . , the OppC and OppF permeases and the glycerol uptake protein ) and stress responses ( e . g . , bile salt hydrolase ( bsh ) , clpC , and heat shock proteins ) ( cluster I ) ( Fig 2 and S1 Table ) . Notably , this cluster includes several virulence-associated genes such as internalin A and B and the glycerol transporter and kinase , which are expressed during L . monocytogenes infection of mammalian cells [15 , 18] . Several metabolic genes/pathways were activated by CodY under this condition , such as arginine biosynthesis , assimilation of ammonia , certain PTS systems and enzymes of the tricarboxylic acid ( TCA ) pathway ( cluster II ) ( Fig 2 and S2 Table ) . Under minimal growth conditions , we identified prfA and its downstream virulence genes ( e . g . , hly , actA , plcA , plcB and mpl ) to be activated by CodY ( cluster IV ) , which is in accordance with our previous results [17 , 18] ( S1 Table ) . Notably , all the other genes within this cluster ( except for one; LMRG_01981 ) were previously found to be induced during L . monocytogenes infection of mammalian cells [18] , implicating a potential role in L . monocytogenes virulence . Some of these genes mediate ammonium transport , nitrogen regulation , cell wall synthesis and certain PTS systems , while others represent conserved hypothetical genes . Of note , it is not known yet whether these genes are under the direct regulation of PrfA . Overall , these findings strengthen the premise that CodY plays an important role in the activation of L . monocytogenes virulence under low BCAAs conditions , which resemble the mammalian cytosolic niche . Under these conditions , we found CodY to repress genes involved in purine metabolism , and iron transport , as well as some genes that encode hypothetical proteins ( Cluster III ) ( S1 and S2 Tables ) . Genes repressed by CodY under both rich and minimal conditions are involved in nitrogen metabolism , arginine deiminase , osmotic and salt stress responses , distinct PTS systems and encode amino acid transport proteins , which most likely reflects nutrient availability within the media and the growth conditions tested ( cluster V ) . Genes/pathways activated by CodY under both conditions include various metabolic genes and transport systems , D-alanine dipeptide synthesis , as well flagella biosynthesis and chemotaxis ( cluster VI ) ( S1 and S2 Tables ) . Notably , several transcription regulators and regulatory proteins were identified within the CodY regulon ( distributed among the different clusters ) , such as sigma-B , GlnR , GntR and certain CRP/FNR transcription regulators , as well sigma-54 regulatory proteins , indicating a hierarchy in CodY gene regulation . Overall , the RNA–Seq analysis highlights the breadth of CodY regulation in L . monocytogenes and its potential regulatory functions under varying metabolic environments . To delineate which genes in the CodY regulon are directly regulated by CodY , a genome wide chromatin immunoprecipitation was performed in combination with DNA sequence analysis ( ChIP-Seq ) using Illumina HiSeq 2500 . An L . monocytogenes codY-6his strain , in which the codY gene was replaced with a 6-histidine tagged codY , was grown in both BHI and LBMM media to mid-exponential phase . Bacteria were then cross-linked using formaldehyde and subjected to ChIP as described in the Materials and Methods . Of note , we have established previously that the CodY-6His protein functions similarly to the native CodY [17] . The ChIP-Seq analysis revealed 302 DNA regions bound by CodY under both conditions ( 270 in LBMM and 131 in BHI ) , with ~30% overlap ( S3 Table ) . Among the 302 binding regions identified , 61 were mapped to transcriptional units ( genes and operons ) that were shown to be differentially regulated by CodY in the RNA-Seq analysis ( corresponding to a total of 127 genes ) , and thus may represent bona fide targets of CodY under the tested conditions . Notably , 48 DNA regions were associated with transcriptional units regulated in LBMM ( activated and repressed ) and 33 DNA regions were associated with transcriptional units regulated in BHI ( activated and repressed ) , whereas 20 DNA regions overlapped the two conditions ( Fig 3A and S4 Table ) . Importantly , the data revealed that CodY directly regulates 33% of the genes within its regulon , and that it directly binds more regulatory regions under minimal growth conditions harboring low levels of BCAAs than under rich growth conditions , which is surprising given what is known about this regulator ( see Discussion ) . Among the 302 CodY-binding regions identified , 71 ( ~23% ) contained a putative CodY-box/s similar to those found in Lactococcus lactis and B . subtilis [37 , 45] , whereas among the 61 DNA regions that coincided with CodY regulated genes , 24 ( ~40% ) exhibited a CodY-box , as determined by the MAST algorithm that predicts the presence of CodY binding motifs ( Fig 3B and S3 Table ) . This observation suggests that CodY may employ additional binding sites or mechanisms to directly regulate genes . As expected , we found that the ilvD promoter region was bound specifically under high BCAA concentrations , whereas the virulence genes region was bound specifically under low BCAA concentrations; these data points serve effectively as positive controls for the ChIP-Seq dataset ( Fig 3B and S4 Table ) . By combining the data of the RNA-Seq and ChIP-Seq analyses , we were able to determine CodY direct-regulated genes , which further fall into the six described clusters . Under rich growth conditions , CodY directly represses the transcription of BCAAs , histidine , methionine , purine and riboflavin biosynthesis genes as well as the transcription of sigma B , clpC , glycerol uptake and phosphorylation , and a few general metabolic genes ( cluster I ) . Under these same conditions , CodY directly activates the transcription of genes that encode for arginine biosynthesis enzymes , peptidoglycan deacetylation enzymes ( on N-acytelglucosamine ) , and several PTS systems ( cluster II ) ( S4 Table ) . Under minimal growth conditions , CodY directly represses purine biosynthesis , iron transport , a gene involved in pyrimidine biosynthesis and an amino acids permease ( cluster III ) . Under the same conditions , CodY directly activates a cysteine transporter and a specific PTS system , in addition to the virulence regulator , prfA ( cluster IV ) . Surprisingly , within this latter cluster we identified a novel CodY binding region upstream to the actA gene , which is responsible for L . monocytogenes intracellular actin based motility , and that is itself under the regulation of PrfA ( S4 Table ) . Of note , this regulatory relationship represents an additional direct role for CodY in L . monocytogenes virulence . Under both rich and minimal growth conditions , CodY directly represses amino acids transport , PTS systems , genes involved in nitrogen ( e . g . , glutamate dehydrogenase , gdhA gene ) , pyruvate and lipids metabolism , and directly activates motility and chemotaxis genes , the GlnR regulator , other PTS systems and additional metabolic genes ( S4 Table ) . Next , as real-time quantitative PCR ( RT-qPCR ) analysis is still considered the gold standard in gene transcription analysis , we employed it together with ChIP RT-qPCR analysis to validate that CodY directly regulates representative genes from each cluster ( clusters I-VI ) . To this end , we chose genes/operons that contain a putative CodY-box in their regulatory region ( S3 Table and S3 Fig ) . From cluster I , comprising genes repressed by CodY in BHI , we chose the BCAAs and the histidine biosynthesis pathways ( genes tested: ilvD , ilvC , hisG , hisA and hisI ) , as well the sigma B regulator ( sigB ) and the glycerol uptake transporter ( glpF ) . From cluster II , comprising genes activated by CodY in BHI , we chose the arginine biosynthesis pathway and the glutamate decarboxylase gene ( genes tested: argH , argF and gadG ) . From cluster III , comprising genes repressed by CodY in LBMM , iron uptake genes were chosen ( e . g . , feoA ) . From cluster IV , comprising genes activated by CodY in LBMM , prfA and actA genes were chosen . From cluster V , comprising genes repressed by CodY under both conditions , the gdhA gene was chosen [39 , 48] and poxB gene encoding a pyruvate oxidase . From cluster VI , comprising genes activated by CodY under both conditions , flagella and motility genes were chosen ( e . g . , motB , flhA and fliP ) as well the glnR gene , encoding the nitrogen metabolism regulator GlnR . In general , we found the RT-qPCR transcription profiles of the tested genes to be similar to those observed using RNA-Seq analysis . Genes predicted to be repressed by CodY were up regulated in the ΔcodY mutant in comparison to WT bacteria , whereas genes predicted to be activated by CodY were down regulated in the ΔcodY mutant ( Fig 4A ) . The only exception is the observation that the ilv genes exhibit a higher transcriptional level in WT bacteria versus ΔcodY mutant during growth in LBMM . These results suggest that CodY further activates these genes when BCAAs concentrations drop significantly , a phenotype that was previously observed at [18] . ChIP RT-qPCR experiments were performed to verify direct binding of CodY to the regulatory regions of the chosen genes/operons ( in the case of an operon the first gene of the operon was tested ) . Similar to the ChIP-Seq experiments , a CodY-6His variant was used to precipitate DNA fragments during L . monocytogenes growth in BHI and LBMM . Amplification of CodY binding regions upstream to the different genes/operons indeed verified that CodY binds all the tested regulatory regions ( Fig 4B ) . In most cases , the binding of CodY correlated with the corresponding conditions in which CodY’s regulatory activity was observed . Next , to affirm that CodY regulation responds primarily to the availability of BCAAs , we repeated the RT-PCR and the ChIP RT-PCR experiments in bacteria grown in minimal defined medium containing high levels of BCAAs ( HBMM , containing 800 μM of BCAAs ) and compared it to BHI and LBMM conditions ( Fig 4A and 4B ) . Interestingly , most of the representative genes were regulated in HBMM as in BHI ( R2 = 0 . 81 ) , and were different from LBMM ( R2 = 0 . 005 ) , indicating that indeed BCAAs represent the primary ligand of CodY under these conditions . An exception , were the genes of the histidine and the arginine biosynthesis pathways , which were regulated by CodY in BHI ( repressed and activated , respectively ) , but not in HBMM , suggesting that in conjunction with CodY , additional factors ( e . g . , GTP ) mediate the regulation of these pathways under rich conditions . To further characterize the binding of CodY to regulatory regions of select genes/operons , an electrophoresis mobility shift analysis ( EMSA ) was performed using DNA probes comprising the upstream intergenic sequences plus a ~100 bp of the gene 5’-coding sequence . In accord with the ChIP data , we observed binding of CodY to all tested probes with varying affinities . Although EMSA reactions are not at equilibrium and should be considered qualitatively , apparent KD values were derived as following: 150 ± 34 nM for hisZ , 151 . 5 ± 0 . 5 nM for rbsV-sigB , 114 ± 13 nM for glpF , 70 ± 11 nM for argG , 25 ± 13 nM for gadC-gadG , 90 ± 48 nM for feoA , 208 ± 19 nM for actA , 57 ± 3 nM for gdhA , 55 ± 2 nM for poxB , 74 ± 32 nM for glnR and 26 ± 9 nM for fliN ( Fig 5 and S4 Fig ) . Of note , binding of CodY to the regulatory regions of genes specifically regulated in LBMM ( Clusters III and IV ) was measured in the absence of BCAAs . Taken together , these experiments corroborate CodY’s ability to directly bind and regulate different genes in a versatile manner and identified hisZ , rbsV-sigB , glpF , argG , gadC-gadG , feoA , actA , gdhA , poxB , glnR and fliN as novel L . monocytogenes CodY direct target genes , representing metabolic and virulence genes . Finally , intrigued by the observation that CodY directly activates genes involved in flagella biosynthesis , we examined the ability of the ΔcodY mutant to swarm on soft agar plates . WT and ΔcodY bacteria were subjected to a swarming assay on plates containing BHI and LBMM media . In accordance with our findings , the ΔcodY mutant was found to be severely impaired in motility in comparison to WT bacteria on both media tested ( Fig 6 ) . On BHI plates , swarming regions with mean diameters of 1 . 07 ± 0 . 07 cm and 0 . 625 ± 0 . 06 cm were measured for WT and ΔcodY bacteria , respectively ( Fig 6A ) , while on LBMM plates , mean diameters of 1 . 3 cm ± 0 . 09 cm and 0 . 76 ± 0 . 04 cm were measured for WT and ΔcodY bacteria , respectively ( Fig 6B ) . Introducing an ectopic copy of the codY gene to the ΔcodY mutant ( ΔcodY+pLIV2-codY ) restored bacterial motility on both media to WT levels ( Fig 6A and 6B ) . Since the flagella plays a critical role in L . monocytogenes attachment to mammalian cells [49 , 50] , we further examined the ability of the ΔcodY mutant to attach to Caco2 epithelial cells . An attachment assay was performed with WT bacteria , ΔcodY , ΔcodY+pLIV2-codY and a ΔflaA mutant as a control . Indeed we observed a reduced attachment of ΔcodY bacteria to Caco2 cells , a defect that was rescued in a ΔcodY complemented strain ( ΔcodY+pLIV2-codY ) ( Fig 6C ) . Overall , these findings support a more central role for CodY in L . monocytogenes pathogenesis , beyond activation of virulence genes per se , a role that encompasses regulation of metabolic and motility genes , functions important for successful mammalian infection . In this study , we applied a genome-wide approach to identify CodY’s regulon , target genes and regulatory functions in L . monocytogenes . Unlike previous studies of CodY , we examined both rich and minimal growth conditions , in attempt to explore further the possibility that CodY retains activity also when BCAAs , its primary ligands , are in limiting amounts . In contrast to the current dogma that CodY functions only in the presence of its ligand BCAAs , our results clearly demonstrate that CodY functions under both conditions , rich and minimal , and that under each condition it can serve as a repressor and as an activator of genes , establishing for the first time a global regulatory role for CodY under low levels of BCAAs ( Fig 7 ) . Furthermore , this study reveals a broader role for CodY in L . monocytogenes physiology , and particularly in regulation of virulence , as novel CodY direct target genes were discovered that are known to contribute to L . monocytogenes infection of mammalian cells ( discussed below ) . Notably , this study not only builds on previous knowledge of how CodY serves to monitor and fine tune bacterial metabolism , but also expands our understanding of CodY’s spectrum of activities and impact on other central bacterial processes . For the first time , CodY is established as an integrator of bacterial motility , stress related and virulence functions and metabolic adaptations . The two systems level analyses that we employed to delineate the role of CodY in L . monocytogenes growth under rich and minimal conditions were RNA-Seq and ChIP-Seq . The RNA–Seq analysis revealed that CodY affects the expression of hundreds of genes , establishing this transcription factor as a central regulator of L . monocytogenes . The ChIP-Seq analysis showed that CodY directly regulates 33% of its regulon , under both rich and minimal growth conditions . In total , more genes were regulated by CodY ( directly and indirectly ) under rich nutrient conditions ( i . e . , in BHI ) , with the majority repressed , essentially validating CodY’s global role as a repressor of metabolic genes . Nevertheless , a third of the CodY genes regulated under this condition were found to be activated by CodY , many of them related to the TCA cycle ( discussed below ) , demonstrating CodY’s ability to serve as an activator as well . As previously reported in other bacteria , we found that CodY represses amino acid biosynthesis ( mainly BCAAs and histidine ) , purine , riboflavin and certain carbon and nitrogen metabolism genes under rich nutrient conditions [37–42 , 48] . However , we found that CodY activates critical enzymes of the TCA cycle , including glutamate/glutamine derivatives and the arginine biosynthesis pathway , in contrast to what was shown in B . subtilis and L . lactis , where CodY was found to repress these genes [37 , 41 , 51] . This intriguing discrepancy suggests L . monocytogenes may have evolved distinct metabolic network/fluxes to fit its unique lifestyle . Specifically , our data predict that under rich nutrient conditions CodY directs metabolic flux from pyruvate to the TCA cycle through pyruvate carboxylase ( PycA ) , which generates oxaloacetate , while blocking pyruvate flux to the BCAAs biosynthesis pathway through direct repression of the pyruvate oxidase gene , poxB and the ilv operon ( both shown to be directly repressed by CodY ) ( Fig 8 ) . This model is based on our findings that CodY upregulates most of the downstream TCA cycle genes ( encoding four consecutive enzymes converting oxaloacetate to 2-oxoglutarate ) . Since the TCA cycle of L . monocytogenes is missing the enzyme that converts 2-oxoglutarate to succinate ( 2-oxoglutarate dehydrogenase ) [52 , 53] , this step may be bypassed by glutamate synthase converting 2-oxoglutarate together with glutamine to two molecules of glutamate , which are then further converted to GABA by glutamate decarboxylase ( GadG ) , and to succinate by the GABA shunt [54] . A support for this metabolic bypass is provided by the observation that the gene encoding glutamate decarboxylase ( gadG ) was also found to be directly activated by CodY , while the enzyme that reverts glutamate to 2-oxoglutarate , namely glutamate dehydrogenase ( encoded by gdhA ) , was found to be directly repressed by CodY under these conditions ( Fig 8 ) . Moreover , further down this pathway , we found that CodY activates expression of fumarate reductase , which converts succinate to fumarate , in agreement with the premise that CodY positively regulates the TCA cycle and the glutamine/glutamate-GABA bypass ( Fig 8 ) . Interestingly , since L . monocytogenes is also missing the enzyme malate dehydrogenase , which converts malate to oxaloacetate [52 , 53] , conversion of fumarate to malate is a dead end reaction . In this regard , our observation that CodY activates expression of arginine biosynthesis genes during growth in BHI may be explained as a way to consume fumarate through a reversed arginine pathway , generating carbamoyl phosphate that can further feed to nitrogen or pyrimidine metabolism ( Fig 8 ) . In general , this alternative metabolic flux may generate energy and essential precursors to other metabolic pathways to support rapid growth of L . monocytogenes in rich medium . In contrast , during growth in LBMM CodY does not up regulate the expression of the TCA cycle or the arginine metabolism genes and most likely diverts the flux of pyruvate to the generation of BCAAs ( Fig 8 ) . This model may explain the observation that the bacteria grow more slowly in LBMM than in rich medium ( S1 Fig ) . Recently , the second messenger molecule c-di-AMP was shown to be involved in regulation of the TCA cycle and was reported to be essential specifically during growth in rich medium conditions , but not during growth in minimal medium [55] . Interestingly c-di-AMP was found to negatively regulate the pyruvate carboxylase , PycA , ( via an allosteric binding ) thus controlling the conversion of pyruvate to oxaloacetate [56] . Therefore , under conditions where c-di-AMP is absent ( simulated by a diadenylate cyclase mutant , ΔdacA ) and it is expected that PycA activity is enhanced , it follows that the TCA cycle is accelerated , resulting in generation and accumulation of intermediates and byproducts that may be toxic to the bacteria . We propose that while c-di-AMP may play a role in regulation of the TCA cycle under rich conditions ( i . e . , by preventing a high flux of pyruvate to the TCA cycle ) , in minimal medium this function may be dispensable , since part of the pyruvate flux is directed by CodY to the BCAAs biosynthesis pathway . A corollary of this hypothesis is that inactivation of CodY , which in turn reduces flux through the TCA cycle , may alleviate the toxic phenotype of the ΔdacA mutant under rich nutrients conditions . Interestingly , a recent study reported an indirect but intriguing link between DacA and CodY , whereby a mutation in the codY gene rescued the virulence defect of a ΔrelA mutant , which in turn rescued the growth defect of ΔdacA [55] . While more research should be done to explore these phenotypes , it is clear that c-di-AMP and CodY play important roles in shaping L . monocytogenes core metabolism and by doing so affect virulence . Notably , the former is a known listerial immunostimulatory ligand , which is recognized by the innate immune system during infection , and thus exerts additional phenotypes within the host [57–59] . Another novel metabolic relationship identified in this study is the direct activation of GlnR expression by CodY . GlnR is a conserved transcription regulator of nitrogen metabolism genes ( such as those involved in glutamine , glutamate and ammonium metabolism ) [60] . Although both CodY and GlnR were previously reported to independently repress nitrogen related genes ( e . g . , gdhA ) [39 , 48] , a direct relationship was not previously documented . The activation of glnR by CodY may underlie CodY’s observed robust regulation of nitrogen metabolism genes . In addition to metabolic regulation , this study identified a role for CodY in regulation of stress responses . Specifically , under rich nutrient conditions we found that CodY represses , directly and indirectly , the stress induced protease HslUV , the chaperon GroEL-GroES , the osmoprotectant transport system OpuCA and the stress responsive alternative sigma factor , SigB ( σB ) . The latter is shown here for the first time to be a direct target of CodY , indicating a hierarchical regulation of stress related genes down-stream to CodY . Notably , σB is a critical sigma factor of L . monocytogenes , playing a major role in regulation of stress related and virulence genes during mammalian infection [61 , 62] . Sigma B itself is directly involved in PrfA regulation , as it binds the prfA promoter and facilitates transcription during infection [61 , 63 , 64] . Our novel finding that CodY regulates sigB raises the possibility that during mammalian infection , conditions in which BCAAs are limited , CodY may promote prfA transcription by at least two mechanisms: directly via binding to the prfA gene and indirectly by relieving sigB repression . In this regard , we identified also several virulence related genes to be indirectly repressed by CodY , e . g . , inlA and inlB , which mediate bacterial internalization into mammalian cells [3 , 4] . These genes were shown previously to be positively regulated by σB [65] and thus may be repressed in BHI as a result of sigB repression by CodY . Similarly , the bile salt hydrolase ( bsh ) and the osmoprotectant system opuCA are indirectly repressed by CodY during growth in BHI and are both known to be positively regulated by σB [61 , 65] . More generally , these findings reveal tight regulatory cross-talk between three central factors , σB , PrfA and CodY that together coordinate L . monocytogenes adaptation to the mammalian niche with regards to stress , virulence and metabolism , respectively . One of the big surprises of this study was the extent of CodY regulation under conditions when BCAAs are limiting . We found that CodY regulates ( activates and represses ) genes involved in metabolism , motility and virulence . Cluster IV is of particular interest , comprising genes activated by CodY in LBMM , as this cluster includes most of L . monocytogenes major virulence factors in addition to some metabolic genes . CodY regulation of Cluster IV strengthens the premise that CodY attunes virulence functions and metabolic requirements to better adapt the bacterium to the intracellular niche . Notably , three novel CodY direct targets were identified within this cluster in addition to the prfA gene: the actA gene; a cysteine transporter gene; and a PTS system operon ( S3 Fig ) . We confirmed CodY binding to the actA regulatory region by ChIP-RT-qPCR and EMSA assays . Nevertheless , it remains somewhat intriguing why CodY directly regulates both actA and prfA genes , as actA is already under the direct regulation of PrfA . One possible explanation is that by regulating actA CodY is serving as a direct regulatory link between metabolism and motility ( in this case intracellular motility ) , in addition to general regulation of virulence . A role for CodY in L . monocytogenes motility is further highlighted by our findings that CodY directly activates flagellar and chemotaxis genes ( i . e . , extracellular motility ) under both growth conditions . Taken together , these observations suggest that CodY plays a major role in coordinating sensing of nutrients with bacterial movement under diverse conditions and niches . CodY regulation of motility was also documented in B . cereus , where it was shown that CodY positively regulates motility genes and that a ΔcodY strain is less motile [40] . Similarly , we showed in the present study that an L . monocytogenes codY mutant is impaired in motility and attachment to mammalian cells . Above all , the present study establishes that CodY regulation is more complex than classically considered . Previous studies depicted CodY as a transcriptional regulator that represses gene expression either by binding to promoter regions to interfere with RNA polymerase binding or by binding to internal sites leading to transcriptional roadblocks [45 , 66 , 67] . In light of our new data , it appears that CodY could function in all possible states and forms , for example under high and low BCAAs levels , as a repressor and as an activator ( under both conditions ) , with the ability to bind multiple binding sites with different affinities around the genome; such complexity will make future study of CodY highly interesting yet challenging . It is most likely that in vivo the different binding sites are subject to genome wide binding competition , which is dependent on the concentrations of the different CodY forms ( for example bound or unbound to a specific ligand ) . The data also suggests that additional factors are involved in mediating CodY binding to the different sites , as many strong sites that are bound under rich conditions ( e . g . , upstream to the ilv operon ) do not appear to be bound in minimal conditions , although BCAAs are still present , while other binding sites are bound under both conditions . A model of cooperative binding was suggested before for CodY to explain such phenomena [45] , where under a given condition CodY may cooperatively bind DNA in high affinity , but if conditions are changed even a bit ( e . g . , a slight drop in BCAA levels ) , binding is completely lost . Under this scenario , other binding sites with lower affinities might now be accessible for CodY binding and thus better compete . This model may explain why we were able to detect many direct binding sites that are specific to LBMM . In addition , it is most likely that CodY does not work alone and that other transcription regulators and factors influence its activity , as was observed in the case of the histidine and the arginine operons . Such factors can affect CodY’s DNA accessibility , binding affinity or conformation and thus modulate gene expression [28 , 68 , 69] . Notably , GTP is another known ligand of CodY that was shown in other bacteria to affect CodY activity [25 , 28] . It is possible that GTP modulates CodY activity also in L . monocytogenes , and that CodY responds to varying concentrations of GTP in a similar manner it responds to BCAAs . Under this scenario , the relative concentrations of the two ligands under changing environments may determine CodY activities and binding affinities , a hypothesis that awaits further investigation . Of note , we did not observe differences in the transcription of the three relA paralogs of L . monocytogenes ( relA , relP and relQ ) in the RNA-seq data , genes that their products are known to affect intracellular levels of GTP . As in other genome wide binding studies , we observed many CodY binding regions that do not appear to be associated with transcriptional regulation . This phenomenon is generally explained by one of the following scenarios: non-specific protein binding that influences DNA topology; an active mechanism that serves to titer the regulator itself; or redundancy with other co-regulators , as was recently shown for CodY and ScoC interactive regulation , where CodY deletion alone did not result in changes in gene expression of the BCAAs permase , braB [69–75] . This notwithstanding , the observation that many of the identified binding regions do not contain a CodY putative binding box as identified in other bacteria [72 , 76 , 77] , suggests that CodY binding sites in L . monocytogenes are more diverged than in B . subitlis and L . lactis and/or that additional CodY recognition sites might exist . Unfortunately , the resolution of our ChIP-Seq data did not allow us to identify such new sites , though new techniques may lead in the future to identification of such sites . Overall , this study expands our understanding of CodY functions in L . monocytogenes , and we expect these insights to impact the study of CodY in other pathogenic and non-pathogenic bacteria .
L . monocytogenes 10403S [78] was used as the wild type strain . Brain heart infusion ( BHI , Merck ) was used as a rich medium , while high BCAAs minimal medium ( HBMM ) was used as defined medium as in [79] with 100 μg ml-1 of BCAAs [isoleucine , leucine and valine] and low BCAAs minimal medium ( LBMM ) was made with 10 μg ml-1 for each BCAA , which is 76 μM for leucine and isoleucine and 85 μM for valine . Bacteria were grown at 37°C with agitation in BHI , HBMM or LBMM . Bacterial strains used in this study are listed in S5 Table . Bacteria ( WT and ΔcodY strains ) were grown to mid-exponential phase in BHI , HBMM or LBMM at 37°C ( OD600 = 0 . 35 ) . RNA was extracted using the RNAsnap method [80] . For RNA-Seq samples , DNase I ( Qiagen ) treatment was performed on Qiagen RNAeasy columns . For RT-PCR analysis , DNase I ( Fermentas ) treatment was followed by phenol-chloroform extraction . For RNA-Seq samples , the RNA integrity number ( RIN ) was evaluated using a TapeStation instrument ( Agilent Technologies ) and then rRNA was depleted using the RiboZero kit ( Epicentere ) . RNA-Seq libraries were prepared using the TruSeq RNA sample Prep kit ( Illumina ) and sequenced ( 50 nt per read ) by HiSeq 2500 instrument ( Illumina ) at the Technion Genome Center ( Haifa , Israel ) . For RT-qPCR analysis , 1 μg of total RNA was reverse transcribed using QScript reverse transcription kit ( Quatna ) . 16 ng of cDNA were used for RT-qPCR analysis with FastStart Universal Green Master Mix ( Roche ) using a StepOnePlus instrument ( Applied Biosystems ) . The transcription level of each gene of interest was normalized to that of the rpoD mRNA . For the ChIP-RT-PCR analysis , each target gene was first normalized by the levels of rpoD and bglA DNA in each sample ( normalization to multiple control genes is recommended and done using StepOne software ) and then the ChIP sample was normalized to its no-ChIP sample using the StepOne V2 . 3 software . First , ΔCt for each sample is calculated as ΔCt = Average Ct—Normalization Factor ( NF ) , while Normalization Factor is the mean of the selected endogenous controls ( single or multiple genes ) , which is used to normalize the Ct value of each sample . Next , Fold enrichment is calculated as RQ = 2 ( –ΔCt ( ChIP sample ) ) / 2 ( –ΔCt ( no-ChIP sample ) ) RT-qPCR primers are described in S5 Table . Bacterial ChIP analysis was performed as described in [17 , 81] . Bacteria were grown in 50 ml of BHI , HBMM or LBMM at 37°C with shaking at 250 rpm to O . D . of ~0 . 35 . 1% formaldehyde was added to the cultures , which were then incubated at room temperature with shaking at 100 rpm for 20 min . 0 . 5 M Glycine was added to quench excess formaldehyde by shaking for 5 min at room temperature at 100 rpm . Afterwards , the samples were centrifuged at 4000 rpm ( 2600 g ) for 10 min at 4°C , washed twice with cold TBS ( 50 mM Tris-HCl pH 7 . 5 , 150 mM NaCl ) and kept at -80°C . Cross-linked samples were resuspended in 0 . 2 ml of lysis buffer ( 10 mM Tris pH 8 , 20% sucrose , 50 mM NaCl , 10 mM EDTA and 10 mg ml-1 of lysozyme ) and incubated for 30 min at 37°C , and then 0 . 8 ml of IP-buffer ( 50 mM HEPES-KOH pH 7 . 5 , 150 mM NaCl , 1 mM EDTA , 1% Triton X100 , 0 . 1% sodium deoxycholate and 0 . 1% SDS ) supplemented with 1 mM PMSF was added . The samples were lysed using sonication ( 6 rounds of 30 s ) and were centrifuged 10 min at 14000 rpm ( 18000 g ) at 4°C and the supernatants were transferred to new 1 . 5ml tubes . Sheered DNA was analyzed by electrophoresis to observe bands between 200–500 bps . 0 . 8 ml of sonicated sample was used for immuno-precipitation by adding 20 μl of magnetic A/G beads ( Millipore , Cat . 16–663 ) and 5 μl of anti-his tag antibody ( Abcam , Cat . 18184 ) . Samples were then incubated with rotation overnight at 4°C to allow immuno-binding . Beads were collected using magnetic stands and the supernatant was transferred to a new tube to be used as control DNA . Beads were washed twice with 0 . 5 ml of IP-buffer , once with 0 . 5 ml of IP-buffer supplemented with 500 mM NaCl , once with 1 ml of washing buffer ( 10 mM Tris pH 8 , 250 mM LiCl , 1 mM EDTA , 0 . 5% IP-40 and 0 . 5% sodium deoxycholate ) and once with TE buffer ( 50 mM Tris pH 7 . 5 and 10 mM EDTA ) . Finally , the samples were resuspended in 0 . 1 ml of elution buffer ( 50 mM Tris pH 7 . 5 , 10 mM EDTA and 1% SDS ) and incubated 10 min at 65°C . Then the beads were removed by magnetic stands and the supernatants were transferred to a new tube . 80 μl of TE buffer and 2 . 5 μl of RNase A ( 8 mg ml-1 ) were added and incubated for 1 . 5 h at 42°C , followed by 2 h incubation at 42°C with 20 μl of Proteinase K ( Fermentas ) . Then , the samples were incubated overnight at 65°C and the next day the DNA was isolated using Qiagen DNA concentration and cleanup kit . For ChIP-Seq analysis , 50 ng of ChIP and control DNA were used to prepare ChIP DNA libraries using the TruSeq ChIP sample Prep kit and sequenced ( 50 nt per read ) on HiSeq 2500 instrument ( Illumina ) at the Technion Genome Center ( Haifa , Israel ) . On average ~10 million reads were obtained per cDNA library in fastq file , providing a 168-fold genomic coverage ( data was deposited in GEO , accession number—GSE76159 ) . The quality of the reads was evaluated using FastQC ( Babraham Bioinformatics ) and if needed , trimming of reads was performed using the fastX tool kit ( http://hannonlab . cshl . edu/fastx_toolkit/ ) . Mapping of reads followed by upper quartile normalization by gene expression and differential expression analysis by the negative binomial distribution as the statistical model was performed using Rockhopper V1 . 3 with default parameters [82] . For curated clustering of differentially expressed genes 3 criteria were used: minimal value of 10 normalized counts in at least one of the samples , minimal fold change between the WT and ΔcodY samples of at least 1 . 8 in either the BHI or LBMM media , and a significant q-value ( <0 . 05 ) of the negative binomial analysis performed by Rockhopper V1 . 3 . Hierarchical clustering of the differentially expressed genes was performed by the average linkage method using Cluster V3 . 0 ( http://bonsai . hgc . jp/~mdehoon/software/cluster/software . htm ) , and a heatmap was generated using TreeView V3 ( http://jtreeview . sourceforge . net ) . fastq files were obtained and processed as mentioned above ( data was deposited in GEO , accession number—GSE76821 ) . Reads were mapped to L . monocytogenes genome using Bowtie2 running with default parameters [83] . Peak calling for ChIP-Seq analysis was performed using MACS V1 . 4 . 2 with default parameters [84] and with SeqMonk . CodY motif search based on L . lactis and B . subtilis CodY motifs was performed using MAST [47] . Differentially expressed genes from each cluster of the RNA-Seq analysis were used as input for the MIPS server [44 , 85] for functional enrichment analysis . L . monocytogenes CodY-6His was expressed in E . coli strain BL-21 from the pET28 expression plasmid . 10 ml of overnight bacterial culture were diluted in 0 . 5 L of LB medium supplemented with 30 μg ml-1 of kanamycin . Bacteria were grown till O . D . 600 = 0 . 3 , and then induced with 1 mM IPTG for 4 h . The bacteria were harvested by centrifugation ( 2600 g , 10 min ) , washed in 50 ml of cold buffer A ( 0 . 3 M NaCl , 50 mM NaH2PO4 , pH 8 ) and resuspended in 15 ml of buffer A supplemented with 10 mM imidazole and 1 mM PMSF . Bacteria were lysed by an ultra high-pressure homogenizer ( Stansted Fluid Power ) at 12000 psi . Cell debris were removed by centrifugation at 16 , 000 g for 20 min and the lysate was incubated with 1ml of Ni-NTA beads ( Sigma ) for 1 h at 4°C with tilting . The Ni-NTA beads were then loaded on a column and washed with 10 ml buffer A supplemented with 10 mM imidazole . The protein was eluted by 250 mM imidazole in buffer A and dialyzed overnight against 100 ml of buffer A . Protein concentration was determined using a Nanodrop 1000 ( Thermo ) spectrophotometer . 0 . 5 μg of purified protein were separated on SDS-PAGE gel followed by Commassie staining to test for the purity of the protein . For EMSA probes of CodY target genes the upstream intergenic region up to ~100 bps into the coding sequence of the target gene was amplified using PCR and labeled using Roche DIG Gel Shift kit . Purified CodY-6His was incubated with 4 ng of DIG-labeled target DNA in binding buffer ( 20 mM Tris-Cl pH 8 , 50 mM KCl , 2 mM MgCl2 , 0 . 5 mM EDTA , 1mM DTT , 0 . 05% NP-40 , 5% glycerol , 25 μg ml-1 salmon sperm DNA ) for 15 min at room temperature . The samples were then loaded onto a pre-run 8% native acrylamide gel ( running buffer composed of: 35 mM HEPES , 43 mM imidazole buffer , pH 7 . 4 ) and separated for 1 . 5 h at 200 V . Detection of DIG labeled probes was performed using Roche DIG detection kit and visualized using Super RX-N FUJIFILM . For calculations of average apparent KD values , the ratio between the free DNA probe and total DNA probe at each CodY concentration ( i . e . , at each lane ) was quantitated by densitometry analysis using ImageJ software for each EMSA gel [86] . Regression analysis was performed for each gel and the apparent KD value was calculated . Average apparent KD values depicted in the manuscript are based on 2–3 regression analyses of each probe . To demonstrate the reproducibility of the EMSA gels , averaged quantifications of 2–3 biological repeats of each probe were fit using exponential least-squares regression analysis ( shown in S4 Fig ) . The apparent KD values were determined as the concentration at which 50% of the DNA probes were unbound , as deduced the average from the regression analyses . List of primers used for the amplification of target DNA sequences is found in S5 Table . BHI and LBMM soft agar plates were prepared ( 0 . 3% agar ) with 1 mM IPTG . 1 μl from the overnight bacterial cultures in BHI was spotted on the soft agar plates and grown for 48 h at 30°C . For the analysis of swarming capability , diameters of the bacterial growth zones were measured using a standard ruler . Pictures were taken using Olympus camera . L . monocytogenes strains were grown overnight in 3 ml BHI cultures at 30°C without shaking . CaCo2 cells were cultured overnight in 6-well plates in CaCo2 medium ( MEM with 20% FBS , 1 mM sodium pyruvate , 2 mM L-glutamine and 5ml MEM non-essential amino acids X100 solution ) supplemented with penicillin and streptomycin in 37°C incubator with 5% CO2 . The next day , the cells were washed twice with PBS and replenished with fresh CaCo2 medium without antibiotics . The bacteria were washed twice with PBS and approximately 1 . 6x107 L . monocytogenes bacteria were used to infect 2x106 CacCo2 cells . Thirty minutes post-infection , the cells were washed 6 times with PBS and lysed with 1 ml cold water . Serial dilutions were plated on BHI plates and colony-forming units ( CFUs ) were counted after overnight incubation at 37°C .
|
Bacterial pathogens sense multiple host-related metabolic signals that alert them of host localization and result in induction of virulence traits . The Gram-positive foodborne pathogen Listeria monocytogenes activates the transcription of its virulence genes in response to low levels of branch-chained amino acids ( BCAAs ) . This phenomenon is dependent on the global transcription regulator CodY , which binds BCAAs as a ligand and this binding affects its regulatory functions . CodY is classically thought to function under rich growth conditions , when bound to its ligands , however we recently reported that CodY directly activates L . monocytogenes virulence when BCAAs are limited . Identifying this novel CodY activity prompt us to further investigate CodY functions under different growth conditions in a genome wide level . For this purpose , we set on analyzing CodY’s regulon in L . monocytogenes in both rich and minimal growth conditions using genome-wide sequencing techniques . Remarkably , we identified for the first time a global regulatory role for CodY when BCAAs are limited , that are similar to those within the mammalian niche . Furthermore , our data establish CodY as a central regulator that integrates metabolism , motility , stress responses and virulence in L . monocytogenes .
|
[
"Abstract",
"Introduction",
"Results",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"gene",
"regulation",
"pathogens",
"microbiology",
"dna",
"transcription",
"regulator",
"genes",
"gene",
"types",
"bacterial",
"pathogens",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"pathogen",
"motility",
"listeria",
"monocytogenes",
"biochemistry",
"virulence",
"factors",
"genetics",
"biology",
"and",
"life",
"sciences",
"biosynthesis"
] |
2016
|
Systems Level Analyses Reveal Multiple Regulatory Activities of CodY Controlling Metabolism, Motility and Virulence in Listeria monocytogenes
|
In recent years , the East African region has seen an increase in arboviral diseases transmitted by blood-feeding arthropods . Effective surveillance to monitor and reduce incidence of these infections requires the use of appropriate vector sampling tools . Here , trapped skin volatiles on fur from sheep , a known preferred host of mosquito vectors of Rift Valley fever virus ( RVFV ) , were used with a standard CDC light trap to improve catches of mosquito vectors . We tested the standard CDC light trap alone ( L ) , and baited with ( a ) CO2 ( LC ) , ( b ) animal volatiles ( LF ) , and ( c ) CO2 plus animal volatiles ( LCF ) in two highly endemic areas for RVF in Kenya ( Marigat and Ijara districts ) from March–June and September–December 2010 . The incidence rate ratios ( IRR ) that mosquito species chose traps baited with treatments ( LCF , LC and LF ) instead of the control ( L ) were estimated . Marigat was dominated by secondary vectors and host-seeking mosquitoes were 3–4 times more likely to enter LC and LCF traps [IRR = 3 . 1 and IRR = 3 . 8 respectively] than the L only trap . The LCF trap captured a greater number of mosquitoes than the LC trap ( IRR = 1 . 23 ) although the difference was not significant . Analogous results were observed at Ijara , where species were dominated by key primary and primary RVFV vectors , with 1 . 6- , 6 . 5- , and 8 . 5-fold increases in trap captures recorded in LF , LC and LCF baited traps respectively , relative to the control . These catches all differed significantly from those trapped in L only . Further , there was a significant increase in trap captures in LCF compared to LC ( IRR = 1 . 63 ) . Mosquito species composition and trap counts differed between the RVF sites . However , within each site , catches differed in abundance only and no species preferences were noted in the different baited-traps . Identifying the attractive components present in these natural odors should lead to development of an effective odor-bait trapping system for population density-monitoring and result in improved RVF surveillance especially during the inter-epidemic period .
Rift Valley fever virus ( RVFV ) is transmitted primarily by mosquitoes and there are periodic outbreaks of this disease in humans and domestic animals in Africa and the Arabian Peninsula [1] , [2] . Key mosquito vectors involved in the enzootic transmission include flood water Aedes spp . as the primary vectors , and other epizootic culicine vectors such as Mansonia , Culex and Anopheles spp . as the secondary vectors [3] . In Kenya , the number of suspected vectors continues to rise with increasing isolation of the virus from additional species [3] . Since human vaccines and therapeutic treatments are not available for RVFV , surveillance is essential for early warning to ensure that devastating outbreaks and/or sporadic infections are prevented . Efficient surveillance is essential for early detection of increased vector abundance and detection of pathogens in trapped mosquitoes . This requires a systematic collection of mosquito samples and routine testing of mosquito pools for arboviruses in order to assess the status of transmission and to allow for informed decision-making [4] . However , fluctuations in mosquito abundance and arboviral infections pose a challenge for mosquito based surveillance programs , since different surveillance strategies are required to detect different arboviral vectors and infection rates and potential and transmission rates . This is particularly problematic in the case of early detection and during the inter-epidemic periods ( IEP ) , when transmission foci are sporadic and mosquito infection rates are low . Therefore , detection of mosquito infections when there is low transmission requires the collection of large samples of mosquitoes . For West Nile virus , 700 mosquitoes are needed for a modest detection probability of 0 . 5 when the natural infection rate is 0 . 1% for mosquito surveillance programs in the early season or in areas of low transmission [5] . Trapping large numbers of mosquitoes for detection of RVFV can be accomplished by improving the efficiency of existing surveillance traps , such as the standard CO2-baited CDC light trap . One way to improve trapping efficiency is by exploiting the host-seeking behavior of female mosquito vectors . Adult female mosquitoes use host-emitted olfactory cues to locate hosts to obtain blood meals [6] . Domestic animals including cattle , sheep , camels and goats serve as hosts for these vectors of RVFV . However , sheep appear to be more susceptible to RVF infections than cattle or camels [7] , [8] . Whether or not animal susceptibility is associated with increased attraction is unclear; however , it is clear that sheep are preferred hosts of these vectors . We hypothesized that body odors from sheep are important cues used by RVF mosquitoes . The present study was carried out to investigate the response of mosquito vectors of RVFV to the CO2-baited CDC light trap combined with sheep skin odors , in a field setting .
All experiments were conducted at two ecologically distinct sites: Ijara and Marigat districts , which are highly endemic areas for epidemic Rift Valley fever ( RVF ) in Kenya [3] , and are currently under active surveillance for arbovirus activities . Ijara District is located in the North Eastern Province of Kenya and is characterized by a semi-arid to arid climate . Mosquitoes were sampled at Kotile ( 1 . 97°S , 40 . 19°E ) ( near Masalani ) and Sangailu ( 1 . 31°S , 40 . 71°E ) , which is around 60 m above sea level . The average annual rainfall is 540 mm with bimodal peaks recorded from March–June and September–December each year . However , the interannual rainfall variability is very high and reaches abnormal levels leading to floods during El Niño years . Minimum temperatures are always above 20°C , and maximum temperatures reach 30°C to 34°C with a high seasonal and interannual variability . The predominant vegetation is Acacia-Commiphora deciduous bushland and thicket ( Savannah , Shrubland , open to very open shrubs ) , which is much degraded due to overgrazing around the settlement areas . The road leading from Masalani to Sangailu demarcates the boundary between these semi-arid landscapes and the more moist Tana River delta and Boni Forest towards the coast . Boni Forest is an indigenous open canopy forest that forms part of the Northern Zanzibar-Inhamdare Coastal Forest Mosaic . The second study site is Marigat district , located in the Kenyan Rift Valley 250 km northwest of Nairobi where traps were set in surrounding villages/communities of N'gambo ( 0 . 50°N , 36 . 06°E ) and Salabani ( 0 . 55°N , 36 . 06°E ) . The study site covers the basin between Lake Baringo and Lake Bogoria with the town of Marigat as an economic center and lies about 1000 m above sea level . The climate is hot and dry with high rainfall variability , both annually and inter-annually . The average annual rainfall is 650 mm with weak bimodal peaks recorded from March–May and June–August . Temperatures vary from 30 to 35°C , but can rise to 37°C in some months . The low lying arid part of the Baringo basin consists of northern Acacia-Commiphora bushlands and thickets but it has experienced severe land degradation caused by uncontrolled grazing and deforestation . Prosopis juliflora ( Sw . ) DC , locally called mathenge , was introduced to Baringo in the early 1980s for fuelwood production and reforestation as a mitigation measure to stop desertification . The plant was introduced at two sites but now covers large areas , i . e . N'gambo village , one of the vector sampling sites . Three indigenous human communities live in this area , the Ilchamus , Pokot and Tugen . They earn their living through pastoralism and agro-pastoralism keeping large numbers of cattle and small livestock such as sheep and goats . The Perkerra irrigation scheme ( growing of vegetables , maize seed production ) , fishing and tourism provide additional income to these communities . Sheep are the most susceptible among livestock hosts afflicted by RVFV [1] , [8] , [11] , and the living animal has been exploited as a lure in trapping mosquito vectors [12] . Its role in the enzootic maintenance of the RVFV [13] is the reason why it is the preferred domestic animal currently being used as sentinels in an ongoing surveillance program for RVF at the two study sites . The study was conducted with the approval of the national ethics review committee based at the Kenya Medical Research Institute ( KEMRI ) and is renewed on an annual basis after a scientific audit . The Animal use component was also given approval by the KEMRI Animal Use and Care committee ( KEMRI-AUCC ) . KEMRI-AUCC complies with the national guidelines for care and use of laboratory animals in Kenya developed by the Kenya Veterinary Association and the Kenya lab animal technicians association 1989 . The KEMRI-AUCC which approved the study protocol has an assurance identification number A5879-01 from the Office of Laboratory Animal Welfare ( OLAW ) under the Kenyan department of health and human services . For purposes of livestock use , funds from the project were used to purchase animals to monitor RVFV seroprevalence and used for all experimental activities described in this study . These animals were owned and maintained for the study by the project . The project bought 492 animals comprising 5 sentinel herds; two in Marigat , three in Ijara district ( one in Kotile and 2 in Sangailu ) . The animals were left with the owners as part of their flocks but they were not allowed to sell or slaughter them because the project was monitoring the animals . The animals were reverted back to the owner at the end of the project activity . Any newborns born out of the tagged animals belonged to the farmers . We worked in collaboration with the department of veterinary services and veterinary doctors mandated by the government to do livestock sampling and research . The above terms were stipulated well in an agreement between the farmers and the International Centre of Insect Physiology and Ecology ( icipe ) , the hosting institution for the AVID Project Consortium . Experiments were conducted in October and December 2010 during the rains to ascertain the presence of mosquitoes . This comprised 10 replicates of 4 treatments per district . The treatment-trap combinations consisted of the standard CDC light trap alone ( L ) and baited with ( a ) animal volatiles ( LF ) , ( b ) CO2 ( LC ) , or ( c ) CO2 and animal skin volatiles ( LCF ) using fur obtained from living sheep . The animal volatiles consisted of fresh sheep ( Ovis aries Linnæus ) hair samples shaved from the belly and back areas of the animals ( avoiding the head and anal regions ) daily . The animal fur was wrapped in five layers of aluminum foil , kept in a cold box ( 10°C ) and immediately transported to the trapping site ( located between 2 to 5 km ) . Once at the trapping site , approximately 19 g of the animal fur were placed in each canister ( cylindrical in shape with a diameter of 9 . 5 cm and height 22 . 5 cm ) designed from Brass mesh wire ( mesh size , 0 . 15 mm , McNichols Co , Tampa FLA ) . With an inter-trap distance of 40±2 m , the traps were hung in trees 1 . 5±0 . 2 m off the ground and activated within 30 min of sunset ( 1800–1830 ) and trap contents collected within 30 min after sunrise ( 0600–0630 hours ) . Treatments and control were assigned to a predetermined similar area following a Latin square design with days as replicates . Traps were rotated on every trapping day to minimize variability due to trap placement . Dry ice ( 1 kg ) was used as the CO2 source , which was delivered in Igloo thermos containers ( ∼2 L ) ( J . W . Hock , Gainesville , FL ) with a 13-mm hole in the bottom center . Treatments with the canisters containing fur ( which released skin volatiles ) were hung at the base of the standard CDC trap ( battery-powered model 512 , John W Hock Co . , Gainesville , FL ) and when in the presence of CO2 directly in the air flow . All bait canisters were boiled in l0% bleach solution after each nightly trapping to eliminate any residual odor . Mosquitoes caught daily from each of the treatments were anesthetized using triethylamine and identified morphologically to species using taxonomic keys 14–16 . When large numbers of mosquitoes were trapped , they were anesthetized , sorted from other insects and immediately stored in 15 or 50 mL centrifuge tubes , and transported in a liquid nitrogen shipper to the laboratory where they were later identified and the total number by species for each treatment-trap were recorded . Trap count data were analyzed per district and were also subdivided into four categories ( i . e . , key primary vectors , primary vectors , secondary vectors and non-vectors ) based on the relative importance and involvement of member species in RVFV transmission 2 , 3 . The four main categories of trapped mosquitoes recorded in the different treatments were further categorized as follows: flood water Aedes species ( key primary vectors; Aedes mcintoshi and Aedes ochraceus ) ; primary vectors ( Aedes sudanensis/Aedes tricholabis ) ; secondary vectors ( Mansonia and Culex spp . ) and non-vectors , which do not fall into any of these categories ( Table 1 ) . Analysis of key primary and primary RVFV vectors was limited to Ijara district where they were mainly encountered and secondary vectors limited to Marigat district where they occurred in substantial numbers ( Table 1 ) . Daily count of mosquitoes recorded in the various trap treatments were analyzed using a generalized linear model with negative binomial error structure and log link using R 2 . 11 . 0 software [17] . Using the treatment L only ( control ) as the reference category , the incidence rate ratios ( IRR ) that mosquito species chose other treatments ( LCF , LC and LF ) , instead of the control , were estimated . The IRR for the control is 1 ( unity ) and values above this indicates better performance and values below under performance of the treatments relative to the control .
The distribution of RVFV mosquitoes captured per treatment-trap combination for the two districts are contained in Table 1 . Mosquito species composition and trap captures differed markedly between the two districts which might suggest varied habitat preferences for each mosquito species . Differences in abundance were observed between the treatments with no clear pattern of preference of any species for a particular trap treatment . Some species were not caught in all replicates , and it is unclear if such variability was due to overall low population densities or the mosquitoes failing to enter ( or to respond to ) the traps . In general , traps baited with CO2 ( LCF and LC ) captured more mosquitoes than those without ( LF and L ) ( Table 1 and Figures 1 and 2 ) . There was a significant effect of treatments compared to the unbaited CDC trap on overall mosquito captures from Marigat ( χ2 = 20 . 68 , df = 3 , p<0 . 001 ) and from Ijara ( χ2 = 37 . 51 , df = 3 , p<0 . 001 ) . Trap catches from Marigat indicate that , compared to L only , LC and LCF traps caught 3–4 times more host-seeking mosquitoes [IRR = 3 . 1 for LC and IRR = 3 . 8 for LCF] . LCF traps recorded higher mosquito catches compared to LC traps ( IRR = 1 . 23 ) although the difference was not statistically significant ( Figure 1 ) . Similarly , the LF trap caught slightly more mosquitoes ( IRR = 1 . 03 ) than L only but was not significantly different ( Figure 1 ) . Similar findings were observed at Ijara where there was a significant treatment effect on mosquito catches ( χ2 = 37 . 51 , df = 3 , p<0 . 001 ) . Carbon dioxide ( LC ) , CO2+fur ( LCF ) significantly increased trap captures by 6 . 5 and 8 . 5 times , respectively , compared to the control . The LF caught more than the control , L , but this was not statistically significant ( Figure 1 ) . Trap catches at Ijara were dominated by flood water aedine mosquitoes categorized as key and primary RVFV vectors; these species were sparse or absent at Marigat . There was a highly significant effect of treatments on key primary RVFV vectors ( χ2 = 199 . 99 , df = 3 , p<0 . 001 ) . For this group , relative to the control , there was a 4 . 0- and 6 . 5-fold significant increase in captures recorded in LC and LCF traps , respectively ( Figure 2 ) . Additionally , LCF capture rates were significantly higher than LC capture rates ( IRR = 1 . 63 ) . A significant effect of treatments on RVFV primary vectors was also evident ( χ2 = 74 . 24 , df = 3 , p<0 . 001 ) . Compared to the control , the treatments LF , LC and LCF caught 2 . 9 , 31 and 42 times as many primary vectors . Interestingly , for this group , there was a 34% significant increase in captures for traps baited with LCF compared to LC ( IRR = 1 . 34 ) ( Figure 2 ) . Marigat yielded very low catches for key primary vectors and there was a total absence of primary vectors . Therefore results are only presented for secondary vectors . For secondary vectors at Marigat , there was a highly significant effect of treatments on the mosquito catches ( χ2 = 22 . 94 , df = 3 , p<0 . 001 ) . Relative to the control , there were 3 to 4-fold increases in captures for LC and LCF traps , respectively ( Figure 2 ) . Comparable captures were recorded for LF and L traps with only a slight increase recorded in LF baited traps relative to the control ( IRR = 1 . 04 ) . Captures rates , although not significant were higher for LCF traps than LC traps ( IRR = 1 . 24 ) ( Figure 2 ) . Mosquito collections within this category at Ijara were low and dominated by Cx . pipiens s . l . with an observed increase in captures in the other treatments compared to L . The non-vectors category included species of the genera Ficalbia , Coquilettidia , Anopheles and Aedes ( Stegomyia ) . Members of these genera occurred in low numbers in both districts , especially at Ijara ( Table 1 ) . However , data for Marigat suggest a bias in trap captures in LCF and LC , compared to L although there were no significant differences in the captures between these treatments , while similar trap captures were observed for the LF and L-baited traps . Non-mosquito species notably beetles and moths were trapped in addition to mosquitoes but were not included in our data .
The results demonstrate that more mosquitoes were caught in traps that contained a release of the combination of sheep odors+CO2 and were in most cases the most attractive bait compared to the conventional CO2-baited light trap . This confirms that odors emanating from sheep fur play a role in host-location by these mosquitoes . The attractive effect was highly evident in captures of flood water aedines comprising key and primary RVFV vectors as well as secondary vectors . The effectiveness of sheep is supported by a study on blood meal patterns during a RVF outbreak where widespread feeding on sheep was observed [18] . Moreover , most mosquitoes belonging to the Culex , Mansonia and Aedes genera have been reported to feed opportunistically and readily on mammals [19]–[21] . The entire animal body emanations comprising breath and skin volatiles influence the outcome of mosquito host-seeking process [22] . Research has indicated that animal skin emanations have a kairomonal ( attractive ) effect on mosquitoes while breath volatiles have an allomonal or repellent effect [23] . Skin body odor may be the primary factor for mosquito attraction and discrimination when mosquitoes are in close proximity of a host . It is therefore not surprising that addition of skin volatiles captured in sheep fur enhanced captures of mosquitoes attracted to sheep hosts when combined with the conventional CO2-baited light trap . The effect of sheep skin odors emanating from fur was not evaluated alone but in combination with CO2 and/or light which are known attractants for mosquitoes and other biting flies . Although animal odors enhanced trap captures when added to either CO2-baited light trap or light trap only , the captures were greater in the CO2-based blend than in the combination without CO2 . However , the crude animal skin odor in traps is imperfect because of possible loss of volatile attractive components over time compared to the dynamic production from live animals . Therefore it would be beneficial to identify the attractive compounds and develop a synthetic blend . In some replicates there was a decrease in trap catches when host odor was added to light or to CO2 . This could be attributed to variation in attractiveness of the batches of animal fur used in the daily trapping experiment as odors used were not from the same animal; low occurrence of targeted mosquitoes , as observed at the districts for certain vector categories; a difference in preferred host other than sheep e . g . Cx . poicilipes between LCF and LC baited traps ( Table 1 ) ; and volatiles from fur are a static system and most volatile compounds evaporate first and therefore the odor profile changes . The effect of host odors did not markedly influence trap catches of the non-vector category . The low abundance of mosquitoes was insufficient to observe a significant preference in trap catches for the different treatments used; even though there was an increase in trap catches for those baited with host odors compared to light only . The effect of CO2 on trap catches was not evaluated independently; however , its effect was evidenced in the difference in trap catches between LC and L baited traps . The data support its role in enhancing trap captures [24] , especially for RVFV vectors . Our experimental setup excluded landing response as a measurement , instead focussing solely on trap catch . The goal was to evaluate animal fur containing skin emanations that provided attractive stimuli . However , the large response of these mosquitoes to CO2 , suggests that it can serve as a good positive control for evaluating candidate synthetic attractants of skin origin for this group of arbovirus vectors of medical and veterinary importance . Our preliminary trials ( data not shown ) and earlier studies highlighted the importance of these attractants in flight activation of mosquitoes towards host odors [25] , [26] . This justifies the inclusion of these well-known long-range attractants in trap design . Our data suggest that host skin odors other than CO2 are important in enhancing mosquito trap captures in concurrence with studies reporting enhanced effect of mosquito attraction to animal skin volatiles in the presence of CO2 or light [22] , [27] , [28] . It is well-known that many nocturnally-active hematophagous insects are attracted to light [29] , [30] . In conformity with earlier findings [31] , our results show that light as a visual cue is enhanced by sheep skin odors and CO2 . Besides being non-specific , previous studies have argued that CO2 activates mosquitoes to initiate host-finding , but may not necessarily attract it and at close range , can actually act as a deterrent [26] and be of limited use in host discrimination [32] . Although this was not the subject of our study , CO2 increased trap captures in the presence of host skin odors , in agreement with previous research [24] , [25] , [33] . The observed trap captures recorded in the LCF traps were generally higher compared to those caught in the LC traps . Nonetheless , among the mosquito species trapped , differences in capture rate were not observed between the LCF and LC traps . Therefore , CO2-baited light traps may be adequate for monitoring and surveillance of these species . However , for effective arbovirus disease surveillance , an improved sampling method is vital especially during the inter-epidemic period where transmission foci are sporadic and infected vectors are rare . Emphasis needs to be placed on increasing the collections with an additional advantage of depicting the dynamics of populations . Beyond the already described finding that animal odor inclusion increases trap catch with CO2 present , there were some cases where it suppressed trap catch . In some cases , lower catches of the LCF trap were noted on days with light showers; therefore , precipitation may have interfered and reduced mosquito attraction to skin odor baits as observed before by Olanga et al . [34] . However there was no record of variation in weather patterns during the study period . Another possibility is the variation from fur samples used in this study . Samples were obtained from various animals without prior assessments of their degree of attractiveness . Animals in a herd are known to vary greatly in their attractiveness to mosquitoes [35] , [36] . Reduced attraction due to loss of important volatile compounds during the fur extraction process remains plausible . A higher number of Cx . poicilipes were collected in CO2-baited light traps than in similar traps baited only with host skin odor , although the difference in trap captures was not significant . This suggests that sheep are not preferred hosts for this species . However , the effect of CO2 in the presence of host-related odors may be variable and a strong attraction response may be observed with often different responses between species [37] , [38] . This observation might emphasize the importance of trap placement in the sampling process , though it is not certain if this species would be attracted to the host that emanates the greatest amount of CO2 in nature . Differential catches of Cx . pipiens s . l . and An . gambiae s . l . to odors from sheep fur were recorded at the different sites ( Table 1 ) . Among the species complexes captured , there are known marked differences in olfactory responses between members of the complexes [39] , [40] . Culex pipiens preferentially feed on birds [41] although they can adapt and readily feed on mammals in proportions possibly based on host abundance [42] . The observed differences in trap responses in the highest trap treatments ( i . e . , LC and LCF ) at both districts may indicate different spatial feeding preferences in geographically separate populations . Related response patterns of discrete populations of mosquito species to host odors has been reported [42] , [43] , as such , it may be worthwhile to include a preference test involving odors from other livestock hosts in field bioassays . Only volatiles from skin emanations captured in the fur were tested in this study . Studies on other volatile sources involved in host attraction to hematophagous flies have been reported from feces [44] and urine [45] , [46] . In this regard , other sources of attractive odors might contribute to the attraction of mosquitoes and combination using these odors may be worth investigating . Laboratory bioassays have commonly been used to evaluate the effect of semiochemicals on mosquito behavior whilst minimizing other environmental variables . However , such an approach is inadequate for predicting effects on natural populations and on ecosystem-level features [47] . Alternatively , insect behaviors have been assessed in the field by baiting traps with extracts of animal volatiles [42] , [48] . Use of whole animals provides another approach but it becomes difficult to delineate individual contributions of attractants from breath or skin emanations or other exogenous compounds to the overall trap catches . Our design followed a field-based approach to evaluate the role of skin emanations on mosquito trap catches . The design can account and provide for an understanding of heterogeneities which dramatically influence dynamics of natural systems . This is similar to the trapping design employed by Njiru et al . [49] and Jawara et al . [50] to investigate mosquito captures in conventional traps baited with human foot odors trapped on nylon stockings . Although the contribution of geographical variability to the total variance was not estimated , possible experimental confounders such as time , location and environmental influence are unlikely to affect the overall observed results as the present experiments were performed at a variety of sites with different animals of the same species and treatment traps treated alike . As such , it is likely that many of the mosquitoes approaching the trap had the opportunity to sample more than one of the treatment-traps , and may have made a choice between them . Albeit the above mentioned challenges , the use of crude volatiles in the field approach presented in this paper can contribute to the evaluation of the effect of host volatiles in the standard CDC light trap . In conclusion , the addition of sheep skin odor to the CO2-baited light trap improved trap catches of RVFV vectors in line with similar findings reporting enhanced effect of animal skin odors and other cues such as CO2 [27] , [28] , [44] . Our results indicate host skin olfactory cues are important signals in mediating mosquito host location . The finding is also in accordance with the consensus that additional compounds other than CO2 from animal skin may be exploited by mosquitoes in host location [51]–[53] . Sheep skin odor contributes to the attraction of host-seeking RVFV mosquito vectors . Identification of chemicals emanated by sheep might provide the basis for the development of improved devices to sample these vectors . However , refinements into an effective monitoring tool requires identifying and understanding the specific behavioral effects of the attractive components present in these skin odors which is currently underway .
|
The East African region is a major epizootic center for endemic and emerging mosquito borne-arboviruses such as Rift Valley fever virus ( RVFV ) , as evidenced by the increasing frequency and magnitude of this disease . The absence of vaccines or prophylactic drugs for most of these diseases emphasizes the need for accurate sampling of mosquito vector populations and testing for arboviruses . Accurate surveillance is crucial for early warning of potential or assessing mitigation of existing outbreaks . However , it is a challenge to sample mosquitoes in adequate numbers during the inter-epidemic periods ( IEP ) because this period is characterized by low mosquito population densities , sporadic transmission foci and low mosquito infection rates . Therefore more efficient tools are needed to increase capture rates so maximized virus detection probability in the mosquitoes can be achieved for assessing risk and outbreak predictions . This can be accomplished by exploiting the host-seeking behavior of adult female mosquitoes and the olfactory cues used to locate a potential host . Here , odors emanating from fur of sheep , a susceptible host for RVFV , is shown to improve trap capture rates of mosquito vectors of RVF in a standard surveillance trap . These data provide for future investigations to identify attractive components present in these natural odors , so that they can be incorporated into existing traps to serve as a population density-monitoring tool for improved arbovirus disease surveillance during IEP .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"chemistry",
"biology",
"veterinary",
"science",
"agriculture"
] |
2012
|
Sheep Skin Odor Improves Trap Captures of Mosquito Vectors of Rift Valley Fever
|
Extracting network-based functional relationships within genomic datasets is an important challenge in the computational analysis of large-scale data . Although many methods , both public and commercial , have been developed , the problem of identifying networks of interactions that are most relevant to the given input data still remains an open issue . Here , we have leveraged the method of random walks on graphs as a powerful platform for scoring network components based on simultaneous assessment of the experimental data as well as local network connectivity . Using this method , NetWalk , we can calculate distribution of Edge Flux values associated with each interaction in the network , which reflects the relevance of interactions based on the experimental data . We show that network-based analyses of genomic data are simpler and more accurate using NetWalk than with some of the currently employed methods . We also present NetWalk analysis of microarray gene expression data from MCF7 cells exposed to different doses of doxorubicin , which reveals a switch-like pattern in the p53 regulated network in cell cycle arrest and apoptosis . Our analyses demonstrate the use of NetWalk as a valuable tool in generating high-confidence hypotheses from high-content genomic data .
An important challenge in the analyses of high throughput datasets is integration of the data with prior knowledge interactions of the measured molecules for the retrieval of most relevant biomolecular networks [1]–[7] . This approach facilitates interpretation of the data within the context of known functional interactions between biological molecules and subsequently leads to high-confidence hypothesis generation . Typically , this procedure would entail identification of genes with highest or lowest data values , which is then followed by identification of associated networks . However , retrieval of most relevant biological networks/pathways associated with the upper or lower end of the data distribution is not a trivial task , mainly because members of a biological pathway do not usually have similar data values ( e . g . gene expression change ) , which necessitates the use of various computational algorithms for finding such networks of genes [1] , [2] , [4] , [5] , [8]–[11] . One class of methods for finding relevant networks utilize optimization procedures for finding highest-scoring subnetworks/pathways of genes based on the data values of genes [2] , [8] . Although this approach is likely to result in highly relevant networks , it is computationally expensive and inefficient , and is therefore not suitable for routine analyses of functional genomics data in the lab . The most popular of the existing methods of extraction of relevant networks from genomic data , however , usually involve a network building strategy using a pre-defined focus gene set , which is typically a set of genes with most significant data values ( e . g . most over-expressed genes ) [1] , [7] . The network is built by “filling in” other nodes from the network either based on the enrichment of interactions for the focus set ( IPA -Ingenuity Pathway Analysis ) [1] , or based on the analysis of shortest paths between the focus genes ( MetaCore ) [7] , [12] . Both methods aim at identifying genes in the network that are most central to connecting the focus genes to each other . Problems associated with these methods have been outlined previously [7] . However perhaps most importantly , the central genes identified by these methods may have incoherent data values with the focus genes ( e . g . the central genes may have reduced expression while the focus genes may have increased expression ) , as data values of nodes are not accounted for during the network construction process using the seed gene list . This may result in uninformative networks that are not representative of the networks most significantly represented in the genomic data ( see Results ) . In addition , these methods do not account for genes with more subtle data values that collectively may be more important than those with more obvious data values [13] . Although powerful data analysis methods for finding sets of genes with significant , albeit subtle , expression changes have been developed ( e . g . GSEA [13] , Molecular concept maps[14] , GenMAPP[15] ) , such an approach has not been incorporated into methods for extracting interaction networks that are most highlighted by the data . In order to overcome these problems , we have employed the method of random walks in graphs for scoring the relevance of interactions in the network to the data . The method of random walks has been well-established for structural analyses of networks , as it can fully account for local as well as global topological structure within the network [16] , [17] and it is very useful for identifying most important/central nodes [16]–[18] . Here , instead of working with a pre-defined set of focus genes , we overlay the entire data distribution onto the network , and bias the random walk probabilities based on the data values associated with nodes . This method , NetWalk , generates a distribution of Edge Flux values for each interaction in the network , which then can be used for dynamical network building or further statistical analyses . Here , we describe the concept of NetWalk , demonstrate its usefulness in extracting relevant networks compared to Ingenuity Pathway Analysis , and show the use of NetWalk results in comparative analyses of highlighted networks between different conditions . We tested NetWalk on experimentally derived genomic data from breast cancer cells treated with different concentrations of doxorubicin , a clinically used chemotherapeutic agent . Using NetWalk , we identify several previously unreported network processes involved in doxorubicin-induced cell death . From these studies we propose that NetWalk is a valuable network based analysis tool that integrates biological high throughput data with prior knowledge networks to define sub-networks of genes that are modulated in a biologically meaningful way . Use of NetWalk will greatly facilitate analysis of genomic data .
Integration of genomic data represented by a vector w with the network data of interactions between genes ( nodes ) is performed by representing each interaction ( edge ) in the network in the form of a transition probability based on the data values ( e . g . mRNA expression change , phenotype score from a genetic screen ) of nodes within the immediate neighborhood: ( 1 ) where pij is the transition probability from node i to node j , wj is the experimental value for node j , and Ni is the set of immediate downstream neighbors ( undirected edges are considered bidirectional ) of node i . If there are no downstream nodes of the node i ( |Ni| = 0 ) , pij is set to pij = 1/|n| for all j , where n is the set of all nodes in the network . The relevance score of each node in the network is defined by the probability of its visitation by the random walker , which is a function of both the local network connectivity as well the data values of nodes . So at any step k of this “random walk” process , the probability of a node being visited by the random walker is ( 2 ) where is the probability of node i at step k , pji is the transition probability from node j to node i and N is the set of interacting neighbors of node i . This can be represented in a matrix form ( 3 ) where gk is the vector of probability values for all nodes in the network at step k , and P is the transition probability matrix of the network . Obviously , since a “walk” can only be performed over adjacent nodes , pij>0 only if nodes i and j directly interact . The expression above can also be written as ( 4 ) where is the transition probability matrix raised to the power k , and is the initial probability distribution over nodes ( all 1/|n| ) . By the Perron-Frobenius theorem for stochastic matrices , as ( infinite random walk ) , the expression above converges to ( 5 ) where g is the left eigenvector of P associated with eigenvalue 1 and contains the final visitation probability values of nodes . The final visitation probabilities of nodes depend on their data values , data values of their neighbors , as well as the local network connectivity . In order to further bias the random walk towards the input data values , we assigned a small probability q that the random walker will return to its starting node . Therefore , the expression for random walk with restart is given by ( 6 ) where q is a vector of all q of length |n| and 1 is a vector of all 1: so that the restart probability is uniform among all nodes . However , we bias the restart probabilities to the data values of nodes , so that the random walker is more likely to return to its initial node if the data value of that node is high . ( 7 ) In this way , the probability that the random walker will restart at another node i is directly proportional on the data value of node i , thereby even more biasing the process of random walk to the biological data . In the case of transcription factor - target gene interactions , these were reversed in the network so that the node values of target genes would contribute to the probabilities of the transcription factors , rather than the other way around . This is because the data values of target genes ( i . e . mRNA expression change ) are more informative of identifying regulation by transcription factors . To find networks of interactions between genes represented in the data , we scored each interaction in the network by ( 8 ) where eij is the flux through edge ij and represents the score of importance of the interaction based on the data . The node visitation frequencies in a random walk directly reflect the relative centralities of nodes in the network , and therefore are highly biased towards the local network topology . Although biasing the random walk to data values skews the visitation frequencies towards the supplied data values , there is still a significantly high correlation with node connectivity values ( Figure S1 ) , which suggests that the random walk process is highly biased to the highly connected hubs in the network . Therefore , it is important to control for topological bias in the network that stems either from its scale-free nature or the historical bias of highly studied genes . In order to control for topological biases in the network , we also calculated background visitation frequencies ( 9 ) which is the same expression as in equation ( 7 ) , with the exception that . Pr is a transition probability matrix formed by letting wi = 1 for all i . Since gr is calculated without considering the data values of genes , it contains all the topological bias in the network . Therefore , to obtain relative visitation frequencies of genes ( g′ ) , we normalize values in g by those in gr , ( 10 ) Relative visitation frequency values in g′ have minimal correlation with node centralities , and have a high correlation with the supplied gene expression measurements ( Figure S2 ) , which indicates that relative visitation frequencies of nodes are highly biased towards the data . Normalization of edge flux values is done by first calculating ( 11 ) where er is the edge score distribution vector calculated by letting wi = 1 for all i . Then , we normalize the data-biased edge flux values to er to obtain normalized Edge Flux of interaction ij ( 12 ) which gives the final normalized score distribution of edges , which reflects edge fluxes of nodes relative to what would be expected by topology alone in the given network . Because of the nature of random walks described above , the input values must be positive , possibly representing ratio of a test versus control sample ( e . g . ratio of mRNA expression levels of treated to untreated samples ) . Missing values in the network are then assigned a value of 1 , which represents a no change case in ratio values . Accordingly , the values of s are centered around 0 , with higher values meaning higher probability relative to what would be expected by chance in the given network ( i . e . networks of high data value nodes , e . g . increased gene expression ) , and lower values meaning lower visitation probability ( i . e . networks with low data values , e . g . reduced gene expression ) ( see below ) . In order to prevent disproportionate skewing of the node probabilities with extreme outliers in the data , the input data is normalized so that all w>k0 . 999 are assigned k0 . 999 , where k0 . 999 is the 99 . 9th percentile value of w . Similarly , all w<k0 . 001 are assigned k0 . 001 . With this procedure , the final normalized visitation frequencies of nodes are highly robust to differences in data distributions and ranges ( see Figure S3 ) . We compiled protein-protein interactions from online databases HPRD [19] BIND [20] , HomoMINT [21] , Gene [22] and IntAct [23] . For directed interactions , we compiled signaling interactions from KEGG [24] , BioCarta ( http://pid . nci . nih . gov/ ) and TRANSPATH [25] , as well as through manual curation of the undirected interactions based on published literature . Transcription factor-target interactions were obtained from ORegAnno [26] and TRANSFAC [27] databases . This resulted in a network of 10 , 473 genes connected by ∼65 , 000 interactions . In network-based analyses of genomic data , the analyses and therefore resultant hypotheses are limited by the gene coverage of the network . Therefore , it is crucial that the interaction network has as much gene coverage as possible . Since our main goal of network-based analyses is identification of relevant biological processes , the interactions represented in the network need not be direct physical interactions . For example , a concordant increase in the expression of genes involved in glucose metabolism will not be captured in network-based analyses of direct physical interactions , as metabolic enzymes within the same pathway rarely engage in direct physical interactions ( with the exception of multifunctional complexes ) . Therefore , inclusion of indirect functional interactions in the network may help identify relevant biological processes that are not captured by direct interactions ( see network plots below ) . In order to increase the coverage of our network , we added functional similarity interactions between genes , where an interaction means that the genes are involved in similar functional processes , such as a metabolic pathway ( e . g . glycolysis ) or a specific enzymatic reaction ( e . g . oxidation/reduction ) . Functional similarity interactions were constructed using Gene Ontology ( GO ) annotations [28] as defined in the Entrez Gene database , and also metabolic pathway annotations in the KEGG database . Any two genes sharing a metabolic pathway annotation ( but not signaling pathways as they are already represented in protein-protein interactions ) from KEGG were assigned an interaction . In the case of GO annotations , two genes were assigned an interaction if the overlap of their GO annotations was significant compared to the rest of the genes:where sij is the significance of overlap between genes i and j; Gk is the set of genes that have the GO term k; N is the set of GO terms common to genes i and j , and n is the total number of genes . If sij<0 . 001 , genes i and j were assigned an interaction . Our final network contains 14 , 506 genes connected by 189 , 901 interactions . Gene coverage of our network of genes in our doxorubicin dataset is comparable to that in the Ingenuity Pathway Analysis ( 13 , 329 in our network versus 13 , 880 in IPA ) . MCF7 cells were grown in DMEM ( Invitrogen ) supplemented with 10% FBS ( Gemini ) to near confluency and treated with 1 or 10 µM Doxorubicin ( Sigma ) . Cells were collected at 0 , 6 , 12 and 24 hours post-treatment . Cell lysis and RNA extraction was done using Mirvana miRNA isolation kit ( Ambion ) and amplification using Illumina TotalPrep RNA amplification kit ( Ambion ) . Equal amount of RNA from each sample was hybridized to Illumina HT12 BeadChip ( Illumina ) . All procedures were performed exactly as described in the respective manuals . The experiments were repeated in triplicate . Networks in IPA were generated using Core analysis with indicated data cutoffs for upregulated genes and using direct interactions with the cutoff for network size to be 70 . Highest scoring 5 networks were merged and exported as text files . All network plottings were done using the gplot function in the sna package for R ( http://erzuli . ss . uci . edu/R . stuff/ ) . Cells were treated as indicated and lysed in a sample lysis buffer ( 50 mM Hepes , 150 mM NaCl , 1mM EGTA , 10 mM Sodium Pyrophosphate , pH 7 . 4 , 100 nM NaF , 1 . 5 mM MgCl2 , 10% glycerol , 1% Triton X-100 plus protease inhibitors; aprotinin , bestatin , leupeptin , E-64 , and pepstatin A ) . Blotting was done using antibodies against p53 ( Cell Signaling ) , p21 ( Cell Signaling ) and Actin ( Sigma ) . The experiment was done in triplicate . FACS: Cells were treated as indicated and after 24 hours trypsinized , fixed with 70% ethanol at −20°C for 10 minutes and resuspended in Propidium Iodide solution . FACS analysis was performed in the Flow Cytometry core facility of M . D . Anderson Cancer Center . Rhodamine 123 assay: Rhodamine 123 staining was performed as described [29] . Briefly , cells were treated as indicated and after 24 hours , trypsinized , spun down and resuspended in 10 µM Rhodamine 123 ( Invitrogen ) in PBS for 30 minutes . Cells were washed in PBS and analyzed by FACS for Rhodamine 123 intensity ( green ) .
Identifying common biological roles of genes whose expression are altered in a microarray experiment is one of the most frequently used strategies to understand the underlying biological processes and derive hypotheses [6] , [13]–[15] , [30] . This strategy is also implicit in NetWalk ( Figure 1 ) , as node visitation frequency values ( hence EF values ) calculated by NetWalk are based on 1 ) data values of nodes , 2 ) data values of their network neighbors and 3 ) the network connectivity among neighbors . Therefore , a node with a high data value that interacts with other nodes with high data values in the network will receive the highest node visitation and EF scores . Similarly , a node with a low data value that interacts with other nodes with low data values in the network will receive the lowest node visitation and EF scores . In order to test the dependency of NetWalk output on the provided data , we performed deletions of portions of data and compared the resultant visitation frequencies to those of the original dataset . Correlation of node visitation frequencies to those of the full dataset closely followed the input data , suggesting that NetWalk output is highly dependent on the supplied data ( Figure S4 ) . However , this may also suggest that NetWalk output is mostly independent of the network connectivity . In order to test the dependence of NetWalk output on the network connectivity , we removed parts of the network and performed NetWalk analysis on the perturbed networks . The resultant node visitation frequencies correlate relatively poorly with those of the original network ( Figure S5 ) , indicating that the network connectivity substantially contributes to node visitation frequency values . We also performed a similar analysis with random deletions and additions of edges , rather than nodes , in the network , and found a similar dependence of the NetWalk output on the network connectivity ( Figure S7 ) . These analyses demonstrate that NetWalk output is highly dependent on both the supplied data as well as the network information . To demonstrate the use of NetWalk in the extraction of relevant networks out of microarray gene expression data , we studied gene expression profiles of MCF7 cells subjected to sub-lethal and lethal doses of doxorubicin . We performed microarray gene expression analysis of MCF7 cells before and after treatment with 1 or 10 µM doxorubicin for 6 , 12 and 24 hours . In these cells , 1 µM doxorubicin causes a cell cycle arrest in S-phase , while a 10 µM dose induces cell death ( Figure 2A–B ) . A NetWalk analysis of the ratio values ( treated/untreated ) for 1 µM treatment was performed using q = 0 . 01 ( see Methods ) . The resulting distribution of edge flux values , and plots of edges with 100 highest and lowest EF values can be seen in Figure 2C–D . EF values are strictly biased towards the data , as the high and low-end networks are entirely composed of genes with , respectively , increased and reduced expression levels . In the Figure 2D , interactions in the cluster made of GLS , GLS2 , P4HA2 , ODC1 and PRODH genes ( arginine and proline metabolism ) have the highest EF scores due to both their high data values and tight interconnections with each other . Similarly , in the low-score network in Figure 2C , interactions in the cluster containing NDC80 , CENPK , CBX1 , CENPA and SGOL1 ( centriole components ) have the lowest EF scores . Nodes with moderate values that are in close proximity to other high value nodes within a tightly connected neighborhood will also get high scores , as is seen with TP53 in Figure 2B . In order to demonstrate that the p53 network extracted by NetWalk is not an artifact of highly connected subnetworks , we performed a NetWalk analysis of baseline expression profile of MCF7 cells relative to other breast cancer cells as reported by Neve et al [31] . The most significantly upregulated networks in MCF7 cells relative to the rest of 53 breast cancer cells are those involved in the Estrogen Receptor signaling ( Figure S6 ) , a well-characterized dominant pathway in the estrogen receptor positive MCF7 cells . This analysis shows that NetWalk output does indeed reflect accurate quantification of highly biologically relevant networks based on the supplied data . Contrary to the seed-based network building methods , NetWalk works with the whole data distribution and so does not require assignment of pre-defined cutoffs or focus gene sets . NetWalk procedure simply translates the gene-centric data values to corresponding interaction scores based on the coherence of the gene values with those in the local network neighborhood as well as the local interaction pattern in the network . Therefore , the results can be viewed at any user defined cutoff value for flexible generation of networks with highly coherent node values . The distribution of input node values and sample networks with different EF cutoffs shows that the node values within networks are highly coherent across a wide range of EF score cutoffs , which allows for high-confidence hypothesis generation about activated and inactivated network processes in response to DNA damage ( Figure 3A–B ) . In comparison , the distribution of data values of nodes in the networks generated by Ingenuity Pathway Analysis , which takes a focus gene list as input to build relevant networks , includes nodes with incoherent data values ( see Figure 3A–C ) , which reduces confidence in the relevance of the generated networks to the data . The network of 124 genes retrieved by IPA using a cutoff of >1 . 5 ( 60 focus genes ) contains many genes with reduced expression values ( Figure 3C ) , which were included in the network by the virtue of their connectivities but not data values . Consequently , the resulting network is not entirely representative of upregulated network processes in response to doxorubicin . Moreover , none of the networks identified by IPA contain all the genes involved in arginine-proline metabolism ( compare Figures 2D and 3C ) or any genes involved in the nucleotide metabolism that were retrieved by NetWalk ( see cluster in Figure 2D containing RRM2B , AK1 , POLR2A and NME2; compare with Figure 3C ) , demonstrating inability of seed-based methods to identify subnetworks with more subtle yet coherent gene expression values . As stated earlier , an important feature of NetWalk is that the result is not a single or a collection of static networks , but a whole distribution of numerical edge scores . In addition to their use for dynamical network construction of different sizes based on the user preference , these can be further subjected to standard statistical tests for a more detailed analysis . The heatmap of interactions with highest and lowest EF scores in each condition in our microarray dataset is shown in Figure 4A . As opposed to clustering with traditional heatmaps of gene expression values where cluster membership of genes is exclusive , here , a gene can appear in several different clusters but all with different interactions . So , analysis of expression with EF scores enables studying specific functions ( i . e . interactions ) of genes rather than their individual expression values . The heatmap shows that the activation and/or inactivation of several networks is specific to low- or high-dose doxorubicin treatment . The cluster K3 , for example , is activated in response to high-dose doxorubicin , while K4 is more specifically activated in response low-dose doxorubicin . A plot of interactions in K3 reveals several metabolic pathways specifically activated in the high-dose treatment , including glycolysis , acetyl coenzyme A synthesis , arginine/proline metabolism and the mitochondrial electron transport chain ( Figure 4C ) . There is also a p53-centered subnetwork containing several previously identified p53 target genes . The plot of interactions in K4 shows an extensive p53-centered network composed mostly of cell cycle regulatory proteins ( e . g . CDKN1A ( p21CIP ) and several GADD45 genes ) ( Figure 4B ) . Interestingly , although p53 appears in both K3 and K4 , its functions seem to be completely different in the low and high dose treatments . In response to low-dose doxorubicin , p53 is involved in the activation of cell cycle regulatory proteins , while under high-dose , it activates other targets , such as TMSB4X . Moreover , p53-target genes in cell cycle regulation in K3 are inactivated in high-dose doxorubicin ( Figure 5A–B ) , which we confirmed by western blotting ( Figure 5C ) , suggesting that p53 may act as a transcriptional activator of these genes during cell cycle arrest but as a repressor during apoptosis . This trend suggests not only that p53 may engage different targets during cell cycle arrest and apoptosis , but also shows dual behavior of p53 under these conditions . In addition , this analysis shows that energy and amino acid metabolisms may play an important role in doxorubicin-induced cell death . Here , clustering analysis using NetWalk results facilitated comparison of networks , rather than genes , between different conditions , leading to the identification of differential activities of p53 under low and high-dose doxorubicin treatment .
Analyses of high content data within the context of biological interactions allow for high confidence hypothesis generation about mechanisms involved in the studied process . While some work has been done on inferring novel causal interactions out of data [32]–[34] , the most popular method is integration of data with prior knowledge on interactions to extract most relevant networks highlighted by the data . Most of the methods for extracting relevant networks rely on finding genes in the network that are most central to connecting the genes of interest identified from the data . The random walk process in NetWalk also scores most central genes in the network . However , rather than working on a small set of focus genes , NetWalk scores centralities of all genes in the network based on the whole data distribution . This is achieved by biasing the random walk transition probabilities between genes to their corresponding data values , which allows for higher visitation probabilities of nodes with high data values and lower probabilities of nodes with low data values . Since visitation probabilities of nodes in a random walk are also dependent on the visitation probabilities of their network neighbors , nodes with relatively moderate data values associated with those with higher values have the potential of high visitation by the random walk . Therefore , NetWalk scores nodes based on their data values , data values of their neighbors and local network connectivity . Unlike most of the existing methods for network extraction , which typically give a set of networks as outputs [1] , [9] , NetWalk gives a distribution of EF values that allows for flexibility in network construction using different EF cutoffs . In addition , EF scores can be subjected to further statistical tests for comparative studies , allowing for network-based comparisons of multiple conditions . Another important feature of NetWalk is its computational efficiency . We implemented a sparse matrix representation and multiplication , which allows for NetWalk to be run on a standard PC equipped with 1 gigabytes of memory . In our case ( PC with Intel Xeon Quad processor ) , NetWalk run of a single dataset in our network ( 14 , 506 nodes and ∼190 , 000 interactions ) took about 2–3 seconds . NetWalk analysis of the experimental data revealed a significant activation of networks involved in energy metabolism , including the glycolytic and mitochondrial electron transport chain components . At least one member of the electron transport chain , SCO2A , has been previously shown to be a p53 target [35] , suggesting that some , if not most , of the metabolic genes activated in response to 10 uM doxorubicin may be p53 target genes . A specific and extensive activation of the energy metabolism during p53-mediated apoptosis has not been previously reported , and therefore it is a novel finding facilitated by NetWalk analysis . Network analysis of experimental data using NetWalk revealed dual behavior of p53 under sublethal and lethal doses of DNA damage . In response to sublethal doses of DNA damaging agents , p53 activates a cell cycle arrest program centered around CDK inhibitors p21 ( CDKN1A ) and GADD45 , as well as several pro-apoptotic genes , such as BAX and APAF1 . However under lethal doses , p53 represses the cell cycle arrest machinery and activates an entirely different program . Use of NetWalk analysis allows network based analysis of genomic data as well as high confidence hypothesis generation and is a valuable tool in post-genomic anlaysis .
|
Analysis of high-content genomic data within the context of known networks of interactions of genes can lead to a better understanding of the underlying biological processes . However , finding the networks of interactions that are most relevant to the given data is a challenging task . We present a random walk-based algorithm , NetWalk , which integrates genomic data with networks of interactions between genes to score the relevance of each interaction based on both the data values of the genes as well as their local network connectivity . This results in a distribution of Edge Flux values , which can be used for dynamic reconstruction of user-defined networks . Edge Flux values can be further subjected to statistical analyses such as clustering , allowing for direct numerical comparisons of context-specific networks between different conditions . To test NetWalk performance , we carried out microarray gene expression analysis of MCF7 cells subjected to lethal and sublethal doses of a DNA damaging agent . We compared NetWalk to other network-based analysis methods and found that NetWalk was superior in identifying coherently altered sub-networks from the genomic data . Using NetWalk , we further identified p53-regulated networks that are differentially involved in cell cycle arrest and apoptosis , which we experimentally tested .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"computational",
"biology/systems",
"biology",
"computational",
"biology/genomics"
] |
2010
|
Use of Data-Biased Random Walks on Graphs for the Retrieval of Context-Specific Networks from Genomic Data
|
The Kaposi sarcoma associated herpesvirus ( KSHV ) latency associated nuclear antigen ( LANA ) is expressed in all KSHV associated malignancies and is essential for maintenance of KSHV genomes in infected cells . To identify kinases that are potentially capable of modifying LANA , in vitro phosphorylation assays were performed using an Epstein Barr virus plus LANA protein microarray and 268 human kinases purified in active form from yeast . Interestingly , of the Epstein-Barr virus proteins on the array , the EBNA1 protein had the most similar kinase profile to LANA . We focused on nuclear kinases and on the N-terminus of LANA ( amino acids 1–329 ) that contains the LANA chromatin binding domain . Sixty-three nuclear kinases phosphorylated the LANA N-terminus . Twenty-four nuclear kinases phosphorylated a peptide covering the LANA chromatin binding domain ( amino acids 3–21 ) . Alanine mutations of serine 10 and threonine 14 abolish or severely diminish chromatin and histone binding by LANA . However , conversion of these residues to the phosphomimetic glutamic acid restored histone binding suggesting that phosphorylation of serine 10 and threonine 14 may modulate LANA function . Serine 10 and threonine 14 were validated as substrates of casein kinase 1 , PIM1 , GSK-3 and RSK3 kinases . Short-term treatment of transfected cells with inhibitors of these kinases found that only RSK inhibition reduced LANA interaction with endogenous histone H2B . Extended treatment of PEL cell cultures with RSK inhibitor caused a decrease in LANA protein levels associated with p21 induction and a loss of PEL cell viability . The data indicate that RSK phosphorylation affects both LANA accumulation and function .
The Kaposi sarcoma associated herpesvirus ( KSHV ) LANA protein is essential for establishment of KSHV latency through its role in replicating the KSHV genome , tethering the episomal genomes to cell chromosomes , interfering with induction of the viral lytic program and creating an environment that is permissive for cell survival and proliferation . Deletion of LANA in KSHV or rhesus rhadinovirus results in a more actively replicating virus [1] , [2] and this outcome derives in part from loss of LANA mediated repression of the lytic RTA transactivator [3]–[6] . LANA promotes cell survival through induction of components of the Notch pathway [7] , [8] , by limiting p53 mediated cell death [9]–[11] and through inhibition of TGF-beta signaling [12] . LANA promotes cell growth by stabilizing beta catenin [13] , deregulating c-Myc [14] , [15] , upregulating survivin and Id-1 expression [16] , [17] and E2F transcriptional activity [18] , [19] and modifying miRNA [20] and cell gene expression [21] . The effects on cell gene expression are due , in part , to LANA mediated de novo promoter methylation [22] and LANA interaction with a variety of transcription factors [14] , [15] , [23]–[31] . LANA serves as the origin binding protein for KSHV latency DNA replication and binds to sequences within the terminal repeats [32]–[34] to support latent DNA replication [35]–[37] and episomal DNA persistence [38] , [39] . LANA appears as nuclear speckles in KSHV infected cell nuclei . This speckling pattern requires the presence of KSHV DNA and in the absence of viral genomes LANA displays a nuclear diffuse staining pattern . LANA links KSHV episomes to host cell chromosomes and maintenance of the KSHV episomes in replicating cells is dependent on this LANA interaction [40] . LANA interaction with histones H2A and H2B through the N-terminal chromatin binding domain is critical for LANA association with chromosomes [41] , [42] . However , both N-terminal and C-terminal regions of LANA bind to chromatin [43]–[45] and LANA also interacts with other chromosome associated proteins such as MeCP2 , Brd4 , DEK , HP-1 alpha and CENP-F [18] , [44] , [46]–[50] . The LANA primary amino acid sequence includes 120 serine , threonine and tyrosine residues that could be subject to post-translational modification . The kinases glycogen synthase kinase 3 , PIM1/3 , ERK1/2 and DNA-PK , [51]–[55] have been shown to phosphorylate LANA and RSK1 has been shown to interact with LANA [55] . However , there has been no global analysis of kinases that are potentially capable of modifying LANA function through phosphorylation . We screened 268 human kinases for the ability to phosphorylate LANA in vitro using a protein array format that also included Epstein-Barr virus proteins . The presence of serine and threonine residues in the N-terminal LANA chromatin binding domain led us to a focus on this motif . The assays validated CSK1 , PIM1 , GSK-3 and RSK3 as kinases that phosphorylated the critical serine 10 and threonine 14 residues in the chromatin binding domain of LANA and RSK as a kinase family whose inhibition affected LANA interaction with histone H2B , LANA protein levels and PEL cell viability .
We have previously described a protein microarray displaying Epstein–Barr virus ( EBV ) proteins purified as GST-fusions from yeast and printed in duplicate [56] . This array was additionally printed with a series of EBV EBNA1 and KSHV LANA N-terminal and C-terminal polypetides printed either as 6xHis-GST fusions , V5-6xHis fusions or 6xHis-Biotin AviTag fusions . The array also contained control proteins that were used for orientation and normalization . This platform was used to globally identify human kinases that phosphorylate the KSHV LANA protein . Two hundred and sixty-eight human kinases were purified from yeast in active form as assayed by dot blot phosphorylation assays using a mixture of histone H3 , myelin basic protein and casein as the substrate . Phosphorylation assays were performed using kinase buffer containing [γ32P]-ATP and individual kinases . As a negative control , two chips per experiment were incubated with kinase buffer containing [γ32P]-ATP but minus the protein kinase . Phosphorylation signals were detected by exposing the arrays to X-ray film . An example of a kinase assay performed on the protein array is shown in Figure 1 . In the subsequent analyses performed with GenePix software , paired signals that were 3 standard deviations ( SD ) above background were considered positive . These assays identified 101 known or predicted nuclear kinases that phosphorylated KSHV LANA ( N+C ) . The EBV EBNA1 protein is the functional homolog of KSHV LANA although the two proteins have no significant amino acid homology . Interestingly , 99 of the kinases that phosphorylated KSHV LANA also phosphorylated EBV EBNA1 , a striking degree of overlap ( Table S1 ) . KSHV LANA and EBV EBNA1 were also phosphorylated by significantly more kinases than any of the other EBV proteins on the array ( Figure 2A ) . The median number of nuclear kinases phosphorylating the other EBV proteins on the array was 2 . The extensive phosphorylation of EBV EBNA1 and KSHV LANA was not due to a disproportionate representation of serine , threonine and tyrosine residues in these proteins relative to the other EBV encoded proteins ( Figure 2B ) . EBNA1 and LANA proteins contain 45 and 120 serine , threonine and tyrosine residues respectively . The median number for EBV encoded proteins is 62 . Amino acids 5–22 are important for KSHV association with the cell chromatin [41] , [42] , [57] . Within this domain are three serine residues and two threonine residues that could be modified by phosphorylation . Previous mutagenesis studies found that mutation of S10 to alanine individually or in the context of a triple mutation prevented chromatin binding , as did mutation of S13 to alanine in the context of a triple mutation . Mutation of T14 to alanine was +/− for chromatin binding , mutation of S22 did not affect binding and mutation of T19 was not examined [41] , [57] . Chromatin binding correlated with binding to histones H2A and H2B [41] . The S13A mutation has been examined only in the context of a triple alanine mutation . We consequently first examined the effect on binding to endogenous histone H2B of individual alanine mutations of LANA residues 5 through 14 , including S13 . In immunoprecipitation assays performed on transfected HEK 293T cells the observed loss of binding of Flag-LANA G5A , M6A , L8A , R9A , S10A , G11A , the retention of binding by R7A and the weak +/− binding of T14A recapitulated published results ( Figure 3A ) . As an individual mutation , conversion of serine 13 to alanine did not affect binding to histone H2B . The mutagenesis data implies that , if phosphorylation impacts on LANA's ability to bind histones , the relevant residues would be S10 and T14 . To assess the potential contribution of phosphorylation of the chromatin binding domain , LANA variants were generated that carried phosphomimetic mutations of S10 , S13 and T14 . The ability of these LANA variants to immunoprecipitate histone H2B was examined . Conversion of S10 to glutamic acid ( S10E ) restored the ability to bind to H2B both in the context of an individual S10E mutation and in the context of an S10E , S13E , T14E triple mutation ( Figure 3B ) . LANA carrying the T14E , S13E double mutation also showed increased histone H2B binding over that seen with T14A . Since mutation at S13 to a non-phosphorylatable alanine residue did not affect binding , this change can be attributed to T14E . These observations suggest that phosphorylation of S10 and T14 may impact on LANA function . The LANA N-terminus contains the chromatin binding domain and we therefore chose to focus on this region of LANA and to limit subsequent experiments to the 63 kinases that phosphorylated both the 6xHis-GST and 6xHis-Biotin AviTag fusions of the N-terminus of LANA ( aa 1–329 ) and are known or predicted to function in the cell nucleus . LANA has a nuclear localization and this subset of kinases is therefore more likely to be biologically relevant ( Table 1 ) . Included in this list are the PIM1 and GSK-3 kinases that are known to phosphorylate LANA [51] , [53] , [54] . We generated GST-fusion proteins of the N-terminal 50 amino acids of LANA along with derivatives in which serine or threonine residues within the chromatin binding domain ( S10 , S13 , T14 ) were mutated to alanine individually or as a triple mutation ( Figure 4A ) . The purified GST-fusion proteins used in the kinase assays are shown in Figure 4B . In vitro kinase assays were performed with 42 of the kinases that phosphorylated LANA ( 1–329 ) on the protein array and 31 of these kinases phosphorylated the N-terminal 50 amino acids of LANA ( Table 1 ) . Examples of the kinase assay using these GST-LANA ( 1–50 ) proteins as substrates are shown in Figure 4C . MAPK14/p38α ( mitogen activated protein kinase 14 ) phosphorylated GST-LANA ( 1–50 ) , each of the S10A , S13A and T14A mutants and the S10A , S13A , T14A triple mutant . MAPK14 either phosphorylates LANA 1–50 outside of the chromatin binding domain or alternatively , if S10 , S13 or T14 are sites , then MAPK4 must also be capable of phosphorylating the N-terminus outside of this domain . CSNK1G2 ( casein kinase 1 , gamma 2 isoform ) phosphorylated GST-LANA ( 1–50 ) and the S13A and T14A mutants but not the S10A or S10A , S13A , T14A mutants . This suggests that S10 within the chromatin binding domain is phosphorylated by casein kinase 1 . The possibility that phosphorylation of sites outside of the chromatin binding domain could complicate interpretation of the kinase assays using GST-LANA ( 1–50 ) as substrate led us to turn to peptide substrates that covered only the LANA chromatin binding domain . A synthetic peptide representing LANA amino acids 3–21 and covering the S10 and T14 amino acids of interest was used to retest 31 of the nuclear kinases that were positive for phosphorylation of GST-LANA ( 1–50 ) . In this assay , 24 of these kinases also phosphorylated the amino acid 3–21 LANA peptide ( Table 1 ) . Examples of the peptide phosphorylation assay are shown in Figure 5A . We next performed in vitro phosphorylation assays using synthetic peptides that carried individual mutations in S10A , S13A or T14A or combined mutations of S10A , S13A , T14A or S10A , S13A , T14A , T19A ( Figure 5B ) . In this set of assays we identified CSNK1G2 , PIM1 and RSK3 as kinases that phosphorylated S10 or T14 within the chromatin binding domain . CSNK1G2 phosphorylated the wild-type peptide and S13A and T14A mutant peptides but not the S10A or S10A , S13A , T14A or S10A , S13A , T14A , T19A peptides ( Figure 5B ) . This is consistent with the data obtained using GST-LANA ( 1–50 ) which also implicated casein kinase 1 in phosphorylation of S10 . PIM1 phosphorylated the wild-type , S10A and S13A peptides but not the T14 , S10A , S13A , T14A or S10A , S13A , T14A , T19A peptides indicating that T14 is a site for PIM1 phosphorylation ( Figure 5B ) . This observation differs from a previous study in which PIM1 phosphorylation of GST-LANA was linked specifically to S205 and S206 [51] . RSK3 ( RPS6KA2 ) phosphorylated the wild-type , S10A , S13A and T14A peptides but not the S10 , S13 , T14 or S10A , S13A , T14A , T19A peptides ( Figure 5B ) . This is consistent with RSK3 phosphorylation of two or more of the S10 , S13 and T14 residues . MAPK14/p38alpha phosphorylated all the peptides except the S10A , S13A , T14A , T19A peptide indicating MAPK14 phosphorylation at T19 ( Figure 5B ) . GSK-3 usually requires a priming phosphorylation by another kinase at the +4 position . The activity of GSK-3 on the LANA proteins purified from yeast implies that some of the protein was being purified in a pre-primed state . However , the synthetic peptide substrate was not phosphorylated by GSK-3 ( Figure 5C ) . Pre-incubation of the peptide with PIM1 in the presence of unlabelled γ-ATP followed by incubation with GSK-3β in the presence of [γ32P]-ATP and a PIM1 inhibitor , SMI-4a , resulted in phosphorylation of the wild-type peptide but not of the S10 mutant peptide ( Figure 5C ) . Thus PIM1 phosphorylation at T14 is able to prime for GSK-3β phosphorylation at S10 . To examine the effect of kinase inhibition on LANA interaction with histone H2B , 293 cells were transfected with Flag-LANA and treated with the CK1 inhibitor CKI-7 , the PIM1 inhibitor SMI-4a , the GSK-3 inhibitor LiCl , the RSK inhibitor BRD 7389 or with DMSO carrier . Short-term ( 4 hr ) treatment with CKI-7 at 30 and 100 µM had no effect on LANA interaction with histone H2B ( Figure 6A ) . Inhibition of GSK-3 or PIM1 , individually or in combination , by treatment for 6 hr with LiCl and SMI-4a also had no impact on LANA binding to H2B ( Figure 6B ) . However , treatment with 1 . 7 or 3 . 4 µM BRD 7389 for 6 hr decreased LANA interaction with histones in a dose responsive manner while having little effect on LANA protein levels ( Figure 6C ) . The effect of a longer exposure to RSK inhibitor was examined firstly in transfected 293 cells . Interestingly , when the cells were treated with 3 . 4 µM BRD 7389 for 24 hr versus 6 hr , there was not only a further decrease in LANA binding to histone H2B but also a decrease in LANA protein levels ( Figure 7A ) . A comparison of LANA protein levels in cells transfected with wt Flag-LANA or the phosphomimetic Flag-LANA [S10E , S13E , T14E] and treated with RSK inhibitor for 24 hr revealed that the decrease in LANA protein levels that occurred with wt LANA was not seen with the triple phosphomimetic mutant ( Figure 7B ) . This result links LANA stability to phosphorylation of the S10 , S13 and T14 residues in the chromatin binding domain and to chromatin binding . Exposure of BC3 and BCBL1 PEL cells to 0 . 85 , 1 . 7 and 3 . 4 µM concentrations of BRD 7389 for 48 hr resulted in decreased LANA protein levels in each case ( Figure 7C ) . To determine whether the observed loss of LANA was mediated at the level of protein turnover , LANA protein levels were examined in BCBL1 cells treated with 1 . 7 µM BRD 7389 for 24 hr with or without the addition of the proteosome inhibitor lactacystin for 6 hr prior to harvesting . LANA levels decreased with BRD 7389 treatment as expected , and were partially restored by proteosome inhibition ( Figure 7D , lanes 3 and 4 ) . To strengthen the conclusion that the loss of LANA was mediated at the post-transcriptional level , rather than via transcription , BC3 and BCBL1 PEL cells were treated with 0 . 85 µM BRD 7389 for 1 , 2 or 3 days and harvested cells were examined for LANA protein levels by western blotting ( Figures 8A and 8B ) and for LANA transcription using RT-PCR ( Figure 8C ) . BC3 and BCBL1 cells differed in the kinetics of LANA protein loss over the 3 day time course . However , at each time point there was a greater decrease in the LANA∶Actin protein ratio than in the LANA∶Actin transcript ratio which remained between 76–100% of that in untreated cells . The data are compatible with RSK inhibition affecting LANA protein stability or turnover . Loss of LANA would be expected to impact on LANA regulated events . LANA is known to bind to p53 and to block p53 mediated responses and release of this inhibition leads to induction of p53 target genes [58] . p53 transcriptionally activates p21 , a cyclin-CDK inhibitor that regulates the G1 phase of cell cycle progression . RT-PCR analysis of p21 mRNA levels was performed on the same samples of BC3 and BCBL1 used to measure LANA expression ( Figure 8D ) . Just as the kinetics of LANA protein loss differed between BC3 and BCBL1 cells , the kinetics of p21 induction also differed . The greatest induction of p21 transcripts occurred at 3 days post treatment , a time when LANA protein levels were lowest . Thus loss of LANA protein can be correlated with a downstream change in a LANA mediated function . To determine the effect of the loss of LANA protein on PEL cell growth , duplicate cultures of BC3 PEL cells and KSHV negative BJAB cells were treated with RSK inhibitor and each sample was measured in duplicate for cell viability using the CellTiter-Glo assay . RSK inhibition has been reported to reversibly inhibit proliferation of tumor derived cell lines and indeed all three inhibitor concentrations stopped the growth of both BC3 PEL cells and BJAB B cells ( Figure 9A ) . While , the BJAB cultures maintained their original cell numbers , the BC3 cultures showed a decrease in viable cell numbers over time suggesting that drug treatment was both cytostatic and cytotoxic for BC3 cells . The IC50 for BRD 7389 was 0 . 27 µM for BC3 PEL cells and 1 . 88 µM for BJAB B cells . To further evaluate the difference in loss of viability between inhibitor treated PEL cells and BJAB B cells , BC3 , BCBL1 and BJAB cells were treated with 0 . 85 µM BRD 7389 for 1 day and a western blot was performed to detect PARP cleavage as a measure of caspase induction and apoptosis . The antibody used is specific for cleaved PARP . BRD 7389 treatment induced PARP cleavage that was substantially greater in the PEL cells than in BJAB cells ( Figure 9B ) . The assay was also repeated on cells treated with 0 . 4 and 1 . 7 µM BRD 7389 for 1 day and in this case the western blot was probed with antibody that detects both cleaved and uncleaved forms of PARP ( Figure 9C ) . Again PARP cleavage was readily detectable in BCBL1 and BC3 cells and was minimal in BJAB cells .
Protein microarrays and high throughput techniques for kinase expression and purification have previously been used to characterize the yeast kinome and analyze yeast phosphorylation networks [59] , [60] . We now describe the use of these technologies to examine phosphorylation of the KSHV LANA protein by human kinases . The ability to screen LANA with 268 different human kinases provided a unique opportunity to examine the potential role of phosphorylation of the chromatin binding domain on LANA function and to identify those kinases able to perform this function . In vitro kinase screens have limitations . Not all of the kinases that phosphorylate the substrate in vitro will have the appropriate intracellular localization , expression pattern or activating signals to perform this function in the cellular environment while kinases that require specialized reaction conditions or partners such as priming kinases for maximum activity will tend to be under represented . It should be noted that substrates purified from yeast exhibit a limited degree of phosphorylation that can partially support the priming activity required by kinases such as GSK-3 . To ensure that we focused on the most biologically relevant LANA partners , kinases that phosphorylated LANA in the initial microarray screen were subjected to database searches for their intracellular localization and only those with a known or predicted nuclear activity were included in subsequent assays . An interesting observation from the initial microarray assay was that the number of kinases phosphorylating LANA and its functional homolog in latency viral DNA replication in EBV , the EBNA1 protein , was an order of magnitude greater than the number detected for most other EBV proteins on the array . This suggests that LANA and EBNA1 functions may be particularly sensitive to regulation by phosphorylation . The extensive overlap in phosphorylating kinases for these two proteins also suggests that they may be subject to regulation by common signaling pathways . Within the LANA chromatin binding domain , three of the amino acids that had been implicated as critical for histone and chromatin binding in mutagenesis studies were S10 , S13 and T14 . We tested S13 as a single mutation , S13A , and did not find any impact on LANA binding to histones . This placed the emphasis on S10 and T14 as residues potentially subject to post-translational modification . The ability of phosphomimetic S10E and T14E mutations to restore LANA binding to histone H2B is suggestive of a role for phosphorylation in the regulation of LANA chromatin binding function . Two of the kinases that we identified as phosphorylating these residues , PIM and RSK have been implicated in KSHV lytic replication . Overexpression of PIM1 and PIM3 led to reactivation of KSHV from latency and the role of these kinases was linked to phosphorylation of LANA at serines 205 and 206 , an event that abolished LANA mediated repression of RTA [53] . Treatment of PEL cells with the RSK inhibitor BI-D1870 or treatment with inhibitors of ERK signaling also inhibited KSHV lytic replication [61] , [62] . The latency and lytic cycles of herpesviruses are usually thought to have opposing requirements . Proteins or pathways that favor latency generally have an inhibitory effect on lytic replication and vice versa . However , a requirement for the activity of certain kinases in both aspects of the KSHV life-cycle need not be contradictory . For example , ERK signaling is also known to be required for establishment of a KSHV infection [63] , [64] . ERK signaling , which activates RSK , can have directly opposing effects depending on the duration and strength of the signal and cross-talk with other signaling pathways [65] . We saw no effect on LANA interaction with histone H2B after a short treatment with CK1 , PIM1 or GSK-3 inhibitors . However , a short treatment with the RSK inhibitor BRD 7389 decreased LANA binding to histone H2B . All of the RSK family members , RSK1 , 2 , 3 and 4 are inhibited by BRD 7389 . Interestingly , a longer BRD 7389 treatment resulted in a decrease in LANA protein levels in LANA transfected cells and in PEL cells . Treatment of cancer cell lines with RSK inhibitors results in growth inhibition [66] , [67] . We observed this same response upon treatment of the BJAB B cell lymphoma cell line and PEL cells . The treated BC3 PEL cells not only stopped growing but also decreased in viable cell numbers suggesting that inhibition of RSK was having an additional impact on these cells . PEL cells are generally wild-type for p53 ( BCBL1 cells are heterozygous [11] ) and BJAB B cells are p53 mutant . One of the functions of LANA is to block p53 mediated apoptosis through interaction with the p53-MdM2 complex [10] , [68]–[70] and through inactivation of GSK-3 [13] , [58] , [71] and interaction with angiogenin [72] . In RSK inhibitor treated PEL cells , p21 induction by activated p53 occurred later in the time course than induction of caspase cleavage suggesting that the loss of viability of treated PEL cells likely occurred by a p53 independent mechanism . Overall , our data provide evidence that phosphorylation affects LANA interaction with chromatin and that inhibition of the RSK kinases that phosphorylate the chromatin binding domain has an additional effect on LANA protein levels . The dual outcomes of reduced histone binding and loss of LANA protein stability imply that chromatin binding protects LANA from degradation and contributes to the long half-life demonstrated by LANA in latently infected cells . ERK signaling positively regulates transcription of LANA upon KSHV infection of endothelial cells [73] but the loss of LANA in inhibitor treated cells was mediated at the level of protein turnover rather than transcriptionally . The Ras/Raf/MEK/ERK signaling pathway that leads to RSK activation regulates cell growth , proliferation and survival . The sensitivity of PEL cells to treatment with inhibitors of this pathway may represent a vulnerability that can be exploited to limit the growth of latently KSHV infected cells .
EBV proteins were purified as 6x-His-GST fusion proteins from yeast using a high-throughput protein purification protocol described previously [74] . Briefly , cultures of 65 yeast strains that express EBV proteins as 6xHis-GST fusions plus yeast expressing 6xHis-GST-LANA ( 1–329 ) and 6xHis-GST-LANA ( 891–1129 ) proteins were grown , harvested , lysed , and the fusion proteins purified using glutathione beads . After extensive washes , the captured GST-fusion proteins were eluted in elution buffer ( 50 mM HEPES [pH 7 . 4] containing 100 mM NaCl , 40 mM reduced glutathione , 0 . 03% Triton X-100 , and 30% glycerol ) . 6xHis-N-LANA-Avi-tag protein and V5-6xHis EBNA1 ( 392–641 ) were purified using the PrepEase His Tagged Protein Purification Kit ( USB ) . The eluate was collected through a filter unit ( NUNC ) and stored in 384-well plates . Protein products that were successfully purified based on immunoblot analysis were then spotted in duplicate onto microscope slides ( Full Moon Biosystems ) using a 48-pin contact printer ( Bio-Rad ) . The quality and quantity of the immobilized proteins on these chips was monitored using anti-GST antibody followed by Cy-5 labeled secondary antibody . Human kinases from 268 recombinant yeast strains were purified as 6xHis-GST-fusion proteins using a previously described high-throughput protein purification protocol [74] , [75] . The captured GST-fusion proteins were eluted in elution buffer ( 50 mM HEPES [pH 7 . 4] containing 100 mM NaCl , 40 mM reduced glutathione , 0 . 03% Triton X-100 , and 30% glycerol ) . The eluate was collected through a filter unit ( NUNC ) and stored at −80°C . Successful purification of human kinases free of contaminating kinase activity from yeast was determined by immunoblot , silver staining and autoradiographic analyses [Newman et al , submitted for publication] . To evaluate the enzymatic activity of each sample , auto- and trans-phosphorylation reactions were performed using a standard liquid-kinase assay with [γ32P]-ATP and a mixture of histone H3 , myelin basic protein ( MBP ) , and casein . To identify potential LANA phosphorylating kinases , a protocol similar to that described by Zhu et al . was used [56] . Briefly , the protein microarrays were blocked in 1× TBST with 1% bovine serum albumen ( BSA ) for 1 hr at room temperature with gentle shaking . The arrays were then incubated with individual kinases in reaction buffer ( 25 mM Hepes at pH 7 . 5 with 100 mM NaCl , 50 mM Tris-HCl pH 7 . 5 , 55 nM [γ32P]-ATP , 1 mM DTT , 10 mM MgCl2 , 1 mM MnCl2 , 1 mM EGTA , 1 mM NaVO4 , 1 mM NaF , and 0 . 1% NP-40 ) for 30 minutes at 30°C . Control slides were incubated with kinase reaction mixture without kinase and processed in parallel . The arrays were subjected to three ten minute washes in , firstly , TBS , 0 . 1% Tween-20 ( TBST ) and secondly in 0 . 5% SDS . Arrays were then rinsed briefly with double-distilled H2O and dried by centrifugation . Each slide was then exposed to BioMax high resolution X-ray film ( Kodak ) for 16 hours . Finally , developed films were scanned using an office scanner at a resolution of 4 , 800 dpi . Captured images from the autoradiographic films were processed using Photoshop . The substrate profile for each image was acquired using GenePix software ( Axon ) . Each protein was printed in duplicate and a protein was scored as a positive hit only if both spots of the same protein showed signal intensity higher than the cutoff value of three standard deviations above the mean . Negative control experiments identified proteins that underwent autophosphorylation on the protein microarrays . The Flag-LANA plasmid pMF24 that is deleted for the central repeat region was cleaved with XmaI and AscI and repaired with oligonucleotide insertions to introduce G5A , M6A , R7A , L8A , R9A , S10A , G11A , R12A , S13A or T14A mutations ( pMS30 to pMS39 respectively ) . Glutamic acid substitutions were introduced by site-directed mutagenesis on plasmid DY52 ( FLAG-LANA ) to generate S10E ( pGL621 ) , S13E , T14E ( pGL628 ) , and S10E , S13E , T14E ( pGL629 ) . These constructions were then used as PCR templates to generate GST-LANA ( aa1–329 ) wt , S10A , S13A and T14A plasmids ( pGL485 to pGL488 respectively ) and GST-LANA ( aa1–50 ) wt , S10A , S13A and T14A plasmids ( pGL502 to pGL505 respectively ) . The GST-LANA ( aa1–50 ) triple mutant S10A/S13A/T14A ( pGL515 ) was generated using XmaI and AscI cleavage and repair with an oligonucleotide insertion carrying mutated sequences . Full length 3xFlag-LANA carrying the triple mutations S10A/S13A/T14A ( pGL628 ) and S10E/S13E/T14E ( pGL629 ) were generated by PCR mutagenesis in the vector p3xFlag-CMV ( Sigma ) . Full length Flag-LANA ( pDY52 ) has been described previously [76] . 6xHis-LANA ( aa1–329 ) -Biotin Avi-Tag ( pGL370 ) and 6xHis-LANA ( aa936–1162 ) -Biotin Avi-Tag ( pGL371 ) were generated in the bacterial expression vector PAC4 ( GeneCopoeia ) . 6xHis-GST-EBNA1 ( 386–641 ) , 6xHis-GST-EBNA1 ( 1–87 ) and EBNA1 ( 392–641-V5-6xHis; pGL451D ) have been described [56] , 6xHis-GST-LANA ( 1–329 ) was generated in the same vector background . LANA amino acid 3–21 polypeptides and variants carrying S/T to A mutations were purchased from Peptide 2 . 0 . The purified GST-LANA fusion proteins ( 0 . 1 µg ) or synthetic polypeptides ( 250 µM ) were incubated with kinase in 25 µl kinase buffer containing 33 . 3 nM [γ32P]-ATP for 30 min at 30°C . Reactions were terminated by heating the mixture at 90°C for 5 min , and the proteins were separated on NuPAGE gels ( 4 to 12% Bis-Tris; Invitrogen or 4 to 20% Tris-HCl; Biorad ) . The gels were dried and exposed to MP Hyperfilm ( GE Healthcare ) . For characterization of GSK-3β phosphorylation , peptide substrates were incubated for 30 min at 30°C in ( i ) kinase buffer with 10 µCi of [γ32P]-ATP , ( ii ) kinase buffer with 10 µCi of [γ32P]-ATP and Pim1 ( Upstate , #14-573 ) , ( iii ) kinase buffer with 10 µCi of [γ32P]-ATP and GSK-3β ( Millipore ) and for primed phosphorylations ( iv ) kinase buffer with 20 µM cold ATP and Pim1 ( Upstate , #14-573 ) followed by incubation with Pim1 inhibitor SMI-4a ( 0 . 1 µM; Enzo Life Sciences ) for 30 min at 30°C ( Millipore ) . Samples were then reincubated in kinase buffer with 10 µCi of [γ32P]-ATP or reincubated for 30 min at 30°C in ( v ) kinase buffer with 10 µCi of [γ32P]-ATP and GSK-3β . Samples were separated by SDS-PAGE ( 16 . 5% Tris-Tricine; BioRad ) and analyzed by autoradiography . HEK 293T cells grown in 10-cm dishes were transfected with 10 ug of total DNA using calcium phosphate precipitation . 48 h after transfection , cells were lysed in 1 ml of lysis buffer ( 50 mM Tris ( pH 7 . 9 ) , 100 mM NaCl , 0 . 5 mM EDTA , 2% glycerol , 0 . 2% NP-40 ) , plus protease inhibitors ( 0 . 5 mM PMSF , 2 ug/ml Aprotinin , and 1 ug/ul leupeptin ) and phosphatase inhibitors ( Phosphatase Inhibitor Cocktail 1&2; Sigma ) , sonicated for 10 s , and cleared by centrifugation . Extracts were precleared using protein A/G PLUS-agarose ( Santa Cruz Biotechnology , Inc . ) and immunoprecipitated with anti-FLAG M2-agarose ( Sigma ) . Beads were washed six times with precipitation buffer and bound H2B was detected by western blotting using rabbit anti-H2B antibody ( Abcam ) . Flag-LANA was detected using rabbit anti-Flag antibody ( Sigma ) or rat anti-LANA antibody ( ABI Advanced Biotechnologies ) . The effect of kinase inhibition on H2B interaction was examined using the inhibitors CKI-7 ( Sigma Aldrich ) , LiCl ( J . T . Baker ) , SMI-4a ( Enzo Life Sciences ) and BRD 7389 ( Tocris Bioscience ) . Cells were treated with the proteasome inhibitor lactacystin ( Peptide Institute Inc ) for 6 hr at a final concentration of 1 µM . PARP cleavage was detected by western blotting using cleaved PARP ( ASP214 ) specific and total PARP antibodies ( Cell Signaling ) . LANA and Actin were detected by rat anti-LANA ( Advanced Biotechnologies Inc ) and mouse anti-Actin ( Sigma ) . BC3 and BCBL1 PEL cells and BJAB B cells were cultured in RPMI 1640 plus 15% fetal bovine serum in 5% CO2 at 37°C . Cell growth in cultures treated with BRD 7389 was measured using the Cell Titer-Glo luminescence assay kit ( Promega ) and luciferase activity was quantified using a Glomax Multi Detection System ( Promega ) .
|
The Kaposi sarcoma associated herpesvirus ( KSHV ) is associated with cancers that have an increased incidence in individuals with compromised immune systems . KSHV expresses a protein , LANA , that is needed to maintain KSHV genomes in infected cells and also promotes the growth of KSHV associated tumors . Kinases regulate protein function through phosphorylation . To identify kinases that may affect LANA function , we performed a screen in which 268 human kinases were isolated and tested for the ability to phosphorylate LANA in vitro . We focused on the region of LANA that contains the chromatin binding domain , a motif essential for tethering KSHV genomes to the cell chromatin and maintaining latent infection . We identified serine 10 and threonine 14 as amino acids within the chromatin binding domain whose phosphorylation was important for histone binding . Serine 10 and threonine 14 were targets of the CK1 , PIM1 , GSK-3 and RSK3 kinases . Treatment with an inhibitor of RSK kinase reduced LANA binding to histones , decreased LANA protein levels and caused a loss of KSHV infected PEL cell viability . Our experiments show that phosphorylation affects LANA function and suggest that KSHV infected cells may be particularly vulnerable to kinase inhibitors .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"virology",
"viruses",
"and",
"cancer",
"biology",
"microbiology"
] |
2012
|
Phosphorylation of the Chromatin Binding Domain of KSHV LANA
|
Acanthamoeba castellanii , which causes keratitis and blindness in under-resourced countries , is an emerging pathogen worldwide , because of its association with contact lens use . The wall makes cysts resistant to sterilizing reagents in lens solutions and to antibiotics applied to the eye . Transmission electron microscopy and structured illumination microscopy ( SIM ) showed purified cyst walls of A . castellanii retained an outer ectocyst layer , an inner endocyst layer , and conical ostioles that connect them . Mass spectrometry showed candidate cyst wall proteins were dominated by three families of lectins ( named here Jonah , Luke , and Leo ) , which bound well to cellulose and less well to chitin . An abundant Jonah lectin , which has one choice-of-anchor A ( CAA ) domain , was made early during encystation and localized to the ectocyst layer of cyst walls . An abundant Luke lectin , which has two carbohydrate-binding modules ( CBM49 ) , outlined small , flat ostioles in a single-layered primordial wall and localized to the endocyst layer and ostioles of mature walls . An abundant Leo lectin , which has two unique domains with eight Cys residues each ( 8-Cys ) , localized to the endocyst layer and ostioles . The Jonah lectin and glycopolymers , to which it binds , were accessible in the ectocyst layer . In contrast , Luke and Leo lectins and the glycopolymers , to which they bind , were mostly inaccessible in the endocyst layer and ostioles . The most abundant A . castellanii cyst wall proteins are three sets of lectins , which have carbohydrate-binding modules that are conserved ( CBM49s of Luke ) , newly characterized ( CAA of Jonah ) , or unique to Acanthamoebae ( 8-Cys of Leo ) . Cyst wall formation is a tightly choreographed event , in which lectins and glycopolymers combine to form a mature wall with a protected endocyst layer . Because of its accessibility in the ectocyst layer , an abundant Jonah lectin is an excellent diagnostic target .
Acanthamoebae , which include the genome project A . castellanii Neff strain , are soil protists named for acanthopods ( spikes ) on the surface of trophozoites [1] . In immunocompetent persons , Acanthamoeba is a rare but important cause of corneal inflammation ( keratitis ) , which is difficult to treat and so may lead to scarring and blindness [2–4] . In immunosuppressed patients , Acanthamoeba may cause encephalitis [5] . Acanthamoeba is endemic in under-resourced populations in the Middle East , South Asia , Africa , and Latin America [6–11] . Acanthamoeba is an emerging pathogen in Europe , North America , and Australia , where 80 to 90% of infections are associated with contact lens use [12–14] . Because water for washing hands may be scarce in places where the free-living protist is frequent , we recently showed that alcohols in concentrations present in hand sanitizers kill A . castellanii trophozoites and cysts [15 , 16] . When A . castellanii trophozoites are deprived of nutrients in solution or on agar plates , they form cysts [17–19] . Transmission electron microscopy ( TEM ) shows cyst walls have two microfibril-dense layers ( outer ectocyst and inner endocyst ) , which are separated by a relatively microfibril-free layer [20] . The endocyst and ectocyst layers are connected to each other by conical ostioles , through which the protist escapes during excystation [21] . The cyst wall of A . castellanii protects free-living protists from osmotic shock when exposed to fresh water , drying when exposed to air , or starvation when deprived of bacteria or other food sources . The cyst wall also acts as a barrier , sheltering parasites from killing by disinfectants used to clean surfaces , sterilizing agents in contact lens solutions , and/or antibiotics applied directly to the eye [22–24] . We are interested in the cyst wall proteins of A . castellanii for three reasons . First , although monoclonal antibodies to A . castellanii have been made , the majority react to trophozoites , and no cyst wall proteins have been molecularly identified [25–27] . Indeed the only cyst-specific protein identified , which was named for its 21-kDa predicted size ( CSP21 ) , is unlikely to be a cyst wall protein , as it lacks a signal peptide [28] . A cyst wall protein that is unique , abundant , accessible , and conserved across many strains of Acanthamoebae would therefore be an excellent target for a new diagnostic antibody . Second , A . castellanii and related species are the only human pathogens that contain cellulose in their wall [29–31] . Dictyostelium discoideum , which also has cellulose in its walls , is not a significant pathogen [32] . In addition , the whole genome of A . castellanii predicts a set of candidate cyst wall proteins that contain two or three carbohydrate-binding modules ( CBM49s ) , which are homologs of a C-terminal cellulose-binding domain ( SlCBM49 ) of the Solanum lycopersicum ( tomato ) endocellulase SlGH9C [1 , 33–36] . Further , the genome predicts a chitin ( a polymer of β-1 , 4-linked GlcNAc ) synthase , a chitinase , and two chitin deacetylases , suggesting the possibility that chitin and chitin-binding proteins are also present in the cyst wall [37] . Note , however , that monosaccharide analysis of cyst wall glycopolymers revealed β-1 , 4-linked glucose and galactose rather than GlcNAc [31] . Third , we are interested in whether abundant cyst wall proteins localize to particular structures in the mature wall: ectocyst layer , endocyst layer , and ostioles . If so , the location of these proteins at numerous time points during encystation might provide insights into how the cyst wall is assembled . Our experimental design was relatively simple . We used TEM , as well as structured illumination microscopy ( SIM ) and probes for glycopolymers , to judge the intactness and cleanliness of purified A . castellanii cyst walls [38] . We used mass spectrometry to identify candidate cyst wall proteins , which were compared to proteins present in walls of other protists , bacteria , fungi , and plants [39] . We used SIM to localize abundant cyst wall proteins , each of which was tagged with a green fluorescent protein ( GFP ) and expressed under its own promoter , in encysting protists and in mature cysts [40 , 41] . We also determined whether each cyst wall protein , expressed as a GFP-tagged protein under a constitutive glyceraldehyde 3-phosphate dehydrogenase ( GAPDH ) promoter in trophozoites or as a maltose-binding protein ( MBP ) in the periplasm of Escherichia coli , binds to microcrystalline cellulose and/or chitin beads [42] . Finally , we used anti-GFP antibodies and MBP-cyst wall protein fusions to test the accessibility of proteins and glycopolymers , respectively , in the ectocyst and endocyst layers of mature walls . In this way , we began to answer five basic questions concerning A . castellanii cyst wall proteins: What are their identities ? When are they made ? Where are cyst wall proteins located in the developing and mature cyst wall ? Why are they located there ? Which cyst wall protein is the best target for diagnostic antibodies ?
Culture and manipulation of A . castellanii were approved by the Boston University Institutional Biosafety Committee . A . castellanii Neff strain trophozoites were purchased from the American Type Culture collection . Trophozoites of A . castellanii MEEI 0184 strain , which was derived from a human corneal infection , were obtained from Dr . Noorjahan Panjwani of Tufts University School of Medicine [16] . Neff strain organisms were used for all experiments with the exception of a few initial mass spectrometric studies . Trophozoites were grown in T-75 tissue culture flasks at 30°C in 10 ml ATCC medium 712 ( PYG plus additives ) ( Sigma-Aldrich Corporation , St . Louis , MO ) [18] . Flasks containing log-phase trophozoites ( free of cysts that form spontaneously in stationary cultures ) were either chilled or scraped with a cell scraper to release adherent amoebae , which were concentrated by centrifugation at 500 x g for 5 min and washed twice with phosphate buffered saline ( PBS ) . Approximately 107 amoebae obtained from a confluent flask were induced to encyst by incubation at 30°C on agar plates containing non-nutrient medium , which contained 2% agarose [16] . After 3 , 6 , 12 , 15 , 18 , 24 , 36 , 72 , or 144 hr incubation , 15 ml of PBS was added to agar plates , which were incubated on a shaker for 30 min at room temperature ( RT ) . Encysting organisms were removed using a cell scraper and concentrated by centrifugation for 10 min at 500 x g for <24 hr cysts or at 1 , 500 x g for >24 hr cysts . Nearly 100% of the organisms formed cysts . Between 5 and 10 million mature cysts ( after 144 hr encystation ) were washed in PBS and suspended in lysis buffer ( 10 mM HEPES , 25 mM KCl , 1 mM CaCl2 , 10 mM MgCl2 , 2% CHAPS , and 1X Roche protease inhibitor ) ( Sigma-Aldrich ) . For SIM , cysts in 500-μl lysis buffer were broken four times for 2 min each with 200 μl of 0 . 5 mm glass beads in a Mini-Beadbeater-16 ( BioSpec Products , Bartlesville , OK ) . For TEM , where glass beads cannot be used , cysts in 200-μl lysis buffer were broken by sonication four times for 20 seconds each in continuous mode in a Sonicator Cell Disruptor ( formerly Heat Systems Ultrasonic , now Qsonica , Newtown , CT ) . Broken cysts were added to the top a 15-ml falcon tube containing 60% sucrose in ddH2O and centrifuged at 4 , 000 x g for 10 min . Bead beating breaks 95 to 100% of cysts . The broken cyst wall pellet , which contained zero to 5% cysts , was suspended in PBS buffer and washed three times at 10 , 000 x g in a microcentrifuge . The cyst wall pellet was used without further modification for SIM or TEM . For mass spectrometry , the cyst wall pellet was placed at the top of gradient containing 2 ml each of 20% , 40% , 60% and 80% Percoll ( top to bottom ) , which was buffered with PBS , and centrifuged for 20 min at 3 , 000 x g . The layer between 60% and 80% Percoll , where the broken cyst walls were located , was collected and washed in PBS . The cyst wall preparation was suspended in 10 ml of PBS , placed in a syringe , and forced through a 25-mm diameter Whatman Nuclepore Track-Etched Membrane with 8-μm holes ( Sigma-Aldrich ) . The cellular debris , which passed through the membranes , was discarded . The membrane was removed from the cassette , suspended in 5 ml of PBS , and vortexed to release cyst walls . The membrane was removed , and cyst walls were distributed in microfuge tubes and pelleted at 15 , 000 x g for 10 min . The pellet was suspended in 50 μl PBS and stored at -20°C prior to trypsin digestion and mass spectrometry analysis . A GST-AcCBM49 fusion-construct , which contains the N-terminal CBM49 of an abundant Luke ( 2 ) lectin minus the signal peptide , was prepared by codon optimization ( 76 to 330-bp coding region of ACA1_377670 ) ( S1 Fig and S1 Excel file ) ( GenScript , Piscataway , NJ ) . It was cloned into pGEX-6p-1 ( GE Healthcare Life Sciences , Marlborough , MA ) for cytoplasmic expression in BL21 ( DE3 ) chemically competent E . coli ( Thermo Fisher Scientific , Waltham , MA ) [43] . Expression of GST-AcCBM49 and GST were induced with 1 mM IPTG for 4 hr at RT , and GST-fusions were purified on glutathione-agarose and conjugated to Alexa Fluor 594 succinimidyl esters ( red ) ( Molecular Probes , Thermo Fisher Scientific ) . Approximately 106 mature cysts or cyst walls were washed in PBS and fixed in 1% paraformaldehyde buffered with 0 . 2 M phosphate , pH 7 . 5 , for 15 min at RT . Pellets were washed two times with Hank’s Buffered Saline Solution ( HBSS ) and incubated with HBSS containing 1% bovine serum albumin ( BSA ) for 1 hour at RT . Preparations were then incubated for 2 hr at 4°C with 10 μl of 0 . 25 μg/μl GST-CBM49 conjugated to Alexa Fluor 594 and 20 μl of 0 . 625 μg/μl wheat germ agglutinin ( WGA ) ( Vector Laboratories , Burlingame , CA ) conjugated to Alexa Fluor 488 in 100 μl HBSS [44 , 45] . Finally , pellets were labeled with 100 μg of calcofluor white M2R ( CFW ) ( Sigma-Aldrich ) in 100 μl HBSS for 15 min at RT and washed five times with HBSS [46 , 47] . Preparations were mounted in Mowiol mounting medium ( Sigma-Aldrich ) and observed with widefield and differential interference contrast microscopy , using a 100x objective of a Zeiss AXIO inverted microscope with a Colibri LED ( Carl Zeiss Microcopy LLC , Thornwood , NY ) . Images were collected at 0 . 2-μm optical sections with a Hamamatsu Orca-R2 camera and deconvolved using ZEN software ( Zeiss ) . Alternatively , SIM was performed with a 63-x objective of a Zeiss ELYRA S . 1 microscope at Boston College ( Chestnut Hill , MA ) , and 0 . 09-μm optical sections deconvolved using Zen software [38] . All SIM images shown were 3D reconstructions using dozens of z-stacks . High-pressure freezing and freeze substitution were used to prepare cysts and cyst walls for TEM at the Harvard Medical School Electron Microscope facility [48] . To make them noninfectious , we fixed mature cysts in 1% paraformaldehyde in PBS for 10 min at RT and washed them two times in PBS . Cyst walls in PBS were pelleted , placed in 6-mm Cu/Au carriers , and frozen in an EM ICE high-pressure freezer ( Leica Microsystems , Buffalo Grove , Il ) . Freeze substitution was performed in a Leica EM AFS2 instrument in dry acetone containing 1% ddH20 , 1% OsO4 , and 1% glutaraldehyde at -90°C for 48 hr . The temperature was increased 5°C/hour to 20°C , and samples were washed 3 times in pure acetone and once in propylene oxide for 10 min each . Samples were infiltrated with 1:1 Epon:propylene oxide overnight at 4°C and embedded in TAAB Epon ( Marivac Canada Inc . St . Laurent , Canada ) . Ultrathin sections ( 80 to 100 nm thick ) were cut on a Leica Reichert Ultracut S microtome , picked up onto copper grids , stained with lead citrate , and examined in a JEOL 1200EX transmission electron microscope ( JEOL USA , Peabody , MA ) . Images were recorded with an AMT 2k CCD camera . Approximately 10 million broken cyst walls , prepared as above , were dissolved into 50 mM NH4HCO3 , pH 8 . 0 , reduced with 10 mM dithiothreithol ( DTT ) for 20 min at 60°C , alkylated with 55 mM iodoacetamide ( IAA ) for 20 min at RT , and then digested with proteomics grade trypsin ( Sigma-Aldrich ) overnight at 37°C . Alternatively broken cyst walls either before or after digestion with trypsin were reconstituted in 1× reducing SDS/PAGE loading buffer and run on a 4–20% precast polyacrylamide TGX gel ( Bio-Rad Laboratories , Philadelphia , PA ) . Bands stained by colloidal Coomassie blue were excised and washed with 50 mM NH4HCO3/acetonitrile ( ACN ) . Reduction , alkylation , and trypsin/chymotrypsin digestion were performed in-gel . Peptides were dried and desalted using C18 ZipTip concentrators ( MilliporeSigma , Burlington , MA ) . Peptides from five biological replicates for both in solution and in-situ hydrolyses were dissolved in 2% ACN , 0 . 1% formic acid ( FA ) and separated using a nanoAcquity-UPLC system ( Waters Corporation , Milford , MA ) equipped with a 5-μm Symmetry C18 trap column ( 180 μm x 20 mm ) and a 1 . 7-μm BEH130 C18 analytical column ( 150 μm × 100 mm ) . Samples were loaded onto the precolumn and washed for 4 min at a flow rate of 4 μl/min with 100% mobile phase A ( 99% water/1% ACN/0 . 1% FA ) . Samples were eluted to the analytical column with a gradient of 2–40% mobile phase B ( 99% ACN/1% water/0 . 1% FA ) delivered over 40 or 90 min at a flow rate of 0 . 5 μl/min . The analytical column was connected online to a QE or a QE-HF Mass Spectrometer ( Thermo Fisher Scientific ) equipped with a Triversa NanoMate ( Advion Inc . , Ithaca , NY ) electrospray ionization ( ESI ) source , which was operated at 1 . 7 kV . Data were acquired in automatic Data Dependent top 10 ( QE ) or top 20 ( QE-HF ) mode . Automated database searches were performed using the PEAKS software suite version 8 . 5 ( Bioinformatics Solutions Inc . , Waterloo , ON , Canada ) . The predicted proteins of Acanthamoeba castellanii Neff strain ( AmoebaDB-33June 30 , 2017 ) was used to predict tryptic peptides for mass spectrometric analyses and was used for bioinformatics analyses ( see below ) [36] . The search criteria were set as follows: trypsin/chymotrypsin as the enzyme with ≤ 3 missed cleavages and ≤ 1 non-specific cleavage , the error tolerances for the precursor of 5 ppm and 0 . 05 Da for fragment ions , carbamidomethyl cysteine as a fixed modification , oxidation of methionine , Pyro-glu from glutamine , and deamidation of asparagine or glutamine as variable modifications . The peptide match threshold ( -10 logP ) was set to 15 , and only proteins with a minimum of two unique peptides were considered . The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium ( http://proteomecentral . proteomexchange . org ) via the PRIDE partner repository with the dataset identifier PXD011826 [49] . Signal peptides and transmembrane helices were predicted using SignalP 4 . 1 and TMHMM , respectively [50 , 51] . Glycosylphosphatidylinositol anchors were searched using big-PI [52] . AmoebaDB , which contains sequence information from the Neff strain and ten other Acanthamoeba strains , was used to identify genome sequences , predict introns , and identify paralogous proteins [35 , 36] . The NR database at the NCBI was used to identify homologs of candidate cyst wall proteins in other species and to identify conserved domains [53] . Carbohydrate-binding modules were searched using CAZy and InterPro databases [34 , 54 , 55] . We used RT-PCR from RNA of encysting protists to obtain the coding sequences of an abundant Luke ( 2 ) lectin ( 840-bp CDS of ACA1_377670 ) , Leo lectin ( 562-bp CDS of ACA1_074730 ) , and Jonah ( 1 ) lectin ( 1596-bp CDS of ACA1_164810 ) . An abundant Luke ( 3 ) lectin ( 1293-bp CDS of ACA1_245650 ) did not contain any introns and so was obtained from genomic DNA . Please see S1 Excel file for a list of primers used to make all the constructs . Using NEBuilder HiFi DNA assembly ( New England Biolabs , Ipswich , MA ) , we cloned each CDS into the pGAPDH plasmid , which was a kind gift from Yeonchul Hong of Kyongpook National University School of Medicine , Deagu , Korea [41] . pGAPDH contains a neomycin resistance gene under a TATA-box promoter ( for selection with G418 ) and a glyceraldehyde 3-phosphate dehydrogenase promoter for constitutive expression of GFP-fusions ( S1 Fig ) . The GFP tag was placed at the C-terminus of each cyst wall protein , and a polyadenylation sequence was added downstream of the GFP-fusion’s stop codon . For expression of cyst wall protein genes under their own promoters , we replaced the GAPDH promoter with 446-bp from the 5 ‘UTR of the Luke ( 2 ) gene , 500-bp from the 5’ UTR of the Luke ( 3 ) gene , 486-bp from the 5’ UTR of the Leo gene , and 571-bp of the 5’UTR of the Jonah ( 1 ) gene , each cloned from the genomic DNA . As controls , GFP alone and SP-GFP , which contains a 60-bp sequence encoding an N-terminal signal peptide of Luke ( 2 ) lectin , were expressed under a GAPDH promoter . As another control , the 470-bp 5’ UTR and 525-bp CDS of CSP21 ( ACA1_075240 ) was made with a GFP tag [28] . Transfections in A . castellanii were performed as described previously [40 , 41] with some modifications . Briefly , 5 x 105 log-phase trophozoites were allowed to adhere to 6-well plates in ATCC medium 712 for 30 min at 30°C . The adherent trophozoites were washed and replaced with 500 μl of non-nutrient medium ( 20 mM Tris-HCl [pH 8 . 8] , 100 mM KCl , 8 mM MgSO4 , 0 . 4 mM CaCl2 and 1 mM NaHCO3 ) . In an Eppendorf tube , 4 μg of Midiprep ( PureLink HiPure Midiprep Kit , Thermo Fisher Scientific ) plasmid DNA was diluted to 100 μl with non-nutrient medium . Twenty microliters of SuperFect Transfection Reagent ( Qiagen Inc , Germantown , MD ) was added to the DNA suspension , mixed gently by pipetting five times , and incubated for 10 min at RT . Six hundred microliters of non-nutrient medium were added to the DNA-SuperFect mix , and the entire suspension was added to the trophozoites adhering to the 6-well culture plate . The culture plate was incubated for 3 hr at 30°C , after which the non-nutrient medium was replaced with ATCC medium 712 and incubated for another 24 hr at 30°C . To select for transfectants , we added 12 . 5 μg/ml of Gibco G418 antibiotic ( Thermo Fisher Scientific ) to the culture after 24 hr , and we changed the medium plus antibiotic every four days . After 2 to 4 weeks , the transfectants were growing robustly in the presence of the antibiotic , and trophozoites and/or cysts expressing GFP were detected by widefield microscopy . Without prior cloning , transfectants were induced to encyst , fixed after 3 to 144 hr , labeled with WGA and CFW , and examined by widefield microscopy and SIM , as described above . MBP-fusion constructs were prepared by cloning the cDNA of an abundant Luke ( 2 ) lectin ( 60 to 843-bp CDS of ACA1_377670 ) and an abundant Jonah ( 1 ) lectin ( 70 to 1599-bp CDS of ACA1_164810 ) without their signal sequences into pMAL-p2x vector ( New England Biolabs ) ( S1 Excel file ) for periplasmic expression in BL21-CodonPlus ( DE3 ) -RIPL ( Agilent Technologies , Lexington , MA ) [42] . For the MBP-fusion , the Leo CDS without the signal sequence ( 67 to 564-bp of ACA1_074730 ) was codon optimized and cloned into pMAL-p2x vector ( S1 Excel file ) . MBP-Luke ( 2 ) was induced with 250 μM IPTG for 5 hr at RT; MBP-Jonah ( 1 ) was induced with 1 mM IPTG for 5 hr at RT; and MBP-Leo was induced with 250 μM IPTG for 3 . 5 hr at 37°C . MBP-fusion proteins were purified with amylose resin following the manufacturer’s instructions ( GE Healthcare , Pierce , Agilent Technologies , and New England Biolabs ) . MBP-fusions ( 1 μg each in 100 μl of 1% NP40 ) were incubated with 0 . 5 μg Avicel microcrystalline cellulose ( Sigma-Aldrich ) or a 50-μl slurry of magnetic chitin beads ( New England Biolabs ) for 3 hr at 4°C with rocking . Cellulose was centrifuged to collect the supernatant ( unbound fraction ) and pellet ( bound fraction ) , while chitin beads were collected with a magnet . The bound fractions were washed three times with 1% NP40 . To solubilize proteins , the input material ( total ) , unbound ( U ) , and bound ( B ) fractions were boiled in SDS sample buffer . MBP-proteins were separated on SDS-PAGE gels , blotted to PVDF membranes , blocked in 5% BSA , and detected using anti-MBP antibodies ( New England Biolabs ) . To test the carbohydrate-binding specificity of the GFP-tagged lectins , we lysed trophozoites expressing Jonah ( 1 ) and Luke ( 2 ) under a GAPDH promoter and then incubated lysates with microcrystalline cellulose or chitin beads , using methods to characterize MBP-fusions . Total , unbound , and bound proteins were released with SDS , separated on SDS-PAGE , transferred to PVDF , and detected with reagents that recognize GFP . A control was GFP alone expressed under a GAPDH promoter . Log-phase trophozoites and 36-hr-old cysts were harvested , and the total protein solubilized in SDS sample buffer , run in SDS-PAGE gels , blotted on PVDF membranes , and blocked in 5% BSA . MBP-cyst wall protein fusions and MBP alone were run in adjacent lanes as positive and negative controls , respectively . The blots were probed with 1:100 dilutions of rabbit polyclonal antibodies ( Li International , Denver , Co ) raised to 16- or 50-amino acid peptides of abundant Luke ( 2 ) lectin ( residues 230–279 of ACA1_377670 ) , Leo lectin ( residues 124–139 of ACA1_074730 ) and Jonah ( 1 ) lectin ( residues 362–411 of ACA1_164810 ) . A 1:1000 dilution of anti-rabbit IgG-HRP ( BioRad ) was used as secondary antibody and Super Signal West Pico PLUS ( Thermo Fisher Scientific ) for chemiluminescent detection . Coomassie stained gels were run in parallel for loading control . We used anti-GFP antibodies to determine the accessibility of GFP-tagged lectins in mature cyst walls . Without prior fixation , mature cysts expressing GFP-fusions under their own promoter were blocked with 1% BSA , incubated with 1:400 mouse anti-GFP IgG ( Roche ) for one hr at RT , washed , and then incubated with 1:800 goat anti-mouse IgG-Alexa Fluor 594 ( Molecular Probes , Invitrogen ) . Preparations were washed , labeled with WGA and CFW , fixed in paraformaldehyde , mounted on glass slides , and observed with widefield microscopy , as described above . To determine the accessibility of glycopolymers in mature cyst walls , we used MBP-fusions to Luke ( 2 ) , Leo , and Jonah ( 1 ) lectins . Mature cysts were fixed , blocked , and incubated with 15 μg of each MBP-cyst wall protein fusion conjugated to Alex Fluor 594 for 2 hr at 4°C . Preparations were labeled with WGA conjugated to Alexa Fluor 488 and CFW , as described above , and visualized with widefield microscopy and SIM . To count the number of ostioles per cyst wall , we rotated three-dimensional SIM reconstructions of mature cysts expressing Luke ( 2 ) -GFP or Leo-GFP or non-transfectants labeled with WGA , MBP-Luke ( 2 ) , or MBP-Leo , all of which clearly outlined conical ostioles .
Cyst wall preparations were made by subjecting mature cysts to sonication ( for TEM ) or bead beating ( for SIM ) , followed by density centrifugation to remove cellular contents . For TEM , mature cysts and purified cyst walls were frozen under high pressure , and fixatives were infiltrated at low temperature [48] . Purified cyst walls had intact ectocyst and endocyst layers , as well as conical ostioles that link them ( Fig 1 ) [20] . The purified walls were missing amorphous material that fills the space between the inner aspect of the cyst wall and the plasma membrane of the trophozoite inside . For SIM , we used probes that bind chitin ( WGA ) and β-1 , 3 and β-1 , 4 polysaccharides ( CFW ) in the walls of fungi and cysts of Entamoeba [44–46] . CFW , a fluorescent brightener , has also been used to diagnose Acanthamoeba cysts in eye infections [47] . In addition , we made a glutathione-S-transferase ( GST ) fusion-protein , which contains the N-terminal CBM49 of a candidate cyst wall protein of A . castellanii ( S1 Fig and S1 Excel file ) [43] . The GST-AcCBM49 expression construct was designed to replicate that used to determine the carbohydrate binding properties of SlCBM49 , the C-terminal carbohydrate-binding module of the S . lycopersicum cellulase SlGH9C [33] . In both mature cysts and purified cyst walls , GST-AcCBM49 predominantly labeled the ectocyst layer , WGA highlighted the ostioles , and CFW labeled the endocyst layer ( Fig 2 ) . A detailed examination of both the mature cyst and the purified wall showed WGA also labeled the endocyst layer and the ectocyst layer ( weakly ) . In summary , TEM and SIM both showed that ectocyst and endocyst layers , as well as ostioles , were intact in purified cyst walls , which were relatively free of cellular material . While GST-CBM49 , WGA , and CFW , as well as abundant cyst wall proteins ( see below ) , were extremely useful for distinguishing structures in the developing and mature cyst walls , their lack of carbohydrate-binding specificity ( again see below ) made it impossible to distinguish whether ectocyst layer , endocyst layer , and ostioles were composed of cellulose , chitin , or both glycopolymers . Indeed , nowhere in this paper have we shown that chitin or chitosan are present in cyst walls of A . castellanii . Trypsin treatment of purified A . castellanii cyst walls , which was followed by LC-MS/MS of the released peptides , gave similar results in five biological experiments ( Table 1 and S2 Excel file ) . While some proteins remained in cyst walls after trypsin digestion , their identities were similar to those detected in the soluble fractions by in gel-digests with trypsin or chymotrypsin . Candidate cyst wall proteins with the most unique peptides identified by LC-MS/MS belonged to three families , which we named Luke , Leo , and Jonah lectins , because each bound to cellulose +/- chitin ( see below ) . Although it was impossible to draw a line that separates actual cyst wall proteins from contaminating proteins , secreted proteins with 18+ unique peptides included six Leo lectins , four Luke lectins , and three Jonah lectins . The vast majority of proteins with <18 unique peptides were predicted to be cytosolic ( including CSP21 ) and so were likely intracellular contaminants of cyst wall preparations . The exception to this hypothesis , we think , were additional Luke , Leo , and Jonah lectins , which were most likely less abundant cyst wall proteins . For readers interested in cytosolic proteins of A . castellanii , we have added S3 Excel file , which contains all the mass spectrometry data , which included a “dirty” cyst wall preparation that was generated without using the Percoll gradient or porous filter . Luke lectins were comprised of an N-terminal signal peptide , followed by two or three CBM49s that were separated by Ser- and Pro-rich spacers ( Fig 3 and S2 Fig ) [33 , 34 , 50] . The N-terminal CBM49 of Luke lectins contained three conserved Trp resides conserved in SlCBM49 from tomato . A fourth conserved Trp is present in the CBM49 of D . discoideum cellulose-binding proteins [56] . The other CBM49s ( middle and/or C-terminal ) of Luke lectins had two conserved Trp residues . Luke lectins were acidic ( pI 5 to 6 ) and had formula weights ( FWs ) from 27 to 29-kDa ( two CBM49s ) or 42 to 44-kDa ( three CBM49s ) . There were no predicted transmembrane helices or glycosylphosphatidylinositol anchors in the Luke or Leo lectins [51 , 52] . LC-MS/MS of the released cell wall peptides identified at least one unique peptide corresponding to all 12 genes encoding Luke lectins , although the number of unique peptides varied from 78 to two ( Table 1 and S2 Excel file ) . In general , Luke lectins with two CBM49s had more unique peptides than Luke lectins with three CBM49s . One to four unique peptides were derived from three CBM49-metalloprotease fusion-proteins , which consisted of an N-terminal signal peptide followed by a single CBM49 with four conserved Trp residues and a metalloprotease ( ADAM/reprolysin subtype ) with a conserved catalytic domain ( HEIGHNLGGNH ) [53] . We used an abundant Luke ( 2 ) lectin ( ACA1_377670 ) with two CBM49s to perform RT-PCR , make rabbit anti-peptide antibodies , and make maltose-binding protein ( MBP ) - and green fluorescent protein ( GFP ) -fusions ( Fig 3 and S1 Fig ) [40–42] . We also used an abundant Luke ( 3 ) lectin ( ACA1_245650 ) with three CBM49s to make a GFP-fusion ( S2 Fig ) . Leo lectins were comprised of an N-terminal signal peptide , followed by two repeats of a unique 8-Cys domain , some of which were separated by a long Thr- , Lys- , and His-rich spacer ( Fig 3 and S2 Fig ) . Leo lectins without a spacer were acidic ( pI ~4 . 8 ) and had FWs from 19 to 24-kDa , while Leo lectins with the TKH-rich spacer were basic ( pI ~8 . 3 ) and had FWs from 36- to 59-kDa . Leo lectins were encoded by 16 genes , of which 14 proteins were identified by our LC-MS/MS analysis . While the number of unique peptides varied from 34 to one , Leo lectins without a spacer generally had more unique peptides than Leo lectins with the TKH-rich spacer . We used abundant Leo lectin without a spacer ( ACA1_074730 ) to perform RT-PCR , make rabbit anti-peptide antibodies , and make MBP- and GFP-fusions ( S1 Fig ) . Jonah lectins were comprised of an N-terminal signal peptide followed by one or three choice-of-anchor A ( CAA ) domains ( Fig 3 and S2 Fig ) [53] . The binding activity of the CAA domain , which is adjacent to a collagen-binding domain in a microbial surface component recognizing the adhesive matrix molecule ( MSRAMM ) of Bacillus anthracis , was not characterized [57] . Jonah ( 1 ) lectins with a single CAA domain were acidic ( pI ~6 ) , had a FW from 44 to 58-kDa and had an N-terminal Thr- , Lys- , and Cys-rich domain . A Jonah ( 3 ) lectin with three CAA domains was basic ( pI ~8 . 8 ) , had a FW of ~146-kDa , and contained Ser- and Pro-rich spacers between CAA domains , as well as hydrophobic regions that may be transmembrane helices [51] . Jonah lectins were encoded by eight genes , of which five were identified by our LC-MS/MS analysis based on one to 147 unique peptides . We used an abundant Jonah ( 1 ) lectin ( ACA1_164810 ) with a single CAA domain to perform RT-PCR , make rabbit anti-peptide antibodies , and make MBP- and GFP-fusions ( S1 Fig ) . Other secreted proteins with 18+ unique peptides detected by LC-MS/MS , which are candidate cyst wall proteins , included a laccase with three copper oxidase domains ( ACA1_068450 ) , a protein with a C-terminal ferritin-like domain ( ACA1_292810 ) , a Kazal-type serine protease inhibitor ( ACA1_291590 ) , a conserved uncharacterized protein ( ACA1_068630 ) , and a protein unique to A . castellanii ( ACA1_145900 ) [53 , 54 , 58] . Interestingly , a bacterial laccase has been shown to bind cellulose [59] . There were also three serine proteases , which have been localized to the secretory system of encysting A . castellanii [60] . These results suggested that the most abundant candidate cyst wall proteins of A . castellanii contain tandem repeats of conserved domains ( CBM49 in Luke lectins and CAA in Jonah lectins ) or a unique domain ( 8-Cys in Leo lectins ) . Peptides corresponding to nearly all members of each gene family were detected by mass spectrometry . However , the relative abundances of unique peptides for each cyst wall protein varied by more than an order of magnitude , suggesting marked differences in gene expression . Because it was not possible to separate cyst walls into component parts ( endocyst and ectocyst layers and ostioles ) prior to LC-MS/MS analysis of tryptic peptides , we used SIM and GFP-tags to localize abundant members of each protein family in cyst walls of transfected A . castellanii ( see below ) . Leo lectins , which had two domains with 8-Cys each , appeared to be unique to A . castellanii , as no homologs were identified when BLAST analysis were performed using the nonredundant ( NR ) database at NCBI ( https://www . ncbi . nlm . nih . gov/ ) [35] . The origin of genes encoding Luke lectins was difficult to infer , because its CBM49s showed only a 31% identity over a short ( 77-amino acid ) overlap with a predicted cellulose-binding protein of D . discoideum ( expect value of BLASTP was just 7e-05 ) [33 , 34 , 56] . In contrast , the CAA domain of Jonah lectins appeared to derive from bacteria by horizontal gene transfer ( HGT ) , as no other eukaryote contained CAA domains , and there was a 28% identity over a bigger ( 263-aa ) overlap with a choice-of anchor A family protein of Saccharibacillus sp . O16 ( 5e-12 ) [35 , 53] . The A . castellanii laccase ( also known as copper oxidase ) , whose signals were abundant in the mass spectra , was likely the product of HGT from bacteria , as there was a 44% identity over a large ( 526-aa ) overlap with a copper oxidase of Caldicobacteri oshimai ( 6e-135 ) [58] . The uncertainty was based upon the presence of similar enzymes in plants , one of which ( Ziziphus jujube ) showed a 39% identity over a 484-aa overlap ( 4e-101 ) with the A . castellanii laccase . No pairs of genes within each lectin family were syntenic as defined by AmoebaDB , indicating duplicated genes are paralogs [36] . With the exception of two Luke lectins ( ACA1_253500 and ACA1_253650 ) that were 98% identical and two Leo lectins ( ACA1_074770 and ACA1_083920 ) that were 85% identical , members of each family of cyst wall proteins differed in amino acid sequence by >40% . Genes that encode cyst wall proteins also varied in the number of introns ( zero to two in Luke , two to four in Leo , and zero to 24 in Jonah ) . Searches of genomic sequences of 11 strains of Acanthamoebae , deposited in AmoebaDB without protein predictions by Andrew Jackson of the University of Liverpool , using TBLASTN and sequences of abundant Luke , Leo , and Jonah lectins localized in the next section , showed four results [35 , 36] . First , although stop codons were difficult to identify using this method , all 11 strains appeared to encode each cyst wall protein . Second , most strains showed 100 to 200-amino acid stretches of each cyst wall protein that were 80 to 90% identical to the A . castellanii Neff strain studied here . These stretches did not include low complexity spacers , which were difficult to align . Third , some of the strains showed greater differences from the Neff strain in each cyst wall protein , consistent with previous descriptions of Acanthamoeba strain diversity based upon 18S rDNA sequences [61] . Fourth , while coding sequences and 5’ UTRs were well-conserved , intron sequences were very poorly conserved , with the exception of branch-point sequences . In summary , genes encoding Jonah lectins and laccase likely derived by HGT , while genes encoding Leo lectins appeared to originate within Acanthamoeba . Although CBM49s of Luke lectins shared common ancestry with plants and other Amoebazoa , their precise origin was not clear . For the most part , gene duplications that expanded each family within the Acanthamoeba genome occurred a long time ago , as shown by big differences in amino acid sequences of paralogous proteins and variations in the number and sequences of introns . Regardless , the set of Luke , Leo , and Jonah lectins identified by mass spectrometry , as well as the sequences of abundant cyst wall proteins localized in the next section , appeared to be conserved among 11 sequenced isolates of Acanthamoebae . To localize candidate cyst wall proteins , we expressed an abundant Leo lectin with no spacer and an abundant Jonah ( 1 ) lectin with a single CAA domain , each with a GFP-tag under its own promoter ( 446- and 571-bp of the 5’ UTR , respectively ) in transfected trophozoites of A . castellanii , using an episomal vector that was selected with G418 ( S1 Fig ) [40 , 41] . We also expressed an abundant Luke ( 2 ) lectin with two CBM49s and an abundant Luke ( 3 ) lectin with three CBM49s , each with a GFP-tag under its own promoter ( 486- and 500-bp of the 5’ UTR , respectively ) . GFP-tagged candidate cyst wall proteins expressed under their own promoter were absent in the vast majority of log-phase trophozoites , while GFP-tagged cyst wall proteins were present in small numbers in trophozoites in stationary cultures , where a few organisms began to encyst spontaneously . As early as three hours after placement on non-nutrient agar , Jonah ( 1 ) -GFP expressed under its own promoter was present in dozens of small vesicles ( Fig 4A ) . The glycopolymer detected with WGA was also made early and was present in vesicles of varying sizes , which did not overlap with those containing Jonah ( 1 ) -GFP . The glycopolymer labeled with GST-CBM49 was also made early in dozens of small vesicles , which were distinct from those labeled with WGA ( Fig 4B and 4C ) . Glycopolymers labeled with CFW were not visible in organisms encysting for 3 and 6 hr , but CFW labeled a thin , spherical wall after 9 hr encystation . At this time , rare protists had one or two small , flat ostioles , but most organisms had none . Finally , neither Luke ( 2 ) -GFP nor Leo-GFP , each expressed under its own promoter , was visible during this first stage of development of the cyst wall . These results showed that the first stage of encystation is an abrupt event in which amoeboid trophozoites rapidly synthesize glycopolymers and a Jonah ( 1 ) lectin in dozens of vesicles that fill the encysting cells . In contrast , Luke ( 2 ) and Leo lectins were not yet made , suggesting encystation-specific proteins are expressed at different times [28 , 62] . GST-CBM49 , WGA , and CFW each labeled primordial cyst walls , which had a single , thin layer and small , flat ostioles ( Fig 5A ) . Ostioles , which labeled with CFW but not with GST-CBM49 or WGA , were at first filled circles but later became rings ( Fig 5B ) . While it was difficult to count these small ostioles because of variable labeling with CFW , they appeared to be in similar number and distribution as conical ostioles of mature cyst walls ( see below ) . Each of the GFP-tagged lectins expressed under its own promoter was present in primordial cyst walls but in markedly different distributions . Jonah ( 1 ) -GFP was homogenously distributed across the surface of the primordial cyst wall ( Fig 5B ) . Luke ( 2 ) -GFP outlined some but not all of early ring-shaped ostioles ( Fig 5C ) . Later , in addition to outlining the ostioles , Luke ( 2 ) -GFP was homogenously distributed across the surface of the primordial cyst wall ( Fig 5D ) . Leo-GFP was latest to the wall and had a patchy distribution , which was , for the most part , independent of the ostioles ( Fig 5E ) . These results showed that in the second stage of encystation the primordial cyst walls contained three abundant lectins , each in a distinct distribution . The presence of small , circular ostioles , which were visualized by the external probe CFW or the internal probe Luke ( 2 ) -GFP , showed these structures are initiated prior to separation of the ectocyst and endocyst layers . In the third stage , the cell body contracted , so that the emerging endocyst layer was made inside the ectocyst layer ( Fig 6 ) . Glycopolymers labeled by CFW moved to the endocyst layer , which was labeled in a variable manner by WGA . Jonah ( 1 ) -GFP remained with the ectocyst layer and had essentially the same appearance in the walls of second and third stage cysts ( Figs 5B , 6A and 6D ) . Luke ( 2 ) -GFP was diffusely distributed in the endocyst layer and dome-shaped ostioles of organisms encysting for 24 and 36 hr ( Fig 6B and 6E ) . For the most part , Leo-GFP localized the endocyst layer of 24 hr cysts , although its distribution remained patchy ( Fig 6C ) . It was not until 36 hr encystation that Leo-GFP began to diffusely label the endocyst layer and outline ostioles ( Fig 6F ) . In summary , during the third stage of encystation , Jonah ( 1 ) -GFP remained in the outer layer of the wall , which is destined to become the ectocyst layer of mature cyst walls ( see next section ) . In contrast , Luke ( 2 ) -GFP and Leo-GFP moved to the inner layer of the wall , which will become the endocyst layer and ostioles of mature cyst walls . Jonah ( 1 ) -GFP expressed under its own promoter was present in the ectocyst layer of mature cyst walls ( ≥ 36 hr encystation ) , which were labeled with WGA and CFW ( Fig 7A ) . In contrast , Leo-GFP , Luke ( 2 ) -GFP , and Luke ( 3 ) -GFP , each expressed under its own promoter , were present in the endocyst layer and sharply outlined the ostioles ( Fig 7B to 7D ) . Jonah ( 1 ) -GFP expressed under a constitutive GAPDH promoter localized to the ectocyst layer of mature walls ( Fig 7E ) , while Luke ( 2 ) -GFP expressed under the GAPDH promoter localized to endocyst layer and ostioles of mature walls ( Fig 7F ) . Because Leo-GFP did not express well under the GAPDH promoter , it was not possible to compare its distribution versus Leo-GFP under its own promoter . These results suggested carbohydrate-binding specificities or protein-protein interactions were more important than timing of expression for localization of Jonah ( 1 ) and Luke ( 2 ) lectins . While the Jonah ( 1 ) lectin localized to the ectocyst layer , Luke and Leo lectins , which do not share common ancestry , both localized to the endocyst layer and ostioles . Finally , Luke lectins with either two or three CBM49s localized to the same place . Numerous control experiments suggested the timing of expression and locations of the GFP-tagged cyst wall proteins in cyst walls were accurate . First , RT-PCR showed that mRNAs of abundant Luke ( 2 ) , Leo , and Jonah ( 1 ) lectins , as well as cellulose synthase ( ACA1_349650 ) , were absent or nearly absent from trophozoites but were present during the first three days of encystation ( S3 Fig ) . In contrast , glyceraldehyde 3-phosphate dehydrogenase ( GAPDH ) , which catalyzes the sixth step in glycolysis , was expressed by both trophozoites and encysting A . castellanii [41] . Second , monospecific , polyclonal rabbit antibodies to a 50-amino acid peptide of an abundant Jonah ( 1 ) lectin and a 16-amino acid peptide of an abundant Leo lectin bound to Western blots of proteins from cysts but not from trophozoites ( S4 Fig ) . We were unable to generate rabbit antibodies to the Luke ( 2 ) lectin , using methods that worked to make antibodies to Jonah ( 1 ) and Leo lectins . Because the rabbit anti-peptide antibodies failed to recognize native proteins , none was useful for localizing cyst wall proteins by SIM . Third , GFP-tagged CSP21 expressed under its own promoter was present in cytosolic accumulations of mature cysts ( S5 Fig ) [28 , 62] . As CSP21 is homologous to universal stress proteins and lacks an N-terminal signal peptide , its presence in the cytosol after nutrient deprivation was expected [50 , 63] . Fourth , a GFP-fusion protein ( SP-GFP ) , which was appended with an N-terminal signal peptide from the Luke ( 2 ) lectin and expressed under a GAPDH promoter , localized to secretory vesicles of cysts but not to cyst walls ( S5 Fig ) . Fifth , GFP alone expressed under the GAPDH promoter was homogenously distributed in the cytosol of cysts ( S6 Fig ) . To test the binding of abundant cyst wall proteins to commercially available glycopolymers , we made MBP-cyst wall protein fusion-proteins in the periplasm ( secretory compartment ) of E . coli [42] . Previously , we used MBP-fusions to characterize carbohydrate-binding properties of cyst wall lectins of Entamoeba , Giardia , and Toxoplasma [64–66] . The targets were microcrystalline cellulose ( used to characterize binding activities of GST-SlCBM49 from tomato cellulase ) and chitin beads ( used to characterize myc-tagged Jacob and Jessie lectins of Entamoeba histolytica ) [33 , 67] . Western blots with anti-MBP antibodies showed MBP-Luke ( 2 ) and MBP-Jonah ( 1 ) each bound to microcrystalline cellulose and somewhat less well to chitin beads ( Fig 8 ) . MBP-Leo bound less completely to microcrystalline cellulose and weakly at best to chitin beads . MBP alone ( negative control ) did not bind to microcrystalline cellulose or chitin beads . As a control , we incubated with Luke ( 2 ) -GFP , Jonah ( 1 ) -GFP , and GFP alone , each obtained from lysates of trophozoites expressing the tagged proteins under a GAPDH promoter with cellulose and chitin . Consistent with the MBP-fusions , Luke ( 2 ) -GFP and Jonah ( 1 ) -GFP bound to cellulose and Luke ( 2 ) -GFP bound to chitin , while GFP alone bound to neither cellulose nor chitin . The one discrepant finding was that Jonah ( 1 ) -GFP failed to bind to chitin , to which MBP-Jonah ( 1 ) bound ( Fig 8 ) . These results showed abundant Luke ( 2 ) , Leo , and Jonah ( 1 ) lectins each bound cellulose well , while binding to chitin was much more variable . The binding patterns of tagged Luke , Leo , and Jonah lectins to cellulose and chitin in vitro , however , were poor predictors for localization of these proteins in mature cyst walls . To determine the accessibility of proteins in the ectocyst and endocyst layers and ostioles of mature cyst wall , we incubated organisms expressing GFP-tagged lectins under their own promoters with anti-GFP antibodies . Widefield microscopy showed that Jonah ( 1 ) -GFP was accessible in the endocyst layer of nearly 100% of mature cysts with a detectable Jonah ( 1 ) -GFP signal ( Fig 9A ) . In contrast , anti-GFP antibodies showed Luke ( 2 ) -GFP and Leo-GFP were accessible in the endocyst layer and ostioles of 3 and 2% , respectively , of mature cysts with detectable GFP signals ( Fig 9B and 9C ) . To determine the accessibility of glycopolymers in two layers of mature cyst walls , we labeled cysts with MBP-lectin fusion-proteins . MBP-Jonah ( 1 ) bound to the ectocyst layer of 100% of mature cell walls ( Fig 9D and S7 Fig ) , which was the same location as Jonah ( 1 ) -GFP expressed under either its own or the GAPDH promoter ( Fig 7A and 7E ) . In contrast , MBP-Luke ( 2 ) and MBP-Leo probes each labeled the endocyst layer and ostioles of 9% mature cyst walls ( Fig 9E and 9F and S7 Fig ) . Although these were the same places in mature cyst walls where Luke ( 2 ) -GFP and Leo-GFP localized under either their own promoters or the GAPDH promoter ( Luke ( 2 ) -GFP ) ( Fig 7B , 7C and 7F ) , these results suggested that glycopolymers bound by MBP-Luke ( 2 ) and MBP-Leo in the endocyst layer and ostioles were , for the most part , inaccessible to external probes . Finally , by rotating three-dimensional SIM reconstructions of organisms expressing Luke ( 2 ) -GFP or Leo-GFP or labeled with WGA , MBP-Luke ( 2 ) , or MBP-Leo , we counted an average of 8 . 8 +/- 2 . 5 ostioles per mature cyst wall ( 24 cysts total ) . To our knowledge , this is the first estimate of the number of ostioles in Acanthamoeba cyst walls , because ostioles have not previously been visualized by light microscopy and are extremely difficult to count by TEM [20] , unless dozens of serial sections are performed .
Although we expected Luke lectins with two or three CBM49s would be present in the cellulose-rich cyst wall , we could not have predicted the other abundant cyst wall proteins , because the 8-Cys domains of Leo lectins are unique to Acanthamoebae and the CAA domains of Jonah lectins were previously uncharacterized [33–35 , 53–57] . While Luke lectins have two or three CBM49s , D . discoideum has dozens of proteins with a single CBM49 ( S8 Fig ) . The Luke lectins bind cellulose and chitin , while the D . discoideum proteins with a single CBM49 bind cellulose [56] . Chitin-binding by DdCBM49 or SlCBM49 was not tested , because this glycopolymer is not present in D . discoideum and tomato walls . Demonstration that CBM49s of the Luke lectin also bind chitin fibrils is new , but is consistent with recent studies showing CBMs may bind more than one glycopolymer [55] . The metalloprotease fused to an N-terminal CBM49 of A . castellanii is absent in D . discoideum , while D . discoideum adds two CBM49s to a cysteine proteinase , which lacks these domains in A . castellanii . The CBM49 may act to localize the metalloproteases to the A . castellanii cyst wall , as is the case for the chitin-binding domain in Entamoeba Jessie lectins or glucan-binding domain in a Toxoplasma glucanase [64 , 66 , 67] . Alternatively , the CBM49 may suggest the metalloprotease cleaves glycopeptides rather than peptides . While the GH5 glycoside hydrolases of A . castellanii lack CBM49s , CBM49 is present at the C-terminus of GH9 glycoside hydrolases of D . discoideum and S . lycopersicum [33 , 34] . Even though A . castellanii Leo lectins and E . histolytica Jacob lectins share no common ancestry , they have 8-Cys and 6-Cys lectin domains , respectively , often separated by low complexity sequences ( S9 Fig ) [67 , 68] . E . histolytica low complexity sequences vary from strain to strain , contain cryptic sites for cysteine proteases , and are extensively decorated with O-phosphate-linked glycans [69] . We have not yet identified any Asn-linked or O-linked glycans on Leo lectins or any of the other cyst wall proteins , but we expect they will be there . A . castellanii and oomycetes ( Pyromyces and Neocallmistic ) each contain proteins with arrays of CAA domains , but the sequences of the CAAs are so different that it is likely that concatenation of domains occurred independently ( S10 Fig ) [53 , 54] . Although A . castellanii is exposed to collagen in the extracellular matrix of the cornea , the protist lacks a homolog of the collagen-binding domain that is adjacent to the CAA domain in the Bacillus anthracis collagen-binding protein [57] . Concatenation of carbohydrate-binding domains in Luke , Leo , and Jonah ( 3 ) lectins , which has previously been shown in WGA , Jacob lectins of E . histolytica , and peritrophins of insects , most likely increases the avidity of the lectins for glycopolymers [67 , 68 , 70] . While the boundaries between the three stages of the development of the cyst wall were somewhat arbitrary ( based upon the times selected for examining cyst walls with SIM ) , each stage had several essential , distinguishing features . In the first stage , encysting organisms rapidly and in an explosive manner transformed from amoeboid trophozoites , which were full of vacuoles and have acanthopods , to immotile , rounded forms making glycopolymers and Jonah ( 1 ) lectins in dozens of distinct vesicles . Because vesicles labeled with WGA and GST-CBM49 did not overlap , it is likely that they contain different glycopolymers . Definitive identification of glycopolymers in vesicles of encysting organisms will depend upon localization of tagged cellulose and chitin synthases , each of which is encoded by a single gene in A . castellanii [1 , 34–37] . Encysting Entamoebae also transform from amoeboid trophozoites to immotile , rounded forms making chitin , chitinase , and the Jacob lectin in distinct vesicles [45 , 64] . Encysting Giardia also transform from flagellated forms with an adherence disc to an spherical , immotile forms making β-1 , 3-linked GalNAc glycopolymer and cyst wall proteins ( CWP1 to CWP3 ) in distinct vesicles [65] . In contrast , no dramatic secretory event occurs in fungi or plants , which remodel their walls with growth , differentiation , or cell division , but never make their walls from scratch , with exception of the septum separating dividing cells [71 , 72] . In second stage , the two most remarkable features of the primordial cyst wall were the distinct distributions of the GFP-tagged lectins and the sets of small , circular ostioles . Early on , Jonah ( 1 ) -GFP was homogenously distributed across the primordial wall , while Luke ( 2 ) -GFP outlined some ostioles . Later , Luke ( 2 ) -GFP spread across the primordial wall , while Leo-GFP appeared in patches , which were not specific to any structure . It was as if Leo-GFP was secreted onto the surface of second stage organisms but had not yet found the glycopolymer to which it binds . Because the ostioles labeled with the least specific external probe ( CFW ) , it was not possible to determine whether ostioles are composed of cellulose , chitin , or another glycopolymer . Indeed CFW has been shown to bind numerous β-1 , 3 and β-1 , 4 polysaccharides , including cellulose , chitin , mixed linkage glucans , and galactoglucomannan [44 , 46 , 47] . The presence of the internal probe Luke ( 2 ) -GFP in the small ostioles did not settle this problem , as Luke ( 2 ) -GFP extracted from lysed trophozoites bound to both microcrystalline cellulose and chitin beads . As ostioles labeled with CFW before they contained Luke ( 2 ) -GFP , it is likely that glycopolymers are the drivers behind the circular or ring-like structures . How the small ostioles simultaneously appear in an even distribution across the surface of the primordial cyst wall and synchronously develop into conical structures is of great interest but is beyond the scope of the present study . Because A . castellanii has almost nine ostioles but uses just one for the excysting trophozoite to escape the cyst wall , it is likely that ostioles serve other functions such as holding layers of the cyst wall together and/or exchanging nutrients or waste products with the environment . In the third stage , the Jonah ( 1 ) lectin and the glycopolymer that it binds remained in the ectocyst layer , while Luke ( 2 ) and Leo lectins and the glycopolymers that they bind move to the endocyst layer and ostioles . In the same way , the outer primary layer of plant cells forms before the inner secondary layer [72] . The distinct distributions of the three GFP-tagged lectins in the primordial , third stage , and mature cyst walls strongly suggests each lectin binds to different glycopolymers ( e . g . cellulose versus chitin ) , glycopolymers modified in different ways ( e . g . unmodified chitin versus deacetylated chitosan ) , and/or glycopolymers with different microfibrillar structures ( e . g . microcrystalline versus amorphous cellulose ) . In support of this idea , the lectins had the same localization in mature cyst walls when expressed as an internal probe with a GFP-tag under its own or under a constitutive GAPDH promoter or when applied externally as an MBP-fusion . There may also be protein-protein interactions and/or lectin-glycoprotein interactions , which determine the localization of cyst wall lectins in the A . castellanii cyst wall . As an example of protein-protein interactions , an E . histolytica Jessie lectin has a chitin-binding domain and a self-agglutinating “daub” domain , which makes cyst walls impermeable to small probes such as phalloidin [64] . As an example of lectin-glycan interactions , the Gal/GalNAc lectin on the plasma membrane of Entamoebae binds to glycans on Jacob lectins , which , in turn , bind to chitin fibrils in the cyst wall [45] . Anti-GFP antibodies and MBP-lectin fusions showed Jonah ( 1 ) lectin and glycopolymers to which it binds are accessible in the ectocyst layer of mature cyst walls , while Luke ( 2 ) and Leo lectins and the glycopolymers to which they bind are , for the most part , inaccessible in the endocyst layer and ostioles . Jonah ( 1 ) lectin , which is unique , abundant , accessible , and conserved across many strains , therefore , is an excellent target for diagnostic antibody to A . castellanii cysts . Diagnostic antibodies bind to abundant cyst wall protein 1 of Giardia and Jacob2 lectin of Entamoeba [73 , 74] . In contrast , Luke ( 2 ) and Leo lectins are inaccessible and so not good targets for diagnostic antibodies . While abundant Luke ( 2 ) and Luke ( 3 ) lectins with two or three CBM49s , respectively , localized to the endocyst layer and ostioles of mature cyst walls , we have no evidence that other less abundant Luke lectins localize in the same place . In the same way , we do not know whether other Leo lectins localize to the endocyst layer and ostioles , or other Jonah lectins localize to the ectocyst layer . The large number of genes encoding Luke , Leo , and Jonah lectins may simply be necessary to increase the quantity of proteins coating glycopolymers in the cyst wall . Alternatively , there may be differences in timing and localization of proteins within the same family , based upon a TKH-rich spacer in Leo lectins or transmembrane helices in Jonah ( 3 ) lectins with three CAA domains ( S2 Fig ) . Finally , other candidate cyst wall proteins , which are abundant but present at lower copy numbers ( e . g . laccase or ferritin-domain protein ) and so not tested here , may have important functions in the cyst wall . While localization of abundant lectins was highly reproducible throughout cyst wall development , identification of the glycopolymers to which they bound was much more difficult . First , while each of the lectins bound well to microcrystalline cellulose , the lectins also bound to varying degrees to chitin beads . Second , the N-terminal CBM49 of a Luke ( 2 ) lectin fused to GST bound to the ectocyst layer of mature cyst walls , while N- and C-terminal CBM49s of the same Luke ( 2 ) lectin fused to MBP bound to the endocyst layer and ostioles . Third , WGA , bound to the ostioles and endocyst layer , the ectocyst layer , or all three structures of mature cyst walls , depending upon the experiment . Fourth and finally , experiments with anti-GFP antibodies and MBP-lectin fusions showed that proteins and glycopolymers in the endocyst layer and ostioles of mature cyst walls are , for the most part , inaccessible to external probes . We , therefore , cannot make any conclusions at this time as to the locations of cellulose and chitin in developing and mature cyst walls . In particular , we do not think it is a simple as cellulose in the ectocyst layer and chitin in the endocyst layer , as suggested by binding of GST-CBM49 and WGA in Fig 2 . However , localization of cellulose and/or chitin in vesicles of encysting organisms and in cyst walls is a solvable problem with 1 ) more specific probes for each glycopolymer , 2 ) GFP-tags for cellulose and chitin synthases that are each encoded by single genes , 3 ) protease or chemical treatments to expose glycopolymers to external probes , cellulases , and chitinases , and/or 4 ) inhibition of chitin and cellulose synthases with pharmacological agents or silencing RNAs . Indeed other investigators have explored the possibility of cellulose synthase inhibitors or cellulases as therapeutics versus A . castellanii cysts [75–78] .
|
A half century ago , investigators identified cellulose in the Acanthamoeba cyst wall , which has two layers and conical ostioles that connect them . Here we showed cyst walls contain three large sets of cellulose-binding lectins , which localize to the ectocyst layer ( a Jonah lectin ) or to the endocyst layer and ostioles ( Luke and Leo lectins ) . We used the lectins to establish a sequence for cyst wall assembly when trophozoites are starved and encyst . In the first stage , a Jonah lectin and glycopolymers were present in dozens of distinct vesicles . In the second stage , a primordial wall contained small , flat ostioles outlined by a Luke lectin . In the third stage , a Jonah lectin remained in the ectocyst layer , while Luke and Leo lectins moved to the endocyst layer and ostioles . A description of the major events during cyst wall development is a starting point for mechanistic studies of its assembly .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"chitin",
"trophozoites",
"parasite",
"groups",
"vesicles",
"chemical",
"compounds",
"cellulose",
"organic",
"compounds",
"parasitic",
"protozoans",
"parasitology",
"acanthamoeba",
"apicomplexa",
"protozoans",
"materials",
"science",
"macromolecules",
"cellular",
"structures",
"and",
"organelles",
"polymers",
"polymer",
"chemistry",
"proteins",
"chemistry",
"biochemistry",
"lectins",
"eukaryota",
"organic",
"chemistry",
"cell",
"biology",
"post-translational",
"modification",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"materials",
"protists",
"signal",
"peptides",
"organisms"
] |
2019
|
The most abundant cyst wall proteins of Acanthamoeba castellanii are lectins that bind cellulose and localize to distinct structures in developing and mature cyst walls
|
Zika virus ( ZIKV ) infection during human pregnancy may cause diverse and serious congenital defects in the developing fetus . Previous efforts to generate animal models of human ZIKV infection and clinical symptoms often involved manipulating mice to impair their Type I interferon ( IFN ) signaling , thereby allowing enhanced infection and vertical transmission of virus to the embryo . Here , we show that even pregnant mice competent to generate Type I IFN responses that can limit ZIKV infection nonetheless develop profound placental pathology and high frequency of fetal demise . We consistently found that maternal ZIKV exposure led to placental pathology and that ZIKV RNA levels measured in maternal , placental or embryonic tissues were not predictive of the pathological effects seen in the embryos . Placental pathology included trophoblast hyperplasia in the labyrinth , trophoblast giant cell necrosis in the junctional zone , and loss of embryonic vessels . Our findings suggest that , in this context of limited infection , placental pathology rather than embryonic/fetal viral infection may be a stronger contributor to adverse pregnancy outcomes in mice . Our finding demonstrates that in immunocompetent mice , direct viral infection of the embryo is not essential for fetal demise . Our immunologically unmanipulated pregnancy mouse model provides a consistent and easily measurable congenital abnormality readout to assess fetal outcome , and may serve as an additional model to test prophylactic and therapeutic interventions to protect the fetus during pregnancy , and for studying the mechanisms of ZIKV congenital immunopathogenesis .
Zika virus ( ZIKV ) is a mosquito-borne virus belonging to the genus Flaviviridae and is closely related to several other arboviruses , including dengue viruses ( DENV ) , yellow fever virus ( YFV ) , West Nile virus ( WNV ) , Japanese encephalitis virus ( JEV ) and tick-borne encephalitis virus ( TBEV ) [1] . ZIKV infection is often asymptomatic or presents as a mild , self-limiting febrile illness accompanied by rash , malaise , conjunctivitis , or muscle and joint pains [1–3] . The epidemic in the Americas first noted in 2015 was temporally related to the occurrence of thousands of fetal abnormalities during pregnancy [1–4] , and the causal relationship between ZIKV infection and adverse pregnancy outcomes ( e . g . congenital ZIKV syndrome , CZS ) has since been established in both humans and following experimental infection of pregnant mice and non-human primates [4–13] . CZS encompasses a number of potential abnormalities to include intrauterine growth restriction ( IUGR ) , fetal demise leading to spontaneous abortion , stillbirth , microcephaly , ocular disorders , and developmental abnormalities yet to be fully characterized [14–17] . Because of the devastating consequences of CZS , vaccine and drug countermeasures are being developed at a rapid pace , and the initial focus is to prevent infection and disease in women of child bearing age . Pre-clinical mouse pregnancy models have been developed to examine the consequences of ZIKV infection on the developing embryo/fetus [6 , 7 , 18–23] . Mouse models of ZIKV infection are constrained by the virus’s inability to replicate and cause overt clinical symptoms and disease manifestations in immunocompetent mice . It has been demonstrated that the murine innate Type I interferon ( IFN ) response abrogates ZIKV infection [22–24] . To overcome this innate response , investigators have employed immunocompromised mice genetically lacking Type I IFN signaling or wild-type mice treated with antibodies to block Type I IFN signals [6 , 7 , 25–27] . Indeed , these models recapitulate several disease manifestations of ZIKV infection observed in humans; however , the utilization of immunocompromised mice prevents the study of potential beneficial or detrimental impact of host immunity on ZIKV disease pathogenesis . Experimental conditions using immunocompetent pregnant mice which have been explored to date include inoculating mice with extraordinarily high doses of virus , or via specialized techniques or surgeries to directly deliver the virus to the maternal-fetal interface ( e . g . vaginal exposure , intrauterine infection ) [7–10 , 18–20] . These studies , too , have had some success . By these manipulations , vertical transmission of ZIKV from the mother to the fetus has been demonstrated in mice [6 , 7 , 18 , 19] , despite the fundamental differences of placental structures in humans and mice . These studies showed that fetal damage coincides with ZIKV disseminated to the placenta and embryo . However , these settings created high and sustained viral titers in the circulation or in the proximity of the placenta [6 , 7 , 19] , opening the possibility that the transplacental transmission and fetal damage may have been driven by unusually high maternal viral loads which would not have been achieved in an immunocompetent host . CZS in human newborns has been reported with asymptomatic maternal ZIKV infection [4 , 28] , suggesting that maternal innate immune responses may suffice to prevent or significantly attenuate illness , yet not suffice to protect the fetus from infection or poor clinical outcome . Little is known about the consequences of ZIKV infection during pregnancy beyond those resulting directly from vigorous ZIKV infection and vertical transmission . ZIKV infection can also damage the placenta leading to placental insufficiency and pathology [6 , 19] , introducing the question of whether CZS is caused by direct viral effects on the fetus , secondary effects resulting from primary placental damage , or both . The impact of ZIKV infection on developing embryos has been shown to vary with the gestational stage [14 , 15 , 29] . Infection during early pregnancy appears to pose the highest risk of fetal damage in both humans and mouse models [4–6 , 19 , 20 , 28–30] . Growing epidemiological data also indicate the risk of adverse fetal outcome when maternal infection occurs at second and third trimesters [4 , 31 , 32] . Animal models suggest that early stage infection prior to completion of placentation often results in higher rates of viral transmission and more severe fetal outcomes than later stage infection when the placenta is fully functional [6 , 19 , 20 , 29] . It is conceivable that the window of susceptibility and the impact of ZIKV infection on fetal outcome may depend on gestational stage-associated changes at the maternal-fetal interface . In this study , we utilized immunologically unmanipulated wild-type C57BL/6J mice , which therefore developed limited ZIKV infection . We found that even pregnant mice competent to generate intact immunity , including type I IFN signaling , nonetheless developed placental pathology and profound fetal abnormalities , including IUGR , embryonic malformation and high frequency of fetal demise . The impact on the developing embryo was caused by live infectious virus and was dependent on the dose and timing of viral administration during pregnancy . However , because viral replication was limited in the immunocompetent mice , vertical transmission was rare and , consequently , fetal abnormality did not reliably associate with the presence of ZIKV in maternal tissues , placenta or embryo . Our results reveal that adverse fetal outcome in the absence of vertical transmission was likely associated with placental pathology , including trophoblast hyperplasia in close proximity to embryonic blood vessels in the labyrinth and trophoblast giant cell necrosis in the junctional zone . Our findings indicate that , in this model , placental pathology leading to placental insufficiency rather than viral transmission can lead to profound adverse fetal outcome . Overall , our findings suggest the importance of a ZIKV pregnancy model that allows detection of placental pathology and fetal abnormality even with very low levels of overall viral infection , in order to test for interventions that can provide protective benefits against the full range of deleterious placental and fetal outcomes beyond merely preventing vertical transmission .
We infected wild-type , timed-pregnant C57BL/6J mice intravenously with 3 . 4 × 105 PFU of ZIKV Puerto Rico strain PRVABC59 at embryonic day 9 . 5 ( E9 . 5 ) of pregnancy , a time just prior to complete placentation ( E10-E10 . 5 ) . Mice were sacrificed 8 days post infection ( dpi ) at E17 . 5 , and uteri and individual embryos were evaluated for morphological appearance and weight ( scheme outlined in Fig 1A ) . Mock-infected or mice given heat-inactivated ZIKV showed no apparent differences compared with untreated mice ( Fig 1B , left 3 panels ) . In contrast , the uteri of live ZIKV-infected mice contained embryos undergoing fetal demise ( Fig 1B , right 3 panels ) . In some cases , the embryos had all undergone complete resorption , leaving only the placental residues and embryonic debris inside a constricted amniotic sac . In other cases , dams carried a mixture of placental residues and/or morphologically abnormal embryos . The few embryos that remained exhibited a range of significant growth restriction and malformations , as well as significantly reduced weight ( S1 Fig ) . All 16 ZIKV-infected dams showed abnormal pregnancies in that they all carried at least one or more dead embryos or placental residues , or morphologically abnormal but live embryos ( Fig 1C , left panel ) . Of the 120 implantation sites in the ZIKV-infected dams , 85% were affected , exhibiting complete embryonic resorption ( i . e . demise ) , or signs of growth restriction , malformation , or anatomical developmental defect by gross morphological examination ( Fig 1C , right panel ) . These data demonstrate that this experimental protocol results in a highly reproducible model of profoundly adverse ZIKV-associated fetal outcomes . To confirm that live infectious ZIKV is the causative agent of these adverse fetal outcomes , we pretreated the inoculum prior to infection with a neutralizing ZIKV-specific mAb , ZK-67 [33] , or the immune serum harvested from ZIKV-infected adult mice ( scheme outlined in Fig 1A ) . Results shown in Fig 1D and quantified in Fig 1E show that all of the dams infected with untreated ZIKV or irrelevant isotype-treated ZIKV showed abnormal pregnancies , and more than 90% of their embryos were affected ( i . e . demise or abnormal ) . In contrast , dams infected with ZIKV pretreated with either ZK-67 mAb or immune serum , and their embryos , showed no apparent morphological differences nor statistical significance compared with those inoculated with Vero cell culture supernatant ( i . e . mock infection ) . Together , these data demonstrate that the profound adverse fetal outcomes result from live infectious ZIKV , rather than other immuno-stimulatory molecules that potentially could be present in the inoculum . To determine whether ZIKV infection-induced fetal demise was dose-dependent , and to assess whether disease severity could be modulated by a lower inoculum of virus , we infected pregnant dams at E9 . 5 with 3 . 4 , 1 , or 0 . 34 × 105 PFU and sacrificed the mice 8 days later ( scheme outlined in Fig 2A ) . As shown in Fig 2B and 2C , we observed a dose-dependent effect of viral inoculum on both the number/percentage of affected embryos ( Fig 2B ) and the extent to which the embryos carried by individual dams were affected ( Fig 2C ) . While 85% of total embryos carried by the dams infected with 3 . 4 × 105 PFU were affected , only 36% of total embryos carried by the dams infected with 1 × 105 PFU were affected , which was still significantly higher than that of uninfected dams ( Fig 2B ) . In contrast , less than 3% of total embryos carried by the dams infected with 3 . 4 × 104 PFU were affected ( Fig 2B ) . As the dose of initial viral inoculation progressively decreased , so did the proportion of morphologically affected embryos to normal embryos carried by each individual dam ( Fig 2C ) . These data indicated a dose-dependent effect of administered viral inoculum on fetal outcome . To investigate the effect of gestational stage and placental maturity on ZIKV-associated adverse effects , we infected dams with 3 . 4 × 105 PFU ZIKV at E9 . 5 or E12 . 5 and compared maternal and embryonic outcomes at 5 dpi ( scheme outlined in Fig 2D ) . Consistent with our previous observation ( Fig 1 ) , we observed adverse fetal outcomes in dams infected at E9 . 5: 90% of the embryos were affected ( Fig 2E ) . In contrast , when dams were infected at E12 . 5 , only 13% of their embryos were affected ( Fig 2E ) . Thus , the gestational age at which the mice encountered ZIKV dramatically affected the severity of fetal outcome . To investigate the infection status of dams after exposure to ZIKV , we measured ZIKV RNA levels in several maternal tissues ( spleen , liver , kidney , brain and serum ) by quantitative real-time reverse transcription polymerase chain reaction ( RT-PCR ) . All ZIKV-infected dams exhibited detectable and comparable levels of ZIKV RNA in their spleens , regardless of the initial inoculum dose ( Fig 3A ) . However , ZIKV RNA was infrequently detected in the liver , kidney and brain , and in the serum of a single dam at a very low level when the dams were infected with the highest dose ( i . e . 3 . 4 × 105 PFU ) ( Fig 3B ) . None of the dams infected with the two lower doses had detectable levels of ZIKV RNA in the liver , kidney or serum . To determine the ability of ZIKV to reach the placenta and infect the embryo , we measured ZIKV RNA levels in the placental residues of resorbed embryos , and in the remaining intact embryos and their corresponding placentas . We detected ZIKV RNA in some but not all placentas of dams infected at all three doses ( Fig 3C , left panel ) , indicating the ability of ZIKV to infect the placenta , despite the dams’ intact immunity , and regardless of the ZIKV inoculum dose . In the cases of fetal demise in the dams that were infected with the highest dose , we detected ZIKV RNA in 15 out of 45 placental residues , some of which exhibited high RNA levels ( Fig 3C , right panel ) . However , the remaining 30 of 45 placental residues did not harbor detectable ZIKV RNA , even though all of their corresponding embryos suffered complete resorption ( Fig 3C , right panel ) . Of the 8 remaining embryos that were not reabsorbed , ZIKV RNA was only detectable in a single embryo ( head ) at a very low level , but not in its corresponding placenta ( Fig 3C , right panel ) . Overall , there was no statistically significant association between presence of detectable ZIKV RNA in the placenta and/or embryo and embryonic morphological outcome ( p = 0 . 97 , chi-square test; Fig 3D ) . When dams were infected at different gestational age ( i . e . E9 . 5 or E12 . 5 ) , ZIKV RNA was detectable in all of the maternal spleens after 5 days of infection ( Fig 3E ) . Notably , the levels of ZIKV RNA were comparable regardless of when the dams were infected , implying that the pregnant adult mice are equally susceptible to ZIKV infection at E9 . 5 and E12 . 5 , even though their embryos show different outcomes ( Fig 2E ) . Interestingly , whereas 32% of the placentas carried by dams infected at E9 . 5 had detectable ZIKV RNA , 68% of the placentas carried by dams infected at E12 . 5 had detectable ZIKV RNA ( Fig 3E ) , exhibiting greater placental penetration at E12 . 5 compared with E9 . 5 . Despite this , no embryos carried by dams infected at E12 . 5 had detectable ZIKV RNA ( Fig 3E ) . This is consistent with our previous observation ( Fig 3D ) that ZIKV RNA detected in the dams , placentas or embryos after infection does not associate with adverse fetal outcome; thus , the residual ZIKV RNA level is an unreliable indicator in this model . The complete fetal demise and resorption observed at 5 or 8 dpi following maternal infection at E9 . 5 underscores the devastating consequences of maternal ZIKV infection , but does not shed light on the pathophysiologic events that lead to the extreme outcome . The fact that we did not detect ZIKV RNA in all affected placentas and embryos at these time points could not rule out the possibility that fetal morbidity was attributable to ZIKV infection of the placenta and embryo that occurred earlier but was no longer detectible at the time of harvest . To investigate whether early vertical transmission occurs , and whether viral vertical transmission or other early ZIKV-associated events , such as placental pathology , contribute to eventual fetal demise , we infected dams at E9 . 5 with 3 . 4 or 1 × 105 PFU ZIKV and sacrificed the mice at 1 through 4 days post infection ( scheme outlined in Fig 4A ) . Consistent with our previous observation at 5 and 8 dpi ( Fig 3E and 3A , respectively ) , all dams harbored ZIKV RNA in the spleens at all earlier time points ( Fig 4B ) . ZIKV RNA levels were the highest in spleens at 1 dpi , then gradually declined . Notably , splenic ZIKV RNA levels did not differ significantly between dams infected with 3 . 4 or 1 × 105 PFU ZIKV at the various time points . Furthermore , ZIKV RNA was only detected in maternal serum at 1 dpi ( Fig 4C ) , suggesting rapid clearance of ZIKV from the maternal circulation in immunocompetent mice , without establishing prolonged viremia . The levels of virus in the serum in the two infected groups did not differ significantly . Although ZIKV RNA was detected , infectious virus , if present in either maternal spleen or serum , was below the limit of detection of standard plaque assay at all time points , further confirming that the ZIKV infection was limited , yet resulted in adverse fetal outcome . Gross morphological examination of the uteri and embryos in ZIKV-infected dams revealed visually obvious fetal abnormalities ( e . g . growth restriction , growth arrest , and delay of development ) as early as 2 dpi , with further deterioration over the next 24–48 h to include severe fetal demise ( Fig 4D ) . At 4 dpi , the embryos carried by dams infected with 3 . 4 × 105 PFU ZIKV were already frequently discolored and evidently bloodless , and likely non-viable ( Fig 4D ) . Consistent with our previous observation ( Fig 2 ) , fetal outcome of dams infected with 1 × 105 PFU ZIKV was less severe than that of dams infected with 3 . 4 × 105 PFU ZIKV ( Fig 4D ) . Notably , embryonic destruction did not occur synchronously for the entire litter , allowing us to observe varying degrees of early fetal abnormality and overall developmental delay . Furthermore , concordant with our previous observation ( Fig 3 ) , ZIKV RNA was detected only sporadically in the placentas ( Fig 4E ) and embryos ( Fig 4F ) at these earlier time points . While all embryos were affected at 3 dpi , 70% harbored no detectable ZIKV RNA in the placentas or the embryos , indicating that significant vertical transmission did not occur in the majority of affected embryos , even at earlier time points . Our analysis thus far revealed that ZIKV RNA level in the placenta or embryo did not predict adverse fetal outcome , even at earlier time points . Subsequently , we looked for histopathological lesions in the placenta that may explain fetal demise at 3 and 4 dpi . We found necrosis of embryonic endothelium within the embryonic blood vessels in the placental labyrinth ( Fig 5A ) . ZIKV envelop ( E ) protein antigen was detected within the endothelium of embryonic blood vessels in the placental labyrinth but not in embryonic cells or tissues at any time studied ( Fig 5B ) . The linear “flat” antigen staining pattern strongly suggested that ZIKV antigen localized predominantly in the long and thin embryonic endothelial cells , which line the embryonic blood vessels and have flat oval-shaped nuclei . In addition , necrotic cell debris within the embryonic vasculature often accompanied viral staining ( Fig 5B ) , suggesting that placental and embryonic blood circulation may have been compromised , most likely due to ZIKV exposure and subsequent destruction of embryonic endothelial cells . This cellular debris may have originated from necrotic endothelium and other cells . General histological examination of the placentas associated with morphologically abnormal embryos , regardless of whether viral antigen was detected or not , revealed that the most prominent placental pathological findings were focal or diffuse labyrinth trophoblast hyperplasia ( Fig 6A ) , and trophoblast giant cell degeneration and necrosis in the junctional zone ( Fig 6B ) . We also observed localized areas of necrotic cells and thrombi in the maternal blood spaces in the labyrinth of at least one placenta ( Fig 6C ) . Trophoblast hyperplasia within the labyrinth was evident , with focal or diffuse labyrinth trophoblast basophilia and mitotic trophoblasts ( Fig 6A ) in close proximity to embryonic blood vessels and maternal blood spaces ( trophoblast-lined sinusoids ) . Necrotic cell debris indicated focal necrosis within the junctional zone ( Fig 6B ) . We consistently observed these features as early as 1 dpi , implying that they may not have been caused by direct viral infection of placental cells . Moreover , while focal necrosis in the labyrinth was seen at 2 dpi , embryos began to show some focal necrosis only at 3 and 4 dpi , indicating that the placental lesions developed prior to those in the embryos . Although the placental necrosis may not have been severe enough to alone cause placental failure , the major anatomical lesions , labyrinth trophoblast hyperplasia and loss of embryonic blood vessels ( Fig 6D ) after endothelial necrosis likely interfered with embryonic nutrition exchange sufficiently to cause hypoxia to the embryo . Taken together , these data suggest that following maternal ZIKV exposure the normal placental architecture and function was substantially disrupted , causing placental insufficiency leading to fetal demise , even without widespread direct viral infection of the embryos . In further support of this view , histological evaluation consistently indicated there were no significant differences in the placentas between uninfected dams and dams infected at E12 . 5 ( Fig 7 ) . This is consistent with their normal embryonic morphology we observed at 5 dpi in these dams ( Fig 2E ) . There were no signs of trophoblast hyperplasia or necrotic/apoptotic cells in the labyrinth of the placentas from the dams infected at E12 . 5 ( Fig 7A and 7B , right panels ) , compared with those from the dams that were infected at E9 . 5 , which showed substantial lesions ( Fig 7A and 7B , left panels ) . These data indicate that placentas and embryos at E9 . 5 may be more vulnerable and thus exhibit greater pathology than those at E12 . 5 , which appear relatively unaffected . While real-time PCR detected ZIKV RNA in the placental tissue infected at E12 . 5 ( Fig 3E ) , we were unable to detect ZIKV antigen by immunohistochemistry ( Fig 7B ) . Placentas in dams infected at E12 . 5 appeared relatively healthy ( Fig 7 ) despite their higher ZIKV RNA levels ( Fig 3E ) . This , combined with the observation that neither placental nor embryonic ZIKV RNA levels were reliable predictors of fetal outcome ( Figs 3 and 4 ) , suggests that placental health is a more meaningful prognosticator of adverse fetal outcome than level of placental or embryonic ZIKV RNA . However , it should be stressed that the placental pathology observed in our model has yet to be directly linked with placental insufficiency as a cause of the high frequency of fetal demise . To demonstrate the utility of our model and to test whether the fetal morphological outcome could serve as a rapid readout for initial screening of candidate vaccines and therapeutics against ZIKV infection , we performed proof-of-principle studies using a potential protective therapy—the immune serum harvested from convalescent ZIKV-infected adult mice . When pregnant dams received immune serum prior to infection ( scheme showed in Fig 8A ) , their uteri and embryos showed no apparent morphological difference compared with mock infected dams , while dams that received saline or naïve serum showed complete fetal demise ( Fig 8B ) . Immune serum significantly decreased the number and percentage of affected dams ( Fig 8C ) and embryos ( Fig 8D ) to the level that was comparable to the mock infection . It also significantly reduced ZIKV RNA in the maternal spleens at 8 dpi ( Fig 8E ) . Moreover , the weights of embryos carried by the dams that received immune serum did not differ significantly from those carried by the dams that received mock infection ( Fig 8F ) . Together , along with results shown previously in Fig 1D and 1E , these results indicate that immune serum could prevent adverse fetal outcome in the pregnancy setting , and demonstrate that in this model , embryonic viability and gross uterine and embryonic morphology can serve as rapid and easy in vivo phenotypic readouts for the early screening of the efficacy of candidate vaccines and therapeutics against ZIKV infection . It is important to mention that despite observing no overt embryonic abnormalities 5 days after an E12 . 5 maternal infection ( Fig 2E ) , we cannot conclude that those embryos will be spared from future pathological consequences . Indeed , when a parallel cohort of dams was allowed to carry their pregnancies to term ( scheme showed in Fig 9A ) , we observed significant reductions in weights of newborns carried by ZIKV-infected dams ( Fig 9B ) , as well as markedly increased neonatal morbidity within 24 h of birth ( Fig 9C ) . We do not yet know whether this was a postnatal effect on the newborns from ZIKV exposure in utero or just maternal negligence and/or destruction of newborns that were undersized , perhaps resulting from maternal virus exposure . Nevertheless , our findings suggest that in this immunocompetent mouse model , maternal exposure to ZIKV at later stages of pregnancy may still pose risks of adverse postnatal sequela , even in the absence of acute fetal pathology in utero or at birth .
We report a highly reproducible pregnancy model in ZIKV-infected immunocompetent C57BL/6J mice , exhibiting a high incidence of adverse pregnancy outcomes including profound fetal morphological abnormalities and fetal demise , in the absence of viral vertical transmission . This model revealed fetal pathology , placental pathology , and potential postnatal impact . There are several key observations from this study . First , ZIKV infection of pregnant immunologically intact mice at E9 . 5 caused dramatic deleterious fetal outcomes . This was caused by infectious virus and depended on initial viral dose administered . Second , placental pathology was consistently observed in affected concepti , independent of detectable ZIKV RNA or protein in the placentas and/or embryos . Third , whereas pregnant immunocompetent females themselves resisted effects of ZIKV infection , their embryos appeared exquisitely sensitive to infection , and this effect depended on the time of infection during gestation . Fourth , even in the absence of acute fetal pathology in utero or at birth , pups born from ZIKV-infected dams may nonetheless be adversely affected . Our findings argue against the concept that ZIKV-associated adverse fetal outcome is simply the result of virus crossing the placenta and having direct effects on the fetus . Rather , we show that adverse fetal outcome can occur in the apparent absence of significant , direct fetal infection . Our data support a model of placental insufficiency as a contributing cause of fetal demise in an immunocompetent setting . Moreover , Yockey et al . showed that placental damage with the potential to harm the embryo can result from Type I IFN-triggered immunopathology [34] . Taken together , these findings are in agreement with studies in pregnant rhesus macaques and humans , which reported that the severity of fetal outcome did not generally correlate with clinical maternal disease severity , the ZIKV RNA level in the serum or urine , or the duration of maternal viremia [28 , 35] , and raise the possibility that maternal disease outcome may not be strictly predictive of fetal outcome . Humans usually contract ZIKV during mosquito feeding , which has been estimated to deliver between 104 to 106 PFU in related flaviviruses WNV and DENV [36 , 37] . Although the doses we administered were within this range , the possibility exists that the responses induced by experimental intravenous infection may differ from those induced by mosquito bites in humans . Indeed , a recent publication found that infection via mosquito bite delays ZIKV replication as compared to needle inoculum and alters ZIKV tissue tropism in rhesus macaques [38] . In addition , flaviviruses are known to antagonize Type I IFN signaling by targeting the key component STAT2 in order to replicate and cause human disease [24] . Unlike humans , mice are resistant to this mechanism , hence ZIKV is unable to replicate efficiently in immunocompetent mice . Our observations that ZIKV was rapidly cleared from maternal circulation and gradually declined in maternal spleen suggest a non-productive infection where after initially infecting host cells ZIKV may not replicate , or may replicate but fail to subsequently release infectious virus . Failure to establish prolonged viremia and cause vertical transmission during pregnancy differs from the common outcome of humans and nonhuman primates . However , our data suggest that the impact of ZIKV on the embryo can result from even a non-productive infection , highlighting the extreme sensitivity of the pregnancy process . Significantly , even in the case of detectable ZIKV RNA in maternal and fetal tissues , replicating infectious ZIKV are not always detectable in maternal or fetal tissues by standard plaque assay , as reported in a pregnant nonhuman primate study where fetal brain damage was observed [12] . Using a two-step enhanced plaque assay where virus is pre-amplified in insect cells prior to performing a standard plaque assay , we have successfully detected infectious viruses in the placentas of dams infected at E9 . 5 with 3 . 4 × 105 PFU ZIKV at 5 and 8 dpi ( 13% and 8% , respectively ) . Although the detection frequency was low , it implies that sporadic or modest productive infection is still possible even in these immunocompetent pregnant mice . Notably , our enhanced plaque assay failed to detect any infectious virus in maternal spleen at 5 and 8 dpi , suggesting that replication-competent infectious viruses may take sanctuary in the placenta , even when they are unable to persist elsewhere in immunocompetent mice . The abovementioned limitations may affect the direct application of our findings to human infection . However , our findings nonetheless demonstrate the potential for even non-productive ZIKV infection to cause critical fetal outcomes during pregnancy , and provide a useful model to investigate the underlying mechanisms . Despite inherent limitations , our data illustrate the importance of an immunocompetent mouse pregnancy model . In prior animal models where vertical transmission was demonstrated and believed to be required for adverse fetal outcome [6 , 7 , 19] , the use of Type I IFN blockade or the localized inoculation directly into the vaginal tract or uterus induced robust maternal viremia and/or direct viral seeding of reproductive tissues proximal to the conceptus . The resulting profound viral replication in the placenta was a likely driver of universal corresponding embryonic infection . The rampant viral infection in those studies presumably obscured the opportunity to identify other determinants of embryonic distress . In contrast , our use of systemically infected immunocompetent mice allowed the establishment of a sub-maximal level of infection that is subtle enough to observe previously unnoticed nuances of placental pathology . Our observation that adverse fetal outcome may occur even in the absence of vertical transmission suggests that ZIKV countermeasures that can effectively block vertical transmission still may not guarantee prevention of adverse fetal outcome . The placenta forms a crucial physical barrier between the maternal and fetal compartments , preventing pathogen transmission during pregnancy . It also mediates exchange of gases , nutrients and wastes . Previous mouse models have demonstrated that ZIKV was detected in the placenta and resulted in trophoblast infection and apoptosis , vascular damage , hemorrhage , and loss of placental structure [6 , 7 , 19] . This pathology could compromise placental barrier function to allow transplacental transmission , as well as disrupt exchange function , leading to placental insufficiency . In cases where we observed virus in the placenta , we also observed that ZIKV antigen was co-localized within normal and necrotic labyrinth embryonic endothelial cells , supporting the possibility that ZIKV infection directly damages the placenta and reduces its circulation function . However , the majority of the placental and fetal pathology observed in our model showed infrequent or no measurable associated viral infection , suggesting that placental function may be compromised despite negligible viral titer . Indeed , a pregnancy mouse model of murine γ-herpes virus 68 ( MHV-68 ) infection suggested that even in the absence of vertical transmission , the fetus could be adversely affected by an inflammatory response induced by viral invasion of the placenta [39] . In immunocompetent mice , it is conceivable that the maternal ( or even fetal ) immune response could cause collateral damage to the placenta in the course of viral clearance , resulting in immunopathology and placental insufficiency without detectable virus . Subsequently , such damage could in turn facilitate viral breach of the placental barrier in cases where the maternal infection has not been resolved . ZIKV infection has been shown to provoke an antiviral immune response in both human Hofbauer cells and mouse placenta [19 , 40] . Much remains to be learned about the possible protective and detrimental roles of maternal and fetal innate and adaptive immunity against ZIKV during pregnancy . A recent publication suggested that fetal Type I IFN signaling may play a detrimental role in mediating fetal demise after ZIKV infection by causing abnormal placental development [34] . Using a sophisticated breeding scheme and vaginal infection , it showed that only the embryos with functional Type I IFN signaling were resorbed despite relatively lower viral titer in their placentas , while their littermates with defective Type I IFN signaling continued to develop . It further showed that concepti with functional Type I IFN signaling had increased apoptotic cells in the placental labyrinth and upregulated hypoxia response genes in the embryo . Consistent with our observations , it supports the notion that placental dysfunction resulting from innate immune response is a possible cause of fetal demise , and taken together may provide insight into the underlying mechanisms of adverse fetal outcome of ZIKV infection in immunocompetent mice . However , whether the Type I IFN signaling-mediated placental pathology in immunocompetent hosts is specific to ZIKV infection or a general collateral damage after viral infection warrants further investigation . The most consistent pathological feature associated with fetal abnormality in our study was trophoblast hyperplasia and embryonic endothelial cell necrosis in the labyrinth leading to loss of labyrinth embryonic blood vessels , and focal necrosis in the junctional zone . Trophoblast hyperplasia in the labyrinth has been shown in an infection-unrelated study to lead to global disruption of labyrinth and vascular architecture , and ultimately fetal death [41] . Our data suggest that trophoblast hyperplasia is not directly caused by viral infection , as placentas from dams infected at E9 . 5 exhibited trophoblast hyperplasia , while those from dams infected at E12 . 5 did not , despite high placental ZIKV RNA levels . Although the direct link between trophoblast hyperplasia and fetal abnormality/demise remains to be established , our study clearly indicates that placental pathology can contribute to ZIKV-associated adverse fetal outcome independent of direct viral presence within developing embryos in immunocompetent mice . The devastating effects of ZIKV infection during pregnancy on fetal development makes it clear that a pregnancy model is necessary for pre-clinical testing of vaccines and therapeutics . Protective efficacy of experimental vaccines against ZIKV-induced fetal demise was reported in a mouse pregnancy model using Type I IFN blockade [42] . Our immunocompetent pregnancy model could provide an additional methodologically simple and high throughput platform for pre-clinical testing . However , as we have demonstrated that the absence of detectable virus does not necessarily predict fetal health , and birth of pups does not guarantee their survival , longer term postnatal follow-up monitoring in both models is necessary to confirm the health or survival of newborns . Moreover , it has been shown that ZIKV can persist at low levels in several anatomically compartmentalized areas , including the central nervous system , reproductive tracts or bodily fluids for up to several weeks [43–45] , which may be difficult to detect with standard RT-PCR or plaque assays . Similarly , placental tissue may provide a refuge where ZIKV may avoid complete clearance and persist at trace levels . Indeed , our ability to detect replication-competent virus at 5 dpi in placenta using an enhanced plaque assay , despite its absence in spleens , implies that the placenta may be a uniquely susceptible sanctuary for ZIKV compared with other tissues . Moreover , the profound harmful effects of infection or immunity on the fetus while leaving the mother essentially spared reveals the exquisite sensitivity of the placental-fetal environment to infection . Therefore , it is critical to determine not only whether a vaccine or drug successfully protects against overt disease/infection in adults , but also prevents fetal sequelae during pregnancy .
All animal studies were conducted in accordance with the Guide for Care and Use of Laboratory Animals of the National Institutes of Health and approved by Trudeau Institute Animal Care and Use Committee ( IACUC protocol # 16–009 ) The ZIKV Puerto Rico strain PRVABC59 was obtained through BEI Resources , NIAID , NIH ( NR-50240 ) . Vero cells ( African green monkey kidney epithelial cells ) were purchased from American Type Culture Collection ( ATCC CCL-81 ) and maintained in DMEM supplemented with 10% heat-inactivated FBS , 2mM L-glutamine , 1mM penicillin-streptomycin and 0 . 3% sodium bicarbonate at 37°C with 5% CO2 . ZIKV stocks were propagated in low passage number of Vero cells infected at a multiplicity of infection ( MOI ) of 0 . 1 . Culture supernatants from both mock-infected and ZIKV-infected cells were harvested 4 days after infection and clarified by centrifugation at 3800g for 15 min at 4°C . Fetal bovine serum was added to 20% final concentration ( v/v ) and aliquots were stored at -80°C . The infectious viral titer of stocks was determined by plaque assay on Vero cell monolayers . Briefly , Vero cells were seeded in 6 well plate to reach confluency in 18–24 h . Cells were infected with 1 ml serial 10-fold dilutions of supernatant prepared in neat DMEM for 1 h at 37°C . Then cells were overlaid with carboxymethyl cellulose medium ( 0 . 75% in Vero cell medium ) and incubated for 5 days at 37°C . Cells were fixed with methanol and stained with crystal violet to visualize the plaques . The most appropriate dilution was chosen to determine the amount of infectious virus in the stocks . Female C57BL/6J mice were purchased from Jackson Laboratory ( Bar Harbor , ME ) . Mice were set up for timed-mating at 8–12 weeks of age and the day the plug was found was considered as 0 . 5 day of gestation ( i . e . E0 . 5 ) . At indicated gestational days , pregnant mice were lightly anesthetized with isoflurane and infected with 3 . 4 × 104 , 1 × 105 , or 3 . 4 × 105 PFU of ZIKV in a volume of 100 μL by intravenous ( retro-orbital ) route . Control mice were left untreated , or inoculated with 100 μL of Vero cell culture supernatant ( mock infection ) or 3 . 4 × 105 PFU-equivalent of heat-inactivated or pretreated ZIKV . Some pregnant mice were allowed to give birth , and both the dams and newborns were examined within 24 h after birth . To prepare heat-inactivated ZIKV , viral stock was heated to 60°C for 1 h . To pretreat inoculum prior to infection , viral stock was incubated with 0 . 5 mg/mL of the neutralizing anti-ZIKV mAb ZK-67 [33] ( Absolute Antibody Ltd , Oxford , UK ) , 0 . 5 mg/mL of mouse IgG2a isotype control ( clone C1 . 18 . 4; BioXCell , West Lebanon , NH ) or 1:4 of ZIKV-immune serum per 3 . 4 × 105 PFU at 37°C for 1 h . An 3 . 4 × 105 PFU-equivalent of pretreated ZIKV was administered intravenously to pregnant mice . The viability and infectivity of ZIKV after treatment was confirmed by plaque assay . To generate convalescent-phase serum for passive transfer , female C57BL/6J mice at >10 weeks of age were infected intravenously with 2 × 105 PFU of ZIKV at day 0 and day 35 , and serum samples were collected at day 50 , pooled , aliquoted and stored at -80°C . Control serum was collected from uninfected naïve mice at the same age . Pregnant mice at E8 . 5 were treated intraperitoneally with 50 μL of immune serum , 250 μL of naïve serum or saline in a total volume of 250 μL . Mice were then infected intravenously with 3 . 4 × 105 PFU of ZIKV . Control mice were inoculated with 100 μL of Vero cell culture supernatant ( mock infection ) . ZIKV-infected pregnant mice were euthanized at indicated times after infection . Spleen , liver , brain , kidney , as well as embryos and their corresponding placentas were harvested , snap-frozen in liquid nitrogen , or homogenized with stainless steel beads in 1 ml of DMEM supplemented with 2% heat-inactivated FBS using a TissueLyzer II instrument ( QIAGEN ) . All samples were stored at -80°C until RNA purification or virus titration . Some placental and embryonic samples were fixed in 10% neutral buffered formalin for further histological analysis . To measure infectious viral particles , tissue homogenates were thawed and clarified by centrifugation at 2000g for 10 min at 4°C . Viral burden was determined by plaque assay on Vero cells as described above . To measure ZIKV RNA levels , frozen tissue samples were disrupted and homogenized in Buffer RLT ( QIAGEN ) containing β-Mercaptoethanol ( β-ME ) using a Tissuemiser homogenizer ( Fisher Scientific ) and total RNA was purified using RNeasy Mini Kit ( QIAGEN ) according to the manufacturer’s instructions . Total RNA was reverse-transcribed to cDNA using random hexamers and ZIKV RNA levels were determined by quantitative real-time PCR on a 7500 Fast Real-time PCR System ( Applied Biosystems ) using standard cycling conditions . The ZIKV-specific primer set and probe were previously published [46]: forward , 5’-CCGCTGCCCAACACAAG-3’ , reverse 5’-CCACTAACGTTCTTTTGCAGACAT-3’ , probe 5’-/56-FAM/AGCCTACCT/ZEN/TGACAAGCAGTCAGACACTCAA/3IABkFQ/-3’ ( Integrated DNA Technologies ) . Levels of PCR product were normalized to the housekeeping gene GAPDH: forward 5’-CTCGTCCCGTAGACAAAATGG-3’ , reverse 5’-AATCTCCACTTTGCCACTGCA-3’ , and probe CGGATTTGGCCGTATTGGGCG ( Integrated DNA Technologies ) . ZIKV RNA levels were interpolated against standard curves prepared by diluting RNA from uninfected tissue spiked with a known quantity of ZIKV ( as determined by plaque assay ) and expressed as PFU-equivalent per gram of tissue . Blood was collected from infected mice , allowed to clot at 37°C and serum was separated by centrifugation at 10600g for 10 min and stored at -80°C . Total RNA in 50 μL serum was purified using QIAmp Viral RNA Mini Kit ( QIAGEN ) . ZIKV RNA levels were determined by TaqMan Fast Virus 1-step Master Mix ( Applied Biosystems ) , interpolated against a standard curve prepared using known quantities of ZIKV and expressed as PFU-equivalent per mL of serum . Tissues were fixed in 10% neutral buffered formalin , embedded in paraffin , sectioned , and stained with hematoxylin and eosin . For immunohistochemistry staining of the ZIKV , 5-micron sections of paraffin-embedded tissues were washed with several changes of xylene to remove paraffin , and rehydrated through decreasing grades of ethanol ( absolute to 70% ) followed by deionized water . Endogenous alkaline phosphatase was blocked using Bloxall ( Vector Labs ) for 10 min . Tissues were washed with PBS containing 0 . 1% Tween 20 , blocked with 5% normal mouse serum ( Jackson Immunoresearch ) for 30 min , and then incubated with the primary Ab ( Zika virus Envelop protein antibody , GeneTex ) diluted in PBS with 0 . 1% Tween 20 and 5% normal mouse serum for 2 h at room temperature . After washing , tissues were incubated with anti-rabbit IgG alkaline phosphatase ( Vector Labs ) for 30 min , then developed using Vector Red alkaline phosphatase substrate ( Vector Labs ) , and counterstained with hematoxylin ( Fisher ) . Slides were imaged using a Nikon Eclipse Ci microscope and Nikon SPOT 2 digital camera . The primary Ab used to stain for ZIKV envelop protein was validated by specific staining of ZIKV-infected Vero cells ( S2 Fig ) . Uninfected and ZIKV-infected Vero cells processed as cell pellets , embedded in paraffin , and stained with the identical reagents as were the tissues were used as negative and positive controls , respectively , and were included in each staining protocol . Tissues from uninfected mice were also used as negative controls . Histology slides were reviewed by a board certified veterinary pathologist . Statistical analyses were performed using Prism 5 ( GraphPad Software ) . ZIKV RNA data were compared by nonparametric Mann-Whitney tests . ZIKV RNA levels that fell below the detection limit of our assays were arbitrarily assigned values 0 . 2 log below the limit of detection for nonparametric statistical analyses . Contingency data were analyzed by Fisher’s exact test using numbers in each group . Association/dissociation was analyzed by chi-square test . Weight data were analyzed by Student t test . *p<0 . 5 , **p<0 . 01 , ***p<0 . 001 , ****p<0 . 0001; ns , not significant .
|
Zika virus ( ZIKV ) infection during human pregnancy may cause severe congenital abnormalities and fetal death . There is currently no licensed vaccine or anti-ZIKV therapeutic to prevent or treat infection and/or disease . To generate an animal model that mimics human ZIKV infections , others have manipulated mice to impair their innate immunity , which allows ZIKV to develop high levels of infection . Here , we present an experimental model using immunologically unmanipulated pregnant mice and show that even limited maternal ZIKV infection nonetheless resulted in profound placental pathology and high frequency of fetal demise . However , neither viral RNA level in the dam , placenta or embryo reliably predicted fetal abnormalities . Our studies suggest that , in this model , placental pathology including trophoblast hyperplasia , focal regions of necrosis , and loss of embryonic blood vessels in the placenta likely promote adverse fetal outcomes . This immunocompetent pregnant mouse model provides clear in vivo phenotypic readouts ( e . g . embryonic viability and gross uterine and embryonic morphology ) to assess the potential for clinical benefit of candidate vaccines and therapeutics in a model not dependent on vertical transmission of virus .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"blastocysts",
"vero",
"cells",
"medicine",
"and",
"health",
"sciences",
"reproductive",
"system",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"animal",
"models",
"of",
"disease",
"biological",
"cultures",
"microbiology",
"animal",
"models",
"viruses",
"developmental",
"biology",
"model",
"organisms",
"rna",
"viruses",
"signs",
"and",
"symptoms",
"experimental",
"organism",
"systems",
"embryos",
"research",
"and",
"analysis",
"methods",
"animal",
"models",
"of",
"infection",
"embryology",
"placenta",
"animal",
"studies",
"medical",
"microbiology",
"microbial",
"pathogens",
"mouse",
"models",
"cell",
"lines",
"necrosis",
"diagnostic",
"medicine",
"anatomy",
"flaviviruses",
"trophoblasts",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"organisms",
"zika",
"virus"
] |
2018
|
Zika virus infection in immunocompetent pregnant mice causes fetal damage and placental pathology in the absence of fetal infection
|
The operation of a health care facility , such as a cholera or Ebola treatment center in an emergency setting , results in the production of pathogen-laden wastewaters that may potentially lead to onward transmission of the disease . The research presented here evaluated the design and operation of a novel treatment system , successfully used by Médecins Sans Frontières in Haiti to disinfect CTC wastewaters in situ , eliminating the need for road haulage and disposal of the waste to a poorly-managed hazardous waste facility , thereby providing an effective barrier to disease transmission through a novel but simple sanitary intervention . The physico-chemical protocols eventually successfully treated over 600 m3 of wastewater , achieving coagulation/flocculation and disinfection by exposure to high pH ( Protocol A ) and low pH ( Protocol B ) environments , using thermotolerant coliforms as a disinfection efficacy index . In Protocol A , the addition of hydrated lime resulted in wastewater disinfection and coagulation/flocculation of suspended solids . In Protocol B , disinfection was achieved by the addition of hydrochloric acid , followed by pH neutralization and coagulation/flocculation of suspended solids using aluminum sulfate . Removal rates achieved were: COD >99%; suspended solids >90%; turbidity >90% and thermotolerant coliforms >99 . 9% . The proposed approach is the first known successful attempt to disinfect wastewater in a disease outbreak setting without resorting to the alternative , untested , approach of ‘super chlorination’ which , it has been suggested , may not consistently achieve adequate disinfection . A basic analysis of costs demonstrated a significant saving in reagent costs compared with the less reliable approach of super-chlorination . The proposed approach to in situ sanitation in cholera treatment centers and other disease outbreak settings represents a timely response to a UN call for onsite disinfection of wastewaters generated in such emergencies , and the ‘Coalition for Cholera Prevention and Control’ recently highlighted the research as meriting serious consideration and further study . Further applications of the method to other emergency settings are being actively explored by the authors through discussion with the World Health Organization with regards to the ongoing Ebola outbreak in West Africa , and with the UK-based NGO Oxfam with regards to excreta-borne disease management in the Philippines and Myanmar , as a component of post-disaster incremental improvements to local sanitation chains .
Outbreaks of specific infectious diseases that may potentially be transmitted by human excreta , including cholera , Ebola and hepatitis A and E present a challenge to existing WASH ( water , sanitation and hygiene ) practices and a greater focus on practical in situ disinfection of human waste may offer an effective first step in the development of a longer-term sanitation ladder to support infection control . The research presented here focuses on an innovative in situ disinfection technique , which to date has been mainly applied in the context of a cholera outbreak , but which could potentially , and in the near future , provide a health protection intervention within the context of other outbreaks of neglected tropical diseases , including Ebola . Ten months after the devastating earthquake of 12th January 2010 , cholera appeared in Haiti for the first time in nearly a century . The outbreak escalated and as of 21st November 2014 , the resulting mortality had reached 8 , 505 and the cumulative morbidity had reached 717 , 203—equivalent to approximately 6 . 9 percent of the national population [1] [2] . According to the WHO [3] , the outbreak accounted for 57% and 53% of global cholera cases , and 58% and 37% of global cholera deaths reported in 2010 and 2011 respectively . Morbidity levels have probably been significantly higher than these figures suggest , as globally only a minority of cholera cases may be reported to the relevant authorities [3] . Cholera is a severe , acute , dehydrating diarrheal disease of humans , which , in the absence of adequate rehydration , can lead to death in both children and adults within twelve hours . The case-fatality rate for severe cholera without treatment can be as high as 50% [4] . The disease results from infection by a pathogenic strain of the bacterium Vibrio cholerae , which is capable of producing a potent toxin . Since the first recorded cholera pandemic , which began in 1816 , the pathogen has spread and evolved rapidly [4] [5] . The ongoing seventh cholera pandemic began in 1961 and there is now good molecular evidence to suggest a close relationship between the Haitian isolates of V . cholerae and variant V . cholerae El Tor O1 strains isolated in Bangladesh in 2002 and 2008 , and a more distant relationship with isolates currently circulating in South America [6] [7] . Established cholera control strategies call for a combination of interventions , including improvements to the quality and quantity of drinking water supplies , provision of consistently functional sanitation chains and promotion of effective hygiene practices . Under certain circumstances , the administration of oral vaccines to ‘at risk’ communities may also be recommended [8] [9] . Treatment of infected individuals is largely based on oral ( or in more serious cases , intravenous ) rehydration [4] . For the most severe cases , a suitable antibiotic , such as tetracycline , doxycycline or azithromycin , may be administered [10] . However , it has been widely recognized that treatment alone will not break the cycle of disease transmission and that improvements of WASH infrastructure are essential to achieving sustained control , elimination , or eradication of many tropical diseases [11] [12] [13] [14] . Doctors Without Borders ( Médecins Sans Frontières , or MSF ) is an international medical humanitarian organization that delivers emergency aid to people affected by armed conflict , epidemics , natural disasters , and exclusion from healthcare . It has operated successfully in numerous cholera emergencies during the past four decades , and has formulated effective field response strategies to cholera outbreaks , including the design and operation of appropriate water and sanitation technologies . The organization has been active in Haiti for over twenty years and , in collaboration with the Haitian Ministry of Public Health ( MSPP ) , has been a leading provider of treatment to cholera patients in the country since the beginning of the outbreak , treating more than 300 , 000 patients by September 2014 [15] [16] [17] . The established MSF protocol for dealing with human fecal wastes in emergencies involves the addition of 2% chlorine solution to each bucket of patient feces or vomit and the construction and operation of soil infiltration pits or trenches to dispose of the large volumes of waste produced by CTC operations [18] . However , this approach is only appropriate when the water table remains at least 1 . 5 meters below the lowest point of the excavated pit or trench . In the densely populated Haitian capital of Port-au-Prince , the water table may be considerably higher ( as little as 30 cm below the surface during periods of heavy rainfall ) , and clearly , under these circumstances , infiltration cannot provide safe disposal of the infectious human wastes arising [19] . From the outset , the response of many of the international organizations operating in the wake of the Haiti cholera outbreak was to instigate road haulage ( by truck ) of all fecal waste originating from cholera patients ( chlorinated or otherwise ) to a centralized waste pit at the Truitier landfill site on the outskirts of Port-au-Prince . This facility is situated close to the impoverished and densely populated community of Cité Soleil on the western outskirts of the capital , a few hundred meters from the coast and on the aquifer of the Cul-de-Sac plain , traditionally a source of raw drinking water for the city of Port-au-Prince [20] . At an early stage of the emergency response , water and sanitation engineers of MSF-OCA ( Médecins Sans Frontières–Operational Centre Amsterdam ) concluded that using the Truitier landfill site for the disposal of cholera wastes represented a clear hazard to human health and the organization therefore decided that it was not prepared to countenance this practice . The organization further decided that the practice of ‘super-chlorination’ followed by disposal to the environment was also unacceptable . The principal reasons for these decisions were: Road transportation of significant quantities of contaminated wastewater was considered hazardous to human health , particularly within the complex , often chaotic , urban context of post-emergency Port-au-Prince . The contaminated wastewater arising from CTC is characterized by extremely high concentrations of readily-oxidizable matter . It would therefore be imprudent to assume that a wastewater disinfection process based on chlorination would consistently disinfect the waste to an adequate degree [21] , given that the ability of these in situ disinfection strategies to reduce target pathogens had not been formally assessed [19] . It has been suggested that certain strains of V . cholerae ( the “rugose” phenotype ) may be more resistant to chlorine-based disinfection as a result of exopolysaccharide production , which promotes cell aggregation . Such strains may therefore pose an elevated risk to human health [22] [23] . Even if ‘super-chlorination’ were able to reduce Vibrio cholerae numbers to levels that did not pose a significant risk to those living downstream of CTC operations , the production of combined chlorine residuals and the relatively high operational costs associated with this process would likely make it both environmentally and financially unacceptable in the medium- to long-term . Moreover , this approach to disinfection does not significantly remove suspended material . By October 2010 , the rapid spread of the Haitian cholera outbreak had resulted in a pressing need for CTC facilities throughout the country and a novel , low-cost and consistently-effective way to treat and disinfect the wastewaters from MSF CTC operations was therefore urgently required . In Port-au-Prince , a partly-commissioned MSF maternity hospital ( ‘Delmas 33’ ) was converted by the organization into a CTC within a matter of days . By the time its operational life ceased in early March 2011 , more than 3 , 000 cholera in-patients had been treated at the facility . It was then converted back into a maternity hospital , and a new MSF CTC was established on nearby tennis courts . In total , at these two CTC , MSF water and sanitation engineers were required to treat and dispose safely of over 320 , 000 liters of wastewaters , which were potentially infected with high levels of Vibrio cholerae . It has been estimated that wastewaters from CTC , especially those components derived from patient stool buckets , may contain more than 107 Vibrio cholerae per 100 ml [24] [25] [26] . Such wastes must therefore be treated and disposed of with extreme caution . Within the Haitian context , rapid intervention to provide effective disinfection of this wastewater was essential in order both to control disease transmission and to respond to the prevailing concerns of the local populace with regard to the management of cholera wastes by international organizations . The onsite treatment of CTC wastewaters within the challenging context of medical emergencies needs to be relatively low-cost , logistically simple , rapid to deploy , immediately effective and capable of removing microbial pathogens significantly more effectively than conventional treatment technologies . Such systems have rarely been established , and no peer-reviewed literature that critically evaluates their operational performance is available . However , the concentration of Vibrio cholerae in CTC wastewaters and the potential risk to public health that the pathogen represents may be estimated from previous studies . During a two-year investigation of cholera carriers in the Philippines , Dizon et al . [27] measured the numbers of Vibrio cholerae per gram of feces among human populations in areas of the country in which the disease was endemic or epidemic . The feces of ‘simple carriers’ contained between 102 and 105 Vibrio cholerae per gram of feces , whereas the feces of patients presenting symptoms of ‘mild cholera’ were shown to contain between 106 and 109 Vibrio cholerae per milliliter of stool ( on their first day of illness ) . Howard et al . [24] [25] examined the wastewater from a hospital operated by the UK-based NGO Oxfam in Bangladesh , which admitted between two and forty confirmed cholera cases per day . The authors recorded levels of Vibrio cholerae between 5 x 105 and 5 x 107 per 100 ml of wastewater . It is worth noting that the level of Vibrio cholerae was demonstrated to exceed that of thermotolerant coliforms in this instance . In the work described here , the authors aimed to use the best available expertise to design , construct rapidly and operate an effective onsite CTC wastewater treatment system that would protect the health of the inhabitants of Port-au-Prince from the potential risk of disease associated with contaminated wastewaters . Further , it was considered essential that this innovative technology should be subjected to a robust critical risk evaluation of each stage of the project cycle . This was designed to maximize human health protection at the time of the emergency and to enable MSF and other NGO ( potentially operating in other parts of the world ) , to gain the fullest possible benefit from the resulting evidence-base . Based on initial estimations of Vibrio cholerae levels in the CTC wastewaters and with reference to the available literature , a wastewater management strategy , involving four consecutive and distinct barriers to the transmission of Vibrio cholerae , was proposed as follows: Initial chlorination of patient feces within stool buckets immediately following collection by MSF health-care professionals , as already practiced according to MSF protocols [18] [28]; Storage of pooled CTC wastewaters in open tanks—in practice for up to twelve weeks ( average six weeks , minimum three ) at relatively high ambient temperatures—resulting in a further reduction in levels of enteric microorganisms as a result of natural biological , chemical and physical processes; The design and operation of a novel batch-operated onsite wastewater treatment and disinfection plant , as described in detail below; and finally Controlled effluent disposal within soil infiltration trenches according to existing MSF protocols [18] . In-house MSF water and sanitation expertise , supported by expert external advice , were used to develop a shortlist of three technologies that might meet the objective of achieving effective , robust and relatively low-cost onsite treatment of the CTC wastewaters: Protocol A: Coagulation/flocculation and disinfection of the wastewater with hydrated ( slaked ) lime ( Ca ( OH ) 2 ) at high pH levels , using a treatment system that was based on the methodology of Taylor et al . [29] [30]; Protocol B: A novel approach involving disinfection at low pH levels using hydrochloric acid , followed by pH neutralization and subsequent coagulation/flocculation , achieved using aluminum sulfate ( or an alternative low-cost coagulant ) ; and Protocol C: Septic tank treatment combined with an anaerobic filter . Protocol C was rejected at an early stage , as it was considered that this approach would take too long to establish and would be insufficiently robust to operate effectively and reliably within an emergency setting . Subsequently , batch treatment systems based on Protocols A and B were designed , operated , and monitored within two CTC operations in Port-au-Prince , over a period of six months . The main goal of the treatment was to achieve a level of disinfection of the highly contaminated fecal waste that was adequate to release the effluent and the sedimented sludge into the environment without introducing a new disease transmission route . Effective disinfection was achieved through the combined and simultaneous action of two mechanisms , namely: The exposure of the pathogens to an alkaline ( ‘protocol A’ ) or acidic ( ‘protocol B’ ) environment , resulting in pathogen deactivation The physical removal of the pathogen as a result of coagulation and flocculation and sedimentation . The sedimented sludge was subsequently treated in drying beds before incineration or controlled infiltration to soil ( see details on the next section ) Gram-negative ( Gram- ) and Gram-positive ( Gram+ ) bacteria are both sensitive to high pH , although Gram- bacteria ( including Vibrio cholerae ) tend to be more susceptible to high pH levels because of their relatively thin peptidoglycan layer: the Gram- layer is in fact only about 2 to 3 nm thick , whereas the Gram+ layer is about ten times thicker [31] . The peptidoglycan layer stabilizes the cytoplasmic membrane of intact bacterial cells against the pressure exerted by the cytoplasm [32] . Therefore , the thinner peptidoglycan layer associated with Gram- bacteria may less effectively prevent the cytoplasmic membrane from bursting once it is weakened by a high pH environment [33] . There are multiple hypotheses as to how strong and weak , organic and mineral acids inhibit or destroy bacteria . In general , acids have antimicrobial activity both in their undissociated and dissociated forms ( although the former has a stronger antimicrobial effect ) [34] . One of the prevailing hypotheses is that strong acids inhibit or destroy microorganisms by interfering with the permeability of the microbial cell membrane . The acidic solution interferes with the substrate transport and with the oxidative phosphorylation from the electron transport system . This results in the acidification of the cell content , which is considered to be the principal ( but not the only ) cause of inhibition and death [35] . It has also been suggested that some acids may also inhibit or kill bacteria by blocking amino-acid uptake through the membrane [36] . Moreover , some acids may enter the bacterial cells as undissociated molecules that are soluble in phospholipid membranes and then acidify the cell interior [37] . For the purposes of this study , thermotolerant coliforms were used as an index of disinfection efficacy . These bacteria primarily originate from the intestines of warm-blooded animals and are widely used as an indicator of the presence of fecal material in water . Although it was not feasible under the conditions of the study described here to enumerate the pathogen Vibrio cholerae directly , all available evidence suggests that for the extreme pH levels achieved during the treatment protocols described here , thermotolerant coliforms represent an acceptable conservative indicator of the presence of Vibrio cholerae in that they exist at high concentrations in human feces and are , like Vibrio cholerae , a Gram- bacterium of primarily enteric origin . Although future work on the behavior of specific pathogens to low-cost on-site disinfection processes is warranted , the authors believe that the approach taken here represents a robust approach to estimating the risk of pathogen transmission . Another aspect of this study that was partly limited by the constraints of an emergency setting was chemical analysis of the wastewater . However , wastewater from cholera treatment centers is commonly composed of human feces and sullage ( graywater ) from personal washing facilities and hygiene practice . Therefore the main components of the wastewater are generally known though the alkalinity , the buffering capacity , the relative concentrations of organic matter and other components are liable to vary between CTC and with time . Several studies have previously determined the composition of wastewater derived from various infectious ( including tropical ) disease hospital departments [38] [39] [40] [41] [42] . However , although the quantity of alkaline or acidic solution required to achieve adequate treatment and disinfection by the protocols described here will depend on a number of wastewater characteristics ( e . g . , organic content and alkalinity ) it is important to note that the protocols are defined by pH ‘end-points’ , in that reagents are added until a prescribed pH level is reached so that variability in wastewater composition ceases to be a factor in treatment efficacy . Since this will be the case in other future applications of the protocols , the authors believe that the inability to obtain detailed data on the wastewater composition under the conditions of this study does not constitute a significant weakness in the research .
Simple jar-test studies , using five-liter beakers , were initially used to investigate the efficacy of Protocol A with regard to the removal of thermotolerant coliforms and suspended solids ( or turbidity ) . At the inception of each jar-test , a small sample of untreated wastewater ( approximately 30 ml ) was taken and the following parameters were tested for: turbidity–recorded as nephalometric turbidity units ( NTU ) ; presumptive thermotolerant coliforms–recorded as colony-forming units ( CFU ) per 100 ml; pH level; and quantities of chemical reagents used–recorded as grams or milligrams per liter . The first step of each jar-test experiment involved the step-wise addition of hydrated lime slurry ( Ca ( OH ) 2 ) to the wastewater , until the pH of the mix reached a level between 11 . 4 and 12 . 2 . This was immediately followed by three minutes of ‘rapid-mixing’ , followed by 15 minutes of ‘slow-mixing’ ( both steps being achieved manually in the absence of a mechanical jar test-rig ) . The contents of the beaker were then left to settle overnight . The supernatant was subsequently removed and its pH level adjusted to approximately 7 by the addition of concentrated hydrochloric acid ( HCl ) . At the end of each jar-test process , a small sample ( approximately 50 ml ) of supernatant was removed and tested for the same set of wastewater quality parameters as mentioned previously . Jar-test studies were similarly performed in order to investigate the efficacy of Protocol B . At the inception of each jar-test , a small sample of untreated wastewater ( approximately 30 ml ) was tested for the same set of parameters as in Protocol A . The first step of the jar-test experiment for Protocol B involved the addition of hydrochloric acid ( HCl ) , at a quantity sufficient to decrease the wastewater pH to a level between 3 . 7 and 3 . 9 , so as to achieve disinfection of the wastewater . This was immediately followed by ‘rapid-mixing’ for one minute . Following overnight sedimentation , the wastewater was adjusted to a pH level of approximately 7 , by the addition of the hydrated lime slurry that was also used for Protocol A . At this point , another small sample ( approximately 30 ml ) of supernatant was removed for analysis as before . Aluminum sulfate ( 75 to 150 mg/L—either as Al2 ( SO4 ) 3 * 16H2O or as Al2 ( SO4 ) 3 * 18H2O ) was next added to the beaker as a coagulating agent , in order to achieve suspended solids removal , and consequently , to achieve a further reduction in microbial levels . The addition of aluminum sulfate was immediately followed by three minutes of rapid-mixing , followed by 15 minutes of slow-mixing . Following the slow-mixing phase , the wastewater was allowed to settle in the five liter beaker reactor for one hour . Once again , a small sample of supernatant ( approximately 30 ml ) was removed for analysis , as before . Laboratory jar-testing of the high pH treatment process ( Protocol A ) using hydrated lime ( Ca ( OH ) 2 ) and , at a later stage , the low pH treatment process ( Protocol B ) using aluminum sulfate , was followed by full-scale batch treatment . Here , wastewater and coagulants ( added at concentrations suggested by the jar-tests ) were combined within regimes that mimicked , as closely as possible , initial rapid-mixing followed by slow-mixing , and finally settlement for a minimum period of fourteen hours , all within a 30 m3 circular open tank . Figs 1 and 2 outline the full-scale treatment procedures adopted for each protocol . The 30 m3 treatment tank ( reactor ) was filled to a maximum level of approximately two-thirds of the total capacity of the tank . The wastewater was then mixed by re-circulation using a gasoline-fueled centrifugal pump so as to obtain a homogenous mix . The established set of bacteriological and physico-chemical parameters measured during the pilot-studies was determined for the wastewater influent from grab samples of approximately 30 to 50 ml . In addition , the COD ( mgO2/L ) of the reactor influent was measured . The lime slurry was prepared by the addition of hydrated lime to chlorinated drinking water , at a concentration of approximately 20 g/L , in a 200 liter drum , placed on a platform above the reactor tank , directly above the influent pipe . Lime slurry was continuously added to the wastewater , in an attempt to achieve rapid-mixing ( with the inflow hose running parallel to the tank wall by means of an ‘elbow-joint’ ) , until the pH level of the circulating wastewater was measured to be greater than , or equal to , 11 . 4 . Once the target pH level had been reached , the pump was operated continuously at a relatively low revolution rate for approximately 15 minutes , in order to achieve slow-mixing , and consequently to aid flocculation of the reactor contents . The pump was then switched off and the wastewater was left to settle for at least fourteen hours . A small grab sample of the resulting ( ‘partially treated’ ) supernatant was removed for analysis using the same set of parameters used to test the untreated wastewater influent . After measuring the depth of sludge in the tank , the supernatant was carefully pumped into a nearby 3 . 8 m3 tank , taking care not to re-suspend the sludge . The contents of this tank were then adjusted to a pH level of between 7 and 8 , by the addition of HCl . A final sample of supernatant was removed for analysis as before . Provided that the effluent had reached a quality considered to be ‘satisfactory’ ( defined as having achieved a turbidity level of less than 50 NTU , a pH level of between 6 and 8 , and containing fewer than 1 , 000 thermotolerant coliform CFU per 100 ml ) , this ‘final effluent’ was then carefully infiltrated into onsite soil trenches . If the effluent quality failed to meet these quality criteria , the entire treatment procedure was repeated before the final effluent was allowed to be infiltrated to the soil . The process of tank filling was identical to that followed under Protocol A and grab samples of the influent were analyzed for the same parameters prior to treatment . HCl was then added to the tank contents until the pH level of the circulating wastewater was recorded to be equal to , or lower than , 3 . 9 . Once the target pH level had been reached , the contents were recirculated slowly by pumping for five minutes to ensure that the pH level within the reactor was as homogenous as possible . The pump was then switched off , and the tank contents were left to stand for a minimum period of no less than fourteen hours . When the depth of sludge in the tank had been measured , the supernatant was carefully pumped ( taking care not to re-suspend the limited quantity of sludge that had been produced at this stage ) into the nearby smaller tank ( 3 . 8 m3 ) . The contents of this tank were adjusted to a pH level of between 6 and 7 , by the addition of lime slurry , before a grab sample was removed for analysis using the same set of parameters as before . The wastewater at this stage was considered to be ‘partially treated’ . It is perhaps worth noting that , while the addition of HCl , as described above , did not in itself result in coagulation/flocculation , it was considered useful to take advantage of unaided overnight sedimentation before the supernatant was removed for subsequent coagulation/flocculation the next day . The remaining , relatively small quantity of ‘low-pH ( disinfected ) sludge’ removed from the bottom of the treatment tank in Protocol B was blended with the much larger volume of ‘high-pH sludge’ produced by Protocol A , in order to produce a pH-neutral blend . A concentrated solution of aluminum sulfate was prepared by dissolving approximately 300 g of the hydrated salt in 1 liter of chlorinated drinking water . Four transparent beakers , each containing 1 liter of wastewater , were used for jar-tests , with the aim of determining the quantity of coagulant needed to achieve adequate sedimentation . This was found to be approximately 100 mg/l . The aluminum sulfate solution was added to each 3 . 8 m3 tank , with manual ‘rapid-mixing’ achieved using a short stirring rod for approximately 5 minutes ( flash-mixing ) , followed by a manual slow-mixing phase of about 15 minutes , using a longer stirring rod ( to improve the formation of flocs ) . The wastewater was then left to settle for approximately one hour and a grab sample of approximately 30 ml of supernatant ( ‘final treated effluent’ ) was tested for the standard parameters , as before . Provided that the effluent had achieved the ‘satisfactory’ quality previously specified under Protocol A , the effluent was carefully infiltrated to the soil . Again , if the quality criteria for satisfactory final effluent had not been met , the coagulation/flocculation procedure , using aluminum sulfate , was repeated . If the effluent quality level had still not met the specified quality standards at this stage , the entire treatment process , including low-pH disinfection and coagulation/flocculation , would have been repeated , but in practice this was never required ( Fig 1 ) . All sludge had been exposed to either the high or low pH environment ( that had each demonstrated more than 3-log reduction in levels of thermotolerant coliforms in the supernatant ) for at least twenty four hours . Bacterial enumeration of sludge is not possible by membrane filtration as the solids block the pores of the nitrose-cellulose filter . The alternative enumeration method by ‘multiple tube’ ( most probable number ) was not feasible under the emergency field conditions at the CTC . Therefore , although it can be assumed from the analysis of the supernatant that significant disinfection had been achieved throughout the contact tank during the coagulation-flocculation and subsequent sedimentation stages , the precautionary principle was used in all subsequent handling of the sludge . All sludge was air dried for at least fifteen days and then either carefully placed in infiltration pits ( as continues to be common practice for the disposal of fresh , untreated human excreta in many CTC operations around the world ) or incinerated along with solid hazardous health-care wastes , as recommended by Gautam et al . [40] . The authors therefore conclude that the hazard of human infection from the sludge was appropriately managed . The following set of physico-chemical and bacteriological analyses was performed on all Protocol A and Protocol B samples ( both during pilot-scale studies and full-scale plant operation ) . The main aim of all analyses was to determine the degree of reduction in turbidity ( NTU ) or total suspended solid ( TSS ) , and thermotolerant coliforms ( CFU per 100 mL ) . Measurements of COD concentration were only achieved during full-scale operation of Protocol A . All analyses were undertaken on grab samples , typically 30–50 ml of the wastewater , taken either from the five-liter beakers ( pilot-scale trials ) or from the full-scale treatment tanks . Initially , turbidity levels were recorded ( as NTU ) following a simplified 'turbidity tube' method [44] . This method was later replaced by a spectrophotometric protocol , using a Hach portable turbidimeter ( model 2100P ) , which operated within a wavelength range of 400 to 600 nm . All turbidity data reported here were recorded spectrophotometrically . Measurement of total suspended solids ( as mg/L ) was achieved by filtration of the sample through a glass-fiber filter , according to standard methods [45] . As an oven was not available in the field , filters were dried at ambient temperature ( normally greater than 30°C ) until constant weight was achieved ( normally within 48 hours ) . pH levels were measured several times during both protocols to minimize the quantity of reagents used to achieve adequate disinfection ( and in the case of Protocol A , to ensure effective coagulation and flocculation ) . The pH level was also frequently measured during the later neutralization phases ( both protocols ) in order to achieve a final pH level of between 6 and 8 . A Palintest Micro 500 pH meter was used for all measurements . pH buffers ( 7 . 0 and 4 . 0 ) were used for pH meter calibration and pH probes were stored in a saturated KCl solution . In addition , because of the potential for damage to the probe at high and low pH levels , simple pH litmus paper strips were frequently used to verify the pH values obtained . COD ( as mgO2/L ) was measured using a simplified spectrophotometric field kit ( Palintest ) . The samples were digested at 150°C for two hours in a strong solution of sulfuric acid , in the presence of chromium and silver salts . The tubes were then cooled and the color was measured using the Palintest photometer . Four test kits were used , with a maximum detection level of either 2 , 000 mg/L or 20 , 000 mg/L , for analysis of the influent , and either 150 mg/l or 400 mg/l , for analysis of partially , or fully-treated wastewaters . During the field conditions encountered , the quantities of all chemical reagents used were recorded as accurately as possible during all operations . Presumptive counts of thermotolerant coliforms were recorded as colony-forming units ( CFU ) per 100 ml , following membrane filtration through a sterile nitrose-cellulose membrane filter ( 0 . 45 μm ) ( using a DelAgua water-testing kit , sterilized by the production of formaldehyde , formed from burning methanol ) . Acidic and alkaline samples were washed through the filter with an excess of distilled water for one minute , to ensure that the pH level of the membrane prior to incubation approximated 7 . Following filtration , the filters were incubated at 44°C ±1°C for 18 to 24 hours , on sterile absorbent pads , soaked in membrane lauryl sulfate broth ( Oxoid ) . Samples were diluted according to their predicted bacterial counts using de-ionized water . Following incubation , all yellow colonies greater than 2 mm in diameter were enumerated and recorded as CFU of presumptive thermotolerant coliforms per 100 ml of the original sample .
Treatment by both Protocols A and B achieved an effectively clarified effluent , with a turbidity reduction consistently greater than 80% and a mean reduction of 93% ( 1 . 1 log ) . Mean TSS reduction was 92% ( 1 . 1 log ) . Removal of thermotolerant coliforms was consistently greater than 99 . 8% ( 2 . 7 log ) , with a mean reduction of 99 . 9% ( 3 log ) . The mean removal of COD ( calculated with reference to an average value for untreated wastewater in the absence of sufficient data ) was consistently higher than 99% ( 2 logs ) . The rate of consumption of chemical reagents during the full-scale treatment operations ( when adequate monitoring equipment had become available in the field ) is summarized in Table 2 . A comparison of the two protocols suggests that , overall , Protocol B was more efficient in terms of the total mass of reagents required to achieve the desired treatment outcome . Protocol B was demonstrated to require on average 1 . 30 L of HCl per m3 of wastewater , compared with 2 . 25 L HCl per m3 wastewater for Protocol A . Additionally , a mean dose of 0 . 47 kg of Ca ( OH ) 2 was required per m3 of wastewater for Protocol B , compared with 3 . 96 kg Ca ( OH ) 2 per m3 of wastewater for Protocol A . The mean residual aluminum level in the treated effluent from Protocol A was shown to be 0 . 05 mg per L and 0 . 07 mg/l for Protocol B . Levels of residual aluminum were never reported to exceed 0 . 1 mg/l . The mean volume of sludge produced was 6% ( vol . /vol . ) . Operating a novel wastewater treatment plant during the Haitian cholera outbreak presented significant logistical problems , the foremost of which were limited access to adequate supplies of good quality chemical reagents ( including lime , alum and hydrochloric acid ) and inadequate provision of resources and facilities to support effective operational research . Notwithstanding these obstacles , a novel CTC wastewater treatment and disinfection system was designed and operated successfully , and provided a potentially very useful knowledge-base for further development and application of the technology in other settings . During the entire operational stage ( rather than solely during the final phase reported above ) , Protocol A demonstrated a greater requirement for chemical reagents than Protocol B ( in terms of mass of chemicals to be transported into the field per m3 of wastewater to be treated ) . It is important to note here that variance in the mass of hydrated lime used per unit of wastewater during the execution of Protocol B was much higher than was predicted by initial laboratory tests . This is probably the result of , not only variations in the characteristics of the wastewater between each batch , but also variations in the purity ( as percentage weight of CaO ) of successive batches of the lime obtained during the challenging circumstances encountered at the time of the study . Additionally , plant operation was undertaken in conjunction with operator training . During initial plant operation ( data not reported here ) operators were trained to prevent excessive use of reagents that was unnecessary to meet the treatment objectives . The residual levels of aluminum recorded in the treated effluent produced under Protocol B suggest that addition of this coagulant ( which is commonly used in drinking water treatment systems ) , is unlikely to represent a significant risk to the health of human populations living downstream of the treatment system [46] . The recorded average volume of produced sludge , at 6% ( vol . /vol . ) , was slightly higher than the values recorded in the literature for coagulation/flocculation using hydrated lime and aluminum sulfate [47] [48] . However , sludge volumes slightly in excess of those stated in the literature may be deemed acceptable for this kind of experimental field-work , especially given the practical constraints observed at the time these trials were undertaken . For example , during certain phases of the project , the removal of supernatant was found to have been performed under sub-optimal conditions . This was because it took up to one month to train personnel adequately so as to optimize the process and minimize the sludge volume . An additional study of the microbiological characteristics of the sludge , before and after its subsequent treatment by solar drying and prior to its incineration or controlled discharge to a protected soil infiltration pit is warranted in the future . This would need to be done using a multiple tube ( most probably number ) approach rather than the membrane filtration method available to the authors during this study . However , it is important to note that , although pathogens would have been concentrated in the sludge during the treatment process , the evidence suggests that the extreme pH levels to which they were subjected ( for an extended period of time ) would have resulted in a highly significant reduction in the concentration of viable organisms and that controlled soil infiltration , according to the protocols used elsewhere for untreated CTC wastewaters , constitutes a rational management of the risk of onward human infection . A relatively simple cost analysis demonstrated that labor costs per unit of treated fecal waste for Protocols A and B are roughly equivalent to those of the super-chlorination approach to disinfection , the efficacy of which has been questioned [21] [19] [22] [23] . Moreover , significant financial savings , in relation to reagent costs , may be achieved using the protocols presented here . Further details are provided in the Supporting Information files . In light of the recent findings of a panel of experts reporting to the United Nations , the research presented here is timely [49] . The report states that “[…] to prevent introduction of contamination into the local environment , United Nations installations worldwide should treat fecal waste using on-site systems that inactivate pathogens before disposal . These systems should be operated and maintained by trained , qualified […] staff or by local providers with adequate oversight […]” [50] . Although the authors of the report do not prescribe an appropriate microbiological quality standard that might be met by disinfection of the wastewater prior to discharge into the environment , it is interesting to note that concentrations of thermotolerant coliforms in the treated wastewater reported in the study reported here consistently met the WHO bacteriological guideline values for agricultural reuse , i . e . , fewer than 1 , 000 CFU/100 ml [51] . In fact , the quality of the final effluent achieved by both full-scale treatment protocols was consistently higher than the minimum standard initially agreed for disposal by direct infiltration . The high rate of disinfection achieved using both physico-chemical treatment protocols described here clearly suggests that this innovative technology may be an appropriate and potentially valuable option for the onsite-disinfection of CTC wastewaters generated in the emergency settings encountered during cholera epidemics and potentially may offer a valuable form of wastewater and human excreta disinfection during outbreaks of other infectious diseases . For example , although the Ebola virus is considered to be ‘fragile’ beyond the environment of bodily fluids ( including feces ) , its potential presence in large numbers in the feces of Ebola patients and its relatively low infective dose [52] [53] present a potent hazard to health care workers . The disinfection options presented here may be readily adapted to provide an important in situ excreta disinfection step as part of an integrated infection control framework . More accurate determination of the chemical consumption for both protocols is currently being achieved through laboratory experimentation , but a key finding of the field work reported here , that chemical consumption during the execution of the low pH treatment process ( Protocol B ) was significantly lower than that during the high pH process ( Protocol A ) , appears to be valid . This consideration potentially has significance for international medical organizations that may wish to use this technology during future disease outbreaks , especially in scenarios where reducing the quantity of chemicals , either purchased locally or imported , may be a high priority . The evidence available from the published literature suggests that the organism Vibrio cholerae is highly likely to respond to extreme levels of pH achieved in the protocols presented here in a similar manner to thermotolerant coliforms ( including Escherichia coli ) . However , a future investigation , which compares the behavior of Escherichia coli and other commonly used indicator bacteria ( such as intestinal enterococci ) , with that of Vibrio cholerae would probably provide valuable additional information to help optimize the treatment technologies presented here . The engineering priority now must be to monitor these treatment systems under more highly-controlled conditions , in order to refine the treatment processes and validate the data reported here , which were achieved under challenging field conditions . A longer-term challenge for microbial ecologists is to develop a better understanding of how toxigenic strains of V . cholerae and other excreta-borne pathogens behave in the environmental niches present in wastewater treatment plants [54] , but the technology outlined here may have broader application to scenarios in which hygienic management of sludges and wastewaters has to be achieved rapidly and at relatively low-cost . The authors are therefore currently exploring its application to other NTD outbreak settings and to the broader issue of urban excreta management in low-income communities . However , it is essential that those actively involved in WASH operational research should take a multi-disciplinary approach to the issue of controlling disease transmission from human excreta and should avoid the tendency to focus exclusively on infrastructural interventions [11]: those responsible for designing and operating new wastewater treatment technologies in emergency settings should always consider the broader and longer-term public health context of their interventions and should fully evaluate all new technologies within the rational risk management framework of ‘sanitation safety planning’ .
|
When an outbreak of infectious disease occurs in a low-resource setting , the rapid construction of emergency healthcare facilities may significantly reduce mortality . The facilities also result in the generation of large volumes of highly contaminated fecal waste that represents a potential basis for further disease transmission . Infection protection and control strategies at healthcare facilities must therefore include measures to establish and maintain good water supplies , sanitation and hygiene ( WASH ) . Even where the pathogen of concern is not waterborne , health-care providers have a ‘duty-of-care’ to protect workers and neighboring communities from all excreta-borne diseases . In this study , the authors successfully demonstrated , for the first time , the in situ disinfection of wastewaters from cholera treatment centers during the Haiti cholera outbreak , using a low-cost physicochemical method . The approach is currently being adapted by NGOs to help manage human excreta in other emergency settings , including the current Ebola outbreak . Although the Ebola virus is relatively fragile , it may exist in high concentrations in the bodily fluids ( including feces ) of those with the disease . The approach to in situ disinfection of excreta described here may therefore support infection control in outbreaks of Ebola and other infectious diseases .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"and",
"Discussion"
] |
[] |
2015
|
Minimizing the Risk of Disease Transmission in Emergency Settings: Novel In Situ Physico-Chemical Disinfection of Pathogen-Laden Hospital Wastewaters
|
The duplication of mammalian genomes is under the control of a spatiotemporal program that orchestrates the positioning and the timing of firing of replication origins . The molecular mechanisms coordinating the activation of about predicted origins remain poorly understood , partly due to the intrinsic rarity of replication bubbles , making it difficult to purify short nascent strands ( SNS ) . The precise identification of origins based on the high-throughput sequencing of SNS constitutes a new methodological challenge . We propose a new statistical method with a controlled resolution , adapted to the detection of replication origins from SNS data . We detected an average of 80 , 000 replication origins in different cell lines . To evaluate the consistency between different protocols , we compared SNS detections with bubble trapping detections . This comparison demonstrated a good agreement between genome-wide methods , with 65% of SNS-detected origins validated by bubble trapping , and 44% of bubble trapping origins validated by SNS origins , when compared at the same resolution . We investigated the interplay between the spatial and the temporal programs of replication at fine scales . We show that most of the origins detected in regions replicated in early S phase are shared by all the cell lines investigated whereas cell-type-specific origins tend to be replicated in late S phase . We shed a new light on the key role of CpG islands , by showing that 80% of the origins associated with CGIs are constitutive . Our results further show that at least 76% of CGIs are origins of replication . The analysis of associations with chromatin marks at different timing of cell division revealed new potential epigenetic regulators driving the spatiotemporal activity of replication origins . We highlight the potential role of H4K20me1 and H3K27me3 , the coupling of which is correlated with increased efficiency of replication origins , clearly identifying those marks as potential key regulators of replication origins .
The faithful duplication of mammalian genomes at each S phase is under the control of a spatiotemporal program that orchestrates and regulates both the positioning and the timing of firing of replication starting points also called replication origins . The molecular mechanisms involved in coordinating of the activation of 50 , 000 to 100 , 000 origins in each cell and at each cell cycle are still poorly understood , despite the need for a comprehensive understanding of these processes . Indeed , defects in the normal sequence of events leading to replication initiation may be directly responsible for genomic instability and/or the deregulation of differentiation programs . Consequently , the first and necessary step towards understanding this regulation is to refine our vision of the spatiotemporal replication program . For this reason several laboratories have chosen to map both the spatial and temporal programs of replication , in different systems and cell lines . The temporal program of replication has been successfully analyzed in many laboratories with no particular controversy . By contrast , attempts to identify replication origins remain a subject of passionate debate in the field , as the intrinsic rarity of replication bubbles makes it difficult to purify the genomic material . The most popular method for mapping replication starting points in mammals is the purification of short nascent strands ( SNS ) . Several laboratories have demonstrated that this purification requires the use of the -exonuclease to remove the high background due to broken genomic DNA [1]–[3] . The debate has been kept alive because previous studies that did not use -exonuclease [4] reported SNS levels incompatible with true initiation events [1] , [5] . The debates then turned to the putative lack of overlap between datasets from different laboratories using SNS enrichment with exonuclease purification , even for the same cell line , suggesting that the method was probably inaccurate . However this argument is almost entirely based on a comparison of two data-sets for the same cell line [1] , [6] , but the results of the first study [6] have repeatedly been shown to display a marked lack of overlap with those of other studies , whereas the results of the second [1] overlap significantly with those of other studies [7] , [8] . Agreement on a consensual protocol for SNS enrichment and quantification has also become a critical issue as the scale of investigation of replication origins has changed profoundly in recent years . Beginning with investigations of individual loci , and continuing with the microarray technology , there has recently been another technological shift in this field towards the use of ultra-deep sequencing [8] . Origin-omics has now become a way of thinking about replication that incorporates tens of thousands of loci embedded within various genomic landscapes . The emphasis also needs to shift from protocols to methods used for the analysis of genome-wide replication data . Indeed , despite a spectacular increase in the sensitivity of detection , Origin-omics is already subject to the same pitfalls as all other types of omics: the difficulty achieving an appropriate balance between the specificity and sensitivity of the analysis method . In a recent study based on the ultra-deep sequencing of SNS , origins were detected using chIP-Seq tools [9] for peak detection . This resulted in 250 , 000 identified origins in different human cell lines [10] . These predictions cover 6% of the human genome , with average origins length of 760 bp , that presumably includes most of the previously reported origins ( except those reported in the above mentioned study [6] , which should now reasonably be excluded ) . We noticed however one possible caveat in the use of chIP-Seq tools for the detection of replication origins based on sequenced SNS . Indeed , prior to the sequencing , SNS are first selected based on their size ( about 1 . 5–2 kb ) . Hence the resolution of detection of replication origins cannot be less than this size . It is therefore possible that chIP-Seq tools tend to split the signal into multiple peaks and hence tend to overestimate the number of replication origins . In this work , we first address this issue of resolution of detection . We propose a peak-detection method that is adapted to the special case of SNS sequencing data , based on the prior control of the resolution of detection of exceptional local enrichments of reads . The method relies on sliding windows , the size of which is imposed by the size of the sequenced SNS fragments . We deal with multiple testing by providing a significance threshold that controls for false-positive detections , and that is adaptive to local coverage variations . The consensus on the SNS purification protocol made it possible to apply our method to our samples ( K562 cells ) and to published data [10] ( from four different cell lines ) , which allows us to compare detection methods on SNS data . Origin-omics shares another difficulty with other omics fields , which is the need for validation of genome-wide detections by an independent method . Very interestingly , the field has recently been enriched by another genome-wide map of replication origins obtained by bubble trapping [11] , which is based on the sequencing of EcoR1 fragment containing at least one replication bubble . This new map consists of ∼125 , 000 EcoR1 fragments that cover 25% of the human genome . We took this opportunity to confront different genome-wide detections of replication origins based on different methods and protocols , which had never been done before . Comparisons between SNS-based origins and bubble trapping based origins on different cell lines show a good agreement between maps . Furthermore these comparisons indicate that the sensitivity and specificity of the detection of origins based on SNS data is significantly improved with our dedicated method compared to previously used chIP-Seq tools . Now that a consensus set of replication origins has been identified , the time has come to unravel the genomic and epigenetic characteristics that make these particular loci replication origins . To proceed we focus on the connections between the spatial and temporal programs of replication at fine scales . It is now well established that genomes are organized into early- , mid- and late-replicating domains , and early domains have been shown to be associated with active epigenetic marks such as H3K4me1 , 2 and 3 , H3K27ac , H3K36me3 and H3K9ac [12] . However , different studies have generated conflicting results , demonstrating the difficulties involved in precisely defining the chromatin landscape of the domains replicated in mid- and late S-phase . The first genome-wide studies showed that late-replicating domains were weakly correlated with the repressive mark H3K9me2 , but not with H3K27me3 [12] . This result conflicted with the finding of a previous study based on 1% of the human genome ( ENCODE regions ) , which reported a strong correlation between late replication and H3K27me3 [13] . Finally , an association of H3K27me3 with mid-S phase-replicating chromosomal domains was recently demonstrated , together with a substantial correlation with early-replicating domains [14] . These results also highlight the difficulties involved in assessing the impact of specific modifications on normal S-phase progression . We hypothesize that the imprecise mapping of origin positions has hampered the search for specific epigenetic signatures . In this study we integrated data collected in several genome-wide studies aiming to map DNA replication timing domains and chromatin states . We provide unique datasets including origin position , efficiency and timing , and the local genomic characteristics of each origin ( histone modifications , sequence characteristics ) . Overall , our findings make it possible to define new mechanisms potentially involved in defining of origin sub types activated sequentially during S phase .
OriSeq data analysis based on SNS material consists in detecting significant read enrichments corresponding to accumulations of SNS throughout the human genome . For a given origin , reads accumulate around the initiation starting point with a span determined by the size of the SNS fragments . It is important to notice that SNS are selected based on their size ( between 1 . 5–2 kb ) , and then fragmented and sequenced . hence , for a given origin the resolution of detection can not be smaller than 1 . 5–2 kb . Tools for the detection of peak-like patterns in ChIP-Seq data , such as SoleSearch [9] , [10] , have been used for detection purposes , without controlling for the size of the peak , which results in peaks smaller than 1 kb on average ( Table 1 ) . In our method we control the resolution of detection by considering sliding windows of size 2 kb . Then we define an appropriate statistical model for discriminating between signal and noise and controlling for false-positive peaks , while accounting for the genome-ordered structure of the data . We used scan statistics by calculating the probability that the richest window corresponded to a false positive [15] . This approach is designed to avoid false-positive detections , and was calibrated adaptively to coverage variations to account for coverage heterogeneities along the genome ( Figure 1-A ) . Using the scan method , we detected between 60 , 000 and 90 , 000 replication origins ( depending on read depth ) , which cover ∼12% of the genome ( Table 1 ) . Details are provided in the Methods Section . We first generated our SNS samples from K562 cell lines ( see Materials and Methods ) and we showed that the reproducibility of our detections was good , as of origins detected in one technical replicate are found in another ( on K562 cells , Supp . Table S1 ) , which actually corresponds to the technical reproducibility of origins detected by bubble trapping [11] . Then we investigated the quantitative properties of our analysis . In OriSeq data , which are obtained from populations of asynchronous cells , the number of reads for a detected origin reflects the percentage of cell cycles using this locus as a starting point for replication . The density of reads within an origin , therefore , constitutes a measurement of the efficiency of that origin . We assessed the precision of our method , by randomly selecting weak , intermediate and strong origins on the basis of read densities . We found that the number of reads detected for a given origin and the efficiency of that origin , as assessed by qPCR on an independent SNS preparation , were correlated ( , , Figure 1-B ) . These experimental validations confirm that the set of replication origins detected by our method is likely to correspond to true positive initiation events . To evaluate the reproducibility of replication origin detections , we compared our results with those of two previously published studies [10] , [11] . The first dataset was obtained with the same SNS purification protocol but in a different laboratory [10] , on different cell lines ( IMR-90 , HeLa , human embryonic stem cell H9 , induced pluripotent stem cells from IMR90 ( IPS ) ) . These datasets were comparable with ours , despite coverage differences ( Table 1 ) . In the original publication , detections were made using SoleSearch on these data [9] . For comparison we re-analyzed them using the scan method ( Table 1 ) . The second dataset corresponds to origins detected using bubble trapping , which is based on the sequencing of EcoR1 fragments containing at least one bubble ( on GM06990 cells ) . In this case detections were based on a background read depth distribution [11] . For the sake of simplicity , the three data sets will be referred to as SNS-scan , SNS-SoleS and Bubble origins . The three methods differ widely in the number of detected origins , with about 2 to 3 times more SNS-SoleS origins than Bubble and SNS-scan origins , even for data from the same cell lines ( Table 1 ) . SNS-SoleS origins are 760 bp long on average and cover ∼6% of the genome , whereas SNS-scan origins are longer ( ∼4 kb on average ) and cover ∼12% of the genome , and Bubble origins ( 6 . 4 kb long on average ) cover ∼25% of the genome ( Table 1 , Figure 1-C ) . The strong contrast between SNS-detected origins and Bubble origins reflects differences in the level of resolution of the methods: whereas SNS data allow the detection of origins at relatively high resolution ( based on 1 . 5–2 kb fragments ) , the resolution of bubble trapping experiments is limited by the genomic density in EcoR1 restriction sites . For SNS data , the main difference between our scan approach and the SoleSearch method is that this latter does not control the level of resolution and hence tends to detect many small peaks . To compare datasets we computed the proportion of origins of a given dataset that overlap with origins of another dataset . In a first step we compared SNS-scan origins with SNS-SoleS origins on 4 different cell lines ( Table 2 ) . On average 70% of SNS-scan origins overlap with SNS-SoleS origins ( on the same cell line ) and 58% of SNS-SoleS origins overlap with SNS-scan origins , compared with 3–6% expected by chance ( Table 2 , the randomization procedure being detailed in the Methods section ) . Visual inspections suggested that in many cases , SNS-SoleS origins corresponded to multiple small peaks located within a same SNS-scan origin ( Supp . Figure S1 ) . To account for this difference in resolution , we clustered neighboring SNS-SoleS results so that origins from both methods have the same length on average ( Table 2 , Supp . Figure S1 ) . By doing so , 71% SNS-scan origins overlap with ( clustered ) SNS-SoleS origins and 63% of SNS-SoleS origins overlap with SNS-scan origins . Thus , when compared at the same resolution , the overlap between methods is between 60 and 70% . Then we compared SNS origins with Bubble origins to assess the overlap between experimental protocols . Here 45–46% of SNS origins ( SNS-SoleS or SNS-scan ) overlap with Bubble origins and 36–37% of Bubble origins overlap with SNS origins ( vs . 5–7% expected by chance , Table 3 ) . Given the strong difference in resolution between the two methods ( Table 1 , Supp . Figure S1 ) , we repeated the comparison after having clustered SNS origins ( so that to obtain a resolution comparable to that of Bubble origins ( Table 3 ) . With this procedure , 65% ( 51% ) of SNS-scan ( SNS-SoleS ) origins overlap with Bubble origins and 44% ( 37% ) of Bubble origins overlap with SNS-scan ( SNS-SoleS ) origins , compared to 6–7% expected by chance ( Table 3 ) . We note that the cross-validation of SNS-detected origins by Bubble origin data is stronger when we used SNS-scan origins than SNS-SoleS origins , which suggests that the scan approach achieves a better sensitivity and specificity than SoleSearch . It should be noticed that the comparisons between replication origins detected by SNS or by bubble trapping were performed on different cell lines , and hence underestimate the true overlap between the different methods . The matter of resolution appears central in a fair comparison between datasets , and was partly assessed by considering clustered SNS origins . However the distributions of origins length still differ between methods , despite comparable on average ( Figure 1-D ) . Nevertheless , this study demonstrates a good agreement between SNS-based and bubble trapping-based replication origin maps , with at least 65% of SNS origins confirmed by bubble trapping . The first attempts to unravel the spatial program of replication showed that replication origins were associated with specific genomic features , such as high GC content , CpG islands ( CGIs ) [1] , [16] , [17] and G-quadruplexes [10] , [18] . Moreover , a recent study showed that origins could be divided into subgroups [10]: constitutive origins , which are common to all cell lines investigated; specific origins , which are found in only one cell type; and common origins , which are neither constitutive nor specific . This suggests that the same genomic and epigenomic constraints may apply to all constitutive origins ( and possibly to other types of origins too ) . We confirmed the strong overlap of datasets from different cell lines , as constitutive origins accounted for 57% , 35% , and 35% of origins in K562 , HeLa and IMR90 cell lines respectively , whereas specific origins accounted for only 15 , 13 , and 9% . We then showed that the clustering of origins previously described [10] was specific to constitutive origins , as the size distribution was skewed towards large sizes ( average sizes of 4 . 1 kb , 2 . 4 kb , 2 . 1 kb for constitutive , common and K562-Specific Oris , respectively , Supp . Figure S2 ) , indicating that constitutive origins may present the highest density of initiation events . We then connected the temporal and spatial programs of replication at fine scales , by assigning a temporal status to each origin on the basis of publicly available replication timing data [19]–[21] , as explained in the Methods section . This made it possible to distinguish between origins activated in early S-phase ( classes 1–2 ) and origins activated in mid- and late S-phase ( classes 3–4 and 5–6 , respectively ) . Most early origins ( 67% ) were found to be constitutive , whereas late origins tended to be more cell type-specific ( Figure 2-A for K562 cells , Supp . Figure S3-A for HeLa and IMR90 cells ) . This suggests that a large proportion of the origins replicated early in S phase are controlled by a highly robust combination of factors common to most cell lines . We then focused on the strong overlap between replication origins and CGIs [1] , [16] , [17] . We showed that this enrichment was not homogeneous throughout S phase , as 32 . 5% , 15% and 8% of early , mid and late origins overlapped with CGIs , this enrichment being significant whatever the timing of replication , in the three cell lines investigated ( Table 4 ) . This suggests that CGIs intrinsically favor initiation activities . In addition , 86% of the origins associated with CGIs were constitutive origins ( Figure 2-B , Supp . Figure S3-B ) . Read accumulation levels were much higher for CGI-constitutive origins ( Figure 2-C , Supp . Figure S3-C ) , consistent with the larger size of the constitutive-CGI origins ( Supp . Figure S2 ) . Moreover , when considering origin efficiency , corresponding to the total number of reads divided by the length of the origin , origins were found to be more efficient in early S phase ( as already reported [10] , [16] ) , but we found that origins were more efficient when associated with CGIs ( Figure 2-C , Supp . Figure S3-C ) . As most CGIs overlap with promoters and transcriptional regulatory elements , we also analyzed the distribution of origins with respect to transcription start sites ( TSSs ) . We also used maps of several chromatin states as a function of transcriptional activity [22] to evaluate the impact of transcription on the replication program . We found that 5 to 37% of origins were associated with a TSS in K562 cells , depending on the timing of replication ( Table 4 ) . Inactive poised promoters were poorly represented , whereas active and weak promoters were evenly distributed and significantly enriched in early S origins ( Table 5 ) . The association of many origins with weakly transcribed regions ( 39% of early origins and 19% of mid-S phase origins were found to be associated with chromHMM-11 , Table 5 ) is also consistent with previous studies [16] , [23] indicating that many initiation events take place within the body of genes . We also found associations between early origins and strong and weak poised enhancers ( Table 5 ) . These results confirm the early data obtained with small fractions of the human and mouse genomes [1] , [17] and indicate that the contribution of CGIs to the establishment of the spatiotemporal program of DNA replication extends to the whole genome , together with potential transcriptional regulatory elements . Moreover , the extension of the study to several cell lines showed that 42% of CGIs were very efficiently recognized as sites of replication initiation in all cell types , thus defining a very robust and efficient subset of replication origins with a tendency to fire early in S phase . The key role of CGIs was also highlighted by the clear increase in the percentage of CGIs overlapping at least one origin in one cell line with the number of cell lines investigated ( Figure 2-D ) . With the datasets for five cell lines used here , we were able to associate ∼76% of CGIs with at least one origin , but the trend clearly suggests that most CGIs are potential origins of replication . ORC1 binding sites were recently mapped genome-wide in HeLa cells [7] . ORC1 is a subunit of the origin recognition complex ( ORC ) used as a landing platform for the assembly of a cascade of components that together form the prereplicative complex ( pre-RC ) in G1 . Pre-RCs , including the core replicative helicase Mcm2–7 , are sequentially activated during S phase . It has now been established that more pre-RCs are formed in G1 than are actually required in S phase . This redundancy has the advantage of providing “dormant origins” , which may be used in a fail-safe mechanism activated in the vicinity of arrested replication forks , to restart replication [24] , [25] . Thus , this model predicts that more pre-RC ( ORC1 ) binding sites than initiation sites should be mapped . However , only 13 , 604 ORC1-enriched peaks were identified [7] , suggesting that the ORC1-ChIP experiment was not sensitive enough to detect every site of replication initiation and/or pre-RC formation . Nonetheless , the strong association with TSSs and known replication origins observed suggests that this new analysis identifies potential initiation sites ( even though restricted to a subset ) . We , therefore , studied the association of ORC1 binding sites with replication origins observed in HeLa cells , and found that 44% ( 5 , 957 ) of ORC1 peaks were located within replication initiation sites ( vs . 3% , as would be expected by chance ) . Given that the overlap between two anti-ORC1 ChIP-Seq replicates in HeLa cells was ∼60% , these results suggest that our SNS method is highly reliable for the determination of origin positions . The association of origins with ORC1 was stronger for constitutive origins ( 66% ( 3917 ) of ORC1 binding sites co-localize with constitutive origins , 30% and 4% for common and cell-specific origins respectively ) , consistent with the stronger Orc1-chIP signal for efficient origins ( Figure 3-A ) . Consistently , ORC1-bound origins were found to be enriched in CGIs ( 60% are CGIs , and CGI-origins account for 16% of all origins ) , supporting a central role for CGIs in replication . Two recent studies suggested that G-rich motifs capable of forming G-quadruplexes ( G4s ) are potential regulators of origin function [10] , [18] . These motifs are able to form four-stranded DNA structures with loops of different sizes . We showed that the association of G4 with origins was dynamic and dependent on the association with CGIs: the enrichment in CGI-origins was higher than expected and remained high for origins of all replication timing groups , whereas the enrichment of non-CGI origins in G4 was also much higher than expected but was lower for late-replicating origins ( Figure 3-B ) . We then assessed the impact of the size of the G4 loops on origin efficiency ( Figure 3-C ) . We showed that the efficiency of replication origins increased with the local density of G4s ( as measured by the number of G4s in a 5 kb window ) , and that G4s with short loops had a greater impact ( Figure 3-C ) . This result is in agreement with the observation that on one model origin two G4 motifs cooperate to drive initiation very efficiently [26] . We investigated the importance of G4s for origin selection further , by determining whether CGIs associated with origins ( Ori-CGIs ) displayed a higher level of enrichment in G4s than CGIs not associated with origins ( nonOri-CGIs ) . The human genome contains a total of 28 , 691 CGIs ( from the UCSC database-hg19 ) , 50% of which overlap with constitutive origins ( 76% overlapped with origins in at least one of the cell lines investigated ) . As mentioned above , constitutive origins are the most efficient and tend to fire in early S phase . We found that CGIs overlapping with constitutive origins displayed a greater enrichment in G4 L1–7 and L1–15 than nonOri-CGIs ( Figure 3-D ) . This result again strongly supports the hypothesis that G4 L1–7 plays an important role in the control of origin selection . Many studies have tried to decipher the roles of histone marks and nucleosomal organization in origin selection . However , our understanding of the complex relationships between chromatin states and replication has been limited by the scale of investigation , as all studies consider replication timing domains of 200 kb to 2 Mb , potentially resulting in a lack of resolution [27]–[30] . Moreover , origins of replication have been found embedded within many types of chromatin substrates [1] , [31] , [32] , suggesting that any regulatory effect of chromatin structure would not be homogeneous across replication initiation sites . This was confirmed by studies in mouse Embryonic stem cells ( ESCs ) and neural precursor cells ( NPCs ) , showing regions that replicate early to be enriched in open chromatin marks , such as H3K4me3 and H3K36me3 [33] , whereas little ( if any ) association was detected with other marks , such as H3K27me3 , H3K9me3 or H4K20me3 . Investigations of the methylation of H4K20 have provided new insight . Several studies have shown that PR-Set7 , which is involved in depositing the histone mark H4K20me1 , plays a role in the control of origin firing [34] . These findings are consistent with recent data indicating that Suv4-20h plays a crucial role in the further methylation of H4K20me1 [35] . Nevertheless the fraction of replication origins that really do carry this mark ( and are therefore potentially regulated by this modification ) remains unknown . Our study provides a unique framework for unraveling the connections between the fine-scale spatiotemporal program of replication and the landscape of chromatin modifications ( links to chromatin data are provided in Supp . Table S3 ) . The H4K20 monomethylation mark thought to control origin licensing has been shown to be associated with 50% of origins , this enrichment being significant for origins activated early or in mid-S ( Figure 4-A and Table 6 ) . The dynamic association of replication origins with open chromatin marks , such as H3K9ac , H3K4me3 and H2AZ , was strong ( and significant ) for origins replicated early in S phase , whereas origins activated in the second part of S phase were less associated with such marks ( Figure 4-A and Table 6 ) . We also found that these marks tended to be absent from late-activated origins , such as K562 cells ( Table 6 ) . Overall , 64% of origins carried none of these three open chromatin , indicating that most origins may not be directly driven by the presence of open chromatin marks , as previously proposed [1] , [32] . The association with heterochromatin marks has been reported to be negatively correlated with replication timing . We , therefore , also investigated two histone marks known to be enriched in facultative and constitutive heterochromatin . Early origins displayed a significant depletion of H3K9me3 , whereas late origins were characterized by a significant enrichment in this mark ( Figure 4-A and Table 6 ) . These results were confirmed by an independent study defining chromatin states ( Table 5 , HMM13 ) . By contrast , we found that origins activated early and in mid-S phase were enriched in H3K27me3 , which was thus associated with a large proportion of replication origins ( 40% ) ( Figure 4-A and Table 6 ) . The association of this mark , deposited by PRC2 complexes , is confirmed by the strong overlap between H3K27me3 and Ezh2 responsible for the deposition of this mark ( Table 6 ) . These results were also confirmed by an independent study in which the polycomb-repressed chromatin state was annotated ( Table 5 ) , although the overlap with replication origins was weaker in this case . We further focused on spatial interactions between marks that might characterize the temporal progression of replication . For each origin detected in K562 cells , we considered its linear distance to the closest mark , H2AZ , H4K20me1 , H3K27me3 , H3K9me3 , H3K9ac or H3K4me3 . We then used a discriminant analysis to identify combinations of chromatin marks that could discriminate ( and thus characterize ) early , mid- and late S-phase origins on the basis of their spatial co-localizations with replication origins . A complete description of the discriminant analysis is provided in the Methods section . A first combination of marks was characterized by the proximity of early origins to open chromatin marks ( H2AZ , H3K9ac and H3K4me3 ) and H4K20me1 . The distance between early origins and open marks increased with the progression of replication , whereas mid-S phase origins remain strongly associated with H4K20me1 . Mid-S phase origins were also characterized by a strong association with H3K27me3 , and the coupling of H4K20me1 and H3K27me3 with the exclusion of other marks constituted a strong characteristics of this category of origins . Finally the association with H3K9me3 was identified as characteristics of late origins , further from H4K20me1 and H3K27me3 . Once we had elucidated the spatiotemporal interactions between origins and histone modifications , we further investigated whether they were associated with functional effects such as efficiency , length and density ( Figure 4 ) . We first investigated the responses to separate associations , and then studied the effect of combinations of marks . The separate analysis identified H4K20me1 and H3K27me3 as potential regulators of the replication program . When associated with CGIs , origins carrying these marks were characterized by a higher efficiency and length ( Figure 4-B , C and Supp . Table S4 ) , suggesting that they were associated with a larger number of initiation events . Colocalization with H4K20me1 and H3K27me3 was also associated with a higher density of origins ( Figure 4-D , Supp . Table S5 ) . By contrast , when associated with open marks , origins were less dense ( Figure 4-D , Supp . Table S4 ) , but their efficiency and length were not affected ( Figure 4-B–C , Supp . Table S4 ) , the slight effect on origin length observed in early S-phase being due to a high proportion of origins carrying both open chromatin marks and H4K20me1 , as shown in Figure 5-A . We then characterized the functional responses associated with marks combinations that we identified for early , mid-S phase and late origins . We found that H4K20me1 and open chromatin marks co-localize on 38% of early origins and 16% of for mid S-phase origins , this proportion being increased for CGI origins to 64% of early and 48% of mid-S phase origins , ( Figure 5-A , Supp . Figure S4 for non-CGI origins ) . Moreover , H4K20me1 also colocalized with H3K27me3 , particularly in origins activated in mid-S phase ( Figure 5-A ) . These highly frequent colocalizations of marks were associated with different functional responses , as the coupling between H4K20me1 and H3K27me3 was the only combination to be associated with a significant increase in efficiency and density whatever the timing of replication ( Figure 5-B–C , Supp . Table S5 ) . The colocalization of H4K20me1 with open chromatin marks had very moderate additional effect over and above the separate effects of each mark ( Figure 5-B–C , Supp . Table S5 ) . The presence in ∼60% of origins of H4K20me1 or H3K27me3 ( or both ) , and the strong functional responses associated with the colocalization of these marks suggests their potential importance in the control of the human genome replication program .
Origins overlapping with a CGI tended to be more efficient than non-CGI origins and were more abundant than would be expected on the basis of chance among the origins active in early S phase . This constitutes a subclass of origins playing an important role in establishing the spatiotemporal program of DNA replication . One key issue to be resolved concerns the way in which origins of this type are regulated . We investigated the characteristics making CGI active origins , by focusing on the differences between CGIs associated with origins ( Ori-CGIs ) and CGIs that were not associated with an origin ( nonOri-CGIs ) . We found that Ori-CGIs were enriched in potential G4s L1–7 , suggesting that G4s might be important cis-regulators of origin activity . This hypothesis was also supported by the observation that nonCGI-Oris also overlapped strongly with G4s . However , not all G4s are origins , suggesting that G4s are therefore not sufficient to induce the formation of an efficient origin . We also performed genetic studies on one model origin , which confirmed that a structured G4 was important for origin activity and that this structured G4 had to cooperate with a 200 bp flanking cis-regulatory element to form a functional origin [26] . The cooperation with a flanking cis-module identified in one model origin added complexity to the origin signature , accounting for G4s not being systematically associated with origin function . We predict that the cooperating cis-module will act by binding transcription factors . Different classes of transcription factors may be involved , resulting in a complex signature motif for replication start sites . Taken together , our genome-wide and genetic studies and other published results [10] , [18] suggest that G4s can be considered consensus cis-regulatory elements for replication origins in vertebrates . Further studies should search for trans-factors capable of recognizing structured G4s and , through this function , regulating origin function . We also deciphered the epigenetic characteristics of the temporal program of replication , providing new insight to improve our understanding of the spatiotemporal regulation of origins . Our work provides the first genome-wide demonstration of the strong association between early-firing origins and open chromatin marks . Our study was mainly based on origins detected in K562 cells , but we also provide similar analysis on HeLa cells ( Supp . Figures S5 , S6 , S7 , S8 , Supp . Table S6 ) . Early-replicated origins are enriched in open chromatin marks ( they have more such marks than origins of other timing categories ) , consistent with the findings of previous genetic studies showing that the deposition of open chromatin marks close to replication origins can impose early firing in vertebrates [36] , [37] . Moreover , a recent study showed that a strong replication origin lying within a region that is naturally replicated in late S phase may be induced to replicate earlier in S phase by the presence of binding sites for the USF transcription factor [37] . This shift is local , because replicons located 50 kb away are not affected , and it is associated with the appearance of open chromatin marks at the shifted origin . We also found that early origins are enriched in binding sites for transcription factors known to recruit open chromatin marks , including USF ( data not shown ) . Overall , the results of genetic and genome-wide studies suggest that the deposition of open chromatin marks may be an important pathway for the regulation of early firing in vertebrates . We also found that early-replicated origins were the most efficient and that most were constitutive . We suggest that the construction of large early-replicated domains is dependent on the overlap of very efficient origins ( CGI origins ) and the recruitment of open chromatin marks by transcriptional regulatory factors , the binding sites of which are highly abundant in these domains . We identified new associations with combinations of chromatin marks for origins replicated in mid-S phase regions , corresponding to the colocalization of the polycomb mark H3K27me3 with H4K20me1 . A positive correlation between the PcG-mediated H3K27me3 mark and late replication has been reported in Drosophila [38] , but no correlation has yet been established in mammalian cells , with the exception of one study that explored only 1% of the human genome , the findings of which conflicted with those of other genome-wide analyses [13] . However an association of H3K27me3 with mid- S phase-replicating chromosomal domains was recently identified , although a substantial correlation with early-replicating domains was also described [14] . Likewise , a recent study demonstrated a direct role for Pc-G proteins in the regulation of late replication in Drosophila [39] , and studies have provided strong support suggesting that this mark is important for the control of DNA replication particularly in mid-S phase [40] , [41] . One study showed that Pc-G-mediated chromatin assembly occurs during the post-mitotic G1 phase in human cells and that the depletion of Suz12 ( the essential non catalytic subunit of the enzyme responsible for the trimethylation of H3K27 ) in G1 impairs the progression of cells in the following S phase and , particularly , in mid/late S phase [40] . In another study on mouse embryos , the depletion of components of the PRC1 complex ( Ring1 and Rnf2 ) , which recognizes H3K27me3 , was shown to block DNA synthesis in most two-cell embryos . Based on the appearance of H2AX foci in two-cell embryos , the authors also concluded that most depleted embryos did not finish S phase , suggesting their arrest in S phase [41] . Our study is the first to demonstrate a strong genome-wide association of H3K27me3 with replication origins activated in mid-S phase , suggesting that this mark is important for the control of DNA replication in mid-S phase in vertebrates . Moreover , we found that origins associated with this mark were generally more efficient and were embedded in regions with a higher density of replication origins ( Figure 4-B ) , suggesting a regulatory role of this mark in origin selection . Further studies should focus on the local effect of this mark on origin firing . Our results also highlight a potential key role of the H4K20me1 chromatin mark . It has already been suggested that the histone H4 Lys 20 methyltransferase PR-Set7 regulates replication origins in mammalian cells , based on the observations that 1 ) the onset of licensing coincides with an increase in H4K20me1 at known replication origins , and 2 ) PR-Set7 is normally degraded in S phase and the PR-Set7 mutant insensitive to this degradation displays the maintenance of H4K20me1 at replication origins and repeated DNA replication [34] . It has recently been shown that the function of PR-Set7 is dependent on the further methylation of H4K20me1 by Suv4-20h [35] . Thus , the regulation and timing of H4K20me1/2/3 is critical for the accurate regulation of origin firing . H4K20me1 mark deposition is the primary and necessary event leading to the trimethylated state . We investigated the statistical association of this monomethylation with replication origins and found a very strong coincidence of this mark with origins , suggesting that many replication origins may have the potential to be controlled by this modification . We also showed that origins carrying this mark were associated with increased efficiency and were located in regions with a higher density of potential origins . The next step in our investigations of this regulation will be the mapping of H4K20me1/2 and 3 , genome-wide , during the different phases of the cell cycle crucial for origin preparation , from early G1 to late S phase . This work will provide insight into the relationship between the dynamics of H4K20 methylation and origin function .
Short nascent strands were purified as previously described [32] , but with minor changes to the protocol . We pooled fractions 18 to 24 . These fractions contained single-stranded DNA molecules of various sizes , from 1 . 5 to 2 . 5 kb . We used 500 U of a custom-made -exonuclease ( Fermentas ( 50 U . l-1 ) ) for each preparation . For the genome-wide mapping of origins , eight SNS preparations were obtained independently from cells each and then pooled . SNS were made double-stranded by random priming with the Klenow exo-polymerase ( EP0421 , Fermentas ) and random primers ( 48190011 , Invitrogen ) . Adjacent strands were then ligated with Taq DNA Ligase ( M0208L , Biolabs ) . Two libraries were constructed with Illumina protocols and five deep sequencing runs were performed with a Solexa/Illumina GA I genome analyzer generating 75 bp reads . SOAP ( v2 ) software was used to map reads to the reference human hg19 genome with the following parameters: r:0 , I:30 and v:5 command-line . Data were deposited in the Gene Expression Omnibus ( http://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? token=pzexhssiceikczw&acc=GSE46189 ) . According to our detection model , read occurrences throughout the genome follow a Poisson distribution with a heterogeneous intensity that can be interpreted as the coverage process . We also assume that at a given position t along the genome , the number of reads follows a geometric distribution . We then consider , which counts the number of reads along the genome and , to detect local exceptional read accumulations , we compute , which quantifies the number of reads within a window of size u = 2 kb . For calculations of the the significance threshold for detection , we used scanning statistic results for compound distributions , making it possible to calculate the probability of the richest window actually being a false positive [15] . The detection is performed at level by setting threshold such that . To account for coverage heterogeneities , we segment the coverage process ( ) into regions of constant intensities ( constant ) . We use a segmentation model for this purpose , based on the Poisson distribution adapted from segmentation models for array CGH data analysis [42] . This segmentation step has two main advantages: First , it automatically detects regions of constant coverage ( constant ) and regions with extremely low coverage that are excluded from the study . Second , it allows our significance thresholds to adapt to coverage variations . An example of detection is provided in Figure 1-A . Our method is available at http://pbil . univ-lyon1 . fr/members/fpicard/research . html . Threshold was calibrated using independent input DNA from public databases since input DNA was not available at the time of the experiment . We applied the detection method to input DNA and we assessed the percentage of nucleotides detected as origins in the SNS data that were also detected as peaks in the input DNA data ( Supp . Table S2 ) . We chose which corresponds to an estimated false discovery rate of 4% , 10% and 18% for K562 , IMR90 , and HeLa cells . This constitutes an overestimation of the false discovery rate of detection since origins detected in peak-assigned regions of the input-DNA are not necessarily false positives . TSS , CpG islands , and chromatin mark positions were downloaded from the UCSC Genome Browser ( http://genome . ucsc . edu/ ) . Details on the datasets are provided in Supp . Table S3 . The positions of G-quadruplexes were determined by applying Quadparser on hg19 [43] , specifying the length of the spacer between 4 tracks of GGG or CCC with spacer of size 1–7 , 1–15 , 1–20 , and 1–30 . All results and tables can be downloaded from http://pbil . univ-lyon1 . fr/members/fpicard/research . html . We determined the mean replication timing profiles throughout the complete human genome from Repli-Seq data [19] , [21] , as previously described [44] . Repli-Seq tags for six FACS fractions were downloaded from the NCBI SRA website ( study accession number: SPR0013933 ) for the erythroid K562 cell line , and from the UCSC ENCODE website http://hgdownload . cse . ucsc . edu/goldenPath/hg19/encodeDCC/wgEncodeUwRepliSeq/for the IMR90 fetal lung fibroblast cell line . For the HeLa cell line , we calculated the mean replication timing ( MRT ) rather than the S50 ( median replication timing ) [19] , [20] . Timing categories were determined by dividing timing values into 6 intervals ( , , early origins , , , mid-S origins , , , late origins ) . We used a randomization procedure to assess the expected overlap between origins of replications detected by SoleSearch , scan , and by bubble trapping . To compute the expected overlap between SNS-SoleS and SNS-scan origins ( SoleS in scan , Table 2 ) , we randomly sampled genomic intervals on the mappable fraction of the human genome that were excluding SNS-SoleS origins . The number of sampled intervals was the same as the number of SNS-SoleS origins , and we sampled 50 sets of such random origins . The overlap of SNS-scan with sampled intervals was used to assess the expected SoleS in scan overlap . To assess the scan in SoleS overlap , genomic intervals excluding SNS-scan origins were sampled . The procedure was similar to compute the expected overlap between Bubble and SNS origins . We also used a randomization procedure to determine whether the association of replication origins with genomic features ( such as CGIs , chromatin marks , Gquadruplex motifs ) was significantly more frequent than would be expected by chance alone . For a given cell line , we compared the observed proportion of replication origins overlapping a given genomic feature with the expected proportion calculated from genomic intervals randomly sampled from throughout the genome . We excluded the replication origins we detected and the non mappable regions of the human genome as provided by the 1 , 000 Genomes Project [45] , and we sampled 100 , 000 random intervals of the same length as origins of replication , on average . This procedure was repeated 1 , 000 times , to account for different sequence characteristics ( such as gene density , or GC content ) . When replication timing was considered , the timing of replication for randomly sampled origins was determined from published timing data [21] . Random intervals were considered to be associated with genomic features if their intervals overlapped . We consider a linear discriminant analysis to find a linear combination of chromatin features which characterize early , mid-S phase and late origins [46] . We consider the data matrix with rows corresponding to origins detected in K562 cells and columns corresponding to linear distances to chromatin marks ( datasets links are provided in Supp . Table S3 ) . The interpretation of a discriminant analysis is based on two key ingredients ( similarly to Principal Component Analysis ) : the position of the origins on the discriminant axis ( Figure 6-A ) and the correlations of the chromatin features with the discriminant axis ( Figure 6-B , C ) . DA1 is the discriminant axis best discriminating between timing categories . It comprises the distances of origins to open chromatin marks ( negative correlation with distances to H2AZ , H3k9ac , H3k4me3 ) and to H4k20me1/H3K27me3 ( Table 7 ) . This indicates that the temporal decrease observed along DA-1 ( Figure 6-A ) corresponded to an increase in the distance to these marks . Consequently , the first combination of chromatin marks that emerged was the proximity of open chromatin marks and H4K20me1 to early origins . The second axis ( DA2 ) illustrates the opposition between H3K9me3 and other marks , and shows a different pattern between mid and late origins . Mid origins have a lower coordinate on DA2 ( Figure 6-A ) , which corresponds to a smaller distance to H3K27me3/H4K20me1 . DA2 was controlled by a positive correlation with the distance to H3K27me3/H4K20me1 ( and to a lesser extent to H2AZ ) , along with a strong negative correlation with the distance to H3K9me3 ( Table 7 ) . Thus the proximity of origins to H3K27me3 and H4K20me1 emerged as a marks combination for mid-S phase origins . Finally late origins have a higher coordinate on DA2 ( Figure 6-A ) , which corresponds to a smaller distance to H3K9me3 .
|
Replication is the mechanism by which genomes are duplicated into two exact copies . Genomic stability is under the control of a spatiotemporal program that orchestrates both the positioning and the timing of firing of about 50 , 000 replication starting points , also called replication origins . Replication bubbles found at origins have been very difficult to map due to their short lifespan . Moreover , with the flood of data characterizing new sequencing technologies , the precise statistical analysis of replication data has become an additional challenge . We propose a new method to map replication origins on the human genome , and we assess the reliability of our finding using experimental validation and comparison with origins maps obtained by bubble trapping . This fine mapping then allowed us to identify potential regulators of the replication dynamics . Our study highlights the key role of CpG Islands and identifies new potential epigenetic regulators ( methylation of lysine 4 on histone H4 , and tri-methylation of lysine 27 on histone H3 ) whose coupling is correlated with an increase in the efficiency of replication origins , suggesting those marks as potential key regulators of replication . Overall , our study defines new potentially important pathways that might regulate the sequential firing of origins during genome duplication .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry",
"genomics",
"cell",
"biology",
"dna",
"replication",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"dna",
"epigenetics",
"computational",
"biology",
"molecular",
"cell",
"biology"
] |
2014
|
The Spatiotemporal Program of DNA Replication Is Associated with Specific Combinations of Chromatin Marks in Human Cells
|
Organ development is directed by selector gene networks . Eye development in the fruit fly Drosophila melanogaster is driven by the highly conserved selector gene network referred to as the “retinal determination gene network , ” composed of approximately 20 factors , whose core comprises twin of eyeless ( toy ) , eyeless ( ey ) , sine oculis ( so ) , dachshund ( dac ) , and eyes absent ( eya ) . These genes encode transcriptional regulators that are each necessary for normal eye development , and sufficient to direct ectopic eye development when misexpressed . While it is well documented that the downstream genes so , eya , and dac are necessary not only during early growth and determination stages but also during the differentiation phase of retinal development , it remains unknown how the retinal determination gene network terminates its functions in determination and begins to promote differentiation . Here , we identify a switch in the regulation of ey by the downstream retinal determination genes , which is essential for the transition from determination to differentiation . We found that central to the transition is a switch from positive regulation of ey transcription to negative regulation and that both types of regulation require so . Our results suggest a model in which the retinal determination gene network is rewired to end the growth and determination stage of eye development and trigger terminal differentiation . We conclude that changes in the regulatory relationships among members of the retinal determination gene network are a driving force for key transitions in retinal development .
During organogenesis , cells undergo progressive cell fate restriction coupled with a loss of pluripotency . This process is hallmarked by the stages of specification , proliferation , and differentiation [1] . The transitions between each of these states mark major changes in developmental competence and plasticity during tissue and organ development . The adult fly eye develops from a larval structure called the eye imaginal disc [2] , [3] . Following specification and growth during early larval development , the retinal field begins to differentiate during the third larval stage , or instar [4] . Drosophila eye differentiation occurs progressively , proceeding from the posterior to the anterior margins of the disc; its progress is marked by a morphologically and molecularly detectable event called the morphogenetic furrow [5]–[7] . Anterior to the morphogenetic furrow , cells are determined and proliferating , while posterior to it cells exit the cell cycle and differentiate . Within the morphogenetic furrow , cells transition from proliferation to differentiation . Thus , the developing Drosophila eye is an ideal system to study how cells regulate the transition from pluripotency to terminal differentiation . Selector genes direct the development of many organs from their primordia [8] . The development of the eye imaginal disc into the adult eye is directed by a conserved network of transcriptional regulators called the retinal determination ( RD ) gene network . The core members of this network , twin of eyeless ( toy ) , eyeless ( ey ) , sine oculis ( so ) , eyes absent ( eya ) , and dachshund ( dac ) , are each necessary for normal eye development and are sufficient to drive ectopic eye development in other imaginal discs [9]–[17] . During normal development , Toy activates ey expression in the first instar [17] . Initially , Ey is expressed throughout the disc and activates the expression of eya , so , and dac [18]–[21] . Once established , So maintains its own expression , as well as that of dac and ey [19] , [22] . Such positive feedback mechanisms within the network are well characterized [17]–[19] , [23]–[25] . The downstream RD network members Eya , So , and Dac are expressed and necessary in cells posterior to the morphogenetic furrow ( Figure 1B–D ) [9]–[13] , [22] , [26] . In contrast , at the morphogenetic furrow , ey expression is sharply down-regulated ( Figure 1A ) , but how the positive feedback loops are terminated remains unknown [14] , [18] . In the region just anterior to the morphogenetic furrow where Dac , Eya , So , and Ey overlap , these proteins cooperate to initiate the expression of low levels of the proneural gene atonal ( ato ) , which is required for the onset of photoreceptor differentiation [27]–[29] . However , without further amplification and refinement by Notch signaling in the morphogenetic furrow , the low level of Ato expression induced in this region of the eye is not sufficient to induce photoreceptor differentiation , and Ey expression persists [30]–[32] . Thus , while RD gene activity is required to initially activate one of the most upstream genes required for the onset of differentiation , this is not sufficient to fully trigger differentiation . In this work , we show that maintaining expression of ey posterior to the morphogenetic furrow blocks photoreceptor differentiation . In addition , we identify a key regulatory switch in the RD gene network required for the repression of ey . Specifically , So directly regulates ey anterior to the furrow to promote high levels of expression , and via the same enhancer binding site blocks high levels of ey expression posterior to the furrow . Our results support a model that ey expression posterior to the furrow is regulated indirectly by eya and dac expression , and is triggered by signaling events in the morphogenetic furrow . These results suggest a model in which rewiring of the RD gene network is a key driving force during retinal organogenesis .
During the third instar , Eyeless ( Ey ) is strongly expressed anterior to the morphogenetic furrow . However , its expression sharply decreases at the morphogenetic furrow , and is detected only weakly in the differentiating eye field ( Figure 1A ) . In contrast , the downstream RD gene network members are expressed not only in undifferentiated cells anterior to the morphogenetic furrow , but also in differentiating cells posterior to the morphogenetic furrow ( Figure 1B–D ) . To determine if reducing Ey expression at the morphogenetic furrow is important for normal eye development , we overexpressed Ey posterior to the furrow using two methods . First , using the Flipout-Gal4 system we generated clones of cells that maintained Ey expression beyond the passage of the furrow [12] , [33] . This caused cells to fail to differentiate , as assayed by expression of the pan-neuronal marker ELAV ( Figure 1E ) . Second , we reactivated Ey expression in cells posterior to the furrow using the GMR-Gal4 and lz-Gal4 drivers [34] , [35] . GMR-Gal4 eventually drives expression in all cells posterior to the furrow , while lz-Gal4 drives expression in cells that generate the future photoreceptors R1 , 6 , and 7 as well as in the cone and pigment cell precursors . ELAV expression is not affected in these genotypes , suggesting that Ey is not sufficient to block differentiation once differentiation has begun ( Figure S1A–C ) . However , adult eyes of lz-Gal4; UAS-ey show defects in ommatidial shape and pigment when compared to wild-type ( Figure S1D , E ) . Sections through lz-Gal4; UAS-ey eyes showed that photoreceptors survive , but that rhabdomere morphogenesis and ommatidial rotation are abnormal , suggesting that terminal differentiation events are disrupted by ectopic Ey expression ( Figure S1F , G ) . From these results we conclude that down-regulation of Ey expression is necessary for normal photoreceptor differentiation . To identify how the change in Ey expression is regulated , we undertook a candidate gene approach based on the literature . Previous studies of the RD gene network member Sine oculis ( So ) indicate that So activates ey expression during the third instar; however , so loss-of-function clones posterior to the morphogenetic furrow contained Ey expression , suggesting either that So is also required to suppress Ey expression or alternatively that these cells are trapped in an earlier developmental state [18] , [22] , [36] . This apparent paradox in the literature led us to examine Ey expression in so3 null clones in different positions of the eye disc during the third instar . In so3 clones anterior to the morphogenetic furrow , Ey expression was reduced , supportive of the model that So positively regulates ey expression anterior to the furrow ( Figure 2A , arrow ) [22] . Posterior to the morphogenetic furrow , we observed strong Ey expression in so3 clones ( Figure 2A ) [36] . We conclude that So promotes Ey expression anterior to the furrow and suppresses Ey expression posterior to the furrow . We investigated the non-uniform appearance of Ey expression in posterior so3 clones , and observed that it is due to the morphology of the clones ( Figure 2A ) . Specifically , orthogonal sections through clones displayed a spherical shape , with Ey expression being restricted to the so mutant tissue ( Figure S2A , B ) . To determine if these cells lie in the interior of the clones that express low levels of Ey or no Ey , we co-labeled so3 clones for both Ey and Lamin , a marker of the nuclear membrane . We observed spaces within the clones that lack nuclei , and these spaces lack Ey ( Figure S2C ) . Therefore , we conclude that Ey is robustly expressed cell autonomously in all so mutant cells posterior to the furrow . Our clonal analyses suggest that So cell autonomously promotes Ey expression anterior to the morphogenetic furrow , and suppresses Ey expression posterior to the morphogenetic furrow . The presence of ey transcript or protein in so loss-of-function clones posterior to the morphogenetic furrow has been interpreted previously as a secondary consequence of failed furrow progression and/or differentiation [26] , [36] . However , it may be that so expression is required posterior to the morphogenetic furrow to negatively regulate Ey . To distinguish between these models , we let Ey undergo normal regulation anterior to and within the morphogenetic furrow and then knocked down so expression specifically in differentiating cells posterior to the morphogenetic furrow . The F2-Gal4 driver , generated by our group with a characterized enhancer of the sens gene [37] , initiates expression in the intermediate clusters within the furrow , posterior to Ey negative regulation , and is ultimately refined to drive expression most strongly in the R8 photoreceptor ( Figure S2D , F ) . This driver permits analysis of the role of so in Ey regulation specifically in differentiating cells . Additionally , changes in expression are easily detectable because normal cells surround the knockdown cells . In F2-Gal4>UAS-so-RNAi discs , we observed Ey expression posterior to the morphogenetic furrow in an R8-like pattern ( Figure 2B ) . Knockdown of So in F2-Gal4>UAS-so-RNAi discs is supported by So staining ( Figure S2E ) and results in a mildly disorganized adult eye ( Figure S2G ) . Based on these results , we conclude that so is required to suppress Ey expression posterior to the morphogenetic furrow and that such suppression is required for normal eye development . So is a homeodomain transcription factor , leading us to ask if So suppresses ey expression at the transcriptional level . To test this , we required a reporter that recapitulates ey regulation anterior and posterior to the morphogenetic furrow . Published ey enhancer reporters [22] , [38] , unlike Ey expression , persist posterior to the morphogenetic furrow , possibly due to perdurance of beta-galactosidase . We therefore constructed a new destabilized GFP ( dGFP ) reporter . To compare wild-type and mutant constructs while avoiding position effects , we utilized a vector that could integrate only at specific sites in our analysis [39]–[41] . We cloned a previously characterized full-length eye enhancer from the ey locus into this new dGFP vector , “ey-dGFP” [38] , [39] . We detected robust expression with ey-dGFP throughout larval development ( Figure 3A–C , Figure S3A–C ) . Similar to ey expression , ey-dGFP is expressed throughout the eye disc in first instar ( not shown ) and is maintained throughout the eye disc until furrow initiation ( Figure 3A ) . During the third instar ey-dGFP is maintained anterior to the morphogenetic furrow and suppressed at the morphogenetic furrow , similar to Ey expression ( Figure 3B ) . This expression pattern is maintained throughout the third instar ( Figure 3C ) . Therefore , this enhancer recapitulates the Ey expression pattern in the eye field . To determine if ey-dGFP can be regulated by So , we generated so3 clones and assayed reporter expression in clones anterior and posterior to the furrow . As with Ey , ey-dGFP reporter expression was reduced in anterior so3 clones , while it was induced in posterior clones ( Figure 3G–H ) . Based on these results , we conclude that So regulates ey expression at the transcriptional level both anterior and posterior to the morphogenetic furrow . To determine if So can regulate the expression of ey-dGFP directly , we mutated a previously well-characterized So binding site in the ey enhancer to generate eymut-dGFP [22] . From early development through initiation of the morphogenetic furrow eymut-dGFP is indistinguishable from ey-dGFP , consistent with published data that early ey expression is independent of So ( Figure 3D , Figure S3D ) [18] . However , during furrow progression , the expression pattern of eymut-dGFP is dynamic . The expression of eymut-dGFP anterior to the morphogenetic furrow is initially strong but weakens throughout the third instar , and eventually becomes barely detectable ( Figure 3E , F , Figure S3E , F ) . This may indicate that additional positive regulators of ey are initially expressed in this domain , consistent with findings that Tsh promotes Ey expression in the same region [42] , [43] . This is also consistent with our observation that Ey expression is diminished but not lost in the anterior so3 clones we observed ( Figure 2A ) . By the time the furrow has progressed 7–8 columns , eymut-dGFP expression is detected posterior to the onset of Sens expression in the furrow . By 14 columns of photoreceptor recruitment , eymut-dGFP is expressed in most cells posterior to the morphogenetic furrow ( Figure 3E , Figure S3E shows a disc at 11 columns ) . Posterior expression is detected weakly even in very late discs where anterior expression is lost ( Figure 3F , Figure S3F shows a disc of 18 columns ) suggesting that the So binding site is required posterior to the furrow to suppress activation of ey by another activator . We conclude that a So binding site is required to suppress expression of the ey enhancer reporter posterior to the furrow and to maintain reporter expression anterior to the furrow . To determine if So can regulate eymut-dGFP expression , we examined eymut-dGFP expression in so3 clones . If mutation of the binding site is sufficient to make the reporter unresponsive to regulation by So , then we should not observe changes in the reporter expression pattern when we compare tissue within versus outside of clones . We chose to assay a time point early in furrow progression when the reporter is still expressed anterior to the furrow and is beginning to express posterior to it . We observed areas of identical reporter brightness both inside and outside of the clones , leading us to conclude that mutation of the binding site makes the reporter unresponsive to regulation by So ( Figure 3I ) . Together with the fact that this binding site has been demonstrated to be bound by So in vitro [22] , our analyses of ey-dGFP and eymut-dGFP lead us to conclude that So directly regulates the expression of Ey both anterior and posterior to the morphogenetic furrow through the same binding site . We next wanted to investigate the mechanism by which So represses Ey posterior to the furrow . Sine oculis interacts with multiple cofactors that affect its function as a transcriptional regulator , including the transcriptional coactivator Eyes absent ( Eya ) and the TLE family corepressor Groucho ( Gro ) [25] , [36] , [44]–[46] . As both cofactors are expressed in the eye disc , we set out to determine which of them , if either , cooperates with So to regulate Ey . We performed loss-of-function analyses for each cofactor and assayed the effects on Ey expression in clones . Our primary candidate was Gro , which cooperates with So in the repression of targets in the eye [25] , [46] . Surprisingly , null loss-of-function clones of gro had no effect on Ey expression anterior or posterior to the morphogenetic furrow ( Figure S4A ) . We conclude that Gro is not necessary for the normal regulation of Ey expression during the third instar , and unlikely to cooperate with So in this process . We next wanted to determine if Eya cooperates with So to regulate ey . Previous studies found that So and Eya physically interact to promote the activation of target genes [28] , [36] , [45] , [47] , [48] . Based on these studies , we predicted that eya would be necessary for the maintenance of Ey expression by So anterior to the morphogenetic furrow . To test this , we generated eya null clones and examined Ey expression . We observed , surprisingly , that Ey expression was normal in eya anterior clones ( Figure 4A ) . As these clones were small and rare , we also used RNAi to knockdown eya expression using the Flipout-Gal4 technique . Even in large knock-down clones we observed that Ey expression was normal in clones anterior to the morphogenetic furrow ( Figure 4B ) . These results indicate that Eya is not required to maintain Ey expression anterior to the furrow . Posterior to the furrow , both null and RNAi knockdown clones of eya expressed Ey strongly ( Figure 4B , C ) . We also observed similar morphology changes in eya clones as in so clones posterior to the furrow ( compare Figure 4C to Figure 2D ) . Based on these results , we conclude that eya expression is required for Ey suppression posterior to the furrow . Eya is necessary for furrow progression and differentiation; therefore , failure of morphogenetic furrow progression through eya clones could result in the maintenance of Ey in these clones [26] , [36] , [49]–[51] . To test if Ey expression in posterior eya clones is an indirect effect of failed furrow progression , we used the F2-Gal4 driver to knock down eya expression specifically posterior to the furrow . We observed Ey expression in eya knockdown cells ( Figure 4D ) . Staining for Eya indicates that the RNAi effectively knocks down eya expression ( Figure S5A ) . Adults of F2-Gal4>eyaRNAi have disorganized eyes ( Figure S5B ) . We conclude that Eya is required for Ey suppression posterior to the furrow . To determine if eya is required for Ey suppression at the transcriptional level and dependent upon the So binding site , we examined ey-dGFP and eymut-dGFP expression in posterior eya clones . In clones posterior to the furrow , ey-dGFP was expressed , similar to so clone phenotypes , suggesting that eya is required for the negative regulation of ey at the transcriptional level ( Figure 4E ) . In contrast to ey-dGFP , the expression of eymut-dGFP is not induced in posterior eya clones , suggesting that it no longer requires eya for its regulation ( Figure 4F ) . From these results we conclude that Eya regulation of ey requires the So binding site . Eya and So each overlap Ey expression just anterior to the morphogenetic furrow , but do not negatively regulate Ey expression in this zone . Therefore , we re-examined the expression of Eya and So in the eye imaginal disc to determine if their expression patterns could suggest how Eya and So could be required to suppress ey expression posterior to the furrow . Quantification of Eya and So expression in orthogonal sections revealed that expression of both factors is increased posterior to the morphogenetic furrow ( Figure 5A , B ) . To test if the increased level is sufficient to repress Ey , we overexpressed both eya and so within the Ey domain using the Flipout-Gal4 strategy . Co-misexpression of eya and so was sufficient to repress Ey expression to background levels within the eye field , while ectopic Ey expression was detected in clones in other discs ( Figure 5C , and data not shown ) . These data suggest that , within the developing retinal field , increased so and eya expression is sufficient to repress Ey expression anterior to the morphogenetic furrow . When we utilized the temperature sensitivity of the Gal4-UAS system to overexpress eya+so at 18°C , which results in lower expression of eya+so than at 25°C , they failed to repress Ey expression in the eye field , but were still sufficient to ectopically activate Ey expression in the antennal disc ( Figure 5D , Figure S6A , white arrow ) . The levels of So and Eya expression increase posterior to the morphogenetic furrow in response to activation of the Hedgehog ( Hh ) and Decapentaplegic ( Dpp ) signaling pathways [26] . Next , we asked if upregulation of Eya and So is sufficient to suppress ey even without the signaling pathways normally required for morphogenetic furrow movement . To test this , we made use of the MARCM system . We overexpressed eya and so simultaneously in smo3 , mad1–2 double mutant clones , which cannot respond to either Hh or Dpp signaling . Clones doubly mutant for these two signaling effectors are known to lack furrow progression: they do not activate Notch signaling , they lack differentiation , and they retain Ey expression [5] , [26] , [49] , [52]–[54] ( Figure 5E ) . We observed that Ey is strongly repressed in clones anterior to the morphogenetic furrow , and is not expressed in clones posterior to the morphogenetic furrow ( Figure 5F , G ) . Therefore , high levels of eya and so are sufficient to repress Ey in the absence of normal morphogenetic furrow signaling . Together , these data suggest that the increased levels of Eya and So induced by signals in the morphogenetic furrow are important for Ey repression . To gain a better understanding of how Eya and So cooperate to regulate ey expression , we tested the response of the ey enhancer in vitro to So and/or Eya . In Drosophila S2 cells , when the ey enhancer is used to drive luciferase expression ( ey-luc ) , reporter expression was induced by co-expression of So with Eya , but not by either factor alone ( Figure 6A , “WT” ) . This suggests that the ey enhancer can be activated by Eya and So , and is consistent with previously published results that they cooperate to activate targets [45] , [47] . Mutation or deletion of the So binding site ( Mut or Short , respectively ) within the reporter strongly reduced its induction by Eya/So ( Figure 6A , B ) . This suggests that the activation of the construct in our assay depends primarily upon the So binding site . Our in vivo results indicate that high levels of Eya and So expression can repress Ey expression . However , even a 10 fold increase of both transfected plasmids did not repress; rather , the reporter was activated more strongly ( Figure 6A ) . To generate additional hypotheses we re-examined the in vivo expression of Ey , So , and Eya . We quantified pixel intensity values for Eya , So , and Ey in orthogonal sections ( as in Figure 5B ) across multiple imaginal discs ( n = 5 ) as a proxy to examine expression levels across the third instar disc . Values were normalized and plotted for each protein to generate a line graph that visually depicts staining intensity across the section ( as shown in Figure 6C , D ) . We observed that So undergoes a greater average positive fold change ( Posterior Max/Anterior max ) than Eya in both apical and basal sections ( Figure 6E ) . While this analysis is only semi-quantitative , it was highly reproducible , and could indicate that So is in excess to Eya in posterior cells . At a minimum it suggests that their relative levels of expression are different in anterior and posterior cells . To test the simple model that excess So can prevent ey expression , we increased the ratio of transfected so plasmid to eya plasmid in our in vitro system . In response , we observed a dramatic decrease of reporter expression ( Figure 6B ) , leading us to conclude that excess So suppresses activation of ey-dGFP by the Eya/So complex in vitro . To test this model in vivo we overexpressed So anterior to the morphogenetic furrow . We observed that in some clones Ey expression was mildly repressed by overexpression of So ( Figure 6F ) . Based on our in vivo and in vitro observations , we conclude that excess So expression can be sufficient to suppress ey expression . Within the morphogenetic furrow , we observed that Eya and So levels are not increased until after the initial decrease of Ey expression , indicating that there must be an additional mechanism that contributes to Ey negative regulation in this domain . The Ski/Sno family member Dachshund ( Dac ) physically interacts with Eya [20] , [55] , and may cooperate to regulate targets of So and Eya [28] . In mammals , the ortholog Dach interacts with the Eya and So orthologs to repress targets [56] , though this interaction has not been confirmed in Drosophila . To test if Dac is involved in Ey repression , we generated dac null clones . Anterior to the furrow , Ey expression was not affected in dac clones , suggesting that dac is not required for Ey expression anterior to the morphogenetic furrow ( Figure 7A ) . As previously reported for clones posterior to the morphogenetic furrow , we observed increased Ey expression in dac clones near the furrow , but not clones distant from it ( Figure 7A , B ) [26] . This overlaps the highest levels of Dac expression posterior to the furrow ( Figure S7A , B ) . This shows that dac is required for negative regulation of Ey specifically in the domain near the morphogenetic furrow . It is known that large dac clones can have delayed morphogenetic furrow progression , making it possible that Ey expression within these clones could be a secondary consequence of a delayed furrow [13] . To address this , we assayed furrow progression through small dac clones and compared this to the Ey expression boundary . Cubitus interruptus ( Ci ) , the effector of Hedgehog signaling , normally accumulates to high levels in a tight band within the morphogenetic furrow , just posterior to the onset of Ey negative regulation ( Figure 1A , Figure S7C ) . In dac clones spanning the furrow , Ci accumulation was not delayed , but Ey overlapped high levels of Ci , which was not observed in wild-type cells ( Figure 7B , compare to Figure 1A ) . This result suggests that the leading edge of the morphogenetic furrow , normally correlating with Ey suppression , moves into and through these dac clones . As Ey suppression is delayed in these clones , it indicates that Dac is required for suppression of Ey near the furrow independent of its role in furrow progression . To further test if dac represses Ey posterior to the furrow , we used F2-Gal4 to drive multiple independent dac RNAi transgenes , and observed that Ey expression is detected in knockdown cells posterior to the furrow ( Figure 7C and data not shown ) . This result shows that Dac is necessary to suppress Ey expression posterior to the furrow . We used the reporter ey-dGFP to determine if Dac suppresses ey at the transcriptional level . Like Ey , ey-dGFP is expressed in dac clones near the furrow ( Figure 7D , orange arrow ) , but not clones far posterior to the furrow ( Figure 7D , blue arrow ) . This indicates that Dac is required to suppress ey transcription near the morphogenetic furrow , consistent with the expression pattern of Dac . We also examined eymut-dGFP in dac clones . First , near the morphogenetic furrow , we did not observe expression of eymut-dGFP in dac clones as we had observed with ey-dGFP ( Figure 7E , orange arrow ) . This result indicates that the elevated levels of wild-type reporter expression observed in dac clones require the So binding site . By extrapolation , this result suggests that So still activates ey expression in dac clones near the MF; this places repression by Dac earlier than suppression by So during development . In clones far posterior to the morphogenetic furrow we observed that eymut-dGFP is expressed in dac clones ( Figure 7E , blue arrow ) , suggesting the repression of the wild-type reporter observed in dac clones requires the So binding site . We conclude that the phenotypes of ey reporter expression in dac clones reflect regulation by So in these domains . Furthermore , we conclude that Dac suppression of ey expression is an earlier developmental event than repression by So . We next overexpressed Dac with Eya or So to see if they were sufficient to suppress Ey expression anterior to the furrow . Overexpression of dac or eya alone did not alter Ey expression ( data not shown ) . Co-overexpression of eya and dac also had no effect on Ey expression ( data not shown ) . However , co-overexpression of so with dac was sufficient to repress Ey expression to modest levels ( Figure 8A ) . We conclude that Dac and So can cooperate to reduce Ey expression in vivo . dac is a downstream target of the So/Eya complex in the eye [19] , [25] , [49] . Therefore , we wanted to determine if Ey repression anterior to the furrow by co-overexpression of Eya and So ( Eya+So ) requires the activation of dac by these genes . To test this , we generated Eya+So overexpression clones that were also null for dac using the MARCM technique [57] . So and Eya reduced Ey expression anterior to the furrow , though less effectively than in cells that can still express Dac ( Figure 8B vs . Figure 5C ) . This suggests that So and Eya can repress Ey expression without Dac , but that full repression anterior to the furrow requires Dac . In MARCM clones spanning the furrow , the phenotype resembles dac null clones and Ey is not repressed , suggesting that Dac is specifically required in this domain ( Figure 8B ) . Finally , in posterior clones distant from the furrow , Ey is not expressed ( Figure 8B ) . This indicates that Eya and So are sufficient to completely suppress Ey in this domain . Together , these results indicate that Dac is required near the morphogenetic furrow to negatively regulate Ey expression , but that So and Eya can cooperate to repress Ey independent of Dac further posteriorly .
In this work , we have found that a switch from high to low levels of Ey expression is required for normal differentiation during retinal development . We also present a mechanism of Ey regulation by the RD gene network members Eya , So , and Dac . Specifically , we report that So switches from being an activator to a suppressor of ey expression , both depending on a So binding site within an ey eye-specific enhancer . We additionally report that the So cofactors Eya and Dac are required for ey repression posterior to the furrow but not for its maintenance ahead of the furrow , and are sufficient to cooperate with So to mediate Ey repression within the normal Ey expression domain . Our results support a Gro-independent mechanism for the suppression of target gene expression by the transcription factor Sine oculis ( So ) . An independent study has also shown that So can repress the selector gene cut in the antenna in a Gro-independent process though the mechanism was not determined [46] . We observe that Ey is expressed at low levels posterior to the morphogenetic furrow . However , when so expression is lost in clones posterior to the furrow , Ey expression and ey-dGFP expression are strongly activated . We show that this is not simply a default response of ey to So loss , as removing So from developmentally earlier anterior cells results in reduced ey expression . We also observe that knockdown of So specifically in differentiating cells using RNAi causes a similar phenotype , suggesting that an activator of Ey expression is expressed in differentiating photoreceptors . Mutation of a known So binding site in ey-dGFP results in activation of the reporter posterior to the furrow , supporting a model that binding of So to the enhancer prevents inappropriate activation of ey expression posterior to the furrow . Finally , in vitro we observe that an excess of So is sufficient to prevent activation of the enhancer and observe that in vivo overexpression of So can suppress normal Ey expression . Our observations are consistent with what in vitro studies have indicated about So function: when So binds DNA without Eya , it can only weakly activate transcription [45 , and this work] . However , our work introduces a novel mechanism of regulation for So targets , in which So occupancy of an enhancer prevents other transcription factors from inducing high levels of target gene expression . Our results also indicate that suppression of robust ey expression is an important developmental event . It is not yet clear if maintaining basal expression of ey , rather than completely repressing it , is developmentally important; however , it is possible that the ultimate outcome of a basal level of ey transcription may be necessary for the completion of retinal development [58] . Our results also show that eya is required for Ey suppression in vivo . However , consistent with its characterization as a transcriptional coactivator , our in vitro analysis does not indicate a direct role for Eya in repression . Previous studies , and our observations , indicate that Eya is required for the expression of So posterior to the furrow in the third instar [18] , [24] , [25] , [36] , and Figure S5 . Additionally , our reporter analysis shows that Eya regulation of ey requires the So binding site . We propose that the simplest model for Eya function in the suppression of ey is through its established function as a positive regulator of So expression , as we observe that overexpression of So alone is sufficient to weakly repress Ey expression and to block reporter activation in vitro . This model could also account for the results reported by us and others regarding the inability of this UAS-so construct to induce ectopic eye formation [16] , [36] , [46] , [59] . Briefly , the primary function of So in ectopic eye formation is to repress the non-eye program [46] . Overexpressing the So construct used in this study alone is not sufficient to induce this program , possibly because the transgene expression level is not sufficient; however , co-expression of the so positive regulator Eya is sufficient to induce robust ectopic eye formation [16] , [36] . In light of our findings , we propose that Eya co-expression is necessary to induce So expression to sufficient levels to block transcriptional activation of non-eye targets to permit the induction of the ectopic eye program; however we cannot rule out that other functions of Eya may play a role . We further demonstrate that dac expression is required specifically near the furrow for Ey repression . In addition , we show that the So binding site is required for strong ey expression in dac clones near the furrow , suggesting that So activates ey in these clones . This suggests that repression by Dac occurs before the transition to repression by So , making Dac the first repressor of ey expression at the furrow , and identifying how the initiation of repression occurs before So levels increase . We further show that Eya and So are sufficient to repress ey expression in dac mutant clones anterior to the furrow , though not as completely as in cells that express Dac . This result indicates that Dac is not an obligate partner with Eya and So in ey repression , but is required for the full suppression of ey . One model would be that Dac and So can cooperate in a complex to modestly repress eyeless directly . This would be consistent with our loss-of-function and reporter data as well as the observation that Dac and So misexpression can weakly cooperate to repress Ey anterior to the furrow . However , while a similar complex has been described in mammalian systems , previous studies have been unable to detect this physical interaction in Drosophila [44] , [45] , [55] , [60] . An alternative model is that Dac suppresses ey expression indirectly and in parallel to Eya and So . A previous study has shown that dac expression is necessary and sufficient near the furrow to inhibit the expression of the zinc finger transcription factor Teashirt ( Tsh ) [26] . Tsh overlaps Ey expression anterior to the furrow , and can induce Ey expression when misexpressed [42] , [43] . Furthermore , tsh repression is required for morphogenetic furrow progression and differentiation [42] , [43] . In light of these previously published findings , we propose that a simpler model based on current knowledge is that Dac repression of tsh at the morphogenetic furrow reduces Ey expression indirectly . Future studies may distinguish between these mechanisms . In addition to the role of the RD gene network in ey modulation , we identify that signaling events within the morphogenetic furrow indirectly regulate the switch to low levels of ey expression . It has been shown that signaling pathways activated in the morphogenetic furrow increase levels of Eya , So and Dac; furthermore , it is proposed that this upregulation alters their targets , creating an embedded loop within the circuitry governing retinal development and allowing signaling events to indirectly regulate targets through the RD network [26] , [28] , [61] . The identification of ey regulation by So posterior to the morphogenetic furrow represents a direct target consistent with this model . In conclusion , we present a model that rewiring of the RD network activates different dominant sub-circuits to drive key transitions in development ( Figure 9 ) . To the interactions previously identified by others , we add that strong upregulation of So , dependent on Eya , results in minimal levels of ey transcription [18] , [25] . We propose that the identification of this novel sub-circuit of the RD network provides a mechanism for terminating the self-perpetuating loop of determination associated with high levels of Ey , permitting the onset of differentiation and the completion of development . Together , these results give us a new view into how temporal rewiring within the RD network directs distinct developmental events .
The enhancer sequences were amplified from ey-dGFP or eymut-dGFP with XhoI and NheI tails . PCR fragments were digested and ligated per the manufacturer's instructions ( NEB , Takara ) directionally into pGL3-Basic ( Promega ) . Correct ligation events were identified by sequencing to generate ey-Luc and eymut-Luc , respectively . eyshort-Luc was amplified from ey-Luc and generates a truncated enhancer that ends 8 bp upstream of the So binding site . Drosophila S2 cells were cultured in Schneider's medium containing 10% fetal bovine serum and antibiotics . Cells were transiently transfected in 48-well plates using Cellfectin ( Invitrogen ) according to the manufacturer's protocol . Cells were transfected with ey-Luc , eyshort-Luc , or eymut-Luc , in the presence or absence of Eya and So in pMT vector ( Invitrogen , a gift from Ilaria Rebay ) , along with tub-Renilla luciferase in pRL vector ( a gift from K Basler ) . 24 hrs after transfection , cells were induced with CuSO4 at a final concentration of 500 µM . Luciferase activity was assayed 2 days after induction using the Dual-Glo kit ( Promega ) according to the manufacturer's protocol . Data were graphed in GraphPad Prism and labeled using Adobe Illustrator . For a list of the genotypes used , please reference Table S1 . All crosses were performed on standard cornmeal agar at 25°C unless otherwise noted . Heat shocks were performed at 37°C . Flipout-Gal4 [63] crosses were heat shocked for 8 min , 48 hrs after egg laying ( AEL ) . For loss-of-function clones or MARCM clones [57] , heat shocks were performed for 1 hr at 48 hrs AEL , or , for so3 and eyacliIID clones , 72 hrs AEL . Wandering third instar larvae were collected and dissected using standard methods as previously described [37] . Staining was performed as previously described [64] . For antibodies used , please reference Table S2 . Imaginal disc images were captured using a Zeiss LSM confocal microscope . LSMs were stacked using ImageJ software and stacks were merged in ImageJ and prepared for figures using Adobe Photoshop . Staining quantification for Eya , Ey and So: orthogonal sections were generated using ImageJ and represent approximately 10 micron wide slices through the full depth of the disc ( n = 5 ) ; sections were resliced in ImageJ to generate XZ stacks which were summed . The apical ROI was measured based on the width of the Eya signal in photoreceptors . The basal ROI was the same ROI , shifted basally to exclude the apical Eya signal . Pixel intensity was calculated using the plot profile function , and values were normalized . Pixel intensity plots and bar graph of average fold change were generated in GraphPad Prism . For adult images , adults were frozen at −80°C for 30 minutes . Light microscopy images of adult heads were captured on a Zeiss Axioplan microscope , and were processed with Adobe Photoshop software .
|
Animals develop by using different combinations of simple instructions . The highly conserved retinal determination ( RD ) network is an ancient set of instructions that evolved when multicellular animals first developed primitive eyes . Evidence suggests that this network is re-used throughout evolution to direct the development of organs that communicate with the brain , providing information about our internal and external world . This includes our eyes , ears , kidneys , and pancreas . An upstream member of the network named eyeless must be activated early to initiate eye development . Eyeless then activates the expression of downstream genes that maintain eyeless expression and define the eye field . Here , we show that eyeless must also be turned off for final steps of eye development . We investigated the mechanism by which eyeless is turned off and we find that feedback regulation by the downstream RD genes changes to repress Eyeless expression during late stages of development . This study shows that tight regulation of eyeless is important for normal development and provides a mechanism for its repression .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"developmental",
"biology",
"cell",
"differentiation",
"biology",
"cell",
"fate",
"determination"
] |
2013
|
Dynamic Rewiring of the Drosophila Retinal Determination Network Switches Its Function from Selector to Differentiation
|
Human T-cell lymphotropic virus type 1 ( HTLV-1 ) infection can increase the risk of developing skin disorders . This study evaluated the correlation between HTLV-1 proviral load and CD4+ and CD8+ T cells count among HTLV-1 infected individuals , with or without skin disorders ( SD ) associated with HTLV-1 infection [SD-HTLV-1: xerosis/ichthyosis , seborrheic dermatitis or infective dermatitis associated to HTLV-1 ( IDH ) ] . A total of 193 HTLV-1-infected subjects underwent an interview , dermatological examination , initial HTLV-1 proviral load assay , CD4+ and CD8+ T cells count , and lymphproliferation assay ( LPA ) . A total of 147 patients had an abnormal skin condition; 116 ( 79% ) of them also had SD-HTLV-1 and 21% had other dermatological diagnoses . The most prevalent SD-HTLV-1 was xerosis/acquired ichthyosis ( 48% ) , followed by seborrheic dermatitis ( 28% ) . Patients with SD-HTLV-1 were older ( 51 vs . 47 years ) , had a higher prevalence of myelopathy/tropical spastic paraparesis ( HAM/TSP ) ( 75% ) , and had an increased first HTLV-1 proviral load and basal LPA compared with patients without SD-HTLV-1 . When excluding HAM/TSP patients , the first HTLV-1 proviral load of SD-HTLV-1 individuals remains higher than no SD-HTLV-1 patients . There was a high prevalence of skin disorders ( 76% ) among HTLV-1-infected individuals , regardless of clinical status , and 60% of these diseases are considered skin disease associated with HTLV-1 infection .
Adult T-cell leukemia/lymphoma ( ATLL ) , HTLV-1-associated myelopathy/tropical spastic paraparesis ( HAM/TSP ) and infective dermatitis associated with HTLV-1 ( IDH ) are the main diseases caused by human T-cell lymphotropic virus type 1 ( HTLV-1 ) infection [1]–[3] . However , several other clinical conditions have been associated with this viral infection , such as uveitis , thyroiditis , arthritis and polymyositis [4]–[6] . There are an estimated 5 to 10 million HTLV-1 infected individuals worldwide and Brazil is considered a highly endemic area for HTLV-1 infection , with the largest absolute number of HTLV-1 infected individuals , with more than one million people living with this virus [7]–[9] . Despite this high prevalence , only a few studies on the dermatological aspects of HTLV-1 infection have been described in this country [10] . There is a lack of surrogate markers to assess the infected patients who have a higher risk for HTLV-1 associated skin disorders . Moreover , there are few immunological studies among HTLV-1-infected persons who are simultaneously suffering from HAM/TSP and skin diseases . The aim of this study is to evaluate the prevalence of skin disorders in HTLV-1-infected individuals and to correlate this prevalence with the initial HTLV-1 proviral load , and initial CD4+ and CD8+T cell count .
In the last 18 years , a cohort of HTLV-infected subjects has been followed in the HTLV-outpatient clinic at the Institute of Infectious Diseases “Emilio Ribas” ( IIER ) , with the support of nurses , nutritionists and physical therapists . From a total 450 HTLV-1-infected individuals , including asymptomatic carriers and HAM/TSP patients , 193 of them were consecutively evaluated for skin disorders from January 2008 to July 2010 by the same dermatologist , blinded for the clinical status to minimize information bias . Demographical and clinical dates were collected , and dermatological examinations were carried out . The HIV co-infected individuals were excluded , but HCV co-infected subjects were included . Ethics Statement: Written informed consent was obtained from all participants , and the IIER ethical board approved the protocol . The patients underwent laboratory studies , including HTLV-1 serological diagnosis , initial CD4+ and CD8+T cell counts , and an initial HTLV-1 proviral load . In accordance with previous studies , the following skin disorders associated with HTLV-1 infection ( SD-HTLV-1 ) were considered: xerosis/acquired ichthyosis , seborrheic dermatitis and infective dermatitis associated with HTLV-1 ( IDH ) [1] , [10]–[12] . HAM/TSP was diagnosed according to previously established criteria [13] . Skin culture and punch skin biopsies were performed when clinical examination was not sufficient for a dermatological diagnosis . Antibodies to HTLV-1/2 were detected by a diagnostic enzyme-linked immunosorbent assay ( ELISA ) and confirmed by Western blot analysis and polymerase chain reaction ( PCR ) , which are capable of discriminating between HTLV-1 and HTLV-2 [14] . To determine the counts of CD4+ and CD8+ T-cell subsets , fresh whole blood specimens were collected in EDTA tubes and subjected to flow cytometry ( Coulter EPICS® XL-MCLÔ Flow Cytometer - Beckman Coulter , Fullerton , CA ) , using human monoclonal antibodies anti-CD3 , anti-CD4 , and anti-CD8 , labeled with fluorochrome . The results of the first HTLV-1 proviral load were available in the database . Quantitative proviral DNA levels were detected by a real-time automated PCR method , using TaqMan probes for the pol gene . The albumin gene served as the internal genomic control , and MT2 cells were used as a positive control . The results are reported as copies/10000 PBMCs , and the detection limit was 10 copies [15] . Data were analyzed using SPSS 17 . 0 software . The association between independent variables and the outcome was analyzed either by Student's t-test or ANOVA ( normal distribution variables ) , or by Mann-Whitney test ( non-normal distribution variables ) , while the association between categorical variables and the outcome was assessed by the X2 test . HTLV-1 proviral load was log-transformed to obtain a normal distribution . Correlations between HTLV-1 proviral load and SD-HTLV were performed using Spearman's rank correlation . Data are expressed as mean ± standard deviation ( normal distribution variables ) or median and interquartile range ( non-normal distribution variables ) . Statistical significance was set at a p value<0 . 05 .
One hundred ninety-three HTLV-1-infected subjects ( 43% of all HTLV-1-infected patients at the HTLV outpatient clinic from the Emilio Ribas Institute cohort ) underwent a dermatological exam . Mean age of patients was 49 . 4±12 . 3 years . Female gender had a higher prevalence of HTLV-1 infection ( 72% ) . Regarding the presence of neurological involvement , 38% of the patients had a diagnosis of HAM/TSP . The dermatological examination revealed a high prevalence of skin disorders among the HTLV-1-infected patients ( 76% ) . Sixty-five individuals ( 34% ) had one dermatological condition , and 42% ( n = 81 ) of the patients had two or more dermatological conditions . Among the 147 patients that had an abnormal skin condition , 79% ( n = 116 ) had a skin disorder associated with HTLV-1 infection ( SD-HTLV-1 ) ( xerosis/ichthyosis or seborrheic dermatitis ) and 21% ( n = 31 ) had other dermatological diagnoses . The most prevalent skin disorder associated with an HTLV-1 diagnosis was xerosis/ichthyosis ( 48% ) , followed by seborrheic dermatitis ( 28% ) . Table 1 shows the prevalence of skin disorders in the HTLV-1 patients based on a diagnosis of HAM/TSP or asymptomatic carriers . SD-HTLV-1 were more prevalent on HAM/TSP patients ( xerosis/acquired ichthyosis ( p = 0 . 007; seborrheic dermatitis ( p = <0 . 0001 ) ; IDH ( p = 0 . 022 ) . Patients with SD-HTLV-1 , including asymptomatic carriers and HAM/TSP , are older , have a higher prevalence of HAM/TSP , and have a higher first HTLV-1 proviral load ( performed for 109 patients , p = 0 . 009 ) , compared with patients without SD-HTLV-1 ( Table 2 ) . Note that 75% of the SD-HTLV-1 group was made up of HAM/TSP individuals . Table 3 depicts the presence of SD-HTLV-1 in subjects that are asymptomatic for neurological symptoms ( absence of HAM/TSP ) . Mean age of SD-HTLV-1 patients was 51 years , as compared with 44 years for HTLV-1 patients without SD ( p = 0 . 002 ) , regardless of gender , CD4+ and CD8+ T cell counts ( p = 0 . 489; p = 0 . 824 ) . The initial HTLV-1 proviral load was significantly higher for the group with SD-HTLV-1 as compared with that for the group without SD-HTLV-1 and asymptomatic for neurological symptoms ( p = 0 . 021 ) .
Notably , 76% of the HTLV-1-infected asymptomatic carriers and 88% of the HAM/TSP patients showed some skin disorder in our study . These findings are similar to those described in two previous studies involving asymptomatic carriers and HAM/TSP subjects [10] , [12] . Thus , it is important to stress that HTLV-1 infection may have an etiological link to skin disease . In fact , skin disorders are highly associated with HTLV-1 infection , regardless of neurological symptoms , and they may represent a clinical warning sign for the diagnosis or progression of this infection [10]–[12] . Excluding HAM/TSP cases , subjects with a diagnosis of SD-HTLV-1 are older than groups with other types of skin disorders or individuals with a normal dermatological exam . However , no significant association was observed with gender . These findings suggest that older HTLV-1-infected individuals , who probably have a longer duration of their viral infection have had more time to develop SD-HTLV-1 . Although IDH is the only skin disease in which HTLV-1 infection is a criterion for diagnosis , other skin disorders could also be associated with HTLV-1 infection , including xerosis/acquired ichthyosis and seborrheic dermatitis , as previously demonstrated [10]–[12] , [16] . In fact , in our study , these illnesses were the most prevalent skin disorders associated with adult HTLV-1-infected individuals , regardless of the clinical status . However , studies with a longer follow-up should be performed to assess the hypothesis that these skin manifestations are related to HTLV-1 infection . The exact pathogenic mechanism of IDH still needs to be made clear , but the current view is that a diagnosis of HTLV-1 infection is necessary , leading in susceptible individuals to immune deregulation , with subsequent immunosuppression and superinfection with Staphylococcus aureus and beta-haemolytic streptococci , what additionally leads to chronic antigenic stimulation and persistent inflammation of the skin . Genetic , host and environmental factors have been shown to be associated [17] . Xerosis and acquired ichthyosis have been described as the main dermatological manifestations associated with HAM/TSP patients [11] , [12] , [18] . Xerosis is characterized by dryness of the skin , and acquired ichthyosis is clinically characterized by cutaneous xerosis and the formation of polygonal thin flat scales of varying sizes , mainly on the extremities [19] . Acquired ichthyosis is a consequence of hypohydrosis that may be secondary to the involvement of the autonomic nervous system , affecting directly the HTLV-1-infected skin cells [12] , [20] . On the basis of histopathological and immunohistochemical analyses of skin fragments of acquired ichthyosis from HAM/TSP individuals , it was concluded that keratinocytes are activated , probably as a result of cytokines that are liberated from HTLV-1-infected lymphocytes . This activation leads to an interference in keratinocyte differentiation and migration , resulting in a defect in the processes of normal desquamation , accumulation and retention of corneocytes [18] . We noticed that more than 50% of the SD-HTLV-1 cases involved xerosis/acquired ichthyosis and decided to include them in the same group because the acquired ichthyosis showed a similar clinical dermatological pattern , comparable with higher degree xerosis , and so they were clinically difficult to differentiate [12] . HTLV-1 proviral load is a laboratorial risk marker for the development of HAM/TSP and other diseases related to HTLV-1 infection [21] , [22] . The initial HTLV-1 proviral load was higher in the group with SD-HTLV-1 ( p = 0 . 009 ) as well as in the SD-HTLV-1 HAM/TSP-free group ( p = 0 . 021 ) , both of which were statistically significant . Although the proviral load may differ greatly among individuals , it is relatively stable during the course of the HTLV-1-related disease [23] . In HAM/TSP patients this finding suggests that proviral load reaches a stable level determined by the relationship between viral expression and the immune response against the virus [23] . As previously shown , HTLV-1 was identified by PCR on skin cells in addition to lymphocytes in HTLV-1-infected persons , regardless of their clinical status 10 . Because of these findings , several authors believe that HTLV-1 can modify the function of infected cells , resulting in skin disorders that are caused directly by the presence of the virus in the infected cells [10] , [16] , [24] . Another possible mechanism of skin disorders among HTLV-1-infected individuals is the production of cytokines in HTLV-1-infected lymphocytes , promoting a functional disturbance on skin cells [18] . This lack of association may be explained by the presence of HTLV-1 in specific sites that occur during HAM/TSP and the proviral load in the cerebral spinal fluid ( CSF ) [25] . Lezin et al . reported that the proviral load quantified in CSF was able to distinguish clearly between healthy groups of HTLV-1 carriers and patients presenting HAM/TSP [25] . Finally , for the first time , a high prevalence of skin disorders ( 76% ) independent of clinical status was disclosed among HTLV-1-infected individuals . These findings may have important implications in the clinical setting in places where this infection is endemic and dermatologists , infectious diseases specialists and general clinicians should be aware of skin presentations of the HTLV-1 infection . Moreover , the influences of demography and co-morbid conditions may be relevant , but have not been fully studied . Thus , data derived from a cohort of referred patients followed in a specialized HTLV clinic may not be a representative sample of the whole population of HTLV-1 patients and therefore unlikely to reflect the prevalence of skin conditions among all HTLV-1 patients .
|
HTLV-1 infection may increase the risk of developing skin disorders . A total of 193 HTLV-1 infected subjects were studied , including asymptomatic carriers and HAM/TSP patients . Of the subjects , 76% had an abnormal skin condition , with a high prevalence both among HTLV-1 asymptomatic carriers and HAM/TSP patients . The most prevalent SD-HTLV-1 was xerosis/acquired ichthyosis ( 48% ) , followed by seborrheic dermatitis ( 28% ) . Patients with SD-HTLV-1 were older ( 51 vs . 47 years ) , had a higher prevalence of myelopathy/tropical spastic paraparesis ( HAM/TSP ) ( 75% ) and an increased first HTLV-1 proviral load compared with patients without SD-HTLV-1 . When excluding HAM/TSP patients , the first HTLV-1 proviral load of SD-HTLV-1 individuals remains higher than no SD-HTLV-1 patients . Thus , skin diseases are highly prevalent among HTLV-1-infected individuals .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[] |
2013
|
High Prevalence of Skin Disorders among HTLV-1 Infected Individuals Independent of Clinical Status
|
Filarial nematodes currently infect up to 54 million people worldwide , with millions more at risk for infection , representing the leading cause of disability in the developing world . Brugia malayi is one of the causative agents of lymphatic filariasis and remains the only human filarial parasite that can be maintained in small laboratory animals . Many filarial nematode species , including B . malayi , carry an obligate endosymbiont , the alpha-proteobacteria Wolbachia , which can be eliminated through antibiotic treatment . Elimination of the endosymbiont interferes with development , reproduction , and survival of the worms within the mamalian host , a clear indicator that the Wolbachia are crucial for survival of the parasite . Little is understood about the mechanism underlying this symbiosis . To better understand the molecular interplay between these two organisms we profiled the transcriptomes of B . malayi and Wolbachia by dual RNA-seq across the life cycle of the parasite . This helped identify functional pathways involved in this essential symbiotic relationship provided by the co-expression of nematode and bacterial genes . We have identified significant stage-specific and gender-specific differential expression in Wolbachia during the nematode’s development . For example , during female worm development we find that Wolbachia upregulate genes involved in ATP production and purine biosynthesis , as well as genes involved in the oxidative stress response . This global transcriptional analysis has highlighted specific pathways to which both Wolbachia and B . malayi contribute concurrently over the life cycle of the parasite , paving the way for the development of novel intervention strategies .
Human filarial infections are currently a leading cause of morbidity in the developing world . Despite the large cost to human health , the chronic and debilitating diseases caused by filarial nematodes remain largely neglected . Two of the most prevalent chronic diseases caused by filaria include lymphatic filariasis , caused by Wuchereria bancrofti , Brugia malayi , and Brugia timori , and onchocerciasis , caused by Onchocerca volvulus [1] . Currently 38 . 5 million people have lymphatic filariasis while 15 . 5 million people have onchocerciasis , representing in 2015 over 300 , 000 years lived with disability ( YLDs ) [2] . While efforts to mitigate the effects of these diseases have been successful in some regions , current medications are insufficient to reach elimination by 2020 , particularly in regions of co-endemicity with loasis , caused by the filarial nematode Loa loa [3] . Current mass drug administration relies on a small arsenal of drugs , increasing the likelihood of development of resistance , a phenomenon already observed in their veterinary applications [4] . One such drug , Ivermectin , the primary control strategy for onchocerciasis , is unsafe to use in regions where lymphatic filariasis or onchocerciasis are co-endemic with loasis due to the risk of severe adverse effects in individuals heavily infected with Loa loa . Most filarial nematodes are hosts for an obligate bacterial endosymbiont , the intracellular bacteria of the genus Wolbachia . As the filariae require these bacteria to develop , reproduce and survive in the human host , they represent an attractive target for intervention . The bacteria reside in the lateral cords of the larval and adult nematodes ( male and female ) as well as in the ovaries and developing embryos of the adult female worms . While the relationship between the nematode and the bacteria is known to be co-dependent , the molecular basis for this relationship remains poorly understood . Wolbachia are required for the parasite to reproduce and develop in the mammalian host , while the parasite likely provides amino acids required for bacterial growth [5] . Analyses show significant degradation of the Wolbachia genome compared to its free-living relatives , yet it appears to have maintained a number of intact metabolic pathways such as riboflavin , heme , and nucleotide synthesis [5 , 6] , three pathways that are deficient in the nematode host [7] . As these metabolites are considered essential to all living things , these deficiencies may underlie the symbiotic relationship . Interestingly , in a genome-wide screen for diversifying selection , genes for heme , riboflavin , and nucleotide biosynthesis were found to be under positive selection , again implying they may be integral to the symbiotic relationship [8] . Curiously , however , the recently sequenced L . loa genome , a Wolbachia-free filarial nematode believed to have lost the endosymbiont , also lacks these metabolic pathways and does not appear to have acquired them through horizontal gene transfer [9 , 10] . This suggests that filarial worms could also be acquiring these essential metabolites from their mammalian hosts . Thus , the basis of the filaria-Wolbachia co-dependency has still not been clarified with the availability of the genomes . Clearance of Wolbachia with the use of antibiotics results in significant apoptosis of filarial germline cells , cells of developing embryos in the female worms , as well as somatic cells of the microfilaria . These effects are non cell-autonomous , meaning cell death is not restricted to cells infected with Wolbachia pre-treatment [9] . It is hypothesized that Wolbachia are preventing apoptosis by one or both of two possible mechanisms: i ) Wolbachia are interfering with the host apoptotic program to prevent cell death , and ii ) Wolbachia secrete some necessary metabolic product ( s ) that prevent cell death . In this study , we profiled the transcriptomes and inferred co-expression of genes in Wolbachia and B . malayi during the development of male and female worms to identify co-expressed pathways necessary for mediating the endosymbiotic relationship .
Parasites were obtained from FR3 where they were isolated and separated by sex from infected gerbils ( Meriones unguiculatus ) at 16 ( L4 ) , 30 , 42 and 120 days post infection ( dpi ) . Worms were flash frozen and shipped to the New York Blood Center for processing . B . malayi worms where homogenized in Trizol ( ThermoFisher ) using a hand-held pestle in 1 . 5mL tubes containing the worms . For extraction , 2 , 000 L4s , 50 male and female juveniles ( at 30 dpi and 42 dpi ) , and 10 male and female adult worms ( 120 dpi ) were used , with two biological replicates for each . Total RNA was extracted by organic extraction using Trizol . A portion of each sample was saved for a DNA extraction while the rest was treated with DNaseI ( New England Biolabs ) . Ribosomal RNA ( rRNA ) depletion was performed using Terminator ( Epicentre ) , a 5’-phosphate-dependent exonuclease that degrades transcripts with a 5’ monophosphate . Libraries were prepared using the NEBNext Ultra RNA Library Prep Kit for Illumina ( New England Biolabs ) according to manufacturer instructions . Library quality was assessed using a D1000 ScreenTape Assay ( Aligent ) prior to sequencing . Library concentrations were assessed using the qPCR library quantification protocol ( KAPA biosystems ) . Libraries were sequenced on the Illumina HiSeq2500 platform with 150bp paired-end reads . To minimize the confounding effects of lane-to-lane variation , libraries were multiplexed and sequenced with technical replicates on multiple lanes . Each developmental stage received an average of 141 million mapped reads . Read quality was assessed using FastQC ( Babraham Bioinformatics ) . Sequence reads from each sample were demultiplexed and analyzed with the Tuxedo suite of tools [11–13] . Reads were mapped to the annotated B . malayi ( WormBase . org ) and Wolbachia [6] genome assemblies with Tophat2’s ( v2 . 1 . 1 ) Bowtie2-very-sensitive algorithm [11] . The resulting BAM files were then used with Cufflinks ( v2 . 2 . 1 ) [11–13] to obtain fragments per kilobase of exon per million fragments mapped ( FPKMs ) for each of the annotated transcripts and with Cuffnorm [11–13] to obtain normalized FPKMs , normalized for library size . The Tophat2 alignment files were also used to determine differentially expressed genes in both organisms by first using HTSeq ( v0 . 6 . 1p2 ) [14] to generate read counts for each gene . Raw read counts were used as input to EdgeR ( v3 . 16 . 5 ) [15] to obtain differentially expressed genes between life stage . Genes were determined as significantly differentially expressed using a threshold of p <0 . 05 and a false-discovery rate ( FDR ) of 5% , standard settings in EdgeR . To make the co-expression network and identify the co-expressed gene modules in the symbiosis between B . malayi and Wolbachia , we normalized the gene expression profiles of B . malayi and Wolbachia using Cuffnorm [11–13] and then performed weighted gene correlation network analysis ( WGCNA ) on the combination of normalized gene expression of B . malayi and Wolbachia using the WGCNA package in R [16] . Hierarchical clustering and dynamic branch cutting were used to identify stable modules of densely interconnected genes . GO term information was downloaded from WormBase . org . Metadata including WSP ( Wolbachia Surface Protein ) and a ratio of wsp to gst ( glutathione-S-transferase ) were all integrated into the co-expression network . To estimate the relative expression of Wolbachia genes over different stages of worm development , the DNAse-treated RNA stored in aliquots that were prepared for library preparation and sequencing ( see above ) was used as a template for cDNA synthesis using the SuperScript III First Strand cDNA Synthesis Kit ( Invitrogen ) . The cDNA was prepared from two biological replicates . Gene expression was estimated using the standard ‘ΔΔCt’ method . For internal control of Wolbachia gene expression , we selected two housekeeping genes ( wBm0291 and wBm0528 ) based on their constitutive expression over the development of the worm according to the RNA-seq data . DNA was extracted from B . malayi worms ( the same samples as RNA ) by taking the non-organic fraction of trizol/chloroform solutions ( see above ) . DNA was precipitated by ethanol and diluted in water . Wolbachia numbers per worm were quantified by qPCR using primers for a Wolbachia single-copy gene ( wsp ) as previously described [17] . Expression data have been deposited in the Sequence Read Archive ( SRA ) under Accession number SRP090644 .
To obtain a global view of the transcriptional programs of both B . malayi and Wolbachia concurrently , over the course of worm development from L4 to adulthood , we performed dual RNA-seq . In total over 988 million ( or 486 paired-end ) RNA-seq reads out of 1 . 5 billion reads ( 65 . 9% ) obtained were mapped to the B . malayi and Wolbachia reference genomes ( Fig 1: Circos plots for Wolbachia ) . Mapped reads per stage ranged from 77 to 216 million for the B . malayi genome and 2 . 1 to 3 . 7 million for the Wolbachia genome ( Table 1: Sequencing summary ) . We found over 96% of B . malayi gene models to be “expressed” in at least one stage ( i . e . a minimum of four cumulative reads across the two independent biological replicates ) ( Fig 2: Clustering of Brugia Expression ) ( Table 1: Sequencing Summary ) . Expression in B . malayi was dominated by sex-biased gene expression , with the 120 dpi adult male and females expressing the most genes at the highest expression levels ( Fig 2 ) . In Wolbachia , 85% of gene models were classified as expressed ( Table 1: Sequencing Summary ) . Sequence reads from technical replicates—i . e . the same library sequenced on different lanes of the HiSeq—were combined per biological replicate as they contained the same insert size distribution . Using multidimensional scaling analysis , we clustered biological replicates for each B . malayi developmental stage ( S1a Fig: B . malayi MDS ) . All biological replicates clustered closely to each other with the exception of the F42 replicates , where F42b clustered more closely with the F30 replicates than with F42a . However , when we clustered biological replicates for Wolbachia reads , F42a and F42b clustered closely together ( S1b Fig: Wolbachia MDS ) . It is thus unlikely that the observed B . malayi disparate clustering for that stage is due to the mislabeling or contamination of the sample and more an effect of natural population variation . Generally , clustering of the stages indicates good reproducibility of the biological replicates , with the 30 dpi samples of both males and females clustering more closely with the mixed-gender L4s than to each other , followed by the 42 dpi samples . As expected , the 120 dpi male and female samples are found to be the most different from each other than the other samples from earlier life stages . To validate the use of RNA-seq for the purpose of transcriptional analysis , seven Wolbachia genes , with ten pair-wise comparisons , were selected for qRT-PCR analysis of their relative expression . Four of the genes ( wsp , Hsp90 , DnaK , and GroEL ) , with seven significant pair-wise comparisons , were chosen based on the criteria that they were found to be significantly differentially expressed and had over 50 read counts per stage . We also included three genes ( RibA , HemA , and AfuA ) that were constitutively expressed , based on an FDR of 1 in EdgeR , which indicates that they were the least likely to be differentially expressed . We observed a spearman correlation between the qRT-PCR and RNA-seq results of 0 . 987 and a p-value < 2 . 2e-16 ( S2 Fig: qPCR validation graph ) . Differential expression in B . malayi was dominated by sex-biased gene expression , as previously observed [18] , with the largest number of sex-biased genes at 120 dpi , with 2 , 753 genes showing male bias , and 3 , 109 showing female biased expression ( S1 Table: Brugia DE Female , S2 Table: Brugia DE Male , and S3 Table: Brugia DE Male to Female ) . We find that 82% of the genes previously determined to be significantly up-regulated in adult male worms and 79% of the genes significantly up-regulated in females worms [18] , were , in our new data set , also up-regulated in male worms ( M120 ) or female worms ( F120 ) , respectively , as compared to worms of the opposite sex . This shows good reproducibility between the two studies , although it should be noted that many additional genes were found to be differentially expressed between F120 and M120 worms in the new data set . This is likely due to the use of biological and technical replicates , as well as a higher depth of coverage . To uncover the role Wolbachia may play in worm development , we analyzed differentially expressed Wolbachia genes in male and female worms at each developmental stage . Pair-wise differential expression analysis was performed using EdgeR , after removing all genes with zero expression in two or more samples per comparison . The percentage of differentially expressed genes in any pair-wise comparison ranged from 0–4 . 8% of Wolbachia genes expressed ( Table 1 ) . We identified a total of 62 differentially expressed ( DE ) Wolbachia genes across a single or multiple pair-wise comparisons ( Fig 3: Clustering of Wolbachia DE genes ) . The largest number of Wolbachia DE genes ( 34 genes ) is in the females from 42 dpi ( F42 ) to 120 dpi ( F120 ) , while there are no DE genes in the 30 to 42dpi comparisons in both males and females ( Table 2: DE summary ) . In comparing stages between genders , there were 17 DE genes between females and males at 120 dpi ( F120 and M120 ) , and no genes differentially expressed between both sexes at 30 or 42 dpi . Comparisons between female stages consistently resulted in more DE genes than did comparisons between male stages ( Table 2: DE summary ) : 40 Wolbachia genes were determined as DE over the course of female growth , but were absent in any male comparisons . Because these genes appear to be differentially regulated during female worm development only , they are potentially required for female-specific processes known to be dependent on Wolbachia infection , such as maturity of female gonads and germline development , as well as embryogenesis . The ten Wolbachia genes determined as DE in both males and females represent potential expression in the lateral cords , required for the development of both the male and female germlines . Twelve Wolbachia genes were determined as DE in males only ( S4 Table: Wolbachia DE genes ) . The pattern of Wolbachia differential expression in the female stages was dominated by chaperone protein expression ( Table 3: DE expression of Chaperones ) . Integral membrane proteins , translation , antioxidants , oxidative phosphorylation , DNA replication , and peptidase function were also highly represented ( S4 Table ) . GO-term enrichment analysis of Wolbachia DE genes during female development using Fisher’s exact test revealed an enrichment of GO terms associated with chaperone function including protein-folding and unfolded protein-binding ( Table 3 ) . We detected seven genes with chaperone function that were significantly up-regulated in F30 as well as in F42 , four of which are also significantly up-regulated in F120 as compared to M120 . The four genes found to be significantly up-regulated in all female stages include wBm0350 groEL and co-chaperonin wBm0349 groES , which work in complex as an integral part of several stress responses in bacteria , including the oxidative stress response , where they recover oxidized proteins [19–23] . Also highly up-regulated in all female stages is the molecular chaperone wBm0533 grpE , shown to assist in protein refolding during oxidative stress , as well as HslU , a subunit of an ATP-dependent protease with chaperone function [24 , 25] . Among the chaperones significantly up-regulated in the F30 and F42 stages , but not F120 , are DnaK and DnaJ , another molecular chaperone system shown to be required for cell division in bacteria as well as for resistance to heat shock [23] . Among the Wolbachia DE genes in adult female worms ( F120 ) are a number of genes involved in combating oxidative stress . wBm0439 coenzyme Q-binding protein , an antioxidant , is significantly up-regulated from F30 to F120 as well as from F42 to F120 . We also find wBm0220 SodA , a superoxide dismutase , to be significantly up-regulated from F42 to F120 . SodA catalyzes the conversion of superoxide radicals to hydrogen peroxide and oxygen and is known to be essential in combating oxidative stress [19] . Additionally , we detect significant upregulation from F42 to F120 of wBm0674 , a malic enzyme responsible for the interconversion of L-malate and pyruvate . This reaction is essential for maintaining cellular pools of NADPH , required for a number of downstream processes including reducing oxidative stress [26 , 27] . A number of genes determined as DE during female development are involved in energy production . It is hypothesized that a key mechanism of the Wolbachia-host symbiosis is aerobic energy production by the bacteria for the worm [28–30] . NADH dehydrogenase subunit B ( wBm0242 ) , which is involved in oxidative phosphorylation , is significantly up-regulated in F30 as compared to L4s . ATP synthase subunit C , which creates ATP using a proton gradient , is also up-regulated in F120 as compared to F30 , and in M120 as compared to F120 . Two proteins involved in iron-sulfur cluster formation are also up-regulated in F120 as compared to earlier female stages: wBm0756 , an iron-sulfur cluster assembly scaffold protein , and wBm0448 , a succinate dehydrogenase flavoprotein . Iron-sulfur clusters are essential co-factors for respiratory chain proteins involved in ATP production [28] . Several glycolytic enzymes were significantly up-regulated in F120 and M120 as compared to earlier stages , including transaldolase ( wBm0686 ) , an enzyme linking the pentose phosphate pathway to glycolysis , and wBm0097 , a fructose-bisphophase aldolase . Notably , the Wolbachia genome lacks two glycolytic enzymes ( 6-phosphofructokinase and pyruvate kinase ) likely rendering the glycolytic pathway defective . Wolbachia may therefore depend on products from the B . malayi glycolytic cycle such as pyruvate , as well as TCA cycle intermediates derived from amino acids . Accordingly , wBm0207 pyruvate dehydrogenase , which transforms pyruvate into acetyl-CoA that can then be used in the citric acid cycle , was differentially expressed during female development . Correspondingly , in B . malayi we see an upregulation of Bm5241 in F120 , involved in the glycogen catabolic process . We also see differential expression of wBm0384 , an extracellular metalloprotease potentially involved in the breakdown of filarial peptides for amino acids during female development [6] . Additionally we see the two Zn-dependent peptidases cluster with most of the TCA cycle enzymes based on expression . Together with the up-regulation of pyruvate dehydrogenase , this co-expression suggests an increased dependence on B . malayi products for energy production . Another functional category represented in the up-regulated genes in F120 as compared to younger female stages is that of DNA replication . DNA polymerase III gamma/tau subunit ( wBm0434 ) and recJ ( wBm0124 ) , a single-stranded DNA-specific exonuclease involved in single-strand break repair , and DNA/RNA helicase ( wBm0708 ) , required for both DNA replication and transcription , are significantly up-regulated at this stage , as is the RNA polymerase omega subunit ( wBm0387 ) , indicating an increase in transcription during this developmental stage . We determined two ribosomal proteins , S4 and S15 , to also be significantly up-regulated at F120 . The ribosomal protein S4 is essential for protein synthesis through its function in RNA binding , leading to fewer errors , while S15 plays an essential role in the assembly of the central domain of the small ribosomal subunit [31] . Our observations are consistent with findings in the Wolbachia populations in the gonads of O . ochengi showing differential regulation of Wolbachia genes required for DNA replication and translation , including ribosomal proteins in the germline [29] . Genes involved in lipid II/ peptidoglycan biosynthesis ( wBm0493 , metC , and wBm0492 , murE ) and wBm0490 , a protein shown to interact with ftsH , a gene required for cell division , were also found to be significantly up-regulated in F120 . Among the genes differentially expressed over the course of female development are three peptidases including wBm0384 , an extracellular metallopeptidase unique to the Wolbachia of B . malayi , that are potentially involved in the breakdown of filarial peptides for amino acids [6] . The other two genes , wBm0772 and wBm0552 , encode ATP-dependent protease subunits . Genomic analysis revealed that Wolbachia has maintained the biosynthetic pathways for purines and pyrimidines while B . malayi has not , suggesting that Wolbachia are potentially provisioning nucleotides to their filarial hosts , especially during times of increased need . In accordance with this hypothesis , wBm0443 guanosine monophosphate synthase , an essential enzyme in de novo purine biosynthesis , is differentially expressed during female worm development . Additionally , we find wBm0255 amidophosphoribosyltransferase , also involved in de novo purine biosynthesis , to be differentially expressed in both males and females . This supports a potential role of nucleotide production by Wolbachia in the lateral cords in male and female worm development . It was proposed that Wolbachia might be inhibiting apoptosis in the worm host through the manipulation of the host apoptotic pathway [9 , 29] . While very little is understood about how this may be occurring , we find significant up-regulation of three genes putatively involved in the manipulation of the apoptotic pathways in F120 . One such gene is wBm0152 , a Wolbachia surface protein , shown to inhibit apoptosis of purified human polymorphonuclear cells in vitro [32] . We also find significant up-regulation of wBm0296 , an ankyrin repeat-containing protein , hypothesized to be an effector protein of the Type-IV secretion system ( T4SS ) able to mediate interactions with the host cells as they are for other intracellular bacteria [33–35] . Lastly , wBm0490 , a gene with high homology to a bax-inhibitor in the Wolbachia of Drosophila , is significantly up-regulated in F120 . Manipulation of the host apoptotic pathway through the expression of bax-inhibitors is believed to be responsible for the suppression of apoptosis of host cells by the obligate intracellular bacteria , Chlamydia trachomatis [36] . The expression in Wolbachia of three biosynthetic pathways ( heme , riboflavin , and FAD ) potentially important for symbiosis as well as Wolbachia transporters , were examined in each developmental stage ( Fig 4: Heatmap of gene expression of pathways of interest ) . Common to all Wolbachia genomes sequenced thus far is the presence of nearly all genes necessary for the synthesis of the iron-containing cofactor , heme , except for hemG , which is missing in many heme-producing bacteria [6 , 28 , 37] . Heme is an essential cofactor for cytochromes , peroxidases , and catalases , which are involved in a number of critical cellular processes including oxidative phosphorylation and electron transport . Heme is a co-factor for peroxidases essential for molting and might possibly also be a co-factor for steroids involved in molting of filarial parasites [28 , 38] . Unlike what was shown for O . ochengi where very low expression of the heme biosynthetic pathway in adult tissues was detected [29] , we found all Wolbachia genes involved in heme biosynthesis to be expressed in all sampled stages , with the highest expression at the L4 and F120 stages . Additionally , we find constitutively high expression of iron ABC transporters responsible for importing iron into the bacterial cell as well as heme ABC-transporters responsible for transporting heme from the bacterial cystoplasm into the periplasmic space and potentially involved in the transport of heme into the cytoplasm of the filarial host cell . Unlike Rickettsia , Wolbachia has maintained the ability to synthesize both riboflavin and FAD [6] . As riboflavin biosynthesis has been lost in B . malayi , it was hypothesized that Wolbachia were provisioning this cofactor to their filarial hosts . In support of this hypothesis we found that wBm0416 , involved in FAD biosynthesis , as well as RibA and RibB are constitutively highly expressed across all stages of worm development . RibA is a bifunctional enzyme that catalyzes the first two essential steps in riboflavin biosynthesis , and is co-regulated with the T4SS [39] . F120 is the only stage in which we find all genes in the FAD/ riboflavin biosynthetic pathways to be classified as expressed , and at particularly high levels . The Sec translocase system is responsible for the majority of protein trafficking across the bacterial cytoplasmic membrane into the periplasm with the use of ATP [40] . SecY is a transmembrane protein constituting the core of the protein-translocating complex . SecY was constitutively highly expressed across all stages of the life cycle . SecG associates with SecY to form a heterotrimeric complex . While not necessary for general function of the system , SecG has been shown to facilitate transport at low temperatures ( 20°C ) , or when the proton-motive-force is reduced [41 , 42] . We find expression of secG only in L4 , M42 , F120 , and M120 where we see particularly high expression . While Wolbachia lack the tatB gene , part of the Sec-independent twin arginine translocation ( Tat ) protein system present in most bacteria , they do maintain TatA and TatC genes . Thus , as in other alpha-proteobacteria , it is likely that this system is still functional [43] . TatA was highly expressed in F120 and M120 exclusively , while TatC was expressed constitutively across all stages . Secretion in Wolbachia requires not only translocation into the periplasmic space by either the Sec or Tat systems , but transport across the outer membrane as well . This is accomplished by the T4SS , a leader-peptide independent mechanism for transporting effector proteins and virulence factors found in many pathogenic and endosymbiotic bacteria [35 , 44–46] . We find constitutive expression of nearly all genes in this pathway at most stages except for wBm0798 in L4 . ATP-binding cassette transporters ( ABC transporters ) are composed of two transmembrane domains and two cytoplasmic ATP-binding domains . They are involved in the uptake of a variety of nutrients and the extrusion of drugs and metabolites [47] . As previously mentioned , we saw constitutive expression of all four heme ABC transporters encoded in the Wolbachia genome , as well as two lipoprotein transporters . Constitutive expression of the lipoprotein transport system LolCDE is required to export lipoproteins to the outer membrane [48] . Lipoproteins have been shown to be agonists of inflammatory pathogenesis in lymphatic filariasis , recognized by the TLR-2 and TLR-6 in the human host [49] . Correspondingly , we see significant up-regulation in adult female worms of the wBm0152 peptidoglycan-associated lipoprotein-like outer membrane protein shown to be localized to numerous sites on the bacterial membrane [50] . We also find constitutively high expression of two phosphate transporters , potentially required for importing phosphate for nucleotide production . Experiments in L . sigmodontis show that when Wolbachia is depleted with tetracycline , expression of a filarial phosphate transporter is significantly increased to compensate for the decrease in Wolbachia-produced nucleotides that are essential for worm embryogenesis and survival [51] . To determine which genes were being co-expressed between B . malayi and Wolbachia , we built a co-expression network for the two organisms using WGCNA ( Fig 5: The co-expression network for B . malayi and Wolbachia ) . WGCNA is a well-established method by which expression data and trait data are integrated to identify co-expressed pathways . We used hierarchical clustering and dynamic cutting to determine modules of co-expression . A module is a cluster of interconnected genes with high correlation based on their expression profiles ( S5 Table: Summary of module membership ) . GO term enrichment of each resulting module was performed to determine which modules were biologically significant ( S6 Table: GO term Enrichment ) . To identify modules of gene co-expression with the most interest based on the symbiotic interaction of the two organisms , we evaluated the correlation of each module to a measure of Wolbachia population , wsp/gst . We found three modules that had the highest negative correlations with the wsp/gst ratio ( brown , -0 . 57 p-value = 0 . 03; yellow and green , -0 . 65 p-value = 0 . 01 ) . A fourth module had the highest positive correlation with the wsp/gst ratio ( salmon , 0 . 63 p-value = 0 . 02 ) . Because the F120 samples have the lowest wsp/gst ratio ratio due to the large size of the female worms , we determined that the three modules with the highest negative correlation were indicative of adult female gene expression . The green module contains 2 , 312 B . malayi genes and 28 Wolbachia genes ( S5 Table ) ; it shows an enrichment of DE genes that are up-regulated in the adult females , with a p-value < 2 . 2e-16 . The most significantly enriched GO terms in the green module include intracellular signal transduction , proteolysis , transmembrane transport , cyclic nucleotide biosynthetic process , cysteine-type peptidase activity , and mitochondrion . Interestingly , this module also contains the cathepsin-like cysteine protease Bma-cpl-6 found to be involved in development and embryogenesis in the worm as well as in Wolbachia expansion ( S6 Table ) [52] . The yellow module contains 2 , 782 B . malayi genes and 39 Wolbachia genes and shows an enrichment of GO terms that include the regulation of transcription , oxidoreductase activity , sequence-specific DNA binding , and cell redox homeostasis . Interestingly , the co-expressed Wolbachia genes in this module include six of the seven chaperones found to be differentially expressed during female development as well as a gene involved in redox sensing and two genes involved in cytochrome c biogenesis . Cytochrome c is known to play a role in the electron transport chain and cell apoptosis , as well as being an antioxidative enzyme by removing superoxide and hydrogen peroxide . The brown module contains 4 , 148 B . malayi genes and 64 Wolbachia genes and shows a significant under-representation of DE genes ( p-values < 2 . 2e-16 ) in F120 as compared to F42 . As one of the modules that are the most highly negatively correlated to the Wolbachia population , the brown module contains a number of genes implicated in the host/endosymbiont relationship including genes involved in riboflavin and nucleotide biosynthesis ( S5 Table ) . Among the most enriched GO terms in the brown module are metabolism and transport . Co-expressed Wolbachia genes include a number of genes also involved in transport , such as three ABC transporters , including an iron importer , a general permease exporter , and a polyamine transporter . Among the many B . malayi genes involved in transport co-expressed in this module is Bm4941 , one of three genes in the genome predicted to have nucleoside transmembrane transporter activity based on protein domain information . In the brown module , we find the GO term glycogen catabolic process ( GO:0005980 ) to be significantly enriched , as well as the genes Bm5241 in B . malayi and wBm0207 in Wolbachia , previously mentioned in relation to pyruvate metabolism , to be co-expressed in this module . Among the Wolbachia genes co-expressed in this module are four genes involved in de novo purine and pyrimidine biosynthesis . Also in the brown module are two Wolbachia genes in the riboflavin biosynthesis pathway , including RibA , which are co-expressed with a number of B . malayi riboflavin or flavin-requiring proteins . Another set of enriched GO terms in the brown module includes DNA repair and replication , a function also represented with the co-expressed Wolbachia genes that include five genes involved in DNA replication and repair , including ribonucleotide reductase , an enzyme integral in controlling the rate of DNA synthesis [53] . Correspondingly , two genes involved in cell division in Wolbachia are co-expressed in this module . Interestingly , the salmon module , the module with the highest positive correlation to Wolbachia population , is enriched for GO terms related to regulation of apoptosis , response to oxidative stress , oxidase activity , and heme binding .
The dual transcriptional profiling that we performed revealed potential stage-specific requirements from Wolbachia during filarial development and embryogenesis . Differentially expressed Wolbachia genes during the course of female development generally fell into the functional categories of chaperone function , energy production , nucleotide biosynthesis , DNA replication , and anti-oxidative defense . These categories include genes that are likely to be required for specific developmental processes , including germline and embryonic development in the nematode and Wolbachia invasion of the gonad . Similar studies have been performed on the filarial nematode Dirofilaria immitis , or dog heartworm , and its Wolbachia endosymbiont , wDi [30 , 54] . While the first study found no wDi genes differentially expressed between adult male and female worms , a second tissue-specific study reported differentially expressed wDi genes in the uterus and female body wall . Our findings including wBm genes differentially expressed over the course of female development correspond to those wDi genes found to be up-regulated in the uterus , including multiple genes encoding ribosomal proteins , DNA replication and repair machinery , a tRNA synthetase , and a component of the purine biosynthetic pathway . This indicates that the role Wolbachia play in certain life stages of the filaria may be well conserved . It has been proposed that the ability of Wolbachia to perform aerobic respiration and metabolize iron whilst responding to oxidative stress may be an essential mechanism of the endosymbiotic relationship with filarial worms [28 , 29] . Studies in Litomosoides sigmodontis , for example , have found that targeting Wolbachia with antibiotics resulted in the up-regulation of components of the mitochondrial respiratory chain [55] . Experiments in O . ochengi showed that worms treated with antibiotics lose motility and that Wolbachia density in infected cells greatly exceeded that of mitochondria [56] . These results point to the potential ATP provisioning by Wolbachia to the filarial host [56] . The up-regulation of genes involved in the ATP transport chain and in iron-sulfur cluster formation—which are essential co-factors for respiratory chain proteins in ATP production—provide support for this hypothesis but it is difficult to prove . Alternatively , the up-regulation of ATP production could be required for the increased propagation of Wolbachia at this stage . The potential production of ATP by Wolbachia for its filarial host likely contributes to oxidative stress of their cellular environment by generating ROS as by-products of aerobic metabolism . Consistent with this hypothesis , many genes that encode proteins known to be involved in combating oxidative stress were highly differentially expressed during female development , including chaperone proteins and a gene involved in single-strand break repair , a potential consequence of an increasingly oxidative environment . A number of these chaperone proteins have been shown to be part of the oxidative stress response in bacteria and to maintain their stability under oxidative conditions [57 , 58] . It was proposed that overexpression of groEL is an important adaptation allowing for the obligate intracellular lifestyle of Wolbachia within a cytoplasmic vesicle [59 , 60] . We find that these requirements appear to be of special importance during female development , potentially as a consequence of the up-regulation of oxidative phosphorylation . Chaperones GroEL , HSP60 , and DnaK were found to be among the proteins with the most abundant peptide counts in proteomic analysis of B . malayi in the adult stages of the worm [61] . This study did not however look at abundance over the course of worm development , L4 through 30–42 dpi . The chaperone HslU forms a complex with the peptidase HslV , which was also found to be significantly up-regulated in the F30 and F42 stages . HslV , or HSP20 , has additionally been shown to be involved in bacteria-host interactions in Helicobacter pylori [62] . Interestingly , the DnaJ/K chaperones are among the Wolbachia genes found to be inserted in the nuclear genomes as well as expressed by the Wolbachia-free filariae A . viteae and O . flexuosa [63] . While many endosymbionts and parasites , including B . malayi , as well as members of the Rickettsia genus have lost the pathways for de novo purine and pyrimidine synthesis , Wolbachia has maintained these biosynthetic pathways . Additionally , Wolbachia was shown to lack the ADP/ATP translocases used by other endosymbionts , including the parasitic Rickettsia and Chlamydia , and the mutualist Buchnera , to scavenge for nucleotides from the host [8] . These observations , combined with the evidence for positive selection on genes in this pathway suggest that Wolbachia produce nucleotides not only for internal consumption but also for the host at times when the requirement for DNA synthesis is particularly high , such as during oogenesis and embryogenesis [6 , 8] . During the mitotic proliferation of the B . malayi oogonia , Wolbachia divides rapidly , requiring increased expression of the replication machinery [64] . Correspondingly , we found in F120 significant up-regulation of the Wolbachia DNA replication machinery and genes involved in transcription and translation . In our quest to capture the basis of the endosymbiotic relationship between B . malayi and Wolbachia , we looked at the expression of genes that are part of biosynthetic pathways in Wolbachia that are missing in B . malayi . The absence of the heme , riboflavin and FAD biosynthetic pathways in filaria led to the hypothesis that Wolbachia could be providing these to the filarial host . Evidence for this , however , remains elusive . Several studies have shown that inhibitors of the heme biosynthetic pathway such as 5-aminolevulinate ( ALAD ) and N-methyl mesoporphyrin ( NMMP ) have adverse effects on B . malayi , causing a marked reduction in motility [38] . While this suggests a role for Wolbachia in provisioning heme , adverse effects were also observed on C . elegans , which also lack the heme biosynthetic pathways and are Wolbachia-free , suggesting non-specific effects . We found constitutive expression of all genes in the heme biosynthetic pathway of Wolbachia as well as of heme ABC transporters at almost all sampled worm stages , with the highest expression at the L4 and F120 stages . This suggests the importance of heme synthesis and transport in the symbiotic relationship at these stages [29] . However , it remains unknown if or how B . malayi might receive heme from its bacterial endosymbiont . Examination of the riboflavin and FAD biosynthetic pathways in Wolbachia revealed F120 as the only stage in which all genes are highly expressed , suggesting an increased need in adult females for riboflavin and FAD for embryogenesis . This finding is consistent with the observation that when adult worms are grown in the presence of doxycycline , causing severe adverse effects in embryogenesis , supplementation with riboflavin is able to rescue embryogenesis in adult female worms by approximately 50% [39] . If Wolbachia are indeed provisioning B . malayi with metabolites or nutrients , they would require active secretion and transport systems to do so . We determined that the Sec-dependent and Sec-independent systems appear to be constitutively expressed , especially in adult females . We also find genes in the T4SS , responsible for transport across the outer membrane , to be constitutively expressed across all sampled stages with the exception of wBm0798 in L4 . This suggests that the T4SS is not only active in Wolbachia , but also important in all stages of development that were included in this study . These results confirm the potential for the filarial dependence on Wolbachia products at all stages of the life cycle . Constitutive expression of the heme , phosphate , and lipoprotein ABC transporters is also consistent with the expression of these biosynthetic pathways as phosphate is an essential molecule in nucleotides . The co-expression network analysis of Wolbachia and B . malayi genes was another approach to define interactions . The use of WGCNA to construct a co-expression network for both Wolbachia and B . malayi genes revealed co-expression of important pathways . A number of resulting modules were significantly correlated with Wolbachia density , either positively or negatively . Analysis of GO enrichment of the modules as well as module membership revealed a number of pathways of interest including redox homeostasis and oxidative stress responses as well as the co-expression of DNA repair and replication between the two organisms . Transport mechanisms were also co-expressed , including a nucleoside transporter in B . malayi co-expressed with Wolbachia genes involved in de novo purine and pyrimidine biosynthesis . We plan to expand this co-expression network to include additional stages of parasite development , including molting larvae and microfilaria , in order to better represent the dynamics of endosymbiosis over the entire parasitic lifecycle . In conclusion , our study provides novel insight into the complexity of the interactions between B . malayi and its endosymbiotic bacteria , Wolbachia . We find that it is unlikely that this obligate symbiotic relationship relies on a single process or pathway , but rather on more complex interactions that likely vary over the life cycle of the parasite . This work paves the way for functional validation of the essential role of these associations through the use of RNAi experiments . Elucidation of essential pathways involved in the endosymbiosis between Wolbachia and B . malayi will allow for the identification of novel drug targets .
|
Filarial nematodes currently infect millions of people worldwide and represent a leading cause of disability . Currently available medications are insufficient in reaching elimination of these parasites . Many filarial nematodes , including Brugia malayi , have an Achilles heel of sorts—that is their obligate symbiotic relationship with the bacteria Wolbachia . While it is known that the nematode and the bacteria are co-dependent , the molecular basis of this relationship remains poorly understood . Using deep sequencing , we profiled the transcriptomes of B . malayi and Wolbachia across the life cycle of the parasite to determine the functional pathways necessary for parasite survival provided by the co-expression of nematode and bacterial genes . Defining the mechanisms of endosymbiosis between these two organisms will allow for the exploitation of this relationship for the development of new intervention strategies .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"heme",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"parasitic",
"diseases",
"animals",
"wolbachia",
"parasitology",
"invertebrate",
"genomics",
"nematode",
"infections",
"developmental",
"biology",
"brugia",
"bacteria",
"proteins",
"gene",
"expression",
"life",
"cycles",
"animal",
"genomics",
"biochemistry",
"brugia",
"malayi",
"post-translational",
"modification",
"genetics",
"nematoda",
"biology",
"and",
"life",
"sciences",
"biosynthesis",
"genomics",
"organisms",
"parasitic",
"life",
"cycles"
] |
2017
|
Defining Brugia malayi and Wolbachia symbiosis by stage-specific dual RNA-seq
|
In the G1 phase of the cell division cycle , eukaryotic cells prepare many of the resources necessary for a new round of growth including renewal of the transcriptional and protein synthetic capacities and building the machinery for chromosome replication . The function of G1 has an early evolutionary origin and is preserved in single and multicellular organisms , although the regulatory mechanisms conducting G1 specific functions are only understood in a few model eukaryotes . Here we describe a new G1 mutant from an ancient family of apicomplexan protozoans . Toxoplasma gondii temperature-sensitive mutant 12-109C6 conditionally arrests in the G1 phase due to a single point mutation in a novel protein containing a single RNA-recognition-motif ( TgRRM1 ) . The resulting tyrosine to asparagine amino acid change in TgRRM1 causes severe temperature instability that generates an effective null phenotype for this protein when the mutant is shifted to the restrictive temperature . Orthologs of TgRRM1 are widely conserved in diverse eukaryote lineages , and the human counterpart ( RBM42 ) can functionally replace the missing Toxoplasma factor . Transcriptome studies demonstrate that gene expression is downregulated in the mutant at the restrictive temperature due to a severe defect in splicing that affects both cell cycle and constitutively expressed mRNAs . The interaction of TgRRM1 with factors of the tri-SNP complex ( U4/U6 & U5 snRNPs ) indicate this factor may be required to assemble an active spliceosome . Thus , the TgRRM1 family of proteins is an unrecognized and evolutionarily conserved class of splicing regulators . This study demonstrates investigations into diverse unicellular eukaryotes , like the Apicomplexa , have the potential to yield new insights into important mechanisms conserved across modern eukaryotic kingdoms .
Protozoans belonging to the Apicomplexa were recently combined with two other groups the ciliates and dinoflagellates to form a new monophyletic group called the Alveolata [1] . Unicellular Alveolates arose early in the evolution of eukaryotes , and while the placement of this branch before or after the separation of the plant and animal kingdoms is controversial [2] , [3] , there is no disputing this large collection of protists has an ancient origin . By some estimates the age of Alveolate divergence exceeds a billion years [2] . Modern Alveolates are found throughout the world where they have successfully adopted free and/or parasitic life styles . Many important species in the Apicomplexan subgroup are human pathogens including five members of the Plasmodium genus that cause malaria . Plasmodium falciparum is responsible for nearly a million deaths annually with this disease concentrated in central Africa affecting many children under 5 years old [4] . Exploring the cell biology of these ancient eukaryotes reveals novel features have evolved to ensure cell growth and proliferation of Alveolates in diverse environments . Nothing illustrates this adaption better than the elaborate counting mechanisms that enable Dinoflagellates to switch their cell cycle from binary to multinuclear in different nutrient conditions [5] or have allowed Apicomplexans to reproduce at scales ( 2 to >25 , 000 divisions ) matched to the choice of host cell [6] . To ensure transmission to the next host , the Apicomplexa devote considerable resources to construct their invasion apparatus at the right time in mitosis , which provides another remarkable example of novel processes integrated into the classic eukaryotic mitosis . Given the peculiar features documented it is tempting to speculate that control of Alveolate replication is different from other eukaryotes . When key checkpoint proteins are not detected in genome sequence [6] this conclusion seems at least partially correct . Yet , most Alveolate cell cycles have some transitions that are similar to division cycles of eukaryotes from multicellular kingdoms . Interphases comprised of conventional G1 and S phases ( G2 maybe less conserved ) are common [5] , [6] , [7] and cyclin-CDK factors present in Alveolate genomes [6] , [8] are assumed to check the fidelity of cell cycle transitions as they do in other eukaryotes [9] . Segregation of nuclear chromosomes in Alveolates requires microtubule organizing structures and the timing of mitosis likely utilizes some version of the anaphase promoting complex whose components are also found in Alveolate genome sequence [6] . These distinct cell cycle views of old and new suggest we have much to learn about the mechanisms working in these protozoa to achieve their diverse replication schemes . Recent reflections on the cell cycles of fungi and animals [9] , indicates we should expect regulatory factor divergence even where protozoan cell cycles appear to have preserved the same network topology working to control multicellular eukaryote division . Not surprising , all expected levels of protein conservation from pan-eukaryote to species-specific growth factors are emerging from unbiased genetic screens now successfully developed for the study of Toxoplasma gondii cell division ( Suvorova and White , unpublished ) [8] , [10] , [11] . Model organisms among these ancient protozoa are valuable because they offer insight into the flexibility possible in the eukaryotic cell cycle and at the same time will help define core cell cycle mechanisms preserved since the first eukaryote . Toxoplasma has emerged as a principal genetic system from the Apicomplexa with particular strengths in the study of cell cycle mechanisms [8] . The binary replication of the Toxoplasma tachyzoite is relatively simple composed of major G1 , S , and mitotic phases [12] . Internal budding that is a hallmark of the apicomplexan division begins late in the tachyzoite S phase and spans the classical mitotic events necessary for chromosome segregation ultimately resulting in two infectious daughter parasites [6] , [7] . There is a rough demarcation of old and new cell cycle processes in the two halves of the tachyzoite division cycle that was borne out in the Toxoplasma , and also the Plasmodium falciparum , cell cycle transcriptomes [13] , [14] suggesting this is a general expression scheme among the Apicomplexa . The cell cycle transcriptome of these protozoa is characterized by a serial progression of cyclical mRNAs with canonical growth factors reaching a maximum in G1 followed by peak expression of dozens of specialized genes needed for building the invasion apparatus in S and mitotic phases [13] . How this transcriptional cascade is regulated is unknown nor do we understand the intersection between cell cycle gene expression and the checkpoint mechanisms orchestrated by cyclin-CDK factors present in these protozoa . Here we describe a cell cycle mutant in Toxoplasma that is defective in a fundamental eukaryotic cell cycle function . This defect results in a rapid growth arrest in the G1 phase , which is not recoverable when cells are shifted back to a permissive temperature . The novel RNA binding protein discovered by genetic rescue is conserved in many modern eukaryotes such that the human ortholog can fully replace the function of the Toxoplasma mutant protein . We show that this new family of proteins is required for gene expression where it promotes proper mRNA splicing .
The cell cycle of apicomplexan protozoa has elements in common with other eukaryotes as well as features such as internal daughter budding that are unique to this parasite family . The molecular basis of cell division in these important protists remains understudied compared to other biochemical processes . To expand research efforts in this area , we recently generated a large collection of temperature-sensitive ( ts ) mutants in order to identify essential mechanisms in Toxoplasma replication [8] . An isolate from this collection ( mutant 12-109C6 ) rapidly stopped cell division when shifted to the non-permissive temperature ( 40°C ) indicating a key growth factor was mutated in this parasite . We compared the growth of mutant 12-109C6 to the parental strain ( RHΔhxgprt ) , and while the rate of division in parental parasites increased with temperature , the mutant clone immediately arrested at temperatures above 37°C ( Figure 1A ) . Many mutant parasites at 40°C were unable to complete a single division in the host cell indicating the defect was expressed quickly following temperature shift . To understand whether clone 12-109C6 was a cell cycle mutant , we examined changes in genomic DNA distributions with respect to incubation temperature . These results revealed the genomic content of the mutant clone grown at 40°C had a higher proportion of haploid ( 1N ) parasites ( Figure 1B ) consistent with arrest in the G1 phase . Similar to other eukaryotes , Toxoplasma tachyzoites duplicate their centrosome at the G1/S boundary [12] providing an internal subcellular marker to further validate the G1 phase arrest ( Figure 1C ) . As expected , mutant parasites incubated at 34°C ( Figure 1C , upper panel ) duplicated their centrosomes consistent with the known S/M distribution of asynchronous populations ( Figure 1D , 48% duplicated ) [15] . By contrast , mutant populations exposed to 40°C ( lower IFA panel ) contained mostly single centrosomes ( Figure 1D , 78% singles ) and the absence of cytokinesis ( i . e . internal daughters ) in these arrested parasites ( Figure 1C ) further supports an arrest in the G1 phase . Altogether , these results confirm clone 12-109C6 is a conditional growth mutant carrying a defect in a mechanism needed for G1 to S phase progression . We have utilized a forward genetic approach to link the G1 defect in 12-109C6 parasites to the responsible chromosome mutation [8] . Mutant parasites were complemented with Toxoplasma cosmid libraries ( RH strain genomic DNA ) under pyrimethamine selection at 40°C followed by identification of the integrated cosmid insert via marker rescue techniques [8] . Recovered cosmid insert fragments were sequenced and mapped to chromosome VIIa between 2 , 771 , 038 bp and 2 , 813 , 240 bp ( Figure 2A ) ; this chromosome region contains six predicted genes ( genes #1-6: TGGT1_017830/dynein 1 , beta heavy chain , TGGT1_017840/citrate synthetase , TGME49_003100/hypothetical protein , TGGT1_017850/hypothetical protein , TGGT1_017860/RRM domain-containing protein and TGGT1_017870/conserved hypothetical protein ) . The locus was resolved to a single gene by a new round of complementation with genomic fragments spanning genes #3 , #4 , or #5 ( Figure 2A ) . High temperature rescue of mutant 12-109C6 was only observed in parasites transfected with fragments containing gene #5 . The protein encoded by gene #5 ( TGGT1_017860 ) is one of 86 genes in the Toxoplasma genome predicted to encode proteins with one or more RRM domains ( see Table S1 for a full list of RRM containing genes in Toxoplasma ) . The current annotation for gene #5 predicts a 302 amino acid ( aa ) polypeptide with a single RRM domain flanked by N- and C-polypeptide tails ( confirmed by cDNA sequencing , not shown ) . As the first functionally described RRM protein in Toxoplasma , we have designated this gene and protein as TgRRM1 . Sequencing of the mutant allele of TgRRM1 ( TGGT1_017860 ) revealed a single thymidine to adenine transversion at nucleotide position 505 of the coding sequence that led to exchange of tyrosine 169 for an asparagine residue ( Figure 2B ) . The mutation lies in the RRM domain affecting one of the conserved aromatic residues in the RNP1 subdomain predicted to involve RNA binding ( Figure S1A ) . A plasmid construct expressing temperature-sensitive allele of gene #5 ( see Dataset S1 and Material and Methods for all construct designs ) failed to rescue mutant 12-109C6 at 40°C confirming the Y/N non-synonymous mutation in this protein is responsible for temperature sensitivity ( Expression constructs , Figure 2A ) . The rapid and lethal growth arrest of mutant 12-109C6 suggests TgRRM1 has a vital , if unknown role in cell division . To build clues to function , we first explored how TgRRM1 is expressed in the parasite cell cycle by introducing an epitope tagged version ( wt-TgRRM1myc ) controlled by the native TgRRM1 promoter ( primers , constructs and strains , Dataset S1 and Material and Methids ) . The wt-TgRRM1myc protein rescued mutant 12-106C6 at 40°C ( reported at the bottom of Figure 2A ) where we observed maximum expression of the factor concentrated in the nucleus of G1 parasites before protein levels dropped below detection in parasites that were undergoing mitosis and early cytokinesis ( Figure 3A ) . The cell cycle profile of wt-TgRRM1myc followed closely the cyclical timing of the mRNA encoding the native protein that also peaked in G1 ( Figure 3B ) [13] . Importantly , wt-TgRRM1myc downregulation coincided with centrosome duplication , which is thought to mark commitment to DNA replication and entry into S phase [16] . The tight “on-off” cell cycle switching of wt-TgRRM1myc was clearly evident in representative vacuoles ( Figure 3A , bottom 4 image panel ) containing parasites with single versus duplicated centrosomes . In this single microscopic field , wt-TgRRM1myc was expressed only in the cells on the G1 side of the G1/S transition . It is important to note that the cell cycle profile of TgRRM1 likely results from transcriptional and post-transcriptional mechanisms as the encoded mRNA levels never fall below the 70th percentile in the cell cycle transcriptome data ( Figure 3B ) , while protein levels are clearly more dynamic based on IFAs ( Figure 3A ) . These observations are similar to other cell cycle proteins we have studied in tachzyoites [13] . To understand whether the cell cycle timing of this factor was conserved in other Apicomplexa , we raised antiserum against the Plasmodium falciparum ortholog ( PF13_0318 , designated PfRRM1 ) . The specificity of this antiserum and cross-reactivity to TgRRM1 was verified by Western analysis ( Figure S2 ) . Western and mRNA analysis of synchronized P . falciparum merozoites demonstrated that PfRRM1 is constitutively expressed ( Figure 3D and 3E ) , which is different from the periodic profile of TgRRM1 . The cell cycle of the P . falciparum merozoite also has an early G1 phase termed the ring form , and it was in this well recognized stage we detected PfRRM1 concentrated in discrete intranuclear bodies ( Figure 3C ) . PfRRM1 was diffusely distributed in the nucleus of the trophozoite or schizont stages that are the equivalent to S and M/C phases , respectively ( Figure 3C ) . It is possible that in P . falciparum nuclear redistribution is the major cell cycle feature of this related factor . Finally , we explored whether the molecular basis for the mutant 12-109C6 cell cycle defects was caused by changes in either the expression and/or alterations in the cellular localization of the temperature sensitive TgRRM1 protein ( ts-TgRRM1myc ) . We confirmed ts-TgRRM1myc was unable to rescue mutant 12-109C6 at the high temperature employing a plasmid construction based on native promoter expression as the wt-TgRRM1myc construct used above ( see Expression constructs , Figure 2A ) . Immunostaining with anti-myc antibody demonstrated both TgRRM1myc isoforms ( wt versus ts ) were concentrated in the nucleus ( see merged anti-myc and DAPI images , Figure 4 ) , although the level of the ts-TgRRM1myc protein was significantly reduced at the restricted temperature indicating the Y to N change may primarily affect protein stability . In the single vacuole of three parasites in the bottom panel ( 24 h at 40°C ) there was a loss of intravacuolar synchrony that correlated with differential ts-TgRRM1myc expression; the parasite negative for ts-TgRRM1myc had not divided , whereas the other two other parasites still positive for ts-TgRRM1myc had progressed into the second cell cycle following invasion . Thus , the timing of TgRRM1 loss at high temperature likely determines whether a parasite arrests within the G1 period of the present or the next division cycle . Due to the conserved protein sequence , the antiserum raised against recombinant PfRRM1 also binds TgRRM1 , and this reagent was used to confirm the instability of the encoded ts-TgRRM1 protein ( in mutant 12-109C6 ) at the restrictive temperature ( Figure S2 ) . Western analysis showed the original 12-109C6 mutant parasites lost ts-TgRRM1 protein quickly upon shift to 40°C ( Figure S2 ) . The single RRM domain in TgRRM1 may provide interaction with nucleic acids , as this class of proteins is known to bind ssRNA or ssDNA [17] , [18] . The RNA binding domain of TgRRM1 can be readily modeled ( Figure S1B ) into one of the resolved RRM folds [18] suggesting TgRRM1 likely also binds RNA . Yet , the protein features required for cellular replication are the key functional question , which we explored by determining the minimal TgRRM1 structures able to complement mutant 12-109C6 ( see Figure 5 and Dataset S1 for construct designs ) . It is assumed the RRM domain is critical to function based on the ts-mutation , therefore our deletion study focused on the extended N- and C-terminal tails . Interestingly , TgRRM1 deletions ( TgRRM1DDmyc series ) that truncated the first 76 amino acids ( ΔN , 77–302 aa ) or removed 45 amino acids from the C-terminal end ( ΔCc , 1–257 aa ) were fully capable of rescuing the mutant 12-109C6 at high temperature ( Figure 5A ) . Likewise , we found combining these deletions in a single construct design ( ΔNCc , 77–257aa ) were also functional indicating the N- and most distal C-terminal residues were dispensable for genetic rescue of mutant 12-109C6 . The deletion of an additional 47 residues ( total deletion of 92 residues ) in the C-terminal tail with or without a N-terminal deletion ( ΔCa or ΔNCa ) failed to complement mutant 12-109C6 , and also changed the protein subcellular distribution . The ΔCa- and ΔNCa-TgRRM1DDmyc proteins were not excluded from the nucleus but also did not concentrate there suggesting residues 210-257 of the TgRRM1 C-terminal tail either carry a signal for nuclear retention or alternatively the loss of function indirectly causes the observed change in cellular distribution . TgRRM1 orthologs can be found by mining genomic sequence of other eukaryotic species , including all Apicomplexa for which there is sequence available , but also in multicellular plants and animals ( see Figure S3 for protein alignment ) . The overall similarity in this protein family is moderate with the RRM domain showing the highest conservation ( 80–90% similarity ) , while N- and C- terminal extensions when present are typically unique even in orthologs from related species . Secondary structure analysis of divergent TgRRM1 orthologs illustrates protein similarity beyond the primary sequence ( Figure S1C ) including the positively charged region C-terminal to the RRM domain we found was critical for function and nuclear retention ( Figure 5A ) . The mutant 12-109C6 offers a unique opportunity to explore whether these minimal structures conserved in TgRRM1 orthologs are sufficient for function . We tested PfRRM1 from the related apicomplexan P . falciparum and RBM42 from human cells ( see construct designs Figure 5B ) . The P . falciparum PfRRM1 differs from TgRRM1 by having shorter N- and C-terminal tails flanking the central RRM domain . PfRRM1DDmyc was unable to fully rescue mutant 12-109C6 , although expression levels of PfRRM1DDmyc and nuclear retention ( Figure 5B ) were comparable to wt-TgRRM1DDmyc ( Figure 5A ) . Interestingly , the mutant isolates expressing PfRRM1DDmyc did not immediately growth arrest at 40°C like the original ts-mutant strain ( Figure 1 ) suggesting PfRRM1DDmyc was partially complementing the defect ( not shown ) . We then added the Toxoplasma N- ( 0–76 residues ) and C-terminal ( 258–302 residues ) tails to the core PfRRM1 domain ( see Figure S4 for chimera designs ) and achieved full genetic rescue of mutant 12-109C6 with the chimeric protein Tg/PfRRM1DDmyc ( Figure 5B ) . These results are consistent with the conservation of the Toxoplasma and Plasmodium RRM domains ( 93% similarity/70% identity , Figure S3 ) , although they also demonstrate nuclear retention alone is not sufficient to achieve functional complementation . Human RBM42 shares 64% identity and 89% similarity to TgRRM1 in the RRM domain . RBM42 has an extended N-terminus that is structurally similar to the N-terminal extension of TgRRM1 in the residues immediately upstream of the RRM domain ( Figure S1C ) . The slightly shorter C-terminal tail of RBM42 nonetheless preserves the charged nuclear retention domain . Remarkably , RBM42DDmyc protein ( residues 33–480 aa ) was able to fully rescue the 12-109C6 mutant at high temperature with a restoration of a wild type growth rate equivalent to complementation with wt-TgRRM1DDmyc ( Figure 5B ) . Rescue of ts-TgRRM1 mutants was specific for RBM42 as genetic complementation of mutant 12-109C6 failed using the non-homologous human RRM proteins , CCR4-NOT transcription complex subunit 4 ( NP_037448 . 2 ) and eukaryotic translation initiation factor 3 ( NP_003742 . 2 ) , despite the proper localization of the CCR4-NOT factor to the parasite nucleus ( data not shown ) . The rapid cell cycle arrest of mutant 12-109C6 , and the potential of TgRRM1 to bind RNA , led to us to look for clues to TgRRM1 function in whole-cell gene expression . Total RNA was isolated in duplicate from mutant 12-109C6 parasites grown at permissive and non-permissive temperatures , converted to cRNA and used to hybridize a custom Affymetrix GeneChip with multiple probes for ∼8 , 000 Toxoplasma genes ( http://ancillary . toxodb . org/docs/array-tutorial . html ) . A total of 473 mRNAs was statistically altered ( fold change ≥4 up or down ) when mutant 12-109C6 was shifted to the higher temperature . Most transcriptome changes involved decreases in mRNA levels occurring by 6 h post-temperature shift and affected many different pathways of cell metabolism ( see Dataset S2 for full gene list ) . Our recent analysis of the Toxoplasma cell cycle transcriptome identified two major waves of transcription with peak mRNA levels associated with G1 or S/M subtranscriptomes [13] . In comparison to this cell cycle transcriptome , we found 261 mRNA profiles altered in the 12-109C6 mutant that were also periodic mRNAs in the tachyzoite division cycle . A heat map of the 261 mRNAs reveals both G1 and S/M transcripts are included ( Figure 6A; G1 mRNAs peak in cluster 1 , S/M mRNAs peak in cluster 2 ) , and in nearly every instance , these mRNAs from either half of the cell cycle were strongly downregulated in temperature restricted 12-109C6 parasites ( Figure 6A compare 34°C versus 40°C ) . A reduction in the levels of mRNAs that peak in the tachyzoite S/M periods was expected given the G1 arrest of mutant 12-109C6 , however , reductions of G1 peak transcripts were a surprise . Importantly , mRNA levels were not extensively downregulated in asynchronously dividing 12-109C6 parasites ( grown at 34°C ) ( Figure 6A , lane 1 ) or in an unrelated mutant ( ts-mutant 12-88A5 ) previously shown to rapidly arrest in the G1 phase at 40°C ( Figure 6A , lane 5 ) [8] , [13] . In the set of altered mRNAs in mutant 12-109C6 , a few transcripts encoded proteins with possible functions related to TgRRM1 ( Figure 6B ) . Among the abundant family of the novel RNA binding proteins in Toxoplasma ( 86 total , Table S1 ) , four mRNAs were downregulated , while two were upregulated ( Figure 6B , TGME49_ numbers of these genes are indicated ) . Two proteins associated with the U2 spliceosome and the splicing factor 3b subunit 10 were also significantly downregulated ( Figure 6B ) . To examine whether splicing was affected in mutant 12-109C6 parasites at high temperature , we surveyed a selection of genes including key RNA polymerase II subunits using primers spanning an intron ( Figure 6C ) . Semi-quantitative PCR analysis showed the accumulation of pre-mRNA for all genes , which was readily detected as early as 6 h following the shift of 12-109C6 parasite cultures to 40°C ( Figure 6C , lane 6 h ) . Pre-mRNA was not amplified from total RNA obtained from 12-109C6 parasites grown at 34°C or from 12-109C6 parasites complemented with wt-TgRRM1myc grown at 40°C ( Figure 6C , lane 24* ) . The stability of unspliced mRNA was variable with pre-mRNA levels equal to the matched spliced mRNA for some genes , whereas pre-mRNA levels were significantly lower than the mature mRNA in others . The accumulation of unspliced mRNA would contribute to hybridization signals on the microarrays suggesting we may have underestimated the influence of TgRRM1 on global gene expression . We have addressed this question by deep sequencing RNA samples from 12-109C6 parasites and the complemented strain ( Figure 7 , Table S2 and Dataset S3 ) . We calculated the total number of RNA reads aligning to either exons or introns ( based on ToxoDB 6 . 1 predictions ) and determined the ratio of intronic to exonic hits ( I/E ) for each spliced gene under conditions of mutant parasites grown at 34°C versus 40°C , and determined whether genetic rescue restored I/E ratios at the higher temperature ( Figure 7 and Table S2 ) . A spreadsheet listing the I/E values for the 5 , 833 intron-containing genes under the conditions examined along with the overall abundance levels for all RNAs detected ( including single exon genes ) is included in Dataset S3 . As expected , the overall I/E ratios were dramatically increased when mutant 12-109C6 parasites were shifted to 40°C ( Figure 7A ) . This effect was genome-wide ( total of 5 , 204 mRNAs affected , Dataset S3 ) and is consistent with a global defect in mRNA intron splicing . Few genes showed an increase in splicing ( 184 total ) and these were largely very low expressed and enriched for genes expressed in other developmental stages . A few genes with decreased I/E ratios had wrong gene models ( not shown ) . Thus , TgRRM1 appears to be a factor required for general mRNA splicing , and consistent with this view , the genetic rescue of the splicing defect at high temperature was nearly complete ( Figure 7B , >98% of I/E ratios in the 40°C complemented sample were fully or partially restored to the values observed in the 34°C RNA samples ) . There was a statistically nonsignificant ( p>0 . 05 ) increase in the complemented strain I/E ratios grown at 40°C compared to the mutant grown at 34°C ( Table S2A ) . The I/E ratios of genes shown to be regulated during the course of the cell cycle were also examined . Genes were categorized as being either up-regulated during either chromosome synthesis and mitosis ( S/M ) phases ( 1 , 217 genes ) or during growth in the G1 phase ( 1 , 635 genes ) [13] . In general , I/E values of G1 and S/M phase genes behave in much the same way as the general population , increasing significantly when grown at 40°C , but recovered by complementation with the TgRRM1 wt-allele ( Figure 7C and 7D , Table S2B ) . Due to the essential nature of this mechanism to mRNA splicing , we were not surprised to find steady state levels for all genes , including those without predicted introns , are impacted by this defect ( Figure S5 and Dataset S3 ) . There is a much larger amount of variance in the mutant parasites grown at 40°C than the parasites grown at 34°C ( Figure S5A ) . In overall agreement with the microarray results ( Figure 6 ) , mRNA expression was severely reduced following the shift of mutant parasites to the non-permissive temperature . This was true for cell-cycle dependent genes as well single exon genes ( Figure S5B ) . The reduction in mRNA was reversed by complementation of the mutant with the wt-TgRRM1 allele ( Figure S5A and S5C ) . The global mRNA splicing defect caused by TgRRM1 downregulation in mutant 12-109C6 raised the possibility this factor has a direct involvement in spliceosome function . The cellular components that assemble into active splicing machinery include a number of RRM proteins , although there is no report of RBM42 or its orthologs serving as an integral component of the splicing machinery . To explore this possibility , co-immunoprecipitation followed by a comprehensive proteomic analysis was performed on a temperature resistant clone rescued by wt-TgRRM1myc complementation ( Figure 2A ) and compared to the original 12-109C6 mutant as a negative control . Purified complexes from whole cell lysates ( Figure 8A ) or nuclear extracts ( see Dataset S4 ) were resolved by electrophoresis and individual gel slices subjected to mass spectrometry analysis . These experiments identified 14 and 21 unique proteins in the whole lysate and nuclear extracts , respectively , including TgRRM1 itself ( see Dataset S4 ) . Proteins interacting with TgRRM1 are encoded by mRNAs with a wide range of abundance from low expression ( 50th percentile ) to highly abundant transcripts ( 90th percentile ) ( see Table 1 for examples ) . TgRRM1 containing complexes were highly enriched for spliceosome factors with multiple peptides recovered and up to 12% sequence coverage for some splicing factors; 15 out of 23 co-precipitated proteins are known components of the U4/U6 or U5 small ribonucleoprotein particles ( see Dataset S4 for full details ) . Five core components of U4/U6 snRNP , seven components of U5 snSNP , and three accessory proteins were identified ( all indicated in solid black in Figure 8B ) . Few proteins and no splicing factors were identified in negative controls . The 14 splicing factors identified in these pull-downs including the 10 proteins recovered from both extracts are listed in Table 1 . The selective co-precipitation of TgRRM1 with U4/U6 and U5 snRNPs suggests a role for this protein in spliceosome function at the level of complex B formation ( Figure 8B ) .
Combining chemical mutagenesis with forward genetics to identify regulators of the eukaryotic cell cycle was introduced more than 40 years ago [19] , [20] and remains a direct approach to uncover novel protein mechanisms affecting cell division . More than a decade ago we adapted this strategy to study unique mechanisms in Toxoplasma division , yet the impact of our studies is not restricted to apicomplexan cell cycles . Important proteins spanning evolution both recent and long conserved are emerging from the study of this collection of ts-mutants . Here we describe the discovery of a conserved protein ( TgRRM1 ) that is required for cell cycle progression through an essential role in mRNA splicing . The chemical mutant that provided these insights rapidly arrests in the G1 phase when shifted just 6°C higher than a safe growth temperature ( 34°C ) due to a point mutation in the TgRRM1 gene sequence . The TgRRM1 gene expression mechanism extends beyond the confines of apicomplexa biologic inventions as the human ortholog , RBM42 fully rescued the growth of this mutant and its splicing defect . Splicing of mRNA is a fundamental process for gene expression in eukaryotes . Recognition of the splicing signals and removal of non-coding regions of pre-mRNA ( introns ) is carried out by a megadalton multi-subunit complex called the spliceosome . More than 150 proteins and five snRNAs ( U1 , U2 , U4 , U5 , and U6 ) are required for a sequential assembly and activation of the spliceosome ( for review see [21] , [22] , [23] , [24] ) . Management of the complexity of spliceosome protein components is solved by pre-packing groups of proteins and RNA into building blocks that are detectable as stable snRNP particles ( see Fig . 8B for diagram of spliceosome assembly ) . U1 snRNP particles are first recruited to 5′ splicing site ( complex E ) , followed by the U2 snRNP association with the 3′ splicing site ( complex A ) . It is known that assembly of spliceosome on the active splicing site requires the tri-snRNP that is formed from the U4/U6 di-snRNP and U5 snRNP . Together U1 and U2 snRNPs and the tri-snRNP form the pre-spliceosome complex B [24] , [25] . Finally , several rearrangements within the complex B complete spliceosome activation ( complex B* ) . The tri-snRNP complex contains a few unique components known to associate only with the assembled tri-snRNP particle prior to integration into the pre-splicesome such as SAD1 and SART1 [26] . Unlike U1 and U2 snRNPs , the tri-snRNP complex also undergoes dramatic rearrangements during each splicing cycle . U4 and U6 RNAs undergo re-annealing and modification before re-assembly into the U4/U6 di-snRNP . This renewed complex in turn is integrated into a functional tri-snRNP complex . We identified many components of the tri-snRNP complex , but not U1 or U2 snRNPs in TgRRM1 pull-downs including the two tri-snRNP-specific factors , the ortholog of hSAD1 ( TGME49_094360 ) and hSART1 ( TGME49_118140 ) . These results indicate TgRRM1 has a role in formation of the pre-splicesome complex B particle that is downstream of assembly of the U4/U6 and U5 snRNPs particles . This role for TgRRM1 could also involve post-spliceosome dissassembly and recycling of components into new active spliceosomes . Recent reports demonstrate PRP4 kinase has a role in tri-snRNP formation through phosphorylation of PRP6 and PRP31 [27] with the loss of the PRP4 kinase activation step leading simultaneously to defects in mRNA splicing and specific cell cycle arrest at G1/S and G2/M transitions [27] , [28] . In our TgRRM1 pull-down experiments , we also detected both of these factors necessary for tri-snRNP formation; U5 snRNP-specific PRP6 ( TGME49_005220 ) and U4/U6 snRNP-specific PRP31 ( TGME49_044100 ) . Importantly , defects in TgRRM1 leads to a specific cell cycle arrest similar to what occurs in yeast PRP4 kinase mutants [28] with the primary block in Toxoplasma at the G1 to S phase transition , which may be related to the absence of a G2 period in the parasite asexual cell cycle [6] , [7] . It is intriguing to speculate that the essentiality of TgRRM1 is related to the limiting step of tri-snRNP assembly thought to link spliceosome formation to cell cycle progression . Although compositions of the specific snRNPs and spliceosomal complexes ( E , A , B and others ) are thought to be well established , new factors are occasionally reported in various proteomics studies revealing the likelihood there is greater complexity in the splicing machinery than currently outlined in reviews of mRNA splicing [29] , [30] , [31] . This study presents a new essential protein that associates with tri-snRNP in Toxoplasma tachyzoites as a case in point . Whether TgRRM1 should be considered an integral component of the U4/U6 . U5 tri-snRNP complex or a transient regulator of complex assembly will require further dissection of the mechanism involved . Interestingly , data mining of proteomics studies in human and Drosophila melanogaster reveal a transient and selective association of the TgRRM1/RBM42 orthologs with spliceosome complex B [32] , [33]; the impact of this interaction on cellular mechanisms was not determined . Here the discovery of a role for TgRRM1 in the assembly of the spliceosome opens a new chapter in understanding how this protein family regulates mRNA splicing . The ultimate goal of cells in G1 phase is to prepare for the next round of cell division , which a cell commits to once chromosome synthesis begins in S phase . The G1 phase is the most variable , and often the longest phase of the cell cycle , and appears to be well conserved in evolution . Such universal steps as first making proteins needed for gene expression such as transcription , splicing , and translation factors followed by building the key DNA synthetic factors appears to be an ancient G1 order [13] , [34] , [35] , [36] . Despite the functional conservation of G1 , the regulatory network controlling G1 progression is not necessarily shared . We noted previously [13] the traditional order of gene expression unfolding in the G1 phase of Toxoplasma tachyzoites was not governed by the retinoblastoma/E2F transcriptional network of higher eukaryotes [37] or by the alternative yeast G1 network involving SBF/MBF transcription complexes [38] as all these proteins are absent in the Apicomplexa . If checkpoint G1 regulators differ between eukaryotes , a more fundamental mechanism must be responsible for the traditional biosynthetic progression in G1 . In this study we have discovered an essential splicing factor conserved across a billion years of evolution whose function is required for G1 progression into S phase in Apicomplexans . It is well documented that splicing and transcription are intimately connected [22] , [39] , [40] and there is a heavy dependence on gene expression in the G1 period ( 50% more mRNAs peak in G1 than S/M in Toxoplasma ) [13] . Nearly all the splicing factors interacting with TgRRM1 follow the G1 peak expression timing of TgRRM1 itself ( see Table 1 ) . This timing is not coincidental as transcription and splicing activity is not equal across the cell division cycle in higher eukaryotes . In the open mitosis of animal cells , transcription is repressed from prophase to early telophase , at which time various splicing components appear to be held inactive in discrete compartments [39] . Transcription recommences in the late telophase with reassembly of the nuclear envelope and this is accompanied by the import of the splicing machinery stored in these compartments . Some details of this mechanism do not apply to Toxoplasma where chromosome segregation is endomitotic . However , we can not rule out the export/import of key splicing factors and there is a significant downregulation of mRNAs during mitosis/cytokinesis [13] providing evidence for temporal regulation of transcription/splicing in the Toxoplasma cell cycle . The results shown here suggest there is another simple model for regulating cell cycle progression through the strict timing of key splicing regulators like TgRRM1 . In the absence of other cell cycle network controls [6] , [7] , [41] , the “just-in-time” delivery of essential proteins appears to be a dominant method for regulating outcomes in Apicomplexa replication [14] . Thus , it is possible that basic timing mechanisms that achieve sufficient coordinate control in the ancient Apicomplexa may have since been modified to more nuanced and complex strategies in higher eukaryotes or were lost when the dependent mechanism was reduced such as in splicing of yeast mRNAs that lacks any ortholog of TgRRM1 .
Parasites were grown in human foreskin fibroblasts ( HFF ) as described [42] . All transgenic and mutant parasite lines are derivatives of the RHΔhxgprt parasite strain . Temperature sensitive clone 12-109C6 was obtained by chemical mutagenesis of the RHΔhxgprt strain [8] . Growth measurements were obtained using parasites pre-synchronized by limited invasion as previously described [10] , [43] . Vacuoles in the infected plates were evaluated over various time periods with average vacuole sizes determined at each time point from 50–100 randomly selected vacuoles . Plasmodium falciparum NF54 parasites were cultured at 37°C in 5% hematocrit ( O-positive blood ) RPMI1640 ( Life Technologies ) , 0 . 5% Albumax , or 10% human AB serum . Confluent HFF cultures on the glass coverslips were infected with parasites for the indicated time . Cells were fixed in 3 . 7% paraformaldehyde , permeabilized in 0 . 25% Triton X-100 and blocked in 1% BSA in PBS . Incubations with primary antibody ( 1 h ) followed by the corresponding secondary antibody ( 1 h ) were performed at room temperature with DAPI ( 0 . 5 µg/ml ) added in the final incubation to stain genomic DNA . The following primary antibodies were used at the indicated dilutions: mouse monoclonal anti-myc ( Santa Cruz Biotechnology , Santa Cruz , CA ) , anti-centrin 26-14 . 1 ( kindly provided by Dr . Jeffrey Salisbury , Mayo Clinic , Rochester , NY ) and anti-IMC1 ( kindly provided by Dr . Gary Ward , University of Vermont , VT ) at 1∶1000 . Serum raised against the conserved human centrin 1 ( 26-14 . 1 ) was previously shown to cross-react with the Toxoplasma centrin ortholog [10] , [44] . All Alexa-conjugated secondary antibodies ( Molecular Probes , Life Technologies ) were used at dilution 1∶1000 . After several washes with PBS , coverslips were mounted with Aquamount ( Thermo Scientific ) , dried overnight at 4°C , and viewed on Zeiss Axiovert Microscope equipped with 100× objective . Images were processed in Adobe Photoshop CS v4 . 0 using linear adjustment for all channels . P . falciparum NF54 parasites were synchronized two times , 8 hours apart , using 5% sorbitol , in two continuous intraerythrocytic cycles . Immunofluorescence assays were performed as described before [45] . In brief , parasite cultures from ring ( 8–16 hours post-invasion ) , trophozoite ( 24–32 hours post-invasion ) , and schizont ( 36–44 hours post-invasion ) stages were fixed overnight in 4% paraformaldehyde and 0 . 0075% glutaraldehyde in RPMI medium , permeabilized in 0 . 1% Triton X-100 in PBS , blocked in 3% bovine serum albumin ( BSA ) , incubated with a 1∶100 dilution of anti-PfRRM1 antibody , probed with a 1∶70 dilution of FITC-labeled goat anti-rabbit secondary antibody ( KPL ) and 10 µg/ml Hoechst 33342 ( Life Technologies ) , and visualized by microscopy . Nuclear DNA content of mutant 12-109C6 parasites was evaluated by flow cytometry using SYTOX Green ( Life Technologies ) staining of tachyzoites as previously described [41] . Briefly , purified parasites were fixed in 70% ethanol and incubated at −20°C for at least 24 h . Fixed cells were stained with 1 µM SYTOX Green in 50 µM Tris pH 7 . 5 and treated with RNase cocktail ( 250 U; dark , room temperature ) at a final concentration of 6×106 parasites/ml . Nuclear DNA content was measured based on fluorescence ( FL-1 ) using a 488 nm argon laser flow cytometer . Fluorescence was collected in linear mode ( 10 , 000 events ) and the results were quantified using CELLQuest v3 . 0 ( Becton-Dickinson Inc . ) . Mutant 12-109C6 was complemented using the ToxoSuperCos cosmid genomic library as previously described [8] , [10] . Briefly , mutant parasites were transfected with cosmid library DNA ( 50 µg DNA/5×107 parasites/transfection ) in twenty independent electroporations . After two consecutive selections at 40°C and than combination of high temperature and 1 µM pyrimethamine , double resistant populations were passed four times before genomic DNA was collected . Cosmid tags were recovered from the genomic DNA by plasmid-rescue protocols [8] . To identify the rescue locus rescued genomic inserts were sequenced using a T3 primer and the sequences mapped to the Toxoplasma genome ( ToxoDB: http://www . toxodb . org/toxo/ ) . To resolve the contribution of individual genes in the locus , Gateway-based entry clones ( Life Technologies ) were build for three ORFs: TGME49_003100 ( predicted by ToxoDB , ver . 4 . 0 , but was not confirmed by later versions ) , TGGT1_017850 and TGGT1_017860 . Each construct included predicted genomic coding region and 1 kb genomic sequence including 5′UTR and 3′UTR regions . Cloning primers with incorporated attB-recombination sites are listed in the Dataset S1 . Constructs were electroporated ( 5 µg DNA/5×107 parasites ) in the mutant 12-109C6 and parasite survival was monitored at 40°C . All ectopic construct designs and primers are listed in Dataset S1 . The wt- and ts-alleles of TGGT1_017860 were cloned into pDEST_gra-myc3X/sag-HXGPRT vector , which provides a C-terminal myc3X tag . Genomic locus including 2 kb of the promoter region was amplified from genomic DNA purified from parent RHΔhxgprt strain and the mutant 12-109C6 using TGGT1_017860_FOR_attB1 and TGGT1_017860_REV_attB2 primers ( Dataset S1 ) . Plasmids were electroporated in the mutant 12-109C6 and selected on the medium with mycophenolic acid and xanthine . Stable clones were established and tested for growth at 40°C . To perform a structure-functional analysis of TgRRM1 , constructs expressing either a full-length protein or its subdomains were designed . PCR products were obtained by amplification from the RHΔhxgprt cDNA library using the following set of primers listed in Dataset S1: TGGT1_017860_FOR_MfeI and TGGT1_017860_REV_SbfI ( wt ) ; TGGT1_017860_FOR_N_MfeI and TGGT1_017860_REV_SbfI ( ΔN ) ; TGGT1_017860_FOR_MfeI and TGGT1_017860_REV_Cc_SbfI ( ΔCc ) ; TGGT1_017860_FOR_N_MfeI and TGGT1_017860_REV_Cc_SbfI ( ΔNCc ) ; TGGT1_017860_FOR_MfeI and TGGT1_017860_REV_Ca_SbfI ( ΔCa ) ; TGGT1_017860_FOR_N_MfeI and TGGT1_017860_REV_Ca_SbfI ( ΔNCa ) . Human RBM42 ( 96–1440 bp of the coding sequence ) was amplified from the cDNA library of HFF cells with RBM42_FOR_MfeI and RBM42_REV_SbfI primers . PCR products were cloned into expression vector ptub- DDL106P-myc3X/sag-CAT using unique MfeI and SbfI cloning sites . The expression constructs had FKBP destabilizing domain ( DDL106P ) [46] and three copies of the myc epitope tag fused to the N-terminus of the polypeptide in an expression context driven by α-tubulin promoter . Chimeric Tg/PfRRM1 protein was designed in a way that the full sequence of PfRRM1 was flanked with the first 76 residues of TgRRM1 at the N-terminus and the last 45 residues of TgRRM1 at the C-terminus ( see Figure S4 ) . Recombinant cDNA of the chimeric protein was synthesized and cloned into ptub- DDL106P-myc3X/sag-CAT vector by GenScript ( GenScript USA ) . Plasmid constructs were transfected in the mutant 12-109C6 and stable transgenic clones were selected with 20 µM chloramphenicol . Presence of the ligand shield 1 stabilized the RRM1 recombinant proteins with DD-domain , however , basal expression from the tubulin promoter ( without shield 1 ) slightly exceeded the endogenous levels of TgRRM1 ( data not shown ) , therefore , shield 1 was not included in any complementation assays and was only used in immunofluorescent analysis to facilitate protein visualization . PfRRM1 coding sequence was amplified from the cDNA library of the blood stage cells of Plasmodium falciparum NF54 strain using primers PF13_0318_FOR_GST_SmaI and PF13_0318_REV_GST_NotI ( Dataset S1 ) . PCR fragment was cloned into expression vector pGEX6T-2 ( GE Healthcare ) that introduced a Glutathione-S-transferase ( GST ) tag to the N-terminus of the protein . Recombinant GST-PfRRM1 was expressed in bacteria , purified on Gluthatione-agarose beads ( Sigma-Aldrich ) , and used to immunize rabbits . Anti-PfRRM1 antibodies were affinity-purified from the serum using nitrocellulose strips with electroblotted PfRRM1 protein . Purified parasites were washed in PBS and collected by centrifugation . Total lysates were obtained by mixing with Leammli loading dye , heated at 95°C for 10 min , and briefly sonicated . After separation on the SDS-PAGE gels proteins were transferred onto nitrocellulose membrane and probed with anti-PfRRM1 antiserum . After incubation with secondary HRP-conjugated anti-rabbit antibody , proteins were visualized in enhanced chemiluminescence reaction . Confirmation of the PfRRM1 rabbit antiserum specificity is shown in Figure S2 . To evaluate PfRRM1 in merozoites , twenty million P . falciparum NF54 parasites each from synchronized ring , trophozoite , and schizont stages were treated with 0 . 1% saponin ( Sigma-Aldrich ) , washed in PBS , resuspended directly in equal volumes of 2× SDS-PAGE sample buffer ( Bio-Rad ) , boiled for 5 min and used for Western analysis . Anti-Histone H3 antibody ( Abcam ) was used as a loading control . Orthologs of TgRRM1 ( TGGT1_017860 ) were identified by BLAST and aligned by ClustalW2 . Analyzed species and corresponding gene ID: Toxoplasma gondii ( TGGT1_017860 ) , Neospora caninum ( NCLIV_021840 ) , Babesia bovis ( BBOV_II004860 ) , Theileria annulata ( TA12210 ) , Eimeria tenella ( ETH_00023710 ) , Cryptosporidium parvum ( cgd1_1070 ) , Plasmodium falciparum ( PF13_0318 ) , Plasmodium vivax ( PVX_114975 ) , Perkinsus marinus ( XP_002778935 . 1 ) , Paramecium tetraurelia ( XP_001433574 . 1 ) , Caenorhabditis elegans ( NP_498090 . 1 ) , Drosophila melanogaster ( NP_649552 . 1 ) , Arabidopsis thaliana ( NP_187100 . 1 ) , Mus musculus ( NP_598454 . 2 ) , Homo sapiens ( NP_077297 . 2 ) . Protein alignments are shown in Figure S3 . RNA was extracted from parasites using the RNeasy kit with β-mercaptoethanol and DNase I treatment ( Qiagen ) . RNA quality was determined using the Agilent Bioanalyzer 2100 ( Santa Clara , CA ) . A total of 500 ng starting RNA was used to produce cRNA using the Affymetrix One-Cycle Kit ( Affymetrix , Santa Clara CA ) . Fragmented cRNA ( 5 µg ) was hybridized to the Toxoplasma gondii Affymetrix microarray according to standard hybridization protocols ( ToxoGeneChip: http://ancillary . toxodb . org/docs/Array-Tutorial . html ) . Two hybridizations were done for each sample type and all data were deposited at NCBI GEO ( GSE43315 ) . Hybridization data was preprocessed with Robust Multi-array Average ( RMA ) and normalized using per chip and per gene median polishing and analyzed using the software package GeneSpring GX ( Agilent Technologies , Santa Clara CA ) . An ANOVA or t-test were run in order to identify genes with significantly greater than random variation in RNA abundance across the data grouped by either temperature or mutant type . Variances were calculated using cross-gene error model , with a p-value cutoff 0 . 05 , and multiple testing correction: Benjamini and Hochberg False Discovery Rate . This restriction tested 8 , 131 probe sets . Ambion MicroPoly ( A ) Purist Kit ( Ambion ) was used for enrichment of transcripts . The SOLiD Total RNA-Seq Kit ( Life Technologies ) was used to construct template cDNA for RNA-Seq following the protocol recommended by Applied Biosystems ( Life Technologies ) . Briefly , mRNA was fragmented using chemical hydrolysis followed by ligation with strand specific adapters and reverse transcript to generate cDNA . The cDNA xfragments , 150 to 250 bp , were subsequently isolated by electrophoresis in 6% Urea-TBE acrylamide gel . The isolated cDNA was amplified through 15 amplification cycles to produce the required number of templates for the SOLiD EZ Bead system ( Life Technologies ) , which was used to generate template bead library for the ligation base sequencing by the SOLiD 4 instrument ( Life Technologies ) . Mapping of SOLiD reads was analyzed using the Whole Transcriptome analysis pipeline in the Applied Biosystems BioScope software ( Life Technologies ) . Mapping was done twice for each sample , once against the genome of T . gondii strain ME49 , and a second time against the genome of T . gondii strain GT1 . Fasta files and corresponding GFF files ( converted to GTF files ) were obtained for both reference genomes from the ToxoDB web site ( www . toxodb . org; Release 6 . 1 ) . BioScope parameter settings were left at the default mapping values . The filter file used consisted of the short adapter sequences . Each pipeline run generated an alignment report and a filtering report . The read counts were summed in a set of * . wig files ( two for each chromosome , corresponding to the strands ) and in an exon rollup file that summed to the exon/gene level based on the coding region locations given in the GTF file . Finally , three BAM files ( binary compressed versions of Sequence Alignment/Map ( SAM ) files ) were created that stored the mapping information for each individual read: one for mapped reads , one for unmapped reads , and one for the filtered reads . An index file was created for the BAM file , storing the location of the mapped reads . Note: the Fasta files from ToxoDB had , in addition to Fasta entries for the 14 chromosomes in T . gondii , Fasta entries for several hundred short floating contigs that could not be placed on a chromosome ( 308 contigs in ME49 strain , 351 such in GT1 strain ) . For use in BioScope , which puts a limit on the number of separate Fasta sequences to which reads can be mapped , these large sets of floating contigs were combined into an artificial 15th chromosome , each contig separated by a set of 60 N's from neighboring contigs . With our SOLiD reads being a maximum of 50 bases long , an interval of 60 N's clearly separated the contigs for mapping use . For each gene , the total number of RNA reads aligning to either exons or introns using the “coverageBed” program was calculated , which is part of the BEDTools suite [47] . In order to estimate mRNA expression levels from the RNA sequencing data ( RPKM normalized according to [48] ) , we determined the number of reads that aligned to the predicted exons for each gene model excluding hits to predicted intron regions ( ME49 strain , ToxoDB release 6 . 1 ) . For those genes predicted to have introns , we calculated the ratio of intronic to exonic reads ( I/E ) . The I/E ratio quantifies the relative abundance of unspliced versus spliced transcripts for each gene . Mature transcripts will have I/E values close to zero , while unspliced message will have relatively higher I/E values . Genomic coordinates of exons were obtained from ToxoDB ( ME49 strain , release 6 . 1 ) and derived intron positions were based on the exon coordinates . If there is no splicing defect , the ratio of I/E values for a particular gene between conditions were expected to be equal ( I/Ey = I/Ex ) . To conservatively measure the extent to which two I/E ratios differ , the shortest line between the point ( I/Ex , I/Ey ) and the diagonal line ( y = x ) was calculated and this value was designated the “distance” between the I/E ratios in Equation 1 . A two-tailed Student's t-test was performed on the distances for each RNA-sequencing experiment in order to test for significant changes between mRNA splicing under different conditions . ( 1 ) For proteomics studies parasites were grown at 34°C for 48 hours . Infected monolayers of HFF cells were washed once with PBS and collected by centrifugation at 4°C for 10 min at 700×g . Cell resuspended in cold PBS were passed sequentially through 20/23/25-guage needles to release parasites from host cells and parasitophosphorous vacuoles . Parasites were pelleted ( 4°C for 15 min at 700×g ) and counted . Whole cell lysates were obtained by lysing 2×109 parasites for 60 min at 4°C in the lysis buffer ( 0 . 1% [v/v] Nonidet P-40 , 10 mM HEPES pH 7 . 4 , 150 mM KCl , with protease inhibitors ) , followed by five cycles of snap freezing in the liquid nitrogen bath and slow thawing in ice-water bath . Lysates were clarified by centrifugation at 12 , 000×g for 30 minutes at 4°C . To make nuclear extracts , 2×109 parasites were first lysed for 5 min on ice in the lysis buffer A ( 0 . 1% [v/v] Nonidet P-40 , 10 mM HEPES pH 7 . 4 , 10 mM KCl , 10% [v/v] glycerol , with protease inhibitors ) , then centrifuged at 6 , 000×g for 8 min at 4°C . Nuclei pellet was further lysed at 4°C in lysis buffer B ( 0 . 1% [v/v] Nonidet P-40 , 10 mM HEPES pH 7 . 4 , 400 mM KCl , 10% [v/v] glycerol with protease inhibitors ) , and subjected to five cycles of freezing-thawing . Nuclear extracts were clarified by centrifugation at 12 , 000×g for 30 minutes at 4°C . Protein extracts were rotated overnight at 4°C with magnetic beads ( MBL International , MA ) containing pre-bound monoclonal anti-myc antibody ( Santa Cruz Biotechnology , Santa Cruz , CA ) . After five washes with cold lysis buffer , bound proteins were eluted in 50 µl Laemmli sample buffer for 5 min at 95°C . Precipitated complexes were separated by SDS-PAGE ( Any kD precast polyacrylamide gel; Bio-Rad ) and stained with Coomassie Blue ( GelCode Blue Stain Reagent , Pierce ) . The entire length of each sample lane was cut into 24 slices that were maintained in MilliQ water for mass-spectrometry . Proteins from a coomassie-stained gel slice were reduced and alkylated with TCEP and iodoacetamide prior to trypsin digest ( 10 ng/µl in 25 mM ammonium bicarbonate , 0 . 1%ProteaseMax ( Promega ) ) for 1 hour at 50°C . Nanospray LC-MS/MS was performed on a LTQ linear ion trap mass spectrometer ( LTQ , Thermo , San Jose , CA ) interfaced with a Rapid Separation LC 3000 system ( Dionex Corporation , Sunnyvale , CA ) . Samples were run sequentially on Acclaim PepMap C18 Nanotrap column ( 5 µm , 100 Å , /100 µm i . d . ×2 cm ) followed by Acclaim PepMap RSLC C18 column ( 2 µm , 100 Å , 75 µm i . d . ×25 cm ) ( Dionex Corp ) . Peptides eluted with gradient mobile phase A ( 2%Acetonitrile/water +0 . 1% formic acid ) and mobile phase B ( 80% acetonitrile/water +0 . 1% formic acid ) were assessed for MS/MS analysis . Raw LC-MS/MS data was collected using Proteome Discoverer 1 . 2 ( Thermo Scientific ) . Created mgf files were used to search the Toxo_Human Combined database using in-house Mascot Protein Search engine ( Matrix Science ) . Final list was generated by Scaffold 3 . 5 . 1 ( Proteome Software ) with following filters: 99% minimum protein probability , minimum number peptides of 2 and 95% peptide probability .
|
The study of eukaryotic cell division has overwhelmingly focused on cells from two branches of evolution , fungal and metazoan , with more distant eukaryotes rarely studied . One exception is apicomplexan pathogens where in the last two decades development of genetic models has been rapid . While not a perfect solution to fill the missing evolutionary diversity , Apicomplexans represent one of the oldest eukaryotic lineages possibly pre-dating the divergence of plant and animal kingdoms . A key to uncovering novel and conserved cell cycle mechanisms in these protists was the development of forward genetic approaches that permit unbiased discovery of essential growth factors . The apicomplexan , Toxoplasma has provided the best resource so far with ∼60 , 000 chemical mutants yielding a collection of 165 temperature-sensitive isolates that arrest in all phases of the parasite cell cycle . Efforts to identify the defective genes in this model are providing insights into the regulatory factors possibly active in the original eukaryote cell cycle , like the mRNA splicing factor discovered in this study .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"infectious",
"diseases",
"model",
"organisms",
"genetics",
"biology",
"genomics",
"microbiology",
"genetics",
"and",
"genomics"
] |
2013
|
Discovery of a Splicing Regulator Required for Cell Cycle Progression
|
In Deinococcus radiodurans , the extreme resistance to DNA–shattering treatments such as ionizing radiation or desiccation is correlated with its ability to reconstruct a functional genome from hundreds of chromosomal fragments . The rapid reconstitution of an intact genome is thought to occur through an extended synthesis-dependent strand annealing process ( ESDSA ) followed by DNA recombination . Here , we investigated the role of key components of the RecF pathway in ESDSA in this organism naturally devoid of RecB and RecC proteins . We demonstrate that inactivation of RecJ exonuclease results in cell lethality , indicating that this protein plays a key role in genome maintenance . Cells devoid of RecF , RecO , or RecR proteins also display greatly impaired growth and an important lethal sectoring as bacteria devoid of RecA protein . Other aspects of the phenotype of recFOR knock-out mutants paralleled that of a ΔrecA mutant: ΔrecFOR mutants are extremely radiosensitive and show a slow assembly of radiation-induced chromosomal fragments , not accompanied by DNA synthesis , and reduced DNA degradation . Cells devoid of RecQ , the major helicase implicated in repair through the RecF pathway in E . coli , are resistant to γ-irradiation and have a wild-type DNA repair capacity as also shown for cells devoid of the RecD helicase; in contrast , ΔuvrD mutants show a markedly decreased radioresistance , an increased latent period in the kinetics of DNA double-strand-break repair , and a slow rate of fragment assembly correlated with a slow rate of DNA synthesis . Combining RecQ or RecD deficiency with UvrD deficiency did not significantly accentuate the phenotype of ΔuvrD mutants . In conclusion , RecFOR proteins are essential for DNA double-strand-break repair through ESDSA whereas RecJ protein is essential for cell viability and UvrD helicase might be involved in the processing of double stranded DNA ends and/or in the DNA synthesis step of ESDSA .
The bacterium Deinococcus radiodurans is extremely resistant to treatments such as ionizing radiation and desiccation . This resistance can be correlated with the ability of D . radiodurans to reconstruct a functional genome from hundreds of radiation or dessication-induced chromosomal fragments , while the genomes of most organisms are irreversibly shattered under the same conditions . The rapid reconstitution of an intact genome is dependent on extended synthesis-dependent strand annealing ( ESDSA ) and recombination [1] , [2] . It was proposed that , following severe DNA damage , the fragmented DNA end is recessed in a 5′–3′ direction , liberating single stranded 3′ overhangs which , through RecA- and RadA-mediated strand invasion , prime DNA synthesis on overlapping fragments [2] . DNA synthesis is initiated by Pol III and elongated by Pol I or by Pol III and the newly synthesized single-strands anneal to complementary single stranded extensions forming long double stranded DNA intermediates which are assembled into intact circular chromosomes by RecA-mediated homologous recombination [2] . Though the dependence of ESDSA on RecA , Pol I , and Pol III activities is well documented [1] , [2] , little is known about the cellular factors required for the first steps of this process ( i . e . the formation of the single stranded 3′ overhangs which promote RecA/RadA - dependent strand invasion to prime DNA synthesis ) . Three enzymatic activities are required for presynaptic processing of double stranded DNA ends in the model bacterium Escherichia coli: a helicase , a 5′-3′exonuclease , and a mediator function for efficient RecA filament formation onto ssDNA ( see for reviews [3]–[5] ) . All these activities are carried out by the RecBCD complex ( or its functional homolog AddAB ) which is the major component for initiation of recombinational repair of DNA double-strand-breaks ( DSB ) in wild-type cells . However , if RecBCD is inactivated , an alternate pathway , the RecF pathway , promotes recombinational DSB repair [6]–[10] in cells containing mutations in sbcB ( suppressor of recBC ) , which encodes the 3′-5′ exonuclease I , and in sbcC ( or sbcD ) [11] . This pathway comprises the 5′-3′ single-strand DNA exonuclease RecJ , the RecQ helicase and the RecF , RecO and RecR proteins that act together to promote loading of RecA onto single stranded DNA . Whereas examination of the phylogenetic distribution of RecBCD and AddAB complexes revealed that one or the other complex is present in most sequenced bacteria , D . radiodurans is naturally devoid of these two complexes but does encode a RecD homologue [12] . RecD protein was shown to be present in the absence of RecBC not only in D . radiodurans , but also in firmicutes and Streptomyces [13] . The deinococcal RecD protein is expressed and active as a DNA helicase [14] . Further work is required to assign RecD protein to a specific DNA repair pathway because conflicting data have been published concerning the in vivo role of RecD in radioresistance [15]–[16] . D . radiodurans possesses homologs of the key components of the RecF pathway: RecJ ( DR1126 ) , RecQ ( DR1289 ) , RecF ( DR1089 ) , RecO ( DR0819 ) , and RecR ( DR0198 ) suggesting that the RecF pathway is the main recombinational repair pathway in this organism , as observed in other bacteria that lack RecBCD homologs [13] . D . radiodurans also lacks homologs of the SbcB nuclease , an inhibitor of the RecF pathway in E . coli . Moreover , it was shown that expression in trans of the SbcB protein from E . coli renders D . radiodurans cells radiation-sensitive [17] . In this paper , we investigate the role of the D . radiodurans proteins belonging to the RecF pathway in ESDSA and/or homologous recombination . We demonstrate that RecJ exonuclease is an essential protein for cell viability . We show that the RecF , RecO , RecR proteins as well as the RecA protein are absolutely required for massive DNA synthesis during DSB repair whereas RecQ appears to be substituted by the UvrD helicase to play a role in this process . We propose that RecJ , in conjunction with UvrD , could generate the single stranded tails on which RecFOR will stimulate RecA loading . Interestingly , an intact genome could be slowly reconstituted in the absence of RecA , RecF , RecO or RecR , suggesting alternate DSB repair through non-homologous end joining ( NHEJ ) and/or single-strand annealing ( SSA ) .
To determine the importance of the RecFOR pathway in DSB repair and radioresistance in D . radiodurans , we replaced the coding regions of key genes belonging to this pathway ( recJ , recQ , recF , recO , and recR ) with an antibiotic resistance cassette . The deletion-substitution alleles were constructed in vitro using the tripartite ligation method [18] and introduced by transformation into D . radiodurans to replace the corresponding wild-type alleles via homologous recombination . Because D . radiodurans contains from 4 to 10 genome equivalents [19] , [20] , the transformants were extensively purified on selective media in order to obtain the mutant homogenotes whose purity was verified by PCR . Whereas only few rounds of purification on selective antibiotic plates sufficed to obtain ΔrecQ , ΔrecF , ΔrecO and ΔrecR homogenotes ( see Figure S1 ) , in the case of recJ , the wild-type allele was present together with the ΔrecJ allele even after seven rounds of purification of three independent candidates ( Figure 1 ) , suggesting that RecJ protein is essential for cell viability . To obtain positive evidence for the essentiality of the recJ gene , we used the new diagnostic assay described by Nguyen et al [21] . For this purpose , the recJ gene was cloned onto a prepUTs vector thermosensitive for replication in D . radiodurans [21] . The sequence of DR1126 ( recJ ) in strain ATCC 13939 ( GenBank , accession number QG856645 ) was found to differ from the DR1126 published sequence [22] . An additional G was found 7 nucleotides upstream the published putative GTG initiation codon and another additional G was found 58 nucleotides before the published putative TGA STOP codon giving rise to a RecJ protein containing 705 aa ( versus 684 aa in the RecJ protein predicted from the previously published sequence ) with 64 additional amino acids in the N-terminal domain and 43 aa missing in the C-terminal domain of the protein . The predicted sequence of the RecJ protein in strain ATCC 13939 displays a better alignment with the published protein sequences of the E . coli , Deinococcus geothermalis and Thermus thermophilus RecJ proteins ( Figure S2 ) . The recombinant plasmid was introduced into a recJ+ recipient and the chromosomal copy of recJ in the resulting merodiploid strain was inactivated ( Figure 1 ) . If recJ is an essential gene , the cells will die upon loss of the complementing plasmid at the non permissive temperature . As can be observed in Figure 2 ( lanes 1–3 ) , the ΔrecJ ( prepUTs-recJ+ ) bacteria grew normally at 28° ( the permissive temperature for the plasmid ) but lose viability at 37° ( the non-permissive temperature for the plasmid ) , demonstrating the essentiality of the recJ gene . The ΔrecF , ΔrecO , and ΔrecR mutants , though viable , showed a greatly impaired growth . Indeed , the mutants had a generation time 4-fold longer than the wild-type ( 5 hours for the mutants versus 80 min for the wild-type ) and comparable to that of a ΔrecA mutant . Furthermore , cells devoid of RecF , RecO or RecA proteins had a 10-fold reduced plating efficiency as compared to the wild-type strain and this defect was even more pronounced in the ΔrecR mutant , displaying a 30-fold reduced plating efficiency ( Figure 3 ) . In E . coli , the RecQ helicase initiates DSB repair via the RecFOR pathway by unwinding duplex DNA in the 3′-5′ direction , while the single stranded DNA exonuclease RecJ hydrolyzes the 5′ strand to provide a DNA-substrate for RecA loading onto the 3′ strand [4] , [23] . We found that inactivation of the RecQ helicase in D . radiodurans had no effect on radioresistance , because the knockout mutant was as resistant to γ-irradiation as the wild-type strain ( Figure 4A ) . This result suggests that other helicase ( s ) might be involved in the initiation step of DSB repair in this organism . We tested the RecD and UvrD helicases for putative roles in DSB repair . We found that a ΔrecD deletion mutant was as radioresistant as the wild-type strain , whereas a ΔuvrD mutant showed a reduction in survival that ranged from 5-fold at 11 . 6 kGy to more than 100-fold at 17 . 8 kGy ( Figure 4A ) . However , the mutant still retained significant radioresistance as compared to a repair-deficient ΔrecA strain ( Figure 4B ) , suggesting that other helicase ( s ) may overlap in function with UvrD and thus lessen the effect of a ΔuvrD mutation . To test this hypothesis , we investigated whether the combined absence of UvrD and RecQ or UvrD and RecD proteins results in a more dramatic effect on radio-resistance . As seen in Figure 4A , the ΔuvrD ΔrecQ double mutant bacteria were not more sensitive to γ-rays than a ΔuvrD single mutant . In contrast the ΔuvrD ΔrecD double mutant bacteria were slightly more sensitive to γ-rays than a ΔuvrD single mutant , suggesting that the RecD helicase may have a partial back-up function in the absence of UvrD . To investigate the possible role ( s ) of the UvrD helicase in DSB repair , we examined whether the ΔuvrD mutant was affected in two key steps of the ESDSA pathway: ( i ) the reassembly of broken DNA fragments and ( ii ) the associated massive DNA synthesis . Cells were exposed to 6 . 8 kGy γ-irradiation , a dose that introduces approximately 200 DSB per genome equivalent in a D . radiodurans cell [24] . Recovery from DNA damage was monitored by the appearance of the complete pattern of the 11 resolvable genomic DNA fragments generated by NotI digestion [25] and de novo DNA synthesis was measured by labelling DNA with a 15 min 3H-TdR pulse at different times post irradiation . As seen in Figure 4 , ΔrecQ and ΔrecD cells repaired DSB with the same kinetics as the wild-type strain , reconstituting an intact genome within 3 h post-irradiation ( Figure 5A ) . In contrast , in ΔuvrD bacteria , this process required approximately 8 h ( Figure 5A ) , the kinetics of DSB repair had an increased latent phase ( 240 min in the mutant versus 90 min in the wild-type ) during which DNA degradation took place and a slower rate of fragment assembly . Moreover , resumption of DNA synthesis was delayed in ΔuvrD mutant bacteria and its rate was 2-fold lower than that observed in wild-type bacteria ( Figure 5B ) . These results suggest that UvrD plays a major role in DSB repair through ESDSA . The ΔrecF , ΔrecO and ΔrecR mutants were as radiosensitive as a ΔrecA mutant ( Figure 4B ) . The radiosensitivity of the ΔrecF and ΔrecO mutants was fully complemented by a plasmid expressing RecF or RecO proteins in trans , whereas , in the case of the ΔrecR mutant , bacteria expressing recR+ in trans only recovered 90% of wild-type survival after γ-irradiation ( Figure 4B ) . Because recR belongs to a putative operon , the radiosensitivity of the knock-out mutant may be due in part to a polar effect of our construct on a downstream gene or to a sub- or overoptimal plasmid-based expression of the RecR protein . The kinetics of DNA double-strand-break repair in the three mutants was very similar to that observed in a ΔrecA mutant ( Figure 6A ) . There was a slight and progressive reassembly of the radiation-induced DNA fragments that culminates at 24h post-irradiation incubation in the restitution of a complete pattern of the 11 NotI resolvable fragments ( Figure 6A ) . However , only very faint bands of reconstituted chromosome were observed 24h post-irradiation incubation suggesting that a complete genome was only present in a small subpopulation of the mutant cells . The initial degradation of the damaged DNA that can be seen in the wild-type during the first hour of post-irradiation incubation ( Figure 5A ) was also markedly reduced in the three recFOR mutants ( Figure 6A ) , as was previously observed for a ΔrecA mutant [2]; Figure 6A ) . The reconstitution of the complete genomic NotI pattern in the irradiated recFOR mutants did not result from the multiplication of rare survivors , because there was no observable increase in the number of CFU during 24 hours of incubation of irradiated cells ( data not shown ) . Pulses of 3H-TdR showed that no DNA synthesis was observed during the 6 hours following γ-irradiation ( Figure 6B ) nor during the late fragment assembly in ΔrecF , ΔrecO and ΔrecR bacteria ( data not shown ) , as observed previously in ΔrecA bacteria [2] . Moreover , the late genome reconstitution in these mutants is not sufficient to ensure cell survival . In conclusion , our results suggest that RecF , RecO and RecR proteins , like RecA protein , play a central role in Deinococcal radioresistance , probably because they are absolutely required for loading RecA onto its DNA substrate to perform efficient double-strand-break repair via ESDSA and recombinational repair pathways .
D . radiodurans is naturally devoid of RecB and RecC proteins but contains homologs of key proteins of the E . coli RecF pathway: RecJ , RecQ , RecF , RecO and RecR . We found that cells devoid of RecF , RecO or RecR proteins were as radiosensitive as cells devoid of RecA . The ΔrecF , ΔrecO and ΔrecR mutants , as previously shown for a ΔrecA mutant [2] , supported a slow and progressive reassembly of the radiation-induced DNA fragments . As in ΔrecA cells , genome reassembly was not accompanied by significant DNA synthesis , suggesting that cells devoid of RecF , RecO or RecR proteins are deficient for ESDSA , with repair of DSB probably mediated by RecA-independent pathways , such as single-strand annealing ( SSA ) or non-homologous end joining ( NHEJ ) . The mutants also showed an important lethal sectoring during normal growth , similar to that observed in a ΔrecA mutant , in which about 90% of the visible cells failed to give rise to colonies [26] . The similarity of the ΔrecFOR and ΔrecA phenotypes supports the hypothesis that RecA activity in D . radiodurans is totally dependent on a functional RecF pathway . Following exposure of D . radiodurans to ionizing radiation , there is a rapid and extensive degradation of chromosomal DNA that plays an important role in the repair process in this species ( reviewed by [24] ) . The initial degradation of damaged DNA can be observed using pulsed-field electrophoresis as a reduction of the amount of the double stranded DNA fragments during the first 90 min of post-irradiation incubation in the wild-type cells , prior to the onset of fragment assembly . Slade et al observed that DNA degradation is markedly reduced in a ΔrecA mutant , leading the authors to propose that RecA itself regulates maturation of double-strand ends by controlling both DNA degradation and DNA synthesis [2] . We found that DNA degradation was also reduced in ΔrecF , ΔrecO or ΔrecR mutants , as well as in a ΔrecA mutant . RecA may play a regulatory role in the control of expression of nuclease-like activities in response to DNA damage , while RecFOR proteins may be indirectly involved in DNA degradation by facilitating the formation of the RecA filament on single stranded DNA . It would be interesting to analyse DNA degradation in the Deinococcal recA424 mutant , which retains the RecA coprotease activity while remaining deficient in recombination activity [27] . Biochemical studies using RecFOR proteins from E . coli indicate that these proteins act together as mediators of the formation of the pre-synaptic RecA filament onto single stranded DNA . Current models agree on the formation of two complexes , RecFR and RecOR . RecOR is generally thought to be responsible for rendering SSB-coated ssDNA accessible to RecA . RecFR targets dsDNA or dsDNA-ssDNA junctions and is responsible for the targeting of RecA to the ssDNA region of gaps [28]–[31] . More recently , it was proposed that RecR is the key component with which RecA interacts , whereas the RecO protein can displace SSB and bind to single stranded DNA independently of RecR , yet does not load RecA until RecR is added [32] , [33] . When RecF is present , a RecFOR loading pathway , independent of RecO-SSB interactions , is preferred [33] . Recent X-ray structural analysis of RecO and RecR proteins from D . radiodurans confirms the existence of a RecOR complex in this organism . RecR molecules form a ring structure that can encircle both dsDNA and ssDNA [34] , [35] . The structure of the RecF protein from D . radiodurans has also recently been elucidated , showing that the RecF protein exhibits extensive structural similarity with the head domain of the eukaryotic Rad50 protein [36] . More recently , a model of recognition of the ds-DNA ss-DNA junction in D . radiodurans through a DNA-protein and protein-protein interaction was proposed: RecR interacts with ssDNA coated by RecO-SSB , which leads to the elevation of the local concentration of RecR and stimulates RecF binding in the adjacent ds-DNA [37] . While inactivation of RecA or RecFOR proteins in D . radiodurans reduced cell viability , inactivation of RecJ resulted in a fully lethal phenotype . In other bacterial species , mutations in recJ are highly synergistic with those in recBCD . In E . coli , recBC recJ mutants are recombination deficient , extremely UV-sensitive and highly growth disrupted [11] , [38] . In Salmonella typhimurium , recB recJ mutants also display a similar phenotype [39] . More recently , it was shown that a recJ knock-out is colethal with recBCD or recD deletions in Acinetobacter baylyi [40] . The strongly reduced viability ( or lethality ) of recBC recJ bacteria was attributed to severe deficiencies in repair of spontaneous DNA damage and inactivated replication forks [39] , [40] . It should be noted that , whereas E . coli and S . typhimurium contain at least three 5′-3′ exonucleases [RecJ , Exo V ( RecBCD ) , Exo VII ( XseAB ) ] , the genome of A . baylyi encodes only Exo V and RecJ , and that of D . radiodurans encodes only RecJ and one of the two subunits of Exo VII . We propose that Exo VII has some back-up activity in E . coli or S . typhimurium when RecJ and ExoV are inactivated , an activity that is missing in A . baylyi and D . radiodurans . In E . coli , RecJ and RecFOR were proposed to be required to restore DNA synthesis after UV-induced damage [41] , [42] . The mechanism by which lesion-blocked replication forks recover in E . coli is thought to involve the formation of reverse replication fork intermediate stabilized by RecA and RecF and degraded by the RecQ-RecJ helicase-nuclease when RecA or RecF are absent [41] . The fork regression allows DNA repair enzymes to remove the blocking lesion , thus restoring processive replication . In the absence of RecJ , the recovery of replication is significantly delayed and both replication recovery and cell survival become dependent on translesion synthesis by DNA polymerase V [42] . D . radiodurans does not encode a bypass DNA polymerase belonging to the Y family , and under these conditions RecJ may be essential for restoration of replication forks after arrest , even in cells not treated by DNA damaging agents . Frequent DNA double-strand-breaks were thought to arise spontaneously ranging from 0 . 2–1 per genome replication in E . coli [4] , [43] . However , a more direct quantification of DNA double-strand-breaks indicated that the rate of spontaneous breakage is 20 to 100-fold lower than predicted , only one percent of the cells having one or more DNA double-strand-breaks per genome replication [44] . Because cells devoid of RecA or RecFOR are viable , the ΔrecJ lethal phenotype cannot be only due to a possible deficiency in DSB repair , leading us to postulate that RecJ is required in D . radiodurans for more than one cellular process and that inactivation of all of these processes ( DSB repair , fork reversion , restoration of a fork structure after regression … ) may be lethal for the cell . In E . coli , the RecJ exonuclease has been mainly associated with the RecQ helicase in recombination and repair ( see , for review , [4] ) . The RecQ protein from D . radiodurans shows unusual domain architecture with three tandem HRDC ( Helicase RNase D C-terminal ) domains in addition to the conserved helicase and RQC ( RecQ C-terminal ) domains . The tandem arrangement of HRDC domains regulates the specificity of the binding of RecQ to DNA substrates [45] , [46] . Here , we found that ΔrecQ mutants displayed a wild-type level of resistance to γ-irradiation , exhibiting the same kinetics as the wild-type strain for fragment reassembly and DNA synthesis after irradiation . In another report , a recQ knock-out mutant was shown to be highly sensitive to H2O2 and slightly more sensitive than the wild-type strain to elevated γ-irradiation doses [46] . It was recently proposed that recQ deletion , by causing transcriptome alteration , would generate ROS accumulation and Fe and Mn alterations [47] . Our findings suggest that the RecQ helicase in D . radiodurans plays only a minor role in DSB repair , probably as consequence of redundant functions provided by other helicase ( s ) . Mutants devoid of RecD behave like ΔrecQ mutants in that they show wild-type radioresistance and repair capacity . The Deinococcal RecD protein has been characterized in vitro as a helicase with 5′-3′ polarity ( opposite to that of RecQ ) and low processivity [14] . In contrast , we found that inactivation of UvrD ( helicase II ) markedly reduced Deinococcal radioresistance and severely delayed the kinetics of DSB repair . UvrD has been largely characterized for its role in nucleotide excision repair ( NER ) and mismatch repair ( MMR ) in E . coli ( reviewed by [48] ) . However , the altered kinetics of repair and the radiosensitivity of ΔuvrD bacteria are unlikely to result from a deficiency in the NER pathway because uvrA deficient mutant bacteria display a wild-type survival pattern following exposure to ionizing-radiation ( Figure S3 ) . The ΔmutS bacteria deficient for MMR were also shown to be as radioresistant as wild-type bacteria [18] . Interestingly , the delayed kinetics of DSB repair in cells devoid of UvrD coincided with DNA synthesis ( albeit significantly less extensive than that observed in the wild-type cells ) suggesting that ESDSA repair could take place but only inefficiently in this mutant . We propose that UvrD is involved in ESDSA and that the redundant activity of other helicase ( s ) is responsible for the residual DNA repair capacity observed in the ΔuvrD mutant . The fact that the ΔrecQΔuvrD and the ΔrecDΔuvrD double mutant bacteria were not as radiosensitive as ΔrecA bacteria suggests that neither RecQ nor RecD can solely fulfil this role , and that other helicase ( s ) may be involved . Helicase IV ( HelD ) has been implicated as partner of the RecJ exonuclease in the RecF pathway in E . coli , together with Helicase II and RecQ [49] . Mutational inactivation of Helicase IV has no effect on the radioresistance of D . radiodurans [50] . Alternatively , RecA itself , by binding to double stranded DNA ends , could unwind DNA and provide a DNA substrate for RecJ or another 5′-3′ exonuclease . Indeed , in vitro , the D . radiodurans RecA protein binds preferentially to double stranded DNA [51] . In E . coli , the UvrD protein was not shown to be required for DNA double-strand-break repair . In contrast , it was shown to possess an anti-recombination activity , which has been related to its capacity to disrupt the RecA nucleoprotein filament [52] , [53] . This activity is conserved among many species [54] . Thus , as in other species , the D . radiodurans UvrD protein might not be involved directly in the maturation of DNA double-strand ends . Several observations suggest that E . coli UvrD may be involved in DNA replication [55]–[58] and it was shown to be required for DNA replication of several different rolling-circle plasmids in E . coli [59] . Thus , the D . radiodurans UvrD protein might also act in the DNA synthesis step of ESDSA . Taking into account our results and those of others [2] , [60] , [61] , we propose a model for the role of the proteins of the RecF pathway in ESDSA ( Figure 7 ) . In this model , RecJ or an as-yet unidentified exonuclease associated with the UvrD helicase , could generate 3′ single stranded DNA ends required for priming of massive DNA synthesis . Alternatively , RecA itself , by binding to double stranded DNA ends , could unwind DNA and provide a DNA substrate for RecJ or another exonuclease . Analysis of the transcriptome of D . radiodurans revealed a large group of genes that are up-regulated in response to either desiccation or ionizing radiation [62] . The deinococcal specific ddrA ( DR0423 ) and ddrB ( DR0070 ) genes were found among the most highly induced in response to each stress and their inactivation promotes sensitization of the mutant cells to ionizing radiation [62] . The DdrA protein is involved in protection of 3′ DNA single stranded ends [60] and presumably ensures long-lived recombinational substrates [63] . The DdrB protein binds single stranded DNA but not duplex DNA and is the prototype of a new bacterial SSB family [61] . The induction of an alternative SSB following irradiation has potentially broad significance for efficient genome reconstitution . We propose that during initiation of ESDSA , DdrA protects the 3′ DNA ends whereas SSB or the SSB-like DdrB binds to single stranded DNA . Our results supporting the idea that RecA activity in D . radiodurans is totally dependant on a functional RecF pathway , lead us to propose that RecFOR renders SSB or DdrB- coated single stranded DNA accessible to RecA and favors formation of a RecA nucleoprotein filament required for invasion of a double stranded homologous DNA . Finally , as described previously [2] , Pol III and Pol I can promote DNA synthesis , eventually with the help of the UvrD helicase . Moreover , the compact D . radiodurans nucleoid structure that remains unaltered after high-dose γ-irradiation may passively contribute to radioresistance by preventing the dispersion of free DNA ends [64] , [65] . Such a condensed genome may provide suitable scaffolds for DNA repair through ESDSA , recombinational and/or DNA end joining processes . In conclusion , we demonstrate the essential role of key components of the D . radiodurans RecF pathway in ESDSA . We show for the first time that ( i ) inactivation of only one exonuclease , RecJ , results in cell lethality ( ii ) cells devoid of RecF , RecO or RecR display greatly impaired growth ( iii ) RecF , RecO or RecR proteins are essential for radioresistance through ESDSA ( iv ) UvrD helicase has an unexpected crucial function in DNA double-strand-break repair through ESDSA .
Bacterial strains and plasmids are listed in Table 1 and Table 2 , respectively . The Escherichia coli strain DH5α was used as the general cloning host and strain SCS110 was used to propagate plasmids prior to introduction into D . radiodurans via transformation [66] . All D . radiodurans strains were derivatives of strain R1 ( ATCC 13939 ) . D . radiodurans was grown in TGY2X ( 1% tryptone , 0 . 2% dextrose , 0 . 6% yeast extract ) or in TGYA ( 0 . 5% tryptone , 0 . 2% dextrose , 0 . 15% yeast extract ) at 30°C with aeration or on TGY1X plates solidified with 1 . 5% agar . E . coli strains were grown in Luria-Bertani ( LB ) broth ( Gibco Laboratories ) . When necessary , media were supplemented with the appropriate antibiotics used at the following final concentrations: chloramphenicol 3 µg/mL for D . radiodurans; kanamycin 6 µg/mL for D . radiodurans; tetracycline 2 . 5 µg/mL for D . radiodurans; hygromycin 50 µg/mL; spectinomycin 40 µg/mL for E . coli and 75 µg/mL for D . radiodurans . Transformation of D . radiodurans with PCR products , genomic DNA , or plasmids was performed as previously described [26] . Plasmid DNA was extracted from E . coli using the QIAprep spin miniprep kit ( Qiagen ) . Chromosomal DNA of D . radiodurans was isolated as previously described [67] . Amplification of plasmid or genomic DNA by PCR was performed with DyNAzyme EXT DNA polymerase ( Finnzyme ) or Extensor Hi-Fidelity PCR enzyme Mix ( ABgene ) . Oligonucleotides used are listed in Table S1 . The recF , recO , recR , recA , uvrD , recD , recQ , recJ disruption mutants were constructed by the tripartite ligation method [18] . The mutated alleles constructed in vitro were then used to transform D . radiodurans to replace their wild-type counterpart by homologous recombination . The genetic structure and the purity of the mutants were checked by PCR using primers described in Table S1 . Plasmid p11869 is a derivative of the thermosensitive plasmid p13840 [21] . To construct p11869 , the recJ gene was amplified by PCR using the primer pair ( PS441/PS442 ) and the product was cloned into plasmid p13840 between the NdeI/XhoI sites . Plasmids p11862 , p11860 and p11870 carrying the recF , recO , recR genes , respectively , under the control of their natural promoter were used to express the recF , recO , recR genes in a ΔrecF , ΔrecO , ΔrecR background . To construct plasmid p11860 , the recO gene was amplified by PCR using the primer pair ( PS402/PS403 ) and the resultant product was cloned into plasmid p11520 [68] between the SacI/BamHI sites . Plasmid p11870 , containing the recR gene , was constructed in a similar manner using the primer pairs PS414/PS415 . The recF gene was cloned into plasmid p11520 between the SpeI/BglII sites in a similar manner using the primers PS410/PS411 to obtain p11862 . All constructions were verified by DNA sequencing . Plasmid p11562 [63] , expressing recA from a PSpac promoter , was used to transform GY12968: ΔrecAΩkan giving rise to strain GY14111 . The expression of recA was induced by adding 10 mM IPTG to the media . Exponential cultures , grown in TGY2X ( supplemented with spectinomycin when necessary ) , were concentrated to an A650 = 10 in TGY2X and irradiated on ice with a 137Cs irradiation system ( Institut Curie , Orsay , France ) at a dose rate of 44 . 7 Gy/min . Following irradiation , diluted samples were plated on TGY plates . Colonies were counted after 3–4 days incubation at 30°C . The essentiality of genes was evaluated in a growth experiment in which the strains grown at 28°C in liquid medium with spectinomycin , were serially diluted , plated on TGY agar and incubated at 28°C or 37°C in the presence or absence of spectinomycin [21] . Non-irradiated or irradiated ( 6 . 8 kGy ) cultures were diluted in TGY2X to an A650 = 0 . 2 and incubated at 30°C . At different post-irradiation recovery times , culture aliquots ( 5mL ) were removed to prepare DNA plugs as described previously [60] . The agarose embedded DNA plugs were digested for 16 h at 37°C with 10 units of NotI restriction enzyme . After digestion , the plugs were subjected to pulsed field gel electrophoresis as described previously [69] . The rate of DNA synthesis was measured according to a modified protocol from Zahradka et al [1] . Exponential cultures , grown in TGYA , were concentrated to an A650 = 20 in TGYA and irradiated as described previously . Non-irradiated or irradiated cultures ( 6 . 8 kGy ) were diluted in TGYA to an A650 = 0 . 2 and incubated at 30°C . At different time 0 . 5mL samples were taken and mixed with 0 . 1mL pre-warmed TGYA containing 4 . 8 µCi [methyl-3H]thymidine ( PerkinElmer , specific activity 70–90 Ci/mmol ) . Radioactive pulses of 15 min were terminated by addition of 2 mL ice-cold 10% TCA . Samples were kept on ice for at least 1 h , and then collected by vacuum filtration onto Whatman GF/C filters followed by washing twice with 5mL 5% TCA and twice with 5mL 96% ethanol . Filters were dried for 10 min under a heat source and placed in 4 mL scintillation liquid . The precipitated counts were measured in a liquid scintillation counter ( Packard , TRI- carb 1600 TR ) .
|
Deinococcus radiodurans bacterium is among the best-known organisms found to resist extremely high exposures to desiccation and ionizing radiation , both causing extensive DNA double-strand breaks . Because a single unrepaired DNA double-strand break is usually lethal , DNA double-strand breaks are considered as the most severe form of genomic damage . The extreme radioresistance of D . radiodurans is linked to its ability to reconstruct a functional genome from hundreds of chromosomal fragments . Genome reconstitution occurs through a two step process: ( i ) an extended synthesis-dependent strand-annealing process ( ESDSA ) that assembles genomic fragments in long linear intermediates that are then ( ii ) processed through recombination to generate circular chromosomes . Here , we demonstrate the essential role of key components of the D . radiodurans RecF pathway in ESDSA . We show that ( i ) inactivation of only one exonuclease ( RecJ ) results in cell lethality; ( ii ) cells devoid of RecF , RecO , or RecR display greatly impaired growth; ( iii ) RecF , RecO , or RecR proteins are essential for radioresistance through ESDSA; and ( iv ) UvrD helicase has an unexpected crucial function in DNA double-strand-break repair through ESDSA .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"molecular",
"biology/dna",
"replication",
"cell",
"biology/cellular",
"death",
"and",
"stress",
"responses",
"molecular",
"biology/recombination",
"genetics",
"and",
"genomics/gene",
"function",
"microbiology/microbial",
"growth",
"and",
"development",
"molecular",
"biology/dna",
"repair"
] |
2010
|
A Major Role of the RecFOR Pathway in DNA Double-Strand-Break Repair through ESDSA in Deinococcus radiodurans
|
Protein chaperones are essential in all domains of life to prevent and resolve protein misfolding during translation and proteotoxic stress . HSP70 family chaperones , including E . coli DnaK , function in stress induced protein refolding and degradation , but are dispensable for cellular viability due to redundant chaperone systems that prevent global nascent peptide insolubility . However , the function of HSP70 chaperones in mycobacteria , a genus that includes multiple human pathogens , has not been examined . We find that mycobacterial DnaK is essential for cell growth and required for native protein folding in Mycobacterium smegmatis . Loss of DnaK is accompanied by proteotoxic collapse characterized by the accumulation of insoluble newly synthesized proteins . DnaK is required for solubility of large multimodular lipid synthases , including the essential lipid synthase FASI , and DnaK loss is accompanied by disruption of membrane structure and increased cell permeability . Trigger Factor is nonessential and has a minor role in native protein folding that is only evident in the absence of DnaK . In unstressed cells , DnaK localizes to multiple , dynamic foci , but relocalizes to focal protein aggregates during stationary phase or upon expression of aggregating peptides . Mycobacterial cells restart cell growth after proteotoxic stress by isolating persistent DnaK containing protein aggregates away from daughter cells . These results reveal unanticipated essential nonredunant roles for mycobacterial DnaK in mycobacteria and indicate that DnaK defines a unique susceptibility point in the mycobacterial proteostasis network .
Proper protein folding is essential for all organisms and assures that the primary sequence of the polypeptide forms its functional tertiary and quaternary structures . Protein chaperones are present in all domains of life and serve multiple functions in protein homeostasis . During translation , chaperones are required to assure proper protein folding and prevent protein aggregation , which can occur as hydrophobic segments of the protein emerge from the ribosome . After synthesis , protein denaturation is a common event due to exogenous proteotoxic stresses such as heat and oxidation , correction of which requires chaperone systems to refold denatured proteins when possible , and facilitate disaggregation and degradation when refolding is not possible . The importance of chaperone function for cellular viability is reflected in the frequent redundancy between chaperones for protein folding and aggregate processing [1]–[4] . The Hsp70 family of chaperones is widely distributed in both prokaryotic and eukaryotic cells [5] . The best studied Hsp70 chaperone of bacteria is E . coli DnaK . DnaK is a central hub for protein folding , shuttling misfolded peptides to other chaperones and proteases for resolution , a function that is essential during the protein denaturation that occurs during heat shock [6]–[9] . In addition to its effector function in the heat shock response , DnaK also regulates this response by destabilizing the alternative sigma factor , σ32 , preventing aberrant induction of the heat shock response during non-stress conditions and turning off the response after heat shock [10] . However , In E . coli , DnaK is nonessential for native protein folding because of redundancy with Trigger Factor , which associates with proteins soon after emergence from the ribosome [11] . Although dnaK/tf double mutants are nonviable , overexpression of GroEL , SecB , or Hsp33 can suppress the synthetic lethality of dnaK/tig double mutants [10] , [12]–[16] . Furthermore , examination of proteins that interact with DnaK indicates that most client proteins that require DnaK for proper folding and/or stability are largely non-essential , suggesting that loss of function of these proteins in the absence of DnaK does not impact viability [7] . However , in the absence of both DnaK and TF , the E . coli cell suffers proteostasis collapse characterized by global insolubility of nascent proteins [7] . In bacteria other than E . coli , the function of DnaK has not been studied extensively . In mycobacteria , DnaK regulates the heat shock response through its interaction with the HspR C terminal tail , which becomes insoluble upon heat shock , thereby relieving repression of chaperone genes [17] . The mycobacterial heat shock response is negatively regulated by the repressors HspR and HscA; and positively through σH [18] , [19] . Deletion of hspR , which derepresses several chaperones involved in the heat shock response , including ClpB , alpha-crystalin , and DnaK-GrpE-DnaJ1 , led to decreased persistence after the initial phase of infection in a mouse model suggesting that dnaK and other HspR regulated genes must be controlled during infection for optimal growth and persistence [19] . Host inflicted proteotoxic stress is likely a significant in vivo stress for M . tuberculosis during infection , yet the function of the mycobacterial chaperone network in native and stress induced proteostasis is incompletely understood . Additionally , Mtb DnaK is found in culture filtrates [20] , [21] and on the bacterial surface [22] , and has a role in pathogenesis by modulating host immune responses [22]–[24] . Despite substantial progress in targeting chaperone function in malignant human cells [25] , [26] , inhibition of chaperone function as an antimicrobial strategy is relatively unexplored , in part because of the redundancy of the chaperone network . Using a model mycobacterial species , M . smegmatis , we characterized the function of DnaK in mycobacteria and find an unanticipated lack of redundancy that places mycobacterial DnaK as a central chaperone in both native and stress induced protein folding .
To study the function of mycobacterial DnaK , we attempted to delete dnaK from the M . smegmatis chromosome . Initial attempts yielded no allelic replacements , suggesting that DnaK may be essential for viability . Provision of a second copy of dnaK at the attB phage integration site allowed replacement of the chromosomal dnaK with an unmarked ΔdnaK allele lacking the first 1765 bp of the 1869 bp dnaK ORF ( Figure S1A ) . We then attempted to remove the second copy of dnaK from attB by marker exchange with either a vector , pMV306kan , or a plasmid encoding DnaK and conferring kanamycin resistance , pAJF447 . Only transformation with pAJF447 yielded transformants that were kanamycin resistant at 30°C or 37°C ( Figure S2A ) . We observed small numbers of kanamycin resistant transformants in vector transformed cells , but these cells continued to express DnaK , indicating that the second copy of dnaK was not lost in these transformants ( Figure S2A ) . This failure to remove the copy of dnaK from attB in our ΔdnaK strain suggested that dnaK was required for growth at 30°C ( low ) and 37°C ( high ) temperatures . We observed a similar essentiality for the DnaK cofactor GrpE . After constructing a strain with a grpE deletion and a second copy of grpE at attB ( MGM6023; Figure S1B ) we were unable to remove this second copy of grpE from attB by marker exchange at either 30°C and 37°C ( Figure S2B ) , demonstrating that both DnaK and GrpE are required for growth of M . smegmatis . Similar results were obtained when we attempted to delete the entire dnaK operon suggesting that all three components of the operon are essential ( data not shown ) . To further study the function of DnaK , we generated a depletion strain , MGM6005 , which encodes an anhydrotetracycline ( ATc ) inducible allele of DnaK with a C terminal StrepTagII ( STII ) . By 9 hours after withdrawal of ATc , DnaK-STII was undetectable by immunoblot with anti-STII antibodies ( Figure 1A ) . Cells lacking DnaK continued to grow for an additional 12 to 15 hours without detectable DnaK , at which point optical density stabilized , in contrast to the continued replication of DnaK replete cells ( Figure 1B ) . The viability of growth arrested cells lacking DnaK was determined by culturing dilutions onto selective media with ATc for both depletion and control cultures . The number of viable cells remained constant in the growth arrested , DnaK depleted population , indicating that loss of DnaK is bacteriostatic rather than bactericidal over the time course of the experiment ( Figure 1B ) . The DnaK chaperone system has been previously implicated in protection and recovery of bacterial cells from heat shock [10] , [27]–[29] . In M . tuberculosis , loss of σH has been shown to result in decreased survival at 53°C , a phenotype that was attributed to attenuation of the mycobacterial heat shock response , including the induction of DnaK [18] . To directly test the contribution of DnaK to heat shock response , we depleted DnaK for 12 hours prior to measuring cell viability at 53°C . Cells lacking DnaK were 100 fold more sensitive to killing by heat compared to DnaK replete cells ( Figure S3 ) . To examine the morphologic correlates of the growth arrest that accompany DnaK depletion , we examined DnaK depleted and control cells by microscopy . Nucleoid morphology ( Figure 1C , Hoechst panel ) and cellular morphology ( Figure 1C , DIC panel ) of DnaK depleted cells were indistinguishable from that observed in wildtype cells , suggesting that DnaK depletion did not alter cell gross morphology . However , membrane staining by the lipophilic dye FM 4-64 was altered in cells lacking DnaK . In contrast to the homogenous distribution of the FM 4-64 membrane staining pattern seen in wild type cells , FM 4-64 in DnaK depleted cells was no longer evenly distributed along the entire periphery of the cell , but rather accumulated in a patchy pattern at midcell ( Figure 1C ) . To examine whether this change in membrane staining pattern was accompanied by changes in membrane protein localization , we fused E . coli MalF transmembrane domains 1 and 2 to three copies of mCerulean . This fusion was evenly distributed in the membrane of cells expressing DnaK ( Figure 1D , bottom ) . In contrast , upon DnaK depletion , MalF-mCerulean accumulated at midcell ( Figure 1D , top ) in a pattern that colocalized with FM 4-64 in the absence of DnaK ( Figure S4 ) . Taken together , these results indicate that loss of DnaK affects membrane structure and/or membrane protein localization with relative preservation of overall cell dimensions and nucleoid morphology . To assess whether the membrane alterations seen with loss of DnaK affected cell permeability , we utilized a previously described assay for measuring ethidium bromide ( EtBr ) permeability [30]–[32] . DnaK depleted cells showed increased accumulation of EtBr as compared to replete cells ( Figure S5 ) . The addition of carbonyl cyanide 3-chlorophenylhydrazone ( CCCP ) , an efflux inhibitor shown to increase EtBr accumulation [31] , led to an increase of EtBr accumulation in both DnaK depleted and replete cells , indicating that DnaK depleted cells were still capable of EtBr efflux . Cells lacking DnaK still had increased accumulation relative to replete cells in the presence of CCCP ( Figure S5 ) indicating that the increase in EtBr in DnaK depleted cells was due to an increase in cell permeability . To examine whether mycolic acid synthesis is altered with DnaK depletion , we analyzed total mycolic acid methyl esters and fatty acid methyl esters made in DnaK replete and depleted cultures . We observed no difference in the amount of total mycolic acid methyl esters or fatty acid methyl esters synthesized within 1 hour in between DnaK replete or depleted cells ( Figure S6 ) . Taken together with the FM 4-64 staining and MalF-mCerulean localization , these results indicate that loss of DnaK affects membrane structure and permeability . The requirement for DnaK to sustain mycobacterial growth and membrane integrity suggested that DnaK may have a critical role in the absence of exogenous proteotoxic stress . To test the function of DnaK in native protein folding , we expressed firefly luciferase , which has been used as a model protein to study the activity of DnaK in E . coli cells after heat shock [33] , [34] , in our M . smegmatis dnaK depletion strain . In cells depleted of DnaK , we observed a rapid loss of luciferase activity ( Figure 2A ) , which occurred prior to growth arrest but coincident with loss of DnaK-STII protein ( Figure 2A and B ) . 14 hours after withdrawal of ATc , >80% of luciferase activity was lost . Although luciferase activity dropped in DnaK depleted cultures , luciferase protein levels remained stable ( Figure 2B , Top panel ) , indicating that the loss of activity was not due to a change in steady state protein levels . By utilizing centrifugation to separate insoluble aggregate proteins , as previously described for E . coli [35] , we observed that upon DnaK depletion , luciferase was depleted from the soluble fraction and accumulated in the pellet fraction ( Figure 2C ) . Taken together , these results indicate that DnaK is required for folding of luciferase in the absence of heat shock , suggesting a nonredunant role in native protein folding . To visualize the kinetics and localization of protein aggregate formation without DnaK , we generated a strain carrying a Luciferase-mCitrine fusion protein , which allowed us to simultaneously track luciferase activity and localization during DnaK depletion . Upon depletion of DnaK , the kinetics of loss of luciferase activity from the Luciferase-mCitrine fusion protein were similar to that observed with luciferase ( Figure 2A ) . In contrast , mCitrine fluorescence was maintained in DnaK depleted cells , suggesting that whereas luciferase requires DnaK for folding , mCitrine does not [36]–[38] . Live cell time-lapse imaging in DnaK replete and depleted cells showed two patterns of localization . In the presence of DnaK , Luciferase-mCitrine was cytoplasmic and evenly distributed throughout the cell ( Figure 2D ) . However , upon depletion of DnaK , Luciferase-mCitrine formed polar fluorescent foci which eventually coalesced into large cytoplasmic aggregates ( Figure 2E ) . Immunoblots detecting both Luciferase and Luciferase-mCitrine during DnaK depletion did not detect any proteolytic cleavage of the protein ( Figure S7 ) , even after hours of DnaK depletion . In E . Coli , DnaK and Trigger Factor cooperate to support the folding of nascent peptides . Their activities are largely redundant such that , at permissive temperatures , loss of either chaperone is tolerated [13] , [14] . To examine potential redundancy between mycobacterial TF and DnaK , we deleted MSMEG_4674 , the gene encoding TF ( Figure S1C ) . In contrast to DnaK , mycobacterial TF was nonessential . Furthermore , overexpression of Trigger Factor did not rescue the loss of luciferase activity observed in DnaK depleted cultures ( MGM6073 ) or allow for the loss of the dnaK gene ( MGM6072 ) ( data not shown ) . Loss of TF did not affect luciferase activity in a wild type background ( Figure S8A ) or bacterial growth ( Figure S8B ) . DnaK depletion in the absence of Tigger Factor led to a modest acceleration in the kinetics of luciferase activity loss ( Figure S8A ) , and in the time to growth arrest ( Figure S8B ) . These data indicate that Trigger Factor makes a minor contribution to nascent luciferase folding and stability in M . smegmatis that is only evident when DnaK is absent , and demonstrate that DnaK is the dominant chaperone for native folding . The insolubility of luciferase in DnaK depleted cells suggests a generalized role for DnaK in maintaining native protein folding and solubility . To test this idea , we examined the solubility of endogenous M . smegmatis proteins in the absence of DnaK . We depleted DnaK for 16 hours , a time point at which cells are still replicating and maintain full viability ( Figure 1A ) , yet have no detectable DnaK ( Figure 1B ) . We fractionated lysates from DnaK depleted and replete cells and compared fractions by SDS-PAGE . The total , soluble , and membrane ( 1% Triton X-100 soluble ) fractions were similar from depleted and replete cells ( Figure 3A ) . In contrast , the protein content of the pelleted ( Triton X-100 insoluble ) fraction was substantially increased in lysates from DnaK depleted cells ( Figure 3A ) . To determine whether protein insolubility was the result of nascent protein misfolding versus aggregation of existing protein pools , we performed short term labeling of newly synthesized proteins with 35S-Methionine and analyzed the relative incorporation of the label into soluble and insoluble fractions with and without DnaK . In the absence of DnaK , nascent peptides accumulated in the insoluble fraction , accounting for approximately a 45% ( ±4 . 8% ) increase in insoluble proteins , indicating that DnaK is required for nascent peptide folding ( Figure 3B ) . To exclude an effect of DnaK depletion on translation rates that might account for these findings , we quantitated the rate of nascent chain synthesis using puromycin labeling and detection with an anti-Puromycin antibody [14] , [39] . Both DnaK replete and depleted cells produced puromycilated chains at equal rates , as determined by immunoblotting ( Figure S9 ) , indicating that DnaK loss does not affect rate of translation . Inspection of the proteins that become insoluble without DnaK revealed several high molecular weight proteins in the insoluble fraction ( Figure 3C ) . Mass spectroscopic identification of tryptic peptides derived from these two high molecular weight proteins identified two polyketide synthetases , MSMEG_0408 ( type 1 modular polyketide synthase ) and MSMEG_0400 ( MtbH , peptide synthase ) ( Figure 3C ) . Mycobacteria are unusual in that they encode many multimodular polyketide sythases for lipid synthesis , which are very large proteins greater than 300 kDa . The M . smegmatis chromosome encodes 8 proteins larger than 2000 amino acids , whereas the E . coli K12 chromosome encodes only 1 . To extend the finding that DnaK is required for solubility of large multimodular enzymes in mycobacteria , we examined one additional essential large multimodular protein , fatty acid synthase I ( FASI , 3089AA ) . We generated a strain expressing a single full-length copy of FASI fused to mCitrine the C terminus , expressed from its endogenous locus in our DnaK depletion strain background ( strain MGM6014 ) . Cell fractionation in DnaK replete or depleted cells revealed that , although some FASI is present in the insoluble fraction with DnaK , FASI accumulated in the insoluble fraction after DnaK depletion ( Figure 3D ) . This demonstrates that FASI , a large , multimodular , essential protein in mycobacteria , requires DnaK for optimal folding and solubility in the absence of proteotoxic stress , suggesting that FASI may be a direct client of DnaK . Taken together our data indicates that DnaK is required for the solubility of at least 3 of the 8 large multimodular proteins in the mycobacterial proteome , all of which are lipid synthases , potentially explaining the disruption of membrane integrity observed in DnaK depleted cells . In addition to the large polyketide synthases that become insoluble in DnaK depleted cells , examination of soluble proteins in SDS-PAGE fractionated lysates from DnaK depleted cells revealed a protein species of approximately 90 kDa that was overrespresented in DnaK depleted cells ( Figure 4A , black arrow ) . Mass spectrometry identified this protein as ClpB . To track both ClpB levels and localization during DnaK depletion , we fused mCitrine to the 3′ end of clpB in the DnaK depletion strain ( strain MGM6008 ) . ClpB levels were stable in DnaK replete cells ( Figure 4B ) , and during the first 12 hours of DnaK depletion . Beginning at 21 hours after DnaK depletion , ClpB accumulated ( Figure 4B ) . By microscopy , ClpB-mCitrine was expressed at low levels and was near the limit of detection ( Figure 4C ) . At early time points of DnaK depletion , when native folding is lost , but before ClpB protein accumulation by immunoblot ( 9 to 12 hours ) , ClpB-mCitrine re-localized to form cytoplasmic foci , suggesting that ClpB relocalizes to protein aggregates that accumulate after loss of DnaK ( Figure 4C ) . In several bacterial species other chaperones have been shown to be upregulated after loss of DnaK to compensate for the defect in chaperone function [10] , [40]–[42] . To assess the potential compensatory upregulation of the chaperone network that may respond to DnaK loss in mycobacterial cells , we performed RT-qPCR on RNA collected from control and DnaK depleted cells . We detected upregulation of the mRNAs encoding ClpB , HspR , and Hsp20 ( Figure 4D ) , all most likely the result of destabilization of HspR in the absence of DnaK [17] . No other chaperones or proteases tested were upregulated in the absence of DnaK , indicating that there is a lack of broad compensatory upregulation of alternative chaperone systems to handle the insoluble proteins that accumulate without DnaK function . Based on our findings that mycobacterial DnaK plays a crucial role in native protein folding and in maintaining membrane protein and/or lipid composition , we hypothesized that it might localize in a peri-membrane pattern . In E . coli , DnaK localizes in a diffuse cytoplasmic pattern at 37°C and relocalizes to foci at 42°C and above [43] , [44] . To assess DnaK localization in M . smegmatis , we generated a fully functional DnaK-mCitrine fusion at the native chromosomal locus such that DnaK-mCitrine was expressed as a stable fusion at the estimated full-length size ( Figure S10 ) . DnaK-mCitrine appeared to be functional for essential DnaK functions as well as for heat resistance at 53°C ( Figure S11 ) as the DnaK-mCitrine fusion is the only copy of DnaK in the cell . DnaK-mCitrine was visible in multiple membrane peripheral foci distributed along the entire length of the cell during logarithmic growth at 30°C and 37°C ( Figure 5A ) . DnaK foci were dynamic , changing in both number and localization within minutes ( Figure 5B , Figure S12A , and Movie S1 ) . The number of foci per micron of cell length varied among cells and within the same cell at different timepoints , however the number of foci per micron had a slightly negative correlation with the length of the cell ( Pearson r −0 . 148 , p-value 0 . 0013 ) ( Figure S12B ) . So while longer cells had more total foci than shorter cells , they had slightly fewer foci per micron . A similar pattern of DnaK localization was observed in M . bovis BCG , indicating that this pattern is conserved across saprophytic and pathogenic mycobacteria ( Figure 5C ) . In stationary phase cells , DnaK-mCitrine dramatically relocalized to form 1 or 2 foci in cells ( Figure 5D ) . This pattern of localization suggested that DnaK re-localized to aggregates formed during stationary phase . To test the role of DnaK function in this localization pattern , we generated an ATPase mutant of DnaK , K70A . Ms DnaK ( K70A ) -mCitrine failed to complement the DnaK deletion strain and also did not form foci in log or stationary phases ( Figure 5E ) despite expression as a stable full-length protein at similar levels as wildtype ( Figure S10 ) . Several components of the DnaK chaperone system have been shown to alter the oligomeric state of DnaK in E . coli including levels of ATP and GrpE , a DnaK co-factor , [45] . Elevated levels of GrpE inhibit DnaK chaperone activity [46] . We tested the effect of GrpE overexpression on DnaK function and localization . Overexpression of GrpE altered the localization pattern of DnaK such that the dynamic , peripheral foci were lost ( Figure 5F and Movie S2 ) . The dispersal of DnaK by GrpE was not due to an effect on DnaK protein levels as the abundance of the DnaK-mCitrine fusion appeared unchanged by immunoblot ( Figure S13A ) . Although prolonged overexpression of GrpE inhibited growth and inhibited luciferase activity , DnaK-mCitrine failed to localize to foci ( Figure S13B ) , indicating that overexpression of GrpE inhibits both the log phase and stationary phase functions of DnaK . The focal relocalization of DnaK in stationary phase cells suggested that DnaK may relocalize to protein aggregates . To test this hypothesis , we induced the formation of protein aggregates by expressing the aggregating protein sequence ELK16 fused to mCerulean and confirmed that mCerulean-ELK16 accumulated in the insoluble fraction ( data not shown ) . We observed mCerulean aggregates accumulating after the induction of mCerulean-ELK16 expression ( Figure 6A ) . With low levels of mCerulean-ELK16 ( 3 hours after addition of inducer ) , aggregates colocalized with DnaK in the peripheral foci characteristic of the pattern of DnaK during log phase growth ( Figure 6A , 3 hour panel ) . However , with accumulation of larger mCerulean-ELK16 aggregates , we observed relocalization of DnaK to these larger central aggregates ( Figure 6A , 20 hour panel and Movie S3 ) , a pattern that resembles that of DnaK in stationary phase cells . When we observed mCerulean-ELK16 aggregates in a strain expressing ClpB-mCitrine , we observed colocalization of ClpB to the cytoplasmic , but not peripheral , aggregates at late time points ( Figure 6B , 20 hour panel ) . Taken together , these data indicate that DnaK has two modes of chaperone function , one in native protein folding in which it is localized in mobile peripheral foci , and one in aggregate processing in which DnaK relocalizes to central immobile foci of protein aggregates , which also contain ClpB . We observed that DnaK relocalized to aggregate proteins and that DnaK-mCitrine relocalized to similar patterns during late stationary phase . We next asked whether this aggregate relocalization was reversible by observing the pattern of DnaK localization during outgrowth from stationary phase . We observed that DnaK foci were largely immobile and persisted through several rounds of outgrowth ( Figure 7A , white arrow and Movie S4 ) . By 3 hours , larger foci were still visible , but peripheral dynamic foci had reformed . By 12 hours , the original aggregate containing cell still had some immobile DnaK-mCitrine , but all daughter cells contained only dynamic peripheral foci ( Figure 7A ) . This suggested that protein aggregates formed during stationary phase are not dissolved prior to re-growth , but rather are tolerated by the mycobacterial cell for several rounds of division . This experiment also reveals that the two modes of DnaK function can coexist in the cell with DnaK dynamically shuttling between a function in aggregate processing and native folding . We utilized live cell time-lapse microscopy to directly observe the fate of protein aggregates during cell outgrowth . We first formed luciferase protein aggregates by transiently depleting DnaK , followed by outgrowth with DnaK reexpression . Luciferase-mCitrine aggregates persisted through several cycles of cell division , but cell growth initiated rapidly from the cell pole opposite the aggregate , eventually producing aggregate free cells . ( Figure 7B and Movie S5 ) . mCitrine-ELK16 aggregates behaved similarly: aggregates of mCitrine-ELK16 were very stable and persisted in cells after several rounds of division ( Figure 7C ) with a similar pattern of growth away from the aggregate , eventually forming aggregate free cells , despite persistence of the aggregate in the original cell . Thus , in three different models of protein aggregate formation ( stationary phase , DnaK depletion , and heterologous protein expression ) protein aggregates are stable during outgrowth and dissolution of aggregates is not required for reinitiation of cell growth . Aggregates were eventually lost in the population by dilution as they remained in the original parent cells but were not divided amongst daughters , suggesting that their loss was passive rather than by an active mechanism such as proteolysis .
Our results indicate that mycobacterial DnaK is the dominant chaperone responsible for folding of native peptides in the absence of exogenous stress such as heat shock . This native folding function is evident both with model protein substrates ( luciferase ) and endogenous mycobacterial proteins . Trigger Factor in mycobacteria is nonessential and cannot compensate for DnaK loss , even when overexpressed . This contrasts with the function of E . coli , in which TF and DnaK have redundant functions in native protein folding and are essential in combination [13] , [14] . Although loss of DnaK is accompanied by broad loss of protein solubility and formation of cytoplasmic protein aggregates , we also identified large multimodular lipid synthases as a specific class of proteins that require DnaK for solubility . We show that 3 of the 8 proteins greater than 2000 amino acids in the M . smegmatis proteome become insoluble in the absence of DnaK . One of these proteins , fatty acid synthase I ( FASI ) , is a eukaryotic type FAS protein that is not found in bacterial taxa except for the Mycolic acid producing Actinomycetales [47] and is essential for viability [48] . The abundance of very large lipid synthases in Actinomycetales may mandate distinct chaperone functions to assure proper folding of these large multidomain proteins , as has been shown for eukaryotic proteomes that are enriched for multidomain large proteins in comparison to E . coli [49] , [50] . Aggregation of large multidomain proteins is a feature of the proteostasis collapse that accompanies combined deletion of DnaK and TF in E . coli [7] . This requirement for DnaK in maintaining the solubility of large multimodular lipid synthases is consistent with our finding that the major morphologic and functional perturbation of mycobacterial cells upon growth arrest during DnaK depletion is perturbed membrane structure , rather than the filamentation phenotype seen in Caulobacter crescentus [51] and E . coli lacking DnaK [6] , [52] . This requirement for DnaK to maintain the solubility of the lipid biosynthetic machinery also fits with prior literature demonstrating that GroEL1 in mycobacteria is a specialized chaperone of the FASII enzymes that elongate FASI fatty acid products to Mycolic acids [53] , indicating that mycobacteria use a series of chaperones to maintain a functional lipid biosynthetic machinery . The essential function of DnaK in mycobacteria is also apparently distinct from the mechanism of essentiality recently reported in C . crescentus . Protein aggregation that accompanies loss of DnaK activates degradation of DnaA through the Lon protease with consequent cell cycle arrest and filamentation [51] . Overexpression of DnaA restores cell growth in the DnaK mutant strain , indicating that this is the sole determinant of growth arrest in DnaK depleted cells . In our experiments , DnaK depletion in mycobacteria is not accompanied by filamentation , although DnaA depletion in mycobacteria does cause a filamentation phenotype [54] . It is also unlikely that the essentiality of DnaK is due to constitutive activation of the heat shock regulon through loss of HspR repression as previous work in mycobacteria has shown that hspR is not essential for growth [55] . In addition to its role in native protein folding in unstressed cells , our data also indicate that mycobacterial DnaK also has a second function as a chaperone during states of protein aggregation , akin to the canonical role documented for DnaK in other bacteria . We observed rapid relocalization of DnaK from the dynamic mobile structures characteristic of rapid cell growth to focal protein aggregates that form during stationary phase or by expression of aggregating proteins . These foci remained fixed in both number and intensity and also contain ClpB , suggesting that DnaK and ClpB cooperate in protein aggregate processing , as has been shown in E . coli [56] . Once formed , we find that DnaK/ClpB containing protein aggregates are quite stable upon resumption of cell growth and that their dissolution or degradation was not necessary to restart growth after DnaK depletion-mediated growth arrest . This observation suggests that protein aggregates per se are well tolerated by the mycobacterial cell and that cytoplasmic aggregates are not per se toxic if the DnaK system is operative . In this regard , it is relevant that recent work has shown that ClpP1/P2 are essential in mycobacteria [57] , [58] and form a mixed heterodimer that constitutes the function ClpP protease [59] . Loss of ClpP leaves cells susceptible to proteotoxic stress and the accumulation of misfolded proteins leading to cell death . Cells depleted for ClpP function have increased susceptibility to streptomycin , an antibiotic that caused mistranslation [58] . The findings presented here identify a second crucial susceptibility point in the mycobacterial chaperone/protein quality control network that could be targeted for antimicrobial development . The nonredunant essential role of Mycobacterial DnaK suggests that small molecule inhibition of this chaperone would be lethal for mycobacteria while sensitizing them to proteotoxic stress induced by the host . Inhibition of chaperone function is an emerging therapeutic strategy in malignant cells that depend on chaperone function and could by similarly targeted as a mechanism to sensitize pathogens to the proteotoxic stress inflicted by the host during infection .
Standard procedures were used to manipulate recombinant DNA and to transform E . coli . M . smegmatis strains were derivatives of mc2155 [60] . Gene deletions were made by homologous recombination and double negative selection [61] . All strains used in this study are listed in Table S1 . Plasmids including relevant features , and primers are listed in Table S2 and S3 . M . smegmatis was transformed by electroporation ( 2500 V , 2 . 5 µF , 1000Ω ) . All M . smegmatis strains were cultured in LB with 0 . 5% glycerol , 0 . 5% dextrose ( LBsmeg ) or 7H9 media for labeling experiments . 0 . 05% Tween80 was added to all liquid media . Heat sensitivity was assayed by incubating cultures at OD600 0 . 4 at 53°C . Aliquots were taken at indicated time points and serial dilutions were plated on selective media containing ATc . Antibiotic concentrations used for selection of M . smegmatis strains were as follows: kanamycin 20 µg/ml , hygromycin 50 µg/ml , streptomycin 20 µg/ml , zeocin 12 . 5 µg/ml . For protein and epitope tag detection , the following antibodies were used: StrepTagII ( Genescript , Rabbit Anti-NWSHPQFEK polyclonal antibody , 0 . 5 mg/ml , 1∶40 , 000 ) , YFP ( Rockland Immunochemicals , Rabbit Anti-GFP polyclonal antibody , 1 mg/ml , 1∶20 , 000 ) , Luciferase ( Millipore , Goat Anti-Luciferase ( Firefly ) polyclonal antibody , 10 mg/ml 1∶20 , 000 ) , Puromycin ( KeraFast , Mouse anti-Puromycin ( 3RH11 ) monoclonal antibody , 1 mg/ml , 1∶2 , 000 ) , and RNAP-β ( Neoclone , 8RB13 Mouse Anti-E . coli RNAPβ monoclonal , 1∶20 , 000 ) . Cultures were grown in the presence of 25 ng/ml anhydrotetracycline ( ATc ) to an OD600 of 0 . 4 . Cultures were washed in equal volume of LBsmeg without ATc , then diluted back to indicated OD600 . This dilution culture was then split and 25 ng/ml ATc was added to one culture , the other was grown in the absence of ATc for depletion . All depletions were carried out in the absence of antibiotic selection . A modified previously described method for detecting ethidium bromide uptake was used [31] , [32] , [62] . Depleted and replete cultures were collected 18 hours after ATc withdrawal , washed in an equal volume of PBS with 0 . 05% Tween80 , then resuspended at an OD600 0 . 4 in PBS with 0 . 05% Tween80 and 0 . 5% glucose . 95 µl of culture was added to 0 . 2 ml pcr tubes . For efflux inhibition a final concentration of 5 µM CCCP was added just prior to the start of the assay . The assay was initiated by addition of ethidium bromide ( 0 . 25 µg/ml final ) in PBS with 0 . 05% Tween80 and 0 . 5% glucose . Tubes were incubated at 37°C and fluorescence was determined every minute for 60 minutes using an Opticon2 instrument ( MJ Research ) . Experiments were preformed twice in triplicate . DnaK depletion was carried out as described above for 16 hours . For labeling , 20 µl of 1-14C-Acetic acid , sodium salt ( Perkin Elmer , 55 . 2 mCi/mol ) was added to 20 ml of culture incubated at 37°C for 1 hour . Cells were harvested by centrifugation and washed once with 10 ml water . A final suspension of washed cells in 3 ml water was then added to 3 ml 40% tetrabutylammonium hydroxide and incubated at 100°C for 4 hours . An equal volume of dichloromethane was added then 300 µl of methyl iodide and the tubes were then rotated for 1 hour at room temperature . After the liquid layers were allowed to separate the bottom layer was removed and dried overnight . To the dried material , 2 ml ethyl ether was added and the supernatant was moved to a new tube and dried . 100 ul of ethyl ether was added just prior to spotting 5 ul samples on an HPTLC plate ( Analtech , 58077 ) . The plate was developed for 4 cycles in a 95∶5 mixture of Hexanes∶ethyl acetate . Plate imaging was preformed using a Phosphor storage cassette and Typhoon Trio ( pixel size 200 microns at best sensitivity ) . Culture volumes were normalized by OD600 . A final concentration 5 mM D-luciferin ( Gold Biotechnology ) was added to each well and Counts per second ( CPS ) was counted using a Victor2 microplate reader ( Perkin Elmer ) set to read each well in triplicates , the mean of which was calculated to yield a CPS value for each well . For DnaK depletions %RLUs was calculated by the following equation on biologic triplicate cultures: 100* ( CPSdepletion culture/CPScontrol culture ) . DnaK depletion was carried out as described above for 16 hours during which the cultures reached an OD600 0 . 4 . Puromycin was added at 50 µg/ml and the culture was incubated at 37°C . At indicated timepoints cells were harvested by centrifugation and immediately stored at −20°C until completion of the timecourse . For chloramphenicol inhibition controls , 10 µg/ml chloramphenicol was added immediately before the addition of puromycin . Experiments were preformed twice in triplicate . DnaK depletion was carried out in 7H9 media for 15 hours at 37°C until cultures reached OD600 of approximately 0 . 4 . Cultures were normalized to OD600 0 . 4 and 25 mls of culture was used for labeling . To label , 12 . 5 µl of Trans35S-LABEL ( MP Biomedicals , 10 . 4 mCi/ml ) was added and cultures were incubated with shaking for 30 minutes at 37°C . After 30 minutes 1 mM methionine was added and cultures were cooled to 4°C prior to harvesting by centrifugation . Collected cultures were fractionated for Soluble/Pelleting protein analysis as described above for the 2 Step Fractionation . After separation by SDS-PAGE , gels were dried and radioactivity quantitated using Phosphor storage cassette and Typhoon Trio ( pixel size 200 microns at best sensitivity ) . ImageJ was used to quantitate the total radioactive signal per lane . All images were acquired using a Zeiss Axio Observer Z1 microscope equipped with Definite focus , Stage top incubator ( Insert P Lab-Tek S1 , TempModule S1 ) , Colibri . 2 and Illuminator HXP 120 C light sources , a Hamamatsu ORCA-Flash4 . 0 CMOS Camera and a Plan-Apochromat 100×/1 . 4 oil DIC objective . Zeiss Zen software was used for acquisition and image export . The following filter sets and light sources were used for imaging: YFP ( 46 HE , Colibri2 . 0 505 LED ) , CFP ( 47 HE , HXP 120 C ) , Hoechst 33342 ( 49 HE , HXP 120 C ) , FM 4-64 ( 20 , HXP 120 C ) . For cell staining 100 µl of culture was used . A final concentration of 1 µg/ml FM 4-64 ( Invitrogen ) and/or 10 µg/ml Hoechst 33342 ( Invitrogen ) was added . Cells were pelleted by centrifugation at 5000 g for 1 minute and resuspended in 50 µl of media . For single time point live cell imaging , 7 µl of culture was spotted onto a No . 1 . 5 coverslip and pressed to a slide . For time-lapse microscopy , cells were added to a 1 . 5% Low melting point agarose LBsmeg pad with or without 25 ng/ml ATc . For pad preparation , LBsmeg agarose was heated to 65°C and poured into a 17×28 mm geneframe ( Thermoscientific , AB-0578 ) adhered to a 25×75 mm glass slide . A second slide was pressed down on top and the set-up was allowed to cool at room temperature for 10 minutes . The top slide was removed and the pad was cut and removed so that a 3–4 mm strip remained near the center . 2–3 µl of M . smegmatis culture was added to the pad and a No . 1 . 5 24×40 mm coverglass was sealed to the geneframe . Slides were incubated in stage top incubator at 37°C . For Luciferase-mCitrine aggregate imaging , DnaK depletion was carried out for 6 hours prior to transferring DnaK depleted cells to the pad . For Luciferase-mCitrine aggregate outgrowth imaging , DnaK depletion was performed for 24 hours prior to transferring DnaK depleted cells to a pad containing ATc to reinduce DnaK expression . 50 ml of cultures normalized to an OD600 of 0 . 4 were cooled to 4°C and harvested by centrifugation . Pellets were washed in 1 ml 10 mM Tris , pH 8 . 0 Pellets were then resuspended in 100 µl TE80 with 1 mg/ml lysozyme and disrupted by bead beating with a FastPrep120 2 times at 5 . 0 m/sec for 25 seconds . This lysate was used for RNA purification with a GeneJet RNA purification kit ( Thermoscientific ) following the manufacturer's protocol . RNA was eluted in 85 µl elution buffer and then treated with DNase I ( Thermoscientific ) for 30 minutes at 37°C . GeneJet purification columns were used to clean RNA from DNaseI reactions . First strand cDNA synthesis was carried out using Maxima Universal First Strand cDNA synthesis kit ( Thermoscientific ) with random hexamers and 500 ng RNA . For each RNA sample a no RT control was used to assess DNA contamination . qPCR was performed using DyNamo SYBR green qPCR kit ( Thermoscientific ) and an Opticon2 instrument ( MJ Research ) . For each gene , normalized cycle threshold , C ( t ) , was calculated using housekeeping gene sigA , and relative expression level was calculated using the equation 2− ( C ( t ) gene – C ( t ) sigA ) . All RT-qPCR experiments were performed 2 times in triplicate . Proteins were resolved using SDS-polyacrylamide gel electrophoresis , followed by staining with Coomassie Blue and excision of the separated protein bands; In situ trypsin digestion of polypeptides in each gel slice was performed as described [63] . The tryptic peptides were purified using a 2-µl bed volume of Poros 50 R2 ( Applied Biosystems , CA ) reversed-phase beads packed in Eppendorf gel-loading tips [64] . The purified peptides were diluted to 0 . 1% formic acid and then subjected to nano-liquid chromatography coupled to tandem mass spectrometry ( nanoLC-MS/MS ) analysis as detailed [65] . Initial protein/peptide identifications from the LC-MS/MS data were performed using the Mascot search engine ( Matrix Science , version 2 . 3 . 02; www . matrixscience . com ) with the Eubacteria segment of Uniprot protein database ( 12 , 115 , 765 sequences; European Bioinformatics Institute , Swiss Institute of Bioinformatics and Protein Information Resource ) . The search parameters were as follows: ( i ) two missed cleavage tryptic sites were allowed; ( ii ) precursor ion mass tolerance = 10 ppm; ( iii ) fragment ion mass tolerance = 0 . 8 Da; and ( iv ) variable protein modifications were allowed for methionine oxidation , cysteine acrylamide derivatization and deamidation of asparagines . MudPit scoring was typically applied using significance threshold score p<0 . 01 . Decoy database search was always activated and , in general , for merged LS-MS/MS analysis of a gel lane with p<0 . 01 , false discovery rate averaged around 1% . Scaffold ( Proteome Software Inc . , Portland , OR ) , version 4_1_1 was used to further validate and cross-tabulate the tandem mass spectrometry ( MS/MS ) based peptide and protein identifications . Protein and peptide probability was set at 99% with a minimum peptide requirement of 1 .
|
All living organisms use protein chaperones to prevent proteins from becoming insoluble either spontaneously or during cellular stress that can damage proteins . The HSP70 chaperone DnaK has been well characterized in E . coli and is important for that bacterium to resist protein denaturation from heat , but is dispensable for cell growth in the absence of stress due to redundancy with other chaperone systems . However , the function of chaperones in bacterial pathogens , which are exposed to protein stress within the host , has received less attention . Here we examine the function of DnaK in mycobacteria , a genus that includes multiple human pathogens , and find that DnaK is required for cell growth . This essential function is due to a lack of redundancy with other chaperone systems for the folding of proteins , even in the absence of stress . These findings expand the paradigm of DnaK function and identify DnaK as a promising target for antibiotic development for mycobacteria .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry",
"mycobacteria",
"proteins",
"protein",
"folding",
"medical",
"microbiology",
"chaperone",
"proteins",
"protein",
"structure",
"microbial",
"pathogens",
"biology",
"and",
"life",
"sciences",
"microbiology",
"bacterial",
"pathogens"
] |
2014
|
An Essential Nonredundant Role for Mycobacterial DnaK in Native Protein Folding
|
Kaposi’s Sarcoma-associated Herpesvirus ( KSHV ) is the etiologic agent of Kaposi’s Sarcoma ( KS ) . KSHV establishes a predominantly latent infection in the main KS tumor cell type , the spindle cell , which is of endothelial cell origin . KSHV requires the induction of multiple metabolic pathways , including glycolysis and fatty acid synthesis , for the survival of latently infected endothelial cells . Here we demonstrate that latent KSHV infection leads to increased levels of intracellular glutamine and enhanced glutamine uptake . Depletion of glutamine from the culture media leads to a significant increase in apoptotic cell death in latently infected endothelial cells , but not in their mock-infected counterparts . In cancer cells , glutamine is often required for glutaminolysis to provide intermediates for the tri-carboxylic acid ( TCA ) cycle and support for the production of biosynthetic and bioenergetic precursors . In the absence of glutamine , the TCA cycle intermediates alpha-ketoglutarate ( αKG ) and pyruvate prevent the death of latently infected cells . Targeted drug inhibition of glutaminolysis also induces increased cell death in latently infected cells . KSHV infection of endothelial cells induces protein expression of the glutamine transporter , SLC1A5 . Chemical inhibition of SLC1A5 , or knockdown by siRNA , leads to similar cell death rates as glutamine deprivation and , similarly , can be rescued by αKG . KSHV also induces expression of the heterodimeric transcription factors c-Myc-Max and related heterodimer MondoA-Mlx . Knockdown of MondoA inhibits expression of both Mlx and SLC1A5 and induces a significant increase in cell death of only cells latently infected with KSHV , again , fully rescued by the supplementation of αKG . Therefore , during latent infection of endothelial cells , KSHV activates and requires the Myc/MondoA-network to upregulate the glutamine transporter , SLC1A5 , leading to increased glutamine uptake for glutaminolysis . These findings expand our understanding of the required metabolic pathways that are activated during latent KSHV infection of endothelial cells , and demonstrate a novel role for the extended Myc-regulatory network , specifically MondoA , during latent KSHV infection .
Kaposi’s Sarcoma-associated Herpesvirus ( KSHV ) is a human γ-herpesvirus and the etiologic agent of several malignancies , including two B-cell lymphomas , primary effusion lymphoma ( PEL ) and Multicentric Castleman’s Disease ( MCD ) , as well as Kaposi’s Sarcoma ( KS ) , an angioproliferative tumor[1 , 2] . KS is the most common tumor of AIDS patients worldwide and also commonly occurs in non-AIDS patients in central Africa and the Mediterranean[2–4] . KS is a highly vascularized tumor comprised predominantly of spindle cells of endothelial origin . In both KS spindle cells and endothelial cells in culture , KSHV establishes a primarily latent infection , with only a small percentage of the tumor cells undergoing lytic replication[5 , 6] . How KSHV alters endothelial cells to lead to cancer is still an open question . Previous work from our lab and others has demonstrated that KSHV , similarly to cancer cells , induces several major metabolic pathways . These alterations in cellular metabolism are imperative to the survival of cells latently infected with KSHV[7–9] . During latent KSHV infection , glucose uptake is induced and lactate production is significantly increased[7] . This switch to aerobic glycolysis is characteristic of the Warburg effect , a hallmark of cancer cell metabolism[10] . Interestingly , KSHV-infected endothelial cells require the Warburg effect for their survival , as latently infected endothelial cells are extremely sensitive to drug inhibition of glycolysis[7] . Recent evidence supports that the viral miRNAs expressed during latency are sufficient for the induction of the Warburg effect in KSHV-infected cells[11] . Our lab has also shown that KSHV induces the production of lipids via fatty acid synthesis ( FAS ) during latent infection[8] . Over half of the long-chain fatty acids detected in our metabolomics screen were elevated following latent KSHV infection . Lipid droplet organelles were also increased by latent KSHV infection of endothelial cells , evidence of increased fatty acid synthesis . Inhibition of FAS leads to apoptosis of KSHV-infected cells , which was rescued with supplementation of palmitate , a downstream metabolic intermediate of FAS . These data indicated that downstream intermediates of FAS are required for endothelial cell survival during latent infection . Induction of both glycolysis and FAS are also required in primary effusion lymphoma cells where KSHV is present[9] . Both the Warburg effect and increased FAS are metabolic signatures found in most cancer cells[12] . In these cells , glucose is primarily being utilized to produce lactic acid and fatty acids and is therefore diverted away from the tri-carboxylic acid ( TCA ) cycle . The TCA cycle metabolizes carbon to produce both bioenergetic and biosynthetic precursors . Importantly , glutamine carbon can be utilized to replenish the TCA cycle through a process termed anaplerosis[13] . Glutamine is the most abundant amino acid available to mammalian cells . Cancer cells induce glutamine uptake to support a glutamine requirement that exceeds the amount that cells can synthesize . Cancer studies have shown that transformed cells become glutamine addicted , or dependent on this exogenous glutamine and its catabolism via glutaminolysis for their survival[14 , 15] . Recent evidence demonstrates that glutamine addiction in some cancers is enabled by the extended Myc network . Together , Myc-Max , with MondoA , a nutrient-sensing transcription factor , and its heterodimerization partner , the Max-like protein X ( Mlx ) , facilitate the reprogramming of cellular metabolism in Myc-overexpressing cells [16–18] . A number of lytically replicating viruses also require glutamine for maximal viral replication[19–21] . Previous studies have shown that poliomyelitis virus and human cytomegalovirus depend on both glucose and glutamine for efficient virus replication[19 , 21] . Interestingly , during vaccinia virus infection , glucose is completely dispensable for viral replication , but viral infection is reliant on glutamine for maximal virion production[20] . However , no studies have examined glutamine dependence during de novo KSHV infection . We show that latent KSHV infection of endothelial cells induces glutamine uptake and that infected cells are dependent on the catabolism of glutamine for their survival . In the absence of exogenous glutamine , a significant percentage of KSHV-infected endothelial cells undergo apoptosis unless supplemented with TCA cycle intermediates such as alpha-ketoglutarate ( αKG ) or pyruvate . Targeted drug inhibition of glutamine uptake or glutaminolysis during latent infection recapitulates the findings from the glutamine-deprived conditions . Additionally , we show that KSHV infection induces protein expression of c-Myc , its dimerization partner Max , MondoA , and its dimerization partner , Mlx . KSHV infection also induces protein expression of the glutamine-transporter protein SLC1A5 . c-Myc coordinately with MondoA/Mlx is essential for regulation of glutaminolysis in cancer cells[16 , 18] and is also necessary for the induction of SLC1A5 in KSHV-infected endothelial cells . Inhibition of MondoA or SLC1A5 induces cell death in KSHV-infected cells , but not mock-infected cells , and can be rescued with supplementation of αKG . Therefore , latent KSHV infection induces and requires glutamine uptake and subsequent glutaminolysis , regulated by MondoA and the glutamine transporter SLC1A5 , for the survival of latently infected endothelial cells .
A global metabolomics screen identified that glutamine levels are significantly elevated at both 48 and 96 hours post latent KSHV infection[8] . Intracellular glutamine abundance is elevated 2 . 2 fold at 48 hours post infection ( hpi ) and 2 . 7 fold at 96 hpi , as compared to mock-infected cells ( Fig 1A ) . To determine if the increased levels of glutamine in infected cells was due to increased uptake during latent infection , a radiolabeled glutamine molecule , [3H]-Glutamine , was added to the media of mock- and KSHV-infected Tert-Immortalized Microvascular Endothelial ( TIME ) cells at 96 hpi . Intracellular radiolabeled glutamine levels were then determined 10 min post treatment by scintillation . Latent KSHV infection induces glutamine uptake by approximately 35% compared to mock-infected cells ( Fig 1B ) . These data validate that elevated levels of glutamine during latent KSHV infection are a result of an increase in exogenous glutamine uptake . To determine if exogenous glutamine is a required carbon source for the survival of endothelial cells latently infected with KSHV , we quantified cell death over time in the presence or absence of exogenous glutamine . TIME cells were mock- or KSHV-infected and allowed to establish latency for 24 hours . Cells were re-seeded into 24-well plates , and overlaid with replete media , which contains 4mM glutamine , or glutamine-free media . Both treatment medias were prepared with dialyzed FBS , depleted of small molecules , including glutamine , and experiments were performed in triplicate . Average cell death over time was measured using the live-cell Essen Bioscience IncuCyte imaging system , which records both phase-contrast as well as fluorescent images over time . Dead cells were identified using the fluorescent nuclear dye YOYO-1 , a cell impermeable dye that only enters cells with compromised membranes . Total cell number was determined by using SytoGreen24 , a cell permeable dye that enters all cell nuclei . Percent cell death was calculated by dividing the total number of dead cells ( YOYO-1 positive ) by the total number of cells ( SytoGreen24 positive ) . Cell death was monitored for 48 hours ( 24 hpi through 72 hpi ) . Fig 2A shows the average percent cell death recorded every 2 hours over 48 hours of monitoring for three biological replicate infections . The bar graph shows the average cell death at 0 , 24 , and 48 hours post treatment for each condition . In mock-infected cells , with replete or glutamine-deprived media , there is less than 5 percent cell death over the time monitored ( Fig 2A ) . KSHV-infected cells in replete media have a slight increase in cell death over the time course . However , glutamine starvation of KSHV-infected cells induces a significant increase in cell death , approximately 25–30% after 48 hours of treatment ( 72 hpi ) . Microscopy images were analyzed for positive cell nuclei based on size and fluorescence intensity for both YOYO-1 and SytoGreen24 . Representative images of YOYO-1 positive cells at 48 hours post treatment are shown in Fig 2B . To ensure that glutamine addiction is not simply due to virus binding and entry , we repeated the experiments with UV-irradiated KSHV . UV-irradiated virus is able to bind and enter cells , but does not support viral gene expression . KSHV-infected cells show a 10-fold increase in cell death upon glutamine starvation , whereas UV-irradiated KSHV-infected cells show similar levels of cell death to mock-infected cells ( Fig 2C ) . Therefore , KSHV viral gene expression is required to induce the dependence on glutamine and establish a state of glutamine addiction in endothelial cells . To show that KSHV glutamine addiction was not limited to TIME cells , we conducted similar cell death experiments upon depletion of glutamine in mock- and KSHV-infected primary human dermal microvascular endothelial cells ( 1° hDMVECs ) . Mock- and KSHV-infected 1° hDMVECs were overlaid with glutamine depleted media at 48 or 72 hpi . Fourty-eight hours post glutamine depletion , YOYO-1 and SytoGreen24 counts were measured on a Typhoon scanner to determine relative fluorescence for cell death . These experiments reveal a significant increase in cell death only in KSHV-infected 1° hDMVECs , substantiating that KSHV infection of 1° hDMVECs also induces glutamine addiction ( Fig 2D ) . We have previously shown that inhibition of glycolysis and fatty acid synthesis leads to cell death via apoptosis of endothelial cells latently infected with KSHV[7 , 8] . It has also been shown that glutamine deprivation leads to apoptosis of cancer cells[17] . To determine if glutamine starvation induces cell death via activation of an apoptotic pathway , we performed the previously described cell death assay using YOYO-1 and SytoGreen24 counts in the presence or absence of the pan-caspase inhibitor QVD . Upon supplementation with QVD , KSHV-infected cells deprived of glutamine are rescued from cell death ( Fig 3A ) , indicating that cell death is due to caspase-dependent apoptosis . To confirm that cells are dying via apoptosis , we utilized a fluorogenic Caspase-3/7 substrate , which contains the caspase cleavage site , a short four amino acid peptide ( DEVD ) , conjugated to a nucleic acid binding dye . This cleavage site is specifically targeted by activated executioner caspases 3 and 7 . When caspase 3 and/or 7 are activated during apoptosis , the DEVD site is cleaved , resulting in the release of the DNA dye , translocation to the nucleus and fluorescence . For these experiments , the Caspase-3/7 substrate was added to mock- and KSHV-infected cells in the presence or absence of glutamine at 24 hpi . After 48 hours of treatment , plates were scanned for relative fluorescence using a Typhoon 9400 variable mode imager . Caspase-3/7-mediated relative fluorescence was normalized to SytoGreen24 relative fluorescence from the same experiment . Only KSHV-infected cells starved of glutamine showed significant detection of fluorescence from the Caspase-3/7 substrate , indicating that latent infection induces Caspase 3 and/or 7 activation , which in turn results in an elevated level of DEVD cleavage and nuclear fluorescence ( Fig 3B ) . We also included samples supplemented with QVD . These samples showed no increased fluorescence even in the absence of glutamine ( Fig 3B ) . Representative microscopy images at 48 hours post treatment were captured with the Cellomics ArrayScan Vti ( Fig 3C ) . Overall , these data indicate that when deprived of glutamine , KSHV-infected endothelial cells activate apoptosis in a Caspase-3/7 dependent manner . Upon entering the cell , glutamine is catabolized via glutaminolysis . Glutaminolysis consists of two consecutive deamination steps . First , glutamine is converted to glutamate by glutaminase ( GLS ) . Second , glutamate is converted to αKG by one of three enzymes: glutamate dehydrogenase ( GDH ) , glutamate pyruvate transaminase ( GPT ) or glutamate oxaloacetate transaminase ( GOT ) [14 , 22] . At this stage , αKG can enter and replenish the TCA cycle . To determine if glutamine is required to maintain the TCA cycle in KSHV-infected cells , we added either membrane-soluble αKG or pyruvate , both of which can enter the TCA cycle , to the treatment medium of glutamine-deprived cells during latent KSHV infection . After mock- or KSHV-infection of TIME cells for 24 hours to allow the establishment latency , cells were re-seeded as before and overlaid with replete media , glutamine-free media , or glutamine-free media supplemented with either 3 . 5 mM αKG or 8 mM pyruvate . Supplementation with αKG completely rescues the glutamine-deprived KSHV-infected cells from cell death and supplementation with pyruvate significantly rescues cell death in the glutamine-deprived infected cells ( Fig 4A ) . These metabolite rescue data support the model that the exogenous glutamine taken up by KSHV-infected endothelial cells is necessary to support glutaminolytic metabolism for replenishment of the TCA cycle . BPTES is a specific inhibitor of GLS , the first enzyme of glutaminolysis . When treated with BPTES in the presence of 4mM glutamine ( replete media ) , KSHV-infected endothelial cells died at similar levels to those deprived of glutamine , while having little effect on mock-infected cells ( Fig 4B ) . These data recapitulate our findings with glutamine-deprived media . Taken together , these data validate that glutamine is essential for glutaminolysis in KSHV-infected cells . Glutamine metabolism is regulated by oncogenic c-Myc in many cancer cells[16 , 17 , 23] . Additionally , there is evidence that c-Myc is regulated by latent KSHV infection[24 , 25] . Recently , it was shown that c-Myc[26] , and N-Myc[18] , manipulate metabolic gene expression coordinately with the Myc-bHLHZ superfamily members MondoA , a nutrient-sensing protein , and its dimerization partner , Mlx . MondoA/Mlx or the paralogue ChREBP/Mlx constitute the “nutrient-sensing” arm of the extended Myc network[27] . Protein expression of c-Myc , Max , MondoA and Mlx are increased during latent KSHV infection of TIME cells , as determined by immunoblot analysis of whole cell lysates harvested at 48 hpi ( Fig 5 ) . Additionally , a known target of activated MondoA/Mlx , TXNIP , is upregulated at the protein level during latent KSHV infection ( Fig 5 ) . It has been shown that Myc/MondoA controls glutamine metabolism by inducing the expression of the major glutamine transporter , SLC1A5[18] . SLC1A5 is a neutral amino acid transporter which localizes to the cellular membrane , and is known to primarily import glutamine[28] . SLC1A5 is upregulated in many cancer cells [16 , 18 , 28] . There is a small , but reproducible , increase in SLC1A5 protein in TIME cells latently infected with KSHV when whole cell lysates are compared by immunoblot analysis at 48 hpi ( Fig 5 ) . Together , these data suggest that latent KSHV infection induces changes to the Max/Mlx-regulation network consistent with coordinate regulation of metabolism , including glutamine uptake through SLC1A5 . To determine the role of the glutamine transporter SLC1A5 during latent infection , the SLC1A5 specific inhibitor L-γ-Glutamyl-p-nitroanilide ( GPNA ) was used [29] . Mock- and KSHV-infected TIME cells were re-seeded at 24 hpi and overlaid with replete media or replete media treated with 5mM GPNA . YOYO-1 or SytoGreen24 were added to compare the relative florescence of dead cells and the relative fluorescence of total cells , respectively , at 48 hours post treatment . These experiments were conducted using the Typhoon 9400 variable mode imager to measure relative fluorescence of all samples . GPNA treatment leads to increased cell death only in KSHV-infected cells but not their mock counterparts ( Fig 6A ) . Importantly , when supplemented with 3 . 5 mM αKG , cell death induced by GPNA treatment of KSHV-infected endothelial cells was rescued to KSHV replete control treatment levels , indicating that the drug-induced cell death was due to the requirement of glutamine metabolism via glutaminolysis and not off-target effects . To further confirm the drug studies , a validated siRNA set directed to SLC1A5 was used to knockdown SLC1A5 expression[18] . SLC1A5 expression was reduced by approximately 70% in TIME cells transfected with a mix of four siRNAs specific for SLC1A5 ( siSLC1A5 ) , as compared to cells transfected with a scrambled non-target control ( siControl ) ( Fig 6B ) . Twenty-four hours post transfection with the SLC1A5 or control siRNA , cells were either mock- or KSHV-infected and subsequently provided replete media containing YOYO-1 for cell death or SytoGreen24 to identify all cells . Plates were scanned at 48 hours post treatment ( 72 hpi ) for relative fluorescence on the Typhoon 9400 variable mode imager . Minimal cell death was observed in both mock- and KSHV-infected cells treated with siControl . KSHV-infected cells , but not mock-infected cells , transfected with the siSLC1A5 show an increase in cell death . The fold change in relative fluorescence for cell death of KSHV-infected cells over mock-infected cells is increased in cells transfected with siSLC1A5 compared to cells transfected with siControl ( Fig 6B ) . Together , these data support that KSHV-infected endothelial cells rely on the expression of the glutamine transporter SLC1A5 for survival . SLC1A5 is directly regulated by the nutrient-sensing Myc extended network member MondoA in many human cancer cells[18] . To determine if MondoA controls SLC1A5 expression during latent KSHV infection of endothelial cells , we examined the expression of SLC1A5 upon siRNA knockdown of MondoA in mock- and KSHV-infected endothelial cells . MondoA protein expression was significantly reduced in both mock and KSHV-infected TIME cells transfected with a mix of four siRNAs specific for MondoA ( siMondoA ) , as compared to cells transfected with a scrambled non-target control ( siControl ) ( Fig 7A ) . While SLC1A5 protein levels are elevated by KSHV infection , loss of MondoA results in a reduction in detected SLC1A5 in all samples . Additionally , protein levels of Mlx , a co-stabilized MondoA binding partner[18] , and TXNIP , a known downstream target of MondoA/Mlx regulation , are also reduced upon loss of MondoA . These data support the hypothesis that MondoA is directly regulating SLC1A5 , the major glutamine transporter , during latent KSHV infection of endothelial cells . To determine if MondoA is required for endothelial cell survival during latent KSHV infection , we examined cell death in the presence of control siRNA or siRNA directed against MondoA . As shown in Fig 7B , only KSHV-infected endothelial cells in the absence of MondoA show a significant increase in cell death at 48 hpi , indicating that MondoA is indeed required for the survival of latently infected cells . Importantly , this significant increase in cell death is fully rescued upon supplementation with αKG , indicating that the cell death that occurs in KSHV-infected cells where MondoA is knocked down is due to a loss of TCA cycle intermediates and not an unrelated function of MondoA .
Transformed cells were first described as ‘glutamine addicted’ in the 1950’s[15] . It is now well established that glutamine , the most abundant amino acid in plasma , is ‘conditionally essential’ for cancer cell growth and survival[13] . More recent evidence shows that lytically replicating viruses orchestrate specific cellular metabolic modifications to support the unique requirements for their viral replication[19 , 20 , 30–32] . We demonstrate that latent infection with KSHV , an oncogenic virus , induces glutaminolysis in endothelial cells . In addition to showing that latent KSHV infection enhances glutamine uptake during infection , we have shown that a significant percentage of latently infected endothelial cells become glutamine addicted , and that glutaminolysis is required for the survival of these cells . Deprivation of glutamine in both TIME cells and 1°hDMVECs leads to significant increases in apoptosis unless they are supplemented with TCA cycle intermediates . Glutaminolysis is an important anaplerotic reaction that produces αKG , which can enter the TCA cycle ( Fig 8 ) . TCA cycle intermediates support the production of both bioenergetic and biosynthetic precursors; therefore , glutamine is potentially required for a variety of downstream cellular processes including ATP and NADPH production and fatty acid synthesis[33] . There is substantial evidence in cancer biology that glutamine metabolism is required to replenish the TCA cycle when glucose is being metabolized to lactic acid as part of the Warburg effect[13] . Previous research from our lab has shown that induction of the Warburg effect is required for the survival of endothelial cells during latent KSHV infection . Therefore , we were interested in the role glutamine metabolism may play in KSHV-infected endothelial cells . Human cytomegalovirus and vaccinia virus require glutamine to support the TCA cycle for maximal virus replication and media supplemented with TCA cycle intermediates , such as αKG or pyruvate , rescued replication in the absence of glutamine[19 , 20] . Our data supports that glutamine is a vital carbon source during latent infection with KSHV . A recent study reported an increase in glutamate secretion during latent KSHV infection[25] . Glutamate is produced intracellularly through the deamination of glutamine ( Fig 8 ) . When glutamate secretion was inhibited , cell proliferation was reduced; however , apoptosis was not reported upon treatment with glutamate secretion inhibitors . Therefore , the increase in glutamine uptake that we observe during latent KSHV infection could be supporting the pleiotropic role of glutamine during infection to support multiple cellular processes , including anaplerosis to support the TCA cycle as well as signaling to the extracellular environment . We demonstrate that the glutamine transporter SLC1A5 is upregulated during latent KSHV infection of endothelial cells , and that specifically the latently infected cells are dependent upon SLC1A5 for survival . This was of specific interest because previous studies have shown that oncogenic c-Myc , or N-Myc , induces increased expression of the glutamine transporter SLC1A5 , and dependency upon it for survival in Myc-activated cells[26] . Multiple studies have reported that c-Myc is upregulated during KSHV infection[25 , 34] . We observed an upregulation in c-Myc during infection of endothelial cells , but also identified a significant upregulation in the related proteins MondoA and Mlx . These proteins are a part of the expanded Myc network , known as the Max/Mlx network . MondoA and Mlx form an important glucose-responsive heterodimer that participates in regulating cellular metabolism , specifically glucose , lipid and glutamine metabolism in collaboration with c-Myc or N-Myc . It was recently described that both Myc overexpression and MondoA expression are required to induce the expression of glutamine transporters , including SLC1A5 , as well as induce glutaminolysis[18] . We find that MondoA regulation is required for the survival of latently infected endothelial cells and that supplementation with αKG , the immediate downstream intermediate of glutaminolysis and TCA cycle metabolite , promotes cell survival , similarly to our findings upon glutamine deprivation . This is the first evidence of the requirement for MondoA metabolic regulation during human viral infection . While we have delineated the cellular mechanism of KSHV-induced glutamine addiction , the latent viral gene or set of genes sufficient to induce the MondoA-mediated metabolic switch to glutamine addiction has not yet been determined . Previous research has identified that the latent KSHV protein LANA collaborates with Myc to stabilize and activate the transcriptional regulator during infection[34] . However , this story may be more complicated . It was recently shown that expression of the latent KSHV miRNA cluster is sufficient to induce glucose uptake and glycolysis[11] . If alterations in glucose and glutamine metabolism are interconnected , such as a requirement for glucose to activate MondoA/Mlx , it is likely that multiple viral genes are involved and more work is needed to identify which latent factors are necessary to activate the overall metabolic signature that is required during latent KSHV infection of endothelial cells . Several major metabolic switches are required during latent KSHV infection; however , the question remains whether induction of cancer cell metabolism is pre-adapting cells for a cancer microenvironment or if these alterations are helping drive oncogenesis when cells are placed in the correct microenvironment . Our findings are in agreement with metabolic signatures described by many cancer studies , which would be predicted if latent KSHV infection is indeed predisposing cells for oncogenesis . However , these models are not necessarily mutually exclusive . Comparing induced metabolic phenotypes , such as the Warburg effect and glutamine addiction in a viral system , where we can include mock controls , provides a unique model to identify the initial drivers of oncogenesis as well as characterize the suitable microenvironment established . Glutamine addiction may be induced early in oncogenesis , yet also be a characteristic of long-term tumor maintenance . Drug inhibitors specifically targeting glutamine-addicted cells could also provide novel therapeutic treatments to specifically target endothelial cells latently infected with KSHV .
Tert-Immortalized Microvascular Endothelial ( TIME ) cells [35] and primary human dermal microvascular endothelial cells ( 1° hDMVECs ) ( Lonza , MD ) were maintained as monolayer cultures in EBM-2 media ( Lonza or Cellgro ) or EndoGrow ( Millipore ) supplemented with a bullet kit containing 5% FBS , vascular endothelial growth factor , basic fibroblast growth factor , insulin-like growth factor 3 , epidermal growth , and hydrocortisone . Millipore EndoGrow media , supplemented with dialyzed FBS ( depleted of small molecules including glucose and glutamine ) was used for all experiments that compare replete ( 4 mM glutamine ) and glutamine-free media . KSHV inoculum from induced BCBL-1 cells was titered and used to infect TIME cells or 1° hDMVECs as previously described[36] . Infections were performed in serum-free EBM-2 media and subsequently overlaid with complete EBM-2 media . Infection rates were assessed for each experiment by immunofluorescence and only experiments where greater than 85% of the cells expressed LANA , a latent marker , and less than 1% of the cells expressed ORF59 , a lytic marker , were used . In a subset of the siRNA transfection experiments where larger quantities of siRNA were used , there was a slight increase in the cells expressing ORF59 , but this always occurred in both the control and gene specific siRNA transfections and did not alter the results of the experiments . YOYO-1 and SytoGreen24 were diluted in DMSO and used at a final concentration of 100 nM and 50 nM respectively ( Life Technologies ) . Dimethyl-α-ketoglutarate ( alpha-ketoglutarate ) and pyruvate were purchased from Sigma and used at 3 . 5 mM and 8 mM respectively . Bis-2- ( 5-phenylacetamido-1 , 3 , 4-thiadiazol-2-yl ) ethyl sulfide , or BPTES ( Sigma ) was solubilized in DMSO , subsequently diluted in methanol and used at a final concentration of 2 . 5 μM . QVD-OPH ( SMBiochemicals ) and was dissolved in DMSO and used at a final concentration of 20 μM . L-γ-Glutamyl-p-nitroanilide ( GPNA ) ( Sigma ) , was prepared in DMSO in a 1 M stock solution and used at a final concentration of 5mM . Twenty-four hours post mock- or KSHV-infection; TIME cells were re-seeded into 12-well plates at equal numbers in triplicate . At 96 hpi , cells were overlaid with serum-free media for 2 hours . Cells were then washed three times with PBS before the addition of 1mL of serum-free media containing 0 . 5μCi ( 10 pmol ) of [3H]-L-glutamine ( Perkin Elmer #NET551 ) . Cells were incubated for 10 minutes at 37°C . Following incubation , the medium was removed and each well was washed twice with 1mL of ice-cold DPBS and 200μL of lysis buffer ( 1% SDS in PBS ) was added to each well and incubated at room temperature with occasional agitation for 5 minutes . Lysates were transferred to microcentrifuge tubes and mixed by vortexing . 150μL of each lysate was transferred to a vial containing 4mL of Biofluor Plus scintillation fluid ( Perkin Elmer ) . Each vial was mixed by vortexing and counted in a Beckman LS6500 liquid scintillation counter . The remaining lysate was quantified by BCA Protein Assay Reagent Kit ( Pierce ) for normalization . At 20 hpi , mock- and KSHV-infected TIME cells were re-seeded into 24-well plates . At 24 hpi , cells treated with Replete ( 4 mM glutamine ) , glutamine-free media or replete media with 2 . 5 μM BPTES in triplicate . Of note , no changes in latent or lytic infection rates were observed after glutamine starvation . YOYO-1 , to identify dead cells , or SytoGreen24 , to mark all cell nuclei , were added at this step . For rescue studies , supplementation with 3 . 5 mM αKG , 8 mM pyruvate or 20 μM QVD were added at this step . Plates were then placed on the IncuCyte ( Essen Biosciences ) , a live-cell phase-contrast and fluorescent imaging system and recorded for cell death and total cell number for 48 hours ( 24 hpi through 72 hpi ) . GPNA experiments were prepared according to the same protocol , but scanned on the Typhoon 9400 variable mode imager ( GE Healthcare ) and analyzed with ImageJ software for relative fluorescence at 48 hours post treatment . Apoptosis experiments conducted with the apoptosis marker , Caspase-3/7 substrate , were prepared according to the same protocol , but the Caspase-3/7 Cell Event reagent was added , plates were scanned at 48 hours post treatment on the Typhoon 9400 and ImageJ software and normalized to relative florescence for Styogreen24 . Primary hDMVEC experiments were re-seeded into 24-well plates and at 48 or 72 hpi were treated with Replete ( 4 mM glutamine ) or glutamine-free media and 48 hours post treatment ( 96 or 120 hpi ) were scanned on the Typhoon 9400 variable mode imager ( GE Healthcare ) and analyzed with ImageJ software for relative fluorescence . All cells were lysed in RIPA and protein was quantified using BCA Assay ( Pierce ) . 30–50ug were subjected to SDS-PAGE in 1xMES Buffer ( Life Technologies ) on a 4–12% NuPAGE Bis-Tris Gel ( Life Technologies ) then transferred to 0 . 2um nitrocellulose membrane ( Bio-Rad ) . The membranes were blocked in 5% Non-Fat Dry Milk in TBS with 0 . 1% Tween ( TBST ) for at least an hour then probed with the indicated primary antibodies diluted in 5% milk in TBST for 2 hours at RT , or overnight at 4C ( anti-c-Myc ( Abcam ) , anti-Max ( Santa Cruz Biotechnology ) , anti-MondoA ( Proteintech ) , anti-Mlx ( Santa Cruz Biotechnology ) , anti-SCL1A5 ( Cell Signaling ) and anti-TXNIP ( MBL , JY1 ) . Blots were washed 3 times in TBST , then probed with HRP-conjugated secondary antibody ( Cell Signaling ) diluted in 5% milk in TBST for 1 hour at RT . Blots were washed 3 times in TBST , then subjected to chemiluminescence and exposed to blue autoradiography film ( GeneMate ) and processed in an autoprocessor . Total RNA was isolated from TIME cells 72 hours post siRNA transfection using the Nucleospin RNA II Kit ( Macherey-Nagel ) . Two-step quantitative real-time reverse transcription PCR ( BioRad ) was used to measure expression levels of SLC1A5 and the housekeeping gene GAPDH . iScript Reverse Transcription Supermix and SsoAdvanced SYBR Green Supermix ( BioRad ) were used according to manufacturer’s protocols . The primers used were: SLC1A5-F ‘5-TTATCCGCTTCTTCAACTCCTT-3’ , SLC1A5-R ‘5-ACATCCTCCATCTCCACGAT-3’ , or GAPDH-F: ‘5-GGACTCATGACCACAGTCCA-3’ , GAPDH-R ‘5-CCAGTAGAGGCAGGGATGAT-3’ . Relative levels of SLC1A5 mRNA were normalized by the delta threshold cycle method to the abundance of GAPDH mRNA . A set of four siRNAs specific to the glutamine transporter SLC1A5 ( siSLC1A5 ) and MondoA ( siMondoA ) were purchased ( Qiagen , Flexitube GeneSolution #GS6510 and #GS22877 respectively ) . A negative-control siRNA ( siControl ) was designed and synthesized by Ambion . TIME cells were transfected with siRNA at a final concentration of 200 nM , using the Amaxa Nucleofector Kit by Lonza according to the manufacturer’s protocol . At 24 hours post transfection , cells were mock- or KSHV-infected . Upon completion of the infection , cells were washed and treated with Replete media containing YOYO-1 or SytoGreen24 . Relative fluorescence was measured 48 hours post treatment using a Typhoon 9400 variable mode imager ( GE Healthcare ) and ImageJ software .
|
KSHV is the etiologic agent of KS , the most common tumor of AIDS patients worldwide . Currently , there are no therapeutics available to directly treat latent KSHV infection . This study reveals that latent KSHV infection induces endothelial cells to become glutamine addicted , similarly to cancer cells . Extracellular glutamine is required to feed the TCA cycle through glutaminolysis , a process called anaplerosis . KSHV induces protein expression of the glutamine transporter SLC1A5 and SLC1A5 expression is required for the survival of latently infected cells . KSHV also induces the expression of the proto-oncogene Myc and its binding partner Max , as well as , the nutrient-sensing transcription factor , MondoA and its binding partner Mlx . MondoA regulates SLC1A5 and glutaminolysis during latent KSHV infection , and its expression is required for the survival of latently infected endothelial cells . These studies show that glutaminolysis and a single glutamine transporter , under the regulation of MondoA , are required for the survival of latently infected cells , providing novel druggable targets for latently infected endothelial cells . This work supports that a cancer-like metabolic signature is established by latent KSHV infection , opening the door to further therapeutic targeting specifically of KSHV latently infected cells .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Latent KSHV Infected Endothelial Cells Are Glutamine Addicted and Require Glutaminolysis for Survival
|
Clinical reports of Zika Virus ( ZIKV ) RNA detection in breast milk have been described , but evidence conflicts as to whether this RNA represents infectious virus . We infected post-parturient AG129 murine dams deficient in type I and II interferon receptors with ZIKV . ZIKV RNA was detected in pup stomach milk clots ( SMC ) as early as 1 day post maternal infection ( dpi ) and persisted as late as 7 dpi . In mammary tissues , ZIKV replication was demonstrated by immunohistochemistry in multiple cell types including cells morphologically consistent with myoepithelial cells . No mastitis was seen histopathologically . In the SMC and tissues of the nursing pups , no infectious virus was detected via focus forming assay . However , serial passages of fresh milk supernatant yielded infectious virus , and immunohistochemistry showed ZIKV replication protein associated with degraded cells in SMC . These results suggest that breast milk may contain infectious ZIKV . However , breast milk transmission ( BMT ) does not occur in this mouse strain that is highly sensitive to ZIKV infection . These results suggest a low risk for breast milk transmission of ZIKV , and provide a platform for investigating ZIKV entry into milk and mechanisms which may prevent or permit BMT .
Zika virus ( ZIKV ) is an enveloped virus with a positive-sense , single-stranded RNA genome [1] . For over half a century , this flavivirus was regarded as an arbovirus leading to self-limiting , febrile disease . However , confirmation of or association with new syndromes , including teratogenesis , adult Guillain Barre Syndrome , genital persistence , and sexual transmission , have begun to emerge since the 2015–2016 Brazil ZIKV outbreak . Due to devastating outcomes associated with infection of the developing brain and ZIKV’s apparent ability to cross intact mucosae [2–4] , a key question arises: can ZIKV be transmitted by breast milk ? Reports of ZIKV RNA detection in breast milk are accumulating [5–10] . Although no epidemiologic data regarding ZIKV in lactating women are currently available , ZIKV RNA has been reported in breast milk from 3 [5 , 9] to 33 [6] days after maternal onset of fever . Reports conflict as to whether isolated ZIKV RNA represents infectious virus [7] . In one study , cytopathic effect ( CPE ) could not be demonstrated in cells cultured with either of the breast milk samples from two mothers who nursed infected infants [9] . In two separate reports , CPE was seen upon culturing of cells with breast milk of mothers with uninfected nursing children [8 , 10] . In another study , CPE was demonstrated in cells cultured with milk from a ZIKV-infected mother , and the nursing child was infected with an isolate with ZIKV genome identity of more than 99% between the infected mother and child [5] . Historically , the epidemiology and mechanisms of flavivirus breast milk transmission ( BMT ) have posed somewhat of a scientific enigma . Hepatitis C virus or Japanese encephalitis virus BMT has not been documented , whereas West Nile virus [11] and yellow fever vaccine strain [12] BMT have been reported . Dengue virus ( DENV ) infects approximately 390 million people annually and DENV RNA has been detected in breast milk [13] , but reports of BMT are rare . Furthermore , in the 1970s , two studies also demonstrated that DENV and Japanese encephalitis virus were neutralized by the lipid fraction of breast milk [14 , 15] . In this study , we explored a mouse model for BMT of ZIKV using AG129 mice that are deficient in both type I and II interferon ( IFN ) receptors , and represent a highly sensitive animal model of ZIKV challenge [16] . Following infection of AG129 dams with ZIKV on the date of parturition , viral RNA was detected in pup stomach milk clots ( SMC ) as early as 1 day post maternal infection ( dpi ) and as late as 7 dpi . In contrast , ZIKV NS2B immunofluorescent immunohistochemistry ( IHC ) and examination for CPE of inoculated Vero cells and focus forming assay did not demonstrate infectious virus in fresh milk or in nursing pups . Enzyme IHC provided evidence of intracellular viral replication ( i . e . ZIKV NS2B expression ) in cells morphologically consistent with epithelial cells , myoepithelial cells , and macrophages within the mammary gland . ZIKV NS2B expression was observed also in the SMC , and infectious particles were observed in fresh milk samples after 3 serial passages in Vero cells . The detection of potentially infectious ZIKV in the milk of this mouse model suggest that infectious virus may be present in human breast milk . However , BMT did not occur in this highly stringent ZIKV challenge system . These results suggest a low risk for human BMT of ZIKV , and set the stage for investigating ZIKV entry into milk and mechanisms by which BMT are prevented or permitted .
129/Sv mice deficient in type I and type II IFN receptors ( AG129 ) were bred and maintained at the La Jolla Institute for Allergy & Immunology ( LJI ) under standard pathogen free conditions . LJI has established an animal care and use program in compliance with The Public Health Service Policy on the Humane Care and Use of Laboratory Animals and maintains an animal welfare assurance with the Office of Laboratory Animal Welfare ( OLAW ) . The animal care and use program is guided by the US Government Principles for the Utilization and Care of Vertebrate Animals Used in Testing , Research and Training and by the 8th edition of the Guide for the Care and Use of Laboratory Animals . As such , all research involving animals is reviewed and approved by the IACUC in accordance with The PHS policy on the Humane Care and Use of Animals and the 8th edition of The Guide . In addition , LJI’s animal care and use program is accredited by AAALAC International . All experiments involving these mice were approved by the Institutional Animal Care and Use Committee under protocol no . AP028-SS1-0615 . Samples sizes: Fig 1 ( 1A to 1D: 3 pups per group from 3 separate mothers , 1E: 3 pups per group from 3 other separate mothers ) , Fig 2 ( 3 mothers per group ) , Fig 3 ( 3 mothers per group ) , Fig 4 ( 4A and 4B: 6 pups per group from 3 separate mothers , 4C and 4D: 3 pups per group from 3 separate mothers , 4E to 4F: other 3 pups per group from 3 separate mothers ) . Fig 5 ( 5A to 5C: images representative from 3 independent experiments , 5D: 3 mothers per group , 5E: 10 pups per group from 3 separate mothers ) . Animal experiments were not randomized or blinded . ZIKV strain FSS13025 ( Cambodia , 2010 ) was obtained from the World Reference Center for Emerging Viruses and Arboviruses ( WRCEVA ) . This strain was isolated from a pediatric case [17] . ZIKV was cultured using C6/36 Aedes albopictus mosquito cells as described previously [18] . Viral titers were determined by using baby hamster kidney ( BHK ) -21 cell-based focus forming assay ( FFA ) [19] . Eight-week-old female mice were infected retro-orbitally ( r . o . ) with 1 x 102 focus forming units ( FFU ) of ZIKV FSS13025 in 200 μl 10% FBS/PBS . African green monkey kidney-derived Vero E6 cells were purchased from ATCC . Vero cells were grown in Dulbecco's modified Eagle's medium ( DMEM , GIBCO ) supplemented with 1% HEPES , 1% penicillin/streptomycin ( GIBCO ) and 10% fetal bovine serum ( FBS , Gemini's BenchMark ) at 37°C , 5% CO2 . Mouse organs were collected in 800 μl RNA later ( Ambion ) and the tissues were transferred to 1% BMe/RLT buffer . Maternal mammary gland , brain , spleen , and the pup body minus the head and stomach were each placed in 800 μl . Before analysis , the skin of the head and rest of the body tissues were removed to avoid contamination from the mother’s saliva . SMC were removed from the pup’s stomach for separate analysis . The pups heads and stomachs tissues were processed in 250 μl ( stomach was washed twice with PBS in order to remove remnant milk ) . The tissues were next homogenized for 3 minutes using Tissuelyser II ( Qiagen Inc . ) and centrifuged for 1 minute at 6 , 000 g . Tissue samples , SMC , and serum from ZIKV-infected mice were extracted with the RNeasy Mini Kit ( tissues ) or Viral RNA Mini Kit ( serum and SMC ) ( Qiagen Inc . ) . Real-time qRT-PCR was performed using the qScript One-Step qRT-PCR Kit ( Quanta BioSciences ) and CFX96 TouchTM real-time PCR detection system ( Bio-Rad CFX Manager 3 . 1 ) . A published primer set was used to detect ZIKV RNA ( Lanciotti , 2008 ) . Fwd , 5’-TTGGTCATGATACTGCTGATTGC-3’; Rev , 5’-CCTTCCACAAAGTCCCTATTGC-3’ and Probe , 5’-6-FAM-CGGCATACAGCATCAGGTGCATAGGAG-Tamra-Q-3’ . Cycling conditions were as follows: 45°C for 15 min , 95°C for 15 min , followed by 50 cycles of 95°C for 15 sec and 60°C for 15 sec and a final extension of 72°C for 30 min . Viral RNA concentration was determined based on an internal standard curve composed of five 100-fold serial dilutions of an in vitro transcribed RNA based on ZIKV FSS13025 . The mammary gland was collected at 5 days post-infection ( dpi ) and was fixed in PFA for 24 hr at 4°C . ZIKV-infected tissues and mock-infected tissues were obtained . Tissues were processed and stained according to standard Visikol HISTO process ( protocol . visikol . com ) for antibody labeling . Tissues were immersed in Visikol Permeabilization Buffer at room temperature overnight . The following day , 2 mm thick tissue sections were processed through a series of washing steps of increasing methanol concentrations ( 50% , 80% , 100% ) , followed by permeation with 20% DMSO in methanol , and subsequently decreasing concentrations of methanol and back into PBS 1X . Tissues were then incubated in Visikol Penetration Buffer for 12 hr , washed with PBS , and incubated at 37°C in Visikol Blocking Buffer™ for 12 hr . Tissues were then transferred to microcentrifuge tubes for antibody labeling . The primary antibodies Smooth Muscle Actin ( αSMA ) ( Invitrogen; goat polyclonal ) and anti-ZIKV NS2B ( GeneTex; rabbit polyclonal ) were diluted at 1:100 in Visikol Antibody Buffer , and tissues were incubated at 37°C for 7 days . Tissue sections were washed in 1X Visikol Washing Buffer and then transferred to the secondary labeling solution . Secondary antibodies ( DyLight 488 conjugated anti-goat and Alexa 594 conjugated anti-rabbit IgG-Invitrogen ) were diluted at 1:200 in Antibody Buffer and the samples were incubated for another 3 days along with DAPI counterstain at 1:1000 dilution ) . Tissues were washed and cleared for imaging using ( LSCM ) . For clearing , both control and infected tissues were dehydrated with sequentially increasing concentrations of methanol ( i . e . 50% in PBS , 80% in H2O , 100% ) for 30 min in each step , followed by incubation in Visikol HISTO-1 for 12 hours , and then into Visikol HISTO-2 . Tissues were mounted in Visikol HISTO-2 and imaged using a Leica SP5 LSCM ( laser scanning confocal microscope ) using DAPI , Argon-488 , and 594 nm lasers with 10X and 20X magnification objectives . The mammary gland was collected at 5 , 7 , 9 and 11 dpi; SMC were collected at 5 dpi; and mock-infected samples were prepared . Tissues were fixed in zinc formalin for 24 hr at room temperature . Tissues were processed for paraffin embedding , and sections for slides were cut at 4 μm thickness . For histopathologic evaluation , slides were stained with hematoxylin and eosin . For IHC , slides were microwaved in Antigen Unmasking Solution ( Vector Laboratories ) , endogenous peroxidase activity was blocked by incubation in Bloxall ( Vector Laboratories ) , and nonspecific protein binding was blocked by incubation in 10% goat serum . Slides were then incubated in rabbit anti-ZIKV NS2B antibody ( Genetex ) diluted at 1:100 . ZIKV NS2B protein is a cofactor of the NS2B-NS3 protease which cleaves the viral polyprotein and is thus present during viral replication . Therefore , detection of NS2B serves as a marker of replicating virus as opposed to incomplete , phagocytosed , or degraded virions . Slides were then incubated sequentially in ImmPRESS HRP anti-rabbit IgG ( Vector Laboratories ) , and NovaRED HRP Substrate ( Vector Laboratories ) . IHC slides were also counterstained with hematoxylin . For each slide , the anti-ZIKV NS2B antibody staining was controlled with a slide using nonspecific Rabbit IgG ( Vector Laboratories ) substituted for the anti-ZIKV NS2B antibody , and control tissues from known infected and uninfected mice were included for each batch . A board-certified veterinary pathologist , who was blinded to each slide’s experimental conditions , read and scored each slide for immunoreactivity . Mammary gland slides were examined for mastitis by a pathologist . Bright field imaging was performed with a Zeiss Axio Scan . Z1 microscope and the images were acquired using Zen 2 software ( Carl Zeiss ) . SMC were frozen on dry ice and sent to the Texas A&M Veterinary Medical Diagnostic Laboratory for transmission electron microscopy . To detect viral NS2B protein expression , Vero E6 cells were grown to 70% confluency on glass coverslips . Cells were either mock-infected or inoculated with ZIKV FSS13025 at a MOI of 0 . 001 or with SMC supernatant . The SMC was collected from the pup’s stomach on d3 after birth from AG129 dams that had been previously infected retro-orbitally with 1 x 102 FFU of ZIKV FSS13025 . The SMC was collected at day 3 post infection because this time point was the peak RNA viral burden in the SMC . At day 5 after SMC treatment , Vero cells were fixed in 4°C methanol and permeabilized with 0 . 1% Triton X-100 . Protein blocking was performed with 10% goat serum , followed by incubation with anti-ZIKV NS2B antibody ( Genetex ) at 1:400 dilution . Coverslips were incubated with Alexa Fluor 594 ( Invitrogen ) at 1:300 dilution and then inverted onto glass slides for mounting . Imaging was performed by confocal microscopy . All data were analyzed with Prism software , version 7 . 0 ( GraphPad Software ) and expressed as means ± SEM . For viral burden and focus forming assay data , Krustal-Wallis test was used to compare more than two groups . This test was performed only in ZIKV-infected samples . Mock was not considered in the analysis . p<0 . 05 was considered a significant difference .
To begin evaluating whether ZIKV could infect the mammary gland and be transmitted to breastfed infants , 8-week-old female AG129 mice were infected with ZIKV strain FSS13025 . Viral burdens in several tissues were first assessed at 5 , 7 , 9 and 11 dpi via qRT-PCR . ZIKV RNA levels in the mammary glands were similar at the four time points ( Fig 1A ) . As expected , high levels of ZIKV RNA were present in the brain , spleen , and serum with no significant difference among the four time points . With the exception of serum , there was a slight reduction on 5 dpi compared with 11 dpi ( Fig 1B–1D ) , indicating ZIKV dissemination into tissues . To test for the presence of infectious virus in the mammary gland , we measured viral titers by FFA ( Fig 1E ) . High levels of infectious ZIKV were present in the mammary gland at 5 dpi with a slight reduction in the subsequent days analyzed , demonstrating that ZIKV establishes productive infection in the AG129 mouse mammary glands . To localize ZIKV replication within the mammary gland of AG129 mice , 8-week-old female AG129 mice were mock-infected or infected with ZIKV strain FSS13025 , followed by visualization of ZIKV infection via laser scanning confocal microscopy . After clearing with Visikol HISTO , 800–1000 μm thick portions of the mammary gland were imaged under laser scanning confocal microscopy . Immunofluoresence staining was performed to assess expression of ZIKV NS2B , a marker for viral replication [22] , and alpha smooth muscle actin ( αSMA ) , present in myoepithelial cells . At 5 dpi , strong expression of NS2B and αSMA was detected in the mammary gland of ZIKV-infected AG129 mice ( Fig 2A and 2B ) . The 3D images from this tissue ( S1 and S2 Figs ) and staining for ZIKV NS2B and DAPI ( Fig 2C ) showed similar results . Thus , ZIKV NS2B colocalizes with αSMA-expressing cells within the mammary glands of AG129 mice , suggesting myoepithelial cells as one of the cellular hosts of ZIKV in the mammary gland . To confirm ZIKV replication within the mammary gland of AG129 mice , tissues were fixed in Zinc formalin and then stained for expression of ZIKV NS2B at 5 , 7 , 9 and 11 dpi . No difference was observed among all times points , and we show 5 and 9 dpi as representative ( Fig 3 ) . NS2B expression was detected in cells morphologically consistent with mammary epithelial cells , myoepithelial cells , and interstitial macrophages ( Fig 3A ) . NS2B expression in cells in the stroma surrounding the teat canal on a nipple cross section ( Fig 3B ) and teat Langerhans cells was also observed ( Fig 3C ) . Additionally , histopathologic evaluation of these tissues revealed an absence of mastitis . Thus , ZIKV replicates locally in the mammary gland , and these enzyme IHC results in combination with z-projection images suggest that myoepithelial cells are major cellular hosts of ZIKV . Having established the presence of ZIKV RNA and infectious viral particles in the mammary gland , we proceeded to examine whether ZIKV was transmitted from infected mothers to neonates through breastfeeding . Neonatal heads , stomach tissues , SMC , and the rest of the bodies ( without skin to avoid contamination from the mother’s saliva ) were examined for the presence of ZIKV RNA by qRT-PCR . No ZIKV RNA was detected in the head and the rest of body in the neonates 1 to 7 days after birth ( Fig 4A and 4B ) . However , viral RNA was present in SMC and stomach tissues at almost all time-points tested from 1 to 7 days after birth ( Fig 4C and 4D ) . As ZIKV RNA does not necessarily indicate production of infectious virus , we next assessed for the presence of infectious ZIKV in SMC and stomach tissues via FFA . No infectious ZIKV particles were detected in SMC and stomach ( Fig 4E and 4F ) . Thus , breastfeeding does not appear to be a significant route of ZIKV transmission into neonates in this mouse model . To further assess the lack of infectious ZIKV in SMC , we inoculated SMC supernatant onto Vero cells . Infectivity of the SMC supernatant was assessed by immunofluorescence staining for ZIKV NS2B expression and CPE in the Vero cells , and plaque assay of the Vero culture supernatants . ZIKV NS2B expression and CPE were observed in the positive control cells infected with ZIKV . However , Vero cells inoculated with SMC supernatant did not show any NS2B protein expression or CPE ( Fig 5A and 5B ) , and plaque assay confirmed the absence of infectious virus in the culture supernatant of SMC supernatant-treated Vero cells ( Fig 5C ) . To assess whether ZIKV NS2B expression is observed in the breast milk were present in the SMC and might also infect the stomach tissue , 8-week-old female AG129 mice were infected with ZIKV strain FSS13025 , followed by sacrificing of pups on d5 after postpartum and examination for the expression of ZIKV NS2B on the pup SMC and stomach tissue by IHC . Of 10 sampled pup stomachs , ZIKV NS2B expression surrounded nuclear material in SMC from 3 pup stomachs ( Fig 5D ) . However , no ZIKV NS2B expression was detected in the full thickness of the gastric walls . These results suggest that replicating ZIKV may be passed in milk and is likely cell-associated; however , breast milk does not contain sufficient replication-competent ZIKV to initiate infection in cell culture and in IFN receptor-deficient mice . Finally , we determined whether there is infectious virus present in fresh breast milk by FFA . To increase the sensitivity of infection in these samples , we performed serial passages in Vero cells of fresh breast milk collected at 5 and 7 days postpartum . Only one fresh milk sample collected at d5 showed a low infectivity in the first passage . However , we observed an increase of infectious particles at the second and third passage in both times points ( Fig 5E ) .
In this study , we were able to detect ZIKV RNA in pup stomach milk clots and maternal mammary glands , and within the latter , ZIKV NS2B antigen localized to cells morphologically consistent with glandular epithelial cells , myoepithelial cells , and macrophages . ZIKV-permissive cells were also identified in the teat stroma and epidermis . Further , low levels of replicating virus were detected in fresh milk and ZIKV NS2B expression was detected in SMC samples . These results provide a framework for investigating ZIKV entry into the milk and raise the additional question of whether normal nursing-associated ingestion of maternal epidermal cells and blood may also play roles in ZIKV transmission . We propose that infectious ZIKV may enter human breast milk but may be subsequently inactivated by endogenous or exogenous factors such as lipid , antimicrobial proteins , or gastric acid . Several studies have shown that the acidic pH and the digestive enzymes present in the stomach inactivate virus [23–25] , and combine with mucus to form a chemical barrier to infection . Because dams were infected on the day of parturition , milk in this experiment should not have contained any ZIKV-neutralizing antibody . Recently in a rhesus macaques model , ZIKV RNA was present in saliva , another potential route of mucosal exposure [3] , but no infectious virus was detected . Another study demonstrated that human breast milk inactivates ZIKV after prolonged storage [26] . Additionally , human breast milk has been reported to reduce the infectivity of HIV , HCV , and dengue virus . Thus , antiviral properties of breast milk may reduce BMT [21 , 27] . Human viruses with known clinically relevant risk of BMT are cytomegalovirus ( CMV ) [28] and HIV-1 [27 , 29–31] . Although mastitis is a risk factor for BMT of both viruses , most cases of BMT occur in the absence of mastitis . Further , most cases of CMV and HIV BMT involve a seroconverted mother rather than infection of a naïve mother in the nursing period . Infectious CMV has been isolated from up to 80% of infected breast milk samples , whereas infectious HIV has been extremely difficult to isolate from breast milk . DNA and RNA from other human viruses including herpesviruses , parvovirus , rubella virus , arboviral flaviviruses , and hepatitis viruses A , B and C have been detected in milk [32] . However , perhaps owing to low clinical relevance of BMT of these viruses , it is largely unknown whether detected nucleic acids were non-infectious viral genetic material or derived from neutralized virions . After over two decades of research , the pathogenesis of HIV BMT remains poorly understood . It is estimated that BMT causes approximately 40% of mother-to-child transmission case of HIV . However , isolation of infectious virus from breast milk is rarely successful . HIV RNA , and rarely infectious virus , have been isolated from whey and cellular fractions of milk [33] . In contrast to CMV , viral loads in cellular fractions of milk correlate to transmission whereas loads in cell-free fractions do not . These findings have suggested that an intracellular location shields HIV from immune defenses such as lactoferrin , tenascin-C , defensins , and mucin [34] . Meanwhile , factors such as antibodies and HIV-gag-specific cytotoxic T lymphocytes may reduce cell-associated virus loads . Our early findings with ZIKV showed nursing mouse pups were not infected following ingestion of milk from infected dams . Therefore , the data are not sufficient to conclude that ZIKV infection can be passed via breastfeeding , and support early data suggesting the same for humans [8] . High ZIKV susceptibility of AG129 mice , which globally lack type I and type II IFN receptors , is often cited as a pitfall for many virology studies . However , in the current state of ZIKV science , in which it is unknown whether BMT is a clinical reality and there are no animal models of ZIKV entry into milk , a highly susceptible dam represents an excellent starting point to begin mechanistic manipulations which may reduce entry of viral RNA into milk . Furthermore , neonates , which are also deficient in IFN receptors , are a highly sensitive detection system for arranging conditions that may enable BMT . Indeed , the absence of infection in neonates in this study provides an early suggestion that infectious ZIKV is not easily transmitted through breast milk or other maternal-neonatal contact . It should also be noted that in humans , the tonsil is one of the first potential entry sites for orally ingested ZIKV [3] , whereas mice do not have tonsils . Because ZIKV is already known to have devastating consequences on the developing brain and there are both benefits to and substitutes for breastfeeding [35] it is imperative to fully understand the mechanisms which enable or prevent BMT . The results of this study provide a mouse model for investigating entry of ZIKV RNA into breast milk , and the pups provide a sensitive system for testing modulations which might permit BMT .
|
Can Zika virus be transmitted from nursing mothers to their children via breast milk ? Only 4 years have passed since the Zika virus outbreak in Brazil , and much remains to be understood about the transmission and health consequences of Zika infection . To date , some case reports have detected Zika virus RNA in the breast milk of infected mothers , but the presence of a virus’ RNA does not mean that intact virus is present . Milk also contains many natural defense components against infection , so even intact virus carried in breast milk may not be infectious to a child . Here we used a mouse that is genetically engineered to be highly susceptible to Zika infection , and tested whether 1 ) we could find intact virus in mouse breast milk and 2 ) infection was passed from mother to pups . We found very low levels of intact Zika virus in mouse breast milk , and found none of the nursing pups to be infected . The model of Zika virus breast milk infection developed in this study establishes a system by which we may learn whether Zika RNA in human breast milk is truly infectious to children , and how Zika virus may enter the milk .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"vero",
"cells",
"medicine",
"and",
"health",
"sciences",
"milk",
"breast",
"milk",
"body",
"fluids",
"reproductive",
"system",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"biological",
"cultures",
"microbiology",
"diet",
"viruses",
"animal",
"models",
"model",
"organisms",
"nutrition",
"rna",
"viruses",
"gastroenterology",
"and",
"hepatology",
"experimental",
"organism",
"systems",
"stomach",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"animal",
"studies",
"medical",
"microbiology",
"microbial",
"pathogens",
"exocrine",
"glands",
"mouse",
"models",
"gastrointestinal",
"infections",
"cell",
"lines",
"breast",
"tissue",
"gastrointestinal",
"tract",
"anatomy",
"flaviviruses",
"beverages",
"physiology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"mammary",
"glands",
"organisms",
"zika",
"virus"
] |
2019
|
Detection of Zika virus in mouse mammary gland and breast milk
|
Amphimerus sp . is a liver fluke which recently has been shown to have a high prevalence of infection among an indigenous group , Chachi , who reside in a tropical rainforest in the northwestern region of Ecuador . Since it is unknown which animals can act as a reservoir and/or definitive hosts for Amphimerus sp . in this endemic area , a study was done to determine the prevalence of infection in domestic cats and dogs . This information is important to understand the epidemiology , life cycle and control of this parasite . In July 2012 , three Chachi communities located on Rio Cayapas , province of Esmeraldas , were surveyed . A total of 89 of the 109 registered households participated in the study . Of the 27 cats and 43 dogs found residing in the communities , stool samples were collected from 14 cats and 31 dogs ( total of 45 animals ) and examined microscopically for the presence of Amphimerus eggs . The prevalence of infection was 71 . 4% in cats and 38 . 7% in dogs , with similar rates of infection in all three communities . Significantly more cats were infected than dogs ( p = 0 . 042 ) . The data show a high rate of Amphimerus sp . infection in domestic cats and dogs residing in Chachi communities . It can be concluded that these animals act as definitive and reservoir hosts for this liver fluke and that amphimeriasis is a zoonotic disease . These findings provide important epidemiological data which will aid in the development and implementation of control strategies against the transmission of Amphimerus .
Amphimerus Barker , 1911 is a genus of parasitic liver fluke which are flat helminths ( Platyhelminthes ) of the Trematoda class and belong to the Opisthorchiidae family . The adult flukes reside within the bile ducts of a definitive host [1] . Infection by liver flukes of this family , which include Clonorchis sinensis and Opisthorchis spp . can occur through the consumption of raw or undercooked , metacercariae infected freshwater fish [1–5] . Liver fluke infection is one of the more important food-borne diseases worldwide and is considered by the World Health Organization as a neglected tropical disease [6] . Affected individuals with liver flukes of the Opisthorchiidae family can suffer from suppurative cholangitis , cholelithiasis and cholangiocarcinoma [3 , 5 , 6] . In the Americas , ten species of Amphimerus which infect mammals have been described: A . pseudofelineus , A . pseudofelineus minutus , A . caudalitestis , A . price , A . lancea , A . parciovatus , A . bragai , A . minimus , A . neotropicalis and A . ruparupa [7 , 8] . In Ecuador , flukes found in the bile ducts of dogs were previously described as Opisthorchis guayaquilensis [9] but later , this species , without any further comparative studies was named as being Amphimerus guayaquilensis . However , Artigas and Perez ( 1964 ) considered to A . guayaquilensis to be synonym of A . pseudofelineus . A few years later , A . guayaquilensis was considered to be distinct from A . pseudofelineus and was instead regarded as a synonym of A . parciovatus [7] . Furthermore , Thatcher ( 1970 ) did not agree with this synonymy and contemplated A . guayaquilensis distinct from A . pseudofelineus because of the extent of the vitellaria . The validity of some species in this genus is controversial , since speciation is based only on morphological and morphometric features present in the adult flukes and the assignment of species names must be regarded as speculative . Amphimerus spp . have been demonstrated to infect birds , reptiles and certain mammals including cats , dogs , ducks , the double-crested cormorant , Amazonian dolphins , opossums ( Didelphis marsupialis , Philander opossum ) and the rodent Nectomys squamipes [1 , 9–16] . For other members of the Opisthorchiidae family known to infect humans e . g . C . sinensis and Opisthorchis spp . , cats and dogs are the most important animal reservoirs for human infection [2 , 3 , 17 , 18] . However , it is currently unknown whether domestic animals may act as a definitive and/or reservoir host for human transmission in the recently reported focus of infection in Ecuador [10] . Given the similarities between C . sinensis , Opisthorchis spp . and Amphimerus spp . , it was hypothesized that these mammals may also act as reservoirs for Amphimerus sp . infection in Ecuador . The indigenous group , Chachi , who live along the Rio Cayapas and its tributaries in the north-western coastal rainforest of Ecuador , have been shown to have a high prevalence of infection ( 15 . 5% to 34 . 1% ) with Amphimerus sp . [10] . Afro-Ecuadorian and mestizo populations living in separate communities but along the same rivers were not found to be infected [10] . The solo infection of the Chachi population was postulated to be related to their cultural tradition of eating smoked fish and food sharing [10 , 11] . In a previous pilot study , conducted by the authors , in the same endemic communities for human infection , both cats and dogs were found to be positive for Amphimerus eggs . There was no evidence of infection in any other animals investigated ( e . g . pigs and chickens ) . The objective of this study was , therefore , to investigate the prevalence of infection of Amphimerus sp . in domestic cats and dogs and to determine their role in the transmission in the Ecuadorian villages endemic for human infection .
The study was conducted in three indigenous Chachi communities along the Rio Cayapas in the Province of Esmeraldas , located in the northwest coastal rainforest of Ecuador ( Fig . 1 ) . The indigenous Chachi , living alongside Afro-ecuadorian and mestizo populations , is the predominant ethnic group in this area , and represent 13% of the inhabitants in this region of Ecuador [19 , 20] . These communities are the same as those studied previously , showing the resident Chachi having a high prevalence of infection with Amphimerus sp . [10] . They live in remote villages where the only way to reach the communities are by boat along the river; the source of their water is mainly from rivers and streams and is consumed untreated . Sanitation facilities are lacking , and flush toilets are uncommon . The people are hunters and eat fish caught in the neighboring rivers almost every day and the meal is accompanied with cooked rice and plantain . The province of Esmeraldas , forms part of the tropical rainforest known as “Choco Biogeográfico del Pacífico” which covers a section of the coast of Ecuador , Colombia and Panamá . This area has been labelled as a biological hotspot; an area with an extraordinary concentration of animal species [21] . The climate of this region is warm and humid , with an average temperature of between 24°C and 28°C and an average relative humidity of 85% [22] . This study was based on a previous census conducted in January 2012 where each household was given an identification number and a total of 109 households were recorded in the three communities . In July 2012 , all house owners were asked to participate in the study by providing a stool sample from any cats and/or dogs residing in the respective household , simultaneously a census of dogs and cats residing in the participating communities was conducted . A team of community health workers informed the villagers of the study in their local Chapalache language and residents were free to refuse entry to their household or access to their domestic animals at any point during data collection . Stool samples were collected by their owners from each animal directly after the deposit was made . Plastic flasks containing stool samples were labelled with type of animal , house code and date . The samples were preserved in 10% formalin and transported to the parasitology laboratory at Centro de Biomedicina in Quito where they were stored at 4°C until they were examined . Samples were concentrated using the formalin-ether technique as previously described [10] and were examined under light microscopy by at least two laboratory technicians for the presence of Amphimerus eggs . Samples were then verified by an expert in animal stool examination who was not involved in the data collection . The primary outcome variable was Amphimerus infection , defined as positive if eggs of the parasite were visualized by light microscopy . The yellow-brown eggs measured 20 to 32 um in length and 14 to 16 um in width; they are pyriform with a visible operculum at the narrower anterior end . In the centre of the posterior end there is a small bud ( Fig . 2 ) . The adult parasites were obtained from the livers of two cats and one dog . Animals were presented by their owners as sick and showed severe emaciation . They were euthanized with ether and necropsied . The livers were collected in saline solution , squeezed and sliced in small pieces and maintained for 30 minutes . Flukes were then removed from the saline and were fixed in both 70% ethanol and 10% formalin . For molecular analysis , genomic DNA samples were extracted from each of the Amphimerus adult specimens from the cats and dogs using a DNeasy Blood & Tissue Kit ( QIAGEN K . K . , Tokyo , Japan ) . The ITS2 region of the ribosomal DNA was then amplified by PCR using Ex Taq DNA polymerase ( Takara Bio , Shiga , Japan ) . The primers used were 3S ( forward , 5′-GGTACCGGTGGATCACTCGGCTCGTG-3′ ) [23] and A28 ( reverse , 5′-GGGATCCTGGTTAGTTTCTTTTCCTCCGC-3′ ) [24] . DNA sequencing of amplicons was performed with a 3100-Advant Genetic Analyzer ( Life Technologies , Foster City , CA , USA ) . Data was analysed using SPSS version 19 ( Statistical Product and Service Solutions , Chicago , IL , USA ) . The data was stratified by village and animal species , and prevalence of Amphimerus infection in the animals calculated for each village . A chi squared test was used to detect any significant differences in the prevalence of infection between animal species and between villages . Ethical approval of the study was given by ethic committee of the Universidad Central del Ecuador ( licence number LEC IORG 0001932 , FWA 2482 , IRB 2483 . COBI-AMPHI-0064–11 ) . All villagers were asked for their verbal consent to access their domestic animals and collect stool samples . Infected animals with any parasite were treated with specific drugs free of charge in the following months . The study was conducted according to the above institution’s guidelines for animal welfare .
Of the 109 houses recorded in the communities , 89 ( 81 . 6% ) agreed to take part in the survey ( Table 1 ) . From these 89 houses , a total of 27 cats and 43 dogs were counted at the time of data collection . Stool samples from 45/70 animals ( 64 . 2% ) , 14 from cats and 31 from dogs were collected and examined microscopically for the presence of Amphimerus eggs ( Table 1 ) . Samples of the remaining 25 animals were not able to be collected because they either did not defecate on the collection day or were not present in the home at the time . In cats the prevalence of Amphimerus infection was 71 . 4% ( 95% confidence interval [CI] = 47 . 7–95 . 1 ) , and in dogs , 38 . 7% ( 95% CI = 21 . 6–55 . 8 ) ( Table 1 ) . The overall prevalence of Amphimerus sp . in the two animal species investigated was 48 . 9% ( 95% CI = 34 . 3–63 . 5 ) . Cats were approximately four times more likely to be infected with Amphimerus sp . than dogs , OR = 3 . 95 ( 95% CI 1 . 01–15 . 6 , p = 0 . 042 ) . There was no statistically significant difference between the communities with regard to prevalence of Amphimerus infection in the animals studied . The eggs found in the stools of cats and dogs demonstrated the morphological characteristics consistent with other members of the Opisthorchiidae family ( Fig . 2 ) . In order to confirm that the eggs were from Amphimerus sp . , two positive cats and one positive dog were euthanized and adult flukes were obtained from the bile ducts and subjected to morphological and molecular characterization . The recovered flukes just after extraction from the bile ducts were flat , leaf-like , reddish , flexible and elongated with active movements in saline , with a thinner anterior than posterior extremity , measuring from 15 to 28 mm in length by 2 to 4 mm in width . When fixed in formalin 10% , the flukes became whitish and shorter measuring around 10 to 18 mm long . Flukes were stained with borax carmine and photographed ( Fig . 3 ) . Adults of Amphimerus sp . can be differentiated from Clonorchis sinensis and Opisthorchis spp . by certain morphological features e . g . : 1 ) The presence of two rounded testes , which lie one behind the other in the posterior portion , 2 ) The vitellaria occupy both lateral sides of the fluke , outside of the intestinal branches and are conspicuously distributed in four groups , 2 anterior and 2 posterior extending backwards nearly to the excretory pore , and 3 ) the ventral sucker is larger than the oral [1 , 10] . Furthermore , sequences of the ribosomal DNA ITS2 region of the flukes obtained from the cats and dog were identical to that obtained from humans in the previous study [10] . Accession numbers , deposited in the GenBank/EMBL/DDBJ nucleotide database , are AB678442 , AB926429 and AB926430 for Amphimerus sp . from humans , cats and dogs , respectively .
This study is of importance in showing that the liver fluke Amphimerus sp . can infect and is common in cats and dogs living in Chachi communities of Ecuador , where human amphimeriasis is prevalent . The key finding of the study is that cats and dogs serve as definitive hosts and represent reservoirs for human infection . It can therefore be concluded that amphimeriasis is a zoonotic disease . These results provide relevant data that could be used for policy makers for conducting effective control strategies and measures against Amphimerus infection . As this study found a high prevalence of infection in cats and dogs , the recommended public health measure to control transmission would be to treat these domestic animals , as well as humans , with a specific drug for flukes such as praziquantel .
|
Amphimerus sp . is a fluke that infects the bile ducts of its definitive hosts . Recently , it has been shown that an indigenous Amerindian group , the Chachi , living in a rural and remote tropical area of Ecuador , are infected with this parasite . The epidemiology and life cycle of this parasite remains elusive , and research is needed to understand the mode of transmission and zoonotic potential of the parasite . It was hypothesized that the domestic animals of the Chachi households may act as definitive and reservoir hosts for Amphimerus infection . Hence , the presence and prevalence of infection in these animals residing in communities endemic for human amphimeriasis was investigated . Some 45 animal stool samples were examined microscopically for the presence of Amphimerus eggs . The results showed an infection rate of 71 . 4% in cats and 38 . 7% in dogs . The data provided evidence that these domestic animals act as both definitive and reservoir hosts for the parasite and that amphimeriasis is a zoonotic disease . The implementation of a mass treatment/control program must target both humans and animals in order to minimize the transmission of this liver fluke .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
High Prevalence of the Liver Fluke Amphimerus sp. in Domestic Cats and Dogs in an Area for Human Amphimeriasis in Ecuador
|
The Roseobacter clade is a ubiquitous group of marine α-proteobacteria . To gain insight into the versatile metabolism of this clade , we took a constraint-based approach and created a genome-scale metabolic model ( iDsh827 ) of Dinoroseobacter shibae DFL12T . Our model is the first accounting for the energy demand of motility , the light-driven ATP generation and experimentally determined specific biomass composition . To cover a large variety of environmental conditions , as well as plasmid and single gene knock-out mutants , we simulated 391 , 560 different physiological states using flux balance analysis . We analyzed our results with regard to energy metabolism , validated them experimentally , and revealed a pronounced metabolic response to the availability of light . Furthermore , we introduced the energy demand of motility as an important parameter in genome-scale metabolic models . The results of our simulations also gave insight into the changing usage of the two degradation routes for dimethylsulfoniopropionate , an abundant compound in the ocean . A side product of dimethylsulfoniopropionate degradation is dimethyl sulfide , which seeds cloud formation and thus enhances the reflection of sunlight . By our exhaustive simulations , we were able to identify single-gene knock-out mutants , which show an increased production of dimethyl sulfide . In addition to the single-gene knock-out simulations we studied the effect of plasmid loss on the metabolism . Moreover , we explored the possible use of a functioning phosphofructokinase for D . shibae .
The Roseobacter clade is a versatile group of Gram-negative α-proteobacteria , which can be found in all oceans worldwide . Especially during phytoplankton blooms they account for a large fraction of the marine bacterial community [1] , [2] . Here , we focus on the aerobic anoxygenic phototroph Dinoroseobacter shibae DFL12T [3] . Although the bacterium has been isolated from the surface of the dinoflagellate Prorocentrum lima , it can also be motile by the means of a single polar flagellum . It harbors five plasmids and needs additional vitamins ( biotin , nicotinate , and 4-aminobenzoate ) to grow in minimal seawater medium . To obtain energy , D . shibae can use oxygen , nitrate or dimethyl sulfoxide ( DMSO ) as terminal electron acceptor . Additionally , energy generation is possible via light dependent aerobic anoxygenic photosynthesis [3] , [4] . Despite of their common taxonomic classification , many members of the Roseobacter clade have adopted a unique life style and accordingly have developed a tailored metabolism [1] , [5] . For instance , D . shibae is believed to live in symbiosis with its host and to provide the algae with vitamin B12 in exchange for carbon sources originating from photosynthesis [6] , [4] . Some members of the Roseobacter clade produce storage compounds belonging to the group of polyhydroxyalkanoates , which are biopolymers with a potential industrial use [7] , [8] . Under optimal conditions , Dinoroseobacter sp . JL1447 , a close relative of D . shibae , has been found to produce large quantities of polyhydroxyalkanoates [9] . Moreover , D . shibae and other Roseobacters produce dimethyl sulfide ( DMS ) during DMSO respiration and if dimethylsulfoniopropionate ( DMSP ) is used as carbon source . This molecule contributes to the characteristic odor of the ocean and affects climate by seeding cloud formation [10] . All these properties demonstrate that D . shibae notably differs from well-studied organisms . For the aforementioned reasons , the Roseobacter clade and its members were subjects of intensive research in the past few years [1] , [2] . Since the initial description of D . shibae in 2005 , further studies on the genome sequence and transcriptome analyses under changing illumination conditions have been published [3] , [4] , [11] . Furthermore , important parts of the metabolism of D . shibae were elucidated by 13C labeling experiments , a study on DMSP catabolism , and a study targeting energy conservation [12] , [13] , [14] . Remarkably , no phosphofructokinase activity has been observed in D . shibae during growth on glucose [12] . Recently , a basic metabolic model of Rhodobacter sphaeroides , another member of the Roseobacter clade , has been created [15] . However , no systematic and detailed computational analysis of the metabolism of any member of this ubiquitous group of marine bacteria has been carried out to date . In this work , a large-scale computational analysis of the metabolism of the marine bacterium Dinoroseobacter shibae DFL12T is presented . Prior to the analysis , we created an elaborate genome-scale metabolic model [16] , [17] of D . shibae , denoted iDsh827 . It has been validated against experimental data and covers 827 open reading frames . Moreover , iDsh827 is the first genome-scale metabolic model which explicitly takes the energy demand of bacterial motility into account . Additionally , our model is the first one which uses aerobic anoxygenic photosynthesis . In total , 391 , 560 distinct simulations featuring a large variety of different growth conditions , e . g . varying carbon and nitrogen sources , aerobic and anaerobic conditions , and the availability of light have been carried out . A large fraction of the simulations is dedicated to plasmid and single gene knock-out mutants . In detail , the aim of this work is to
A fundamental assumption made in most constraint-based simulations is that the organism tries to maximize its growth rate as much as possible under the given environmental conditions . Therefore , the model iDsh827 contains a biomass reaction , which consumes appropriate quantities of 113 different metabolites needed for the reproduction of D . shibae . The flux through this reaction corresponds to the growth rate and is the objective function of all simulations presented here . Hence , to obtain precise adjustments , the contribution of each metabolite to the biomass composition was either quantified experimentally for D . shibae or estimated based on literature values of related organisms ( Table 1 ) . For a detailed description of the experimental procedures used to determine the different fractions see the materials and methods section . Literature values originating from Roseobacter denitrificans were used to estimate the amount of lipopolysaccharides produced by D . shibae . Furthermore , the fraction of peptidoglycan was presumed to be approximately equal to the values determined for Escherichia coli [18] . Moreover , compounds whose proportion of the biomass is still unknown were estimated to make up 0 . 1% of the total dry weight all together . Finally , the soluble pool was assumed to account for the remaining biomass fraction . Since some biomass compounds could not be separated by the analytical procedures used during the laboratory experiments , we corrected the measured values for these fractions . As the original lipid fraction also contained the lipopolysaccharide and the bacteriochlorophyll α fractions , we subtracted these values to obtain the actual lipid content . Moreover , we corrected the original protein fraction by subtracting the peptidoglycan value . The relative portions of the nucleotides and amino acids were estimated from the genome sequence as suggested by the reconstruction protocol [19] . The predominant respiratory lipoquinone ( ubiquinone-10 ) and the predominant cellular fatty acid ( 18:1ω7c ) were chosen to represent their corresponding compound group [3] . Furthermore , the ratio of Kdo2-lipid A to the O-antigen in the lipopolysaccharide was calculated from values measured for Roseobacter denitrificans [20] . Under anaerobic conditions , the oxygen atoms in the singlet oxygen quencher spheroidenone neither originate from water nor from CO2 but probably from other cell components [21] . Hence , we included its precursor methoxyneurosporene in the biomass reaction . A detailed measurement of the content of the soluble pool is only available for E . coli [22] . Nevertheless , we used these values to constitute the soluble pool content in iDsh827 . Therefore , we removed compounds which were either specific for the growth conditions used in the study or which were not part of our model . As the genome annotation did not give any indications for spermidine production in D . shibae , we did not include this compound into the biomass reaction . Due to the fact that aerobic anoxygenic phototrophs produce bacteriochlorophyll α exclusively in the dark under aerobic conditions [23] , we included two biomass reactions in the model: one with and one without bacteriochlorophyll α . The former reaction contains a term corresponding to 4 nmol/mgprotein as determined for D . shibae [3] . Before running a simulation , the appropriate biomass reaction is enabled automatically based on the preset conditions . Both biomass reactions are normalized to one gram dry weight . To elucidate the usability of carbon sources by D . shibae , we conducted Phenotype MicroArray experiments . Respiration occurred on the carbonic acids succinate , fumarate , 2-oxoglutarate , L-lactate , acetate , propionate , ( R ) -3-hydroxybutanoate , glycolate , glyoxylate , pyruvate , the carbohydrates α-D-glucose , D-fructose , L-rhamnose , maltose , D-xylose and on the sugar alcohols D-ribitol , myo-inositol , D-arabitol , and D-xylitol . Out of 190 different carbon sources tested , D . shibae was able to utilize only 19 , which confirms the poor variety of potential nutrients used by the organism [3] . Most of the mentioned carbon sources were applied for growth experiments in batch cultures and gave a positive biomass yield . The maximal growth rate on the reference substrate succinate was 0 . 25 h−1 , on glyoxylate 0 . 15 h−1 . As not all cells in a bacterial culture display the same degree of motility [24] , we tracked the movement of D . shibae cells grown on glucose and succinate experimentally to determine the distribution of swimming velocities . The resulting frequency distributions , shown in Figure 1 , slightly differ from each other . The average velocity of cells is 5% higher ( 1 . 68 µm/s ) on glucose than on succinate ( 1 . 59 µm/s ) . However , the shapes of the distributions differ significantly as a two-sample Kolmogorov-Smirnov test yielded a p-value of 1 . 06×10−5 . The reconstructed metabolic network of D . shibae gives insight into the metabolism of this representative of the cosmopolitan Roseobacter clade . The metabolic model iDsh827 consists of 1488 reactions covering 827 open reading frames ( Table 2 ) and is based upon an aggregated genome annotation provided by the EnzymeDetector database [25] . To fill gaps in important pathways , we included a few enzymes with low evidence scores into the model . A graphical representation of the distribution of evidence scores can be seen in Supplementary Figure S1 and the complete list of all genes covered in iDsh827 is given in Supplementary Dataset S1 . Moreover , 199 genes retrieved from the annotation database were not taken into consideration either because their gene product was a non-metabolic enzyme or because the annotation was ambiguous . To reproduce growth on minimal medium as accurately as possible , multiple refinements guided by experimental results found in the literature were made . Important refinements are highlighted in the next sections . The final metabolic model in the SBML format [26] can be found in Supplementary Dataset S2 and in the BioModels database ( accession ID: MODEL1308180000 ) . In total , we carried out 391 , 560 simulations to study the metabolic network of D . shibae under various environmental conditions ( Table 4 ) and the effect of genetic perturbations in terms of single gene and plasmid knock-outs . Figure 2A gives a rough summary of all simulations . The leftmost bar corresponds to physiological states with no or only very little growth . Furthermore , the modes visible in the figure are mainly caused by the different carbon uptake rates and other environmental conditions . As visualized in Figure 2B , some simulations of distinct states yielded exactly the same flux distribution . This happens if the changed parameter has no effect on the metabolism . For instance , an additional nutrient may remain unused or the function of a knocked-out gene can be carried out by another one . Hence , we observed only 55 , 390 distinct flux distributions in our simulations ( Figure 2B ) . Prominently , the most common distribution is the one where no fluxes are active at all and hence no growth occurs . This is often due to insufficient carbon uptake or a lethal knock-out . In contrast , more than 23 , 000 flux distributions are unique for only one physiological state . In the following sections , we will take a closer look at the fluxes in some selected simulations .
Although the precise connection between swimming velocity and the corresponding energy demand depends on many unknown variables ( e . g . viscosity of the medium and efficiency of the motor protein ) , it is very likely that the energy demand increases with velocity . As this demand has a great influence on the energy metabolism and hence on growth , this parameter is worth to be considered in genome-scale metabolic models . We suggest to evaluate values within physiological reasonable ranges in multiple simulations to account for the different swimming velocities occurring within a culture and under different environmental conditions . Our simulations showed that the phenotype of D . shibae growing on glycerol or glyoxylate can only be reproduced if these compounds are allowed to pass the membrane without active transport . Otherwise , the total energy requirement of the cell , including the energy needed for the active transport , cannot be covered by the degradation of these compounds . Hence , we suppose that other transporters requiring less energy exist in D . shibae . Indeed , the existence of a glycerol-conducting channel in E . coli is known [43] . Glycerate is already in a highly oxidized state so that the oxidation to CO2 does not yield enough energy to sustain growth if the import is ATP-dependent . In most physiological states shown in Figure 3 , the TCA cycle operates in forward direction and thus is used to obtain energy . However , in the simulations with glycolate as sole carbon source the isocitrate lyase converts glyoxylate to D-threo-isocitrate . At this point the flux splits: 80% follow the TCA cycle in forward direction and 20% are routed to citrate . Hence , the citrate synthase runs backwards in the glycolate simulations . According to our results , the activity of the TCA cycle and the oxidative phosphorylation in general is significantly reduced in light . This is supported by our simulations , the experimentally observed decrease of the specific carbon dioxide production rates , and metabolome analyses during light shift experiments in continuous cultivations of D . shibae . Hence , we conclude that the aerobic anoxygenic photosynthesis satisfies the energy demand of the organism in large parts . Moreover , under illuminated conditions the production of oxaloacetate is increased to supply anabolic reactions . Another effect of the reduced TCA cycle activity is a decreased production of NADPH by the isocitrate dehydrogenase . Hence , the organism must increase the flux through other reactions for compensation under illuminated conditions . For instance , our simulations showed an increased usage of the pentose phosphate pathway in comparison to dark conditions if glucose is used as carbon source . In general , a flexible energy metabolism is probably very important for D . shibae because the amount of energy generated by the aerobic anoxygenic photosynthesis can vary greatly . This can be due to changing illumination conditions but also due to the degradation of the light harvesting complex in the course of time [44] . Recent experimental results suggest , that the activity of the aerobic anoxygenic photosynthesis is reduced or even stopped under anaerobic conditions [45] . If this holds for D . shibae and all environmental conditions , the illuminated anaerobic simulations would coincident with the dark anaerobic simulations . As we pointed out in the results section , the anaerobic glycolate simulation in light is the only one displaying variance . This is probably due to an energy overflow , which occurs if more energy ( external protons ) is generated by the aerobic anoxygenic photosynthesis than needed for growth . Thus , energy-wasting futile reactions can take place in any part of the metabolic network . Futile reactions waste energy and produce heat and thus are usually disadvantageous for the organism . Hence , they are probably tightly regulated . However , the uncertainty stems from the fact that the simulations cannot predict which set of futile reactions take place . A possibility we did not test in our simulations is a leakage of protons through the membrane . Another uncertainty in energy balance is the precise number of periplasmic protons used for motility . This number may vary greatly depending on environmental conditions and the average velocity of the motile bacteria [38] . Furthermore , our simulations confirmed that formation of phosphoenolpyruvate from pyruvate via pyruvate orthophosphate dikinase neither needs to be active during the glucose simulations [12] nor in any other physiological state studied here . However , we found that the reverse reaction is active in all physiological states . According to our results , light and the presence of oxygen stimulate the usage of the DMSP demethylation pathway . Interestingly , the effect depends on the carbon uptake rate . While oxygen always enhances the activity of the demethylation pathway , the effect of light is much more pronounced in physiological states with a medium uptake rate . On average , the cleavage pathway is used to a greater extend but the fraction of demethylation decreases as the uptake rate increases . This is consistent with measurements made using fifteen different Roseobacter strains [46] . Although DMSP may have become an abundant compound only recently [10] , D . shibae and other Roseobacters seem to be well-suited to use it efficiently under various conditions . Whether the regulation of these pathways is really as sophisticated as predicted by our model remains an open question . Our results support the hypothesis proposed in a recent review stating that under certain conditions , nitrate respiration is not only being used for energy generation but has some other function , which may be redox balancing [47] . As described above , the energy demand of the organism is more than covered by the aerobic anoxygenic photosynthesis under low nutrient conditions . Nevertheless , our simulations showed that a terminal electron acceptor is still required for optimal growth to be retained in the light . This is due to an excess of reducing equivalents produced by different metabolic processes . These reducing equivalents need to be reoxidized to retain redox homeostasis . Again , flux balance analysis is not able to exactly determine which terminal electron acceptor is preferably used to achieve this goal . However , we used flux variability analysis to identify alternate optimal solutions . Indeed , some solutions involve denitrification under aerobic illuminated conditions . Intriguingly , this behavior has been reported for Roseobacter denitrificans [48] . We speculate that aerobic denitrification might be the preferred way to dispose electrons in vivo because it does not waste carbon and produces multiple oxidized redox equivalents . D . shibae is very likely adapted to an oligotrophic marine environment and a regular day/night cycle [49] . Hence , the organism may benefit only moderately from phosphofructokinase activity . This is especially true during the day when the energy demand is covered by the aerobic anoxygenic photosynthesis in large parts . On the one hand , this suggests a permanent shutdown of the phosphofructokinase . On the other hand , another possible explanation is that light induces a down regulation of the phosphofructokinase in D . shibae because the energy generated in lower glycolysis is not needed . This would also explain the inactivity of the enzyme observed by Fürch et al . as their cultures were grown in constant light [12] . In general , it can be informative to simulate a great variety of physiological states for each knock-out mutant . Since some effects do not occur in all physiological states , more subtle differences between the mutants can be revealed this way . This would not be possible with the commonly used method , which relies on one or two minimal media states [50] , [33] . The simulated loss of plasmids brought up two interesting aspects regarding the 153 kb and the 86 kb plasmid . The 153 kb plasmid harbors the only copy of a catalase ( Dshi_3801 ) , which decomposes hydrogen peroxide to oxygen and water . The reason for the decreased growth is that the hydrogen peroxide created during the production of pyridoxal 5′-phosphate and the operation of the glycine oxidase now have to be degraded by a cytochrome-c peroxidase . The reaction catalyzed by this enzyme depends on reducing equivalents and hence is metabolically more expensive . Albeit hydrogen peroxide is created by other metabolic reactions in our simulations , singlet oxygen is also created during aerobic anoxygenic photosynthesis , which imposes an elevated oxidative stress level on the organism [51] . Hence , the catalase gene might be a beneficial acquisition for D . shibae . Furthermore , this hypothesis is supported by the fact that the catalase gene is upregulated in response to light [11] . The 86 kb plasmid contains the complete synthesis pathway leading to dTDP-α-L-rhamnose , which is an important compound of the outer membrane . Hence , our simulations predicted no growth in case of a loss of this plasmid for an unmodified carbolipid content . However , pseudogenes and transposases located on the plasmid may indicate it is no longer needed by the organism [4] . An explanation might be that D . shibae is able to change its surface structure or biofilm formation capabilities . For the first time , we demonstrated that in theory a single gene knock-out is sufficient to significantly enhance the production of the cloud-seeding molecule DMS . In case of the mutants with a blocked demethylation pathway , the DMSP must be degraded inevitable by the cleavage pathway to retain growth . Hence , the more than 60% increase of the DMS production can be explained by the fact that 39 . 0% of the DMSP degraded by the demethylation pathway ( Figure 6 ) is now redirected to the DMS-producing cleavage pathway . A similar effect occurred when the gene coding for the L-serine ammonia-lyase was removed . Although the demethylation pathway was kept intact , its activity was greatly reduced . The reason is that the carbon atom bound to tetrahydrofolate during the demethylation of DMSP , can no longer be routed to the central carbon metabolism . This is due to the fact that the glycine hydroxymethyltransferase binds this carbon atom to glycine producing L-serine whose degradation to pyruvate and ammonia is now inoperative . Alternatively , a salvage via the methionine synthase would be possible but we found no methionine degradation pathway in D . shibae . As the growth on the other carbon sources is not affected , the mutants could be grown on succinate for example and transferred to DMSP for DMS production afterwards .
The D . shibae DFL12T wild-type strain [3] was inoculated in complex medium ( 40 g l−1 Marine Bouillon medium , Carl Roth , Karlsruhe , Germany ) and incubated in darkness at 30°C and 150 rpm for 48 h before being diluted 1∶50 in freshly prepared defined artificial sea water medium ( SWM ) [12] containing the following components per liter of medium: 4 . 0 g NaSO4 , 0 . 2 g KH2PO4 , 0 . 25 g NH4Cl , 20 . 0 g NaCl , 3 . 0 g MgCl2°6 H2O , 0 . 5 g KCl and 0 . 15 g CaCl2·2 H2O , 0 . 19 g NaHCO3 , 1 ml trace element solution ( 2 . 1 g Fe ( SO4 ) ⋅7 H2O , 13 ml 25% ( v/v ) HCl , 5 . 2 g Na2EDTA⋅2 H2O , 30 mg H3BO3 , 0 . 1 g MnCl2⋅4 H2O , 0 . 19 g CoCl2⋅6 H2O , 2 mg CuCl2⋅2 H2O , 0 . 144 g ZnSO4⋅7 H2O and 36 mg Na2MoO4⋅2 H2O per liter ) and 10 ml vitamin solution ( 0 . 2 g biotin , 2 . 0 g nicotinic acid and 0 . 8 g 4-aminobenzoic acid per liter ) . Succinate ( 2 to 4 g l−1 ) and glucose ( 3 . 6 g l−1 ) were used as sole carbon sources . Main cultures were inoculated from these overnight cultures in fresh artificial sea water medium ( SWM ) with an OD600 of 0 . 05 . Main cultures of D . shibae DFL12T were cultivated in 300 ml baffled shaking flasks filled with 50 ml of SWM ( 2 gl−1 succinate as sole carbon source ) . Cells were harvested at an OD600 of 1 ( mid-exponential phase ) . Quantifications of macromolecules ( lipids , proteins , DNA , RNA ) were performed at least in three independent experiments sampling triplicates . Average amounts of total macromolecules were calculated in relation to cell dry weight of D . shibae DFL12T . The experiments were carried out following the manufacturers' instructions ( Biolog Inc . , USA ) with modifications . The inoculation and incubation solutions IF-0a GN/GP were adapted to the artificial sea water medium by adding a 10-fold concentrated SWM solution including vitamins and trace elements without carbon source ( final concentration of components equal to SWM ) . Cultivation of D . shibae DFL12T was performed as described above using 2 g l−1 succinate as sole carbon source . Sampling took place every hour until cultures ( three in parallel ) entered the early stationary growth phase ( approximately 16 h ) . Sampling was performed by transferring 1 ml cell culture into 2-ml tubes . Cells were separated from the medium by centrifugation ( 10 , 000 g , 4°C , 5 min ) . Supernatants were transferred into fresh tubes and stored at −20°C until further processing with GC/MS . For the comparison of bacterial motility D . shibae DFL12T was cultivated as described above using 2 g l−1 succinate as sole carbon source and 1 . 3 g l−1 glucose , respectively . During exponential growth phase subsamples were taken and transferred to a stage micrometer . To prevent false positive movements due to evaporation of medium the cover slip was sealed with nitrocellulose . Cells were monitored through a microscope ( Zeiss Axiostar plus ) equipped with a 40× objective . The bacterial movement was digitally recorded for 60 seconds with a Canon PowerShot A640 camera ( 640×480 pixels resolution , 16× zoom ) which was connected to the microscope . Five still images per second were extracted from the videos and the hqdn3d filter of FFmpeg version 0 . 8 . 5-6 was used to filter noise . Subsequently , all images were converted to grayscale and bright spots were mapped to black pixels with ImageMagick; version 6 . 7 . 7-10 . Spot detection and tracking was performed using Icy version 1 . 3 . 1 . 0 [58] . Only tracks with a length of at least two were kept for further analysis . The starting point of our reconstruction process was the genome annotation of Dinoroseobacter shibae DFL12T provided by the EnzymeDetector database [25] . This database aggregates annotations from different sources and assigns a relevance score to each entry indicating the level of confidence . To exclude poorly annotated enzymes , we selected only entries with a minimum relevance score of 9 . Occasionally , multiple enzymes are annotated for one open reading frame . In such cases the two best entries were compared with each other . Both were kept if their score was equal or above 13 . Otherwise only the entry with the best score was kept . Furthermore , manual additions were made to the annotation during the reconstruction to fill gaps in the metabolic network . The final annotation used for the creation of the model can be found in the Supplementary Dataset S1 . To create the metabolic model , the enzymes from the annotation were mapped to the corresponding chemical reactions via their EC number . This mapping was based on the MetaCyc database version 16 . 0 [60] . Next , spontaneous reactions were added under the condition that all educts of the reaction were already part of the model . This step has been repeated until no new reactions were found . Transport and boundary reactions were added manually to allow certain nutrients , additional vitamins and waste products to enter or leave the system . This preliminary model was iteratively refined by adding additional enzymes and reactions to fully reproduce growth under different conditions . Stoichiometric balancing was performed computationally for all reactions . The non-growth-associated maintenance requirement ( nGAM ) and the growth-associated maintenance requirement ( GAM ) were assumed to be the same as in E . coli ( 3 . 15 mmol ATP/ ( gDW h ) and 53 . 95 mmol ATP/ ( gDW h ) [33] . While the first value models the energy demand of processes not related to growth like DNA repair and preservation of turgor pressure , the second value accounts for the energy needed for reproduction . Most of this energy is needed for the synthesis of proteins , DNA , and RNA . The number of protons in mmol/ ( gDW h ) needed to drive the flagellar motor Pmotility has been estimated based on the number of protons needed for one rotation of the motor N = 1200 [61] , the average number of rotations per second v = 10 s−1 [62] , and the average dry weight of one Roseobacter cell m = 300fg [63]:However , only about 10% of the organisms in a culture of marine bacteria are motile during the early exponential growth phase modeled here [24] . Hence , we constrained the motility proton flux to 24 mmol/ ( gDW h ) . We simulated the physiological states using flux balance analysis [64] , [65] . Furthermore , each flux was tested for variability under the additional constraint of optimal biomass production by ( fast ) flux variability analysis [66] , [67] . All computational analyses were carried out on a computer equipped with a 2 . 67 GHz Intel Core i7 CPU and 4 GB of RAM . The software in use was the metano toolbox ( Riemer et al , in preparation , http://metano . tu-bs . de ) . Altogether , the simulations took about four days to finish . For further evaluation , the resulting fluxes were stored in a relational database . Single gene knock-outs were simulated by constraining the flux through all reactions associated with that particular gene to zero .
|
The oceans are home to a large variety of microorganisms , which interact in several ways with world-wide metabolic cycles . A representative of an important group of marine bacteria called the Roseobacter clade is Dinoroseobacter shibae . This organism is known to use a variant of photosynthesis to obtain energy from light . Another feature of D . shibae and many other Roseobacters is the ability to degrade an abundant compound in the ocean called dimethylsulfoniopropionate . Importantly , one degradation pathway of dimethylsulfoniopropionate releases a side product , which affects climate by seeding cloud formation . In this work , we constructed a genome-scale metabolic model of D . shibae and carried out a detailed computational analysis of its metabolism . Our model simulates the light-harvesting capabilities of D . shibae and also accounts for the energy needed to swim . Thanks to our exhaustive simulations we were able to elucidate the effect of light on the growth of D . shibae , to study the consequences of genetic perturbations , and to identify mutants which produce more cloud-seeding compounds . Foremost , our computational results help to understand an important part of the complex processes in the ocean in greater detail . Besides , they can be a valuable guide for future wet-lab experiments .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
Swimming in Light: A Large-Scale Computational Analysis of the Metabolism of Dinoroseobacter shibae
|
Severe fever with thrombocytopenia syndrome ( SFTS ) is an emerging infectious disease that is caused by a novel bunyavirus SFTSV . Currently our knowledge of the host-related factors that influence the pathogenesis of disease is inadequate to allow prediction of fatal outcome . Here we conducted a prospective study of the largest database on the SFTS patients , to identify the presence of comorbidities in SFTS , and estimate their effect on the fatal outcome . Among 2096 patients eligible for inclusion , we identified nine kinds of comorbidities , from which hyperlipidemia ( 12 . 2%; 95% CI: 10 . 8%–13 . 6% ) , hypertension ( 11 . 0%; 95% CI: 9 . 6%–12 . 3% ) , chronic viral hepatitis ( CVH ) ( 9 . 3%; 95% CI: 8 . 1%–10 . 5% ) , and diabetes mellitus ( DM ) ( 6 . 8%; 95% CI: 5 . 7%–7 . 9% ) were prevalent . Higher risk of death was found in patients with DM ( adjusted OR = 2 . 304; 95% CI: 1 . 520–3 . 492; P<0 . 001 ) , CVH ( adjusted OR = 1 . 551; 95% CI: 1 . 053–2 . 285; P = 0 . 026 ) and chronic obstructive pulmonary diseases ( COPD ) ( adjusted OR = 2 . 170; 95% CI: 1 . 215–3 . 872; P = 0 . 009 ) after adjusting for age , sex , delay from disease onset to admission and treatment regimens . When analyzing the comorbidities separately , we found that the high serum glucose could augment diseases severity . Compared to the group with max glucose < 7 . 0 mmol/L , patients with glucose between 7 . 0–11 . 1 mmol/L and glucose ≥11 . 1 mmol/L conferred higher death risk , with the adjusted OR to be 1 . 467 ( 95% CI: 1 . 081–1 . 989; P = 0 . 014 ) and 3 . 443 ( 95% CI: 2 . 427–4 . 884; P<0 . 001 ) . Insulin therapy could effectively reduce the risk of severe outcome in DM patients with the adjusted OR 0 . 146 ( 95% CI: 0 . 058–0 . 365; P<0 . 001 ) . For CVH patients , severe damage of liver and prolongation of blood coagulation time , as well as high prevalence of bleeding phenotype were observed . These data supported the provocative hypothesis that treating SFTS related complications can attain potentially beneficial effects on SFTS .
SFTS is an emerging infectious disease that was first reported in 2009 in rural areas in central China . SFTSV is principally transmitted to human by tick bites in natural foci , with a possible human-to-human transmission through contacts with infected blood or body fluid [1] . After infection , patients experienced an extensively wide clinical spectrum , with some experiencing self-limiting clinical course , while approximately 16 . 2% ( 95% CI: 14 . 6%-17 . 8% ) developing fatal outcome [2] . Until now , the risk factors for the fatal outcome and pathogenesis mechanisms underlying the fast proceeding to fatal SFTS remained sparsely investigated [2–4] . As has been displayed widely , numerous factors can influence the outcome of viral infections , including but not restricted to pathogen factors , host genetic susceptibility , host immunity response , host comorbidity conditions , and the effects of therapy [5–7] . All of these factors play in a complex way to determine the final outcome after viral infection . For SFTSV infection , our knowledge of the host-related factors that influence the pathogenesis of disease is far from adequate to allow prediction of fatal outcome . Only older age has been associated with higher risk of fatal outcome with consistent conclusion [8–10] . Preexisting chronic conditions , on the other hand , although long been considered to increase risk of death in a range of viral diseases , were rarely studied in SFTS [11 , 12] . Moreover , the specific classification of the diseases was not assigned , therefore making the identification of high-risk population unlikely to attain . In the current study , we are designed to characterize the prevalence of the common preexisting comorbidities related to the metabolic syndromes-associated diseases , including hyperlipidemia , hypertension , chronic viral hepatitis ( CVH ) , diabetes mellitus ( DM ) , cerebral ischemic stroke , heart diseases , chronic obstructive pulmonary diseases ( COPD ) , pulmonary tuberculosis and cancer in SFTSV infections and to evaluate their effect on diseases outcome . Considering the potential interaction between SFTS and comorbidities such as DM , hyperlipidemia , hypertension on resulting in endothelial dysfunction [13–17] , we further measured the level of adhesion factors in patients with or without comorbidities as an indicator of endothelial activation/dysfunction .
The study was performed on a prospectively observed cohort of SFTSV infected patients who were recruited in People's Liberation Army ( PLA ) 154 hospital ( now named as The 990 Hospital of Chinese People's Liberation Army Joint Logistic Support Force ) . The basic information on the cohort has been described in our previous study [2] . Briefly , the hospital is located in Xinyang city , which is located at the southern part of Henan Province , bordering the provinces of Anhui and Hubei to the east and south respectively , representing one of the most severely inflicted cities in central China . Since the beginning of SFTS surveillance in 2011 till 2017 , the hospital had diagnosed and treated the largest number ( over 30% ) of total SFTS cases in China [18] . All the 2096 laboratory-confirmed SFTSV patients were used for the analysis , who met one or more of the following criteria: ( 1 ) a positive SFTSV culture ( 2 ) a positive result for SFTSV RNA by real-time RT-PCR assay ( 3 ) seroconversion or ≥4 fold increase of antibody titers between 2 serum samples collected at least 2 weeks apart . The medical record of all the hospitalized patients was maintained in an electronic system with logic error correction function , ensuring the credibility of the data . For the current research , a medical record review was performed to collect the information on demographic characteristics , preexisting comorbidities , clinical information , laboratory test results and treatment regimens during the entire hospitalization . The clinical information mainly included symptoms and signs that were recorded from the daily physical examination . The extracted laboratory results mainly included hematology , clinical chemistry , urinalysis and live function examination , which were prescribed on hospital admission and during the hospitalization . Other laboratory indicator included blood cultures , HIV , HBV , HCV-specific antigen and antibody testing , electrocardiogram as well as chest radiograph test , which were prescribed on hospital admission . These data were drawn from the database by a group of trained physicians using a standardized format and entered into an EpiData database . The data were further reviewed for accuracy and consistency by a second group of epidemiologists . For the patients who had missing information , a trained study staff interviewed the patients or their family using a standardized supplemental questionnaire . The comorbidities that were used for the current analysis included diabetes mellitus ( both type I and type II , ICD-10 E14 . 8 ) , hypertension ( ICD-10 I10 . X02 ) , hyperlipidemia ( ICD-10 E78 . 500 ) , chronic virus hepatitis ( both HBV and HCV , ICD-10 B18 . 951 ) , cerebral ischemic stroke ( ICD-10 I64 . X04 ) , chronic obstructive pulmonary diseases ( ICD-10 J44 . 900 ) , pulmonary tuberculosis ( ICD-10 B90 . 901 ) , cancer ( ICD-10 C00-C97 ) , heart diseases ( ICD-10 I51 . 900 , due to the small sample size of individual conditions , cardiac heart failure , coronary atherosclerotic heart disease , arrhythmia and other heart diseases were combined into this category ) . The virus load was determined using real-time reverse transcriptase polymerase-chain-reaction ( RT-PCR ) targeting the same gene segment . Serum levels of ten adhesion factors were determined by the ProcartaPlex multiplex immunoassays panels ( Affymetrix , USA ) according to the manufacturer instructions . The measurement of 25 cytokines levels was performed for the serum samples of the survived patient by using Cytokine Human 25-Plex Panel ( Life Technologies , USA ) . The serum samples tested adhesion factors and cytokines were collected on admission and all were within seven days after the onset of disease . Continuous variables were summarized as means and standard deviations ( SD ) or as medians and interquartile range ( IQR ) . Categorical variables were summarized as frequencies and proportions . An independent t test , a χ2 test , a Fisher exact test , or a nonparametric test was used where appropriate to calculate the differences between groups . Logistic regression model was applied to explore the association between comorbidities and clinical information or fatal outcome . The generalized estimating equation ( GEE ) was constructed to compare the laboratory parameters that were evaluated over time . Cytokine and adhesion factors were compared between groups after performing 10 logarithmic transformations using generalized linear model ( GLM ) . Age , sex , time from disease onset to admission and treatment regimens ( ribavirin , corticosteroid and immunoglobulin ) were adjusted in the above models . Odds ratios ( ORs ) and their 95% confidence intervals ( CIs ) were estimated . A two-sided P < 0 . 05 was considered statistically significant . All analyses were performed using Stata 14 . 0 ( Stata Corp LP , College Station , TX , USA ) . The study protocol was approved by the human ethics committee of the hospital PLA 154 . Written or verbal informed consent had been obtained from all the patients or from parents/guardians on behalf of all pediatric participants .
A total of 2096 laboratory-confirmed SFTS patients who were hospitalized from 2011 to 2017 were used for analysis [2] . The mean ( SD ) of the age was 61 . 4 ( 12 . 2 ) years old , and 1239 ( 59 . 1% ) were female . Overall , the presence of preexisting comorbidities was reported in 779 ( 37 . 2% ) of the patients . Hyperlipidemia was the most prevalent comorbidity ( n = 256; 12 . 2%; 95% CI: 10 . 8%-13 . 6% ) , followed by hypertension ( n = 230; 11 . 0%; 95% CI: 9 . 6%-12 . 3% ) , CVH ( n = 195; 9 . 3%; 95% CI: 8 . 1%-10 . 5% ) , and DM ( n = 142; 6 . 8%; 95% CI: 5 . 7%-7 . 9% ) and other diseases ( Fig 1 ) . The SFTS patients with the comorbidities had older age and longer time from disease onset to admission compared those without ( both P<0 . 001 ) ( Table 1 ) . For the 195 patients with CVH , 179 ( 91 . 8% ) were infected with HBV , 15 ( 7 . 7% ) with HCV and 1 ( 0 . 5% ) with both infection . Among the patients , 179 had two kinds of comorbidities and 41 had three or more , mostly observed between hyperlipidemia and others ( Table 2 ) . The case fatality rate ( CFR ) among the patients with any kind of comorbidity was 22 . 5% ( 175/779 ) , significantly higher than those without ( 12 . 5%; 165/1317; adjusted OR = 1 . 628; 95% CI: 1 . 265–2 . 096; P<0 . 001 ) ( S1 Table ) . When the comorbidities were assessed individually , hypertension , DM , CVH and COPD were significantly associated with the development of fatal outcome ( all P<0 . 05 ) . However , after adjusting the effect from age , sex , delay from disease onset to admission and treatment regimens ( ribavirin , corticosteroid and immunoglobulin ) , the significance only remained for DM ( adjusted OR = 2 . 304; 95% CI: 1 . 520–3 . 492; P<0 . 001 ) , CVH ( adjusted OR = 1 . 551; 95% CI: 1 . 053–2 . 285; P = 0 . 026 ) and COPD ( adjusted OR = 2 . 170; 95% CI: 1 . 215–3 . 872; P = 0 . 009 ) ( Fig 2 and S1 Table ) . The presence of over one kind of comorbidity was associated with enhanced risk of death , with the coexistence of DM & CVH having significantly higher odds ratio of developing fatal outcome ( adjusted OR = 4 . 792; 95% CI: 1 . 345–17 . 077; P = 0 . 016 ) , in comparison with those without any comorbidity , thus indicating an interaction between them ( Table 2 ) . We applied a logistic regression model with stepwise method to adjust the potential interaction effects that might be derived from inter-comorbidities . The analysis showed three significant comorbidities in the model that attained higher risk of death: the presence of DM ( adjusted OR = 2 . 328; 95% CI: 1 . 534–3 . 532; P<0 . 001 ) , CVH ( adjusted OR = 1 . 557; 95% CI: 1 . 056–2 . 296; P = 0 . 025 ) , and COPD ( adjusted OR = 2 . 138; 95% CI: 1 . 195–3 . 825; P = 0 . 010 ) . Clinical manifestations and laboratory assessments were compared between two groups . At presentation , three of the commonly seen signs or symptoms , including dizziness , headache , chills and gastrointestinal symptoms were more frequently observed in SFTS patients with comorbidities than those without . Severe complications , including respiratory symptoms , neurological symptoms and haemorrhagic symptoms developed with higher frequency in SFTS patients with any kind of comorbidities than those without ( Table 1 ) . The GEE analysis displayed levels of alanine transaminase ( ALT ) , aspartate aminotransferase ( AST ) , white blood cell ( WBC ) , creatine kinase ( CK ) , globulin ( GLB ) and lactate dehydrogenase ( LDH ) were significantly elevated , while the level of albumin ( ALB ) significantly decreased among the SFTS patients with comorbidities . No differences in SFTSV viral load or platelet counts were observed between the two groups ( Fig 3 ) . Altogether 142 patients with DM and 1954 without were compared for their clinical manifestations and laboratory indicators . Most of the initial symptoms and signs were reported from two groups with similar frequencies ( S2 Table ) . On the other hand , respiratory and neurological complications were more likely to develop in the patients with DM than in patients without . The dynamic profiles of laboratory parameters during the whole course were similar between two groups , except three higher levels of laboratory indicators ( GLB , LDH and SFTSV viral loads ) and lower levels of ALB in the patients with DM than those without ( Fig 4 ) . When the patients were further grouped according to the maximum glucose level during the whole hospitalization ( S3 Table ) , a dose dependent effect was displayed as the decrease in ALB , together with elevation in AST , ALT , WBC , CK , GLB , LDH and SFTSV viral loads were negatively correlated with the glucose level ( Fig 5 ) . Moreover , the glucose level significantly affected the risk of death ( Fig 6A ) . Compared to the group with glucose < 7 . 0 mmol/L , patients with glucose between 7 . 0–11 . 1 mmol/L and glucose ≥11 . 1 mmol/L had higher death risk , with the adjusted OR estimated to be 1 . 467 ( 95% CI: 1 . 081–1 . 989; P = 0 . 014 ) and 3 . 443 ( 95% CI: 2 . 427–4 . 884; P<0 . 001 ) , respectively ( Table 3 ) . The DM patients who received insulin therapy over four times had significant lower glucose level ( S4 Table and Fig 6B ) , which conferred a significantly reduced risk of fatal outcome ( adjusted OR = 0 . 146; 95% CI: 0 . 058–0 . 365; P<0 . 001 ) ( Table 3 ) . Totally 139 patients had ten adhesion factors evaluated , including 50 patients with glucose level ≥ 7 . 0 mmol/L on admission ( S5 Table ) . Only serum amyloid antigen 1 ( SAA-1 ) was found to be elevated in the patients with glucose ≥ 7 . 0 mmol/L than those with glucose <7 . 0 mmol/L ( Fig 7 ) . Altogether 64 SFTS patients had their serum cytokines measured on admission , including 17 patients with glucose exceeding 7 . 0 mmol/L ( S6 Table ) . Six of the 25 tested cytokine , including Interleukin-1RA ( IL-1RA ) , Interleukin-2 ( IL-2 ) , IL-4 , IL-6 , Granulocyte macrophage-stimulating factor ( GM-CSF ) , Interferon-γ ( IFN-γ ) were significantly higher in SFTS patients with high glucose level ( Fig 8 ) . Altogether 195 patients with CVH and 1901 without were compared for their clinical manifestations and laboratory indicators . Most of the non-specific signs or symptoms were comparable between the CVH and non-CVH groups ( S7 Table ) . Higher frequency of neurological symptoms ( 33 . 3% vs . 25 . 6%; adjusted P = 0 . 045 ) and haemorrhagic symptoms ( 45 . 1% vs . 34 . 0%; adjusted P = 0 . 017 ) were disclosed in the CVH group . Higher ALT , AST , CK , GLB concentration , lower platelet counts and ALB level were also related to the presence of CVH in SFTS patients ( Fig 9 ) . Totally 98 CVH and 867 non-CVH patients with SFTSV infection had blood coagulation function test on admission available for analysis ( S8 Table ) . The patients with CVH developed more prolongation of the prothrombin time ( PT ) and activated partial thromboplastin time ( APTT ) and thrombin time ( TT ) , all indicating occurrence of disseminated intravascular coagulation ( DIC ) . The pattern of these parameters corresponded with decreased platelet and high prevalence of bleeding phenotype ( Fig 10 ) .
Over the past few years , several lines of evidence have supported the notion that cardiovascular disease , stroke , diabetes , respiratory diseases and renal disorders may contribute , together with old age , to severe dengue disease [19–21] . Studies on West Nile virus [22] , Japanese encephalitis virus [23] infections , and responses to Yellow fever virus vaccination [24] , have also supported the pathogenic role of chronic comorbidities in the prognosis of infections . Since the discovery of SFTS , although clinical phenotypes have been developed to differentiate the patients with high risk of death , host factors remained sparsely investigated . In this study , we demonstrated the frequency of underlying conditions in SFTSV infected patients and determined their role in developing fatal outcome . Hyperlipidemia and hypertension are the most prevalent comorbidity , while DM , CVH and COPD were more prominent in their association with fatal outcome , with 1 . 551–2 . 304 fold increase in their risk of death than the SFTS patients without comorbidities . From the perspective of clinical features , neurological manifestation and hemorrhagic signs were more frequently seen in patients with underlying diseases , both contributing to the final fatal outcome . A common pathogenic feature of SFTS infection is their ability to inhibit the host immune response , characterized by significantly reduced CD3-positive and CD4-positive T lymphocytes than normal [25] . This is consistent with the clinical phenomenon that most infection occurred in the elderly , who are considered to possess compromised T-cell function [26] . The association between DM , CVH and severe diseases might also be related to immune dysfunction . Abnormal innate and adaptive immunity used to be disclosed in DM patients , reflected by alterations in proliferation of T cells and macrophages , and impairment in function of NK cells and B cells in DM patients [27] . CVH , either hepatitis B or hepatitis C , was capable of inhibiting the adaptive or innate immune response [28] . This supported the hypothesis that DM and CVH , in combination with SFTSV infection , might impair the immune system and attenuate anti-inflammatory responses , subsequently resulting in increased level and prolonged duration of viremia , which predispose patients to higher risk of death . In addition , the preexisting DM is often linked to vascular complications featured by an activation of the inflammation cascade and endothelial dysfunction [13] , which are also identified in SFTSV infection [17] . In line with this mechanism , we observed a remarkably enhanced expression of SAA-1 , a biomarker of endothelial dysfunction [17] , in DM-SFTS than SFTS alone . Cytokine storm had been extensively found to play roles in the pathogenesis of SFTS [29] . Based on the current findings , IL-1RA , IL-2 , IL-4 , IL-6 , GM-CSF and IFN-γ were elevated to remarkably high levels in DM-SFTS patients , likely contributing to the fatal outcome , together with the development of depressed immunity and endothelial dysfunction . All these indicators showed potential to predict adverse outcome . The insulin therapy , on the other hand , had obviously reduced the disease severity . Therefore , it’s justified to actively identify and treat high glucose in SFTSV infection , in order to attain extra benefit of reducing viremia and enhancing disease outcomes . Differing from DM-SFTS , the CVH-SFTS patients were prone to have higher incidence of bleeding manifestation than those without . In line with these findings , abnormal coagulation factors , including platelet and others , were more frequently seen in CVH-SFTS . An interactive effect on liver damage from CVH and SFTS was observed , as liver function related enzymes , especially AST , ALT and ALB , demonstrated remarkable deviation from normal level , which was indicative of progressive hepatic involvement in those patients . As liver is the primary source for producing coagulation factors [30 , 31] , it is reasonable to deduce that the interactive effect from SFTSV and hepatitis virus can predispose the patient to more frequent bleeding . It’s noteworthy that ALB was constantly reduced in patients with DM , or CVH or any kind of comorbidity . ALB is the most abundant protein in plasma , representing the main determinant of plasma oncotic pressure and the main modulator of fluid distribution between body compartments . ALB plays an import role of in plasma leakage that could be a parameter to predict the severity of diseases [32 , 33] . Recently , the endothelial dysfunction and plasma leakage had been identified in SFTSV infection and most viral hemorrhagic fever , manifested by fluid loss from the vascular compartment and by decreased level of ALB [20 , 34 , 35] . It is hypothesized that hypoalbuminemia could be induced from a synergetic effect of comorbidity and SFTSV infection , eventually contributing to the high morbidity and mortality . As such , albumin administration in SFTS patients might be effective in improving the disease outcome . The study is subject to major limitation that when assessing comorbidities , we did not allow for differentiating between those diagnosed before , after or during the infectious episodes . Therefore , the causality between the conditions and adverse outcome cannot be inferred . However , even in the absence of causal inference between the non-communicable and infectious diseases , these findings may guide clinicians to predict complications , at least partially , based on the presence of comorbidity . In addition , we made no efforts to distinguish type 1 from type 2 diabetes for separate analysis , despite of their differential clinical features and etiological factors . Instead we used the glucose level as major variable to explore the dose effect of glucose on the disease severity . Moreover , the clinical status that were acquired from these patients were only partially used , and due to the high cost of testing adhesion factors and interleukins , we did not evaluate the entire population for these indicators , which might have caused selection bias for the inter-group comparison . In conclusion , we provided evidence for a higher prevalence of DM , CVH and COPD in fatal SFTS patients , elucidating the possible mechanism that underlies their interactive effect in resulting in adverse outcome . This knowledge might allow clinical physicians to identify the patients with preexisting comorbidities who may progress to a severe course , thereby to adopt aggressive interventions at early infection .
|
SFTS now brings about a substantial global public health concern . Preexisting chronic conditions were thought to increase risk of severe SFTSV infections , however with sparse data mining efforts . In this study , we quantified the frequency of chronic comorbidities in SFTS , estimated their contribution to disease severity , and separately evaluated the effect from diabetes mellitus and chronic viral hepatitis on resulting in fatal outcome .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"cytokines",
"chemical",
"compounds",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"immunology",
"microbiology",
"carbohydrates",
"organic",
"compounds",
"glucose",
"pulmonology",
"diabetes",
"mellitus",
"chronic",
"obstructive",
"pulmonary",
"disease",
"developmental",
"biology",
"endocrine",
"disorders",
"signs",
"and",
"symptoms",
"molecular",
"development",
"antibodies",
"viral",
"load",
"immune",
"system",
"proteins",
"proteins",
"endocrinology",
"chemistry",
"hypertension",
"blood",
"pressure",
"hyperlipidemia",
"metabolic",
"disorders",
"immune",
"system",
"biochemistry",
"diagnostic",
"medicine",
"organic",
"chemistry",
"virology",
"physiology",
"monosaccharides",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"vascular",
"medicine"
] |
2019
|
Preexisting chronic conditions for fatal outcome among SFTS patients: An observational Cohort Study
|
Novel strategies are required to control mosquitoes and the pathogens they transmit . One attractive approach involves maternally inherited endosymbiotic Wolbachia bacteria . After artificial infection with Wolbachia , many mosquitoes become refractory to infection and transmission of diverse pathogens . We evaluated the effects of Wolbachia ( wAlbB strain ) on infection , dissemination and transmission of West Nile virus ( WNV ) in the naturally uninfected mosquito Culex tarsalis , which is an important WNV vector in North America . After inoculation into adult female mosquitoes , Wolbachia reached high titers and disseminated widely to numerous tissues including the head , thoracic flight muscles , fat body and ovarian follicles . Contrary to other systems , Wolbachia did not inhibit WNV in this mosquito . Rather , WNV infection rate was significantly higher in Wolbachia-infected mosquitoes compared to controls . Quantitative PCR of selected innate immune genes indicated that REL1 ( the activator of the antiviral Toll immune pathway ) was down regulated in Wolbachia-infected relative to control mosquitoes . This is the first observation of Wolbachia-induced enhancement of a human pathogen in mosquitoes , suggesting that caution should be applied before releasing Wolbachia-infected insects as part of a vector-borne disease control program .
Efforts to control vector-borne pathogens have been hindered by evolution of insecticide resistance and failing drug therapies . Evidence suggests bed nets and indoor residual spraying with insecticides are losing efficacy in developing countries [1] , [2] . To improve the sustainability and efficacy of control efforts , alternative vector control strategies are being considered , including methods that suppress the pathogen instead of the vector [3] , [4] . Wolbachia are a genus of maternally-inherited bacterial endosymbionts that naturally occur in numerous arthropod taxa [5] . Wolbachia can inhibit viruses and parasites in fruit flies and mosquitoes [6]–[11] and influence reproduction of their host to facilitate spread through populations [12] . Mosquito-borne disease management programs that use Wolbachia are currently under investigation [13] . In field trials in Australia , Wolbachia reached fixation in naturally uninfected populations of Aedes aegypti [11] and the DENV blocking phenotype has been maintained [14] , but the impacts of Wolbachia on reducing the incidence of disease are yet to be investigated . Pathogen interference conferred by Wolbachia depends on various factors , including Wolbachia strain , pathogen type , infection type ( natural versus artificial ) and host and is not a guarantee [7] , [15] , [16] . For example , Wolbachia increases Plasmodium berghei , P . yoelii and P . gallinaceum oocyst loads in Anopheles gambiae , An . stephensi , and Aedes fluviatilis , respectively [17]–[19] , and P . relictum sporozoite prevalence in Culex pipiens [20] . These Wolbachia-mediated pathogen enhancement studies suggest that careful examination of Wolbachia is required , since the bacterium influences insect-pathogen interactions in ways that may negatively impact pathogen control efforts . Few studies have investigated the effect of Wolbachia on pathogen transmission by Culex mosquitoes , despite the fact they transmit viruses impacting human health [9] , [21] , [22] . Culex tarsalis is a mosquito species associated with agriculture and urban areas in the western United States [23] and is highly competent for West Nile virus ( WNV ) , St . Louis encephalitis virus ( SLEV ) and western equine encephalitis virus ( WEEV ) [24]–[26] . Cx . tarsalis are naturally uninfected with Wolbachia [27] . We established Wolbachia infections in this mosquito by intrathoracic injection of purified symbionts into adult females , characterized the extent of the infection by fluorescence in situ hybridization and quantitative PCR , and assessed the ability for Wolbachia to block WNV infection , dissemination and transmission at multiple time points . We found that , in contrast to other systems , Wolbachia infection enhanced WNV infection rates 7 days post-blood feeding . This is the first observation of Wolbachia-induced enhancement of a human pathogen in mosquitoes and suggests that caution should be applied before using Wolbachia as part of a vector-borne disease control program .
Mosquitoes were maintained on commercially available bovine blood using a membrane feeder . WNV infection experiments were performed under biosafety-level 3 ( BSL3 ) and arthropod-containment level 3 ( ACL3 ) conditions . The Cx . tarsalis YOLO strain was used for experiments . The colony was originally established from Yolo County , CA in 2009 . Mosquitoes were reared and maintained at 27°C±1°C , 16∶8 hour light∶dark diurnal cycle at approximately 45% relative humidity in 30×30×30 cm cages . The wAlbB Wolbachia strain was purified from An . gambiae Sua5B cells according to published protocols [28] . Viability and density of the bacteria was assessed using the Live/Dead BacLight Kit ( Invitrogen ) and a hemocytometer . The experiment was replicated twice; wAlbB concentrations were: replicate one , 5 . 3×109 bacteria/mL; replicate two , 1 . 3×1011 bacteria/mL . Two- to four-day-old adult female Cx . tarsalis were anesthetized with CO2 and intrathoracically ( IT ) injected with approximately 0 . 1 uL of either wAlbB or Schneider's insect media ( Sigma Aldrich ) as a control . Mosquitoes were provided with 10% sucrose ad libitum and maintained at 27°C in a growth chamber . WNV strain WN02-1956 ( GenBank: AY590222 ) was originally isolated in African green monkey kidney ( Vero ) cells from an infected American crow in New York in 2003 [29] and amplified in Aedes albopictus cells ( C6/36 ) to a final titer of 5 . 0×109 PFU/ml . WNV was added to 5 mL defibrinated bovine blood ( Hema-Resource & Supply , Aurora , OR ) with 2 . 5% sucrose solution . Replicate titers were: replicate one , 8 . 0×107 PFU/mL; replicate two , 3 . 0×107 PFU/mL . Seven days post Wolbachia injection mosquitoes were fed a WNV infectious blood meal via Hemotek membrane feeding system ( Discovery Workshops , Accrington , UK ) for approximately one hour . Partially- or non-blood fed females were excluded from the analysis . To characterize Wolbachia infections in Cx . tarsalis tissues , we performed fluorescence in situ hybridization ( FISH ) on mosquitoes at 12 dpi according to published protocols [10] with slight modifications . Briefly , mosquitoes were fixed in acetone , embedded in paraffin wax and sectioned with a microtome . Slides were dewaxed with three successive xylene washes for 5 minutes , followed by two 5-minute washes with 100% ethanol and one wash in 95% ethanol before treatment with alcoholic hydrogen peroxide ( 6% H2O2 in 80% ethanol ) for 3 days to minimize autofluorescence . Sectioned tissues were hybridized overnight in 1 ml of hybridization buffer ( 50% formamide , 5× SSC , 200 g/liter dextran sulfate , 250 mg/liter poly ( A ) , 250 mg/liter salmon sperm DNA , 250 mg/liter tRNA , 0 . 1 M dithiothreitol [DTT] , 0 . 5× Denhardt's solution ) with Wolbachia specific probes W1 and W2 labeled with a 5-prime rhodamine fluorophore [30] . After hybridization , tissues were successively washed three times in 1× SSC , 10 mM DTT and three times in 0 . 5× SSC , 10 mM DTT . Slides were mounted with SlowFade Gold antifade reagent ( Invitrogen ) and counterstained with DAPI ( Roche ) . Images were captured with a LSM 510 META confocal microscope ( Zeiss ) and epifluorescent BX40 microscope ( Olympus ) . Images were processed using LSM image browsers ( Zeiss ) and Photoshop 7 . 0 ( Adobe ) software . No-probe , competition probe and RNAse treatment controls were conducted ( Figure S1 ) . Virus infection and transmission assays were performed as described at 7 and 14 days post blood feeding [31]–[33] . Female mosquitoes were anesthetized with triethylamine ( Sigma , St . Louis , MO ) , legs from each mosquito were removed and placed separately in 1 mL mosquito diluent ( MD: 20% heat-inactivated fetal bovine serum [FBS] in Dulbecco's phosphate-buffered saline , 50 ug/mL penicillin/streptomycin , 50 ug/mL gentamicin and 2 . 5 ug/mL fungizone ) . The proboscis of each mosquito was positioned in a tapered capillary tube containing 10 uL of a 1∶1 solution of 50% sucrose and FBS to induce salivation . After 30 minutes , the contents were expelled into 0 . 3 mL MD and bodies were placed individually into 1 mL MD . Mosquito body , legs and salivary secretion samples were stored at −70°C until tested for WNV presence and Wolbachia titers . Mosquito bodies and legs were homogenized for 30 seconds utilizing Qiagen Tissue Lyser at 24 cycles/second , followed by clarification via centrifugation for one minute . Mosquito samples were tested for WNV infectious particles by plaque assay on Vero cells [34] . Infection was defined as the proportion of mosquitoes with WNV positive bodies . Dissemination and transmission were defined as the proportion of infected mosquitoes with WNV positive legs and salivary secretions , respectively . Proportions were compared using Fisher's exact test . The experiment was replicated twice . To evaluate Wolbachia density in individual mosquitoes from vector competence experiments , DNA was extracted using DNeasy Blood and Tissue kits ( Qiagen ) and used as template for qPCR on a Rotor Gene Q ( Qiagen ) with the SYBR green PCR kit ( Qiagen ) . Wolbachia DNA was amplified with primers Alb-GF and Alb-GR [35] and was normalized to the Cx . tarsalis actin gene [36] ( Table 1 ) . Wolbachia to host genome ratios were calculated using Qgene [37] . PCRs were performed in duplicate . Comparisons of Wolbachia titers between treatments were analyzed using Mann-Whitney U test . To explore Wolbachia effects on mosquito immune gene expression , one- to four- day old adult female Cx . tarsalis were anesthetized with CO2 and injected as described above with Wolbachia ( wAlbB ) or Schneider's insect media as control . Mosquitoes were provided with 10% sucrose ad libitum and maintained at 27°C in a growth chamber . At 7 dpi , mosquitoes were blood fed on bovine blood via glass membrane feeder . At 2 dpf , five mosquitoes per treatment were harvested and RNA extracted using RNeasy mini kits ( Qiagen ) . Extracted RNA was DNase treated ( Ambion #AM1906 ) and converted to cDNA using Superscript III with random hexamers ( Invitrogen #18080-51 ) according to the manufacturers' protocols . qPCRs were performed using the Rotor Gene Q ( Qiagen ) and SYBR Green qPCR kit ( Qiagen ) according to the manufacturer's protocol . Five target immune genes in the Toll and IMD innate immune pathways ( REL1 , REL2 , cactus , defensin and diptericin ) were selected , primers designed based on homologous genes in the Anopheles gambiae , Aedes aegypti and Culex pipiens genomes and normalized to host actin ( Table 1 ) . Gene expression was analyzed by calculating ratios of target to host gene and tested for significance using Mann-Whitney U test . All qPCRs were technically replicated twice .
Using fluorescence in situ hybridization , we observed that wAlbB establishes an infection in both somatic and germline tissue in Cx . tarsalis 12 days post injection . Wolbachia disseminated to various tissues including the head , proboscis , thoracic flight muscles , fat body and ovarian follicles ( Figure 1 ) . Cx . tarsalis appeared heavily infected , suggesting that adult microinjection is an effective method to experimentally infect this mosquito species . We evaluated the vector competence of Wolbachia-infected and uninfected Cx . tarsalis for WNV in mosquito bodies , legs and salivary secretions to determine infection , dissemination and transmission rates , respectively . Replicate results were similar , and results from pooled replicates or analysis of individual replicates were identical , so the pooled analysis is presented for clarity; results from individual replicates are available as Table S1 . wAlbB-infected Cx . tarsalis displayed significantly higher WNV infection rates 7 days post-feeding ( dpf ) ( P = 0 . 04 ) . A similar but non-significant trend was observed 14 dpf ( Figure 2 ) . If mosquitoes were infected , virus dissemination and transmission rates did not differ statistically ( Table S1 ) . To determine if there was a Wolbachia density effect on WNV phenotype , qPCR was used to compare Wolbachia titers in mosquitoes either positive or negative WNV . Wolbachia titers in WNV-infected versus uninfected Cx . tarsalis did not differ statistically; similarly , no significant titer differences were found in individuals that disseminated versus non-disseminated or transmitted vs . non-transmitted ( Figure 3 ) . To elucidate the mechanism behind Wolbachia mediated WNV infection enhancement in Cx . tarsalis , we evaluated mosquito immune gene expression in response to Wolbachia using qPCR . Unlike other systems [38]–[40] , Wolbachia did not induce a significant immune response in Cx . tarsalis females compared to the control . In contrast , REL1 ( the NF kappa B activator of the antiviral Toll pathway ) was significantly reduced in Wolbachia-infected mosquitoes compared to controls ( one-tailed P = 0 . 008 ) ( Figure 4 ) .
It should be noted that these experiments were performed with mosquitoes transiently infected in the somatic tissues with Wolbachia , rather than a stable maternally inherited infection . It remains to be seen whether a stable wAlbB infection will enhance WNV in a similar way . Wolbachia density in mosquito somatic tissues ( as opposed to germline ) was found to explain differences in virus infection in Aedes mosquitoes [41] . Thus , it seems likely that if stable infection in Cx . tarsalis has a similar somatic tissue distribution to a transient infection it may induce a similar virus enhancement phenotype . However , this must be tested empirically . It is also unknown whether virus enhancement is limited to WNV or occurs more broadly . Finally , we tested a single Wolbachia strain , and it is unknown whether virus enhancement is specific to wAlbB or occurs with diverse Wolbachia strains . Previous studies have shown that pathogen suppression by Wolbachia has the potential to be a novel method for controlling vector-borne diseases [4] , [42]–[44] . Not all mosquito species are naturally infected with Wolbachia , but non-infected species may support infection once introduced and these novel infections often effectively inhibit various pathogens [5] , [45] . Our experiments indicate that following adult microinjection , Wolbachia is capable of establishing both somatic and germline infection in Cx . tarsalis but does not inhibit WNV infection , dissemination or transmission . In contrast with other studies showing pathogen inhibition by Wolbachia , our data suggest that Wolbachia may in fact increase WNV infection rates in Cx . tarsalis , particularly at early time points . Increased early infection has the potential to shorten the extrinsic incubation period of the pathogen , which can dramatically increase the reproductive rate of the virus [19] . It has become increasingly clear that Wolbachia does not always suppress pathogens in insects [46] . For example , the cereal crop pest Spodoptera exempta is more susceptible to nucleopolydrovirus mortality in the presence of Wolbachia [47] . In the mosquitoes An . gambiae An . stephensi , Ae . fluviatilis and Cx . pipiens , Wolbachia enhances Plasmodium berghei , P . yoelii , P . gallinaceum and P . relictum , respectively [17]–[20] . Enhancement may be dependent on the host-Wolbachia strain-pathogen system of interest , as Wolbachia strains that block one pathogen yet enhance another have been documented [9] , [17] . Wolbachia-mediated pathogen enhancement may be a common yet often ignored phenomenon , which merits attention when designing Wolbachia-based strategies for disease control [46] . Intracellular infection with bacteria may alter the cellular environment in multiple ways , including bacterial manipulation to avoid host immune defenses [48] . Though the exact Wolbachia-mediated inhibition mechanism is unknown , studies have suggested that Wolbachia indirectly modulates mosquito immunity [40] , [49] . Wolbachia can activate the Toll pathway , stimulating a cascade of events that have been correlated with inhibition of dengue and Plasmodium in mosquitoes [39] , [50] , [51] . In contrast , in Cx . tarsalis , wAlbB infection significantly downregulated REL1 ( the activator of the Toll pathway ) , suggesting that in this system Wolbachia may down regulate antiviral Toll-based immunity leading to increased virus infection . However , while statistically significant , this decrease in REL1 expression was modest , and further study is required to determine the precise mechanism of Wolbachia-based WNV enhancement in this system . To our knowledge this is first study showing Wolbachia can potentially enhance a vector-borne pathogen that causes human disease . Our results , combined with other Wolbachia enhancement studies [17]–[20] , [46]–[47] , suggest that field deployment of Wolbachia-infected mosquitoes should proceed with caution . Wolbachia effects on all potential pathogens in the study area should be determined . Additionally , several studies have shown that Wolbachia is capable of horizontal transfer to other insect species which could have unforeseen effects on non-target insects [52]–[54] . A lack of understanding of Wolbachia-pathogen-mosquito interactions could impact efficacy of disease control programs . Cx . tarsalis is a competent vector for many human pathogens , and further studies that assess alternative Wolbachia strains and viruses in Cx . tarsalis may elucidate the importance of host background on pathogen interference phenotypes in this medically important mosquito species .
|
Current methods to control mosquitoes and the pathogens they transmit are ineffective , partly due to insecticide and drug resistance . One novel control method involves exploiting naturally occurring Wolbachia bacteria in insects . Wolbachia are bacterial symbionts that are attractive candidates for mosquito-borne disease control due to their ability to inhibit pathogens infecting humans . Additionally , Wolbachia affects insect reproduction to facilitate its own transmission to offspring , which has been exploited to establish the bacterium in naturally uninfected field populations . Most Wolbachia pathogen control research has focused on Aedes and Anopheles mosquitoes , but Culex mosquitoes also transmit pathogens that affect human health . We evaluated impacts of Wolbachia infection on West Nile virus ( WNV ) in the naturally uninfected mosquito Culex tarsalis . Wolbachia was able to efficiently establish infection in Cx . tarsalis but contrary to other studies , Wolbachia enhanced rather than inhibited WNV infection . Enhancement occurred in conjunction with suppression of mosquito anti-viral immune gene expression . This study indicates that Wolbachia control strategies to disrupt WNV via pathogen interference may not be feasible in Cx . tarsalis , and that caution should be used when releasing Wolbachia infected mosquitoes to control human vector-borne diseases .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"biotechnology",
"ecology",
"and",
"environmental",
"sciences",
"medicine",
"and",
"health",
"sciences",
"environmental",
"impacts",
"immunology",
"microbiology",
"technology",
"regulations",
"parasitic",
"diseases",
"parasitology",
"population",
"biology",
"zoology",
"veterinary",
"science",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"bioethics",
"molecular",
"biology",
"systems",
"biology",
"ecology",
"science",
"policy",
"genetics",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"evolutionary",
"biology"
] |
2014
|
Wolbachia Enhances West Nile Virus (WNV) Infection in the Mosquito Culex tarsalis
|
Mass drug administration ( MDA ) is , and has been , the principal method for the control of the schistosome helminths . Using MDA only is unlikely to eliminate the infection in areas of high transmission and the implementation of other measures such as reduced water contact improved hygiene and sanitation are required . Ideally a vaccine is needed to ensure long term benefits and eliminate the need for repeated drug treatment since infection does not seem to induce lasting protective immunity . Currently , a candidate vaccine is under trial in a baboon animal model , and very encouraging results have been reported . In this paper , we develop an individual-based stochastic model to evaluate the effect of a vaccine with similar properties in humans to those recorded in baboons in achieving the World Health Organization ( WHO ) goals of morbidity control and elimination as a public health problem in populations living in a variety of transmission settings . MDA and vaccination assuming different durations of protection and coverage levels , alone or in combination , are examined as treatment strategies to reach the WHO goals of the elimination of morbidity and mortality in the coming decade . We find that the efficacy of a vaccine as an adjunct or main control tool will depend critically on a number of factors including the average duration of protection it provides , vaccine efficacy and the baseline prevalence prior to immunization . In low prevalence settings , simulations suggest that the WHO goals can be achieved for all treatment strategies . In moderate prevalence settings , a vaccine that provides 5 years of protection , can achieve both goals within 15 years of treatment . In high prevalence settings , by vaccinating at age 1 , 6 and 11 we can achieve the morbidity control with a probability of nearly 0 . 89 but we cannot achieve elimination as a public health problem goal . A combined vaccination and MDA treatment plan has the greatest chance of achieving the WHO goals in the shorter term .
Schistosomiasis inflicts significant levels of human morbidity and mortality in regions of the world with endemic infection . It is estimated that nearly 258 million people are infected worldwide with up to 700 million at risk of being infected , leading to an estimated 280000 deaths annually [1–3] . Schistosomiasis is an intestinal or urogenital disease caused predominantly by infection with Schistosoma mansoni , S . japonicum or S . haematobium , and is one of the diseases included within the World Health Organization ( WHO ) 2020 goals for neglected tropical diseases ( NTD ) control . Individuals become infected when cercariae ( larval forms of the parasitic worm ) , released by an intermediate host ( various freshwater snail species ) , penetrate the skin during contact with contaminated water [4] . Control programmes are at present based on mass drug administration ( MDA ) using the drug praziquantel , and behaviour modification directed at reducing water contact and improvements in sanitation . MDA has to be repeatedly used , since clearing infection does not result in acquired immunity and treated individuals can be re-infected . Age-related water contact behaviour results in most infection residing in school-aged children ( SAC; 5–14 years of age ) , since age intensity of infection profiles are convex in shape . Treatment is therefore specifically focused on this age group . At present , pre-school aged children ( pre-SAC ) are not eligible for treatment with praziquantel [5] due to the absence of clinical data on the drug effects and safety in the very young . In the coming years a new formulation of praziquantel may be approved for very young children [6] . In areas of high transmission , WHO guidelines also recommend treatment of adults at risk [1] , [7] . By 2020 , WHO aims to increase coverage in areas of endemic infection such that 75% of SAC at risk will be regularly treated [2] , but progress to date in reaching this target has been poor in many regions . Currently WHO recommends using prevalence of infection in SAC to determine how often to treat in a given endemic area [1] . The recommended treatment strategy for schistosome infection is dependent upon whether the community has a low ( < 10% ) , moderate ( 10–50% ) or high ( ≥ 50% ) prevalence at baseline before the implementation of MDA . The strategy for low-risk communities is to treat all SAC twice during their primary schooling age , generally once every three years , and supply praziquantel in local health centres to treat suspected cases . For moderate-risk communities , the recommendation is to treat all SAC and at-risk adults once every two years . For high-risk communities , the recommended approach is to treat all SAC and at-risk adults once a year . At present in national NTD control programmes , schistosomiasis has one of the lowest levels of MDA coverage of all helminth diseases [8] , [9] . Given that MDA needs to be administered to individuals frequently , and that it does not provide long-term protection against the infection in the absence of a strong acquired immunological response to infection , a vaccine is ideally needed for control in the longer term . At present , there is no vaccine for use in humans that can protect against the schistosome infection . However , recent experimental studies by Afzal Siddiqui and colleagues on a candidate vaccine against Schistosoma mansoni infection in a baboon animal model have produced some encouraging results . In four independent , double-blinded studies , a Sm-p80-based vaccine exhibited potent prophylactic , anti-egg induced pathology and transmission-blocking efficacy against S . mansoni in the baboon ( Papio ursinus ) animal model [10] . The vaccine reduced female worm establishment by 93 . 45% and significantly resolved the major clinical manifestations of hepatic/intestinal schistosomiasis by reducing the tissue-egg load by 91 . 35% . A 40-fold decrease in faecal egg excretion by those few female parasites that established in the vaccinated animals , combined with a 79 . 21% reduction in hatching ability of eggs ( the release of viable miracidia ) , suggests the vaccine may have a high transmission blocking potential . The study showed comprehensive evidence for the effectiveness of a Sm-p80-based vaccine for schistosomiasis and provided support for the need to move beyond animal models to human studies . Based on the baboon experiments by Siddiqui and colleagues , and assuming efficacy would be similar in humans , published epidemiological analyses based on mathematical models have predicted that the Sm-p80-based vaccine could potentially block infection in areas of low and moderate transmission provided the duration of protection provided by the vaccine is 5 years or more [11] , [12] . These models were simple in structure and built on a deterministic framework . This study extends these analyses using an individual based stochastic model to look at the impact of a vaccine , with varying durations of protection , employed in different community-based vaccination programmes involving either vaccinating young children in a cohort-based approach or vaccinating the whole community across all age classes ) . Analyses are also presented of the impact on transmission and the prevailing levels of infection using either vaccination alone , MDA alone ( the current most commonly used intervention to control morbidity ) and or using both in different combinations . A description of the impact of MDA , alone on the prevalence and intensity of S . mansoni infection in various transmission settings , is covered in a series of recent publications , as is model structure , model assumptions and data sources for the key transmission and biological parameters [3] , [4] , [7] , [8] . The focus in the present analyses is on the relative merits of vaccination versus MDA , alone or in combination , as a tool for the community control of the morbidity induced by S . mansoni and the likelihood of transmission elimination .
Past work on the impact of MDA on Schistosoma mansoni has employed a hybrid deterministic model ( with deterministic and stochastic components ) based on sets of partial-differential equations to describe changes in the mean worm burden M ( t , a ) , for host a over time t [13–15] . Stylianou et al , developed an age independent deterministic model to explore the effect of community vaccination programmes [11] . We extend this deterministic model and develop an individual-based stochastic model ( an earlier version is described in [4] ) , where an individual of age a can be in one of the two categories; ( i ) unvaccinated group or ( ii ) vaccinated group , denoted by Nu ( a , t ) and Nv ( a , t ) respectively . We assume that the number of births is the same as the number of deaths ( constant size for the human host ) , hence the total population of age a , at time t is N ( a , t ) = Nu ( a , t ) +Nv ( a , t ) . The unvaccinated and vaccinated host dynamics can be described by the following system of partial differential equations ( PDEs ) : ∂Nu ( a , t ) ∂t+∂Nu ( a , t ) ∂a=−q ( a , t ) Nu ( a , t ) +ωNv ( a , t ) −μ ( a ) Nu ( a , t ) ( 1 ) ∂Nv ( a , t ) ∂t+∂Nv ( a , t ) ∂a=q ( a , t ) Nu ( a , t ) −ωNv ( a , t ) −μ ( a ) Nv ( a , t ) ( 2 ) Here q ( a , t ) is the fraction of the population of age a vaccinated at time t , ω=1durationofvaccineprotection is the vaccine decay rate and μ ( a ) is the host mortality rate . The vaccine candidate is assumed to act on the following variables [cf . Eqs ( 3 ) and ( 4 ) ]; ( i ) parasite establishment within the human host by reducing the rate of infection , β , ( ii ) parasite survival and growth within the human host , by reducing adult worm life expectancy , σ and ( iii ) reducing the rate of egg production , λ , due to a reduced growth rate in humans . We assume that the vaccine’s impact on worm death rate , eggs per gram ( EPG ) and age-specific contact rates are v1 , v2 and v3 respectively , where the values range from 0 to 1 . The total worm burden in the unvaccinated and vaccinated hosts are denoted by Mu and Mv and the changes in Mu and Mv , over time for host a are described by the following equations: ∂Mu ( a , t ) ∂t+∂Mu ( a , t ) ∂a=Lβ ( a ) Nu ( a , t ) −q ( a , t ) Mu ( a , t ) +ωMv ( a , t ) − ( μ ( a ) +σ ) Mu ( a , t ) ( 3 ) ∂Mv ( a , t ) ∂t+∂Mv ( a , t ) ∂a=Lv3β ( a ) Nv ( a , t ) +q ( a , t ) Mu ( a , t ) −ωMv ( a , t ) − ( μ ( a ) +v1σ ) Mv ( a , t ) ( 4 ) Here L represents the concentration of the infectious material in the environment , namely , how each individual of age a , contributes to the pool of released eggs . This is discussed in detail in [14] and [16] . It is assumed that the rates of turn over for the miracidia , snail intermediate host and cercaria are much faster ( life expectancies days to weeks ) than the adult worm in the human host ( life expectancy 4–6 years ) , so the dynamics of these life cycle stages are collapsed into the equations for the adult worms in humans as detailed in Anderson & May [15] . The total worm burden in the population is given by the sum of the total worm burden in the unvaccinated and vaccinated hosts . If we denote the total worm burden in the population as the sum of the total worm burden in the unvaccinated and vaccinated hosts by M ( a , t ) ¯=Mu ( a , t ) +Mv ( a , t ) and add Eqs ( 3 ) and ( 4 ) together we obtain the following , ∂M ( a , t ) ¯∂t+∂M ( a , t ) ¯∂a=Lβ ( a ) Nu ( a , t ) +Lv3β ( a ) Nv ( a , t ) −σM ( a , t ) ¯−μ ( a ) M ( a , t ) ¯ ( 5 ) In Eq ( 5 ) we have assumed v1 = 1 . We can express M ( a , t ) ¯ in terms of the mean worm burden , M ( a , t ) , as M ( a , t ) ¯=N ( a , t ) M ( a , t ) . Then we obtain; ∂M ( t , a ) ∂t+∂M ( t , a ) ∂a=Lv3β ( a ) Nv ( a , t ) +Lβ ( a ) Nu ( a , t ) N ( a , t ) −σM ( a , t ) ( 6 ) The egg output ( from the vaccinated and unvaccinated populations ) is given by E=ψL¯∫a=0∞{Nu ( a , t ) F ( Mu ( a , t ) Nu ( a , t ) ;λ ) +Nv ( a , t ) F ( Mv ( a , t ) Nv ( a , t ) ;v2λ ) }ρ ( a ) da ( 7 ) given dLdt=E−μ2L ( 8 ) where the death rate is that of infected snails . In the above equation ψ describes the flow of the infectious material into the reservoir while the function F ( M ( a , t ) ; λ ) generates the egg output as a function of mean worm burden and ρ ( a ) represents the age-specific relative contribution of infectious stages to the environmental reservoir . In our simulations we assume the host contribution to the reservoir to be the same as the age-specific contact rates , β ( a ) . This model has a full age structure for the human host where the outputs are grouped into three age categories , pre-SAC ( 0–4 years of age ) , SAC ( 5–14 years of age ) and adults ( 15+ years of age ) . We use these age groupings based on WHO definitions of treatment groups [1–3] to calculate the necessary coverage levels ( MDA or vaccination ) for each category in order to interrupt transmission . This is typically defined as the overall R0 <1 in infectious disease epidemiology , but as shown by Anderson and May [14] , the system of equations defined above has three possible equilibria; namely , a stable endemic state , an unstable boundary ( transmission breakpoint ) and a stable state of parasite extinction . This model is hybrid in the sense that assumes a negative binomial form for the distribution of parasite numbers per host with a fixed aggregation parameter k , density dependent fecundity , and assumed monogamous sexual reproduction among worms . The mean expected behavior of the individual based stochastic model is identical to the predictions of a deterministic version of the model . However , an individual-based stochastic model permits the examination of the probability distribution of a given event occurring , such as transmission elimination , in a defined period of time during which control measures are applied . Autopsy data show that worms tend to aggregate more in some individuals than in others , due to poorly understood factors such as environmental , social , host genetic or immunological effects [17] . Epidemiological studies also show that those heavily infected are predisposed to this state [18] . To take account of such effects in our model , individuals in each age category are assigned a contact rate drawn from a gamma distribution with shape parameter α , which , via compounding across individual distributions , leads to a negative binomial distribution of worms within the total host population . It is important to note that the aggregation parameter , k , within the stochastic model , fluctuates in value over time , as a result of changes in the mean worm burden . In the deterministic model k is held fixed in value . The stochastic model more accurately mirrors observed patterns where k tends to decrease in value as prevalence declines under the impact of control measures [19] . The egg contribution to the infectious reservoir depends on the age-specific contact rate for each individual and is governed by a deterministic formulation . Treatment events are predetermined , they occur at time tj and the time step to the next treatment event is randomly drawn from an exponential distribution . The rate parameter for this distribution is given by the overall rate that any event happens . Which event occurs is drawn at random , on the basis of the relative magnitude of each individual event relative to the combined rate of all events . Table 1 provides a description of these rates . In this paper we consider 15 years of MDA and vaccination administration . Most of the parameter values used in this paper are taken from within the ranges found in the literature ( Table 2 ) . However , the data for the age-specific contact rates of hosts within the infectious reservoir ( β ) and age-specific contribution of hosts to the reservoir are unknown . They are estimated by using MCMC method in parameter estimation from age intensity and prevalence curves as described in references detailed in the text and Table 2 . Precise details of the model fitting procedure are described in previous publications [4 , 14 , 15 , 17] . In the numerical evaluations of the model’s behavior ( stochastic simulations ) , we follow the WHO guidelines for the implementation of MDA . Starting with an untreated population , we administrate MDA over a 15-year period with coverage levels and treatment intervals based on the baseline prevalence . For low baseline prevalence in SAC , we treat once every 3 years; for moderate baseline prevalence in SAC , we treat once every 2 years and for high baseline prevalence in SAC , we treat once a year . The intensity of transmission is determined by R0 ( the basic reproductive number ) which varies for different baseline settings . When MDA alone is used as the treatment strategy , we simulate the following treatment strategies: ( i ) the WHO recommended treatment coverage of 75% SAC only; ( ii ) 60% of SAC only; ( iii ) 40% of SAC only and ( iv ) 85% of SAC and 40% of adults . In this paper we consider an ideal case-perfect vaccine , meaning that the rate of infection and the rate of egg production are essentially reduced by 100% , which is comparable to the efficacy of the Sm-p80 vaccine in the baboon model . This efficacy considers the prevention of worm establishment , the fecundity falling dramatically in those few worms that establish , and the inability of eggs from these worms to hatch and release viable miracidia . Vaccination is given annually to the children with the pre-specified age of administration , and the coverage levels depend on the age group that is treated and the duration of vaccine protection . In various experimental settings Sm-p80 has demonstrated robust antibody titres in baboons for up to 5–8 years [10] suggesting a reasonably long duration of protection . In this paper we simulate scenarios where ( i ) the vaccine gives a 5 year duration of protection ( from [10] ) and ( ii ) an ideal scenario where the vaccine gives a 20 years of protection which is longer than the duration of treatment ( 15 years ) . It should be noted here that the same results will be obtained for vaccines with a duration of protection longer than 20 years as we are only calculating the probability of achieving the WHO goals within 15 years of initiating vaccination . Also , it should be noted that the vaccine decay rate is given by 1/ ( duration of protection ) . Duration of vaccine protection has a direct impact on the vaccine administration schedule and the coverage levels required to have a significant impact . Here we consider the epidemiology of schistosome infections and the human host age-groups contributing most to parasite transmission . The aim is to cover children from ages 5–15 by vaccinating children in cohorts . We also analyze control strategies where the vaccine is given to younger children in their first year of life . The schistosomiasis vaccine will very likely be administered in conjunction with other vaccines already present in traditional immunization programmes ( HPV , DTP ) . Therefore , the achievable coverage will typically match that achieved for one of the other co-administered vaccines . Vaccination coverage in the first year of life ranges between 85% and 91% at global level and reduces significantly in the following years ( Table 3 ) . The coverage levels for school age children vary between 60% and 70% and for out of school individuals this range is 40%-50% [24–27] . Based on these coverage levels , for a vaccine that provides a 20-year protection against schistosomiasis , we vaccinate at age 1 ( early start ) or age 5 ( school start ) , with coverage levels of 85% and 60% respectively . For a vaccine that provides a 5-year duration of protection against infection , to ensure continuous protection , we vaccinate either at ages 1 , 6 and 11 with coverage levels 85% , 60% and 70% respectively , or at ages 5 , 10 and 15 with coverage levels 60% , 70% and 45% respectively . In this case ( 5-year duration of protection ) we have a 3-dose schedule of vaccination , similar to the HPV administration schedule . We consider MDA and vaccination , alone or in combination , as control strategies , where treatment is delivered at random at each round within the population with a given coverage . In other words , we do not consider individual compliance to treatment [19] in these analyses and just assume the individuals treated or vaccinated are chosen at random at each round . At the end of the treatment period , we calculate the probability of reaching WHO morbidity and elimination as a public health problem goal , by evaluating the fraction of SAC heavy-intensity infection prevalence ( ≤5% heavy-intensity infection in SAC for the morbidity goal and ≤1% heavy-intensity infection in SAC for the elimination as a public health problem goal ) . In our results we include the prevalence of infection ( population having egg count threshold > 0 ) and prevalence of heavy-intensity infections ( population having egg count threshold > 16 ) . The probability of reaching the 5% and 1% WHO goals are calculated as the fraction of repetitions that reach the target , by averaging across 300 simulations ( to ascertain the mean expectation of the stochastic model ) . A summary of the treatment strategies is presented in Fig 1 .
First MDA alone is examined as the treatment strategy , using the WHO targets for treatment of 75% coverage for SAC . The results are presented in Fig 2 and Table 4 . Model simulations ( based on the parameter values listed in Table 2 ) suggest that for low prevalence regions , the 5% morbidity goal in SAC can be achieved within 5 years of treatment , while the elimination as a public health problem goal in the total population can be achieved within 10 years of treatment . Similarly , for moderate-prevalence regions , the 5% morbidity goal in SAC can be achieved within 5 years of treatment , whereas the 1% elimination as a public health problem goal can be achieved within 15 year of MDA treatment . Again , both goals will be achieved within 15 years with a probability of unity . In high transmission regions , we can achieve the SAC 5% morbidity goal in 85% of the simulations . However , the 1% elimination as a public health problem goal in such high transmission ( large R0 values ) settings can be achieved in 35% of our simulations . In these settings , increasing the SAC coverage to > 75% and/or include other age bands in the treatment is highly desirable . In low to moderate transmission settings , using the recommended target coverage of 75% for SAC , the SAC 5% morbidity goal can be achieved within 5 years of MDA . Given the difficulties countries with endemic infection are experiencing in achieving this level of coverage , SAC coverages between 40% and 60% were also examined to explore if it is still possible to achieve the WHO goals with 15 years of MDA treatment . The impact of MDA decreases as SAC coverage declines as indicated in Table 4 . The SAC 5% morbidity goal can be achieved within 5 years at 60% SAC coverage ( in low to moderate settings ) . However , for the <1% heavy infection in the total population goal ( = elimination as a public health problem ) to be achieved within 15 years the probabilities of achieving this are 90% and 70% , respectively , in low and moderate transmission regions . Lowering the SAC coverage to 40% is predicted to achieve the WHO goals in low transmission settings . However , in moderate transmission settings , the SAC 5% morbidity goal can be achieved within 15 years of treatment with probability of 0 . 9 , but the 1% elimination as a public health problem goal is only achieved with probability 0 . 4 in that time . These results highlight the importance of using different MDA coverage levels in different transmission settings , as opposed to following the recommended 75% SAC coverage for all transmission levels . In stochastic ( and deterministic ) models ( and in the real world ) there is always a chance that the prevalence of infection will bounce back after control measures cease since in some simulation runs the breakpoint in transmission is not crossed . It is therefore important to analyze the probability of true elimination ( also known as ‘transmission interruption’ ) which results in the prevalence within the whole community in which control measures are introduced going to zero . As in previous studies [28] it is assumed that if the overall prevalence is less than 1% it is almost certain that transmission interruption has been achieved . We find that treating only 75% of SAC cannot interrupt transmission ( see Fig 2A , 2C and 2E ) , since the reservoir of untreated people in the adult age classes is able to seed the whole population once control ceases at year 15 . As discussed earlier , in high transmission settings it is necessary to treat both SAC and adults . Here we present the simulation results for the scenario 85% of SAC and 40% of adults are annually treated with MDA . These results are summarized in Fig 3 which shows that with this approach the WHO goals can be achieved , although the probability of complete elimination by year 15 is still low ( <0 . 3 ) . Longer durations of treatment and/or more frequent treatment are required to increase this probability . In this section , the effects of both vaccination coverage , and the average duration of protection provided by the vaccine , are examined . It should be noted that , based on the animal model results , we assume the vaccine is 100% efficacious . In the previous two sections it is shown that the WHO 5% morbidity control goals can be achieved in low to moderate transmission settings if either MDA alone or vaccination alone are administrated in endemic regions . However , these goals , particularly the 1% elimination as a public health problem goal , are unlikely to be achieved in high transmission settings . Whether it is beneficial to combine both treatments together is examined in this section . In practice , this is a likely scenario since MDA will remain the main control options for many years to come ( possibly 10 to 15 years ) even if Phase I , II and III trials in humans of the new vaccine go smoothly . The simulation results suggest that giving MDA to 75% of SAC and administrating vaccination with a wide range of coverage levels ( see Figs 6 and 7 , Tables 7 and 8 ) , can reach the 1% elimination as a public health problem goal in high settings with a probability of nearly 0 . 55 and 0 . 82 for vaccines with durations of protection of 20 and 5 years , respectively . The 5% SAC morbidity goal is achieved in all transmission settings . Therefore , a vaccine that provides 5 years of protection and covers three age groups , can achieve the WHO 5% morbidity control and 1% elimination as a public health problem goals . However , for a vaccine that provides 20 years protection we need to increase MDA and vaccination coverage levels , or include other age categories in the vaccination programme , to increase the probability of achieving elimination as a public health problem ( <1% ) in high transmission settings . However , do note that the short duration vaccine must be delivered to multiple age groups . Over 15 years an individual may need three vaccinations ( or 3 short courses of vaccination ) to maintain protection . As such costs and delivery may be important issues with a short duration of protection vaccine .
The results presented in this paper are very sensitive to the values of certain parameters . The two most important are the negative binomial aggregation parameter k and the magnitude of transmission before control measures are initiated ( the magnitude of R0 ) . Using k = 0 . 24 , λ = 0 . 24 in low transmission settings , the model cannot support endemic parasite populations when R0 is low . As a result , the model typically cannot reproduce endemic prevalences less than about 49% . The two possible causes are: ( i ) Diagnostic; due to poor sensitivity in the standard diagnostic test , measured prevalences may be much lower than the real values and ( ii ) model transmission structure; transmission may be confined to specific age groups as elimination is approached , giving a low community-level prevalence . To manage this limitation , we use k = 0 . 04 value for low transmission setting and k = 0 . 24 for moderate to high transmission settings . We have chosen the extreme baseline prevalences ( just below 10% for low transmission settings and just below 50% for moderate transmission settings ) . For these values there is a high probability to achieve the WHO goals and hence lowering the baseline prevalence does not alter the outcome . For a baseline prevalence between 50% and 58% ( high transmission settings ) we obtain qualitatively similar results with the ones produced in moderate settings . Therefore , for high transmission settings , we consider endemic regions with a baseline prevalence of around 62% ( R0 = 3 . 5 ) which is a realistic upper bound of prevalence for S . mansoni in most endemic regions [29] , [30] . In this study , we have used parameter values fitted to data collected in Iietune village in Kenya ( refer to Table 2 ) , but the same model and analysis can be used for other endemic regions . We should note here , that if the age-related contact rates and death rates are similar to the ones we have used , the results will be similar . If the prevalence of intensity is higher ( lower ) in SAC , the probability of achieving the WHO goals will be lower ( higher ) in these regions . These results are based on data for S . mansoni , but the analysis can be easily extended to S . haematobium . A possible key parameter in the analysis and not included in our study is the buildup of acquired immunity . To date , there aren’t enough evidences to show the presence of immunity in S . mansoni and we have assumed that the shape of age-intensity of infection is influenced only by rate of exposure to infection . It will be of great importance , in the future , to extend our model so that we can explore the effect of acquired immunity on morbidity .
Currently schistosome control strategies suggested by WHO and widely implemented in endemic regions include mass drug administration of school aged children and adults in high transmission settings . The primary goal is morbidity prevention in SAC or morbidity elimination in populations in areas of endemic infection . Snail control , snail habitat alterations and improving water , sanitation and hygiene ( WASH ) are also recommended ( there is little information on their efficacy ) , but MDA is the major route for morbidity control at present . In this paper , we have extended the individual based stochastic age structured model developed by Anderson and colleagues , which is constructed on the template of an age structured deterministic model [13–15] where its predictions have been validated using observed infection trends under defined levels of MDA in a number of field settings [31] . We specifically extend past work to include the effect of a vaccine on parasite establishment . The aim has been to explore the impact a vaccine with an efficacy of 100% might have on control efforts to attain the WHO goals for morbidity control in SAC and morbidity elimination in the total population ( but not infection ) . Different treatment and vaccination strategies have been considered in numerical analyses; namely: MDA alone , vaccination alone , or MDA plus vaccination combined . Analyses are conducted for three different transmission settings as defined by WHO on the basis of prevalence; low ( <10% baseline prevalence among SAC ) , moderate ( 10–50% baseline prevalence among SAC ) and high ( ≥50% baseline prevalence among SAC ) settings . These transmission conditions at baseline are determined by the magnitude of R0 , and , concomitantly , by the overall prevalence of infection and the average intensity of infection in defined community . We find that the optimal strategy to control or eliminate morbidity depends on the transmission setting , vaccine coverage level achieved , the duration of vaccine protection and the timeline of vaccination in different age groupings of the human host . In low prevalence settings , MDA alone or vaccination alone , with different levels of protection , can achieve the WHO goals with a probability of close to unity . Furthermore , our results show that treating just 40% of SAC with MDA alone can achieve the morbidity control goal and potentially elimination as a public health problem goal . This is an encouraging prediction considering the difficulties endemic regions are having in achieving the WHO recommended treatment coverage for SAC at 75% . In moderate prevalence settings , treating 60% of MDA can achieve the morbidity goal with probability of unity and possibly the elimination as a public health problem goal with probability of 0 . 7 . Increasing the SAC coverage to 75% increases the probability of elimination to 0 . 96 . Vaccination with a duration of protection of 5 years can achieve the morbidity control goal within 5 years of treatment and elimination as a public health problem goal within 15 years . However , a vaccine with a longer duration of protection ( 20 years ) achieves the morbidity goal with a probability of near unity , but the probability of elimination as a public health problem goal decreases to nearly 0 . 55 . In high transmission settings , we obtain the following outcomes: ( i ) the WHO recommended MDA treatment coverage for SAC at 75% can achieve the morbidity control goal with a probability of 0 . 85 , but there is only a 0 . 35 chance that we can achieve the elimination as a public health problem goal . ( ii ) Vaccinating 85% of 1-year olds with a vaccine that provides 20 years of protection , can achieve the morbidity control goal with probability of 0 . 61 , but it is very unlikely that the elimination as a public health problem goal will be achieved . ( iii ) increasing the vaccination coverage levels ( vaccinating 85% of age 1 , 60% of age 6 , 70% of age 11 or vaccinating 60% of age 5 , 70% of age 10 and 45% of age 15 ) and decreasing the duration of protection to 5 years , increases the probability of achieving the WHO goals . For the morbidity control this probability increases from 0 . 61 to 0 . 89 , while the probability of the elimination goal increases from 0 , 06 to 0 . 22 . Thus , in high transmission settings , vaccination alone or MDA alone cannot achieve the elimination as a public health problem target . We can modify this outcome by vaccinating across bands of age classes ( i . e . including adults ) . However , this may risk a high frequency of adverse effects due to past or present infection in vaccinated individuals . The best strategy in these circumstances is intensive MDA plus vaccination . Treating 75% of SAC with MDA and vaccinating 60% of age 5 , 70% of age 10 and 45% of age 15 ( duration of vaccine protection is 5 years ) can achieve the morbidity control goal with probability of unity and the elimination as a public health problem with a probability of nearly 0 . 84 . Alternatively , increasing the SAC coverage to 85% and including 40% of adults in the treatment plan , could achieve the WHO goals with a high probability . This outcome is in line with previous results found in [3] , which has reported that including adults in the treatment strategy and increasing SAC coverage levels can lower the prevalence of heavy-intensity to below 1% in SAC . Analysing the vaccine’s administration schedule ( early start versus starting vaccination on entry to school ) , vaccinating 5-year olds may arguably be an easier strategy to implement than vaccinating 1-year olds . Relatively few individuals will have become infected by 4 years of age , but some have . The main argument in favour of the latter age is that it is easier to reach children for vaccination via school infrastructure/attendance . Alternatively , if the vaccine is safe for very young children ( < 1 year of age ) then the vaccine could just be part of the national immunization schedule for infants and young children . The other benefit of vaccinating at age 1 is to avoid morbidity induced by early infection in infancy . Given the long duration of vaccine protection , model simulations suggest little difference between the two strategies . This suggests that programmatic and cost issues will be most important in public health policy formulation for the use of the vaccine . Comparing vaccination with a long duration of protection and MDA alone , we find that good coverage of MDA across bands of age classes ( i . e . SAC ) is predicted to have a greater and quicker impact than cohort immunization in all settings . However , we have used different coverage levels between these two treatment strategies with less people being vaccinated than are treated with MDA . On the other hand , a vaccine with a shorter duration of protection performs better ( in terms of achieving the WHO goals ) because we are treating more age groups . Unless true elimination of transmission is achieved , treatment should not cease as there is a chance that the prevalence of infection will bounce back after cessation . True elimination is not achieved in any of the scenarios considered . Unless the treatment frequencies and coverage levels are increased considerably from the scenarios examined it is very unlikely that this goal will be achieved . Factors such as individual adherence to treatment is not taken into consideration and we have assumed a random treatment adherence at each round for a given coverage level . The simulations may therefore be on the optimistic side since a proportion of the chosen individuals for a given coverage are likely to be nonadherent over many rounds of MDA [13] , [21] , [23] , [32] , [33] . It will be of great importance to have the relevant adherence data to make more accurate predictions . The predictions presented in this paper depend on the assumptions made concerning the precise nature of the manner in which the intensity of infection varies by age in a given endemic region , the magnitude of R0 ( = transmission intensity ) reflected by the baseline prevalence prior to the introduction of control measures . It will be harder to achieve the WHO targets if infection in the very young ( pre-SAC ) and adults is high . We have used data for S . mansoni but the same methods of analysis can be applied for S . haematobium infection . In summary , vaccination alone or in combination with MDA , proves to be an effective method to control or eliminate schistosomiasis as a public health problem . Achievement of the WHO goals for morbidity control and elimination depends on vaccine efficacy , on the duration of vaccine protection and on the coverage levels achieved in different age classes .
|
Nearly 258 million people are infected worldwide by schistosome parasites . The World Health Organization ( WHO ) has set control guidelines to combat the morbidity and mortality induced by infection , defined by reaching ≤5% and ≤1% prevalence of heavy-intensity infections in school-aged children ( SAC ) , respectively . Mass drug administration ( MDA ) is the major route for morbidity control and elimination . However , MDA does not provide long-term protection against schistosome parasites and frequent drug administration is therefore required to control morbidity . Infection does not induce lasting acquired immunity to reinfection . Drug resistance is another issue with MDA which , if it arises , could possibly make drug treatment ineffective over time as drug-resistant genes in the parasite population increase in frequency . A vaccine is ideally needed to both reduce the possibility of reinfection and to achieve transmission elimination within a feasible time frame . Based on the recent results obtained for a new candidate vaccine in the baboon animal model , we employ an individual-based stochastic model to assess the impact of a vaccine with an efficacy of 100% when applied in endemic regions with different intensities of transmission . Simulations suggest that the probability of achieving morbidity control and elimination as a public health problem depends on the duration of protection provided by vaccination , the age categories of the human host population vaccinated , and the coverage levels achieved . In order to achieve elimination as a public health problem , model simulations suggest that combining vaccination ( with 5 years of protection ) with MDA ( treating 75% of school-aged children , 5–14 years of age ) is the best option , particularly in high transmission settings .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Sensitivity",
"analysis",
"and",
"model",
"limitations",
"Discussion"
] |
[
"schistosoma",
"invertebrates",
"schistosoma",
"mansoni",
"medicine",
"and",
"health",
"sciences",
"helminths",
"immunology",
"vertebrates",
"parasitic",
"diseases",
"animals",
"mammals",
"health",
"care",
"vaccines",
"preventive",
"medicine",
"age",
"groups",
"primates",
"infectious",
"disease",
"control",
"vaccination",
"and",
"immunization",
"morbidity",
"old",
"world",
"monkeys",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"monkeys",
"baboons",
"people",
"and",
"places",
"health",
"statistics",
"eukaryota",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"amniotes",
"organisms"
] |
2019
|
Modelling the impact of a Schistosoma mansoni vaccine and mass drug administration to achieve morbidity control and transmission elimination
|
Microsporidia comprise a phylum of over 1400 species of obligate intracellular pathogens that can infect almost all animals , but little is known about the host response to these parasites . Here we use the whole-animal host C . elegans to show an in vivo role for ubiquitin-mediated response to the microsporidian species Nematocida parisii , as well to the Orsay virus , another natural intracellular pathogen of C . elegans . We analyze gene expression of C . elegans in response to N . parisii , and find that it is similar to response to viral infection . Notably , we find an upregulation of SCF ubiquitin ligase components , such as the cullin ortholog cul-6 , which we show is important for ubiquitin targeting of N . parisii cells in the intestine . We show that ubiquitylation components , the proteasome , and the autophagy pathway are all important for defense against N . parisii infection . We also find that SCF ligase components like cul-6 promote defense against viral infection , where they have a more robust role than against N . parisii infection . This difference may be due to suppression of the host ubiquitylation system by N . parisii: when N . parisii is crippled by anti-microsporidia drugs , the host can more effectively target pathogen cells for ubiquitylation . Intriguingly , inhibition of the ubiquitin-proteasome system ( UPS ) increases expression of infection-upregulated SCF ligase components , indicating that a trigger for transcriptional response to intracellular infection by N . parisii and virus may be perturbation of the UPS . Altogether , our results demonstrate an in vivo role for ubiquitin-mediated defense against microsporidian and viral infections in C . elegans .
The Microsporidia phylum contains over 1400 species of obligate intracellular pathogens most closely related to fungi [1] . These pathogens can infect a wide variety of animal hosts including humans , where they can cause significant disease . Infections in humans can cause lethal diarrhea in immunocompromised people such as AIDS patients , and microsporidia are considered priority pathogens at the National Institutes of Health [2] , [3] . Microsporidia can also plague agriculturally significant animals such as fish and honeybees [4] , [5] , [6] . Treatment options for microsporidia infections are limited and often ineffective [7] , [8] . In mammals , studies have shown that T cells and dendritic cells provide protection against infection , but little is known about the innate and/or intracellular responses to these pathogens [9] , [10] , [11] . Previously , we described Nematocida parisii , a microsporidian species isolated from a wild-caught C . elegans near Paris , which causes a lethal intestinal infection in its host [12] , [13] . N . parisii infection of the simple nematode C . elegans provides a convenient system in which to investigate host responses and defense against microsporidia infection . Interestingly , canonical C . elegans defense pathways , such as the conserved PMK-1 p38 MAPK pathway that provides defense against bacterial and fungal infections , are not important for defense against N . parisii [12] , [14] . Thus , distinct immunity mechanisms may be involved in the C . elegans response to microsporidia . In addition to microsporidia , another natural intracellular infection has recently been described in C . elegans: wild-caught animals from Orsay , France , were shown to harbor a viral infection [15] . The Orsay virus is a positive strand RNA virus of the family Nodaviridae , and like N . parisii it appears to undergo its entire replicative cycle inside C . elegans intestinal cells . The RNAi pathway has been shown to provide defense against viral infections in C . elegans [15] , [16] , [17] , [18] , [19] , but little else is known about host defense against this natural intracellular pathogen of C . elegans . Defense against intracellular pathogens in diverse animal hosts is increasingly appreciated to involve ubiquitin-mediated degradation pathways [20] , [21] , [22] , [23] . Ubiquitylation is the process by which an E3 ubiquitin ligase catalyzes the conjugation of a ubiquitin tag onto substrates , which can be further ubiquitylated to generate poly-ubiquitin chains [24] . Ubiquitylated substrates have a number of different fates , two of which involve degradation . The most well characterized fate is degradation by the proteasome , but larger substrates can be targeted for degradation by the process of autophagy , which is termed 'xenophagy' when it involves degradation of intracellular microbes [25] , [26] . Recently , ubiquitin ligases that mediate ubiquitin targeting to human bacterial pathogens Salmonella enterica [21] and Mycobacterium tuberculosis [22] have been identified , and they , together with the autophagy pathway , are important for controlling levels of these intracellular pathogens [23] , [27] , [28] , [29] . However , while several ubiquitin-mediated defense components and mechanisms have been defined , there are many unanswered questions about which host ubiquitin ligases are involved in targeting ubiquitin to different pathogens , how these systems are regulated , and their overall importance for defense in vivo . One major class of E3 ubiquitin ligases includes the Skp1−Cul1−F-box protein ( SCF ) multi-subunit RING-finger type , which is a modular complex found throughout eukaryotes [30] . SCF ligases are usually composed of three core components ( a cullin protein , Skp1 , and a RING-containing subunit ) and a variable F-box protein component , which enables recognition of different substrates depending on which F-box protein is associated with the complex [31] . Interestingly , the C . elegans genome has a greatly expanded and diversified family of F-box proteins ( ∼520 genes compared to 69 genes in humans ) , as well as other SCF components ( 21 Skp1-related genes compared to 1 in humans ) , suggesting they use SCF ligases to recognize an extremely diverse array of substrates [32] , [33] . In particular , it has been proposed that C . elegans uses these SCF ligases to target toxins and intracellular pathogen proteins for degradation , and that the expanded C . elegans SCF ligase repertoire is the manifestation of a host/pathogen arms race between nematodes and their natural intracellular pathogens [32] . At the time this intriguing idea was proposed however , there were no known intracellular pathogens of C . elegans to test the role of ubiquitin-mediated responses in defense . Here we describe the C . elegans host response to the natural intracellular pathogens N . parisii and the Orsay virus , and find a role for ubiquitin-mediated defense against both infections . We perform gene expression analyses of the transcriptional response to microsporidia infection and find that the response is strikingly similar to the response to viral infection , but not to extracellular pathogens . We see upregulation of SCF ligase components , which help to restrict microsporidia growth , and find that defense against microsporidia appears to rely on the proteasome , as well as the autophagy pathway . We find a subset of parasite cells targeted by host-derived ubiquitin , which relies partly on the SCF cullin component CUL-6 . Notably , this ubiquitin targeting , as well as the role for ubiquitin-mediated defense , increases upon inhibition of microsporidia growth by anti-microsporidia drugs . These results suggest that N . parisii may suppress or evade ubiquitin-mediated host defenses . Interestingly , expression of specific infection-upregulated SCF ligase components is also upregulated by genetic or pharmacological inhibition of UPS function , suggesting that stress placed upon the UPS may be a hallmark of intracellular infection , and that hosts monitor UPS function to upregulate appropriate defenses during intracellular infection . Finally , we show that SCF ligase components , in particular CUL-6 , promote defense against viral infection in C . elegans . Altogether , these studies show the involvement of ubiquitin-mediated defense and xenophagy against natural intracellular pathogens in a whole animal host , and provide insight into their regulation in response to infection in vivo .
We examined the C . elegans transcriptional response over the course of an infection with N . parisii using strand-specific deep sequencing of RNA ( RNA-seq ) . Like other microsporidia , the life cycle of N . parisii is complex and its growth and replication takes place entirely inside the host cell ( Figure 1A ) . Microsporidian spores initiate an intracellular infection by firing an infection apparatus called a polar tube , which pierces the host cell membrane and then injects into the host cell a nucleus and sporoplasm , which replicates as a stage called a meront . In the case of N . parisii , meronts become very large , multi-nucleate cells that replicate in direct contact with the cytoplasm . Meronts will eventually differentiate into spores and these spores then exit from infected cells to infect new hosts . We collected and sequenced cDNA from age-matched uninfected controls and infected animals at 8 , 16 , 30 , 40 and 64 hours post inoculation ( hpi ) ( Figure 1A , B ) , which are timepoints that correspond to specific stages of N . parisii infection as described in our previous study [34] ( Table S1 ) . A large number of C . elegans genes had significantly altered expression during N . parisii infection ( edgeR , FDR<0 . 05 , Table S2 ) . The overall number of upregulated genes was relatively stable throughout infection , while the number of downregulated genes increased markedly with time ( Figure 1C ) . To validate our RNA-seq studies , we also performed Affymetrix microarrays , which had substantial agreement in the genes found to be regulated by N . parisii infection ( see Supplemental Text S1 , Table S3 ) . Notably , we found that a significant number of genes upregulated by infection were associated with the intestine , which is the site of N . parisii infection ( Figure 1D ) . Next , we compared genes regulated by N . parisii ( Table S4 ) to gene sets regulated by infection with other pathogenic microbes , by treatment with non-biotic stressors , and by known immunity and stress-response pathways in C . elegans [17] , [35] , [36] , [37] , [38] , [39] , [40] , [41] ( Figure 1E , Table S5 , Table S6 ) . Here , we used a well-established analytical method called Gene Set Enrichment Analysis ( GSEA ) , which analyzes gene expression data at the level of gene sets instead of individual genes ( see Materials and Methods ) [42] . We found limited but significant correlations with gene sets upregulated by heat shock treatments , the pore-forming toxin Crystal protein-5B ( Cry5B ) , and Drechmeria coniospora fungal infection , predominantly at the 30 hpi timepoint ( Figure 1E ) . The heat shock pathway has been shown to play a role in resistance to bacterial pathogens as well as other stresses [43] , [44] . However , despite the overlap between genes induced by heat shock and microsporidia , we found that N . parisii infection upregulated only two canonical heat shock protein-encoding genes , hsp-17 at 30 hpi and hsp-16 . 1/hsp-16 . 11 ( which have identical sequence and are indistinguishable in RNA-seq data ) at 64 hpi ( Table S7 , Figure S1A ) . Notably , there was almost no correlation between C . elegans genes upregulated in response to N . parisii infection compared to infections with the extracellular bacterial pathogens Pseudomonas aeruginosa and Staphylococcus aureus , the fungal pathogen Harposporium , or to genes affected by known C . elegans immunity regulators ( Figure 1E ) . However , there was extensive correlation between genes downregulated by N . parisii and genes downregulated by other pathogens - for further discussion of this correlation , and other comparisons to previously published gene expression analyses see Supplemental Text S1 and Figure S2 . Strikingly , we found a very strong correlation between genes most strongly upregulated by N . parisii , ( e . g . genes of unknown function C17H1 . 6 and F26F2 . 1 ) and genes upregulated by viral infection ( Figure 1E , Figure S1B ) . Thus , N . parisii induces robust gene expression changes that are largely distinct from changes induced by extracellular pathogens , but share similarity to changes induced by the Orsay virus , which is another natural intracellular pathogen of C . elegans . To understand the nature of the C . elegans response to microsporidia infection , we analyzed the enrichment of gene ontology ( GO ) and Kyoto Encyclopedia of Genes and Genomes ( KEGG ) terms for the significantly induced and repressed genes [45] , [46] ( Table 1 and S8 ) . Early during infection upregulated genes were enriched for GO terms associated with regulation of growth , while at later timepoints they were enriched for GO terms associated with the nucleosome , defense response , and structural components . At 30 hpi , upregulated genes were enriched for association with the ubiquitin-mediated proteolysis KEGG pathway ( Table 1 ) . To extend our analysis , we also identified specific enrichment of Pfam protein domains among N . parisii regulated genes ( Table 1 and S8 ) . At early times following infection these included two Caenorhabditis domains of unknown function , DUF713 and DUF684 . Notably , genes upregulated at 8 , 16 and 30 hpi were also enriched for the F-box , FTH ( fog-2-homology ) , and MATH ( meprin and Traf homology ) protein-protein interaction domains , which are domains associated with ubiquitin-mediated proteolysis . For more details on regulated proteins containing these domains , analysis of gene enrichment at later time points , and analysis of downregulated genes , see Supplemental Text S1 . Previously it had been hypothesized that F-box and MATH domain-containing proteins could function in C . elegans to target foreign pathogen proteins for proteasomal degradation , as part of SCF multi-subunit E3 ubiquitin ligases [32] . Indeed , we found that C . elegans SCF ligase components , Skp1-related ( skr ) genes skr-4 and skr-5 , were significantly upregulated at 30 hpi with N . parisii ( Table S2 ) , while skr-3 and the cullin gene cul-6 were also upregulated at 30 hpi over 6 . 5- and 5 . 5-fold respectively , although the difference was not significant ( Table S4 ) . While these SCF ligase components were not reported to be significantly upregulated in a published dataset of the wild-type C . elegans response to viral infection [17] , we found that in the virus-susceptible rde-1 strain of C . elegans , the SCF ligase components cul-6 , skr-3 , skr-4 , and skr-5 were upregulated in response to viral infection ( data not shown ) . Overall , this increased expression of genes encoding SCF ligase components ( see Table S9 for list of significantly upregulated ubiquitylation-associated genes ) is consistent with ubiquitylation being upregulated in virus and microsporidia-infected animals . To examine a functional role for genes induced by N . parisii infection we used RNAi to knock-down expression of specific genes , then infected these animals with N . parisii and measured pathogen load at 24 hpi by quantifying N . parisii rRNA FISH signal ( Figure 2A , S3 ) . We tested several genes highly induced by infection , as well as genes that belong to gene classes identified through our GO term and Pfam domain analysis . Knock-down of most genes showed little to no effect on pathogen load ( see Supplemental Text S1 and Figure S4 ) . When we examined whether the upregulated SCF ligase components have a functional role in defense against N . parisii infection we found a more substantial role . Because there are a large number of F-box proteins in the C . elegans genome ( ∼520 proteins ) , we focused on the core SCF ligase components that belong to smaller families , namely the Skp1-related skr family ( 21 proteins ) and the cullin family ( 6 proteins ) , which likely have less functional redundancy than F-box proteins . In particular , we knocked down expression of cul-6 , skr-3 , skr-4 and skr-5 because these were upregulated upon N . parisii infection . Here , we found a modest but significant increase in pathogen load in cul-6 , skr-3 and skr-5 RNAi-treated animals ( Figure 2B ) , suggesting that these SCF ligase components limit the growth of N . parisii during infection . After substrates have been ubiquitylated by ubiquitin ligases , they can either be degraded by the proteasome or by the autophagy pathway . First , we examined whether components of the proteasome may be acting downstream of SCF ligase components in limiting growth of N . parisii . We reduced expression of ubiquitin itself with RNAi against ubq-2 , as well as two components of the proteasome: pas-5 and rpn-2 . In order for animals to develop properly , we introduced the RNAi in a diluted form at a late larval stage , infected animals and measured pathogen load . We found that reducing expression of any of these three genes led to an increase in pathogen load , suggesting that the UPS is important for defense against N . parisii ( Figure 2C ) . Because the effect of ubiquitin knock-down on pathogen load was modest , we hypothesized that , like other intracellular pathogens , N . parisii may suppress this defense system or subvert some aspects of the UPS to promote its replication . To test this hypothesis , we treated animals with drugs that block N . parisii growth but have minimal effects on adult C . elegans ( see Supplementary Text S1 ) [47] , [48] . First , we treated animals with a low dose of the anti-microsporidia drug fumagillin [49] , [50] , [51] , [52] , which limits N . parisii growth ( Figure 2D and data not shown ) . After fumagillin treatment we found that ubq-2 RNAi had a more robust effect on pathogen load ( 150% increase ) than in the absence of this drug ( 50% increase ) ( Figure 2D ) . Similarly , ubq-2 RNAi had a stronger effect on pathogen load when N . parisii growth was repressed with a DNA synthesis inhibitor , FUdR , ( 320% increase ) than in the absence of this drug ( 70% increase ) ( Figure 2E ) . Taken together , these results suggest that the host UPS plays a greater role in controlling infection when pathogen growth is inhibited . We next investigated a role for the autophagy pathway in response to N . parisii infection . We used RNAi to knock-down expression of different autophagy components , infected these animals with N . parisii , and quantified pathogen load . Similar to the effects of knocking down components of the UPS , we found a modest but significant increase in pathogen load when expression of several key autophagy components was reduced ( Figure 2F ) . Furthermore , RNAi of the C . elegans nutrient sensor TOR ( Target Of Rapamycin ) ortholog let-363 , which activates autophagy in C . elegans [53] , caused a dramatic 70% decrease in pathogen load ( Figure 2G ) . To determine whether autophagy machinery was directed toward N . parisii cells , we examined localization of GFP-tagged LGG-1 ( homolog of Atg8/LC3 in yeast/mammals ) [54] , a protein whose distribution is often used to assess autophagy [55] . We found that early during infection only 7% of parasite cells ( 25/360 parasite cells , n = 6 animals ) were targeted by GFP::LGG-1 ( Figure 3A–C ) . When animals were treated with let-363 RNAi , we found that there was a greater than 2-fold increase in parasite cells targeted by GFP::LGG-1 ( Figure 3D ) , consistent with this treatment causing an upregulation of the autophagy machinery directed toward N . parisii cells . Thus , the autophagy machinery appears to be targeted to N . parisii cells , and promotes resistance against infection . One potential caveat to the results described above is that specific RNAi treatments might affect the feeding rates of nematodes , which could then result in changes in pathogen load simply due to differences in the initial dose of N . parisii spores ingested by these animals . To address this concern , we measured the accumulation of fluorescent beads in the intestinal lumen of animals fed dsRNA against the genes described above ( Figure S5A–C ) . Importantly , RNAi against let-363/TOR , which causes decreased pathogen load , did not cause a decrease in the accumulation of fluorescent beads . In addition , RNAi against most autophagy genes that caused increased pathogen load did not cause an increase in accumulation of fluorescent beads . Furthermore , UPS RNAi , which increases pathogen load , did not increase fluorescent bead accumulation , and to the contrary , knock-down of ubq-2 or pas-5 marginally inhibited accumulation . Finally , feeding rates as measured by pharyngeal pumping were not affected by RNAi treatments , with the exception of ubq-2 RNAi , which caused a decrease in feeding ( Figure S5D–F ) . For further details on these controls , see Supplemental Text S1 . Altogether , our data support the model that defense against N . parisii infection involves ubiquitylation components , the proteasome , and the autophagy pathway , although microsporidia appears to partially evade or suppress this ubiquitin-mediated response . To examine whether N . parisii itself is targeted by ubiquitin , we stained infected animals with the FK2 antibody , which recognizes ubiquitin that is conjugated to a substrate , and with a FISH probe against N . parisii rRNA to label the pathogen . Because N . parisii is a eukaryote , it contains its own ubiquitin , which is recognized by the FK2 antibody . However , distinct from this staining , we observed very strong accumulation of conjugated ubiquitin surrounding a subset of N . parisii meronts , with signal far above background of the microsporidia-derived ubiquitin ( Figure 4A ) . To confirm that this ubiquitin was host-derived , we created a transgenic C . elegans strain that expresses a GFP::ubiquitin fusion protein under the control of an intestinal-specific promoter . Using these transgenic animals , we observed targeting of GFP::ubiquitin to parasite cells ( Figure 4B ) . In contrast , we did not observe significant targeting to parasite cells by a conjugation defective GFP::ubiquitinΔGG fusion protein ( Figure 4C , Figure S6 ) . Altogether these experiments demonstrate that host ubiquitin is specifically targeted to N . parisii cells , where it is conjugated to a substrate . The percentage of N . parisii cells specifically targeted by ubiquitin was relatively low: using the FK2 antibody we found only about 5% of pathogen cells were targeted by ubiquitin at 12 hpi ( Figure 4D ) . Similarly , we found only about 7% of pathogen cells were targeted by GFP::ubiquitin ( Figure 4E ) . Therefore , we examined whether N . parisii is suppressing or evading ubiquitin targeting by the host . If so , inhibiting the growth/vigor of the pathogen should cause an increased level of ubiquitin targeting . Indeed , we found increased targeting of ubiquitin to parasite cells after fumagillin treatment , with 16–18% of cells targeted ( Figure 4D , E ) . This effect was dose-dependent , and was apparent both with the FK2 antibody , as well as the GFP::ubiquitin fusion protein . These results support the hypothesis that N . parisii is actively suppressing or evading ubiquitin targeting by C . elegans , and that after inhibition of N . parisii growth with an anti-microsporidia drug , the host is better able to target pathogen cells with ubiquitin . Because the SCF ubiquitin ligase components cul-6 , skr-3 and skr-5 serve to limit N . parisii growth ( Figure 2B ) we hypothesized that they could be responsible for ubiquitin targeting of parasite cells . Thus , we examined ubiquitin targeting to N . parisii cells in animals that had been treated with cul-6 RNAi compared to the RNAi control ( Figure 4F ) . Indeed , we found that cul-6 RNAi had significantly reduced targeting of ubiquitin to N . parisii cells ( two-tailed unpaired t-test , p<0 . 05 ) . Thus , cul-6 is important for efficient ubiquitylation of parasite-associated proteins , suggesting that cul-6-containing SCF ligases may mediate recognition of N . parisii infection by the host . The ubiquitin targeting of parasite cells described above was only observed at early timepoints of infection , when pathogen cells were small and mono-nucleate . When the pathogen cells grew bigger and became multi-nucleate meronts , we observed virtually no parasite cells targeted by ubiquitin or by autophagy ( data not shown ) . Similarly , once meronts have differentiated into spores at later stages of infection , we found exceedingly few spores targeted by ubiquitin ( Figure 5A ) . Although there was virtually no specific ubiquitin targeting to the parasite at these later stages of infection , we did observe an increased number of clusters of ubiquitylated proteins ( Figure 5B–F ) . These clusters were dispersed throughout the infected intestinal cells , but in some cases were closely associated with N . parisii , although not encircling the parasite cells ( Figure 5D ) . In addition , we found that infection caused increased clustering of the autophagy marker GFP::LGG-1 in regions distinct from the pathogen cells ( Figure S7A–C ) and found that GFP::LGG-1 partially colocalized with ubiquitylated protein clusters ( Figure S7E–F ) . In order to determine whether this is a specific response , we examined GFP::LGG-1 upon infection with the extracellular bacterial pathogen P . aeruginosa , and did not find a significant increase in clustering ( Figure S7D ) . Thus , as infection proceeds , an increased amount of conjugated ubiquitin and GFP::LGG-1 clusters accumulate in the host cytosol , and these markers are almost never seen specifically surrounding the pathogen cells . Recent studies have indicated that host cells monitor the functioning of core processes that are commonly perturbed by pathogen infection and that disruption of these processes can trigger defense-related gene expression by the host [56] , [57] , [58] , [59] , [60] . Because intracellular infection by N . parisii leads to an increase in ubiquitylated protein clusters , which may reflect an increase in demand on the UPS , we investigated whether perturbation of the UPS might be responsible for inducing gene expression changes upon N . parisii infection . To conveniently monitor gene expression in vivo and to examine where genes are induced upon N . parisii infection , we made promoter-GFP fusions for C17H1 . 6 and F26F2 . 1 , two genes of unknown function that are among the most highly upregulated genes at all infection timepoints ( eg . at 8 hpi , C17H1 . 6 and F26F2 . 1 are upregulated 1 . 2×1011- and 1441-fold , respectively ) ( Table S2 , S3 ) . Expression of GFP driven by promoters of these genes was strongly induced in intestinal cells of infected animals by 8 hpi and even more robustly by 24 hpi ( Figure 6A ) . These GFP reporters indicated that N . parisii infection drives expression of genes in intestinal cells of infected animals and provided convenient tools for monitoring expression of infection response genes . To disrupt UPS function , we first performed RNAi knock-down of ubiquitin , pas-5 and rpn-2 in C17H1 . 6p::gfp and F26F2 . 1p::gfp transgenic animals . Strikingly , RNAi against the UPS components dramatically induced GFP expression in the intestine in both of these strains ( Figure 6B ) . To confirm these results we performed qRT-PCR and saw levels of endogenous C17H1 . 6 and F26F2 . 1 mRNA transcripts also increased by UPS RNAi ( Figure 6C ) . To perturb UPS function pharmacologically , we used the proteasome inhibitor MG-132 and similarly saw that this led to dramatic increase in C17H1 . 6 and F26F2 . 1 expression ( Figure S8A–C ) . Because C17H1 . 6 and F26F2 . 1 are genes of unknown function , we extended these analyses to genes upregulated by intracellular infection that have predicted function , namely the genes that encode the SCF ubiquitin ligase components skr-3 , skr-4 , skr-5 and cul-6 ( Figure 6C , Figure S8C ) . Similar to other infection response genes , we found that these genes were also induced by RNAi against the UPS , while another SCF component , skr-1 , whose expression was not altered during microsporidia infection , was not affected ( Figure 6C , Figure S8C ) . Thus , C . elegans appears to monitor efficacy of the UPS , and when this core process is disrupted it can trigger expression of a number of specific genes , including SCF components such as cul-6 that are used by C . elegans to limit intracellular infection . The C . elegans gene expression response to N . parisii was most similar to its response to viral infection , including the upregulation of SCF ligase components ( Figure 1E , Table S5 , Table S6 ) . Because of this similarity , we investigated whether the SCF ligases implicated in response to N . parisii also played a role in response to viral infection . Indeed , we found that cul-6 RNAi caused a 13-fold increase in viral load , and skr-3 and skr-4 RNAi caused 5- and 4-fold increases in viral load respectively ( Figure 7A ) , indicating that these SCF ligase components promote anti-viral defense . However , contrary to N . parisii infection , global inhibition of the UPS by RNAi-mediated knockdown of UPS components drastically reduced viral replication ( Figure 7B ) . Many viruses exploit host UPS in order to replicate , for example to degrade host RNAi and immune signaling machinery , or to control function and stability of viral proteins [61] , [62] , [63] , [64] , [65] , and thus the Orsay virus may likewise be hijacking this host pathway . Importantly , this result also suggests that the increased susceptibility to N . parisii infection of UPS-compromised nematodes ( Figure 2C ) is not likely just a result of general 'sickness' in these animals . Thus , the UPS appears to play two different roles in response to the Orsay virus , involving an unknown ligase ( s ) that promotes susceptibility to viral infection , and the cul-6 , skr-3 and skr-4 SCF ubiquitin ligases promoting anti-viral defense . Because N . parisii infection caused increased clustering of ubiquitylated proteins in C . elegans intestine , and robust gene expression changes in response to infection appeared to be a reflection of increased demand on the UPS , we investigated whether similar host responses occurred upon viral infection . Indeed , we found that infection with Orsay virus caused clustering of ubiquitylated proteins ( Figure 7C ) , and clustering of GFP::LGG-1 ( Figure S7A–D ) . Thus , infection with the Orsay virus induces similar cell biological changes as N . parisii infection . Furthermore , we found that viral infection induced the GFP reporter F26F2 . 1p::gfp ( Figure 7C ) , which is also induced when the UPS is perturbed . Thus , it appears that the C . elegans transcriptional response to viral infection , like the response to N . parisii infection , involves surveillance pathways that detect perturbation of the UPS caused by infection , to upregulate defense gene expression .
Based on our results we propose a model for the C . elegans intestinal response to intracellular infection ( Figure 8 ) , which highlights an important role for ubiquitin-mediated defense . In response to N . parisii infection , C . elegans upregulates expression of SCF ligase components , which restrict growth of the microsporidian pathogen N . parisii , as well as the Orsay virus . Restriction of N . parisii growth appears to also depend on the proteasome , as well as the autophagy pathway . While SCF ligase components such as CUL-6 have a substantial role in restricting growth of the virus , their more modest role in defense against N . parisii may be due to functional redundancy and/or the relatively inefficient targeting of ubiquitin to this pathogen . Inefficient targeting may be a result of suppression or evasion of host defenses by the parasite , as we find increased ubiquitin targeting of pathogen cells and a greater role for ubiquitin-mediated defense after treatment with drugs that inhibit N . parisii growth . Furthermore , we observe an increase in autophagy machinery targeting to N . parisii cells after activation of autophagy by inhibition of the TOR pathway . Interestingly , the increased demand on the UPS caused by intracellular pathogens like N . parisii and the Orsay virus may induce gene expression in response to infection , because genetic or pharmacological perturbation of the UPS upregulates expression of SCF ligase components and other genes that are induced by these intracellular infections . SCF ligases comprise one of the major classes of E3 ubiquitin ligases that catalyze transfer of ubiquitin onto substrates . These ligases have very well characterized roles in controlling levels of endogenous proteins that regulate the cell cycle and development . Intriguingly , the expanded and diversified repertoire in C . elegans and plants of ubiquitin ligase adaptors such as F-box and BTB-MATH domain proteins , as well as other SCF components , has led to the hypothesis that these ligases may also be involved in recognition of foreign substrates . Our study with microsporidia and virus infection provides the first experimental support for this hypothesis . In particular , we see that the C . elegans SCF ligase components cul-6 and skr-3 , skr-4 , and skr-5 , mediate a defense response of C . elegans to N . parisii and virus infection . Previous reports indicated that CUL-6 and SKR-3 interact physically in a yeast two-hybrid assay , indicating these components could assemble in vivo to produce a functioning SCF ligase [33] . Moreover , we see targeting of ubiquitin to N . parisii cells that depends on the cul-6 cullin component of the SCF ligase , which could conjugate host ubiquitin onto pathogen proteins or to host proteins that are associated with the pathogen cell . SCF ligases may also be involved in processing of proteins distinct from the pathogen cells , such as virulence factors that are secreted out of the pathogen cell into the cytosol . Another intriguing possibility is that SCF ligases are important for degrading inhibitory host proteins to trigger host innate immunity , analogous to ubiquitin-mediated degradation of IκB in NFκB signaling in mammals . However , the actual signaling proteins in C . elegans would be different because the NFκB transcription factor has been lost in this lineage [66] . Processing of host signaling proteins could occur in the clusters of conjugated ubiquitin we see later during infection , which are not associated with pathogen cells . Indeed , all of these possibilities are not mutually exclusive , and there are likely many roles for the SCF ligases and ubiquitin-mediated responses to intracellular infection in C . elegans . Our analysis of the gene expression response to N . parisii infection indicated that C . elegans has a very distinct response to this pathogen compared to previously described extracellular pathogens . Responses to extracellular pathogens like S . aureus and P . aeruginosa are marked by upregulation of secreted anti-microbials and detoxifying enzymes [37] , [41] , [67] , [68] , [69] , which did not comprise a substantial part of the gene sets upregulated by N . parisii . Instead we found enrichment for genes associated with ubiquitylation ( Table 1 , S9 ) , and that the response to N . parisii shared greatest similarities with the response to Orsay virus infection . The commonality of transcriptional response to these two very distinct pathogens ( N . parisii is a eukaryotic organism with 2661 genes and the Orsay virus has only 3 genes ) is quite striking , and our data indicate that some genes induced by infection such as SCF ligases can also be induced by perturbation of UPS function . Indeed , inhibition of the proteasome has been shown to induce stress response genes in other C . elegans studies as well [58] , [59] . These results fit with the growing theme that C . elegans epithelial defense relies on monitoring of core host processes as an important cue to indicate the presence of pathogen attack [56] , [57] , [70] , [71] . Such surveillance pathways are increasingly appreciated in mammalian defense as well , and may constitute a major mode by which hosts discriminate pathogens from other microbes [72] , [73] . It is possible that surveillance of UPS function is responsible for controlling the transcriptional response to intracellular infection , although it is possible that UPS perturbation and infection are distinct triggers that converge to upregulate the same response genes . Intracellular infection as well as perturbation of the UPS would be expected to cause substantial stress on the protein homeostasis ( proteostasis ) network of intestinal cells [74] , [75] , [76] . Intracellular infection by both N . parisii and virus should introduce a suite of foreign proteins into the host cell , may also cause damage to host proteins , and lead to activation of inducible immune responses . Any and all of these physiological changes may cause stress on the protein degradation and/or chaperone/folding systems of the host . This stress could explain the partial overlap we saw between the transcriptional response to intracellular infection and prolonged heat shock , a condition known to disrupt cellular proteostasis , although we saw an upregulation of only two hsp chaperones in response to infection ( Table S7 ) . In particular , hsp-16 . 1 , which was significantly upregulated at 64 hpi when animal intestines are filled with parasite spores and large vacuoles ( Figure 1A ) , has been shown to act in the Golgi where it helps to maintain cellular Ca2+ balance and protects cells against necrotic cell death triggered by heat as well as insults unrelated to thermal stress [43] . Further comparison between the responses to UPS stress and intracellular infection will likely shed light on mechanisms of cytosolic quality control and how they regulate defense against intracellular infection . While ubiquitin-mediated defense does play a role in limiting N . parisii growth , it appears to be only a minor one . There are several reasons that could account for this small effect . First , because UPS components are essential for animal development and overall health we relied on partial knockdown of UPS components to compromise UPS function . Second , in analyses of genes that are not essential , such as SCF ligase components , there may be redundancy in the proteins involved in defense . Third , we anticipate that like other intracellular pathogens [77] , [78] ( for example the Orsay virus in this study ) , N . parisii may subvert host ubiquitylation machinery to promote its own growth . In this case , compromised host UPS would negatively impact both the replication of N . parisii as well as the ability of C . elegans to clear infection , yielding a small net change in pathogen load . Lastly , it is possible that N . parisii suppresses or evades the host ubiquitin-mediated defense . Consistent with this idea , C . elegans is better able to target ubiquitin to pathogens and induce their degradation when N . parisii is treated with drugs that slow its growth . Additionally , if N . parisii were suppressing C . elegans ubiquitin-mediated defenses , then genetic inhibition of these processes in the context of infection would only have a minor effect on pathogen resistance , while genetic activation could have a greater effect . Indeed , we found that activating autophagy through RNAi against let-363/TOR led to improved targeting and clearance of N . parisii cells , with a greater effect on resistance than autophagy inhibition . However , it is important to note that let-363/TOR is upstream of several other processes , including protein synthesis [79] , which may also account for the increased resistance of this strain . Other pathogens have been shown to actively suppress ubiquitin-mediated defenses of other eukaryotic hosts [21] , [78] , [80] , [81] , [82] . For example , in human cells , the bacterial pathogen Salmonella enterica suppresses ubiquitin-mediated host defenses with the GogB effector , which inhibits a human SCF ligase by interacting with Skp1 and the human F-box only 22 ( FBXO22 ) protein , an interaction that impedes NFκB signaling and limits inflammation in infected cells [83] . Similarly , N . parisii might deploy effectors that block ubiquitylation of meronts , which are in direct contact with the cell cytosol of C . elegans intestinal cells and should be accessible to host ubiquitylation machinery . N . parisii might also evade ubiquitylation by the host by masking or simply lacking host-recognizable cues present during other intracellular pathogen infections . In particular , because N . parisii is itself a eukaryote , it may possess fewer pathogen-associated molecular patterns ( e . g . bacterial peptidoglycan or lipopolysaccharide ) , which can be used by eukaryotes to recognize pathogens . Microsporidia are increasingly recognized as natural pathogens of nematodes [84] , [85] , and Nematocida strains in particular have been isolated from multiple wild-caught Caenorhabditis nematodes [12] . It will be interesting to examine the interaction between other Nematocida pathogens and Caenorhabditis hosts to determine whether ubiquitin-mediated defenses have a greater or lesser role in those encounters , as part of the ever-shifting landscape of the host/pathogen arms race . Because microsporidia are obligate intracellular pathogens ( which by definition cannot grow outside of host cells ) , it is imperative that they evade or suppress host defense pathways such as ubiquitylation to propagate the species . Thus suppression or evasion of host defense , together with extremely rapid intracellular replication [34] , may be at the heart of why the Microsporidia have grown to be such a large and successful phylum able to infect virtually all animal hosts .
All C . elegans strains were maintained on nematode growth media ( NGM ) and fed with E . coli strain OP50-1 , as described [86] . N . parisii spores were prepared as previously described [87] . Briefly , N . parisii was cultured by infecting large-scale cultures of C . elegans , followed by mechanical disruption of worms and then filtering to isolate spores away from worm debris . The temperature-sensitive sterile strain CF512 fer-15 ( b26 ) ;fem-1 ( hc17 ) was used for RNA-seq and other experiments to prevent internal hatching of progeny at later infection time points . This strain was maintained using standard laboratory techniques at the permissive temperature of 15°C and shifted to 25°C for pathogen infection experiments [39] . The DA2123 adIs2122[lgg-1p::gfp::lgg-1] strain was a kind gift from Dr . Malene Hansen [88] , [89] . Promoter-GFP fusions for the N . parisii induced genes C17H1 . 6 and F26F2 . 1 were made using overlap PCR . Briefly , genomic DNA upstream of the predicted start for these genes was amplified ( 1273 bp for C17H1 . 6 and 796 bp for F26F2 . 1 ) with PCR and then fused in frame to GFP amplified from pPD95 . 75 . These promoter-GFP fusions were co-injected with the myo-2p::mCherry marker that labels pharyngeal muscle . Several independent transgenic lines carrying extrachromosomal arrays for these fusions were isolated and these lines induced GFP upon infection with N . parisii . One line for each fusion was integrated using psoralen/UV-irradiation to generate the integrated transgenic strains ERT54 jyIs8[C17H1 . 6p::gfp; myo-2p::mCherry] × and ERT72 jyIs15[F26F2 . 1p::gfp; myo-2::mCherry] . A GFP-tagged ubiquitin construct pET341 was generated using three-part Gateway recombination by fusing the intestinal-specific vha-6 promoter to GFP at the N-terminus of ubiquitin ( amplified from the C . elegans ubq-1 gene ) , with a unc-54 3'UTR , introduced into destination vector pCFJ150 that encodes for a wild-type copy of C . briggsae unc-119 gene under the control of the unc-119 promoter . This construct was injected into EG6699 ttTi5605 II; unc-119 ( ed9 ) III mutant animals and transgenic progeny were recovered , to generate a multi-copy array strain ERT261 jyEx128[vha-6p::gfp::ubiquitin cb-unc-119 ( + ) ];ttTi5605 II; unc-119 ( ed9 ) . Likewise , construct pET346 was generated , which contains a mutant version of ubiquitin without its last two C-terminal glycines . This construct was injected into EG6699 to generate multi-copy array strain ERT264 jyEx131[vha-6p::gfp::ubiquitinΔGG cb-unc-119 ( + ) ]ttTi5605 II; unc-119 ( ed9 ) . C . elegans infections , RNA isolation , and library construction are previously described [34] . Briefly , synchronized fer-15 ( b26 ) ;fem-1 ( hc17 ) L1s were grown for 24 hours at 25°C on 10-cm NGM plates seeded with OP50-1 E . coli and then infected with N . parisii ERTm1 spores . Infected and control C . elegans were harvested at appropriate times and total RNA was extracted using TriReagent ( Molecular Research Center , Inc . ) . RT-qPCR and the Bioanalyzer assessed quality of RNA samples . Strand-specific libraries were constructed using the dUTP second strand marking method [90] , [91] . Reads were aligned using Bowtie[92] and transcript abundance estimated using RSEM [93] . Differentially expressed transcripts were identified using the edgeR Bioconductor package ( Empirical analysis of digital gene expression data in R , v 3 . 0 . 8 ) [94] . FDR [95] cutoff was set to <0 . 05 , which yielded lists of genes with >4-fold difference in expression . C . elegans reads comprised the majority of the infected sample reads , ranging from over 99% early during infection ( 8 and 16 hpi ) to 71 . 6% at 40 hpi ( Table S1 ) . The progressive reduction in the fraction of C . elegans reads corresponded to replication of microsporidia in the C . elegans intestine resulting in increased contribution of parasite RNA to total RNA of each infected sample [34] . The number of expressed C . elegans genes in all samples ranged from 55 . 4% ( 64 hpi ) to 62 . 1% ( 16 hpi ) of the total genome ( Table S1 ) . Despite the growing input of parasite RNA , global C . elegans gene expression remained comparable between infected samples and uninfected controls , with the greatest absolute difference ( 3 . 61% ) in total number of expressed genes , which occurred at 64 hpi ( infected vs uninfected control ) . Based on previous studies , genes were classified as either intestinal-associated ( as determined by fluorescence-activated nuclei sorting ) [96] , germline-associated ( as determined by SAGE ) [97] , or neither . Very few germline specific/enriched genes were among the differentially expressed genes ( Table S2 ) and therefore we used all genes expressed in germ lines detected by SAGE as the germline-associated class . We then compared the number of differentially expressed genes from each category to the number expected from the classification using the chi-squared test . Gene Set Enrichment Analysis ( GSEA ) v2 . 0 [42] was used to compare gene sets from relevant C . elegans expression studies to our RNA-seq data . The RNA-seq expression dataset file used to generate ranked gene lists ( from most upregulated to most downregulated ) based on changes in expression between infected and uninfected conditions is summarized in Table S4 while the compiled gene sets used for analysis are described in detail in Table S5 . Genes from other studies were converted where necessary to WBGeneIDs according to Wormbase version WS235 . Five independent analyses were performed , one for each infection timepoint , with 1000 permutations for each analysis . Results for gene sets with FDR<0 . 25 and nominal p-value<0 . 05 were compiled into a graphical representation based on their NES-values , and for gene sets where the NES was not considered significant a value of zero was assigned ( Table S6 ) . Experiments were performed at 25°C and for each condition two biological replicates were included . About 200 synchronized fer-15 ( b26 ) ;fem-1 ( hc17 ) L1s were grown on 6-cm plates for two days , feeding on a lawn of E . coli RNAi clones from the Ahringer library or the skr-4 RNAi clone generated through amplification of C . elegans skr-4 genomic sequence ( using primers 5′ CCGAATTCGTCTCACGAAAAGTGATC - and 5′- CCGAATTCGGCGTTATACATTTATTCAA ) and cloned into the L4440 RNAi vector using EcoRI restriction sites . Animals were then infected with 2 million spores , fixed in 4% paraformaldehyde ( PFA ) 24 hpi , and stained with MicroB FISH probe against N . parisii rRNA as previously described [12] , [34] . Stained animals were mounted on glass slides in Vectashield with DAPI ( Vector Laboratories ) and imaged using a Zeiss AxioImager microscope with a 10× objective . Exposure times were kept the same for all samples within a single experiment . For all experiments except for ones in Figures 2B , 2G and S4 , where a custom fully automatic method for estimating pathogen load written in Matlab was used ( see Figure S2 and Supplemental Methods in Text S1 ) , images were analyzed semi-manually using ImageJ software , where the nematode body area , and the area of pathogen contained within were determined using two different thresholds of the MicroB FISH signal ( a relaxed threshold to recognize the background staining of the animal body , and a stringent threshold to specifically recognize the pathogen ) . Due to developmental defects caused by knockdown of UPS components , for experiments targeting the UPS , animals were first grown for one day on E . coli strain OP50-1 , and then transferred to plates seeded with UPS RNAi clones diluted with the L4440 RNAi vector control ( 1∶10 for ubq-2 , 1∶5 for pas-5 , and 1∶20 for rpn-2 ) . C . elegans has two genes encoding for ubiquitin , ubq-1 and ubq-2 . The ubq-2 RNAi clone was chosen for majority of experiments because it had less pronounced developmental defects then animals fed with RNAi against ubq-1 ( data not shown ) . After one day on RNAi , animals were infected and processed as described above . For fumagillin and FUdR experiments , animals were grown , infected , and processed as described above , except at 8 hpi , 0 or 25 µM of fumagillin ( Medivet Pharmaceuticals Ltd . ) or 0 or 2 . 6 µg/µL of FUdR ( Acros Organics ) in 250 µL of M9 with 0 . 1% Triton-X was spread onto plates containing the animals for a final concentration of 0 to 0 . 26 µg/mL ( fumagillin ) and 0 to 59 µg/mL ( FUdR ) present for the remainder of the experiment ( an additional 16 hours ) . To quantify ubiquitin colocalization with microsporidia , about 200 synchronized fer-15 ( b26 ) ;fem-1 ( hc17 ) L1s were grown on 6-cm plates for 2 days at 25°C , and then were infected with 5 million N . parisii spores . At 8 hpi , the infected animals were treated with 250 µL of 0 µM , 25 µM , or 150 µM of fumagillin in M9 with 0 . 1% Triton-X ( fumagillin final plate concentrations of 0 µg/mL , 0 . 26 µg/mL , or 1 . 56 µg/mL ) . At 12 hpi , animals were anesthetized with 10 mM levamisole , their intestines dissected out , and fixed for 15–30 min in 4% PFA . The intestines were stained with MicroB FISH probe against N . parisii rRNA , followed by staining with FK2 antibody ( Millipore ) , and secondary antibody staining with FITC goat anti-mouse IgG ( Jackson ImmunoResearch ) . Stained intestines were mounted in Vectashield with DAPI ( Vector Laboratories ) and imaged . For each condition , z-stacks spanning the width of twelve intestines were taken , and colocalization between each imaged parasite cell and the FK2 antibody was determined . All images , unless specified otherwise , were captured using a laser scanning confocal microscope with a 40× oil immersion objective ( Zeiss LSM 700 , equipped with an AxioCam digital camera and Zen 2010 acquisition software ) . Images were imported into Adobe Photoshop and assembled using Adobe Illustrator . For ubiquitin immunofluorescence at different stages of infection , animals were infected with N . parisii as described for RNA-seq . After 30 or 40 hpi , animals were anesthetized with 10 mM levamisole , their intestines dissected out , and fixed for 30 min in 4% PFA . The intestines from the 30 hpi infected and uninfected control samples were stained as described above . Intestines from the 40 hpi infected and control samples were stained directly with antibodies without FISH staining . Stained intestines were mounted in Vectashield with DAPI ( Vector Laboratories ) and imaged . To quantify GFP::ubiquitin colocalization with microsporidia , about 200 synchronized ERT261 or ERT264 L1s were grown on 6-cm plates , seeded either with OP50-1 E . coli or control L4440 and cul-6 RNAi clone , for 36 hours at 20°C and then infected with 5 million N . parisii spores . At 10 hpi , the infected animals were treated with 250 µL of 0 µM , 25 µM , or 150 µM of fumagillin in M9 with 0 . 1% Triton-X , and at 15 hpi animals were fixed in 4% PFA , stained with MicroB FISH probe against N . parisii rRNA , mounted in Vectashield with DAPI , and imaged as described above . For each condition and experiment , z-stacks spanning the width of twenty to eleven ERT261 and seven to ten ERT264 intestines were taken , and colocalization between each imaged parasite cell and GFP was determined . For RNAi experiments , eight to ten ERT261 animals were imaged for each condition and experiment . For imaging of GFP::ubiquitin in live animals , synchronized ERT261 animals expressing the intestinal GFP::ubiquitin construct were grown and infected at 20°C to minimize ubiquitin aggregate formation in uninfected controls . Synchronized animals were grown for 24 hours on 6-cm plates prior to inoculation with 2 million N . parisii spores and 48 hpi were mounted on agarose pads , anesthetized with 1 mM levamisole and imaged . For quantification of GFP::ubiquitin aggregates , synchronized ERT261 or ERT264 animals were grown at 20°C for 31 hours on 6-cm plates prior to inoculation with 1 million spores . At 10 , 30 and 45 hpi , animals were fixed with PFA and stained with MicroB FISH probe as described above . Stained animals were mounted in Vectashield with DAPI and viewed directly with a laser scanning confocal microscope with a 40× oil immersion objective ( Zeiss LSM 700 ) . To image promoter-GFP reporter strains , synchronized ERT54 and ERT72 L1s were grown for 24 hours at 25°C and infected with 10 million N . parisii spores on 6-cm plates . Infected and control worms were anesthetized with 1 mM levamisole , mounted on agar pads , and imaged at 8 and 24 hpi using a Zeiss AxioImager microscope . For RNAi experiments , synchronized ERT54 and ERT72 L1s were grown for 48 hours at 20°C on plates seeded with RNAi clones and imaged as described above . For MG-132 experiments , synchronized ERT54 and ERT72 L1s were grown for 24 hours at 20°C , incubated on a nutator at room temperature for six hours in M9 with 0 . 1% Triton-X and 0 µM , 500 µM , or 1mM MG-132 , and then imaged as described above . To measure endogenous mRNA expression changes due to UPS component knockdown , synchronized fer-15 ( b26 ) ;fem-1 ( hc17 ) L1s were grown at 20°C for 48 hours on RNAi bacteria , and then collected in TriReagent ( Molecular Research Center , Inc . ) for RNA extraction . To measure endogenous mRNA expression changes due to pharmacological proteasome inhibition , synchronized fer-15 ( b26 ) ;fem-1 ( hc17 ) L1s were grown 24 h at 20°C , incubated on a nutator at room temperature for six hours in M9 with 0 . 1% Triton-X and 0 µM or 500 µM MG-132 , and then collected in TriReagent for RNA extraction . RNA extraction , reverse transcription , and qRT-PCR were performed as previously described [41] . qRT-PCR primer sequences are available upon request . Each biological replicate was measured in duplicate and normalized to the snb-1 control gene , which did not change upon conditions tested . The Pffafl method was used for quantifying data [98] . Virus stock for infections was prepared as described previously [5] , with minor modifications . Briefly , the virus-susceptible rde-1 ( ne219 ) nematodes were grown in large-scale cultures until just starved , mechanically disrupted , and filtered through a 0 . 2 µm filter to separate the virus away from nematode debris . When spread on a 6-cm plate in a 250 µL volume , the 1∶50 dilution of this filtrate was the maximum dilution tested that turned on the F26F2 . 1p::gfp reporter in all animals 24 hpi at 25°C ( data not shown ) . These conditions were used for all viral infections . To measure changes in viral load upon RNAi-mediated knockdown of C . elegans genes of interest , the viral RNA1 levels were measured using primers GW195 and GW194 [5] and compared to those found in L4440 controls . For these experiments , fer-15 ( b26 ) ;fem-1 ( hc17 ) animals were grown and treated with RNAi the same as for the N . parisii pathogen load experiments , except about 300 synchronized L1 animals were used per 6-cm plate and following 24 hours of infection with the virus , animals were collected for RNA extraction and qRT-PCR . Intestine dissections from ERT72 animals and immunofluorescence with FK2 anti-conjugated-ubiquitin antibody were performed as described above , except the secondary antibody used was the Cy3 goat anti-mouse IgG ( Jackson ImmunoResearch ) . C . elegans genes analyzed: cul-6 , skr-3 , skr-4 , skr-5 , ubq-1 , ubq-2 , pas-5 , rpn-2 , lgg-1 , lgg-2 , atg-18 , sqst-1 , let-363 , C17H1 . 6 , F26F2 . 1 , skr-1 , C17H1 . 14 , F26F2 . 4 , Y39G8B . 5 , sdz-6 , T08E11 . 1 , W08A12 . 4 , ZC196 . 3 , Y94H6A . 2 , his-10 , his-16 RNA-seq data are part of NCBI BioProject #PRJNA163569 .
|
Microbial pathogens have two distinct lifestyles: some pathogens live outside of host cells , and others live inside of host cells and are called intracellular pathogens . Microsporidia are fungal-related intracellular pathogens that can infect all animals , but are poorly understood . We used the roundworm C . elegans as a host to show that ubiquitin pathways provide defense against both a natural microsporidian infection of C . elegans , as well as a natural viral infection . Our study shows that ubiquitin , the proteasome and autophagy components are all important to control intracellular infection in C . elegans , although microsporidia seem to partially evade this defense . We also show that SCF ubiquitin ligases help control both microsporidia and virus infection . Furthermore , we find that C . elegans upregulates expression of SCF ligases when ubiquitin-related degradation machinery is inhibited , indicating that C . elegans monitors the functioning of this core cellular process and upregulates ligase expression when it is perturbed . Altogether , our findings describe ubiquitin-mediated pathways that are involved in host response and defense against intracellular pathogens , and how this machinery is regulated by infection to increase defense against intracellular pathogens such as microsporidia and viruses .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genome",
"expression",
"analysis",
"invertebrates",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"caenorhabditis",
"immunology",
"microbiology",
"animals",
"parasitology",
"animal",
"models",
"caenorhabditis",
"elegans",
"model",
"organisms",
"emerging",
"infectious",
"diseases",
"genome",
"analysis",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"viral",
"clearance",
"pathogenesis",
"immune",
"response",
"immune",
"system",
"immunity",
"host-pathogen",
"interactions",
"virology",
"transcriptome",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"nematoda",
"genomics",
"computational",
"biology",
"organisms"
] |
2014
|
Ubiquitin-Mediated Response to Microsporidia and Virus Infection in C. elegans
|
Zika virus ( ZIKV ) is a little known flavivirus that caused a major outbreak in 2007 , in the South-western Pacific Island of Yap . It causes dengue-like syndromes but with milder symptoms . In Africa , where it was first isolated , ZIKV is mainly transmitted by sylvatic Aedes mosquitoes . The virus has also been isolated from Ae . aegypti and it is considered to be the vector involved in the urban transmission of the virus . Transmission of the virus by an African strain of Ae . aegypti has also been demonstrated under laboratory conditions . The aim of the present study is to describe the oral susceptibility of a Singapore strain of Ae . aegypti to ZIKV , under conditions that simulate local climate . To assess the receptivity of Singapore's Ae . aegypti to the virus , we orally exposed a local mosquito strain to a Ugandan strain of ZIKV . Upon exposure , fully engorged mosquitoes were maintained in an environmental chamber set at 29°C and 70–75% RH . Eight mosquitoes were then sampled daily from day 1 to day 7 , and subsequently on days 10 and 14 post exposure ( pe ) . The virus titer of the midgut and salivary glands of each mosquito were determined using a tissue culture infectious dose50 ( TCID50 ) assay . High midgut infection and salivary gland dissemination rates were observed . By day 5 after the infectious blood meal , ZIKV was found in the salivary glands of more than half of the mosquitoes tested ( 62% ) ; and by day 10 , all mosquitoes were potentially infective . This study showed that Singapore's urban Ae . aegypti are susceptible and are potentially capable of transmitting ZIKV . The virus could be established in Singapore should it be introduced . Nevertheless , Singapore's current dengue control strategy is applicable to control ZIKV .
Zika virus ( ZIKV ) is an emerging mosquito-borne pathogen belonging to the genus Flavivirus of the Family Flaviviridae [1] . It is a positive single stranded RNA virus with a 10 , 794 nucleotide genome that is closely related to the Spondweni virus ( Flavivirus , Family Flaviviridae ) [2] , [3] . The virus was first isolated in 1947 from a febrile rhesus monkey in the Zika forest of Uganda [4] . Non-human primates were implicated as the reservoir host of ZIKV in Africa and Asia [5] . In humans , ZIKV causes a mild infection manifested by a rash , fever , joint and muscle pain , headache and peri-orbital pain , which are characteristic signs and symptoms of flavivirus infections [6] , [7] . The first human ZIKV infection was reported in Uganda in 1964 [6] . Although the isolation of ZIKV has so far been confined to the African continent [8] , [9] , serological evidence has shown widespread distribution of the virus even in Asian countries such as Malaysia , India , Philippines , Thailand , Vietnam , Indonesia , and Pakistan [10] , [11] , [12] , [13] , [14] , [15] . The first major outbreak of human ZIKV infection was reported in the Pacific island of Yap and its adjoining islands in the Federated State of Micronesia in 2007 [3] , [7] , [16] , [17] . The outbreak lasted four months infecting approximately 73% of the islands' population [7] . In 2011 , ZIKV was first reported in the western hemisphere in travellers returning from Senegal [18] . Most recently , ZIKV was isolated from a 3-year old boy in Cambodia in 2010 [19] . ZIKV is transmitted to humans by Aedes spp . mosquitoes . The earliest evidence of ZIKV in a pool of Ae . africanus from Uganda in 1948 coincides with its first isolation from a rhesus monkey in the same location [4] . Subsequent documents reported further isolation of the virus from Ae . africanus and Ae . apicoargenteus caught in the Zika forest [20] , [21] , [22]; from Ae . luteocephalus in Nigeria in 1969 [23]; and from Ae . vitattus , Ae . furcifer , and Ae . aegypti in Ivory Coast in 1999 [24] . High prevalence of ZIKV antibodies in the urban population of Nigeria has led Fagbami [23] to suspect that Ae . aegypti may play an important role in the urban transmission of ZIKV . Further evidence came from Asia , when ZIKV was isolated from a pool of Ae . aegypti caught in Bentong , Peninsular Malaysia [25] . This finding provided evidence of ZIKV transmission outside Africa . In Indonesia , the peak of human ZIKV infections coincides with peak Ae . aegypti population which is by the end of rainy season [14] . Apart from field surveillance data , early experimental studies conducted by Boorman and Porterfield [26] and Cornet et al . [27] have also demonstrated the competency of Ae . aegypti to transmit ZIKV . Considering the geographic spread and the possible impact on susceptible human populations , mosquito-borne diseases are currently considered as a major threat to global health in both developing and developed world [28] , [29] . According to Gushulak et al . [30] , the threat of emerging infectious diseases is mainly influenced by the migration and mobility of the human populations . The dengue , chikungunya and malaria situations in Singapore clearly demonstrate the role of importation in shaping the epidemiology of these diseases [31] , [32] , [33] . Introduction of ZIKV into Singapore , a travel and trading hub , is plausible . Coupled with the local presence of Ae . aegypti , local transmission of the virus is likely . Furthermore , as ZIKV has never been reported in Singapore , the local population is presumed to be immunologically naive and vulnerable to the infection . Although experimental studies conducted in the past have shown that Ae . aegypti is a competent vector for ZIKV , these studies used African strains of Ae . aegypti that were caught in Nigeria [26] and Senegal [27] and had been maintained in the laboratory for years . Furthermore , experimental methods used in these studies differed from those of the current study . Although Boorman and Porterfield [26] infected the mosquitoes using the oral route , the average incubation temperature was 24°C , which is low in the tropical context and resulted in an extrinsic incubation period that suggested low vectorial capacity . While Cornett et al . [27] incubated their infected mosquitoes between 27 to 28°C , the method of infection was by intrathoracic route which can artificially lead to shorter extrinsic incubation period and higher number of mosquitoes infected . In addition , the geographical variations in terms of oral susceptibility of mosquitoes to different viruses are also well documented [34] , [35] , [36] , [37] , [38] , [39] . The present study describes the oral susceptibility of a Singapore field strain Ae . aegypti to ZIKV , under condition that simulate local climate .
Ae . aegypti , used for the experimental infection , were derived from eggs collected in the Western part of Singapore during a weekly ovitrap surveillance study to determine mosquito population density . Ovitraps were placed in public areas , mostly along the common corridors of public housing . The surveillance study was conducted by colleagues from the Environmental Health Institute . F0 adults were allowed to emerged and were maintained under standard insectary condition at 28±1°C and 75–80% relative humidity ( RH ) , with a photoperiod of 12h∶12h light∶dark ( L∶D ) cycles . They were allowed to mate randomly and fed with pathogen-free pig's blood ( A*star Biomedical Resource Center , Singapore ) using a Hemotek membrane feeding system ( Discovery Workshops , Lancashire , United Kingdom ) . F1 eggs were collected using filter paper ( Whattman , USA ) . Eggs were then allowed to hatch using de-chlorinated water and larvae were reared in 25 cm×30 cm×9 cm enamel pans containing 800 mL of water and fed with crushed dog food . Pupae were placed in 30 cm×30 cm×30 cm ( HxWxL ) cages before emergence . Prior to the infectious feed , adult mosquitoes were provided with 10% sugar/Vitamin B complex solution ad libitum . Ugandan MR766 ZIKV strain obtained from the American Type Culture Collection ( Manassas , VA , USA ) was used to expose the mosquitoes to ZIKV . This virus was originally isolated from the blood of an experimental sentinel rhesus monkey in 1947 [4] and passaged in suckling mouse brains . The stock virus used in the current study has been passaged thrice in Vero cells prior to the infectious feed . Five- to 7-day-old female mosquitoes ( n = 120 ) were transferred to 0 . 5 L containers and starved for 24 hours prior to the infectious blood meal . The blood meal consisted of 1∶1 100% swine-packed RBC ( Innovative Research , USA ) and fresh virus suspension at a final concentration of 7 . 0 Log10 tissue culture infectious dose50 ( TCID50 ) /mL . Adenosine Triphosphate ( Fermentas , USA ) , at a final concentration of 3 mM , was added to the blood meal as a phagostimulant . Mosquitoes were fed with an infectious blood meal that was constantly warmed to 37°C using a Hemotek membrane feeding system ( Discovery Workshops ) housed in a feeding chamber . Thirty minutes after exposure to the infectious blood meal , mosquitoes were cold anesthetized at −20°C . Fully engorged females were transferred to 300 mL cartons and were maintained in an environmental chamber ( Sanyo , Japan ) at 29°C and 70–75% RH with a 12h/12h L∶D cycle and provided with 10% sugar/vitamin B complex ad libitum . All experiments were carried out in an arthropod containment level 2 ( ACL-2 ) facility . To determine the ZIKV infection and dissemination rates in Ae . aegypti , eight mosquitoes were sampled daily from day 1 to day 7 , and subsequently on days 10 and 14 post exposure ( pe ) . To prevent cross-contamination of virus between midgut and salivary glands of each mosquito , these organs were carefully dissected using different dissecting needles and the organs were rinsed in Medium 199 ( M199 ) ( Gibco , USA ) supplemented with amphotericin B ( Sigma Aldrich , USA ) . The midguts and salivary glands from each mosquito were individually transferred to 2 mL microtubes containing 250 µL of M199 . These organs were then homogenized using five mm stainless steel grinding balls ( Retsch , Germany ) in a MM301 mixer mill ( Retsch , Germany ) set at frequency of 12/sec for 1 min . The supernatant of the homogenate was applied in the viral titer assay . All dissecting needles were dipped in 80% ethanol and cleaned before being re-used . All experiments were conducted inside an ACL-2 facility . Viral titers in this study were determined with a tissue culture infectious dose50 assay , an endpoint dilution technique , using Vero cells as described by Higgs et al . [40] . Briefly , 100 µL of 10-fold serial dilutions of each sample were titrated ( in duplicate ) in 96-well microtititer plates and incubated with Vero cells at 37°C and 5% CO2 . At the end of day-7 incubation , the cells were examined microscopically for ZIKV-induced cytopathic effect ( CPE ) . A well is scored positive if any CPE is observed compared to the uninfected control cells . All virus titers were expressed as Log10 TCID50/mL . Proportion infected was calculated by dividing the number of infected midguts ( or salivary glands ) by the total number of miguts ( or salivary glands ) sampled . To compare viral titers at different time points , raw data was subjected to a normality test using SPSS Ver 18 ( IBM , USA ) . Data that passed the normality test were analyzed by analysis of variance using the above mentioned software .
Presence or absence of blood in the midgut was verified during dissection under a Stereoscope ( Olympus , USA ) . By Day 3 , when blood had been completely digested , seven ( 87 . 5% ) of the analyzed mosquitoes were positive for ZIKV ( Figure 1 ) . From day 6 pe onwards , all midguts were positive for ZIKV except for one of the mosquitoes that was negative for the virus at day 7-pe . The presence of viable ZIKV in the salivary glands ( n = 1 ) was first observed on day 4 pe ( Figure 1 ) and 62% of mosquitoes sampled on day 5-pe showed detectable virus in the salivary glands . ZIKV was observed in salivary glands of all infected mosquitoes sampled at days 10 and 14 pe . Figure 2 presents ZIKV midgut titers at different days pe . Although remaining blood meal in midgut was not removed , an eclipse phase typically associated with low virus midgut titer can be seen on day 1 pe , with only one of the midgut showing detectable ZIKV . Virus titers in day 2 pe were higher than that observed for day 1 pe , mirroring the results obtained on percentage of midguts infected ( Figure 1 ) . These suggest that midgut ZIKV titer observed during day 2 pe was most probably due to virus replication in the midgut rather than to the remaining amount of blood observed in some of the mosquitoes . A significant increase ( P<0 . 026 ) in mean viral titers was observed between days 3 pe ( 3 . 9 Log10 TCID50/mL ) and day 5 pe ( 5 . 6 Log10 TCID50/mL ) . From day 6 pe onwards , mean viral titers showed a decreasing trend from fluctuated between 5 . 4 Log10 TCID50/mL and 5 . 9 Log10 TCID50/mL but the differences observed were not statistically significant ( P≥0 . 91 ) . ZIKV titers in the salivary glands increased from day 4 pe onwards ( Figure 3 ) . Although the difference in mean viral titers from day 5 pe ( 2 . 7 Log10 TCID50/mL ) to day 7 pe ( 3 . 7 Log10 TCID50/mL ) was not significant ( P = 0 . 68 ) , the mean viral load increased significantly ( P<0 . 001 ) by day 10 pe ( 6 . 4 Log10 TCID50/mL ) , achieving the highest mean viral load of >8 . 0 Log10 TCID50/mL by day 14 pe .
Recent unprecedented spread of chikungunya virus ( CHIKV ) in many parts of the world , with millions of people affected , exemplifies how arboviruses can adapt and affect human health on a global scale [31] . Singapore's vulnerability to emerging and re-emerging arboviruses is accentuated by the country's location as a popular tourist and business hub , high dependency on migrant workers , tropical climate , dense population , and the presence of potential mosquito vectors . An outbreak of chikungunya in Singapore during the 2008–09 period attests the country's vulnerability to mosquito-borne diseases [31] , [41] . The outbreaks of ZIKV on Yap Island and the worldwide spread of CHIKV have shown the propensity of arboviruses to spread outside their known geographical range and their potential to cause large-scale epidemics . Unlike CHIKV which has received much scientific attention , ZIKV is a little-known flavivirus despite its outbreak potential [42] . Most studies on ZIKV were conducted more than two decades ago and there is a dearth of information on mosquito-ZIKV interactions that are salient to a better understand virus transmission . In 1956 , Boorman and Porterfield [26] successfully transmitted the virus to both mice and monkeys using ZIKV-infected laboratory strains of Ae . aegypti . Cornet et al . [27] further demonstrated that a high percentage ( 88% ) of intrathoracically infected Ae . aegypti can transmit ZIKV to experimental mice within 7 days and transmission rates increased up to 95% on day 21 pe . The current study , using a field strain of mosquitoes , showed that Singapore's Ae . aegypti are highly susceptible to ZIKV , with high midgut infection and salivary gland dissemination rates . By day 5 pe , 62% of the mosquitoes had detectable ZIKV in their salivary glands and by day 10 pe all mosquitoes were potentially infective . Based on the studies of Cornet et al . [27] , nearly all mosquitoes with ZIKV in their salivary glands are assumed to be able to transmit the virus . This is supported by previous studies that have shown oral transmission of dengue ( DENV ) [43] , [44] and West Nile ( WNV ) [45] viruses were correlated with the proportion of mosquitoes with infected salivary glands . The decrease in midgut viral titer at day 14 pe observed in our study was consistent with other published DENV and WNV studies [46] , [47] , [48] and were probably due to virus clearance by the mosquito immune system [47] , [49] , [50] . Despite a decrease in midgut viral titer , ZIKV infection in salivary glands was found to be higher than that observed in midgut . This suggests that the proliferation of ZIKV in Ae . aegypti salivary glands is not attributed to direct dissemination from the midgut , but rather a result of viral dissemination and amplification within the glands or other organs or tissues such as hemocytes , ganglion , fat bodies etc [49] , [51] , [52] . Salivary gland dissemination rates obtained from our current study is similar to that observed for a local highly epidemic DENV-2 in the same strain of Ae . aegypti ( Tan et al . , unpublished data ) . A phylogenetic analysis , based on the NS5 region , of ZIKV revealed three branches: West African ( Nigeria ) , East African ( Uganda ) and those from Yap island ( ZIKV 2007 EC ) , with the latter virus being the most distally related [17] . The strain used in our current study , MR766 , is the Ugandan prototype strain and the only strain available to our laboratory . It would be very interesting to study and compare the recent epidemic ZIKV 2007 EC strain in Ae . aegypti , especially in the light of a four amino acid motif found in the viral envelope genes of the ZIKV 2007 EC strain that are absent in the MR766 strain [17] . Unfortunately , no ZIKV 2007 EC was isolated during the outbreak in Yap Island . The four amino acid motif found in the ZIKV 2007 EC strain correspond to an envelope protein 154 glycosylation motif and the loss of this motif in the ZIKV prototype strain is thought to have been due to extensive passage in mice [17] . Studies have showed that loss of glycosylation motif due to mutation has been found to affect the replication rates of tick-borne encephalitis virus . DENV , and WNV in both vector hosts and insect cell lines and the dissemination rate of WNV in different Culex spp . mosquitoes [53] , [54] , [55] , [56] . Despite the absence of this aa 154 glycan , the present study has shown that ZIKV MR766 has a high dissemination rate in Singapore's Ae . aegypti . This could probably be due to the high midgut pH found in Ae . aegypti [57] , a characteristic shared by Cx . tarsalis , which rendered it susceptible to WNV virus lacking the aa 154 glycan [53] . Future studies with other strains will take these observations into consideration . Timely detection of the causative agents and implementation of effective control strategies during an epidemic or outbreak are always challenging . A fully-integrated vector control program incorporating advances in laboratory techniques and surveillance programs designed to address all components of the virus life cycle is considered the best approach in detecting and controlling any vector-borne disease as they emerge [42] . Such was the case of the successful control of the CHIKV outbreak in Singapore in 2008 [41] . The use of rapid and sensitive diagnostic and effective field surveillance tools and good coordination between field and laboratory personnel coupled with an understanding of mosquito-virus relationship assisted in the situation assessment and operational decision-making in controlling the outbreak . The present study revealed the potential role of local Ae . aegypti as a vector of ZIKV . Given the presence of the virus in the region , the Environmental Health Institute screened febrile cases not attributable to DENV and CHIKV for ZIKV and other arboviruses . Among the 690 cases screened between 2009 and 2010 , none was found positive for flaviviruses other than DENV . While there is currently no evidence of its circulation in Singapore , regular screening will be performed to monitor the situation . Based on the information gathered from this study ( e . g . viral dissemination rate ) , the threat of ZIKV can be addressed by the existing dengue control programme . However , there is also a need to determine the susceptibility of other common mosquito species , in order to design a comprehensive vector control strategy for Zika infection .
|
Zika virus ( ZIKV ) is an emerging mosquito-borne zoonotic pathogen that causes dengue-like syndromes . Despite its high epidemic potential , little is known about the virus . Although the isolation of the virus was confined to the African continent , serological evidences have shown the widespread distribution of ZIKV , particularly in Asia . In 2007 , it caused a major outbreak on the Pacific Island of Yap , infecting more than 70% of the island’s inhabitants . The propensity of the virus to spread outside its known geographical range was again demonstrated when it was detected in the US from travellers coming back from endemic countries . Several species of Aedes spp . mosquitoes have been incriminated as vectors of ZIKV , including Ae . aegypti . The current study showed that local Ae . aegypti are highly susceptible to ZIKV and by day 5 post-infectious blood meal , more than 50% of mosquitoes were potentially infective . Singapore being a tourist and a business hub , coupled with the presence of susceptible vector and a population that is immunologically naive and vulnerable , the local transmission of the ZIKV is plausible . Nevertheless , Singapore's current dengue control strategy is applicable to control ZIKV .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"infectious",
"diseases",
"mosquitoes",
"vector",
"biology",
"neglected",
"tropical",
"diseases",
"biology",
"microbiology",
"vectors",
"and",
"hosts",
"viral",
"diseases"
] |
2012
|
Oral Susceptibility of Singapore Aedes (Stegomyia) aegypti (Linnaeus) to Zika Virus
|
Computational neuroimaging methods aim to predict brain responses ( measured e . g . with functional magnetic resonance imaging [fMRI] ) on the basis of stimulus features obtained through computational models . The accuracy of such prediction is used as an indicator of how well the model describes the computations underlying the brain function that is being considered . However , the prediction accuracy is bounded by the proportion of the variance of the brain response which is related to the measurement noise and not to the stimuli ( or cognitive functions ) . This bound to the performance of a computational model has been referred to as the noise ceiling . In previous fMRI applications two methods have been proposed to estimate the noise ceiling based on either a split-half procedure or Monte Carlo simulations . These methods make different assumptions over the nature of the effects underlying the data , and , importantly , their relation has not been clarified yet . Here , we derive an analytical form for the noise ceiling that does not require computationally expensive simulations or a splitting procedure that reduce the amount of data . The validity of this analytical definition is proved in simulations , we show that the analytical solution results in the same estimate of the noise ceiling as the Monte Carlo method . Considering different simulated noise structure , we evaluate different estimators of the variance of the responses and their impact on the estimation of the noise ceiling . We furthermore evaluate the interplay between regularization ( often used to estimate model fits to the data when the number of computational features in the model is large ) and model complexity on the performance with respect to the noise ceiling . Our results indicate that when considering the variance of the responses across runs , computing the noise ceiling analytically results in similar estimates as the split half estimator and approaches the true noise ceiling under a variety of simulated noise scenarios . Finally , the methods are tested on real fMRI data acquired at 7 Tesla .
Computational modelling approaches applied to functional magnetic resonance imaging ( fMRI ) measurements aim to explain and predict the brain responses by expressing them as a function of model features that describe the sensory ( or cognitive ) stimuli [1–5] . By doing so , computational neuroimaging methods have been proposed as a means to test the ( brain ) validity of the algorithm being evaluated and eventually its refinement . At the single voxel level , two different approaches , population Receptive Fields ( pRF ) modelling [3] and linearized encoding models [6 , 7] , have been developed to link computational models and fMRI responses . In the following , we will refer to both these approaches indiscriminately as encoding models ( see e . g . [8] for the relation between linearized encoding models and pRF approaches ) . The performance of a computational model that describes fMRI responses is evaluated in terms of its accuracy in predicting new ( test ) data . The prediction accuracy is not only affected by inaccuracies in the definition of the algorithm ( i . e . mismodelling ) but also by other sources of variance in the brain responses that are not expressly modelled ( e . g . attention and adaptation ) and , most importantly , by physiological ( e . g . respiration ) and measurement noise . These effects are evidenced by the fact that in real data , presenting multiple times the same stimulus does not result in the same measured brain response . Commonly tested models of sensory ( or cognitive ) stimuli do not account for the variability in the response between repetitions of the same stimulus which imposes a bound to the ability to encode computational models in fMRI responses . This bound can be interpreted as the performance of the computational model underlying the generation of the responses ( i . e . the true underlying model ) conditional to the noise ( experimental , physiological or other ) that is present in the test data ( under the assumption of infinite training data ) . It should be noted here that this represents one of many possible definitions of a bound to the performance of a computational model . This bound is imposed exclusively by the measurement noise in the test data ( i . e . test-data-noise ceiling ) . A more realistic definition of the noise ceiling would also consider the influence of the size of the training set and the algorithms used for estimating the computational model , however it has not been proposed yet . Reporting the test-data-noise ceiling allows assessing the quality of the predictions obtained when using computational modelling approaches relative to the quality of the data , and thus comparing modelling efforts on different datasets across labs . In the neuroimaging community , it has been recommended to report the performance of a computational model with respect to the ( test data ) noise ceiling and , in some cases , these recommendations have led to the use of normalized accuracy scores ( e . g . dividing the accuracy by the noise ceiling , [9–13] ) . Different estimation procedures have been proposed for the test-data-noise ceiling but the properties of these different estimators have not been compared . The main purpose of this article is to provide a framework in which the concept of the test-data-noise ceiling can be clarified for the users of encoding models and in which different estimators can be compared on the basis of their assumptions . To illustrate the concept of test-data-noise ceiling we can consider the two-step procedure that in many cases is used when fitting a computational model . In the first step , responses to the stimuli are estimated from the whole fMRI time series and , in the second step , the computational model is fit to the stimulus response series . While this two-level procedure is not used by all encoding approaches in practice , the unmodelled variability of the response between repetitions of the same stimulus limits the performance of a computational model that predicts the whole fMRI time series as well ( see e . g . pRF models [3] ) . As a consequence of this hierarchical estimation framework , the prediction accuracy of the model ( step 2 ) is bound by the uncertainty in the estimation of the response ( step 1 ) . This bound corresponds to the intraclass correlation coefficient [14] a well-known result in multilevel modelling . The noise ceiling can be obtained considering the variability across subjects ( see e . g . [15] ) , but here we will focus on estimation procedures at the single subject level , where two approaches have been proposed to estimate the test-data-noise ceiling of single voxels . The first models the response of a voxel as a univariate normal distribution with two variance components [16] . The first variance component corresponds to the variability of the signal around its mean due to genuine differences in the brain response between different stimuli ( excluding the effects of measurement noise ) . The second variance component corresponds to the variability in the brain response due to measurement noise . Having an estimate of the measurement noise allows generating new samples for both the signal without noise ( genuine brain response ) and the measurement ( i . e . signal plus noise ) using Monte Carlo simulations . The noise ceiling ( measured with correlation or predictive R2 ) is then computed using the simulated signals and measurements ( i . e . considering the performance in predicting the noisy measurements of a model whose prediction is the clean signal ) . In what follows we will refer to this approach as the Monte Carlo noise ceiling ( MCnc ) . Alternatively , the noise ceiling can be estimated as the correlation between the estimates of the responses in two independent repetitions of the same experimental procedure [17 , 18] . In absence of two repetitions of the test set , the split-half noise ceiling estimator ( SHnc ) can be estimated by splitting the available test data in two disjoint sets ( i . e . splitting the trials of all test stimuli in two sets to obtain two estimates of the test data ) , computing the split-half correlation and applying a correction factor that accounts for the reduced number of trials in each half of the data compared to the full dataset . In this article , we describe the differences between these two noise ceiling estimators using simulated data and derive an analytical solution to the calculation of the test-data-noise ceiling obviating the need of computationally demanding procedures ( i . e . Monte Carlo simulations ) or splitting the data in two sets . Importantly , the MCnc and the analytical noise ceiling we propose are based on an estimate of the variability of the estimated responses due to the measurement noise . In simulations , we show how different estimators for this variability impact the resulting test-data-noise ceiling depending on the structure of the noise in the data . When using linearized encoding approaches , regularization is often required because of the dimensionality of the model with respect to the number of stimuli and because of collinearity between the features of the computational model . Here we evaluate how the bias variance tradeoff introduced by the regularization influences the performance of an ideal model and thus the relationship with the noise ceiling ( which is model independent ) by imposing a second constraint . Finally , we evaluate the differences between the noise ceiling estimators using real fMRI data , obtained from 7 Tesla acquisitions . Our results are discussed in terms of their implications for the evaluation of computational models and their comparison using fMRI data .
The observed fMRI response is assumed to be linearly dependent on the stimuli ( design matrix ) [19 , 20] and the estimation ( for every voxel ) is achieved using generalized least squares ( GLS ) : β^= ( ΦTΩ^−1Φ ) −1ΦTΩ^−1y ( 1 ) The covariance matrix for β^ is: Vβ^= ( ΦTΩ^−1Φ ) −1ΦTΩ−1Φ ( ΦTΩ^−1Φ ) −1 ( 2 ) For every voxel , β^ is the vector of the estimated responses to the stimuli , y is vector of the voxel time course ( the observed fMRI signal ) and Φ is the design matrix describing the timing of presentation of the stimuli in the experiment , including the effect of the hemodynamic response . The matrix Ω^ is the estimated covariance matrix of the fMRI noise , while the Ω is the true value of this covariance . This general formulation can accommodate a variety of estimators for computing Vβ^ which depends on the assumptions made with respect to Ω . Note that Vβ^ depends of the true Ω which is unknown . In practice , estimators of Ω should be obtained which leads to the estimator of Vβ^ [21]: V^β^= ( ΦTΩ^−1Φ ) −1 ( 3 ) The assumption of identically independent distributed ( i . i . d ) noise ( which implies Ω^=I ) leads to the ordinary least squares estimator ( OLS ) [20] . Violations of the i . i . d . assumptions ( i . e . presence of temporal dependences and lack of stationarity of the fMRI noise ) involve computing Ω^ which is an ill conditioned problem , which is usually solved by imposing some form of regularization . Typically , this is achieved by parametrizations of the noise covariance matrix that accommodate the assumptions about the noise structure ( e . g . first order autoregressive model ) [22–24] . Linearized encoding models assume the estimated response vector β^ to be linearly dependent on the description of the stimuli on the basis of a computational model represented by a matrix X that projects each of the n stimuli onto the model space described by a model with f features ( Fig 1A ) . The ( linear ) weights that link the computational model to the fMRI response are referred to as the population receptive field of the voxel . If the objective of the encoding approach is to model the differences in the brain responses between stimuli as a function of the computational model then , voxels that differ only in their means , but otherwise represent stimuli in the same way , should have the same estimated population receptive field . To do this , the mean of the brain responses ( across all stimuli ) can be removed from the response vector β^ before fitting the model X , or a column of ones can be added to the model X before fitting the un normalized responses [25] . When there is collinearity across the features or when the number of stimuli n is smaller than then number of features f , regularization is used . Here we consider the use of ridge regression [26] for estimating the population receptive field P^ linking the computational model ( represented by the training data matrix X ) to the estimated voxel response vector on the training data set β^: P^= ( XTX+λI ) −1XTβ^ ( 4 ) where λ is the regularization parameter . P^ allows predicting the responses to new ( test ) stimuli by considering β*=X*P^ , where X* contains the representation of the stimuli in the test set in the space of f features . Other approaches use grid search or more sophisticated optimization algorithms when a non-linear relationship between the features and the response is assumed [3 , 27] . The performance of a computational model can be assessed on test stimuli using e . g . the sample correlation coefficient between the responses predicted by the computational model β* and the estimated brain responses in the test data β^ ( see Eq 1 ) : ρ=1 ( n−1 ) ∑i=1n ( β^i−β^¯ ) ( βi*−β¯* ) σ^β^2σ^β*2 ( 5 ) Where βi* and β^i are the predicted and estimated response to stimulus i respectively . Estimated variances refer to the variability between the components of the vector of responses around the mean of the vector: σ^β^2=1n−1∑in ( β^i−β^¯ ) 2 . The estimated mean β^¯=1n∑inβ^i corresponds to the sample mean of the estimated response across its components ( each component corresponds to one presented stimulus ) , with consistent definitions for σβ*2 and β¯* . Alternatively , predictive R2 is frequently used for describing the performance of an encoding model: R2=1−∑i=1n ( β^i−βi* ) 2∑i=1n ( β^i−β^¯ ) 2 ( 6 ) Note that we are computing the explained variance between the observed and predicted brain responses β^ and β* , which is different than the explained variance at the level of the fMRI time series y . When computed on independent test data , R2 is defined in the interval [−∞ , 1] . It is important to note that , while R2 is sensitive to scaling transformations of the estimated response , the correlation coefficient measures the similarity between the predicted and observed responses in term of covariations around their mean and is insensitive to scaling transformations . This difference between the metrics is relevant when regularization is used . The relation between predictive R2 and ρ ( see S1 Text ) can be rendered explicit considering that ( without loss of generality ) the estimated responses were centred to have zero mean ( β^¯ = 0 ) [25]: R2=2ργ ( λ ) −γ ( λ ) 2−n ( n−1 ) β¯*2σ^β^2 ( 7 ) Where γ2 ( λ ) =σ^ ( λ ) β*2σ^β^2 is the ratio between the estimated variances of the predicted and observed response vectors and dependent on the amount of regularization used to estimate the computational model ( see Eq 4 in [5] ) . The β¯* represents the bias in the mean of the predicted response vector ( i . e . how much the mean of the predicted response vector differs from zero , See S1 Text ) . Note that the optimal λ for the maximization of R2 does not necessarily corresponds to the λ which maximizes ρ . The predictive squared Euclidean distance D2 between the vectors β^ and β* , which is also a frequently used metric , is closely related to the explained variance: D2= ( n−1 ) σ^β^2 ( 1−R2 ) . As a consequence of the estimation procedure highlighted above ( see Eq 1 ) , in linearized encoding models , the estimated response vector β^ is assumed to be multivariate normally distributed around the true response vector β and with covariance matrix that reflects the variability of the β^ estimator: β^∼N ( β , Vβ^ ) ( 8 ) For linearized encoding models , β ( i . e . the expected value of the estimated response β^ ) can be considered to be generated on the basis of the computational model defined by the matrix X . In particular we can consider: β=XP ( 9 ) This assumes a fixed linear relation between matrix of model features X and the underlying “true” brain response β ( not influenced by measurement noise ) which is mediated by the population receptive field vector P . Note that the true response β can have any shape and is not limited to be a standard gaussian variable . The noise ceiling estimated under such fixed relationship is based on the assumption that the total variance of the underlying β can be explained by the features contained in X . For infinite training data ( and without the use of regularization ) the noise ceiling can be defined as the expected performance ( measured as correlation or predictive R2 ) of the model underlying the generation of the responses ( i . e . the “true” model X ) . Such model uses the true pRF ( P ) and produces correct predictions for a voxel , i . e . it assumes that the predicted responses β* are the true β: ρNC=E ( ρ ) β*|λ=0 , ntr=∞ ( 10 ) This definition of the noise ceiling is a function of the variability of the test data , is by construction independent of the computational model , and thus is better referred to as test-data-noise ceiling . Different assumptions underlying the univariate modeling of fMRI data result in different estimates of the responses ( β^ ) and their variance . Thus , the ( true ) noise ceiling is affected by the fMRI response estimation procedure . More sophisticated noise ceiling definitions might also consider the performance of the true model conditioned to the actual amount of data available , or conditioned on the particular algorithm that was used to fit the computational model ( e . g conditioned on the value of λ ) . As highlighted above , the MCnc and the analytical approach rely on the estimation of the covariance matrix of the estimated regression coefficients ( V^β^ ) . Different V^β^ estimates will therefore result in different expected values for the noise ceiling . In particular , as highlighted in Eq 3 , V^β^ depends on the assumptions made about the structure of the noise in the fMRI time series ( i . e . the structure of Ω^ in Eq 3 ) . Different analysis software use different covariance constrains for computing Ω^; to account for autocorrelation in the noise , here we followed the SPM-12 ( http://www . fil . ion . ucl . ac . uk/spm/software/spm12 ) approach which estimates the brain response in two passes . In the first pass , estimated brain responses are computed assuming Ω^=I , which corresponds to OLS . Next , the noise covariance matrix is computed assuming the same correlation structure for each voxel within each session . A parametrized model of the noise covariance matrix is fit to the pooled covariance matrix using restricted maximum likelihood ( spm_reml . m ) . The covariance constraint for Ω^ are obtained with the function ( spm_Ce . m ) with autocorrelation coefficient of 0 . 2 ( empirically determined ) . In the second pass the β^ are obtained with Eq 1 which uses the estimate of Ω^ derived in the first pass . To account for non-stationarity we followed the RobustWLS toolbox approach ( http://www . diedrichsenlab . org/imaging/robustWLS . html ) . In particular , the estimation of Ω^ is performed using a Newton-Raphson algorithm ( spm_rwls_reml . m in the WLS toolbox ) [30] . Note that autocorrelation and the non-stationarity constraints can be combined in the same estimation of Ω^ [30] . However , we do not perform this combined estimation here since we used WLS as implemented in the RobustWLS toolbox . When data are acquired across multiple runs ( or sessions ) , additional sources of variability across runs ( sessions ) , can have an effect on the variance of the β^ [31] . See [32] for a proper partitioning of the variance of the fMRI into runs and sessions components . Ignoring this source of variability resulted in the underestimation of the variability of the brain responses with the consequence of overestimating the NC . A solution for this problem is to estimate , for each voxel , the β^ responses using a mixed effect model . Two reasons limit the use of mixed models in fMRI: first , estimating such model for each voxel has a large computational cost ( impractical for a large number of voxels ) ; second , and more relevant , the design may not allow a reliable estimation of the model when the number of presentations of each stimulus across the runs is too small . An alternative is to compute the variability of β^ as the variance of the mean of β^ across the fMRI runs ( sessions ) : V^β^ii=1 ( nr−1 ) nr∑r=1nr ( β^ir−β^¯ir ) 2 ( 17 ) where nr is the number of runs ( or sessions ) , β^ir is an estimate of the response for stimuli i in the run r and β^¯ir is the mean of the response for stimulus i across the nr runs . Note that the variance of the mean of β^ across nr runs is the estimator of the variance of β^ divided by the number of runs . Since the noise ceiling is only a function of the measurement error of each component of β^ , ( and not of the covariance between components ) only the diagonal elements of the matrix V^β^ are relevant ( See Eq 16 ) . This estimator of the variance does not rely on assumptions on the particular form of Ω^ made by the parametric estimation procedure and can be derived directly from the run to run variability of the β^ . Such estimator has the attractive of not depending of the assumptions regarding the structure of the fMRI noise , however in fMRI experiments with only few runs it could make the estimator less robust .
In the i . i . d noise scenario , and using an estimation based on the assumption of noise covariance matrix is equal to the identity matrix , we verified that the expected value of the analytical NC corresponds to its definition of Eq 14 by comparing ( mean and [5 95] percentiles across 100 simulations ) the estimated noise ceiling value ( using Eq 16 ) with the true noise ceiling that in simulations can be computed directly knowing the true simulated β using Eq 14 ( red line in Fig 2 ) . This comparison is reported in Fig 2 for both correlation and predictive R2 . The mean value of the analytical NC estimator ( blue in Fig 2 ) matched the mean value and variance of the true noise ceiling . The R2 showed stronger dependence with the experimental noise than the correlation coefficient . The noise ceiling estimator showed larger variability ( blue dashed line ) than the true noise ceiling ( shaded region ) due to the uncertainty associated with the estimation of Vβ^ ( see Eq 3 ) . Using simulated data under i . i . d scenario and estimating V^β^ assuming that the noise covariance matrix is equal to the identity matrix we validated the equivalence between the Monte Carlo estimator ( MCnc ) [16] and the analytical estimator ( Eq 16 ) . The results are reported in Fig 3 for the correlation as evaluation metric ( for simplicity ) . Both NC estimators resulted in the same mean and variance across the 100 simulations ( the mean and the [5 95] confidence bands are superimposed in Fig 3 left panel ) . For both NC methods , the variability of the estimated NC increased with decreasing mean value ( mean and variance are not independent ) . The relation between the two NC estimators is presented as a scatter plot in the right panel of Fig 3 for all levels of experimental noise ( without averaging across simulations ) . The advantage of using the analytical noise ceiling estimator is that it did not require a computationally expensive resampling procedure at each voxel . The difference in computation time was: analytical NC ( 7x10-5 seconds/voxel ) vs MCnc ( 0 . 06 seconds/voxel , for 1000 Monte-Carlo samples ) , on a PC with intel ( R ) , i7-6700HQ , CPU 2 . 6 GHz , 16Gb of RAM processor ) . We have verified equivalence between the MCnc and the analytical NC also in the other simulation scenarios and using different assumptions about the structure of the covariance matrix of the noise during estimation . As both methods proved to be equivalent we report only the analytical NC estimator for the subsequent analyses . Fig 4 summarizes the results ( for different simulated noise levels ) obtained in estimating the noise ceiling in the i . i . d . noise scenario . The first three panels from the left represent the analytical NC obtained using three different parametric estimates of V^β^ ( i . e . assuming different structures for the noise covariance matrix during estimation , see Methods section ) . The analytical NC based on parametric estimators of V^β^ are compared to the noise ceiling using the run to run variability for estimating the diagonal entries of V^β^ ( referred here as R2Rnc ) and the split-half noise ceiling estimator ( SHnc ) . In each panel , the true value of the noise ceiling ( mean across 100 simulations–red in Fig 4; shaded gray area represents the [5 95] percentile of the true value across the simulations ) is compared to the estimated mean ( blue in Fig 4 –dashed blue lines represent the [5 95] percentiles ) . All estimators showed the same mean across simulations ( blue line ) . However the parametric noise ceiling estimators showed less variability ( dashed lines ) than the non parametric noise ceiling estimator ( SHnc and R2R ) . Between the non parametric noise ceiling estimators the SHnc showed larger variability than the R2Rnc . This simulation proves that all the NC estimators are equivalent when the noise is i . i . d . Fig 5 summarizes the results obtained in estimating the noise ceiling in the case of autocorrelated noise . Each panel reports the result obtained with a different estimation procedure ( parametric estimators–first three panels from the left; R2R noise ceiling and SHnc ) for different levels of the lag-1 autocorrelation in the noise . The true noise ceiling is also reported ( red–mean value across 100 simulations; gray shaded areas [5 95] percentiles ) . The analytical noise ceiling assuming i . i . d noise ( Ω^=I - first panel in Fig 5 ) or using a parametric model of the non-stationarity of the noise ( Ω^NST , third panel in Fig 5 ) result in overestimating the NC ( with larger overestimation at higher levels of autocorrelation in the noise ) . The reason for this overestimation is the underestimation of the variance of the brain response ( V^β^ ) . Note that for Ω^NST we consider only non-stationarity of the noise , but in principle non-stationarity and autoregressive assumptions can be combined in one estimation [30] . In this combined NST-AR ( 1 ) estimation we would expect the estimated noise ceiling to approach the true noise ceiling . All other estimators resulted in unbiased ( i . e . estimated noise ceiling equal to the true noise ceiling ) NC values . Interestingly , violating the noise assumption ( e . g . assuming i . i . d . noise instead of autocorrelated noise ) also results in a lower true noise ceiling . This effect is visible when comparing the true noise ceiling ( red curves ) across the panels in Fig 5 , and is caused by the increased variance in the β^ obtained with e . g . an OLS model compared to estimates obtained with AR ( 1 ) . The performance of different estimators of the noise ceiling under violations of the non-stationarity assumption for the noise is presented in Fig 6 for an SNR level of one ( i . e . when the non-stationarity factor [x-axis in the panels of Fig 6] is equal to one , the simulation is identical to an i . i . d . scenario with SNR = 1 ) . Interestingly , the non-stationarity affected both the true noise ceiling ( red curve and shaded grey area in Fig 5 ) and the estimated noise ceiling ( blue curve and dashed blue curves in Fig 5 ) , that are both underestimated when i . i . d . or AR ( 1 ) assumptions are used . The SHnc showed the larger variability across simulations . This results from the increased variance for the β^ compared to estimates obtained with weighted least squares Ω^NST [30] . The last set of simulations tested the presence of autocorrelation combined with run to run variability in the true β responses . Fig 7 reports results for the different noise ceiling estimation methods at different levels of variance of the noise ( i . e . decreasing SNR ) and for a fixed level of autocorrelation in the noise ( 0 . 25 ) which corresponds to the level of autocorrelation observed in the real fMRI data presented in this article ( see below ) . In this scenario only the analytical noise ceiling based on the run to run variability ( R2Rnc ) and the split-half noise ceiling ( SHnc ) showed mean values consistent with the true noise ceiling ( red curves ) . The SHnc noise ceiling showed the larger variability across all the noise ceiling methods . The left panel of Fig 8 shows , on data simulated with i . i . d . noise , the performance of an encoding model trained using the true underlying computational model ( i . e . an encoding model trained using the matrix X used for simulating the responses ) . The performance is measured as the correlation between predicted β* and the estimated β^ vectors of the test data , for different sample sizes in training ( from ntr = 126 up to 10 times more the number of features ntr = 1260 ) and two levels of regularization ( λ = 100 and λ = 103 ) ( green and red curves in Fig 8 ) . The variance of the noise was fixed to 0 . 5 . As expected , prediction accuracy increased asymptotically with sample size . However , as a consequence of the bias-variance trade-off introduced by the regularization procedure , the performance did not reach the noise ceiling even for a very large number of samples . High regularization implies less variability in the estimated linearized model weights ( i . e . the pRF in fMRI encoding approaches ) as depicted by the narrower [5 95] variability bands . Note that the reduced variability in data poor scenarios comes at the cost of an increased bias highlighted by the larger distance to the noise ceiling in data rich scenarios . The right panel of Fig 8 depicts in more detail the influence of the regularization parameter . By keeping the experimental noise variance and sample size constant ( 0 . 5 and ntr = 126 respectively ) we evaluated the prediction accuracy ( measured as correlation ) of the model underlying the generation of the data and considering the effect of regularization on the difference between the actual model performance and the noise ceiling . The performance did not reach the noise ceiling even when the optimal regularization parameter was selected ( the one that results in the highest performance ) . Using a real fMRI data set we tested the analytical noise ceiling ( using parametric and non-parametric estimates for V^β^ ) as well as the SHnc . Fig 9 shows the results obtained with all the tested approaches in one single individual ( 50000 randomly selected ) . The column on the left in Fig 9 reports the results for the analytical estimator of the noise ceiling using different estimation of β^ and V^β^ . Similarly to simulations with autocorrelated noise , the results indicate that assuming the noise structure to be either i . i . d or non-stationary ( without autocorrelation ) results in a higher estimate of the noise ceiling compared to the estimation obtained using an AR ( 1 ) assumption ( See Fig 5 ) . The second and third column from the left in Fig 9 show estimate of the R2Rnc ( analytical NC using run to run variances ) and SHnc obtained using different estimates for the response vector β^ ( i . e . assuming Ω^=I , Ω^AR ( 1 ) and Ω^NST from top to bottom respectively ) . While small differences between the rows can be observed , the R2R noise ceiling and the SHnc are less affected by the method used for the estimation of the response vector compared to the analytical solution that uses a parametric model of the V^β^ ( left most column in Fig 9 ) . The relationship between the SHnc and the R2Rnc ( analytical NC using run to run variances ) is presented at the upper panel of Fig 10 for the data of one participant ( all the other participants in the dataset showed a similar behaviour ) . The two NC estimators converged to similar values for high signal to noise ratio ( SNR; i . e . when the estimated NC is high ) . This is in line with what we observed in the simulations where the variance of the estimated noise ceiling was low when the SNR was high . In those voxels where the SNR is low , the variability of the estimators results in a low correlation between the two NC methods . The lower panel of Fig 10 , confirmed the equivalence , in real data between the analytical NC of the noise ceiling and the MCnc . Fig 11 shows the results obtained for the R2Rnc ( red ) and the SHnc ( blue ) in ten subjects ( first two rows and the two left most panels in the bottom row ) . These results are obtained on 50000 voxels ( randomly selected ) and assuming i . i . d . noise for estimating the voxel responses . The SHnc showed slightly higher mean values compared with the R2Rnc . These results were obtained including the noise regressors in the fMRI design matrix for estimation of the β ( see section Functional data analysis ) . Removing the noise regressors from the estimation of the responses did not affect the results ( see supplementary figures ) .
The data and the matlab codes are achieved and described at zenodo . org . The voxels time series for one subject , the fMRI design matrix , the β^ images ( assuming Ω^=I ) and the Matlab codes for computing the noise ceiling can be downloaded from the same zenodo url at: https://zenodo . org/deposit/1489531 , doi: 10 . 5281/zenodo . 1489531 . Additional data and codes can be obtained by request to the authors ( a . lagecastellanos@maastrichtuniversity . nl ) .
|
Encoding computational models in brain responses measured with fMRI allows testing the algorithmic representations carried out by the neural population within voxels . The accuracy of a model in predicting new responses is used as a measure of the brain validity of the computational model being tested , but the result of this analysis is determined not only by how precisely the model describes the responses but also by the quality of the data . In this article , we evaluate existing approaches to estimate the best possible accuracy that any computational model can achieve conditioned to the amount of measurement noise that is present in the experimental data ( i . e . the noise ceiling ) . Additionally we introduce a close form estimation of the noise ceiling that does not require computationally or data expensive procedures . All the methods are compared using simulated and real fMRI data . We draw conclusions over the impact of regularization procedures and make practical recommendations on how to report the results of computational models in neuroimaging .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"statistical",
"noise",
"diagnostic",
"radiology",
"functional",
"magnetic",
"resonance",
"imaging",
"data",
"acquisition",
"statistics",
"random",
"variables",
"neuroscience",
"covariance",
"magnetic",
"resonance",
"imaging",
"simulation",
"and",
"modeling",
"mathematics",
"brain",
"mapping",
"neuroimaging",
"research",
"and",
"analysis",
"methods",
"gaussian",
"noise",
"computer",
"and",
"information",
"sciences",
"imaging",
"techniques",
"mathematical",
"and",
"statistical",
"techniques",
"monte",
"carlo",
"method",
"probability",
"theory",
"radiology",
"and",
"imaging",
"diagnostic",
"medicine",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"statistical",
"methods"
] |
2019
|
Methods for computing the maximum performance of computational models of fMRI responses
|
Cryptosporidium is a leading cause of moderate-to-severe diarrhea ( MSD ) in young children in Africa . We examined factors associated with Cryptosporidium infection in MSD cases enrolled at the rural western Kenya Global Enteric Multicenter Study ( GEMS ) site from 2008-2012 . At health facility enrollment , stool samples were tested for enteric pathogens and data on clinical , environmental , and behavioral characteristics collected . Each child’s health status was recorded at 60-day follow-up . Data were analyzed using logistic regression . Of the 1 , 778 children with MSD enrolled as cases in the GEMS-Kenya case-control study , 11% had Cryptosporidium detected in stool by enzyme immunoassay; in a genotyped subset , 81% were C . hominis . Among MSD cases , being an infant , having mucus in stool , and having prolonged/persistent duration diarrhea were associated with being Cryptosporidium-positive . Both boiling drinking water and using rainwater as the main drinking water source were protective factors for being Cryptosporidium-positive . At follow-up , Cryptosporidium-positive cases had increased odds of being stunted ( adjusted odds ratio [aOR] = 1 . 65 , 95% CI: 1 . 06–2 . 57 ) , underweight ( aOR = 2 . 08 , 95% CI: 1 . 34–3 . 22 ) , or wasted ( aOR = 2 . 04 , 95% CI: 1 . 21–3 . 43 ) , and had significantly larger negative changes in height- and weight-for-age z-scores from enrollment . Cryptosporidium contributes significantly to diarrheal illness in young children in western Kenya . Advances in point of care detection , prevention/control approaches , effective water treatment technologies , and clinical management options for children with cryptosporidiosis are needed .
The Global Enteric Multicenter Study ( GEMS ) was undertaken to assess the burden and etiology of moderate-to-severe diarrhea ( MSD ) in seven countries , three in South Asia and four in sub-Saharan Africa . In all African sites , Cryptosporidium was the second-highest enteric pathogen attributable to infant MSD; in GEMS Kenya , Cryptosporidium was a major pathogen across all age groups ( 0–11 , 12–23 , and 24–59 months ) [1] . Cryptosporidium was also identified as one of five pathogens with the highest attributable burden of infant diarrhea in a study of malnutrition and enteric disease ( MAL-ED ) , a cohort study that compared diarrheal and non-diarrheal stools in children under two years old collected at community surveillance visits at 8 sites in South America , Africa , and Asia [2] . Based on GEMS data , it has been estimated that there are nearly three million annual diarrhea episodes attributable to Cryptosporidium in young children in sub-Saharan Africa [3] . Globally , acute Cryptosporidium infections are estimated to cause 48 , 000 annual deaths in children under five years old [4] . Cryptosporidium infections in young children in low- and middle-income countries have been associated with excess mortality [5] , an excess burden of diarrhea later in life [6] , and growth faltering , the deficits of which may not be recovered for those children infected during infancy [7] . Cryptosporidium has been associated with decreases in height-for-age z-scores in children , even in the absence of diarrhea symptoms [4] . Cryptosporidium infections have been associated with persistent diarrhea in Kenya [8 , 9] . Cryptosporidium is highly tolerant to disinfection with chlorine [10] . Nitazoxanide can treat cryptosporidiosis in immunocompetent children 1–11 years old [11]; however , it is not often available in developing countries [12] and is presently not approved for infants [11] . There is currently no vaccine available for Cryptosporidium; however , the evidence of acquired immunity suggests that one could be effective [12] . Although outbreaks of Cryptosporidium in developed countries have been studied in detail , less is known about risk factors for cryptosporidiosis in countries where it is endemic [10] . Reviews of risk factors for Cryptosporidium infection identified malnutrition , contact with domestic animals , non-exclusive breastfeeding in infants , lack of sanitation facilities , and crowded living conditions as possible risk factors for infection in low- and middle-income countries [13 , 14] . Few studies have examined risk factors for Cryptosporidium infection in Kenyan children [15–17] . In Kenya , risk factors for Cryptosporidium in children include being HIV-positive [17] , or having an HIV-positive mother [15] . We describe the prevalence of Cryptosporidium infections in Kenyan children under five years old with MSD , assess clinical , environmental , and behavioral characteristics associated with Cryptosporidium infection , and describe the outcomes and consequences of cryptosporidiosis .
We evaluated data collected in Kenya from cases enrolled in GEMS , a four-year , prospective , age-stratified , health facility-based matched case-control study of MSD among children aged 0–59 months residing within a defined and enumerated population . The rationale , study design , clinical and microbiologic methods , and assumptions of GEMS have been described elsewhere [18 , 19] . Briefly , GEMS enrolled MSD cases from selected sentinel health facilities in each of three age strata ( 0-11 , 12-23 , and 24-59 months old ) , along with 1–3 matched community controls who had not had diarrhea in the week before enrollment . MSD was defined as having three or more loose stools in the previous 24 hours , with onset within the 7 days prior to enrollment , and having one or more of the following illness severity characteristics: loss of skin turgor , sunken eyes , required intravenous fluid rehydration , dysentery ( blood in stool ) , or required hospitalization [18] . At enrollment , demographic , clinical , epidemiological information , and stool samples were collected . Cryptosporidium oocyst antigens were detected in whole stool specimens by enzyme immunoassay ( EIA; TechLab , Inc , VA , USA ) . Detailed laboratory methods are described elsewhere [19] . DNA was extracted from a subset of stools that were EIA-positive for Cryptosporidium . Restriction fragment length polymorphism analyses and DNA sequencing of polymerase chain reaction ( PCR ) products were used to identify Cryptosporidium genotypes for these specimens at the U . S . Centers for Disease Control and Prevention ( CDC ) [20] . To assess each child’s health status , a home visit including focused physical exams and anthropometric measurements was conducted ~60 days ( acceptable range 50–90 days ) following enrollment . Mortality that occurred at any time between enrollment and this follow-up was recorded . In Kenya , children were enrolled between January 31 , 2008 and January 29 , 2011 , and again from October 31 , 2011 to September 30 , 2012 . The study was conducted in Siaya County , in the areas of Gem and Asembo , and during the second enrollment period , in the areas of Asembo and Karemo due to the Kenya Medical Research Institute ( KEMRI ) /CDC health and demographic surveillance system moving activities . This health and demographic surveillance system has been operating in these communities since 2001 . The study setting has high rates of child mortality , malaria , HIV , and tuberculosis , and has been described elsewhere [21 , 22] . Analyses were performed in SAS 9 . 4 ( SAS Institute , Inc . , Cary , NC ) and R 3 . 4 . 0 ( R Foundation for Statistical Computing , Vienna , Austria ) . To assess variables associated with Cryptosporidium positivity , univariable logistic regression models were used to compute odds ratios ( ORs ) and 95% confidence intervals ( CIs ) . Since Cryptosporidium risk factors may be modified by age and the sample size might limit detection of interactions , we assessed for effect modification by age category for all variables in separate models with a p<0 . 05 cutoff for significance . To assess whether each risk factor was confounded by socioeconomic status ( SES ) we ran models with and without SES and considered confounding if effect sizes changed >10% . Two multivariable models were generated to identify clinical , demographic , environmental , and behavioral characteristics associated with being Cryptosporidium-positive . The first model examined the clinical presentation at enrollment of case children , including age strata and sex ( see Table 1 ) and all variables in Table 2 ( except duration of diarrhea , which includes information collected post-enrollment ) . The second model sought to identify demographic , environmental , and behavioral characteristics that may be risk factors for Cryptosporidium infection ( all variables in Table 1 , and the following caretaker-reported water , sanitation , and hygiene characteristics collected at enrollment: primary source of drinking water , whether water was always available from the main drinking water source , whether the child was given stored water in the two weeks prior to enrollment , whether the caretaker boiled or filtered drinking water , whether there was a facility for feces disposal , whether the caretaker uses soap when washing hands , and whether the caretaker washes their hands at the following times: before eating , after defecating , before nursing , before cooking , after cleaning a child , and after touching an animal ) . Breastfeeding was not considered in either model as collinearity with age was identified , and information on breastfeeding is only available for children under two years old in the first three years of GEMS ( n = 1 , 083 ) ; questions on breastfeeding changed during the fourth study year . Model selection was performed using the Least Absolute Shrinkage and Selection Operator ( LASSO ) method using the minimum error lambda [27 , 28] . While LASSO methods were used to identify variables for inclusion , parameter estimates and CIs are derived by standard logistic regression maximum likelihood methods . Written informed consent was collected from all parents of children who participated in GEMS . The GEMS protocol was approved by the Scientific and Ethical Review Committees of KEMRI ( Protocol #1155 ) and the Institutional Review Board ( IRB ) of the University of Maryland School of Medicine , Baltimore , MD , USA ( UMD Protocol #H-28327 ) . CDC ( Atlanta , GA , USA ) formally deferred its review to the UMD IRB ( CDC Protocol #5038 ) .
Among the 1 , 778 MSD case children enrolled , Cryptosporidium was identified in 195 cases ( 11 . 0% ) . Cryptosporidium infections were more frequently identified in infants ( <12 months old ) , with a peak in Cryptosporidium infection at 6–11 months old ( Table 1 and Fig 1 ) . Compared to case children aged 24–59 months , infants had over triple the odds of being Cryptosporidium-positive ( OR = 3 . 32; 95% CI: 2 . 08–5 . 31 ) . Other demographic and household characteristics were similar between Cryptosporidium-positive and Cryptosporidium-negative cases ( Table 1 ) . A non-statistically significant relationship between Cryptosporidium status and having agricultural land was confounded by SES . As no other variable was confounded by SES , only unadjusted effect measures are shown in Table 1 . The clinical presentation of Cryptosporidium-positive and Cryptosporidium-negative cases was similar ( Table 2 ) . Mucus in the stool was significantly associated with being Cryptosporidium-positive ( OR = 1 . 72 , 95% CI: 1 . 21–2 . 51 ) . Only age category and mucus in stool remained in the final multivariable clinical model ( not presented ) . The findings were the same when children who were enrolled multiple times as a case were excluded . Having mucus in the stool remained significantly associated with Cryptosporidium infection controlling for age ( aOR = 1 . 50; 95% CI: 1 . 05–2 . 20 ) . Approximately two-thirds ( 66% ) of Cryptosporidium-positive cases with daily information on diarrhea experienced prolonged or persistent diarrhea , compared to approximately half ( 51% ) of Cryptosporidium-negative cases ( Table 2 ) . Compared to cases experiencing acute diarrhea , cases experiencing prolonged diarrhea were significantly more likely to be Cryptosporidium-positive ( OR = 1 . 68; 95% CI: 1 . 18–2 . 37 ) ; cases experiencing persistent diarrhea were also significantly more likely to be Cryptosporidium-positive compared to cases experiencing acute diarrhea ( OR = 3 . 43; 95% CI: 1 . 97–5 . 98 ) . At enrollment , sex was a significant effect modifier of the relationship between Cryptosporidium and stunting/severe stunting . Among girls , Cryptosporidium-positive cases had significantly greater odds of being stunted at baseline than Cryptosporidium-negative cases ( OR = 1 . 82 , 95% CI: 1 . 10–3 . 01 ) . There were no other statistically significant differences in malnutrition indicators between Cryptosporidium-positive and Cryptosporidium-negative cases at enrollment ( Table 3 ) . At the 60-day follow-up , Cryptosporidium-positive cases had significantly greater odds of being stunted ( aOR = 1 . 65 , 95% CI: 1 . 06–2 . 57 ) , underweight ( aOR = 2 . 08 , 95% CI: 1 . 34–3 . 22 ) , or wasted ( aOR = 2 . 04 , 95% CI: 1 . 21–3 . 43 ) compared to Cryptosporidium-negative cases , controlling for baseline status for each measure ( Table 3 ) . Cryptosporidium-positive cases had significantly larger negative changes in HAZ and WAZ measures from baseline to follow-up . When considering HAZ by sex , female Cryptosporidium-positive cases had significantly larger negative changes in HAZ compared to female Cryptosporidium-negative cases ( Table 3 ) . HIV status was available for 58 . 8% of GEMS-Kenya cases . Of the 114 Cryptosporidium-positive cases with available HIV test results , 5 ( 4 . 4% ) were HIV-positive , compared to 3 . 0% ( 28/932 ) of Cryptosporidium-negative cases ( p = 0 . 39 ) . There was no significant association between being Cryptosporidium-positive and having an HIV-positive biological mother ( n = 1 , 194 tested; OR = 1 . 17; 95% CI: 0 . 77–1 . 77 ) . Most children under two years old ( 81 . 4% ) were partially breastfed . Breastfeeding was similar between Cryptosporidium-positive and Cryptosporidium-negative cases ( Table 4 ) . There was no significant difference in the proportion of Cryptosporidium-positive cases and Cryptosporidium-negative cases who were hospitalized at enrollment ( 13 . 3% vs . 10 . 5% , p = 0 . 24 ) . Among those with 60-day follow-up information , 9 ( 4 . 8% ) of 187 Cryptosporidium-positive cases and 53 ( 3 . 5% ) of 1 , 531 Cryptosporidium-negative cases died by the time of follow-up ( p = 0 . 35 ) . The cause of death , as per verbal autopsy , for the 9 children who died and had Cryptosporidium identified in their stool was as follows: HIV/AIDS related ( n = 5 ) , diarrhea/gastroenteritis ( n = 1 ) , pneumonia ( n = 1 ) , and anemia ( n = 1 ) ; one child did not have a verbal autopsy completed . The most common caretaker-reported primary drinking water sources were rainwater ( 35% ) , surface water ( 31% ) , other improved water sources ( 23% ) , and other unimproved water sources ( 11% ) ; ( Table 5 ) . Compared to cases in households that used rainwater as the primary source of drinking water , case children living in households using other improved sources or unimproved sources ( other than surface water ) had significantly higher odds of Cryptosporidium infection ( OR = 1 . 72; 95% CI: 1 . 14–2 . 58 and 2 . 12; 95% CI: 1 . 31–3 . 41 , respectively ) . Few caretakers of GEMS-Kenya case children reported boiling or using a ceramic filter to treat drinking water; however , those reporting one of these methods had significantly lower odds of Cryptosporidium infection compared to those who didn’t ( OR = 0 . 52 , 95% CI: 0 . 25–0 . 96 ) , predominantly driven by those who boiled water . Only two households reported filtering . Reported handwashing behavior was similar among the caretakers of Cryptosporidium-positive and Cryptosporidium-negative cases ( Table 5 ) . Only age category remained in the final multivariable demographic , environmental , and behavioral characteristics model; thus , this model is not presented . DNA was extracted from a random subset of 64 ( 40% ) of the 160 Cryptosporidium-positive stool specimens from GEMS-Kenya case children enrolled in the first three years . Nested 18S PCR detected Cryptosporidium in 43 ( 67% ) of these specimens . Of the 43 specimens , 35 ( 81% ) were of the species C . hominis and 6 ( 14% ) were C . parvum . C . meleagridis and C . canis were found in one specimen each . Of the 195 Cryptosporidium-positive cases , 142 ( 72 . 8% ) also had one or more additional enteric pathogens identified in their stool; enteric co-infections were common throughout the study population ( Fig 2 ) . The characteristics of case children with only Cryptosporidium detected in their stool were generally similar to Cryptosporidium-positive case children with multiple enteropathogens ( S1 and S2 Tables ) ; the only significant clinical difference was in the child’s mental state at enrollment ( S1 Table ) . Variables chosen for multivariable models were unchanged when excluding children who were enrolled more than once as an MSD case .
This study evaluated the clinical , environmental , and behavioral characteristics associated with Cryptosporidium infection among children under five years old with MSD in rural western Kenya . Overall , 11% of children with MSD had Cryptosporidium identified in their stool; the majority ( 81% ) of genotyped samples were C . hominis . Among MSD cases , being an infant , having mucus in stool , and having prolonged or persistent duration diarrhea were associated with being Cryptosporidium-positive . Boiling drinking water and using rainwater as the main drinking water source appeared to protect against Cryptosporidium infection in MSD cases . Among girls , Cryptosporidium-positive cases were more likely to be stunted at baseline compared to Cryptosporidium-negative cases . Cryptosporidium-positive cases had longer-term consequences in terms of malnutrition , as these children were more likely to stunted , underweight , or wasted at follow-up ( controlling for baseline status ) , and have significantly larger negative changes in height- and weight-for-age z-scores . Except for having mucus in stool , which could be associated with Cryptosporidium adhering to the small intestine mucosa , possibly causing inflammation [29] , the clinical presentation of children with MSD was similar for Cryptosporidium-positive and Cryptosporidium-negative cases , as was observed in another study of Kenyan children with diarrhea [16] . This finding highlights the difficulty in clinically diagnosing Cryptosporidium among children with MSD in this setting and underscores the need for point of care rapid diagnostics for Cryptosporidium . Infants were over three times more likely to have Cryptosporidium identified in their stool compared with children aged 24–59 months . The peak of infection at 6–11 months in this study is similar to the age pattern of Cryptosporidium infections previously reported in sub-Saharan Africa , though an earlier peak than other studies in Kenya [9 , 16] . This timeframe may coincide with the introduction of complementary foods or drinking water . The high prevalence of Cryptosporidium infections in young children is concerning as Cryptosporidium infections in early childhood have been associated with numerous poor outcomes , sometimes lasting beyond the initial infection [6 , 7] , as evident in our findings . Prolonged and persistent duration diarrhea , and growth shortfalls subsequent to enrollment were significantly more pronounced among Cryptosporidium-positive cases compared to other children with MSD . Prolonged and persistent diarrheal episodes occurring in infants have been previously associated with growth shortfalls [30] . The proportion of Cryptosporidium-positive cases who were underweight and wasted increased from baseline to follow-up . This could result from many days of diarrhea experienced by these children . There was also an increase in the proportion of Cryptosporidium-positive cases who were stunted from baseline ( 29% ) to follow-up ( 39% ) . Undernutrition and stunting among children in low- and middle-income countries have predicted decreased performance in school and on cognitive tests in previous research [31] , thus even longer-term consequences could be appreciable although unexplored in the current study . It is estimated that growth faltering contributes substantially to the overall global burden of disease from Cryptosporidium infections in children [4] . By the time of follow-up , 4 . 8% of Cryptosporidium-positive and 3 . 5% of Cryptosporidium-negative cases had died . Although the difference was not statistically significant , this warrants close future attention since other research has shown an association between Cryptosporidium and excess mortality for children who became infected in infancy [5] . Like other studies in Kenya [9 , 16 , 32] , our findings indicate that person-to-person transmission is likely the predominant route for Cryptosporidium infection in rural western Kenya , since the main host for C . hominis is humans [10] . Infections may thus more commonly result from exposure to human feces than animal feces . The presence of animals in the compound was examined in univariable analyses but was not reported in detail or included in the risk factor model selection , as the ownership of many types of animals was not associated with Cryptosporidium infection ( S3 Table ) , though it may be associated with unmeasured confounders ( e . g . , higher income ) . Notably , in another study , C . hominis was associated with more severe clinical symptoms in Kenyan children compared to C . parvum [32] , although we have too little data in GEMS Kenya to examine this . Using rainwater as the main drinking water source was common and was significantly protective against Cryptosporidium infections . Rainwater may be less contaminated with Cryptosporidium , or this finding could be related to the seasonality of Cryptosporidium infections . The proportion of households using rainwater as the main drinking water source varied by month , ranging from 3%-74% . We did not explore Cryptosporidium infections by month/season , as the biweekly enrollment targets for GEMS make interpretation of pathogen-specific seasonal analyses challenging . However , the fact that using rainwater as the primary drinking water source and boiling drinking water were both protective against Cryptosporidium infections indicates that drinking water source choices and certain treatment options may be effective in reducing Cryptosporidium infections and signals that water may play a role in transmission . A limitation of this work is that a comparison could not be made between Cryptosporidium-positive GEMS cases and GEMS controls , as reliable population weights were not available at the time of analysis . The factors associated with Cryptosporidium compared to other individuals with MSD may be different from those risk factors that would be seen when compared to healthy controls . Using MSD as a condition for inclusion for our study may lead to spurious associations , as it is potentially a common effect of both the exposures and the outcome of interest [33] . Our ability to explore data on those with only Cryptosporidium infections was limited due to the small number of single-pathogen Cryptosporidium infections; however , those who presented with Cryptosporidium alone had similar characteristics to those who presented with multiple-pathogen Cryptosporidium infections . We were not able to assess the association between breastfeeding and Cryptosporidium because of ( 1 ) the collinearity between breastfeeding and age , and ( 2 ) the small number of children who were either exclusively or not breastfed in certain age groups . We also could not examine anthropometric outcomes by age or by the number of enteric pathogens isolated in stool , due to the small number of Cryptosporidium-positive children in some age groups and the small number of cases who presented with single-pathogen Cryptosporidium infections . We examined HIV status and malnutrition in our analyses , and performed a sensitivity analysis related to enteric co-infections; however , we did not have information on other co-morbidities that may be associated with Cryptosporidium infection . The burden of diarrhea attributable to Cryptosporidium differed between GEMS and MAL-ED , especially for children 1–2 years old; however , GEMS generally considered more severe cases of diarrhea than MAL-ED . Other differences between the studies have been described elsewhere [34] . While GEMS and MAL-ED found Cryptosporidium to be significantly associated with diarrhea among infants , there were differences between study sites; in other studies Cryptosporidium has been isolated in non-diarrheal stools as often as in diarrheal stools [35 , 36] . Dissimilarities in study design ( e . g . , the time from diarrhea onset to stool collection ) or laboratory methods may partially explain the differences observed , and host susceptibility and other risk factors are likely to vary across settings [35] . However , Cryptosporidium infections have been associated with growth shortfalls in asymptomatic children without diarrhea , thus identification and treatment of Cryptosporidium should remain a priority for young children in settings where it is endemic [35 , 36] . The high prevalence of cryptosporidiosis among young children in our study , coupled with other research that shows extended long-term effects of Cryptosporidium infections and diarrhea early in life , underscores the need for preventive measures aimed at households with young children , as well as improved diarrhea case management . Early diagnosis and management of cryptosporidiosis may mitigate subsequent growth deficits and other long-term consequences . Increased availability of nitazoxanide or new treatments , point of care rapid diagnostics for Cryptosporidium , additional insights into the role of appropriate WASH practices and technologies in childhood cryptosporidiosis , and vaccine development could reduce the burden of disease in such settings . Since Cryptosporidium-positive cases experienced more days of diarrhea and subsequent malnutrition than other MSD cases , increased promotion of the use of zinc in the management of diarrhea , and continued feeding of children with diarrhea should be undertaken , per WHO/UNICEF guidelines [37] . As rotavirus vaccine coverage increases , potentially leading to an altered enteric pathogen landscape , continuing to examine the impact and relative importance of Cryptosporidium infection among infants should remain a priority .
|
Cryptosporidium is an important cause of childhood diarrhea . Research on cryptosporidiosis in countries where it is endemic remains limited; few studies have comprehensively examined risk factors for children in Kenya and similar settings . We examined characteristics associated with Cryptosporidium in children with moderate-to-severe diarrhea in rural western Kenya . We found there is little to clinically distinguish cryptosporidiosis from other childhood diarrhea in the absence of point of care diagnostics . Infants had the highest odds of Cryptosporidium infection; it has been previously established that Cryptosporidium infections in infancy can have severe consequences . Prolonged/persistent duration diarrhea and growth shortfalls were significantly more pronounced among cases with Cryptosporidium . Undernutrition and stunting in children in low- and middle-income countries have predicted decreased cognitive and school performance , thus long-term consequences could be appreciable . Using rainwater as the primary drinking water source and boiling drinking water were protective against Cryptosporidium infection , thus certain water sources may contribute to transmission . Like other studies in Kenya , we predominantly identified Cryptosporidium hominis , an anthropogenic species . Advances in point of care detection , prevention and control approaches , effective water treatment technologies , and clinical management options are needed to mitigate the potentially severe and long-term consequences of Cryptosporidium infection in children .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"neonatology",
"children",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"water",
"resources",
"surface",
"water",
"pathology",
"and",
"laboratory",
"medicine",
"maternal",
"health",
"geographical",
"locations",
"cryptosporidium",
"parasitic",
"protozoans",
"pediatrics",
"health",
"care",
"diarrhea",
"age",
"groups",
"protozoans",
"signs",
"and",
"symptoms",
"women's",
"health",
"gastroenterology",
"and",
"hepatology",
"sanitation",
"africa",
"families",
"hydrology",
"public",
"and",
"occupational",
"health",
"natural",
"resources",
"people",
"and",
"places",
"breast",
"feeding",
"kenya",
"environmental",
"health",
"eukaryota",
"diagnostic",
"medicine",
"earth",
"sciences",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"organisms"
] |
2018
|
Clinical, environmental, and behavioral characteristics associated with Cryptosporidium infection among children with moderate-to-severe diarrhea in rural western Kenya, 2008–2012: The Global Enteric Multicenter Study (GEMS)
|
Intuitively , higher intelligence might be assumed to correspond to more efficient information transfer in the brain , but no direct evidence has been reported from the perspective of brain networks . In this study , we performed extensive analyses to test the hypothesis that individual differences in intelligence are associated with brain structural organization , and in particular that higher scores on intelligence tests are related to greater global efficiency of the brain anatomical network . We constructed binary and weighted brain anatomical networks in each of 79 healthy young adults utilizing diffusion tensor tractography and calculated topological properties of the networks using a graph theoretical method . Based on their IQ test scores , all subjects were divided into general and high intelligence groups and significantly higher global efficiencies were found in the networks of the latter group . Moreover , we showed significant correlations between IQ scores and network properties across all subjects while controlling for age and gender . Specifically , higher intelligence scores corresponded to a shorter characteristic path length and a higher global efficiency of the networks , indicating a more efficient parallel information transfer in the brain . The results were consistently observed not only in the binary but also in the weighted networks , which together provide convergent evidence for our hypothesis . Our findings suggest that the efficiency of brain structural organization may be an important biological basis for intelligence .
Researchers have long studied the biological basis for intelligence and have found increasing evidence relating high performance on intelligence quotient ( IQ ) tests to the coordination of multiple brain regions , utilizing both structural and functional brain imaging techniques [1]–[11] . Our hypothesis , inspired by these earlier findings , is that higher IQ test scores may correspond to more efficient information transfer in the brain . However , no direct evidence has been provided from the perspective of brain networks . In particular the relationship between individual intelligence and topological properties of the brain anatomical network has never been investigated , leaving the impact of brain structural organization on intelligence largely unknown . It is well accepted that the human brain , which can be viewed as a large , interacting and complex network with nontrivial topological properties [12]–[17] , especially with small-world attributes , characterized by a high clustering index and a short average distance between any two nodes [18] , is one of the most challenging systems found in nature . Noninvasive investigation of human brain networks has been enabled by recent advances in modern neuroimaging techniques . Small-world attributes have been found in brain functional networks using electroencephalography , magnetoencephalography and functional magnetic resonance imaging [13]–[17] , [19] . Also , recent progress has been made in the investigation of brain anatomical networks by He et al . [20] , who investigated patterns of anatomical connections in cerebral cortices in vivo using cortical thickness measured from structural magnetic resonance imaging ( MRI ) . Their findings supported the view that human brain anatomical networks manifest small-world attributes . However , only one binary anatomical network could be generated from a group of subjects by their method , which made it inapplicable for investigating the network properties of an individual brain . In addition to He et al . 's cortical thickness measurements , an anatomical network was derived from the inter-regional covariation of the gray matter volume by Bassett et al . using MRI data from 259 healthy volunteers [21] . In this data classical divisions of the cortex ( multimodal , unimodal and transmodal ) showed distinct topological distributes . Diffusion imaging is a relatively new MRI technique , which can visualize brain white matter fiber tracts in vivo [22]–[28] , and has been recently used to investigate human brain anatomical networks . Hagmann et al . made the first attempt by applying diffusion spectrum imaging to two healthy volunteers and was thus the first to confirm small-world topology in the anatomical networks of individual brains [29] . They further extended their investigation into the dense network of cortico-cortical axonal pathways and revealed a structural core in the human cerebral cortex [30] . Another study performed by Iturria-Medina et al . established a weighted anatomical network for individual brains using diffusion tensor imaging ( DTI ) and graph theory; they also found small-world properties of the networks across 20 subjects [31] . However , their approach will sometimes result in assigning a nonzero connection probability value to brain region pairs which are unlikely to be connected ( e . g . , left frontal and right occipital cortex ) [31] . In a recent study by Gong et al . [32] , a macro scale anatomical network was established across 80 healthy volunteers using diffusion tensor tractography ( DTT ) . The entire cerebral cortex was subdivided into 78 regions , not including the subcortical structures , using automated anatomical labeling ( AAL ) . Their findings suggested prominent small-world attributes which are generally compatible with the findings of previous studies . However , only one group-based binary network was generated from all subjects using their approach , leaving the investigation of individual brains and the construction of weighted brain networks unstudied . In the present study , we tested the hypothesis that individual intelligence is associated with the individual's brain structural organization . Specifically , higher intelligence test scores correspond to a higher global efficiency of the individual's brain anatomical network . We performed our study on 79 healthy young adults , basically using the DTT method proposed by Gong et al . [32] with some modifications to allow the method to fit our goal . First , we constructed a binary anatomical network of the individual brain of each subject using a modified method , in which subcortical structures ( i . e . the thalamus ) were included and a robust algorithm for fiber tracking was employed . Secondly , we developed the binary networks into weighted ones by introducing an appropriate index to achieve a more complete picture for our investigation . Thirdly , topological properties of the binary and the weighted anatomical networks of each subject were calculated and used for the small-world evaluation . Fourthly , depending on their IQ tests scores , all healthy adults were divided into general intelligence ( GI ) and high intelligence ( HI ) groups , and a two-sample t-test of network properties was performed between the two groups . Finally , partial correlation analyses were performed between the IQ scores and the topological properties of brain anatomical networks across all subjects while controlling for age and gender . To obtain convergent evidence from the test of our hypothesis , both inter-group comparisons and partial correlation analyses were performed on the binary and the weighted networks; we also reproduced our investigation utilizing different brain parcellation schemes for network construction as well as different indices for weighted network construction .
We successfully constructed binary and weighted anatomical networks for each of the 79 subjects in the form of symmetric connectivity matrixes using our method ( see Materials and Methods , Fig . 1 , Tables 1 and 2 ) . Figures 2 and 3 show the mean map which was obtained by averaging across the binary connectivity matrixes of all 79 subjects ( Fig . 2 ) as well as a 3D representation of the network in anatomical space ( Fig . 3 A , B and C ) . The network is primarily comprised of intra-hemispheric connections with a few major inter-hemispheric connections . This connection pattern is generally comparable with previous brain anatomical network studies utilizing MRI and diffusion imaging data [20] , [30]–[32] . Please note that we constructed the network showed in Figs . 2 and 3 using a threshold value of 3 ( see Materials and Methods ) . In addition , six well-known white matter fiber tracts - the genu of the corpus callosum ( CC ) , the body of the CC , the splenium of the CC , the cingulum , the corticospinal tract and the inferior frontooccipital fasciculus - were further constructed in three randomly selected subjects utilizing our fiber tracking method and are presented in Fig . 4 . We used the AAL regions as seed regions and some extra ROIs as filters which are necessary for correctly reconstructing the six fiber tracts . In detail , the filter ROIs for the corpus callosum were placed on the midsagittal planes; the ROIs for the cingulum were placed through the genu-trunk junction and the trunk-splenium junction of the corpus callosum in coronal planes; the ROIs for the corticospinal tract were placed in the posterior limb of the internal capsule and the pre- and postcentral gyri respectively; and the ROIs for inferior frontooccipital fasciculus included large part of the entire frontal and occipital lobes [33] , [34] . The trajectories of these major white matter tracts are consistent with the existing anatomical knowledge-base [35] as well as with a previous DTI study [36] . This consistency with anatomical and DTI information may provide further support for the validation of our constructed network . Network measures included the total number of edges , absolute clustering coefficient , mean characteristic path length and global efficiency of the network as well as the small-world indices and ( see Materials and Methods ) . The average value of these topological properties of the binary and the weighted networks across all the 79 subjects are listed in Table 3 along with the results of previous studies on functional and anatomical human brain networks at a macro scale level [15] , [16] , [20] , [31] , [32] . Our results are very compatible with these previous findings . In particular , a prominent small-world attribute was consistently observed in the binary networks of all 79 healthy volunteers . In addition , we examined the hub regions and degree distributions of the binary networks we constructed . These examinations showed consistent results with previous studies of functional or anatomical networks , providing further support for our current study ( Details can be found in Text S3 ) . As shown in Table 4 , significant differences in network properties were found between the GI and HI groups by a two-sample t-test ( see Materials and Methods ) : was significantly larger in the HI group; the of the binary and the weighted networks was significantly shorter in HI group; the of the binary and the weighted networks was significantly higher in HI group; no significant difference in was observed between the two groups in the binary and weighted networks . In most cases , the weighted networks showed a much smaller P-value than the binary networks , suggesting that the differences in network properties between these two groups were more significant in the weighted networks . Please note that these results were observed using a threshold value of 3 to construct the network ( see Materials and Methods ) . To explore the dependence of our results on our choice of threshold , we reproduced the two-sample t-test between the GI and HI groups on binary and weighted networks that we constructed using five different threshold values ranging from 1 to 5 . Similar results were consistently observed , suggesting that our findings are relatively robust ( Table 4 ) . Intelligence test scores included full scale IQ ( FSIQ ) , performance IQ ( PIQ ) and verbal IQ ( VIQ ) ( see Materials and Methods ) . As shown in Table 5 , significant correlations between the intelligence test scores and the topological properties of the binary and the weighted anatomical brain networks were found by partial correlation analyses in all 79 subjects , when the data were controlled for age and gender ( see Materials and Methods ) : was found to be positively correlated to FSIQ and PIQ ( Fig . 5 ) ; for the binary networks , was found to be negatively correlated to FSIQ and PIQ , and for the weighted networks , was found to be negatively correlated to FSIQ , PIQ and VIQ ( Fig . 6 ) ; was found to be positively correlated to FSIQ , PIQ and VIQ in the binary and the weighted networks for all subjects ( Fig . 7 ) ; no significant correlation was found between and the intelligence tests scores . In most cases , the weighted networks showed a much larger absolute value of the partial correlation coefficient and a much smaller P-value than the binary networks , suggesting that the correlations were stronger and more significant in the weighted networks . Having established that changing the threshold values did not change our overall conclusions , we will use a threshold value of 3 throughout the rest of the Results section . To further localize the association with intellectual performance , the local efficiency ( ) of each node region was calculated for each subject ( see Materials and Methods ) . As shown in Tables 6 and 7 , when we controlled for age and gender , we found significant correlations ( , uncorrected ) using partial correlation analyses performed across all subjects between their intelligence test scores and the local efficiency ( ) of multiple brain regions , which were located in widely-distributed areas across the brain . These involved cortical areas in the parietal , temporal , occipital and frontal lobes as well as subcortical structures such as the thalamus , amygdala and hippocampus . We reproduced our investigations utilizing different brain parcellation schemes for network construction ( see Text S1 ) as well as different indices for weighted network construction ( see Text S2 ) . In each of these situations , we calculated the topological properties of brain networks for small-world evaluation and performed statistical analyses , including inter-groups comparisons and partial correlation analyses between IQ scores and brain network properties across all subjects as well . The results of these analyses showed that , in most of the tested situations , prominent small-world attributes were consistently observed across all the 79 subjects ( see Text S1 ) . More importantly , significantly higher global efficiencies of the brain networks were consistently observed in the HI group ( see Text S1 and Text S2 ) , and significant correlations were consistently found between specific IQ scores and brain network properties ( see Text S1 Text S2 as well as Figs . S1 , S2 , S3 , S4 and S5 ) . In particular , higher intellectual performance corresponds to better global efficiency of the brain networks . These comprehensive analyses provide convergent evidence for the validity of our findings .
In this study , global efficiency of the brain anatomical network was higher in the HI groups than in the GI groups , and positive correlations between intelligence tests scores and the global efficiency of the networks were found in all the healthy young adults while controlling for age and gender . These findings were consistently observed in the different situations we tested , including the binary and the weighted networks we constructed , the different brain parcellation schemes we employed ( see Text S1 ) and the various indices we used for weighted network construction ( see Text S2 ) . Many previous studies have related intelligence to different structural and functional properties of the brain . Positive correlations between IQ and total brain volume have been reported by several research teams who used structural imaging techniques on different populations with different scan protocols and different intelligence measures [41]–[46] . Utilizing voxel-based morphometry methods , recent studies have revealed correlations between IQ and certain specific brain regions involving the frontal , parietal , temporal and occipital lobes [3]–[5] , [47]–[50] . Several previous functional imaging studies , using intellectually demanding tasks ranging from working memory to a variety of verbal and non-verbal reasoning , have also shown that people who performed well on intelligence related tests recruited multiple brain regions [1] , [2] , [9] , [51] . Although none of these previous studies investigated the issue from the perspective of brain networks , they can nonetheless provide support for our current findings . Partial correlation analyses performed across all subjects while controlling for age and gender revealed significant correlations between intelligence test scores and the local efficiency ( ) of multiple brain regions , including cortical areas located in the parietal , temporal , occipital and frontal lobes as well as subcortical structures such as the thalamus , amygdale and hippocampus ( Tables 6 and 7 ) . Please note that the significance level for our partial correlation analyses of the local efficiency ( ) was set at and was uncorrected for multiple comparisons across all the 90 brain regions . An even higher level of significance might be able to be achieved in future studies by including more subjects . However , although the interpretation of our results must be cautious , our findings appear to provide new evidence for the biological basis of intelligence from a network perspective . In particular , in one recent review of 37 neuroimaging studies associated with the neural basis of intelligence [11] , Jung and Haier found that individual differences in intelligence were closely related to variations in a distributed brain network which included multiple brain regions located in the dorsolateral prefrontal cortex , the inferior and superior parietal lobe , the anterior cingulate , the temporal and the occipital lobes . Our investigations may provide evidence for their findings from a brain anatomical network perspective , and more importantly , our findings may indicate that the efficient organization of the brain anatomical network may be important for individual intellectual performance . In a recent study performed by our group [34] , a partial correlation analysis on the same 79 healthy volunteers together with 15 mental retardation patients controlling for age and gender showed that FSIQ scores were significantly correlated with the FA value of the bilateral uncinate fasciculus , the genu and truncus of the corpus callosum , the bilateral optic radiation and the left corticospinal tract . Significant correlation was also found between the FSIQ scores and the FA of the right UF when further controlling for group identity between patient and normal control [34] . The findings of this earlier research provide structural evidence for our current investigation by showing that the integrity of the major white matter bundles , which was measured by the FA value , may be an important biological basis for human intelligence . The results of our current study show that higher intelligence test scores are related to a larger global efficiency ( ) of the brain anatomical network ( Table 5 and Fig . 7 ) , which may indicate better parallel information transfer in the brain [52] . According to the DTT method , in which the propagation of fiber tracking depends on white matter integrity as measured by the FA value , we may speculate that the more efficient network organization associated with better intellectual performance may relate to increased white matter integrity , not only in the major fiber bundles investigated in our previous study but also in the white matter connectivity across the whole brain . Our findings support the previous finding that cognitive processes are dependent upon the fidelity of the underlying white matter to facilitate the rapid and error-free transmission of data between different brain regions [11] . In another resting state functional MRI study on a subset of the same 79 healthy adults ( 59 subjects ) performed by our group [6] , brain regions in which the strength of functional connectivity significantly correlated with intelligence scores were distributed in the frontal , parietal , occipital and limbic lobes . This gives increased credence to our current study by supporting a network view of intelligence from functional imaging evidences , thus revealing that brain activity may be relevant to differences in intelligence even in the resting state [6] . Subjects with higher IQ scores consistently showed more edges ( ) and shorter characteristic path lengths ( ) in the various situations which we tested . This is consistent with previous findings that short paths in brain networks assure effective integrity or rapid transfer of information between and across remote regions that are believed to constitute the basis of cognitive processes [12] . A previous study performed by Kaiser and Hilgetag [53] demonstrated that neural systems are not optimized exclusively for minimal global wiring length , but for a variety of factors including the minimization of processing steps . Although not completely comparable in data types and analysis methods , our finding of shorter characteristic path lengths ( ) in the subjects with higher IQ scores may reflect fewer signal processing steps between brain regions . As reviewed by Roth and Dicke [54] , no universally accepted definition of animal intelligence exists; nor has any procedure for measuring it come to dominate the field . One view that has emerged from previous studies of comparative and evolutionary psychologists and cognitive ecologists is that animal intelligence can be defined as the degree of mental or behavioral flexibility resulting in novel solutions , either in the wild or in the laboratory [54]–[57] . According to review studies of previous intelligence investigations [11] , [54] , various brain properties such as brain volume , relative brain volume and encephalization quotient have been assumed to be relevant for intelligence . However , although humans are generally considered to be the most intelligent species , they do not have the largest brain or cortex , either in absolute or relative terms . But they do have the largest number of cortical neurons and a relatively high conduction velocity between those neurons , which appears to correlate better with intelligence as the basis for information processing capacity [54] . Significantly , myelinated cortical fibers are relatively thin in elephants and cetaceans , but particularly thick in primates [58] , [59] , contributing to a better conduction velocity . This supports the idea that an increase in information processing capacity is of great importance for intelligence [54] . In our study , intelligence test scores were found to be significantly correlated to the complex brain network topological properties derived from a fiber tracking method based on DTI . Our results appear to support previous findings since DTI is currently the only noninvasive brain imaging technique that can explore the structure of white matter in vivo and provide information about the white matter integrity of cortical fibers , a topic which is obviously closely related to fiber myelination [28] , [60] , [61] . However , more extensive future analyses are necessary to clarify more clearly the relationship between the complex brain network topological parameters that we calculated and the conduction velocity between neurons and to determine how these are related to the information processing capacity of the human brain . In conclusion , we successfully constructed binary and weighted anatomical networks of the individual brains of 79 healthy adults . These networks showed topological properties that included a prominent small-world attribute that was quite comparable with the findings of previous human brain network studies . More importantly , extensive analysis consistently revealed significant correlations between intelligence test scores and brain anatomical network properties across all subjects , providing convergent evidence for our hypothesis that a more efficient brain structural organization may be an important biological basis for higher intelligence . Our study may provide new clues for understanding the mechanism of intelligence .
It should be noted that the healthy adults included in this current work have been used in previous studies performed by our group for different purposes [6] , [34] , [62] . However , we will again present the description of these adults in detail here in order to clearly present our current investigation . Seventy-nine normal subjects ( 44 males and 35 females , mean age = 23 . 8 years , range = 17–33 years ) were recruited by advertisement . Each subject was examined using the Chinese Revised Wechsler Adult Intelligence Scale ( WAIS-RC ) [63] . Across all subjects , the mean FSIQ was 113 . 7 ( range = 71–145 ) ; the mean test score of PIQ was 110 . 6 ( range = 64–153 ) ; and the mean test score of VIQ was 114 . 4 ( range = 76–140 ) . All subjects were right-handed and Han Chinese in origin . After a full explanation , all subjects gave voluntary written informed consent according to the standards set by the Ethical Committee of Xuanwu Hospital of Capital Medical University . Diffusion tensor images of all the subjects were obtained on a 3 . 0-T Siemens MRI scanner . A single shot echo planar imaging sequence ( TR = 6000 ms , TE = 87 ms ) was employed . Diffusion sensitizing gradients were applied along 12 non-collinear directions ( b = 1000 s/mm2 ) , together with a non-diffusion-weighted acquisition ( b = 0 s/mm2 ) . An integrated parallel acquisition technique was used with an acceleration factor of 2 , which can reduce the acquisition time with less image distortion from susceptibility artifacts . From each subject , 45 axial slices were collected . The field of view was 256 mm×256 mm; the acquisition matrix was 128×128 and zero filled into 256×256; the number of excitations was 3; and the slice thickness was 3 mm with no gap , which resulted in a voxel-dimension of 1 mm×1 mm×3 mm . A 3D T1-weighted image for each subject was obtained using a magnetization prepared rapid gradient echo sequence . The imaging parameters were a field of view of 220 mm×220 mm , TE of 2 s , TR of 2 . 6 ms , flip angle of 9° , and a voxel-dimension of 1 mm×1 mm×1 mm . Both the DTI data and T1-weighted data were visually inspected by two radiologists for apparent artifacts arising from subject motion and instrument malfunction . Distortions in the diffusion tensor images caused by eddy currents and simple head motions were then corrected by FMRIB's Diffusion Toolbox ( FSL 4 . 0; http://www . fmrib . ox . ac . uk/fsl ) . After correction , three-dimensional maps of the diffusion tensor and the FA were calculated using the DtiStudio software [64] . T1-weighted images of each subject were co-registered to the subject's non-diffusion-weighted image ( b = 0 s/mm2 ) using the SPM2 package ( http://www . fil . ion . ucl . ac . uk/spm ) , resulting in a co-registered T1 image ( rT1 ) in DTI space . Subsequently , DTT was performed on every subject . Seed points were selected as voxels with an FA value greater than 0 . 3 in each node region [66] . The AAL template is not a pure cortical grey matter mask but includes tissues from both cortical grey matter and subcortical white matter [65] . Selecting seed voxels with the criteria of FA>0 . 3 in every node region helped to ensure that the trajectories we got originated from the white matter tissue underlying the cortical region or adjacent to subcortical structures . A tensorline tracking algorithm , which approximates the direction of fiber propagation by combining the major eigenvector of the tensor , the vector of previous propagation step and the entire tensor itself [67] , [68] , was implemented using an in-house program developed in the Matlab 7 . 0 platform . Several previous studies have demonstrated that tensorline tracking methods can achieve robust and reproducible results for fiber bundles reconstruction [68]–[71] . This was helpful when subcortical structures were included for examination in our current study . The tracking procedure was terminated at voxels with an FA value of less than 0 . 15 or when the angle between adjacent steps was greater than 45° [66] . Two AAL node regions i and j were considered to be connected if the reconstructed fiber bundles with two end points located in these two regions respectively were present [32] . However , considering the limited resolution of DTI and the capacity of the deterministic tractography method we employed , there is a risk that some false-positive connections will be included . This possibility may increase if only a few fiber bundles are reconstructed between two node regions . In this situation the apparent connections may be the result of noise . To address this issue , a threshold value for the number of presented fibers was utilized to exclude connections between regions that have too few reconstructed fiber bundles to be certain of their validity . On the other hand , some false-negative connections ( that is , connections that are real , but are rejected as false ) might be excluded when a relatively large threshold value was used . To determine the most appropriate threshold , we tested values from 1 to 5 and calculated the topological properties for the resultant networks of every subject at each tested value . Based on the results showed in Table 2 , we chose a value of 3 , which was the highest threshold that maintained the average size of the largest connected component at 90 across all subjects , meaning that the 90 brain regions in the network were all connected at this threshold value in the majority of the 79 subjects . A binary symmetric connectivity matrix was obtained for each subject using the above procedures . Please note that to further examine how dependent the results of our study are on the choice of different threshold values , most of the subsequent statistical analyses were also performed on the topological properties of networks constructed using each of the different threshold values ranging from 1 to 5 . A two-sample t-test on the properties of binary and weighted networks was performed between the GI and HI groups using SPSS13 . 0 , and a threshold value was set at for significance . Please note that our database of healthy adults was divided into GI ( 70<FSIQ<120; 22 men and 20 women; age , 22 . 8±4 . 1 years ) and HI ( FSIQ> = 120; 22 men and 15 women; age , 24 . 9±3 . 3 years ) groups according to their FSIQ scores in the same manner as in the previous study by our group [34] , which was performed on the same dataset , for the sake of methodological consistency . We believe that an explanation for our choice of an FSIQ score of 120 as the cut-off value for general and high IQ groups division will be helpful for clarifying this study . In the Chinese Revised Wechsler Adult Intelligence Scale ( WAIS-RC ) we used , IQ classification in educational use is defined as: ( 1 ) Extremely Low ( 69 and below ) ; Borderline ( 70–79 ) ; ( 3 ) Low Average ( 80–89 ) ; ( 4 ) Average ( 90–109 ) ; ( 5 ) High Average ( 110–119 ) ; ( 6 ) Superior ( 120–129 ) ; ( 7 ) Very Superior ( 130 and above ) . The IQ score of 120 is the cutting point which can be used to identify the subjects with “superior” and “very superior” intelligence . In addition , there are two previous studies which support this cutoff . In Waldmann's et al . [78] study , subjects between the ages of 18 and 30 were divided into groups based on their Satz-Mogel Wechsler Adult Intelligence Scale-Revised FSIQ scores: ( a ) Borderline ( 70 to 79 ) ; ( b ) Low Average ( 80 to 89 ) ; ( c ) Average ( 90 to 109 ) ; ( d ) High Average ( 110 to 119 ) ; ( e ) Superior ( 120 to 129 ) . In another study by Karande et al . [79] , ninety-five children with specific learning disabilities ( aged 9–14 years ) were divided into groups based on their nonverbal IQ scores obtained on the Wechsler Intelligence Scale for Children test: ( i ) average-nonverbal intelligence group ( IQ 90–109 ) , bright normal-nonverbal intelligence group ( IQ 110–119 ) , and ( iii ) superior-nonverbal intelligence group ( IQ 120–129 ) . In both studies , an IQ score of 120 was used as the cutoff for identifying the “superior group” . Because these two studies are basically comparable to our current study ( although differing in populations , intelligence scale editions and IQ scores ) they add credibility to the IQ cutoff in our investigation . Partial correlations between intelligence test scores and global brain network properties ( E , , , ) were performed across all subjects using SPSS 13 . 0 , while controlling the effects of age and gender . The threshold value was set at for significance . Furthermore , to localize the association with intellectual performance , partial correlations were also performed between the local efficiency ( ) of each node region and the intelligence test scores across all subjects , while controlling for age and gender . The threshold value was set at for significance ( uncorrected ) . There are several methodological issues in our present study that need to be addressed . First , a deterministic tractography method was utilized for network construction . We realize that this kind of fiber tracking method has a limited capacity for resolving crossing fiber bundles [71] , which may lead to the loss of some existing fiber connections between brain regions or to the inclusion of some non-existent fibers . A probabilistic tractography method may be a better solution for future work as recent studies have demonstrated that this method is advantageous for overcoming the fiber crossing problem [27] , [80] , [81] . However , it is not applicable in our current investigation as only 12 diffusion directions were employed for data acquisition , an insufficient number for performing a valid probabilistic tracking method . However , the tensorline tracking method we used has been shown to be able to achieve robust and stable tracking results [68]–[71] , which would help to increase the validity of our network construction . Second , in contrast to a population-based network analysis , which may tend to exclude false-negative connections [32] , our analysis of individual brain networks may lead to false-positive connections in each individual subject as a result of limitations that may arise from the image resolution and the tracking method . To increase the reliability of our work , we employed a threshold value on every individual brain network to exclude regional connections that have too few existing fiber bundles to be valid . Since the threshold value was carefully tested ( see Table 2 ) and since consistent , stable results were obtained across all the different situations we tested , we believe that our investigation of individual brains was basically valid . However , more datasets using different populations should be tested in the future for further evaluation of our method . Third , we developed our investigations from a binary to a weighted anatomical network by introducing different weighted indices . Although no existing studies can directly validate our method , our results showed that either using the number or the average FA value of the existing fiber bundles between two regions can lead to a network topology similar to that found in previous human brain network studies ( see Text S2 ) . The results of the statistical analyses indicated that using the number of fiber bundles that link two regions as an edge weight may be more appropriate than using the average FA when investigating the network properties associated with intellectual performance ( see Text S2 ) . We realize that the number of fiber bundles that we used here cannot represent the actual number of axonal fibers , but rather indicates the strength of the white matter connectivity between different brain regions . Although our findings provided relatively good support for this weight index , further examinations on other datasets are necessary . Finally , a risk of this study is that some of the fiber tracts reconstructed by our method may not belong to the specific AAL region . This could happen if the white matter voxels included in the fiber tracking procedure were not truly adjacent to the cortex . Additionally , the choice of the relatively high FA threshold of 0 . 3 for the seed voxel in our current study might increase this possibility , since it may exclude low FA sub-cortical white matter areas as seed regions . To address this issue Gong et al . [32] removed white matter voxels from the unanalyzed AAL cortical mask if no cortical voxels existed within 2 mm3 of them . We believe that they have made an original and creative contribution to this issue . On the other hand , because no gold standard for identifying the nature of the removed white matter voxels exists , their method could lead to a risk of excluding fiber tracts that actually belong to the specific AAL region . This exclusion of potentially significant fiber tracts could subsequently affect the topological properties of the resulting brain anatomical network . Here , we would like to point out that the FA threshold of 0 . 3 we used in the current study was selected based on a somewhat similar study performed by Thottakara et al . [66] , in which an FA threshold of 0 . 3 was used for selecting seed voxels to reconstruct fiber tracts originating from or terminating in different Brodmann areas utilizing the streamline tracking method . Although the details of our current study and theirs are not completely comparable , we believe that the FA threshold of 0 . 3 we used is basically valid , considering that the DTI images in our study were obtained from a 3 . 0-T MRI scanner using 12 non-collinear diffusion encoding directions , which are the same as those used in their study . Nevertheless , future investigations using a more sophisticated brain template will be necessary to better address this methodological limitation of our current study .
|
Networks of interconnected brain regions coordinate brain activities . Information is processed in the grey matter ( cortex and subcortical structures ) and passed along the network via whitish , fatty-coated fiber bundles , the white matter . Using maps of these white matter tracks , we provided evidence that higher intelligence may result from more efficient information transfer . Specifically , we hypothesized that higher IQ derives from higher global efficiency of the brain anatomical network . Seventy-nine healthy young adults were divided into general and high IQ groups . We used diffusion tensor tractography , which maps brain white matter fibers , to construct anatomical brain networks for each subject and calculated the network properties using both binary and weighted networks . We consistently found that the high intelligence group's brain network was significantly more efficient than was the general intelligence group's . Moreover , IQ scores were significantly correlated with network properties , such as shorter path lengths and higher overall efficiency , indicating that the information transfer in the brain was more efficient . These converging evidences support the hypothesis that the efficiency of the organization of the brain structure may be an important biological basis for intelligence .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"neuroscience/cognitive",
"neuroscience",
"radiology",
"and",
"medical",
"imaging/magnetic",
"resonance",
"imaging",
"computational",
"biology/computational",
"neuroscience"
] |
2009
|
Brain Anatomical Network and Intelligence
|
Common diseases such as type 2 diabetes are phenotypically heterogeneous . Obesity is a major risk factor for type 2 diabetes , but patients vary appreciably in body mass index . We hypothesized that the genetic predisposition to the disease may be different in lean ( BMI<25 Kg/m2 ) compared to obese cases ( BMI≥30 Kg/m2 ) . We performed two case-control genome-wide studies using two accepted cut-offs for defining individuals as overweight or obese . We used 2 , 112 lean type 2 diabetes cases ( BMI<25 kg/m2 ) or 4 , 123 obese cases ( BMI≥30 kg/m2 ) , and 54 , 412 un-stratified controls . Replication was performed in 2 , 881 lean cases or 8 , 702 obese cases , and 18 , 957 un-stratified controls . To assess the effects of known signals , we tested the individual and combined effects of SNPs representing 36 type 2 diabetes loci . After combining data from discovery and replication datasets , we identified two signals not previously reported in Europeans . A variant ( rs8090011 ) in the LAMA1 gene was associated with type 2 diabetes in lean cases ( P = 8 . 4×10−9 , OR = 1 . 13 [95% CI 1 . 09–1 . 18] ) , and this association was stronger than that in obese cases ( P = 0 . 04 , OR = 1 . 03 [95% CI 1 . 00–1 . 06] ) . A variant in HMG20A—previously identified in South Asians but not Europeans—was associated with type 2 diabetes in obese cases ( P = 1 . 3×10−8 , OR = 1 . 11 [95% CI 1 . 07–1 . 15] ) , although this association was not significantly stronger than that in lean cases ( P = 0 . 02 , OR = 1 . 09 [95% CI 1 . 02–1 . 17] ) . For 36 known type 2 diabetes loci , 29 had a larger odds ratio in the lean compared to obese ( binomial P = 0 . 0002 ) . In the lean analysis , we observed a weighted per-risk allele OR = 1 . 13 [95% CI 1 . 10–1 . 17] , P = 3 . 2×10−14 . This was larger than the same model fitted in the obese analysis where the OR = 1 . 06 [95% CI 1 . 05–1 . 08] , P = 2 . 2×10−16 . This study provides evidence that stratification of type 2 diabetes cases by BMI may help identify additional risk variants and that lean cases may have a stronger genetic predisposition to type 2 diabetes .
Common diseases such as type 2 diabetes are highly phenotypically heterogeneous . Few studies have performed genome wide association studies in subsets of patients defined by more stringent phenotypic characteristics . It is possible that reducing the heterogeneity of disease cases may increase power to detect associations over and above the loss of power resulting from reduced numbers . To address these questions we hypothesized that the genetic predisposition to Type 2 diabetes may be different in two strata of cases defined by well-accepted cut-offs for body mass index , the strongest known risk factor for type 2 diabetes . Genome-wide association ( GWA ) studies have identified ∼50 independent loci robustly associated with type 2 diabetes [1] , [2] , [3] , [4] , [5] , [6] , [7] . These studies have highlighted new candidate pathways involved in the disease [8] , [9] , identified overlap with monogenic forms of the disease [1] , and provided genetic links with correlated phenotypes [10] , [11] . The GWA studies of type 2 diabetes have not so far provided a greatly improved understanding of the clinical heterogeneity of the disease . Type 2 diabetes cases vary appreciably in their clinical characteristics , particularly age of diagnosis and body mass index ( BMI ) . There is also a group of patients who may present with evidence of an autoimmune component to their diabetes , but who are not insulin dependent [12] . In contrast , the identification of the genetic component to monogenic forms of diabetes has often explained the clinical heterogeneity observed [13] . Previous studies have provided some evidence of genetic heterogeneity between non-obese and obese type 2 diabetic cases [14] , [15] , [16] , [17] . For example , the variant with the strongest effect on type 2 diabetes risk , in TCF7L2 , has a stronger effect in non-obese cases ( odds ratio = 1 . 53 [0 . 37–1 . 71] compared to obese cases ( OR = 1 . 21 [1 . 09–1 . 35] ) [14] . The effect of FTO variation on type 2 diabetes risk depends on how cases and controls are ascertained by BMI status , but this was expected given FTO's known primary effect on BMI . In the most recent GWA studies of type 2 diabetes [1] , risk variants tended to have stronger effects in non-obese compared to obese individuals – of 30 loci examined , 23 showed stronger associations in non-obese compared to obese individuals . We designed the present study in an attempt to understand better the genetic heterogeneity of type 2 diabetes . Type 2 diabetes GWA studies tend to be enriched with cases with stronger family histories and lower average BMIs compared to community based studies . Nevertheless , there is a wide spectrum of BMI amongst type 2 diabetes cases used in GWA studies , with more cases being obese than lean . In this study we tested the hypothesis that we would identify new genetic variants by limiting the clinical heterogeneity of type 2 diabetes . By stratifying cases by their BMI status and performing separate GWA studies for each strata of BMI we identified two signals of association not previously reported in the largest GWA studies in Europeans [1] , although one signal has been identified in a South Asian study [7] . In addition we confirmed with additional data that the majority of known type 2 diabetes genetic associations have stronger effects in lean type 2 diabetic cases compared to obese cases .
To test the hypothesis that we would identify new variants associated with type 2 diabetes in different BMI strata , we used the following study design . We used two separate strata of type 2 diabetes cases defined by the two arbitrary , but well established , cut-offs for classifying people as overweight or obese . The first stratum consisted of non-overweight cases , here defined as “lean” ( BMI<25 kg/m2 ) . The second strata consisted of obese cases ( BMI≥30 kg/m2 ) . For each stratum we used all controls , not selected on BMI to increase statistical power and provide a more robust estimate of the population allele frequency . We did not correct for BMI as BMI was not available in all controls . To check whether or not associations were being driven primarily by effects on BMI we assessed novel variants in an existing GWA studies of BMI using 123 , 865 individuals from the GIANT consortium [18] . Finally , we performed sensitivity analyses , confirming our findings by stratifying controls by BMI as well as cases . We chose to include the largest set of studies available . These studies differed in the proportion of total cases defined as lean ( 8 . 4–30 . 4% ) , the proportion of total cases defined as obese ( 21 . 2–77 . 8% , plus one GWA study , DGDG , that only selected non-obese cases ) . Some studies were specifically designed as case control studies and some as case-cohort studies , and we note that the extent of phenotyping performed to exclude autoimmune processes was different across studies , ranging from not requiring insulin treatment in the first year of diagnosis and GAD autoantibody negative , to general practitioner diagnosis of type 2 diabetes . Descriptions of the participating studies are available in the most recent DIAGRAM manuscript [1] , with summary statistics also presented in Table 1 and in Tables S1 and S2 . The two discovery GWA study meta-analyses comprised 2112 lean type 2 diabetes cases or 4123 obese type 2 diabetes cases , compared against up to 54 , 412 controls . For a subset of SNPs available on the Metabochip ( a custom Illumina iSelect SNP array that included the SNPs identified by GWA studies for several diseases and traits including type 2 diabetes loci ) we included data from an additional 263 lean type 2 diabetes cases , 1735 obese type 2 diabetes cases , and 3691 controls from the GoDARTs study [19] . With the exception of the BMI-stratification of cases , the meta-analyses , individual study quality control , and analytical methods were the same as those recently reported [1] . A genomic control inflation factor was calculated for each study for each analysis , and their test statistics were adjusted accordingly . Inverse-variance fixed effect meta-analyses were performed on imputed SNP datasets , testing for an additive genetic effect . All single point effect estimates are given with their [95% confidence intervals ( CI ) ] . Only autosomal SNPs with imputation quality scores >0 . 5 and a minor allele frequency >1% were included from each study . A SNP was excluded from the meta-analysed dataset if it was present in less than half of the studies . Given the use of two strata , we used a p-value threshold of 2 . 5×10−8 as the criterion for genome-wide significance . An additional 4 studies , totalling 2881 lean cases , 8702 obese cases , and 18957 controls were available for de novo genotyping of SNPs ( Table S2 ) . For the DGDG replication , all polymorphisms were genotyped using the KASPar system ( KBiosciences ) . For Malmo CC , ADDITION-Ely , and Norfolk Diabetes Case Control Study ( NDCCS ) , Taqman assay genotyping was performed . For all four studies genotyping success rate was >95% , the genotyping error rate was 0% based on re-genotyping of 384 individuals , and all SNPs were in Hardy-Weinberg equilibrium ( P>0 . 05 ) . We re-performed the inverse-variance weighted meta-analysis for the replication SNPs using data from all the discovery and replication datasets . To test whether or not type 2 diabetes associations could be primarily driven by effects on BMI , we assessed the association of novel SNPs with BMI using data from the GIANT consortium consisting of 123 , 865 individuals . There are two possible reasons why a variant may be associated with type 2 diabetes in a stratified sample compared to using all data . First , the variant may have a genuinely larger effect in that stratum compared to the overall sample . Second , chance will influence which SNPs are most strongly associated in different subsets of data . To distinguish between these two possibilities we performed a case only analysis in which we tested whether variants associated with lean or obese type 2 diabetes were also associated with BMI within type 2 diabetes cases . We analysed BMI as a quantitative trait in cases from the GWA studies and meta-analysed the summary statistics . If a variant is genuinely associated with type 2 diabetes with stronger effects in the lean stratum , for example , we would expect the risk allele to be associated with lower BMI within cases . This phenomenon was previously reported for the variant in TCF7L2 [14] . SNP association statistics on glyacemic traits in healthy individuals were provided by the Meta-Analyses of Glucose and Insulin-related traits Consortium ( MAGIC ) . Phenotypes available were fasting insulin ( N = 38 , 238 , fasting glucose ( N = 46 , 186 ) , beta-cell function ( HOMA-B , N = 36 , 466 ) , insulin resistance ( HOMA-IR , N = 37 , 037 ) , HbA1C ( N = 46 , 368 ) and 2 hour glucose ( N = 15 , 234 ) after an oral glucose challenge . All traits are naturally log transformed , besides fasting glucose , 2 hour glucose and HbA1c . The studies and methodology for these GWA study data are described in their recent publications [2] , [20] , [21] and available online at www . magicinvestigators . org . We also had access to data from joint meta-analyses of SNP and SNPxBMI interaction on fasting glucose ( N = 58 , 074 ) , insulin ( N = 51 , 570 ) , and 2-hr glucose ( N = 15 , 141 ) , also provided by MAGIC ( Manning et al , in press ) . Identified SNPs were searched against a collected database of expression SNP ( eQTL ) results including a range of tissues [22] , [23] , [24] , [25] , [26] , [27] , [28] , [29] , [30] , [31] , [32] , [33] , [34] , [35] , [36] , [37] , [38] . In addition to identifying new loci , we tested the impact of BMI stratification on SNPs previously identified as associated with type 2 diabetes . We calculated the individual SNP association statistics using the lean and obese meta-analyses described above . To assess the effects of combining information from all known type 2 diabetes SNPs , we next used a single study , the GoDARTs [19] study , independent from the discovery GWA studies . In GoDARTs there were a total of 263 lean type 2 diabetes cases , 1735 obese type 2 diabetes cases , and 3691 controls . Known SNPs ( N = 36 on the metabochip ) were defined as those reaching genome-wide significance in studies using samples of European descent ( excluding FTO due to primary effect on BMI , and DUSP9 not present on the chip ) [1] , [2] , [3] . We also combined the 36 SNPs into a single allele count model . This analysis consisted of a logistic regression model comparing the count of an individual's type 2 diabetes risk alleles , against case-control status . Each risk allele count was weighted by the point estimate effect size of that SNP from the DIAGRAM meta-analysis [1] . We repeated this analysis using stratified controls ( BMI<25 kg/m2 versus lean cases and BMI≥30 kg/m2 versus obese cases ) instead of all controls . Finally , individuals were binned into quintiles based on their weighted allele score and per-quintile odds ratios calculated .
Three independent association signals reached P<2 . 5×10−8 in the lean case genome wide meta-analysis ( Table 2 ) . Two represented previously reported loci - TCF7L2 ( OR = 1 . 58 [1 . 47–1 . 68] , P = 2×10−40 ) and CDKAL1 ( OR = 1 . 26 [1 . 17–1 . 35] , P = 7×10−10 ) . One novel locus reached genome-wide significance , lead SNP positioned ∼25 kb from the HLA-DQA2 gene ( OR = 1 . 3 [1 . 19–1 . 42] , P = 1×10−8 ) . Three further independent signals reached P<5×10−7 , two of which were previously identified ( SNPs in or near ADCY5 , OR = 1 . 25 [1 . 15–1 . 35] P = 6×10−8 , and SLC30A8 , OR = 1 . 23 [1 . 15–1 . 33] P = 4×10−8 ) and one of which was novel ( SNPs in LAMA1 , OR = 1 . 22 [1 . 12–1 . 30] P = 1×10−7 ) . Rs numbers are given in Table 2 . In the obese case genome wide meta-analysis , five signals reached genome-wide significance ( Table 2 ) , all in or near known loci TCF7L2 , FTO , CDKAL1 , HHEX , and IGF2BP2 . A further three signals reached P<5×10−7; SNPs in or near the MC4R gene ( previously associated with BMI ) , and two other signals; in HMG20A ( previously reported in South Asians -OR = 1 . 14 [1 . 09–1 . 19] P = 2×10−7 ) and in ANKS1A ( OR = 1 . 3 [1 . 18–1 . 43] P = 5×10−7 ) . We sought to replicate the signals reaching P<5×10−7 not previously reported in Europeans . SNPs representing the LAMA1 ( rs8090011 ) , HLA-DQA2 ( rs3916765 ) , HMG20A ( rs7178572 ) , and ANKS1A ( rs16896390 ) signals were genotyped in up to 2 , 881 lean cases , 8 , 702 obese cases and 18 , 957 control individuals . Combined discovery and follow-up association statistics for these SNPs are shown in Table 2 . In the lean case analysis , the LAMA1 variant was associated with type 2 diabetes ( combined P = 8 . 4×10−9 , OR = 1 . 13 [1 . 09–1 . 18] , total lean cases N = 4 , 993 , controls = 70 , 515 ) compared to an OR = 1 . 03 [1 . 00–1 . 06] in the obese case analysis ( Figure 1 and Figure 2 ) . In the obese case analysis , the HMG20A signal was associated with type 2 diabetes ( combined P = 1 . 3×10−8 , OR = 1 . 11 [1 . 07–1 . 15] , total obese cases N = 8 , 583 , controls = 62 , 063 ) compared to an OR = 1 . 09 [1 . 02–1 . 17] , P = 0 . 015 , in the lean analysis ( Figure 3 and Figure 4 ) . In previously published studies including 8 , 130 cases not stratified by BMI [1] , the LAMA1 and HMG20A variants reached only nominal levels of significance of P = 0 . 002 ( OR = 1 . 07 [1 . 03–1 . 12] ) and P = 0 . 003 , OR = 1 . 07 [1 . 02–1 . 12] respectively ( both in the same directions as reported here ) . Considering a random-effects model [39] for both LAMA1 and HMG20A signals gave similar evidence for association ( LAMA1 lean analysis: P = 5×10−10 , obese analysis: P = 0 . 02; HMG20A lean analysis: P = 0 . 04 , obese analysis: P = 2 . 7×10−8 ) . Evidence for association at the HLA-DQA2 and ANKS1A signals was reduced when follow-up data were included . We next attempted to understand further the associations between SNPs in the LAMA1 and HMG20A loci and lean and obese type 2 diabetes cases respectively . Our study design , together with the associations between the FTO and MC4R variants in the obese strata , suggested that variants that primarily operate through BMI could drive our newly identified associations . We therefore assessed the two signals in the existing GWA studies of BMI performed by the GIANT study and consisting of 123 , 865 individuals [18] . The LAMA1 SNP was not associated with BMI ( P = 0 . 19 ) whilst the type 2 diabetes risk allele at the HMG20A SNP was nominally associated with increased BMI ( P = 0 . 02 ) . If the associations at the LAMA1 and HMG20A loci are genuinely stronger in one strata of diabetic cases compared to the other , we should observe an association of those variants with BMI within cases only . This phenomenon has previously been reported for the variants in TCF7L2[14] . The LAMA1 type 2 diabetes risk allele was associated with lower BMI within cases alone ( P = 2×10−6 when analysing BMI as a quantitative trait in 26 , 366 cases ) , a result consistent with its association being stronger in the lean case analysis . The HMG20A risk allele showed no evidence of association ( P>0 . 05 ) . Next we used data from MAGIC to assess potential roles of variants in normal glycaemia . The SNP representing the novel LAMA1 association showed no association with fasting glucose ( P = 0 . 48 , beta ( se ) = 0 . 0027 ( 0 . 004 ) N = 46 , 186 ) , fasting insulin ( P = 0 . 87 , beta ( se ) = 0 . 0006 ( 0 . 004 ) N = 38 , 238 ) , HbA1C ( P = 0 . 19 , beta ( se ) = 0 . 005 ( 0 . 004 ) N = 46 , 368 ) , 2-hour glucose response ( P = 0 . 43 , beta ( se ) = −0 . 016 ( 0 . 02 ) , N = 15 , 234 ) , or any of the SNP×BMI-interaction models . However , LAMA1 isn't unique amongst type 2 diabetes loci in showing no effect on glycemic traits in the MAGIC study . The HMG20A diabetes risk allele was associated with higher fasting glucose ( P = 0 . 04 , beta ( se ) = 0 . 008 ( 0 . 004 ) , N = 46 , 186 ) , higher HbA1C ( P = 0 . 002 , beta ( se ) = 0 . 01 ( 0 . 004 ) , N = 46 , 368 ) and higher fasting glucose after accounting for BMI and SNPxBMI interaction ( P = 0 . 008 , N = 58 , 074 ) . In an attempt to gain further insight into likely functional genes in the LAMA1 and HMG20A loci , we tested the lead SNPs at for association in a number of eQTL datasets . Tissues tested included various blood , brain , liver and fat samples ( see Methods ) . Only ‘cis’ associations were considered ( eQTL effects on a transcript within 1 Mb of the signal SNPs ) . The rs7178572 SNP in the HMG20A region was significantly associated with mRNA expression levels of HMG20A in the liver ( P = 4×10−5 ) , supported by two separate expression probes , and was the strongest known regional SNP for both the liver eQTL and type 2 diabetes . No other study-wide significant results were observed ( N = 14 tissues , 24 datasets/analyses ) . For each of 36 published type 2 diabetes loci ( identified in European studies and available on the metabochip ) we compared the effect sizes between the lean and obese GWA study meta-analyses ( Table 3 ) . Among the 36 independent variants , 29 had a larger point estimate odds ratio in the lean analysis compared to the obese analysis ( binomial test of 29/36 versus 50% under the null hypothesis of no difference , P = 0 . 0002 ) . We next assessed the combined effect of these SNPs in a case control study independent of the GWA studies - GoDARTs ( Figure 5 ) . In the lean stratum , we observed a weighted per-risk allele OR = 1 . 13 [1 . 10–1 . 17] , P = 3 . 2×10−14 . This was larger than the same model fitted in the obese strata where the OR = 1 . 06 [1 . 05–1 . 08] , P = 2 . 2×10−16 . Results were very similar when stratifying the controls as well as the cases by BMI: lean weighted per risk-allele OR = 1 . 13 [1 . 09–1 . 17]; obese weighted per risk-allele OR = 1 . 08 [1 . 05–1 . 10] ( heterogeneity of odds ratios P = 0 . 036 ) . We also observed a difference between lean and obese cases when removing controls and fitting a regression model of lean cases vs obese cases ( P = 0 . 0001 ) . None of these 36 variants were associated with BMI in 28 , 000–32 , 000 individuals from GIANT [1] , [2] . We next divided the case/control samples into risk quintiles , based on the number of risk alleles they carry , weighted by the relative effect sizes of those alleles from the larger DIAGRAM meta-analysis . The risk of being in each quintile relative to the median quintile is shown in Figure 6 . For the lean group , we observed an OR = 2 . 1 [1 . 47–3 . 01] for the quintile of individuals carrying the most risk alleles compared to the middle quintile . This effect was larger than that in the obese group where the equivalent OR = 1 . 37 [1 . 15–1 . 64] .
We have confirmed our hypothesis that it is possible to identify genetic associations in previously tested samples by constraining the phenotypic heterogeneity of disease cases . By stratifying type 2 diabetes into two well accepted definitions of lean and obese cases , we identified and replicated one locus in each BMI stratum , each previously unreported in European studies: a signal in the LAMA1 gene in the lean stratum and a signal in the HMG20A gene in the obese stratum . Lack of evidence for association with BMI for these two signals in 123 , 000 individuals [18] argues that these associations are not driven by a primary association with BMI . There are two reasons why previously undetected genetic associations may be observed in stratified data . First chance , in this context “sampling error” , may occur – new signals may reach statistical thresholds in subsets of data due to a combination of real association and chance . Second , the signal may represent genuine heterogeneity . The enrichment of the LAMA1 signal in lean type 2 diabetes cases compared to obese cases is likely to be a real effect but the enrichment of the HMG20A signal in obese cases is more likely to be due to chance . Whilst we observed some regression to the mean ( or “winner's curse” ) for the LAMA1 signal , the effects remained different in lean compared to obese cases in the replication samples alone ( Figure 1 ) . In addition , the LAMA1 type 2 diabetes risk allele was associated with lower BMI within cases alone ( P = 2×10−6 when testing BMI as a quantitative trait in cases ) – a similar result was previously reported for the TCF7L2 risk allele [16] . In contrast there is no evidence that the HMG20A signal is stronger in obese replication strata compared to lean replication strata ( Figure 3 ) and there was no association with increased BMI within cases alone ( P>0 . 05 when testing BMI as a quantitative trait in cases ) . The LAMA1 signal falls in a recombination block within the LAMA1 gene ( Figure 2 ) , with the lead SNP positioned within intron 61 . Searching for correlated SNPs ( r2>0 . 5 ) using 1000 Genomes Project data identified only additional intronic SNPs . Previous cell biology studies support a role for LAMA1 , encoding laminin-1 , in diabetes etiology - inhibition of LAMA1 expression reduced glucose-stimulated secretion in INS1E cells [40] . Several studies observed the beneficial effects of laminin-1 , and extracellular matrix ( highly enriched with laminin-1 ) preparations on pancreatic islet development and function [41] , [42] , [43] , [44] , [45] , [46] . Laminin-1 is expressed in intra-islet capillaries [47] and a role for laminin receptor 1 was proposed in angiogenesis [48] . The confidence in the HMG20A association is enhanced by several lines of evidence from other studies . The HMG20A signal was previously identified in a GWA study of South Asian individuals [7] and was nominally associated with fasting glucose ( P = 0 . 04 , N = 46 , 186 ) and HbA1C ( P = 0 . 002 , N = 46 , 368 ) in non-diabetic individuals analysed by the MAGIC consortium . The association with fasting glucose became stronger when adjusting for BMI in an interaction model ( P = 0 . 008 ) . We initially discovered a genome-wide significant signal near the HLA-DQA2 locus , which subsequently failed to replicate ( rs3916765 , P = 1×10−6 ) . This variant is not in the same gene or in linkage disequilibrium with previously reported associations between HLA loci and type 2 diabetes [1] , [49] . Concerned with the prospect of this association being due to auto-immune diabetes case admixture , we assessed the association of the strongest known type 1 diabetes signals in our lean meta-analysis . None of these showed any significant evidence of association – including the lead signals from the WTCCC type 1 diabetes study in the HLA region ( rs3129941 , P = 0 . 08 ) , or near the INS ( rs3842748 , P = 0 . 64 ) or PTPN22 ( rs2476601 , P = 0 . 38 ) genes . This study has provided the most robust evidence to date that lean type 2 diabetic cases are likely to carry a disproportionately high load of known type 2 diabetes risk alleles . More than 80% ( 29/36 ) of type 2 diabetes variants established in Europeans had stronger effects in lean compared to obese cases and the odds ratio for the 20% of lean cases carrying the most risk alleles was more than twice that of the 20% of obese cases carrying the most risk alleles . The corollary of these findings is that obese cases on average carry a disproportionately low load of confirmed type 2 diabetes risk variants , but their diabetes risk will likely be more heavily influenced by their genetic and environmental predisposition to gaining weight in adulthood . Despite this enrichment of stronger effects in lean versus obese cases , analyses focused only on lean cases is not a more powerful study design compared to using all cases . For each of the known loci tested , the power gained by increased effect sizes is easily offset by the reduced power of having a case sample size of ∼25% . Nevertheless our data indicate that , given limited resources , recruitment strategies that target leaner type 2 diabetes cases will have more power than those that target a similar number of cases but without enrichment for lower BMI . There are several limitations to our study . First , the use of an unstratified control group made testing the significance of differences between lean and obese cases difficult in the context of a genome wide meta-analysis . However , several lines of evidence support our conclusions that lean individuals are enriched for known type 2 diabetes genetic effects . This evidence includes: the very large differences between the upper and lower 95% confidence intervals of the weighted per allele effects in lean and obese , the consistency of the weighted per allele results when stratifying controls as well as cases , and the 80/20 proportion of SNPs showing stronger effects in lean compared to obese individuals respectively . Second , after stratifying by BMI , we did not use other criteria to reduce the clinical heterogeneity of type 2 diabetes . Of note , cases within the BMI strata differed appreciably in their age at diagnosis and the degree to which autoimmune or monogenic diabetes had been excluded . Instead , having stratified by BMI , we opted to use the largest available sample sizes . It is possible that a small number of monogenic or autoimmune forms of diabetes amongst our cases could have reduced our power to detect novel variants . Further studies may help refine how known and novel diabetes signals operate in more clinically homogenous settings . Finally , known type 2 diabetes signals are likely to account for only a small fraction of all risk variants that exist in the genome and any inferences we make are limited to the known signals . In conclusion , we report associations with the LAMA1 and HMG20A ( not previously associated at genome-wide significance in Europeans ) gene regions with type 2 diabetes risk . We have demonstrated that lean diabetic cases are enriched for known type 2 diabetes risk alleles compared to obese cases . This enrichment is consistent with the observation that many of the variants with the strongest effects on diabetes are associated with reduced beta cell function [1] . At the opposite end of the spectrum , obese cases presumably need fewer diabetes risk variants to push them towards diabetes , as they are already under strain from the physiological impact of obesity and insulin resistance . These data suggest a disease model where type 2 diabetes cases lie across a continuous distribution with regards to genetic/environmental risk , and beta-cell dysfunction versus insulin resistance aetiologies .
|
Individuals with Type 2 diabetes ( T2D ) can present with variable clinical characteristics . It is well known that obesity is a major risk factor for type 2 diabetes , yet patients can vary considerably—there are many lean diabetes patients and many overweight people without diabetes . We hypothesized that the genetic predisposition to the disease may be different in lean ( BMI<25 Kg/m2 ) compared to obese cases ( BMI≥30 Kg/m2 ) . Specifically , as lean T2D patients had lower risk than obese patients , they must have been more genetically susceptible . Using genetic data from multiple genome-wide association studies , we tested genetic markers across the genome in 2 , 112 lean type 2 diabetes cases ( BMI<25 kg/m2 ) , 4 , 123 obese cases ( BMI≥30 kg/m2 ) , and 54 , 412 healthy controls . We confirmed our results in an additional 2 , 881 lean cases , 8 , 702 obese cases , and 18 , 957 healthy controls . Using these data we found differences in genetic enrichment between lean and obese cases , supporting our original hypothesis . We also searched for genetic variants that may be risk factors only in lean or obese patients and found two novel gene regions not previously reported in European individuals . These findings may influence future study design for type 2 diabetes and provide further insight into the biology of the disease .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"genome-wide",
"association",
"studies",
"medicine",
"public",
"health",
"and",
"epidemiology",
"diabetes",
"mellitus",
"type",
"2",
"diabetic",
"endocrinology",
"genetic",
"epidemiology",
"molecular",
"epidemiology",
"epidemiology",
"endocrinology",
"diabetes",
"and",
"endocrinology",
"biology",
"phenotypes",
"heredity",
"genetics",
"human",
"genetics",
"genetics",
"and",
"genomics",
"complex",
"traits"
] |
2012
|
Stratifying Type 2 Diabetes Cases by BMI Identifies Genetic Risk Variants in LAMA1 and Enrichment for Risk Variants in Lean Compared to Obese Cases
|
A powerful way to separate signal from noise in biology is to convert the molecular data from individual genes or proteins into an analysis of comparative biological network behaviors . One of the limitations of previous network analyses is that they do not take into account the combinatorial nature of gene interactions within the network . We report here a new technique , Differential Rank Conservation ( DIRAC ) , which permits one to assess these combinatorial interactions to quantify various biological pathways or networks in a comparative sense , and to determine how they change in different individuals experiencing the same disease process . This approach is based on the relative expression values of participating genes—i . e . , the ordering of expression within network profiles . DIRAC provides quantitative measures of how network rankings differ either among networks for a selected phenotype or among phenotypes for a selected network . We examined disease phenotypes including cancer subtypes and neurological disorders and identified networks that are tightly regulated , as defined by high conservation of transcript ordering . Interestingly , we observed a strong trend to looser network regulation in more malignant phenotypes and later stages of disease . At a sample level , DIRAC can detect a change in ranking between phenotypes for any selected network . Variably expressed networks represent statistically robust differences between disease states and serve as signatures for accurate molecular classification , validating the information about expression patterns captured by DIRAC . Importantly , DIRAC can be applied not only to transcriptomic data , but to any ordinal data type .
Molecular signatures based on the measured abundance of biomolecules ( e . g . , mRNA , proteins , metabolites ) have the potential to discriminate among disease subtypes , to predict clinical outcomes , or to provide insights into the mechanistic underpinnings of disease progression . Moreover , with sufficient data , these signatures begin to enable the identification of perturbed networks that reflect core aspects of the disease process—and thus could provide insights into functionally relevant drug targets as well as new approaches to diagnostics [1] , [2] . However , distinguishing signal from noise in high-throughput data such as mRNA microarray experiments presents a significant challenge . This noise commonly results from technical issues in data production and the integration of datasets from different platforms , laboratories , or even experiments within a lab . Noise in high-throughput data also stems from biological variability in the sources , such as genetic polymorphisms , different stages of the biological process , disease stratification , and stages of disease progression . In the study of human disease processes , this variability poses a unique hurdle as there are often only data for a single point in time; when comparing data between individuals who appear to have the same disease , one does not know whether the observed differences reflect disease subtypes or different stages of a single disease type . A fundamental tenant of systems approaches to biology and medicine is that dynamically changing biological networks mediate physiological , developmental , and disease processes , and that the key to understanding these processes is translating network dynamics into phenotypes . As such , a powerful method to mitigate some forms of biological noise ( hence increasing the utility of high-throughput data as a diagnostic and scientific tool ) is to convert the molecular data from individual genes or proteins into an analysis of comparative biological network behaviors . Typically , studies search for a small number of individual genes whose differential expression is highly correlated with phenotypic changes . However , malignant phenotypes in many diseases arise from the net effect of interactions among multiple genes and other molecular agents within biological networks . For example , cooperating oncogenes interact synergistically to evade tumor suppression mechanisms such as cell-cycle arrest and apoptosis [3] , [4] . The combinatorial nature of such disease-induced perturbations leads to a highly complex picture of the underlying biological processes . As such , the biological insight gleaned from the expression patterns of individual genes is often limited . Other pitfalls associated with individual gene expression analysis have been proposed and discussed elsewhere [2] , [5] , [6] . The importance of studying network behavior—evident in most phenotypes , disease or otherwise—is particularly well-documented for cancer . Research has linked modulated function on the level of either metabolic networks [7]–[9] and/or signaling networks [10]–[12] to cancer hallmarks including angiogenesis , increased growth , metastasis , and evasion of immune detection . Similarly , recent global genomic analyses in glioblastoma multiforme [13] , [14] and pancreatic cancers [15] have revealed both varying numbers and frequencies of genetic alterations within distinct core networks of each disease . In light of these findings , microarray data analysis methods have begun to shift towards identifying biologically meaningful pathways or networks . We consider all pathways to in fact be part of interconnected biological networks , and henceforth use the term network rather than pathway . In general , network regulation controls the expression levels of related genes responding to specific conditions . Existing tools for network-based expression analysis commonly investigate informative patterns of up-regulation or down-regulation ( i . e . , increases or decreases in expression ) of genes in different disease states . For example , the widely-used gene set enrichment analysis ( GSEA ) platform identifies networks that are significantly enriched for individual genes that are highly correlated with a phenotype [5] , [16] . Other methods employ a single statistic to represent the collective activity of a network ( e . g . , mean or median gene expression ) [2] , [17]; perturbed levels of network activity ( i . e . , collective up- or down-regulation ) are then examined to identify those networks most differentially expressed between phenotypes . These frameworks have been applied to diverse cancer systems and serve as a robust source of biological discovery [2] , [18] . Studying cellular regulation of networks in terms of “unidirectional” changes may , however , overlook subtle , yet influential , changes in the relationships among the genes within a network . This drawback directly reflects the combinatorial operation of genes in networks , in which the actions of one gene greatly influences the actions of other genes . By accounting for these combinatorial interactions we can begin to alleviate the signal-to-noise issues in disease-perturbed networks ( as well as dynamically changing networks mediating physiology or development ) . In particular , even the elementary interactions captured by the relative orderings among two or three genes have been shown to provide powerful biomarkers for separating phenotypes [19]–[21] . With methods that aim to identify statistically significant up- or down-regulation of genes or networks , results will also depend largely on the context of the microarray experiment . Cellular regulation in a case with a number of up- or down-regulated genes in one phenotype versus another manifests as an increase in absolute expression levels above some threshold , relative to all other genes on the microarray . Even when thresholds are tuned to produce statistically significant results , the findings are still based on indirect measurements , ( i . e . , fluorescence ) and therefore may depend heavily on the experimental set up , type of data normalization , and other factors . In addition to the technical limitations of microarray experiments , biological context can greatly impact results . For instance , if nearly all genes are differentially expressed between two phenotypes , then no single network will be statistically “enriched” for change . It is also possible that neither individual network genes nor any network as a whole will display notable over- or under-expression in response to environmental or disease-related stimuli . The importance of accounting for combinatorial gene interactions—and to do so without need to reference all of the genes on the microarray—again becomes clear . We have developed a new method called Differential Rank Conservation ( DIRAC ) which considers combinatorial behavior , and provides quantitative measures of how network expression differs within and between phenotypes . The DIRAC approach assesses cellular regulation of a network in the context of the relative levels of expression for participating genes . For each microarray , the expression values of the network genes are ordered from highest expression ( ranked first ) to lowest expression ( ranked last ) ; regulation is then quantified entirely by the rankings of genes within a selected network . Consequently , DIRAC identifies and measures network-level perturbations from a completely novel perspective , namely the “combinatorial comparisons” of network genes as opposed to increases or decreases alone , allowing one to study how this ordering changes in different conditions—and thus begin to infer the consequences of combinatorial gene interactions . As a result , this approach has two key advantages over tools that measure absolute changes in expression levels . First , it accounts for gene-gene interactions; second , the results do not depend on the other genes on the microarray or on the method of normalization used . These are both critical points in dealing with signal-to-noise issues . Notably , as DIRAC treats each network independently , it can still identify perturbed networks even when every gene on the microarray is differentially expressed ( in contrast to enrichment measures ) . Our strategy for representing network rankings uses pairwise comparisons of gene expression levels . Such pairwise comparisons can yield two-gene predictors with simple decision rules for classification of expression profiles [22] , [23] . These decision rules have in turn resulted in highly accurate two-gene diagnostic classifiers based on relative expression reversals that have proven effective for molecular identification of cancer [19]–[23] . We extend the relative expression reversal concept to networks . However , analyzing sample-to-sample changes for every possible distinct ordering of gene expression values within a network is not computationally feasible; there are simply too many possible orderings , i . e . , permutations . Knowing the states of all pairwise orderings is equivalent to knowing the full ranking , which motivates our representation . For each distinct pair of genes within a network , we consider a binary variable indicating whether or not the mRNA abundance of the first gene is less than that of the second gene; in fact , we restrict attention to the probability of this event within a phenotype for each pair of genes . In this way , we avoid the combinatorial complexity of permutations and represent the “expected” ordering of network genes for a given phenotype as a binary template . Unlike the probabilities of full orderings , pairwise frequencies are reliably estimated with typical sample sizes , while still capturing a great deal of information about network regulation . We subsequently compute a matching score to signify how closely each sample's network ordering matches a phenotype-specific template . We can use DIRAC at the population level to quantify conservation differences between networks for a given phenotype . Specifically , DIRAC allows us to use rankings to identify and contrast tightly and loosely regulated network types of a single phenotype: Tightness of regulation for a selected network is best understood as the allowed variation in gene expression levels observed across the population . This offers an advantage over studying up- or down-regulation only because it indicates the level of control across samples in a population . In this work we use the DIRAC approach to identify networks that are tightly regulated in a number of human cancers and neurological disorders . Since networks under tight control in a particular phenotype may be necessary to maintain a specific cellular function , tightly regulated networks that change across phenotypes may provide insight into processes such as disease progression . Additionally , DIRAC can be applied at the sample level to identify conservation differences between phenotypes for a specified network . At this level the DIRAC method can identify variably expressed networks that reveal statistically robust differences between disease states , leading to highly accurate classification of expression profiles from various diseases . When used to separate expression profiles , the DIRAC method is noteworthy because it ( i ) is independent of microarray data normalization; ( ii ) results in a simple yet efficient classifier for phenotype distinction; and ( iii ) appears to be comparable in accuracy to state-of-the-art classification methods . Learning the regulation of gene rankings within different states allows us to discover molecular signatures composed of related genes that distinguish phenotypes , identify networks most involved in disease transitions , and assist identification of potential therapeutic targets . Importantly , while we focus on gene expression in the present study , the method can be generalized to any ordinal dataset , and thus can be applied to such biological data types as proteomics , gene copy number , chromosomal position , and so forth .
The DIRAC approach was used to evaluate regulation of gene ordering within networks in different diseases . For each microarray sample in each phenotype studied , we characterized the ordering of network genes ( i . e . , network ranking ) in terms of comparisons between the expression values of pairs of genes . Based on the comparison statistics , we defined a rank template for each network and phenotype representing the expected ( i . e . , most common ) pairwise ordering of gene expression for that network in that phenotype . We employed a simple measure—a rank matching score ( R ) —to determine how well the network ranking in each individual sample ( i . e . , expression profile ) matched the ordering defined in the rank template . Averaging R over all samples within a phenotype yields a network-specific rank conservation index ( μR ) which represents how well , on average , all samples in the same phenotype match the corresponding rank template . Alternatively , comparing two rank matching scores for the same sample leads to a highly-discriminating rank difference score ( Δ ) that allows one to determine the most variably expressed networks between two phenotypes . The calculation of these quantities is illustrated in Figure 1 . Several prototypical scenarios arise from these measures . In one scenario ( Figure 2 , left ) , conservation indices are used to measure the consistency with which network rankings are maintained in a population , and are used to identify tightly regulated networks in each phenotype . One situation , where all samples have similar network rankings , yields a large rank conservation index and indicates the network is tightly regulated . A second situation , where the ordering of network genes is highly varied , yields a small rank conservation index and indicates the network is loosely regulated . In a second prototypical scenario , the DIRAC method detects changes in ranking ( i . e . , shuffling of gene expression values ) between phenotypes for a selected network ( Figure 2 , right ) . The top networks selected by DIRAC based on the difference score can be used to classify gene expression profiles by phenotype . We first applied DIRAC to investigate network rankings using gene expression profiles obtained from patients with different stages of prostate disease . The gene expression data , originally reported by Yu et al . [24] and publically available in the NCBI Gene Expression Omnibus ( GDS2545 ) , contains 108 human prostate samples: 18 samples of normal prostate tissue ( NP ) from organ donors , 65 primary prostate tumor ( PT ) samples , and 25 metastatic prostate tumor ( MT ) samples . The findings for normal prostate and prostate cancer samples presented below represent the main features of the DIRAC method , and can be similarly obtained for any disease expression data . In addition to the more detailed prostate cancer analysis , we examined a number of other disease phenotypes including cancer subtypes and neurological disorders , and identified both tightly regulated and variably expressed networks in each . For each dataset , we grouped expression levels of genes into 248 human signaling networks , defined according to the BioCarta gene sets collection in the Molecular Signatures Database ( MSigDB ) [5] . In order to ensure that the networks examined were as complete as possible , we used gene synonym information from NCBI to replace unmatched names in each dataset with those belonging to networks in the BioCarta collection . This step led to an average increase of 5% in the fraction of network genes ( 1296 total across 248 networks ) for which a corresponding expression value was found ( Table S1 ) . The population-level analysis is centered on the rank conservation index ( μR ) -defined for each network and each phenotype . This index represents the degree of conservation in the rankings of the expression levels of the network genes , averaged over samples of the phenotype . In order to identify variably expressed networks between two selected phenotypes , we designed a rank difference score ( Δ ) , calculated for each sample based on rank matching scores . For a particular network , this measure indicates the similarity between the ordering of network genes in a sample to the template of one class versus the template of the other . The difference score ranges from -1 to 1 , with positive values suggesting the first phenotype , and negative values suggesting the second , culminating in simple rules for classifying an expression profile . Our purpose in introducing the rank difference score was two-fold: ( i ) to identify variably expressed networks between two selected phenotypes; and ( ii ) to validate the DIRAC approach to network identification , and the emphasis on combinatorial interactions , by demonstrating the discriminative power of the networks identified . Systems medicine approaches assume that disease arises from disease-perturbed biological networks in the relevant organ or organs . These disease-perturbed networks alter the envelopes of information that they express—and these changes encode the pathophysiology of the disease . Moreover , the altered patterns of information can elucidate new strategies for diagnosis or therapy . Future drugs will likely be designed to re-engineer disease-perturbed networks to behave in a more normal fashion , or at least to abrogate their most deleterious consequences . This will require a new drug target identification approach , and re-engineering disease-perturbed networks appropriately will almost always require multiple drugs . Likewise , the perturbed nodal points in disease-perturbed networks can be expressed as proteins in the blood—where the disease-altered levels of expression may reflect the disease process . These disease-altered blood proteins will create unique blood fingerprints specific for each disease process , and thus provide powerful diagnostics . These advances rely upon the proper identification of disease-perturbed networks . To date , most of the evaluation of networks has employed lists of transcripts that are perturbed from the levels of their counterparts in normal organs . This listing , as with genome-wide association ( GWAS ) studies , misses the key fact that disease-perturbed networks must be assessed in the context of the combinatorial interactions of their nodal components . Our method is the first approach that begins to account for the combinatorial behavior of interacting genes , mRNAs and/or proteins . Using DIRAC-based calculations allows us to begin to assess the key disease-perturbed networks that may aid in the approach to diagnosis and therapy . We also stress that these methods will almost certainly prove powerful in the stratification of disease types . The example of gastrointestinal stromal tumors ( GIST ) and leiomyosarcomas ( LMS ) , histologically indistinguishable , but clearly classifiable by a primitive version of DIRAC , is striking . We believe this will be a powerful approach in , for example , distinguishing various types of neurodegenerative diseases , as well as the stratification of complex diseases such as Alzheimer's . Notably exciting , some of the key transcripts used in this classification process actually encoded proteins secreted into the blood . Findings of this nature could lead to the use of altered blood levels of proteins for diagnosis without the need to sample disease tissues . Emerging technologies will make these measurements possible at the single cell level , exposing other exciting possibilities for diagnosis using the strategies outline above . We predict the application of DIRAC as a powerful clinical tool in the advancing proactive , rather than reactive , new medicine—the so-called P4 medicine ( predictive , personalized , preventive and participatory ) —where blood and single-cell diagnostics will be the foundation of the P4-medicine revolution . In this study we demonstrate a novel method to identify highly discriminative biological networks based on differing patterns of gene expression ranking within networks . These results provide a coarse , but meaningful , glimpse into patterns of network regulation for different phenotypes based on combinatorial relationships between the involved genes . For example , when comparing two disease states , it appears to be very common ( although not universal ) for network rankings to be more varied—or less tightly regulated—in the more pathological state . This increased disorder associated with malignancy might be expected , as mutations and other altered behavior of biomolecules lead to breakdown of typical functioning in biological networks; rank conservation index values calculated in DIRAC represent a quantitative means to study and further verify this notion . Importantly , this method not only identifies perturbed networks , but does so in such a way that it can classify samples . Thus , predictive accuracy becomes a strong measure for the validity of the perturbed network as a reproducible hallmark of the disease phenotype . Such high predictive accuracy in classification adds much stronger evidence that biologically meaningful network differences are found than only a low P-value or FDR , which simply measure how likely the result derives from chance . Measures of global regulation can also give useful information for designing research to identify expression-based classifiers of disease . For instance , it would be more fruitful to search for clear molecular signatures with tightly regulated phenotypes . In cases with mostly loosely regulated networks , the greater variation from sample-to-sample would pose a more difficult challenge for identifying reliable classifiers . Studying rank regulation of biologically relevant networks thus offers a promising tool for measuring network behavior within and across different populations . Looking forward , the results obtained through this approach should provide increased insight into phenotypic processes of importance in biology and medicine .
Given the list {g1 , … , gGm} of Gm genes within a network m on a microarray , we let X = ( X1 , … , XGm ) denote the corresponding expression profile , where Xi is the expression level of gene gi . Our data then consists of a Gm x N matrix; the nth column represents the expression profile xn of the nth sample , n = 1 , … , N . In addition , each sample is labeled by a phenotype Y∈{A , B , … , K} . The labeled training set is F = { ( x1 , y1 ) , … , ( xN , yN ) } . Expression profiles X and phenotype labels Y are regarded as random variables , and the elements of F represent independent and identically distributed samples from some underlying probability distribution of ( X , Y ) . Our analysis is based entirely on the ranks within each expression profile . With Gm genes , there are Gm ! possible orderings for the expression values . The networks we consider typically have tens or hundreds of genes; consequently , working directly with individual permutations is not feasible . For example , any estimated distribution over permutations using training data would be highly singular . Instead , we base the analysis entirely on pairwise comparisons . Knowing the ordering of the gene expressions within each network expression profile is equivalent to knowing all of the pairwise orderings , i . e . , whether Xi<Xj or Xi>Xj for each distinct pair of genes 1≤i , j≤Gm within the network m . Evidently , there are Gm ( Gm–1 ) /2 such pairs . For example , if there are Gm = 4 genes , then there are six distinct ordered pairs: { ( 1 , 2 ) , ( 1 , 3 ) , ( 1 , 4 ) , ( 2 , 3 ) , ( 2 , 4 ) , ( 3; 4 ) } . In order to define a template representing the expected ranking of network genes within a phenotype , we consider the probabilities Pr ( Xi<Xj |Y = k ) for each pair of genes gi<gj and for each phenotype k . We estimate these probabilities from the training set by computing the fraction of samples in each phenotype for which gene gi is expressed less than gene gj . The rank template for a fixed network m and phenotype k is the binary vector T ( m , k ) of length Gm ( Gm–1 ) /2 where the i , jth component is 1 if Pr ( Xi<Xj |Y = k ) >0 . 5 and 0 if Pr ( Xi<Xj |Y = k ) ≤0 . 5 . The calculation of a rank template is illustrated in Figure 1 . Given an expression profile xn for the network m , there is then a natural measure for how well the sample matches the template T ( m , k ) . The rank matching score of sample n is denoted by R ( m , k ) ( xn ) and is defined to be the fraction of the Gm ( Gm–1 ) /2 pairs for which the observed ordering within xn matches the template—the orderings expected for phenotype k . See Figure 1 for an illustration of a rank matching score . Averaging the rank matching score over all the samples in a phenotype k yields a rank conservation index denoted by μR ( m , k ) = E ( R ( m , k ) |Y = k ) . This index is estimated by averaging the scores R ( m , k ) ( x ) over all the samples ( x , y ) in the training set for which y = k . Whereas the rank matching score is a sample-based statistic , i . e . , it is defined for each expression profile , the rank conservation index is a population statistic . The rank conservation index can be seen as a measure of the stability in rankings among the network genes in the phenotype . Two extreme cases correspond to ( i ) pure random shuffling of the expression values in the phenotype from sample to sample , in which case μR ( m , k ) ≈0 . 5; and ( ii ) all samples displaying exactly the same ordering , in which case μR ( m , k ) ≈1 . In general , there are many gene pairs gi and gj which are expressed on different scales , and hence xi<xj across nearly all samples and phenotypes . As a result , one generally finds μR ( m , k ) ≫0 . 5 . This index is similar to entropy in the sense that values of μR ( m , k ) ≪1 indicate a highly disorganized state in which there is a great deal of variation among the rankings in phenotype k from sample to sample and values of μR ( m , k ) ≈1 indicate a highly ordered state in which samples have very similar , and hence predictable , orderings among the genes . Consider two phenotypes Y = A , B , and a fixed network m . If network m is tightly regulated in one phenotype , the samples from that phenotype , say Y = A , will have high R ( m , A ) values on average . But if μR ( m , k ) is large for both k = A and k = B , and if the two rank templates T ( m , A ) and T ( m , B ) are significantly different , then the samples from phenotype Y = A will generally have low values for the statistic R ( m , B ) as well as high values for the statistic R ( m , A ) , and vice-versa for the samples from phenotype Y = B . We capture this phenomenon , namely low variance of network ranking within a phenotype , but high variance between phenotypes , with a single statistic calculated for each sample: the difference Δ ( m ) ( xn ) = R ( m , A ) ( xn ) –R ( m , B ) ( xn ) . Clearly , –1≤Δ ( m ) ( xn ) ≤1 with positive ( respectively , negative ) values providing evidence that the phenotype of sample n is Y = A ( resp . , Y = B ) . As a result , the difference score provides a classifier for phenotype identification based on the degree of regulation of the genes in network m . A new sample n is predicted to belong to phenotype Y = A if Δ ( m ) ( xn ) >0 and to phenotype Y = B if Δ ( m ) ( xn ) ≤0 . The classification rate for network m is then: η ( m ) = Pr ( Δ ( m ) ( X ) >0|Y = A ) *Pr ( Y = A ) +Pr ( Δ ( m ) ( X ) ≤0|Y = B ) *Pr ( Y = B ) . The calculation of a rank difference score was shown in Figure 1 . For example , if Y = A denotes prostate cancer and Y = B denotes normal prostate , and if we assume that the two phenotypes are a priori equally likely , then η ( m ) is simply the average of sensitivity and specificity relative to identifying cancer . In order to determine the most variably expressed networks between two given phenotypes , we calculate rank templates for each phenotype , evaluate the differential score for each sample in the training set and choose the networks with the largest estimated classification rate . One previously reported method , k-TSP , classifies expression profiles based on k pairs of genes with the most significant expression reversals among all assayed genes [22] . The classifier based on the rank difference score is also based on k pairs of genes , with k equal to the distance between the two rank templates . To see this , notice that upon computing the difference Δ ( m ) ( xn ) for pathway m and phenotypes A and B , the gene pairs ( i , j ) for which T ( m , A ) ( i , j ) = T ( m , B ) ( i , j ) cancel out . The DIRAC-based classifier therefore reduces to voting among the gene pairs whose probabilities straddle 0 . 5—i . e . , satisfy Pr ( Xi<Xj |Y = A ) <0 . 5<Pr ( Xi<Xj |Y = B ) or vice versa . However , these k pairs of genes are those in the “top-scoring network” as determined by DIRAC rather than the most discriminating k pairs overall ( as would be identified by k-TSP ) . Procedures for estimating statistical significance are described below for metastatic prostate tumors ( MT ) and normal prostate ( NP ) . Identical procedures were used for all binary phenotype datasets studied . We used leave-one-out cross validation to estimate the ( generalization ) error rate of each classification method studied . Importantly , for each classification method tested , all processes were done using only the training samples without including any information from the test sample . Within each iteration of the cross validation loop , expression profiles in the original training data F = { ( x1 , y1 ) , … , ( xN , yN ) } are divided into two groups: a training set ( Ftrain ) and a test set ( Ftest ) . The classifier is trained on the N–1 samples of Ftrain and then used to predict the phenotype of the remaining “left out” sample in Ftest . The overall cross validation classification rate after N total train-test divisions and predictions is calculated as the average of sensitivity and specificity . Details for training and testing with each type of classifier are described below .
|
The systems approach to medicine derives from the idea that diseased cells arise from one or more perturbed biological networks due to the net effect of interactions among multiple molecular agents; by measuring differences in the abundance of biomolecules ( e . g . , mRNA , proteins , metabolites ) we can identify reporters of network states and uncover molecular signatures of disease . However , a major limitation of previously published network analyses is the focus on small numbers of individual , differentially-expressed genes , hence the failure to take into account combinatorial interactions . We report a new technique , Differential Rank Conservation , for identifying and measuring network-level perturbations . Our rank conservation index is based entirely on the relative levels of expression for participating genes and allows us to detect differences in network orderings between networks for a given phenotype and between phenotypes for a given network . In examining cancer subtypes and neurological disorders , we identified networks that are tightly and loosely regulated , as defined by the level of conservation of transcript ordering , and observed a strong trend to looser network regulation in more malignant phenotypes and later stages of disease . We also demonstrate that variably expressed networks represent robust differences between disease states .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"biology/gene",
"expression",
"genetics",
"and",
"genomics/gene",
"expression",
"molecular",
"biology/bioinformatics",
"computational",
"biology/signaling",
"networks",
"computational",
"biology",
"computational",
"biology/systems",
"biology"
] |
2010
|
Identifying Tightly Regulated and Variably Expressed Networks by Differential Rank Conservation (DIRAC)
|
Gerodermia osteodysplastica ( GO ) is characterized by skin laxity and early-onset osteoporosis . GORAB , the responsible disease gene , encodes a small Golgi protein of poorly characterized function . To circumvent neonatal lethality of the GorabNull full knockout , Gorab was conditionally inactivated in mesenchymal progenitor cells ( Prx1-cre ) , pre-osteoblasts ( Runx2-cre ) , and late osteoblasts/osteocytes ( Dmp1-cre ) , respectively . While in all three lines a reduction in trabecular bone density was evident , only GorabPrx1 and GorabRunx2 mutants showed dramatically thinned , porous cortical bone and spontaneous fractures . Collagen fibrils in the skin of GorabNull mutants and in bone of GorabPrx1 mutants were disorganized , which was also seen in a bone biopsy from a GO patient . Measurement of glycosaminoglycan contents revealed a reduction of dermatan sulfate levels in skin and cartilage from GorabNull mutants . In bone from GorabPrx1 mutants total glycosaminoglycan levels and the relative percentage of dermatan sulfate were both strongly diminished . Accordingly , the proteoglycans biglycan and decorin showed reduced glycanation . Also in cultured GORAB-deficient fibroblasts reduced decorin glycanation was evident . The Golgi compartment of these cells showed an accumulation of decorin , but reduced signals for dermatan sulfate . Moreover , we found elevated activation of TGF-β in GorabPrx1 bone tissue leading to enhanced downstream signalling , which was reproduced in GORAB-deficient fibroblasts . Our data suggest that the loss of Gorab primarily perturbs pre-osteoblasts . GO may be regarded as a congenital disorder of glycosylation affecting proteoglycan synthesis due to delayed transport and impaired posttranslational modification in the Golgi compartment .
Bone mass is highly heritable and largely determined by bone growth during childhood and adolescence leading to the so-called peak bone mass , and the rate of subsequent bone loss at older ages [1] . Gerodermia osteodysplastica ( GO; OMIM #231070 ) belongs to the group of autosomal recessive cutis laxa ( ARCL ) syndromes characterized by lax , wrinkled skin , a generalized connective tissue weakness , and a progeroid appearance [2–4] . GO features pronounced osteoporosis leading to pathological fractures already in childhood . GORAB , the gene product defective in GO , is a coiled-coil containing peripheral membrane protein that is recruited to the Golgi compartment via a specific , GTP-dependent interaction with the small GTPases ARF5 and RAB6 [5] . Due to this fact , GORAB has been suggested to belong to the group of golgins , small GTPase effector proteins involved in different steps of Golgi-related transport processes . Nevertheless , the physiological role or GORAB in development and homeostasis of the skeleton and of connective tissues is not well understood . The Golgi compartment is a central hub for protein trafficking and posttranslational modification within the secretory pathway , among which glycosylation processes are most prominent [6] . While the classical disorders of glycosylation ( CDGs ) affect N-glycosylation leading to a prototypical combination of neurological , hepatic , and gastrointestinal symptoms , impairment of the different types of O-glycosylation often causes musculoskeletal phenotypes [7] . Glycosaminoglycans ( GAGs ) , mostly attached to proteoglycan core proteins through glycanation processes in the Golgi apparatus , importantly contribute not only to tissue elasticity and organization of the ECM , but also regulate growth factor signaling [6] . One example are the small leucine rich proteoglycans decorin and biglycan , which carry dermatan or chondroitin sulfate GAG chains and regulate collagen fibrillogenesis and signaling mediated by diverse ligands [8] . Several known disorders of GAG synthesis are characterized by a prematurely aged appearance and fragile bones [9–11] . However , the molecular background of these pathologies has only partially been unraveled . We here report on the characterization of different constitutive and conditional mouse models indicating that GO is due to osteoblast dysfunction and that Gorab is most relevant in early stages of osteoblast differentiation . Gorab inactivation reduces dermatan sulfate levels and proteoglycan glycanation in skin and bone tissue . Loss of GORAB in fibroblasts leads to decorin retention and lower GAG levels in the Golgi compartment . Altered proteoglycans are not only associated with disorganization of the collagen matrix , but also with aberrant TGF-β activation , which likely perturbs differentiation and function of osteoblast lineage cells .
The majority of GORAB mutations found in GO patients lead to a loss of the GORAB protein [3] . Therefore , we first constitutively inactivated Gorab in a genetrap mouse line and in a complete knockout after removing the floxed exons 2 and 3 from the Gorabflox locus in the germline to study the unknown cause for bone fragility in GO ( S1A–S1D Fig ) . Both mouse lines , which are in the following referred to as GorabNull , showed an identical phenotype with absence of skin changes reminiscent of cutis laxa , but early lethality due to respiratory distress , most likely secondary to decreased alveolar airspace ( S2A–S2D Fig ) . This is in line with an independent description of GorabNull mice [12] . Apart from enlarged fontanels no significant skeletal abnormalities were identified ( S3A–S3E Fig ) . Especially the cortical bone porosity was in the normal range for this developmental stage ( S3C and S3D Fig ) . We took advantage of the β-galactosidase expressed from the Gorab genetrap locus to visualize the expression pattern in the developing skeleton since available Gorab antibodies lack specificity in immunohistology ( Fig 1A ) . Strongest signals were seen in the perichondrium and the periosteum , which governs formation of the cortical bone . Furthermore , we assessed Gorab expression in comparison to several marker genes in calvarial osteoblasts differentiated in vitro for 12 days until Dmp1 , a marker for late osteoblasts/osteocytes , was robustly expressed ( S4A Fig ) . Gorab expression peaked at day 6 of differentiation , together with type 1 collagen ( Col1a1 ) and decorin ( DCN ) , which was induced about 40-fold compared to day 0 ( S4A Fig ) . We also differentiated osteoblast precursors isolated from E18 . 5 and P0 GorabNull calvariae into mineralizing osteoblasts in vitro . Neither alkaline phosphatase activity nor mineralization as measured by alizarin red showed any differences correlating with the dramatic bone changes found in GO ( S3G and S3H Fig ) . We therefore hypothesized that development of the bone phenotype in our murine GO model occurs mainly postnatally when the embryonic woven bone is converted into mature lamellar bone , a process which is not faithfully recapitulated in the usual 2D osteoblast cultures . To investigate the role of Gorab in postnatal skeletal development we generated conditional Gorabflox mice to selectively prevent Gorab expression in bone tissue while preserving expression in the lung ( S1C and S1D Fig ) . We inactivated the gene in the limb bud mesenchyme , pre-osteoblasts , and late osteoblasts/osteocytes by crossing Gorabflox mice with Prx1- , Runx2- , and Dmp1-cre mice , respectively , to investigate at which differentiation stage osteoblast lineage cells are most sensitive to a loss of Gorab ( S4A Fig ) [13–15] . Gorab expression in cortical bone was reduced to a similar degree in all three conditional mouse lines ( S4B Fig ) . The resulting GorabPrx1 and GorabRunx2 mutants showed a retardation of long bone growth and a dramatic loss of cortical and trabecular bone ( Fig 1B–1D ) . In contrast , only a mild loss of trabecular bone was evident in tibiae from GorabDmp1 mutants ( Fig 1B–1D ) ( S5A Fig ) . Since Prx1 is not expressed in the axial skeleton no changes were observed in GorabPrx1 vertebral bone structure , but GorabRunx2 and GorabDmp1 mutants showed a clear vertebral osteopenia ( S5B Fig ) . These data suggest that , at least for cortical bone development , Gorab function is most important during differentiation of pre-osteoblasts from mesenchymal progenitor cells . We then focused our analysis on the GorabPrx1 model , which corresponded well to the GO long bone phenotype . The biomechanical and material properties of bone tissue from this mouse model have been described elsewhere [16] . Histological analysis of different stages of postnatal development revealed that the bone anomalies were most striking in the tibia at four weeks of age . Thinning and porosities of the cortical bone were most pronounced in the posterior metaphyses ( Fig 2A ) . Several cells were observed in the large cortical pores suggesting that they represent merged osteocyte lacunae . The overall number of osteocyte lacunae was elevated and the periosteum was strongly thickened at the expense of cortical bone ( Fig 2B ) . The metaphyseal cortical bone changes in GorabPrx1 culminated in very high bone fragility as mirrored by spontaneous fractures in up to 80% homozygous mutants within the first four weeks of postnatal development ( Fig 2C ) . Humeri and tibiae were most frequently affected , often leading to deformations . We postulate that Gorab deficiency in early osteoblast lineage cells during the first weeks of postnatal development cannot be compensated due to the immense rate of ECM production , which is known to exert stress on the cells secreting ECM components . We therefore next wanted to know more about the impact of loss of Gorab function on the cells of the bone multicellular unit . Histomorphometric analysis of GorabPrx1 tibiae revealed higher osteoblast and osteocyte numbers ( Fig 2D and 2E ) . In spite of more abundant osteoblasts , mineral apposition rate was reduced in GorabPrx1 mutants and an increased osteoid surface indicated impaired mineralization ( Fig 2F and 2G ) . A reduced cortical bone mineral to matrix ratio was also detected in twelve week old GorabPrx1 mutants by Fourier transform infrared ( FTIR ) imaging [16] . These findings were confirmed in a bone biopsy from a GO patient , demonstrating that our mouse model closely recapitulates the human condition ( S6 Fig ) . To elucidate the basis for these osteoblast lineage anomalies we performed expression analyses in tibial cortical bone tissue . Individual qPCR assessment of osteoblast marker genes revealed an upregulation of Spp1 and Sp7 ( Fig 2H ) . In addition , genome-wide expression analysis of GorabPrx1 diaphyseal cortical bone by array hybridization and qPCR verification showed upregulation of the osteocyte marker genes Dmp1 and Fgf23 , ( Fig 2H ) ( S1 Table ) . The upregulation of Dmp1 was already evident in E18 . 5 GorabNull bones , together with a slight suppression of Sost , which could indicate a delay in osteocyte maturation during prenatal development ( S3F Fig ) . Immunohistochemistry confirmed increased levels of the proteins osteopontin , osterix , and dentin matrix protein 1 in GorabPrx1 cortical bone ( Fig 2I ) . Interestingly , also the genes Ank , Enpp1 , and Mepe known to inhibit mineralization were induced in GorabPrx1 mutants ( S1 Table ) , possibly contributing to the observed osteoid mineralization defect ( Fig 2G ) . With the exception of normal Sost expression levels the gene expression profile in GorabPrx1 mutant cortical bone resembled that of osteoid osteocytes indicating that terminal osteocyte differentiation was impaired as a consequence of Gorab inactivation in osteoblast precursors [17] . Osteoclast numbers were not significantly changed in GorabPrx1 trabecular bone ( Fig 2J ) . Cell counts in the available human bone biopsy neither did show significantly elevated osteoclast numbers ( S6 Fig ) . Although ECM disorganization often induces increased osteoclast numbers and activity , this is probably suppressed in GorabPrx1 mutants by an elevated Opg/Rankl ratio detected by expression analysis ( Fig 2K ) [18 , 19] . Taken together , these data suggest low turnover kinetics due to functional impairment of the osteoblast lineage . This is in contrast to osteogenesis imperfecta ( OI ) , the most common type of congenital bone disease with fractures , which is characterized by high turnover kinetics [18] . Gerodermia osteodysplastica is characterized by a congenital ligamentous laxity indicating collagen abnormalities , and dermal elastic fiber changes . Investigating the dermis of newborn GorabNull mice by electron microscopy we found a disorganization of collagen fibers ( Fig 3A ) . In contrast to control skin almost no fibril formation was seen . Elastic fibers are not yet formed at this developmental stage and could therefore not be studied . GorabPrx1 cortical bone was thinner and showed reduced stiffness and increased fragility ( Fig 2B ) [16] . The mechanical properties of bone are also determined by collagen organization . Similar to the skin we also found missing alignment of collagen fibers surrounding the osteocytes in four week old GorabPrx1 mutants ( Fig 3B ) . Picrosirius red staining and polarized light microscopy confirmed globally impaired collagen fibrillogenesis in tibial cortical bone from GorabPrx1 mutants and in a pelvic bone biopsy from a nine year old GO patient ( Fig 3C ) . Our results show the presence of impaired collagen fibrillogenesis in GO , at least during development of bone and connective tissues . The Golgi localization of GORAB is suggestive of a function in protein secretion and/or modification . Metabolic labeling of fibroblasts from GO patients showed no influence of GORAB on global protein or collagen secretion ( S7A and S7B Fig ) . However , collagen and elastic fiber changes similar to GO were described in other progeroid connective tissue disorders secondary to impaired glycanation of proteoglycans [9–11] . To test a possible involvement of proteoglycans in GO pathology , we measured glycosaminoglycans ( GAGs ) in tissues from E18 . 5 GorabNull embryos . We observed a significant reduction in the amount of dermatan sulfate , but not of other GAGs in skin and cartilage , indicating a specific defect ( Fig 4A ) ( S7C Fig ) . In GorabPrx1 tibial cortical bone the total GAG levels as well as the relative amounts of dermatan sulfate levels were strongly reduced ( Fig 4B and 4C ) . Biglycan and decorin are major dermatan- and chondroitin sulfate-carrying proteoglycans in bone and connective tissues . Their fully glycanated form can be difficult to detect in immunoblots with tissue lysates due to their apparent size ( decorin 100 kDa , biglycan >200 kDa ) and low accessibility of the epitope residing in the core proteins that have a size of 42 kDa and 45 kDa , respectively . Immunoblot detection of both proteoglycans in skin tissue lysates showed a partial loss of the fully glycanated forms and stronger core protein bands in GorabNull mutants suggesting absent/shortened GAG chains ( Fig 4D ) ( S7D Fig ) . Furthermore , after treatment with chondroitinase ABC the core proteins of decorin and biglycan both gave stronger signals indicating lesser glycanation ( S7E and S7F Fig ) . No upregulation of mRNA expression was detected , which could alternatively explain the stronger core protein bands ( S7G and S7H Fig ) . Also in bone lysates from four week old GorabPrx1 animals the fully glycanated form of decorin was less abundant compared to littermate controls and bands with lower apparent weight were enhanced ( Fig 4E ) . Reduced decorin glycanation has been demonstrated to be a consequence of aging in human skin [20] . We found a similar reduction in decorin glycanation with aging in cortical bone from wildtype mice ( Fig 4F ) . Decorin glycanation in 4 week old GorabPrx1 bone was reduced to a level similar to that of 26 week old controls ( compare Fig 4E and 4F ) . Taken together , loss of Gorab caused a general reduction of proteoglycan glycanation , exemplified by the changes in glycanation status of decorin and biglycan , similar to that found in the aged ECM . Decorin and biglycan play a vital role in collagen fibrillogenesis [21] , are highly expressed in osteoblasts ( S4A Fig ) and were linked to alterations of matrix mineralization , growth factor signaling , and bone fragility [22–25] . In immunohistology we found enhanced signals for biglycan and decorin in the ECM of the thickened periosteum in GorabPrx1 tibiae , most likely due to enhanced accessibility of the epitope as a consequence of lower glycanation ( Fig 4G and 4H ) . This suggested a role of these pathologically glycanated proteoglycans in the abnormal gain of periosteum thickness . In order to investigate whether the observed proteoglycan abnormalities are due to a perturbation of the producing cells we investigated decorin production in cultured fibroblasts . In lysates from confluent GorabNull and control fibroblasts we observed a downward shift of the band corresponding to the glycanated form ( Fig 5A ) . A similar finding was obtained for biglycan ( S7E Fig ) . Also decorin secreted by fibroblasts from GO patients displayed a significant reduction of the fully glycanated form ( Fig 5B ) . These data demonstrate that the reduced glycanation of proteoglycans in Gorab-deficient mouse mutants can be recapitulated in vitro in both mouse and human fibroblast cell lines . The GORAB protein is associated with the medial/trans Golgi compartment , where it supposedly regulates transport processes [5] . Therefore , we hypothesized that the decorin abnormalities could be due to impaired intracellular trafficking . In order to prevent influence of extrinsic factors and to ensure comparable immunofluorescence stainings , the following experiments were done in co-cultures of fibroblasts derived from control and GO individuals . Both cell types could be distinguished by presence or absence of GORAB immunofluorescence staining ( Fig 5C ) . GORAB-deficient cells showed increased co-localization of decorin and the Golgi marker GM130 in immunofluorescence compared to GORAB-expressing control cells ( Fig 5C ) . We also assessed the dermatan sulfate levels at the Golgi compartment detected by antibody GD3A12 through immunofluorescence co-staining with the Golgi marker TGN46 [26] . Although decorin accumulated at the Golgi compartment , the signals for dermatan sulfate chains were less intense ( Fig 5D ) . Thus , loss of GORAB seems not only to reduce anterograde trafficking of decorin within the Golgi , but also dermatan sulfate attachment to its core protein . Given the abovementioned changes in total GAG levels in bone and in biglycan glycanation ( S7F Fig ) , this effect does not seem to be limited to decorin , but probably affects proteoglycans in general . Besides collagen , decorin and biglycan also bind TGF-β with high affinity and modulate its bioavailability and interaction with receptors [27 , 28] . TGF-β is a central regulator of bone remodeling produced by osteoblasts . It is secreted predominantly in its latent , inactive form and deposited into the matrix from where it gets activated by osteoclast activity and proteolytic cleavage [29 , 30] . Excessive TGF-β signaling has been shown to be crucial in the pathology of osteogenesis imperfecta and several connective tissue disorders [19 , 31] . Using a TGF-β reporter cell line , we found an increase of active TGF-β in GorabNull skin lysates , while total TGF-β levels remained constant ( Fig 6A ) . Evidence for elevated TGF-β signaling was also indicated by the upregulation of TGF-β responsive genes such as Serpine1 in cortical bone of GorabPrx1 mutants ( Fig 6B ) . Increased nuclear staining of p-Smad2 was found in GorabPrx1 bone tissue , in particular in cells located in the enlarged periosteum ( Fig 6C ) , indicating activation of TGF-β signaling in osteoblast lineage cells . In low passage ( 5–10 ) GO skin fibroblasts we also observed increased levels of p-SMAD2 , and elevated expression of TGF-β responsive genes ( Fig 6D and 6E ) . These findings suggest that GO can be regarded as a congenital disorder of glycosylation ( CDG ) leading to a disorganized collagen network and TGF-β activation due to abnormal function of the underglycanated proteoglycans , including decorin and biglycan .
Our results suggest that the loss of Gorab in the Golgi compartment causes reduced proteoglycan glycanation , abnormalities in collagen networks , and subsequent TGF-β overactivation . GAGs are essential for the function of proteoglycans . This is highlighted by several congenital disorders of glycosylation ( CDGs ) caused by deficient attachment of GAGs to the proteoglycan core protein in the Golgi compartment [9–11] . Besides generalized connective tissue problems and a progeroid appearance some of these disorders also show a dramatic bone fragility . In contrast to GO , these disorders are readily classified as CDGs since they affect enzymes involved in GAG synthesis . The typical readout used in functional studies for these disorders is decorin glycanation status [32] . Our data imply a strong clinical and biochemical overlap of these disorders with GO and suggest that GO might be regarded as a CDG . We used different cre-expressing mouse lines to learn more about the stage of osteoblast differentiation at which Gorab is most essential . Gorab expression peaks in mature osteoblasts , together with ECM proteins like type 1 collagen . It seems that only a loss of Gorab expression before this osteoblast differentiation stage in GorabPrx1 and GorabRunx2 mice gives rise to the full phenotypic picture , while deletion in late osteoblasts in the GorabDmp1 line has milder effects . Late osteoblasts are characterized by reduced ECM production and start to convert into osteocytes [17] . Likewise , the GorabPrx1 and GorabRunx2 mouse models develop age-related phenomena like cortical porosity and low bone turnover already during bone growth , thus hampering bone mass accrual . We speculate that the reason for this might be the strong ECM production during the postnatal growth spurt . From these aspects we conclude that loss of GORAB primarily perturbs strongly ECM secreting cells , which might explain the selective affection of bone and connective tissues in GO in spite of the ubiquitous expression of GORAB . The small leucine-rich proteoglycans ( SLRPs ) are important for ECM homeostasis by regulating collagen fibrillogenesis and binding to TGF-β and other ligands and receptors in a complex manner [8] . It has been hypothesized that upon alteration of the collagen network SLRP binding to the ECM is loosened leading to increased release of active TGF-β [19] . Diffusible biglycan and decorin might even enhance binding of TGF-β to its receptor , thereby further inducing collagen production to repair the defective collagen network [33] . The knockout of decorin and also the exchange of its GAG-carrying serine residue have only mild phenotypic effects [34 , 35] . In contrast , loss of biglycan leads to osteopenia and cortical thinning quite similar to GorabPrx1 and GorabRunx2 mutants [25] . Human biglycan loss-of-function mutations cause aortic dilatation , malar hypoplasia , and osteopenia with thin cortices , a clinical picture somewhat reminiscent of GO [36] . Defective decorin glycanation , which we show in aging bone tissue , has also been demonstrated to be a consequence of skin aging [20] . Generally , the sulfated GAG content of many tissues is reduced with increasing age [37] . From an evolutionary standpoint it was suggested that age-related changes in glycosylation patterns might shield the organism from cancer by lowering proliferation signals and enhancing differentiation signals like TGF-β [38] . Binding of biglycan and decorin to TGF-β and to type 1 collagen is mediated by the core protein and can be lowered by the attached GAG chains [39] . On the other hand , dermatan sulfate was shown to bind to collagen fibrils in a reproducible pattern , which might explain the observed differences in the binding patterns of glycanated and non-glycanated decorin to collagen fibrils [40 , 41] . We therefore assume that reduced glycanation of SLRPs plays a leading role in the pathogenesis of GO . In our mouse models and also in cultured cells we found evidence for elevated TGF-β activation , which we propose to be a consequence of the disrupted ECM . TGF-β signaling plays pivotal role in bone homeostasis [30] . TGF-β stored in the bone matrix is released and activated during bone resorption . This is turn attracts bone osteoprogenitor cells to the resorption sites , which subsequently proliferate and differentiate into osteoblasts , thus coupling bone resorption to bone formation [29 , 42] . A dampening of TGF-β signaling has been shown necessary for terminal osteoblast differentiation [43] . Furthermore , persistently high doses of TGF-β have been shown to inhibit osteoblast differentiation in a physiologically relevant manner as shown for osteogenesis imperfecta and other disorders of the ECM [19 , 31 , 44] . Transgenic overexpression of TGF-β2 in murine osteoblasts results in osteopenia , cortical thinning , and elevated osteocyte numbers [45] . Moreover , constitutive overexpression of Sp7/osterix in osteoblasts causes an accumulation of abnormal osteocytes in the absence of elevated osteoclast numbers [46] . Both mentioned mouse models phenotypically closely overlap with the GorabPrx1 mutants . Interestingly , a direct upregulation of Sp7/osterix by TGF-β has been demonstrated [47] . Furthermore , the expression of Dmp1 , which is strongly induced in Gorab-deficient bone tissues , was found to depend on Tcf11 , which is also regulated by TGF-β [48 , 49] . TGF-β has been demonstrated to promote Rankl-induced osteoclastogenesis [50] . In contrast to other models with elevated TGF-β signaling there is no consistent increase in osteoclast numbers in Gorab-deficient mice [19 , 44] . This can be possibly attributed to the elevated Opg/Rankl ratio in our mouse model . One explanation for this might be the timing of exposure of osteoblast lineage cells to elevated TGF-β . In fibrillin 1-deficient mice TGF-β signaling is altered at the level of the stromal mesenchymal stem cell niche , while in GorabPrx1 later stages of osteoblast differentiation are affected [44] . The localization of GORAB in the medial/trans Golgi is perfectly in line with an impairment of GAG chain elongation , which is carried out in this compartment [51] . The function of GORAB interaction partners ARF5 , RAB6 , and SCYL1 in retrograde Golgi trafficking suggests that the transport of factors relevant for glycanation might be impaired , similar to the regulation of EXT proteins by the Golgi protein GOLPH3 [3 , 5 , 52] . Golgi trafficking defects are often difficult to investigate in vitro . Even in the lethal phenotypes caused by mutations in COG7 or TRIP11 , encoding a component of the conserved oligomeric Golgi complex and the golgin Gmap-210 , respectively , only relatively mild abnormalities were found in cultured mutant cells [53 , 54] . Only after acute loss of Gmap-210 a global impairment of protein secretion and a retrograde trafficking delay became visible [55] . These results together with the clear glycanation impairment in vivo underline the significance of the proteoglycan alterations observed in GORAB deficient fibroblasts . Although multiple lines of evidence point towards a central role of a perturbed proteoglycan-TGF-β axis in osteoblast lineage dysfunction in GO , it has been described that loss of Gorab impairs hedgehog signaling leading to reduced hair growth in GorabNull mutants [12] . Although GO patients show no hair phenotype and the bone phenotype is not typical for altered hedgehog signaling it is possible that this mechanism also contributes to the GO bone pathology [56] . Interestingly , mice deficient for the golgin Gmap-210 not only had a lung pathology very similar to GorabNull mutants , but were also reported to show cilia abnormalities [57] . On the other hand , also knockout of the Golgi enzyme gPapp impairing GAG sulfation leads to a lung phenotype closely resembling that of GorabNull mutants , which supports our hypothesis of a GAG-driven pathomechanism [58] . Further studies are needed to disentangle the contributions of these different pathways . In summary , our study provides a link between the Golgi compartment , intracellular proteoglycan transport and glycanation , ECM disorganization and porosity , and TGF-β overactivation . Some aspects of this pathomechanism are also seen in chronological bone aging , but seem to occur already during postnatal growth in GO leading to the progeroid phenotype . The mechanism described here places GO in close proximity to congenital disorders of glycosylation with impaired proteoglycan synthesis .
Permission for work with human cells was granted by the Charité Ethics Committee ( approval number EA2/145/07 ) . We or the referring clinicians obtained oral informed consent from patients for genetic testing and the use of fibroblasts derived from skin biopsies . Healthy control individuals also gave written informed consent for use of fibroblasts . All animal experimental procedures were approved by the Landesamt für Gesundheitsschutz und Technische Sicherheit ( LaGeTSi ) , Berlin , Germany ( approval number G0213/12 ) . Experiments using animal-derived materials were conducted according to the German law for animal protection ( TierSchG ) . All analyses were done on homozygous GorabGt genetrap or Gorab-/- ( GorabNull ) embryos , or on female mice homozygous for the conditional Gorabflox allele and heterozygous for the Prx1-cre transgene ( GorabPrx1 ) . Homozygous conditional Gorabflox mice without the cre allele from the same generation served as controls . All animals had been backcrossed with C57/Bl6 mice for at least 5 times . GorabNull animals were generated using genetrap ES-cell clone XG183 purchased from Bay Genomics , San Francisco , CA , USA . Germline deletion of Gorabflox allele by crossing with CMV-cre mice resulted in homozygous mutants that were phenotypically identical to GorabNull mice . The construction strategy was as shown in S1C Fig . A BAC clone containing the mouse Gorab locus , BAC clone bMQ-373H11 ( 129S7Ab2 . 2 ) was obtained from Geneservice Ltd ( Cambridge , UK ) . A ~11 . 1kb region containing exons 2 to 4 of Gorab and the flanking introns was extracted into a pBluescript vector by recombination . LoxP sites were inserted into the targeting vector flanking exons 2 and 3 by homologous recombination . Gene targeting of mouse ES cells was done with the help from the Transgenic Core Facility of The University of Hong Kong . Correctly targeted clones were identified by Southern blot and injected into blastocysts from C57BL/6 mice to generate chimeric mice . Mice with germline transmission of the conditional cassette were crossed with ß-actin-flp mice to remove the neomycin selection cassette . The resulting animals were crossed with Prx1-cre mice to yield GorabPrx1 mutants . Crossing with CMV-cre mice resulted in Gorab-/- mutants through germline deletion . microCT analysis for the GorabPrx1 and GorabDmp1 mutants and corresponding littermate controls were done with Scanco μCT40 ( Scanco Medical , Brüttisellen , Switzerland ) at 10μm resolution; while GorabRunx2 and littermate controls were analyzed with Skyscan 1172 ( Bruker microCT , Luxemburg , Belgium ) at 5μm resolution . Tibiae and vertebra were first fixed in 4% paraformaldehyde ( PFA ) in 4°C for 24 hours then scanned in 70% ethanol at 10μm resolution . The trabecular bone measurement was done for a region of 700 μm in the secondary spongiosa of proximal metaphysis of the tibia and the entire vertebral body . 1 mm of cortical bone was measured at a region 1200 μm below the proximal metaphysis of the tibia . E18 . 5 GorabNull embryos were sacrificed and dissected under PBS for lung biopsies to prevent collapse of the lung . Samples were then fixed in 4%PFA at 4°C for 24hours and then dehydrated through gradient ethanol from 70% , 80% , 90% and 100% ethanol at 4°C for 24hours at each step . The samples were then cleared in xylene twice at room temperature for 15min and then infiltrated with paraffin for 60 minutes three times followed by embedding . The embedded samples were sectioned and stained with hematoxylin/eosin for histology . Undecalcified mouse tibiae from E18 . 5 GorabNull embryos and four week old GorabPrx1 mice were first fixed in 4% PFA and subsequently embedded in methylmethacrylate , MMA ( Cat#00834 , Polysciences , Eppelheim , Germany ) and sectioned for histological studies . The fixed bone samples were dehydrated through gradient ethanol from 70% , 80% , 90% to 100% and twice in xylene for 24h in each step . The samples were then infiltrated with infiltration MMA ( 1%v/v polyethylene glycerol ( Sigma-Aldrich , Munich , Germany ) and 0 . 33% w/v benzoyl peroxide in MMA ) for at least 24hours at 4°C . The polymerization was carried out at 4°C in polymerization solution ( 1%v/v polyethylene glycerol , 0 . 55% w/v benzoyl peroxide , 0 . 5% v/v N , N-dimethyl-p-toluidine in MMA ) . The embedded samples were then sectioned using a Leica RM2255 microtome ( Leica , Wetzlar , Germany ) at 5 μm thickness and subjected to Von Kossa/Van Giesson staining , Von Kossa/hematoxylin staining , Goldner trichrome staining or Pircosirius red staining . Histomorphometric analysis of the secondary spongiosa of proximal tibia of 4 weeks old GorabPrx1 mice was carried out using the software Osteomeasure ( Osteometrics , Atlanta , USA ) . For mineral apposition rate determination , GorabPrx1 and control animals were injected with calcein ( 10μl per gram body weight ) at P23 and P26 and sacrificed at P28 . The tibiae of the animals were fixed in 4% PFA and embedded in MMA as mentioned previously and then sectioned at 5μm thickness . The distance between the two lines of calcein labels at the mid bone shaft was subsequently imaged and measured for mineral apposition rate calculation . All specimens were fixed for at least 2h at room temperature in 3% glutaraldehyde solution in 0 . 1M cacodylate buffer pH 7 . 4 and then cut into pieces of 1mm3 . The samples were then washed in buffer and postfixed for 1h at 4°C in 1% osmium tetroxide , followed by dehydration through graded ethanol solutions and embedded in epoxy resin ( glycidether 100 ) . Semithin and ultrathin sections were cut with an ultramicrotome ( Reichert Ultracut E ) . Ultrathin sections were treated with uranyl acetate and lead citrate , and examined with an electron microscope Philips EM 400 . GAGs were prepared from E18 . 5 embryo skin , lung , cartilage and brain as described previously [59] . Each extract was digested with chondroitinase AC ( Seikagaku Corp . , Tokyo , Japan ) , chondroitinase B ( IBEX Tech . , Montreal , Canada ) , or heparinase ( IBEX Tech . , Montreal , Canada ) for analyzing the disaccharide composition of CS , DS , or HS , respectively . Each digest was labeled with a fluorophore 2-aminobenzamide ( 2-AB ) ( Nacalai tesque , Kyoto , Japan ) , and analyzed by anion-exchange HPLC . To recover GAGs from bone specimens , tibia diaphysis from 4 week old GorabPrx1 and wildtype mice were decalcified and digested with 20U of papain ( Sigma-Aldrich , Milano , Italy ) in 100mM sodium acetate , pH 5 . 6 , 100mM EDTA and 5mM cysteine at 65°C for 48h . GAGs were purified by cetylpyridinium chloride precipitation and hyaluronic acid was removed by digestion with Streptomyces hyaluronidase ( Seikagaku Corp . , Tokyo , Japan ) followed by ultrafiltration as described previously [60] . For CS and DS disaccharide composition , GAG aliquots were digested with chondroitinase ABC ( AMSBIO , Abingdon , UK ) ( digesting chondroitin sulfate A , B and C ) or chondroitinase ACII ( Seikagaku Corp . , Tokyo , Japan ) ( digesting chondroitin sulfate A and C ) . Each digest was labeled with 2-aminoacridone ( Thermo Fisher Scientific , MA , USA ) and analyzed by reverse phase HPLC as described previously [61] . The dermatan sulfate ( chondroitin sulfate B ) fraction was determined as disaccharide fraction undigestable by chondroitinase ACII vs . disaccharides digested by chondroitinase ABC . Cell lysates from in vitro cultures were extracted with RIPA buffer followed by sonication . Mouse tissue samples were pulverized in liquid nitrogen and lysates were extracted with 8M urea buffer , 1% SDS with sonication . For chondroitinase ABC digestion the buffer was exchanged to 0 . 1M Tris-HCl pH8 . 0 using microcon 10kDa centrifugal filter units and subsequently incubated at 37°C for 18 hours with or without chondroitinase ABC ( ABCase , 0 . 3U/100μg protein ) . Mouse bone biopsies were first fixed in 4% PFA and then decalcified in Morses’ solution ( 10% Sodium Citrate , 20% Formic Acid ) . The decalcified bone were then embedded in paraffin and sectioned for immunostainings . Antibodies used are as follows: Anti-p-Smad2 ( #3101 , Cell Signaling , Leiden , The Netherlands ) , anti-SMAD2 ( #3102 , Cell Signaling , Leiden , The Netherlands ) , ( Anti-Gapdh ( #sc6215 , Santa Cruz , Heidelberg , Germany ) , Anti-DCN ( #AF143 and #AF1060 , R&D , Abingdon , UK ) , Anti-BGN ( #ENH020 , Kerafast , Boston , USA ) , Anti-Sp7 ( #ab22552 , Abcam , Cambridge , UK ) , Anti-Spp1 ( #MPIIIB10 , DSHB , Iowa , USA ) , Anti-Dmp1 ( #AF4386 , R&D , Abingdon , UK ) . For immunofluorescence on cells we used the following antibodies: anti-decorin ( #14667 Proteintech ) anti-GM130 ( clone 35 , BD Biosciences ) , anti-VSV-tag ( clone P5D4 , Sigma ) . Sheep and rabbit GORAB antibodies were raised against GST-tagged N-terminal ( 1–130 aa ) and C-terminal ( 301–369 aa ) regions of human GORAB respectively and affinity purified against these same proteins . Serum was pre-cleared against GST before affinity purification on immunogen . Antibodies against TGN46 were provided by Dr S . Ponnambalam ( University of Leeds , United Kingdom ) , while GD3A12 anti-DS antibody was a kind gift of Dr Toin van Kuppevelt ( Radboud University Nijmegen Medical Center , The Netherlands ) [26] . For immunohistochemistry , Vectastain elite ABC kit ( Vector Laboratories , Burlingame , CA ) was used for signal detection . For immunofluorescence on sections , the Tyramide signal amplification system ( Perkin Elmer , Baesweiler , Germany ) was used for signal development . Calvarial osteoblast progenitor cells were isolated as previously described [62] . Cells were seeded on 6-well plates in Alpha-MEM ( Lonza , Basel , Switzerland ) containing 10% fetal calf serum ( FCS; Gibco , Life Technologies , Carlsbad , California , USA ) as well as Pen/Strep ( 100U/mL , Lonza ) and 2 mM ultra-glutamine ( Lonza ) . Osteogenic differentiation was induced by with 50 μM L-ascorbate-2-phosphate and 10 mM beta-glycerophosphate . Quantitative AP and matrix mineralization assays were performed as previously described [62] . Shortly , AP activity was determined by homogenizing 3 replicates separately in ALP-buffer1 ( 0 . 1 M Glycine; 1% NP-40; 1 mM MgCl2; 1 mM ZnCl2 ) . After the addition of 1 volume of ALP-buffer2 ( 5 mM p-nitrophenyl phosphate [p-NPP] , 0 . 1 M glycine , pH 9 . 6 , 1 mM MgCl2 , and 1 mM ZnCl2 ) , reactions were incubated at 37°C for 30 min and stopped by addition of 1 M NaOH . The amount of p-NP released from the substrate p-NPP was recorded at 405 nm . AP activity is given as unit of absorption / μg protein / 30 min . Cells were fixed with 4% PFA prior to staining with Alizarin Red . Skin biopsies from E18 . 5 GorabNull embryos were homogenized in liquid nitrogen and soluble protein was extracted with 1X PBS , 1X complete protease inhibitor cocktail ( Roche , Mannheim , Germany ) for 16h at 4°C with agitation . Protein content of the lysates was then quantitated by BCA assay ( Thermo Fisher , Dreieich , Germany ) and 500μg of protein ( either heat activated at 80°C for 10min for the total TGF-ß or without activation ) was added to Plasminogen activator inhibitor-1–luciferase reporter mink lung epithelial cells ( gift from Prof . Petra Knaus , Freie Universität , Berlin , Germany ) and incubated for 16h . The lysates of the reporter cells were than collected for the luciferase activity using the Dual luciferase reporter system ( Promega , Mannheim , Germany ) and the amount of TGF-ß in skin lysates was quantitated by comparing with reporter cells treated with known amounts of recombinant TGF-ß ( eBioscience , Frankfurt , Germany ) . Post-confluent human skin fibroblast were labelled for 18 h with 10μCi/ml L-[2 , 3 , 4 , 5-3H]-Proline in medium containing 0 . 15mM ascorbic acid . The secreted collagen in the medium was precipitated with 25% ammonium sulfate and then resuspended in 50mM Tris , 150mM NaCl pH7 . 5 , which was then subjected to pepsin digestion ( 50μg/ml ) at 4°C for 16h . The digested samples were then lyophilized and separated by SDS-PAGE and the labelled collagen chains were detected by fluorography . WT and GO human skin fibroblasts were seeded into 6-well plates and cultured for 8 days under standard conditions with media change on day 4 . After 8 days cells were washed twice with PBS and the ECM was extracted by incubation with 300 μL of urea buffer ( 6 M urea , 25 mM dithiothretol ( DTT ) in 25 mM ammonium bicarbonate ) for 10 min at RT . ECM samples were then subjected to SDS-PAGE and decorin was analyzed by WB . WT and GO MEFs were seeded into 6-well plates and cultured under standard conditions until they reached confluency . Cells were then stimulated to produce ECM components by incubation in fibroblast-specific serum-free medium ( Lifeline Cell Technology ) for 7 days . Cells were then lysed , samples were subjected to SDS-PAGE and decorin was analyzed by western blotting . WT and GO fibroblasts were seeded as a co-culture on glass coverslips to ensure fair comparison of signal intensities between controls and mutant cells and to exclude the influence of extrinsic factors . The cells were grown till 90% confluency . Cells were washed twice with PBS , fixed with 3% ( wt/vol ) PFA in PBS for 25 min at RT . Cells were then washed with PBS and the excess of paraformaldehyde was quenched with glycine . The cells were permeabilized by 4 min incubation in 0 . 1% ( wt/vol ) Triton X-100 in PBS . Cells were incubated with primary antibody diluted in PBS for 1 hour at RT and incubated three times with PBS for 5 min . Then coverslips were incubated for 1 h with secondary antibody diluted in PBS and incubated three times with PBS for 5 min and twice in ddH20 for 5 min . In case of the GD3A12 antibody detecting dermatan sulfate , the primary antibody was recognized by P5D4 clonal antibody against VSV-tag prior to incubation with fluorescently-conjugated secondary antibody . Coverslips were dried before mounting in Mowiol 4–88 and images were acquired on a Ti inverted microscope ( Nikon ) using a x60/1 . 4 Plan Apo objective , Proscan II motorized stage ( Prior Scientific ) and R6 CCD camera ( QImaging ) . A SpectraX LED light engine ( Lumencore ) , quad dichroic ( Semrock ) and motorized emission filter wheel ( Prior Scientific ) with single bandpass filters for FITC , TRITC and Cy5 ( Semrock ) were used to collect image sequences at each position in the tile . Images were acquired and then aligned and stitched using NIS Elements software ( Nikon ) . These stitched images were then exported as a single TIFF image for further processing in Fiji software [63] . The amount of intra-Golgi decorin and GAG was measured by comparing fluorescence intensity levels with reference to the Golgi markers GM130 and TGN46 . GORAB staining was employed to discriminate between WT and GO fibroblasts . All values are shown as mean ± SD . Data were analyzed with unpaired-two-tailed Student’s t test . Microarray data was analyzed by One-Way-ANOVA . A p-value smaller than 0 . 05 was considered statistically significant . In all figures , * p-value < 0 . 05 , ** p-value < 0 . 01 , *** p-value < 0 . 001 , n . s . = not significant
|
Gerodermia osteodysplastica ( GO ) is segmental progeroid disorder affecting connective tissues and bone , leading to extreme bone fragility . The cause are loss-of-function mutations in the Golgi protein GORAB , whose function has been only partially unravelled . Using several mouse models and patient-derived primary cells we elucidate that loss of Gorab elicits a defect in proteoglycan glycanation , which is associated with collagen disorganization in dermis and bone . We also found evidence for TGF-β upregulation and enhanced downstream signalling . If these changes occur in mesenchymal stem cells or early osteoblasts they impair osteoblast differentiation resulting in cortical thinning and spontaneous fractures . We thus match GO mechanistically with also phenotypically overlapping progeroid connective tissue disorders with glycanation defects .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"tibia",
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"salts",
"osteoblast",
"differentiation",
"fibroblasts",
"collagens",
"cell",
"differentiation",
"animal",
"models",
"developmental",
"biology",
"model",
"organisms",
"osteoblasts",
"proteoglycans",
"connective",
"tissue",
"cells",
"experimental",
"organism",
"systems",
"skeleton",
"research",
"and",
"analysis",
"methods",
"musculoskeletal",
"system",
"animal",
"cells",
"proteins",
"sulfates",
"connective",
"tissue",
"biological",
"tissue",
"mouse",
"models",
"chemistry",
"biochemistry",
"cell",
"biology",
"anatomy",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"glycobiology"
] |
2018
|
Impaired proteoglycan glycosylation, elevated TGF-β signaling, and abnormal osteoblast differentiation as the basis for bone fragility in a mouse model for gerodermia osteodysplastica
|
The 2013–2016 Ebola virus outbreak in West Africa was the largest and deadliest outbreak to date . Here we conducted a serological study to examine the antibody levels in survivors and the seroconversion in close contacts who took care of Ebola-infected individuals , but did not develop symptoms of Ebola virus disease . In March 2017 , we collected blood samples from 481 individuals in Makeni , Sierra Leone: 214 survivors and 267 close contacts . Using commercial , quantitative ELISAs , we tested the plasma for IgG-specific antibodies against three major viral antigens: GP , the only viral glycoprotein expressed on the virus surface; NP , the most abundant viral protein; and VP40 , a major structural protein of Zaire ebolavirus . We also determined neutralizing antibody titers . In the cohort of Ebola survivors , 97 . 7% of samples ( 209/214 ) had measurable antibody levels against GP , NP , and/or VP40 . Of these positive samples , all but one had measurable neutralizing antibody titers against Ebola virus . For the close contacts , up to 12 . 7% ( 34/267 ) may have experienced a subclinical virus infection as indicated by detectable antibodies against GP . Further investigation is warranted to determine whether these close contacts truly experienced subclinical infections and whether these asymptomatic infections played a role in the dynamics of transmission .
There are six antigenically distinct species in the genus Ebolavirus that vary in viral pathogenesis . Infections caused by Zaire ebolavirus result in the highest lethality in humans with case fatality rates during outbreaks ranging from 41% to 90% ( average rate , 78% ) . Ebola virus ( EBOV ) is typically introduced into human populations through direct contact with or the consumption of infected nonhuman primates or other intermediate mammalian hosts or through bats , a potential natural reservoir of EBOV [1] . Human-to-human transmission occurs through direct contact with virus-laden secretions or fluids [2] . Initial symptoms of EBOV infection include fever , cough , rash , and abdominal pain , which occur within 2 to 21 days of contact with the virus , and progress to fatigue , headache , vomiting , diarrhea , shock , organ failure , and potential death . A total of 14 documented EBOV outbreaks have been reported in Central Africa . The 2013–2016 EBOV outbreak in West Africa was the first for this region of Africa; it was also the largest and most devastating EBOV outbreak to date resulting in over 28 , 600 identified human cases and 11 , 300 deaths . These figures include 881 cases of infected health care workers , including 513 deaths . The outbreak was located primarily in the West African countries of Sierra Leone , Liberia , and Guinea , but seven other countries experienced imported cases . Although the highly pathogenic nature of EBOV is well-established , several studies have assessed the incidence of asymptomatic infections that result in seroconversion in the absence of symptoms of disease [3–11] . These studies report a wide variability of seroprevalence , ranging from 1 . 0% to 45 . 9% , which precludes an accurate summary estimate of asymptomatic human cases . In addition to human cases , asymptomatic cases have been documented experimentally in animal models such as ferrets [12] and nonhuman primates [13] . Limited information is available regarding the antibody status of survivors of the West African outbreak and the number of asymptomatic cases that occurred in Sierra Leone . To address this lack of information , we obtained samples from EBOV survivors and from individuals who cared for virus-infected patients either at home or in treatment centers . We assessed antibody levels in these samples by using an ELISA against the three major viral antigens , GP , NP , and VP40; we also evaluated neutralizing antibody titers .
The study was carried out in Makeni ( estimated population of 112 , 428 in 2013 ) , the capital of the Bombali District of Sierra Leone , which experienced 1 , 050 confirmed EBOV cases during the 2014–2016 outbreak . Recruitment of adult volunteers ( survivors and close contacts ) was performed by the Sierra Leone Association of Ebola Survivors of Makeni . Demographic data and information were collected using a questionnaire . The study also included a control cohort of 38 individuals with no known exposure to EBOV and no relationship to EBOV-infected individuals . A peripheral blood sample ( ~3 ml ) was collected in an EDTA vacutainer tube ( Becton Dickinson , Franklin Lakes , NJ ) by local , experienced technicians at the Makeni medical center . Blood samples were stored at 4°C for less than 24 hours prior to isolation of the plasma fraction . The plasma was then treated at 55°C for 30 minutes , divided into aliquots , and stored for use at -80°C . By using commercial , quantitative ELISAs ( Alpha Diagnostics International , San Antonio , Texas ) , we tested the plasma obtained from the blood samples at a 1:400 dilution at least in duplicate for levels of IgG-specific antibodies against the three major viral antigens: the surface GP , the only virally expressed protein on the virion surface; NP , the most abundant viral protein; and VP40 , a major structural protein of EBOV ( strain Mayinga ) . Using the calibrators provided in the ELISA kits , a threshold index for each ELISA run was established to discriminate between positive and negative antibody responses and determine antibody levels ( expressed in units/ml ) . Distribution of antibody levels and determination of mean antibody levels were determined using GraphPad Prism 7 . Neutralizing antibody titers were determined using our replication-defective EBOVΔVP30 system that lacks the essential VP30 protein , but undergoes efficient replication in cell lines that are genetically engineered to stably express VP30 [14] . The virus system is approved for biosafety level-2 containment at the University of Wisconsin and is excluded from the CDC’s Select Agent registration . Serial 2-fold dilutions ( 1:4 to 1:1 , 024 ) of heat-inactivated plasma samples were mixed with an equal volume of ~1 , 000 focus-forming units of EBOVΔVP30 containing the Renilla luciferase reporter gene that resulted in an additional 2-fold dilution of the plasma sample . The virus-antibody mixture ( in duplicate ) was used to infect 96-well plates of VeroVP30 cells , a Vero cell lines that express the EBOV VP30 gene in order to facilitate EBOVΔVP30 replication . Three days after infection , a live-cell luciferase reagent was added to the wells , and luciferase activity ( a measurement of virus replication ) was determined as relative light units ( RLU ) . Neutralizing antibody titers were defined as the highest plasma dilution that resulted in a 50% reduction in RLU compared to a plasma control . The study was approved by the Ethical Review Board of the Ministry of Health and Sanitation of Sierra Leone and the Human Subjects Institutional Review Boards at the University of Wisconsin and the University of Tokyo . All participants in this study were adults . Inclusion in the study was voluntary with written informed consent provided by obtaining a signature or fingerprint from each participant and a signature from a witness .
In March 2017 , a total of 481 blood samples were collected from a cohort in Makeni , Sierra Leone , consisting of survivors of the 2013–2016 EBOV outbreak ( n = 214 ) and individuals who had close contact with EBOV-infected individuals during their illness , but did not develop symptoms of Ebola virus disease ( n = 267 ) . In the cohort of close contacts , health care workers ( n = 56 ) reported working at an Ebola treatment unit for a time period of 1–2 years and consistently used personal protective equipment during their interactions with EBOV-infected individuals . Four health care workers took part in an Ebola vaccine clinical trial and were not included in this study . Also in the cohort of close contacts were relatives ( n = 211 ) who took care of a sick family member on average for 10 days before the infected family member was taken to a treatment unit . Only 66% of relatives reported using personal protective equipment while taking care of sick family members due to the limited access to these items . The blood samples were collected from survivors 15–32 months ( median = 28 months ) after recovery from infection and release from a treatment unit . In this cohort , 85 . 0% , 10 . 7% , and 1 . 9% of samples were positive for antibodies against all three , two , or one viral antigen by ELISA , respectively , while 2 . 3% of the samples were negative against all three antigens ( Table 1 ) . In the survivor cohort , the mean NP antibody level was 2 , 015 units/ml , which was higher than the mean antibody levels against the other two antigens ( GP and VP40 ) ( Fig 1 and Table 2 ) . In the cohort of close contacts , 6 . 7% of samples were positive for antibodies against all three viral antigens ( 0 health care workers and 18 relatives ) , 7 . 5% were positive for antibodies against two viral antigens ( 3 health care workers and 17 relatives ) , and 25 . 8% were positive for antibodies against one viral antigen ( 13 health care workers and 56 relatives ) ( Table 1 ) . Of the close contacts who were positive for antibodies against one or two antigens , a majority of samples ( n = 71 ) possessed antibodies against VP40 . In general , the mean antibody levels against each viral antigen measured in the close contact cohort were lower than those in the survivor cohort ( Fig 1 and Table 2 ) . As a comparison , we examined antibodies against GP in a limited control cohort from Makeni ( n = 38 ) , but we were unable to detect measurable antibody levels ( S1 Table ) . Next , we examined the samples for neutralizing antibodies . Survivors that were positive for GP antibodies by ELISA ( n = 209 ) also had detectable neutralizing antibody titers ( except for one survivor ) with the majority of survivors having titers ranging from 1:128 to 1:512 , but some survivors had titers of great than 1:2048 ( Fig 2A ) . Close contacts that were positive for GP antibodies by ELISA ( n = 34 ) had a range of neutralizing antibody titers from 1:8 to 1:1024 while 3 of these samples had no detectable neutralizing titer ( Fig 2B ) . For close contacts that were antibody positive for other viral antigens ( NP and/or VP40 , but not GP; n = 73 ) , the majority of these close contact samples had an undetectable neutralizing antibody titer; however , titers were detected in 7 individuals at 1:8 and 1:16 ( Fig 2C ) .
While one research group has examined the T-cell response in EBOV survivors of the West Africa outbreak [15] , little is known about the antibody status of survivors of the West Africa outbreak . Here , we demonstrate that two years after infection , 97 . 7% of EBOV survivors in our study have measurable antibody levels against GP , NP , and/or VP40 , and all of the survivors , but one had measurable neutralizing antibody titers against EBOV . The lack of detectable antibody levels in some survivors ( 2 . 3% , n = 5 ) could reflect immune defects resulting in low and/or short-lived antibody responses , or could be due to technical errors or miscommunication during sample collection . Serology studies have examined the incidence of asymptomatic EBOV cases in populations living in endemic and non-endemic areas as well as in populations that have known and unknown contacts with infected individuals ( reviewed in [4] ) . In these studies , EBOV-specific antibody levels were assessed by using different techniques ( e . g . , immunofluorescence assay and commercial or ‘home-made’ ELISA kits ) , which most likely contributed to the wide variation in the seroprevalence rate of 1%–45 . 9% , depending on the method of antibody detection used [4] . In our study of asymptomatic cases , we determined antibody levels against three major viral antigens by using commercial ELISA kits and supplemented our findings by determining neutralizing antibody titers . Given protein homology of EBOV NP and VP40 to related viruses such as paramyxoviruses or rhabdoviruses , there could be cross-reactivity resulting in false-positive results in the NP and VP40 ELISAs [16] . Therefore , we based our incidence of asymptomatic infections on antibodies against GP such that 34 close contacts ( 12 . 7% ) were positive for GP antibodies , and a majority of individuals ( 91 . 2% ) also had measurable neutralizing antibody titers . However , it is unknown when or how these individuals were exposed to EBOV or if they were exposed to a filovirus antigenically similar to EBOV . Asymptomatic cases have been documented for different viral infections including influenza virus and Zika virus [17–19] . For EBOV , these asymptomatic cases may be influenced by the route of infection , the exposure dose , or both . A recent nonhuman primate study demonstrated that a low challenge dose of EBOV ( 10 virus particles ) by oral inoculation resulted in virus shedding , but never resulted in any clinical signs of infection [13] . Similar subclinical cases have been observed in the ferret model of EBOV infection . While infected ferrets developed clinical symptoms of EBOV infection , non-experimentally infected , cage mates that had direct contact with infected animals developed antibodies against EBOV , but never showed any signs of illness [12] . Given that our knowledge of EBOV transmission between individuals is incomplete , it is important to study these asymptomatic cases further to clarify their potential role in the transmission dynamics of an EBOV outbreak .
|
As the causative agent of an often lethal hemorrhagic fever disease in humans and nonhuman primates , Zaire ebolavirus typically causes high fever , severe diarrhea , and vomiting which results in case fatality rates as high as 90% . The 2013–2016 outbreak in West Africa was the largest and most devastating Ebola outbreak to date resulting in over 28 , 600 identified human cases and 11 , 300 deaths . Though our knowledge of virus transmission is incomplete , we do know that transmission occurs through direct contact with virus-contaminated body fluids ( blood , secretions , or other body fluids ) , materials such as bedding contaminated with these fluids , and through the handling and preparation of contaminated food . Asymptomatic Ebola virus infections that result in seroconversion in the absence of disease symptoms have been observed both in humans and experimentally in animal models . In the present serology study , we determined a majority of Ebola survivors in our cohort had measurable antibody levels against at least one viral antigen , as expected . In our cohort of close contacts , relatives and health care workers who took care of Ebola-infected individuals during the outbreak , we observed a rate of seroprevalence of 12 . 7% as indicated by detectable GP antibody levels . Given that Ebola virus is typically associated with a highly lethal disease in humans , it is of great interest to determine the host-virus interactions and transmission dynamics associated with asymptomatic cases .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"enzyme-linked",
"immunoassays",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"body",
"fluids",
"pathogens",
"immunology",
"sociology",
"microbiology",
"social",
"sciences",
"vertebrates",
"animals",
"mammals",
"viruses",
"social",
"systems",
"primates",
"filoviruses",
"rna",
"viruses",
"antibodies",
"immunologic",
"techniques",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"serology",
"proteins",
"medical",
"microbiology",
"immunoassays",
"microbial",
"pathogens",
"blood",
"plasma",
"biochemistry",
"eukaryota",
"blood",
"ebola",
"virus",
"anatomy",
"physiology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"amniotes",
"hemorrhagic",
"fever",
"viruses",
"organisms"
] |
2019
|
Serological analysis of Ebola virus survivors and close contacts in Sierra Leone: A cross-sectional study
|
We develop a new powerful method to reproduce in silico single-molecule manipulation experiments . We demonstrate that flexible polymers such as DNA can be simulated using rigid body dynamics thanks to an original implementation of Langevin dynamics in an open source library called Open Dynamics Engine . We moreover implement a global thermostat which accelerates the simulation sampling by two orders of magnitude . We reproduce force-extension as well as rotation-extension curves of reference experimental studies . Finally , we extend the model to simulations where the control parameter is no longer the torsional strain but instead the torque , and predict the expected behavior for this case which is particularly challenging theoretically and experimentally .
The mechanical and topological properties of DNA and protein-DNA assemblies are of primary importance in many biological processes , including transcription , replication , chromatin organization and remodeling . Since techniques have become available enabling the manipulation of single-molecules [1] , [2] , a large amount of experimental data have been accumulated on the mechanical response of DNA and protein-DNA assemblies under stretching forces and twisting torsions , in particular from optical and magnetic tweezers experiments [2]–[5] . In magnetic tweezers experiments , a DNA molecule is grafted at one end to a coverslip and at the other end to a magnetic bead . The bead is trapped in the magnetic field of a pair of magnets that may be translated , thus exerting a varying force on the bead . Moreover the pair of magnets may be rotated at a certain number of turns , thus constraining the linking number of the DNA molecule . After the stretching force and the number of turns are applied to the bead , the only physical variable that can be directly measured is the DNA extension , i . e . the distance between its two ends . Therefore the interpretation of the experimental results requires an important modeling effort , particularly in the more complex cases where DNA is associated with proteins , as for instance in chromatin assemblies [6] , [7] . Although theoretical approaches may be successful in some cases [8]–[10] , simulations are often crucial tests of the proposed model validity , when they are not the unique possible way of dealing with the system complexity . In this spirit , we aim to develop an efficient tool to manipulate single-molecules in silico reproducing optical and magnetic tweezers experiments . This task is challenging since the DNA model should have precise specifications to reproduce the behavior of DNA accurately . We need to: ( i ) model a polymer , i . e . an articulated chain; ( ii ) reproduce the effective diameter of DNA ( depending on electrostatic conditions ) and , when proteins are present , have the possibility to model their shape and steric hindrance; ( iii ) deal with collisions , especially in order to reproduce DNA supercoiled structures ( plectonemes ) and steric effects in DNA-protein assemblies; ( iv ) reproduce DNA twisting and bending elasticities; ( v ) include statistical mechanics features to account for temperature and thermal motion . Beside these essential points , we also wish to simulate the system dynamics , which may be important in some cases , e . g . when hysteresis is observed under magnetic tweezers [7] or for in vivo chromosome dynamics experiments in the cell nucleus [11] , [12] . This ambitious list of specifications is beyond the reach of traditional simulation approaches where particles interact through 2-body potentials ( as in Molecular Dynamics or Monte Carlo simulations [13] , [14] with a given force field ) . The need to deal with frozen degrees of freedom in coarse grained modeling may be addressed through holonomic constraints , as in the SHAKE algorithm [15] , [16] , where an iterative approach is adopted . However , collision detection and steric hindrance may only be accounted for in this scheme by introducing additional steps . More recently , non iterative algorithms have been developed [17] , that subsequently led to the development of new powerful tools , called “physics engines” . These have been designed by the engineering and robotics communities to reproduce the dynamic behaviour of articulated systems of rigid bodies . Physics engines are acquiring an increasing importance , notably in the fields of computer graphics and video games , where they are now widely used to simulate rigid body motion under realistic conditions and in real-time . Open Dynamics Engine ( ODE ) is one of the most popular rigid-body dynamics open source library for robotics simulation applications [18] . As other physics engines , ODE simulates the kinematics of articulated systems by using permanent joints that impose holonomic constraints , instead of bond potentials . The same method is used to manage collisions: when overlapping between bodies is detected , a temporary joint is locally created that reproduces the action of the contact forces , without the need for explicit permanent 2-body interaction potentials ( see section “Materials and Methods” for details on how ODE manages joints and collisions ) . These extremely efficient simulators haven't , up to now , been used in statistical mechanics . Although well adapted to mechanical simulations , physics engines lack coupling to a thermal bath . The main novelty of our approach is the implementation of Langevin-Euler equation in the ODE software . Moreover we improve the simulation efficiency of this Langevin dynamics by extending the “global thermostat” algorithm designed by Bussi and Parinello in 2008 [19] to physics engines . This algorithm allows much faster yet unbiased sampling of the phase-space . As a first step toward simulating DNA-protein assemblies , we focus here on bare DNA and show how to perform in silico single molecule manipulation of DNA .
In rigid body dynamics simulations run with ODE , the state of a system consisting of rigid bodies is described by the positions of their centres of mass , a quaternion representation of their orientations , and their linear and angular velocities and respectively . These velocities are collected in the column vector . We use the superscript T to denote the transpose of a vector or a matrix everywhere in this article . The vector then collects all linear and angular momenta , where is a block diagonal matrix whose elements are the mass matrices and inertia matrices of the N bodies ( with the identity matrix ) . The Newtonian dynamics equation then reads where the generalized force is a vector collecting forces and torques applied to the system . These forces and torques may be external , due for example to gravity or magnetic fields , or internal , as a consequence of the mechanical constraints between the rigid bodies that make up the system . Most notably , in articulated systems , as is the case of polymers , rigid bodies are connected by mechanical joints . A joint is a relationship that is enforced between two bodies so that they can have only certain positions and orientations relative to each other , and ODE provides different types of joints according to the kind of articulation that has to be implemented , e . g . ball-and-socket , hinge , slider or universal . Mathematically a joint imposes some holonomic constraint between both connected bodies . Such a constraint is an equation that reads where is the distance between both joint bearings , e . g . the center of the ball of one body and the the center of the socket of the other one . The constrained distance is purely geometrical , depending only on the relative position and orientation of both jointed bodies . The position and orientation of each of the bodies the articulated system is composed of depend on time . Therefore the constraints of the articulated system can be derived with respect to time to get the kinematic constraints in the form where we introduce the jacobian matrix of constraints ( see subsection “Exact solution for when there are no collisions” for a detailed example ) . This velocity-based description is used in ODE as in most game/physics engines . So , mechanical joints exert reaction forces and torques on the joint bearings . These internal mechanical constraints can be collected into a generalized constraint force which , by virtue of the principle of virtual work , reads where is a vector of Lagrange multipliers that precisely accounts for the reaction forces and torques coming from the joint bearings [16] . The Newtonian dynamics equation therefore reads where and stand for the external and internal contributions to the generalized force respectively . As the constraint force reads , the Newtonian dynamics equation becomes an equation for in the form: . Solving this equation for should moreover satisfy the holonomic constraints at every timestep . However the discretization used in the numerical calculation results in errors on so that is generally not equal to 0 . Then , in order to have at the next timestep , the kinematic constraint should be adapted accordingly . Indeed according to the Euler semi-implicit integration scheme which is used in velocity-based algorithms . Hence . But then this implies that the kinematic constraint is not equal to zero at time , i . e . , so that the joint bearings will continue to move apart afterwards . In order to keep both and close to zero at every timestep , ODE introduces an error reduction parameter in the kinematic equation [18] . This parameter has to be adjusted to some optimal value between 0 ( no correction at all ) and 1 ( complete correction of in one timestep ) . However setting is not recommended since , as said above , this would imply that the joint bearings will continue to move apart afterwards with maximal velocities . ODE recommends values between and . In addition to , ODE introduces a second ingredient to soften the rigid constraints by allowing the violation of the constraint equation by an amount proportional to the restoring force . More explicitly , a “constraint force mixing” diagonal matrix is defined , such that ( implicit integration ) [18] . This is equivalent to introducing a spring-damper system ( spring constant of and damping constant of ) with implicit integrator between the joint bearings; this can be understood as analogous to a bead-spring model . Nevertheless there is a major difference between this effective spring and a regular spring: the term constrains the velocity whereas a regular spring constrains the acceleration . As a result , no energy is stored in this effective spring , at odds with a regular spring which stores an averaged energy ( see below subsection “Preliminary tests of validity and performance of the global thermostat” along with the histograms of energy in Figure 1 ) . In particular , ODE uses a powerful software called libccd [20] to detect collisions between two convex shapes . Whenever overlapping is detected between two rigid bodies , ODE attaches a temporary joint between them called a “contact joint” . Defining vector ( resp . ) that connects the center of mass of body 1 ( resp . 2 ) to the contact point and denoting the common normal to both bodies at the contact point ( directed from 2 to 1 ) , the kinematic constraint imposed by the contact joint would read in the perfect case when the holonomic constraint imposed by the contact joint reads exactly . However , in practice is not equal to 0 because of discretization errors , hence the kinematic constraint imposed by the contact joint actually reads: ( 1 ) with ( 2 ) ( 3 ) The right hand side of Eq . ( 1 ) deals with the already existing overlapping of the two bodies in contact at time when collision is first detected , or with their residual overlapping while the contact joint exists . By inserting the constraint force into the equation of motion and taking the first-order discretisation of this equation , one can easily get the following expression to be solved for : ( 4 ) This equation is of the form . Importantly , the addition of the term to each diagonal term of the matrix provides a symmetric positive definite matrix , thus greatly increasing the solution accuracy of Eq . ( 4 ) . From this equation , the vector of Lagrange multipliers , hence the constraint force , can be determined . Then the motion solver ( semi-implicit Euler integrator ) gives the new positions and orientations of the articulated bodies at time . It is advantageous to choose the Exponential Map parametrization [21] for the quaternion integration . In general Eq . ( 4 ) has to be solved numerically and ODE has two algorithms to do so , one based on the Successive-Over-Relaxation ( SOR ) method [22] and the other based on the Linear Complementary Problem ( LCP ) [23] . LCP time complexity is of order and space complexity ( memory ) of order where is the number of constraint rows [18]; whereas SOR time complexity is of order where is the number of successive-over-relaxation and space complexity of order [18] . Both algorithms have equivalent performances when . But in general LCP is more accurate , although much more time consuming , than SOR . We compared these two algorithms for a chain of length without noticing significant differences in the errors on ( error on the colocalization of joint bearings ) . In order to save computational time , we preferentially run the SOR method with a value of for the relaxation factor and . These values are different from the default values in ODE and work well for a linear chain of rigid bodies connected with ball-and-socket joints . But in some cases , the SOR method does not converge and we then switch to the LCP method , which always converges . However in the case when there are no collisions between the rigid bodies the articulated system is composed of , we were able to derive an exact solution for ( see next subsection ) . Therefore we solve Eq . ( 4 ) according to the following scheme: For the sake of clarity , let us first consider the example of four rigid cylinders of length each connected with ball-and-socket joints at the extremities with the first one anchored to some fixed point , taken as the origin of the coordinates . The vector is the tangent to the cylinder . The jacobian matrix associated with this system is tridiagonal when there are no collisions , in which case it reads: ( 5 ) For each cylinder , the antisymmetric matrix is associated with the cross product and has the property : ( 6 ) The transpose Jacobian matrix and mass matrix are given by: ( 7 ) Then we get: ( 8 ) and we deduce the final result for the symmetric matrix : ( 9 ) where we define the matrix . We then write in the associated principal axis body frame as : ( 10 ) The vector collects the Lagrange multipliers associated to each joint respectively . The equation gives us a system of coupled equations on . Note that ODE solves in one time all the constraints of this articulated system . This is not the case with the SHAKE algorithm where an unconstrained step is first performed , before correcting the positions and orientations iteratively to get the constraints satisfied eventually . The term from Eq . ( 4 ) is given by: ( 11 ) where we write and . We can then write the constraint forces and torques ( 12 ) We can now generalise the previous example to the case of a linear chain of rigid cylinders connected with ball-and-socket joints with the first one anchored to the ground . We denote the matrix with the following properties: ( 13 ) ( 14 ) ( 15 ) ( 16 ) Using the following decomposition for the matrix where is a block lower matrix with block identity matrix on the diagonal and where is the block diagonal matrix it is easy to show that for . From these we get the following equations: ( 17 ) ( 18 ) ( 19 ) In order to solve the linear system of equations we define and solve the problem in an iterative way: ( 20 ) ( 21 ) and we get the final solution for by solving the problem : ( 22 ) ( 23 ) The method explained here is the exact solution of the problem where no collisions are present in the system . With this exact resolution the simulation is faster than the SOR algorithm ( and ) with a gain of . To model a DNA molecule , we build a linear chain of rigid cylinders of length each , corresponding to 10 base pairs ( ) , which amounts to the double helix pitch . We connect the cylinders to each other by ball-and-socket joints . The radius of the cylinders is set according to the salt buffer concentration of the experimental data we compare with . Indeed , since the DNA molecule is highly negatively charged , DNA-DNA electrostatic repulsion affects the double helix response in single molecule experiments [9] , [24]–[27] . This effect can be easily and implicitly included in simulations and theoretical models by introducing an effective DNA radius where is the crystallographic radius of the DNA double helix and accounts for the DNA-DNA electrostatic repulsion [28]–[31] . It turns out that may be set equal to the Debye length of the salt buffer solution . As with c the salt concentration given in and in , we set the effective radius to in mmol monovalent salt buffer for comparison with the reference experimental data of Mosconi et al [32] . Alternatively we use an effective radius to fit the experimental data obtained in mmol monovalent salt buffer by Smith et al [1] . We performed all our in silico single molecule experiments with a DNA molecule of contour length . The corresponding number of DNA cylinders in the chain is therefore . The DNA molecule is anchored , at one end , to a planar surface ( mimicking the microscope coverslip ) , and at the other end , to a rotatable bead ( mimicking the magnetic bead ) . We set the bead radius to in order to prevent the DNA from looping around it . At both ends of the DNA chain , the rigid cylinders are tangent to their attachment surface . The final problem that remains to be addressed is how to obtain a correct definition of the bending and twisting behavior of DNA . We have solved this problem by a special choice of the connecting joints and by introducing appropriate restoring torques reacting to the bending and twisting deformations . This has been done based on the bending and twisting energies that are defined according to the usual expressions and respectively . The rigidity constants and are related to the bending and twisting persistence lengths and respectively , through the following equations: ( 24 ) ( 25 ) ( 26 ) where is the Langevin function ( see supplementary Text S1 ) . The bending angle and twisting angle are related to the standard Euler transformation ZXZ and are given by ( 27 ) ( 28 ) ( 29 ) where is the tangent vector of cylinder i , a vector normal to and . These three vectors are the principal axis of cylinder i . We finally get the following expression for the global restoring torque between two connected DNA segments ( see supplementary Text S1 for the complete derivation of this equation ) : ( 30 ) We recall that , for DNA , estimates of the bending persistence length give for salt buffer ( see Refs . [1] , [33]–[35] ) ; whereas estimates of the twisting persistence length give for salt buffer [9] , [32] . According to the size of the unit cylinder we find and . Although well adapted to mechanical simulations , ODE lacks coupling to a thermal bath . As physics engines impose to deal with dynamics equations including inertial terms , in particular for computing constraint forces ( collected in ) , we need to turn to some implementation of stochastic isothermal molecular dynamics in order to thermalize the system: isothermal to simulate the system at constant temperature , stochastic to ensure ergodicity . The corresponding algorithms are all related to Langevin dynamics and can be cast into local and global thermostats . In local thermostats , such as standard Langevin dynamics , a correction force including both a frictional term and a stochastic term is exerted on each particle to drive the system to the canonical distribution at a prescribed temperature . Global schemes of Langevin dynamics are designed to minimize the perturbation introduced by the thermostat on the Hamiltonian trajectory ( so called “disturbance” as defined originally in [36] ) , hence on the dynamical properties , such as autocorrelation functions , and related quantities , such as diffusion coefficients . In these globally applied thermostats the stochastic term of the correction force acting on each particle is proportional to the momentum of that particle . Two main global algorithms have been designed so far: ( a ) Stochastic Velocity Rescaling methods , most notably the “global thermostat” introduced by Bussi and Parrinello [19] , ( b ) the Nosé-Hoover Langevin thermostat [37] . Here we first show how to implement Langevin-Euler equation in the ODE software . Moreover we show that the global thermostat introduced by Bussi and Parrinello is so remarkably adapted to this implementation that it improves quite significantly the sampling efficiency with respect to local Langevin dynamics ( by two orders of magnitude in typical situations ) , while preserving the time-dependent properties such as autocorrelation functions . The sampling efficiency is defined as usual as the number of independent configurations generated during the time necessary to reach thermal equilibrium . To begin with , we add to the “mechanical” forces an additional , thermal contribution containing a frictional term and a random force vector . is the matrix of the coupling frequencies to the thermostat , the matrix of white noise amplitudes and a generalized vector of normalized and independent Wiener processes . and are related through the fluctuation-dissipation theorem , which reads here ( 31 ) where with the temperature of the thermal bath and where the superscript denotes that the matrices and are chosen to be diagonal in the principal axis body frame ( where the matrix is diagonal by definition ) . For simplicity , we choose to fix all the to a common frequency . Note that is the relaxation time of the thermostat , i . e . the autocorrelation time of the kinetic energy ( see supplementary Figure S1 ) . We then improve the sampling efficiency of this Langevin dynamics by extending the “global thermostat” algorithm designed by Bussi and Parinello in 2008 [10] to physics engines . This algorithm allows faster yet correct sampling of the phase space in the canonical ensemble . However , it is designed for the translational degrees of freedom only . In order to apply it to an articulated rigid body system , we therefore have to extend it to the rotational degrees of freedom and adapt it to the ODE software . To this aim , we replace the traditional Langevin-Euler correction force ( local thermostat ) by a corresponding global version , which reads ( 32 ) Eq . ( 32 ) shows that is proportional to , so that the stochastic force and torque globally associated with the thermostat is in the same direction as . Hence , a free particle , i . e . a particle not connected to any other particle , will move on a straight line between two collisions . Note that , nevertheless , the particle will undergo Brownian motion along this straight trajectory . The global version of the Langevin dynamics minimizes the disturbance induced by the thermostat on the Hamiltonian trajectory ( equal to according to its definition in Ref . [19] , but extended here to the rotational degrees of freedom ) , nevertheless retaining the same thermalization speed as usual Langevin dynamics ( see supplementary Figure S1 ) . When used in the framework of a velocity-based algorithm such as ODE , the global thermostat presents a remarkable advantage . This is because , in this case the global Langevin contribution is decoupled from the constraint forces , in the sense that it cancels out in the equations for . More precisely , with our definition of ( see Eq . ( 32 ) ) , the contribution to the term in Eq . ( 4 ) is always zero . In other words , not only minimizes the disturbance of the Hamiltonian trajectory , but also does not disturb the generalized constraint force . Both effects cooperate to achieve a dramatic acceleration of the simulation sampling , that is , in the case of our model , approximately times faster than with the local thermostat ( see below “Preliminary tests of validity and performance of the global thermostat” and Figure 2 ) . Importantly this acceleration is compatible with the correct computation of dynamical properties , such as autocorrelation functions . In all DNA simulations presented in this article , we choose the length of the cylinders as the unit length , the mass of the cylinders as the unit mass and the unit of thermal agitation as the unit of energy , from which we deduce the unit of time . The complete set of parameters of our simulations is given in Table 1 . We also choose to deal collisions with a restitution coefficient equal to without surface friction . Hence , when two rigid bodies collide , the constraint force imposed by the contact joint that temporarily connects them is directed along their common normal at the contact point . We finally choose an error reduction parameter and a constraint force mixing parameter .
We start to validate our methodology by simulating a DNA molecule without any constraint applied on the bead ( neither stretching nor twisting ) . To this aim , we first check the equipartition theorem . When the system is at thermal equilibrium , its temperature is related to the kinetic energy through the equation ( 33 ) where is the number of non-redundant holonomic constraints . This relation is standard since is just the number of degrees of freedom ( dof ) of the system . Moreover the distribution of the kinetic energy of the system at thermal equilibrium follows a Boltzmann law and therefore reads: ( 34 ) with . We checked this relation for a DNA molecule of length µm coupled to the two different Langevin thermostats , local and global respectively . The resulting histograms are shown in Figure 1 , confirming that: ( i ) the kinetic energy is correctly sampled at thermal equilibrium with both thermostats , ( ii ) there is indeed no energy stored in the joints , although these have been softened by effective springs ( see above the error reduction parameter in subsection “Introduction to physics engines” ) . We then quantified the simulation sampling efficiency by means of the autocorrelation function of the end-to-end distance of a DNA molecule of length µm coupled to the two different Langevin thermostats , local and global respectively . Here the average is performed over the time and denotes the lag-time of the autocorrelation function . A demonstration of the performance of the global thermostat in terms of relaxation rapidity is given in Figure 2 . Fitting the exponential decrease of both relaxation curves shows that with use of the global thermostat we reach the saturation value at whereas this value is reached at in the case of the local thermostat , thus resulting in an acceleration factor of about for this system composed of articulated rigid bodies . Note that the dimensionless lag-time is equal to ( see Table 1 ) , with the corresponding number of time steps . Then a typical run using the global thermostat is of the order of tens of millions of time steps , whereas it is of the order of billions of time steps with usual Langevin dynamics . A striking illustration of the sampling acceleration provided by the global thermostat is also given in supplementary Video S1 . We also compute the tangent-tangent correlation function along the polymer with both local and global thermostats . No significant deviations were found between both thermostats . Results obtained with the global thermostat are plotted in supplementary Figure S2 along with the corresponding theoretical curves . A simple calculation shows that the tangent-tangent correlation function decreases as , from which one can calculate the average bending . With the DNA persistence length , this quantity amounts to , to be compared to the result from a fit of the simulation curves , giving . The same comparison can be done for the twist angle ( with ) , for which the simulation average matches the theoretical value . These comparisons show that our simulation results are in very good agreement with the analytical formulae , thus validating ( i ) our implementation of the bending and twisting rigidities and ( ii ) the correct sampling of the DNA conformation space by means of the global Langevin thermostat . We then simulate reference force-extension curves , both theoretical and experimental . We thus perform simulations at given stretching force along the z-axis ( normal to the DNA anchor surface ) , and without torsional constraints on the magnetic bead . In order to fit the experimental data obtained by Smith et al [1] in monovalent salt buffer , we set here the DNA radius ( i . e . the radius of the unit cylinders ) to . The resulting force-extension curve is given in Figure 3 where we plot ( red circles ) the dimensionless stretching force as a function of the dimensionless mean relative extension . Here denotes the mean DNA extension , i . e . the mean distance between the bead and the anchor surface , at zero torsional constraint . For comparison , we also plot ( black solid line ) the analytical Worm-Like-Chain ( WLC ) interpolation fitting curve proposed by Bouchiat et al [35] as well as the numerical solution of the WLC model ( black triangles ) obtained by Marko and Siggia with the same persistence length [34] . The simulation reproduces pretty well the WLC behavior , thus validating our implementation of the DNA bending rigidity . Note that , at low forces , the extension saturates at a value greater than zero because of the impenetrable ground and magnetic bead that both confine the DNA molecule . This effect is more pronounced than in the experimental curve [1] because the ratio of the bead radius to the DNA length is higher in our simulations . In Figure 3 , we also show the results obtained in the limit case , for which ( see Eqs . ( 24–25 ) ) , and when there are no collisions . In this case we expect to observe a Freely-Jointed-Chain response ( with segments of each ) . The analytical force-extension relation for FJC is given by the well-known expression as a Langevin function and it is also reproduced in Figure 3 . Again , the simulation results are in very good agreement with the theoretical formula . More interestingly , magnetic tweezers also allow the application of a torsional strain on a single DNA molecule at constant stretching force . This torsional strain is equal to the number of turns of the magnetic bead around the z-axis due to the rotation of the magnets . The number of turns of the bead is also equal to , the variation of the linking number of the DNA double helix with respect to the intrinsic twist of the DNA double helix with the pitch of the DNA . And we define as usual the DNA relative overtwist as . The method developed here can conveniently describe all physiological situations involving DNA positive supercoiling , which are of main importance for DNA transcription or replication . One limitation of the modeling developed so far is that extensive modifications of the double helix structure are not accounted for , e . g . base pair opening that occurs when negative supercoiling is applied ( DNA denaturation ) , or the S-DNA transition observed at extremely high force , or the P-DNA transition under very positive torque . Nevertheless this methodology gives an unparalleled opportunity to study more complex biological systems , such as protein-DNA complexes: in particular , we are currently addressing the modeling of magnetic tweezers response of chromatin fibres [7] , by associating our in silico DNA model with a rigid body representation of the histone octamer . More recently , in vivo experiments also enable the measurement of the dynamics of chromosome loci in the cell nucleus [11] , [12] , [46] .
|
Video game techniques are designed to simulate rigid body dynamics of macroscopic bodies , e . g . characters or vehicles , in a realistic manner . However they are not able to deal with temperature effects , hence they are not able to deal with molecules . In order to extend these powerful techniques to molecular modeling , we implement here Langevin Dynamics in an open source library called Open Dynamics Engine . Moreover we add a “global thermostat” to this Langevin Dynamics , which accelerates the simulation sampling by two orders of magnitude . With these radically new simulation techniques , we prove that we can accurately reproduce single-molecule manipulation experiments in silico , in particular force-extension as well as rotation-extension curves of reference experimental studies . The method developed here represents an unparalleled tool for the study of more complex single molecule manipulation experiments , notably when DNA interacts with proteins . Furthermore the simulation technique that we propose here has all the functionalities required to tackle the nuclear organization of chromosomes at every length scale , from DNA to whole nuclei .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results/Discussion"
] |
[
"physics",
"computer",
"science",
"computer",
"animation",
"biophysic",
"al",
"simulations",
"computing",
"methods",
"biology",
"computational",
"biology",
"biophysics",
"simulations",
"biophysics"
] |
2014
|
In Silico Single-Molecule Manipulation of DNA with Rigid Body Dynamics
|
Trypanosoma brucei rhodesiense ( Tbr ) and T . b . gambiense ( Tbg ) , causative agents of Human African Trypanosomiasis ( sleeping sickness ) in Africa , have evolved alternative mechanisms of resisting the activity of trypanosome lytic factors ( TLFs ) , components of innate immunity in human serum that protect against infection by other African trypanosomes . In Tbr , lytic activity is suppressed by the Tbr-specific serum-resistance associated ( SRA ) protein . The mechanism in Tbg is less well understood but has been hypothesized to involve altered activity and expression of haptoglobin haemoglobin receptor ( HpHbR ) . HpHbR has been shown to facilitate internalization of TLF-1 in T . b . brucei ( Tbb ) , a member of the T . brucei species complex that is susceptible to human serum . By evaluating the genetic variability of HpHbR in a comprehensive geographical and taxonomic context , we show that a single substitution that replaces leucine with serine at position 210 is conserved in the most widespread form of Tbg ( Tbg group 1 ) and not found in related taxa , which are either human serum susceptible ( Tbb ) or known to resist lysis via an alternative mechanism ( Tbr and Tbg group 2 ) . We hypothesize that this single substitution contributes to reduced uptake of TLF and thus may play a key role in conferring serum resistance to Tbg group 1 . In contrast , similarity in HpHbR sequence among isolates of Tbg group 2 and Tbb/Tbr provides further evidence that human serum resistance in Tbg group 2 is likely independent of HpHbR function .
Trypanosomiasis , a deadly disease of humans and livestock in sub-Saharan Africa , is caused by protozoan parasites of the genus Trypanosoma , which are transmitted between mammalian hosts by insect vectors of the genus Glossina ( tsetse ) . Human-infective members of the Trypanosoma brucei complex cause the human form of the disease , Human African Trypanosomiasis ( HAT ) , or sleeping sickness . T . b . rhodesiense ( Tbr ) causes an acute form of HAT in eastern Africa , while T . b . gambiense group 1 ( Tbg1 ) causes a chronic form of the disease in western and central Africa and accounts for over 90% of reported cases ( Figure 1a ) . T . b . gambiense group 2 ( Tbg2 ) , a rare form described from West Africa in the 1970s and 1980s , also causes human disease but the trait of human-infectivity is not stable [1] , [2] , [3] . The final member of the brucei complex , T . b . brucei ( Tbb ) , is not infective to humans , but , together with other animal trypanosome species , causes the livestock wasting disease , Nagana , across a range that overlaps with that of the human-infective parasites . Humans possess an innate resistance to some trypanosomes through the action of trypanosome lytic factors ( TLFs ) in their serum [4] . TLF-1 is a high-density lipoprotein complex that includes the active toxin apolipoprotein L-I ( apoL-I ) in association with haptoglobin-related protein ( Hpr ) . In the primary immune pathway [5] , [6] , [7] , TLF-1 is bound and internalized via a haptoglobin haemoglobin receptor ( HpHbR ) on the surface of susceptible trypanosomes . Uptake of TLF-1 is followed by disruption of the lysosomal membrane by apoL-I and eventual cell lysis . While Tbb is susceptible to lysis by human TLF-1 , Tbr , Tbg1 and Tbg2 are resistant . In Tbr , the serum-resistance associated ( SRA ) protein confers resistance to TLF-1 [8] by binding directly to apoL-I after it has been internalized into the cell , inhibiting its lysosome-lytic capacity [9] . Tbg1 and Tbg2 , on the other hand , lack the gene encoding SRA and are thought to have evolved an independent mechanism to prevent lysis by TLF . In Tbg2 , apoL-I is also internalized , but lysis is prevented by an unidentified mechanism [10] . In Tbg1 , the mechanism is better understood and appears to involve reduced expression and altered function of the parasite HpHbR [11] . Sequencing of a few isolates of Tbg and Tbb led Kieft et al . [11] to suggest that mutations in HpHbR may have altered TLF-1 binding in Tbg1 . Specifically , the authors identified five non-synonymous substitutions shared by the four sequenced isolates of Tbg1 , but not present in two Tbb isolates . This work has helped to narrow the universe of possible structural differences in HpHbR that could , for example , eventually be exploited to design novel drugs to overcome Tbg1 resistance . However , the small number of isolates examined to date is not sufficient to determine whether the mutations are really Tbg1-specific . While genetic variation in Tbg1 is extremely limited [12] , [13] , the remainder of the T . brucei complex exhibits relatively high variation , most of which does not partition into neatly defined geographic or taxonomic units [14] , [15] , [16] , [17] . Thus , characterizing the genetic differences that contribute to a critical epidemiological trait such as human infectivity requires that those differences be evaluated in a comprehensive geographical and taxonomic context . In the present study , we tested if the five non-synonymous substitutions previously hypothesized to alter HpHbR activity in Tbg1 [11] are both conserved in Tbg1 isolates and also absent from other T . brucei subspecies by examining HpHbR gene variation in T . brucei s . l . sampled across the entire range of the species complex . By narrowing the pool of substitutions that are specific to Tbg1 , we expect to facilitate future functional studies aimed at understanding the contribution of HpHbR to conferring human serum resistance .
Isolates of Tbb , Tbg1 , Tbg2 and Tbr , were selected to incorporate representative genetic diversity from the entire geographic range of the T . brucei complex ( Table S1 , Figure 1b ) . When available , we included isolates of all co-occurring taxa from each country sampled ( Figure 1b ) . For each isolate , DNA was extracted as described in [17] . PCR was performed using primers designed from Tbb ( TREU927 ) and Tbg1 ( Dal972 ) TriTrypDB database sequences ( Tb927 . 6 . 440 and Tbg972 . 6 . 120 , respectively ) to amplify a 1297 base pair ( bp ) fragment that encompassed the entire HpHbR gene ( HpHbR_F 5′ CGGGAAAGTTGTACGCAAG , HpHbR_R2 5′ CGACCACTTAATGTTACGAGG ) . For each PCR , 2–4 µL of a 1∶10 dilution of DNA extract were used . PCR reactions were performed using the reagents provided with GoTaq® DNA Polymerase and Green Master Mix . Difficult templates were amplified using Failsafe PCR 2X PreMixes Buffers ( Epicentre Biotechnologies , Madison WI ) . All PCR reactions used the following cycle: Initial denaturation 95°C for 2 min , 50 cycles of 95°C for 35 s , 58°C for 35 s , and 72°C for 1 min 20 s and a final extension at 72°C for 7 min . PCR success was verified with 1% agarose gel electrophoresis . PCR products were purified and then sequenced ( Yale DNA Analysis Facility ) using two internal primers located in the middle of the sequence ( HpHbR_F2in 5′ TGCTCGAGATATTCCTCAAG , HpHbR_Rin 5′ CTCCCACTGAAGCATTAGAC ) . The sequenced fragment included 22 nucleotides upstream of the HpHbR start codon , the entire HpHbR gene and 62 nucleotides downstream of the HpHbR stop codon . Sequences generated using the internal primers overlapped by approximately 200 bp permitting the assembly of an entire contiguous sequence of the HpHbR gene . Contiguous sequences were constructed and chromatograms from each isolate were manually examined for double peaks using the CLCBio DNA Workbench 5 . 7 ( Cambridge , MA ) . Sites with double peaks were assigned the appropriate nucleotide ambiguity code . Sequences were aligned manually using MacClade 4 . 08 [18] . Samples with double peaks were considered heterozygotes . We used the programs SeqPhase [19] to format files and PHASE 2 . 1 . 1 [20] to resolve individual alleles from heterozygous sequences . To assess evidence for recombinant alleles and to relax the assumption of a stepwise mutation model , we employed the recombination model ( MR ) and the parent-independent models , respectively . Each run used 1000 iterations and a burnin of 500 generations and thinning interval = 1 . The dataset was run twice with different random starting seeds and checked for consistency . The replicate with the best average goodness-of-fit was selected for subsequent analyses . Nucleotide DNA sequence alignments were generated from phased alleles in MacClade 4 . 08 . Haplotype networks were constructed in the program TCS [21] . DNA sequences were translated to amino acids and aligned in MacClade 4 . 08 . Non-coding regions were removed from sequences and amino acid sequences were compared to those generated by [11] .
We collected 1296 bp of sequence from each of 65 T . brucei isolates: 32 from Tbb , 15 from Tbg1 , five from Tbg2 and 13 from Tbr . In addition , we generated sequence for one isolate each of Trypanosoma equiperdum and Trypanosoma evansi ( Table S1 ) , both of which are also members of the subgenus Trypanozoon but are not human infective ( reviewed in [3] ) . Of the 67 isolates sequenced in this study , 30 were heterozygous at the locus sequenced . PHASE 2 . 1 . 1 inferred a total of 34 alleles present in the 67 isolates . For all heterozygotes , allele pairs had Bayesian posterior probabilities of 1 . 0 across replicate runs , indicating that no alternative allele sequences could be inferred from the heterozygotes . The 34 alleles recovered in this study exhibited a total of 40 variable sites , of which four were located outside the HpHbR coding region . Each of these four sites occurred in a distinct allele ( f2 , c , u3 , z1 ) across a total of five isolates ( Boula ( Tbg1 ) , STIB338 ( Tbr ) , STIB386 ( Tbg2 ) , STIB777AE ( Tbb ) , and KP13 ( Tbb ) ) . The remaining 36 variable sites were found within the coding region of HpHbR ( File S1 ) . Most allelic diversity ( 28 alleles ) was found in isolates of Tbb , Tbr and Tbg2 and much of this diversity was common to two or more taxa . Five of the seven alleles recovered from Tbr were identical to those found in Tbb . Likewise , four of the five alleles recovered from Tbg2 were also identical to alleles in Tbb . The most common allele in this study ( u1; Table S1 ) was recovered from Tbb , Tbr and Tbg2 . In contrast to these observations , we recovered four distinct alleles from Tbg1 , but none of these were shared with any member of the subgenus Trypanozoon . Allelic diversity in Tbg1 was relatively low . Allele z1 ( identical to the Tbg1 sequence reported in Kieft et al . [2010] ) was the most common Tbg1 variant and was recovered from 26 of 30 sampled chromosomes . The remaining Tbg1 alleles differed by only one nucleotide from this common variant , z1 . Trypanosoma equiperdum sequences were more similar to Tbb and Tbr sequences , though both alleles from T . equiperdum were unique . In T . evansi , alleles were identical to the most common allele found in Tbb , Tbr and Tbg2 ( Fig . 1b ) . The HpHbR protein consists of 403 amino acids . In silico translation of the DNA sequences of the 34 alleles described above yielded 25 unique protein sequences ( Figure 1 , Figure S1 ) . Notably , the single nucleotide difference in HpHbR that distinguished all isolates of Tbg1 from all other T . brucei isolates sampled in this study was non-synonymous , resulting in the replacement of leucine with serine at position 210 ( L210S; Figure 1b , Figure S1 ) . With one exception , all Tbg1 isolates possessed two copies of HpHbR that coded for just this single amino acid sequence ( Z ) . In the exception , isolate ITMAP020578 , one allele coded for amino acid sequence Z and the second allele coded for a second peptide ( Y ) differing from Z by just one amino acid at position 212 . All other variation in HpHbR amino acid sequences partitioned to differences within and among Tbb , Tbr and Tbg2 .
The primary goal of this study was to examine the genetic diversity of HpHbR in a broad geographical and taxonomic context to better characterize the mutations that potentially give rise to differences in HpHbR function and that may contribute to the phenotype of human serum resistance observed in Tbg1 . An earlier study of HpHbR genetic diversity in a limited sample of parasite isolates identified five non-synonymous substitutions shared by Tbg1 , but not found in Tbb , suggesting that these five differences could play an important functional role [11] . By sampling more broadly across the subgenus Trypanozoon and across Africa , we have demonstrated that just one of these substitutions ( L210S ) is conserved in Tbg1 and also absent from the most closely related trypanosome taxa , all of which are either susceptible to human serum ( Tbb ) or known to possess an alternative resistance mechanism ( Tbr or Tbg2 ) . Although our sample size remains relatively limited compared to the vast number of parasites distributed widely across Africa , the extremely low genetic diversity observed in Tbg1 HpHbR is consistent with prior population genetic studies [12] , [13] , [17] and we hypothesize that the mutation L210S is likely fixed in the taxon . This could be extended to field-circulating Tbg1 by using either allele specific PCR primers or a restriction fragment length polymorphism ( RFLP ) that targets the single nucleotide substitution ( e . g . , enzyme PleI ) . To the extent that the unique substitution in Tbg1 HpHbR prevents the uptake of TLF-1 , this single amino acid change may play a key role in conferring serum resistance to this parasite . A role for HpHbR in facilitating lytic activity of human serum was originally established by experiments demonstrating that loss of HpHbR in Tbb ( through RNA interference or gene knockout ) conferred resistance to TLF-mediated lysis [22] . Later work demonstrated that Tbb selected to be TLF-1-resistant exhibited reduced HpHbR expression . Furthermore , the ectopic expression of Tbg1 HpHbR ( using an allele identical to the most common Tbg1 allele identified in our study ) in these serum resistant Tbb was not sufficient to restore human serum susceptibility , providing evidence for the altered function of Tbg1 HpHbR [11] . Our data indicate that this altered function likely stems from the L210S mutation in Tbg1 , a substitution that effects an approximate 20-fold reduction in the affinity of HpHbR for HpHb [23] . Given that L210S appears to be the single mutation that distinguishes Tbg1 HpHbR from the HpHbR of all closely related members of the Trypanozoon subgenus , we hypothesize that this single mutation could play a major role in the serum resistance of Tbg1 . However , this mutation is unlikely to be the sole factor . As noted previously , reduced expression levels of HpHbR are also likely to play a role in Tbg1 serum resistance [10] , [11] . Also , while HpHbR is likely to be the main route of entry into the cell for TLF-1 , poorly characterized alternative routes appear to exist for both TLF-1 and TLF-2 , a second HDL particle that also exhibits trypanolytic activity [6] . Finally , an in vitro study has demonstrated that , regardless of receptor function , Tbg1 may be inherently resistant to apoL-1 , the active trypanolytic factor in human serum [10] . While HpHbR may only be one component of Tbg1 serum resistance , the possible benefit of designing new drugs targeted to this receptor variant warrants further functional study to fully circumscribe its effect on serum resistance . In contrast to Tbg1 , the mechanism of Tbg2 resistance to human serum is thought to be independent of HpHbR , based on the finding that HpHbR from Tbg2 internalizes TLF-1 at a rate similar to that observed in Tbb and Tbr [10] . While that study included just a single strain of Tbg2 ( STIB386 ) , our results , which include data for several additional strains , suggest that this conclusion is likely to hold more broadly in Tbg2 . Sequencing of HpHbR indicated that several isolates of Tbg2 shared sequence identity with isolates of both Tbb and Tbr , while exhibiting no overlap with isolates of Tbg1 , a result that is consistent with previous surveys of neutral genetic markers [13] , [17] . The genetic similarity of HpHbR observed among a large collection of isolates of Tbb , Tbr , and Tbg2 suggests that the function of HpHbR in Tbg2 is more likely to reflect that of Tbb and Tbr than Tbg1 and further supports the conclusion that Tbg2 serum resistance is independent of HpHbR . Our study surveyed only five strains of Tbg2 , but even these five strains exhibited substantially more diversity than Tbg1 at both the nucleotide and amino acid level . The genetic variability of HpHbR in Tbg2 reiterates the fact that Tbg2 , unlike Tbg1 , is not genetically homogeneous and suggests that future studies should consider this diversity when examining functional differences among parasite subgroups .
|
Human African Trypanosomiasis , or sleeping sickness , is caused by two different parasites: Trypanosoma brucei gambiense ( Tbg ) and T . b . rhodesiense ( Tbr ) . Each parasite employs a different mechanism to resist trypanosome lytic factor ( TLF ) , the active innate immune component of human serum . In Tbg group 1 , which causes the vast majority of disease cases , the mechanism is thought to involve the reduced activity of a receptor involved in binding and internalizing TLF . In this study , we investigate genetic variation in this receptor across a broad geographic sample of Tbg and closely related trypanosomes to test whether unique polymorphisms in the receptor from Tbg may explain its altered function . We identified a single mutation in all copies of the receptor gene sequenced from Tbg but not in any other closely related species . This finding suggests that this single mutation could play a key role in conferring human infectivity to Tbg . Given the possible consequences for drug development and diagnostics , we suggest that future functional studies target this mutation to fully elucidate its role .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"public",
"health",
"and",
"epidemiology",
"haplotypes",
"genetic",
"mutation",
"genetic",
"polymorphism",
"epidemiology",
"genetics",
"population",
"genetics",
"biology",
"population",
"biology",
"genetics",
"and",
"genomics"
] |
2012
|
Trypanosoma brucei gambiense Group 1 Is Distinguished by a Unique Amino Acid Substitution in the HpHb Receptor Implicated in Human Serum Resistance
|
In women , oocytes arrest development at the end of prophase of meiosis I and remain quiescent for years . Over time , the quality and quantity of these oocytes decreases , resulting in fewer pregnancies and an increased occurrence of birth defects . We used the nematode Caenorhabditis elegans to study how oocyte quality is regulated during aging . To assay quality , we determine the fraction of oocytes that produce viable eggs after fertilization . Our results show that oocyte quality declines in aging nematodes , as in humans . This decline affects oocytes arrested in late prophase , waiting for a signal to mature , and also oocytes that develop later in life . Furthermore , mutations that block all cell deaths result in a severe , early decline in oocyte quality , and this effect increases with age . However , mutations that block only somatic cell deaths or DNA-damage–induced deaths do not lower oocyte quality . Two lines of evidence imply that most developmentally programmed germ cell deaths promote the proper allocation of resources among oocytes , rather than eliminate oocytes with damaged chromosomes . First , oocyte quality is lowered by mutations that do not prevent germ cell deaths but do block the engulfment and recycling of cell corpses . Second , the decrease in quality caused by apoptosis mutants is mirrored by a decrease in the size of many mature oocytes . We conclude that competition for resources is a serious problem in aging germ lines , and that apoptosis helps alleviate this problem .
As women age , the quality and quantity of their oocytes decline , resulting in a decreased chance of becoming pregnant and an increased chance of having a child with birth defects [1] , [2] . A major cause of this decline is the increasing fraction of oocytes with chromosomal abnormalities , such as those that cause Down's syndrome . These abnormalities are caused , at least in part , by defects in recombination and chromosome cohesion during meiosis [3] , [4] . In theory , the accumulation of other types of mutations , a decreased ability to eliminate defective oocytes , or fewer resources to nurture developing oocytes might also contribute to the decline in oocyte quality . While this decline in quality is occurring , many other oocytes are undergoing apoptosis [5] . It is not known what role these apoptotic deaths play in oogenesis and the maintenance of oocyte quality . The nematode Caenorhabditis elegans is one of the leading models for studying germ cell development [6] and apoptosis [7] . The XX animals are self-fertile hermaphrodites and the XO animals are males . At 15°C , the first 60–80 germ cells in hermaphrodites develop as spermatocytes , resulting in 240–320 sperm , and subsequent germ cells develop as oocytes . In other respects , the hermaphrodites are similar to females from related species; in particular , they have female gonads , which consist of two symmetrical U-shaped tubes connected by a central uterus . The distal end of each tube contains a stem cell niche , where germ cells proliferate under the influence of the Distal Tip Cell [6] . As they move farther down the tube , germ cells enter the transition zone and begin meiosis . Soon afterward , the developing germ cells appear to pause or arrest during the pachytene stage of Prophase I . To complete pachytene , they require a signal from the Ras/MAPK pathway [8] , [9] . At this point , many germ cells begin to increase in size [8] , but more than half of the developing oocytes undergo apoptosis [10] , [11] . The remaining germ cells move into the proximal gonad , progress to diakinesis of prophase I , and arrest until activated by Major Sperm Protein to begin meiotic maturation [12] . After maturation , oocytes are fertilized and ovulated . They quickly complete meiosis , acquire an egg shell , begin embryogenesis , and are laid . In hermaphrodites , about half of all germ cells [10] , [11] and 10% of all somatic cells [13] , [14] undergo apoptosis . All of these deaths are controlled by a common genetic pathway [7]—CED-3 is an executioner caspase that causes apoptosis , the Apaf-1 homolog CED-4 binds to and activates CED-3 , and the Bcl-2 homolog CED-9 binds to and antagonizes CED-4 . Additional genes mediate the engulfment and removal of dying cells , but do not cause programmed cell death [7] . Finally , the BH3-containing protein EGL-1 inactivates CED-9 in the appropriate somatic cells , causing the release of CED-4 , which initiates apoptosis . Although many oocytes die , apoptosis does not occur during spermatogenesis and is not seen in the male germ line [10] . Physiological germ cell deaths require both ced-3 and ced-4 , but are not affected by mutations in egl-1 or by the ced-9 ( gf ) mutation [10] , which prevents EGL-1 from causing the release of CED-4 [15] , [16] . Although the majority of oocytes die in wildtype animals , both ced-3 and ced-4 mutants reproduce in large numbers even though no germ cell deaths occur . Thus the importance of these deaths has been unclear . Germ cells can also undergo apoptosis in response to DNA-damage [17] . These deaths require CEP-1 ( C . elegans p53 homolog ) , which acts through egl-1 and ced-13 to regulate ced-9 activity [18] , [19] . Loss-of-function mutations in any of these genes , as well as the ced-9 ( n1950gf ) mutation , prevent cell death in response to DNA damage , but have no effect on physiological germ cell deaths . Unsynapsed chromosomes also initiate apoptosis in germ cells through a process that does not require cep-1 but does use the AAA ATPase PCH-2 [20] . Finally , other kinds of stress induce germ cell deaths that occur independently of cep-1 [21] , [22] . Recent studies have shown that nematodes show an age-related decline in the number of progeny they produce [23] , but it is not clear what factors underlie this decline . In this paper , we focus on how oocyte quality is influenced by aging , and how apoptosis affects oocyte quality . Since hermaphrodites produce sperm , each new oocyte is fertilized soon after it matures , which makes it difficult to study changes during aging . Thus , we have been using fog-2 ( q71 ) females , which do not make sperm [24] , causing the oocytes to accumulate or “stack” within the gonads of virgin females . By delaying fertilization , we could study the quality of oocytes produced at different ages . We show that nematode oocytes decline in quality during aging , much as in mammals . Furthermore , we demonstrate that physiological germ cell deaths play a key role in maintaining oocyte quality , and that they function by promoting the efficient allocation of resources into developing oocytes .
Nematodes were handled using standard methods [25] . Animals were maintained on NG plates at 15°C ( NG Agar: 6 g NaCl , 9 g KH2PO4 , 1 . 5 g K2HPO4 , 12 g tryptone , 60 g Agar , and 1 ml cholesterol in ethanol ( 15 mg/ml ) are added to 3 L dH2O and autoclaved ) . To score progeny , the animals were raised on Low Growth NG plates ( as above , without tryptone ) at 15°C so that the bacterial lawn remained thin enough to allow accurate counts of eggs and larvae . To age the animals , females were first collected in the late L4 stage and checked for adulthood approximately 5 hours later . Those that had reached adulthood were then aged for 24 hours , 72 hours , or 144 hours for further assays . Females still in L4 were used for zero hour assays . All strains were derived from the wild-type Bristol strain N2 , and included the fog-2 ( q71 ) mutation to prevent sperm production in hermaphrodites , unless otherwise indicated . The alleles used in this study were: LG I: ced-1 ( e1735 ) [26] , fog-1 ( q250 ) [27]; LG III: ced-4 ( n1162 ) , ced-4 ( n2274 ) [28] , ced-6 ( n1813 ) [29] , ced-9 ( n1950 ) [30]; LG IV: ced-3 ( n718 ) , ced-3 ( n2439 ) , ced-3 ( n2921 ) [31] , [32]; LG V: fog-2 ( q71 ) [24] , egl-1 ( n1084n3082 ) [33] egl-1 ( n3330 ) ( B . Conradt and H . R . Horvitz , personal communication ) . Females were allowed to mate with 5 males for 10–14 hours before the males were removed . The females were then transferred to new plates every 12 hours until egg production ceased or until the female was unable to continue laying eggs . Eggs and larvae were counted at 0 hours , 12 hours , 24 hours , and 48 hours after the female had been transferred . Eggs that had not hatched by 48 hours were scored as dead; larvae that had not progressed past the L1 stage by 48 hours were scored as terminally arrested in development . We considered eggs less than approximately 1/3 normal size to be inviable and did not include them in our assays , but did note how frequently they occurred . Eggs greater than 1/3 normal size but still undersized were included in all of our assays , and their frequency was also noted . Line graphs only include time points for which at least 10 eggs for each age and genotype were available . To measure the effects of aging on specific oocytes , females were aged as described , anesthetized and mounted on slides [13] immediately prior to mating . Differential interference contrast microscopy was used to examine the germlines and count the number of full-grown oocytes stacked in the female gonad . The females were recovered and crossed with males for 12 hours , or until egg production began . Males were then removed and the females were transferred to new plates every 2–3 hours until all stacked oocytes had been fertilized and laid .
Nematode hermaphrodites reproduce early in adulthood and quickly exhaust their supply of self-sperm . To see if their oocytes change during aging , we began studying mated females , which have a larger supply of sperm and reproduce for a longer time . We found that almost all of the eggs produced by hermaphrodites or mated females during the first four days of adulthood were viable ( Figure 1A , Dataset S1 ) . However , viability declined when we assayed eggs made during the entire reproductive lifespan of females or mated hermaphrodites . Furthermore , this result was independent of the mutation we used to induce female development . Since the viability of eggs reflects the quality of the oocytes that produced them , we conclude that the quality of oocytes produced later in life is lower than that of earlier ones . Two simple models could explain this decline in quality: ( 1 ) these females might only be able to produce a limited number of high quality oocytes , and all additional oocytes would be inferior , or ( 2 ) regardless of the number of oocytes already produced , older mothers might produce oocytes of lower quality than younger mothers . To distinguish between these models , we allowed females to age before crossing them with males . We found that older mothers produced oocytes of significantly lower quality than younger ones ( Figure 1B , Dataset S1 ) , and also produced fewer fertilized eggs altogether ( Figure 1C , Dataset S1 ) . Thus , the decline in oocyte quality was determined by maternal age , rather than by the absolute number of eggs produced during an animal's lifespan . In these assays , all of the viable eggs yielded healthy larvae , regardless of the mother's age ( >99 . 5%; data not shown ) , so the main effect of low oocyte quality was on embryos . Since fog-2 ( q71 ) and fog-1 ( q250 ) females gave comparable results , the influence of aging on oocyte quality was independent of genetic background . To learn when the effect of maternal age on oocyte quality was most pronounced , we followed groups of eggs laid during 12-hour intervals over a female's entire lifespan , and determined what fraction survived and hatched ( Figure 2 ) . We found that two groups of oocytes were of lower quality than the rest: ( 1 ) the first oocytes fertilized in aging females , and ( 2 ) oocytes that developed more than 6 days after a female had matured , regardless of her age at mating . To determine if the initial decline in quality was due to ‘stacked’ oocytes that had spent an extended period of time arrested in diakinesis , we counted the number of stacked oocytes in 72-hour females just before mating , and then observed how many produced viable eggs . We found that 12% of the eggs that developed from stacked oocytes died before hatching ( 17 dead eggs from 146 stacked oocytes ) , compared with 3% of the eggs produced from oocytes that matured shortly afterward ( 3 dead eggs from 86 oocytes ) . Thus , an extended arrest in diakinesis was detrimental to oocyte quality . Previous studies reported that some ced-3 larvae arrest development before reaching adulthood [34] . In fact , we found that 16 . 5% of all ced-3 ( n718 ) larvae did not develop past the L1 larval stage ( Figure 3A , Dataset S1 ) , although the rest grew normally . Because we suspected that this problem might be caused by variations in maternal oocyte quality , we looked at heterozygous offspring produced by crossing ced-3 females with wild-type males and vice versa . Since the offspring from both of these crosses showed less than 1% larval arrest ( Figure 3A ) , the maternal ced-3 genotype did not cause the lethality . We repeated these crosses using older females and found similar results ( data not shown ) . Further experiments using ced-3 ( n2439 ) , ced-3 ( n2921 ) , ced-4 ( n1162 ) , ced-4 ( n2274 ) , egl-1 ( n1084n3082 ) and egl-1 ( n3330 ) showed the same effect ( Figure 3 and data not shown ) . Finally , analysis of ced-9 ( n1950gf ) females showed that the severity of the cell death defect correlated with the rate of larval lethality ( Figure 3A ) . The n1950 mutation is semi-dominant and has a maternal effect [30] . Heterozygous progeny from wild type mothers show low levels of cell survival [30] and low levels of larval arrest ( Figure 3A ) ; heterozygous progeny from n1950 mothers show higher levels of cell survival [30] and higher levels of larval arrest ( Figure 3A ) . Thus , blocking cell death causes some larvae to halt development and eventually die in the L1 stage , but oocyte quality plays no role in this process . We suspect that certain surviving cells occasionally interfere with normal development . During these studies , we found that ced-3 ( n718 ) eggs died before hatching more frequently than wildtype eggs ( Figure 4 , Dataset S1 ) . To see if this effect was due to decreased oocyte quality in the ced-3 mothers , we again looked at heterozygous offspring . All ced-3 mothers produced more dead eggs than did wildtype mothers , irrespective of the offspring's genotype ( Figure 4 ) , indicating that this effect was indeed maternal , and thus reflected a decrease in oocyte quality . Furthermore , the fraction of eggs from ced-3 mothers that died increased with maternal age , and this increase was more dramatic in ced-3 mothers than in wildtype mothers ( Figure 4 ) . We repeated these experiments using ced-3 ( n2439 ) and ced-3 ( n2921 ) females and observed the same effect ( Figure 4 ) , so it was caused by a decrease in ced-3 activity , rather than by a linked mutation . This age-related decline is specific to the germline , since ced-3 ( lf ) animals have normal lifespans [35] , [36] . We also determined the number of eggs laid by ced-3 females . Since no germ cells were being eliminated by apoptosis , we had expected that ced-3 females might produce more eggs than the wild-type . We found that the total number of eggs they laid was substantially lower , particularly as the females aged ( the average brood size of all 72 hour- and 144 hour-aged ced-3 mothers was 126 and 27 eggs respectively , compared with 180 and 65 eggs for wildtype mothers ) . However , we also saw an increase in the number of tiny , egg-like objects laid by these mothers , which suggested that ced-3 mutants had difficulty producing full-sized oocytes ( the average number of small egg-like objects for all 72 hour- and 144 hour-aged ced-3 mothers was 15 in both cases , compared with 2 and 8 respectively for wildtype mothers ) . Finally , we asked if the effect on oocyte quality we observed in ced-3 females was exclusive to ced-3 , or if other apoptotic genes were involved . Thus , we repeated our experiments using two alleles of ced-4 , n1162 and n2274 . We found that all ced-4 mothers produced more dead eggs than did wildtype mothers ( Figure 4 ) , implying that apoptosis itself is needed to maintain oocyte quality . As with ced-3 , the ced-4 effect increased with age , causing a severe , early decline in oocyte quality , as well as lower total numbers of eggs and increased numbers of tiny , egg-like objects ( data not shown ) . To determine when these mutants were producing defective oocytes , we monitored embryonic lethality in groups of eggs laid at 12-hour intervals over the course of the females' lifespan . We found that oocyte quality in both ced-3 and ced-4 mothers began to deteriorate 3–4 days after sexual maturation ( Figure 5 , boxed areas ) , compared with 6–7 days for wild-type mothers . Thus , blocking all cell deaths strongly influenced the quality of newly formed oocytes in older females . We also observed a modest decrease in viability among eggs produced from oocytes that had spent a prolonged period of time in diakinesis for both ced-3 and ced-4 mothers ( Figure 5 ) . Mutations in ced-3 and ced-4 prevent cell deaths in both the soma and the germline . However , the gain-of-function mutation ced-9 ( n1950 ) blocks somatic cell deaths , but does not affect physiological germ cell deaths [10] , [30] . To see if the decline in oocyte quality we observed in ced-3 and ced-4 mutants was caused by the lack of cell death in the soma , we studied eggs laid by ced-9 ( n1950 ) females . We found that their oocytes were , on average , as healthy as those from wildtype females for all time points and for mothers of all ages ( Figures 5 , 6A ) . Thus , somatic cell deaths are not required to maintain oocyte quality . To confirm these results , we also studied two loss-of-function mutations in egl-1 , a gene that negatively regulates ced-9 . These egl-1 mutations also block somatic cell deaths , but do not affect physiological germ cell deaths [10] . As with ced-9 ( gf ) , the egl-1 ( lf ) mutants produced oocytes that were at least as healthy as those from wild-type females ( Figures 5 , 6A , Dataset S1 ) . We conclude that blocking somatic cell deaths does not influence oocyte quality . Since the mutations in ced-9 and egl-1 also prevent germ cells from undergoing cell death in response to DNA damage [37] , our results imply that apoptosis does not normally maintain quality by eliminating oocytes that contain damaged DNA . Instead , high oocyte quality is maintained by the physiological germ cell deaths that occur in aging females . Two models could explain how germ cell deaths maintain oocyte quality . In the first , apoptosis eliminates defective oocytes from the germ line . In the second , apoptosis modulates the number of developing oocytes to help allocate resources properly . To distinguish between these models , we looked at the effect of mutations that do not block cell death , but do prevent the engulfment and metabolism of cell corpses . We found that mutations in both ced-1 and ced-6 decrease the quality of oocytes , although not as severely as do mutations in ced-3 or ced-4 , which block cell death altogether ( Figure 6A , Dataset S1 ) . When we examined these mutants , we found that their germ lines were often less well-organized in older females ( Figure 6B ) , although not as severely compromised as in ced-3 mutants . Since virgin ced-1 females had an average of 16 corpses per gonad arm at 72 hours ( n = 14 ) and an average of 21 corpses at 144 hours ( n = 30 ) , cell death is still ongoing in those worms . Thus , physiological germ cell deaths appear to maintain oocyte quality by regulating the allocation of resources in the aging germ line . These deaths might act directly by decreasing competition between oocytes , or indirectly by nourishing the somatic gonad , which engulfs each corpse and regulates meiotic maturation in surviving oocytes . Mutations in ced-3 or ced-4 cause germline hyperplasia during aging , resulting in more germ cells but fewer fully grown oocytes [10] . If additional resources are directed to these extra cells , this defect could lead to the production of small eggs that lack the resources needed for embryogenesis . Thus , we noted each time we observed eggs that were smaller than normal . If germ cell deaths were indeed needed to allocate resources in the germ line , low oocyte quality might correlate with the frequency of small eggs . When we plotted the relationship between oocyte quality and the frequency of small eggs , using data points for females of each age and genotype we had examined , we observed a roughly linear relationship between these traits , with a correlation coefficient of 0 . 90 . ( Figure 7A ) . In this compound data set ( n = 216 , 875 eggs ) , 6 . 0% of all eggs died before hatching , but we only scored 1 . 8% of the eggs as small . Some of the eggs that appeared normal might have had more subtle differences in size or composition . To determine if oocyte volume changes during aging , we graphed the frequency of small eggs among the progeny of wild-type mothers first mated at 0 , 24 , 72 or 144 hours ( Figure 7B ) , and of apoptosis defective mothers first mated at each of these times ( Figure 7C ) . For each graph , we also plotted the frequency of eggs that died before hatching . Finally , we plotted the frequency of tiny , egg-like objects we had observed in our studies but not included in our calculations of oocyte quality , since they appeared too small to be viable . We found correlation coefficients of greater than 0 . 997 for each of these markers of oocyte size with the quality of oocytes made by these females . In theory , small eggs could be produced either by the fertilization of small oocytes , or by the breakdown of larger oocytes . However , at the same time that the frequency of small eggs produced by ced-3 or ced-4 mutants was increasing ( Figure 7C ) , the number of maturing oocytes in their gonads was also increasing [10] and the average size of these oocytes was decreasing . In the wild type , the stacked oocytes each occupied an entire slice of the gonad ( Figure 7D ) , whereas in ced-3 mutants , the oocytes were smaller and lay along the surface of the germline , with additional oocytes located underneath in the same region ( Figure 7E ) . In fact , we observed an average of 47 oocytes in diakinesis in ced-3 mutants ( n = 6 gonad arms ) but only 20 in the wild-type ( n = 6 gonad arms ) , even though these cells were distributed over volumes that were equivalent in size . Thus , we conclude that aging animals have difficulty providing sufficient resources to nurture full-sized oocytes , and that defects in apoptosis aggravate this problem , contributing to the decline in oocyte quality .
In this paper , we show that oocyte quality declines in aging nematodes . This characteristic has not been described previously , because of the nature of hermaphrodite reproduction . In C . elegans , a wild-type hermaphrodite can produce about 2000 germ cells during its lifetime . However , the first 60–80 germ cells develop into 240–320 sperm , which fertilize each new oocyte soon after it is fully grown . Thus , hermaphrodites tend to reproduce exclusively at young ages , and over 99% of their self-fertilized eggs survive and hatch into healthy larvae . By working with female nematodes , we found that the oocytes produced and fertilized later in life , after a hermaphrodite's normal sperm supply would have been depleted , are of lower quality than oocytes produced at a younger age . Thus , aging nematodes show a decline in fertility , as occurs with humans and many other animals . This similarity makes C . elegans an attractive model for studying how oocyte quality is controlled during aging . The age-related decline of oocyte quality and quantity in humans is well documented [1] , [2] , [38] , [39] , [40] , [41] . By the time a female reaches puberty , the millions of oocytes she was born with will have dwindled to about 300 , 000 , with hundreds vanishing each month thereafter [2] . This rate of loss increases dramatically when a woman reaches about 37 years of age and continues until menopause , at which point less than 1000 follicles remain [40] . Oocyte quality decreases in parallel with a decrease in the oocyte population , resulting in an increased rate of defective mature oocytes as a woman ages . As a result , an older female is less likely to become pregnant than a younger female , and those that do become pregnant have a higher risk of having a child with birth defects , or losing the pregnancy altogether by miscarriage . Our studies show that the aging process affects two populations of oocytes in nematodes: ( 1 ) oocytes that were arrested in diakinesis while awaiting a signal to mature and be fertilized , and ( 2 ) developing oocytes in older mothers . Rather than compare human and nematode oocytes based only on their progression through meiosis , we suggest that the most important features to consider involve their underlying biology . In particular , oocytes in the first group have reached a mature size and are no longer susceptible to cell death [10] , so they might not be comparable to any stage of human oogenesis . However , the second group contains many cells that are arrested in prophase of meiosis I , awaiting a MAPK signal to grow or initiate apoptosis . Thus , they might be analogous to the vast population of human primordial follicles , which are also arrested in prophase of meiosis I , awaiting signals to develop into primary follicles or undergo cell death [42] , [43] . Do similarities in the aging process reflect similar biological causes ? Although they differ in many ways , nematode and mammalian oogenesis share several common steps . First , germ cells in both groups proliferate in a stem cell niche created by the somatic gonad [6] , [44] . Second , during early meiosis developing oocytes in both groups share cytoplasm—in the early stages of human follicle development , oocyte nuclei cluster together but are not separated by membrane boundaries [44] , [45] , [46] , and in nematodes young oocytes are part of a large syncytium [6] . Third , oocytes in both species make a major transition near the end of pachytene . In humans , folliculogenesis begins after oocytes exit from pachytene of meiosis I and form complete cell boundaries , each surrounded by somatic granulosa cells [47] , and in nematodes , oocytes exit from pachytene of meiosis I and form contacts with a new set of somatic gonadal sheath cells [48] . Finally , in both species oocytes arrest near the end of prophase in meiosis I , and wait for a signal to mature . Although these steps display many species-specific peculiarities , the underlying pathways that regulate somatic/germ cell interactions , that control the progression through meiosis , and that detect and respond to problems could be similar . Many studies done in mammalian systems suggest that the age of the father might also influence the viability of a developing embryo [49] , [50] , [51] . Specifically , offspring from older fathers are at higher risk of developing autism spectrum disorders or other neurodevelopmental problems [52] , [53] . Two observations suggest that paternal age is relatively unimportant in worms . First , we see no difference in embryo viability between a hermaphrodites' normal brood of 240–320 eggs and the first 240–320 eggs laid by females mated at sexual maturation ( Figure 1A ) . Second , we performed a pilot study using young adult ( zero hour ) females mated to males aged 0 hours , 144 hours , or 216 hours past sexual maturation and found no differences in embryonic lethality ( 1 . 9% , 2 . 6% and 1 . 6% respectively ) or normal larval development ( >99% in all cases ) . In nematodes , a hermaphrodite's health is harmed by mating with males [54] , [55]; in theory , this effect could influence oocyte quality . Nonetheless , we suspect it is unimportant , since hermaphrodites and young females mated at sexual maturation produce oocytes of similar quality for at least four days ( Figure 1A ) . Other studies also documented an age-related decline in progeny production [23] . Does this decline in the number of eggs produced by an aging nematode mirror the decrease in oocyte numbers seen in aging mammals ? We suspect not , since nematodes continue to produce germ cells throughout their adult life , and some mature oocytes are visible even in very old animals that no longer lay eggs . Thus , the decline in the number of fertilized eggs is a complex phenomenon that does not merely reflect the number of oocytes . Programmed cell death plays a major role in oogenesis in most animals [44] , [56] , [57] . Indeed , this is the predominant fate of developing oocytes in both humans and nematodes . In humans , approximately 7 million oocytes are produced during embryogenesis , but most of these die , and only about 400 follicles are normally ovulated during a woman's lifetime [47] . Similarly , a hermaphrodite nematode produces about 1700 oocytes over its lifetime , yet fewer than half survive to mature [10] , [11] , and a normal hermaphrodite only lays about 300 eggs [58] . Furthermore , in both groups a massive wave of cell death occurs around the time that oocytes exit from pachytene in prophase of meiosis I [10] , [44] . In mammals , a second wave of cell death affects the follicles that contain oocytes arrested in diplotene . We have shown that apoptosis plays a critical role in maintaining oocyte quality , but how it does so remains uncertain . Three popular theories explain how germ cells might be selected to die [57]: ( 1 ) the unfit oocyte theory , in which apoptosis removes defective oocytes from the pool , allowing only healthy oocytes to mature , ( 2 ) the nurse cell theory , in which some germ cells are selected to nourish maturing oocytes and later undergo apoptosis , as in Drosophila [59] , and ( 3 ) the neglected oocyte theory , which maintains that nutrients and other factors are in short supply in the germline , requiring some cells to die so that others have the resources to develop properly . These theories are not mutually exclusive , and germ cells might die for any one or a combination of these reasons . We propose that in C . elegans , most oocyte deaths function to help redistribute resources in the germ line . One potential role for apoptosis is the elimination of oocytes with certain types of genetic defects . In particular , unsynapsed chromosomes can trigger apoptosis in human oocytes [60] and in nematodes [20] . In addition , DNA-damage can act through an independent pathway to induce apoptosis in both species [37] , [61] . Since maternal age is the most important factor in trisomy formation [1] , [62] , and spindles in the meiotic oocytes of older females are less organized and the chromosomes are less firmly attached [2] , these types of death might maintain quality during aging by eliminating oocytes with chromosomal abnormalities . At least half of the developing oocytes die in nematodes [10] . If germ cell deaths removed only oocytes with damaged chromosomes , then preventing these deaths should result in a high rate of embryonic lethality . However , only 12% of eggs produced by ced-3 or ced-4 mothers died before hatching ( n = 46 , 678 eggs ) . Furthermore , using ced-9 ( n1950gf ) or egl-1 ( lf ) mutations to block germ-cell deaths induced by DNA damage did not lower oocyte quality ( Figures 6 , 8 ) . Thus , we suspect that the numerous physiological germ cell deaths in these animals serve another function . The key feature of nurse cells in Drosophila is that they are set aside from birth to produce materials for a developing oocyte , and then eventually undergo apoptosis . Similarly , in humans , maturing follicles contain numerous somatic cells which nourish the developing oocyte until its fate is determined , and many of these cells die [2] . If nematodes used immature oocytes as nurse cells , then blocking cell death should cause a population of these nurse cells to accumulate . To date , no one has reported a surviving population of nurse cells in ced-3 or ced-4 mutants [10] . Instead , our results show that the first 100–300 oocytes produced by young ced-3 or ced-4 females produce healthy eggs , extending previous observations by Gumienny et al . [10] . Thus , we propose that all germ cells in pachytene have an equal potential to develop into mature oocytes , and that none are specifically designated as nurse cells . This model agrees with studies showing that all germ cells in pachytene contribute cytoplasm to developing oocytes [63] . We propose that apoptosis helps allocate resources among developing oocytes , with some surviving and growing , and others dying and being recycled . This model rests on the following observations: ( 1 ) Oocytes produced by aging ced-3 or ced-4 females , which lack all cell deaths , had the poorest quality of any group we studied ( Figure 4 ) . ( 2 ) These ced-3 or ced-4 mothers also produced more small oocytes than the wild type ( Figure 7 ) . ( 3 ) These problems became more severe as the females aged . ( Figures . 5 , 7 ) . ( 4 ) Oocytes produced by aging ced-1 and ced-6 females , which cannot engulf and recycle cell corpses , were also of lower quality than wild-type oocytes ( Figure 5 ) . We infer that in the absence of germ cell deaths , there is a premature depletion of resources caused by too many competing oocytes . When germ cells undergo apoptosis but are not engulfed and recycled , a smaller depletion of resources occurs . Another result supports this model—the quality of the oocytes arrested in diakinesis is lower in ced-3 or ced-4 mutants than in the wild type . This decline in quality cannot be due to defects in the oocytes themselves , since these oocytes are of higher quality if fertilized immediately . ( Compare ced-3 data for females mated at zero hours with that for females mated at older ages ) . Instead the quality of these oocytes declines over time , and declines faster in ced-3 or ced-4 mutants than in the wild type . Thus , we suspect that a competition for resources in the ced ( lf ) mutants affects the ability of the aging germ line to nurture and maintain arrested oocytes . Indeed , there are often smaller oocytes interspersed with full-grown oocytes in the proximal gonads of older ced-3 and ced-4 females ( Figure 7 and data not shown ) . Although physiological germ cell death plays a major role in maintaining oocyte quality , other factors could be involved . Stresses such as food deprivation [21] , [64] , pathogen infection [65] , and exposure to toxins [21] can trigger germ cell death by acting through CED-9 [10] . These signals might maintain oocyte quality when animals are raised in adverse conditions . In our studies , we found that older ced-3 or ced-4 mutants stopped producing fertilized eggs at a younger age than the wild-type , even though they continued to produce oocytes . This result suggests that some mechanism regulates the ability of developing oocytes to mature and be fertilized , and that this mechanism plays a major role when quality is declining . If so , altering oocyte quality might also influence an animal's overall fertility . One possibility is that some characteristic of oocytes that directly reflects quality determines if an oocyte is able to mature and be fertilized . Alternatively , the accumulation of extra oocytes in these mutants might interfere with the normal rhythm of maturation and fertilization . We have shown that oocyte quality declines in aging nematodes , as it does in humans , and that apoptosis prevents a premature decline in quality . Could some mutations actually improve oocyte quality and reproductive success ? We were surprised to find that both of our egl-1 mutants produce oocytes of higher quality than the wild type in very old animals ( Figure 6A ) . This difference is significant at a 99% confidence level . Furthermore , these egl-1 animals reproduce for a longer period of time than the wild type or than other ced mutants ( Figure 5 ) . Mutations in other genes also extend the reproductive span in nematodes [23] , [66] . For comparison , studies with mice show that mutations in Bax , which block many germ cell deaths , extend the reproductive lifespan of females [67] . However , the analysis of oocyte quality is difficult in mammals , since embryos develop in utero , and it is not known how quality changes in Bax mice during aging . Nonetheless , it seems likely that some genetic changes can improve reproductive success in older mothers . We are now studying the relationship between the control of physiological germ cell deaths and the induction of cell death in response to DNA damage . Furthermore , we are exploring the link between these processes and the DAF-2/insulin-like signaling pathway , which is a conserved regulator of lifespan and development in all tissues [68] , [69] .
|
As women age , the quality of their oocytes declines , causing the eggs they make to have a higher chance of producing a miscarriage or a child with birth defects . We used the roundworm C . elegans to study this problem . We show that oocyte quality declines in these small animals during aging , much as in mammals . Furthermore , our results show that the programmed deaths of many developing oocytes help maintain the quality of the oocytes that survive , resulting in better eggs . These cell deaths appear to regulate the way resources are allocated in the aging germ line . Since many oocytes die in humans as well as in nematodes , our studies point to the possibility of improving oocyte quality by manipulating cell death in the germ line .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"developmental",
"biology/germ",
"cells",
"genetics",
"and",
"genomics/animal",
"genetics",
"cell",
"biology/cellular",
"death",
"and",
"stress",
"responses",
"developmental",
"biology/aging",
"developmental",
"biology/developmental",
"molecular",
"mechanisms"
] |
2008
|
Apoptosis Maintains Oocyte Quality in Aging Caenorhabditis elegans Females
|
Most fungal pathogens of humans display robust protective oxidative stress responses that contribute to their pathogenicity . The induction of enzymes that detoxify reactive oxygen species ( ROS ) is an essential component of these responses . We showed previously that ectopic expression of the heme-containing catalase enzyme in Candida albicans enhances resistance to oxidative stress , combinatorial oxidative plus cationic stress , and phagocytic killing . Clearly ectopic catalase expression confers fitness advantages in the presence of stress , and therefore in this study we tested whether it enhances fitness in the absence of stress . We addressed this using a set of congenic barcoded C . albicans strains that include doxycycline-conditional tetON-CAT1 expressors . We show that high basal catalase levels , rather than CAT1 induction following stress imposition , reduce ROS accumulation and cell death , thereby promoting resistance to acute peroxide or combinatorial stress . This conclusion is reinforced by our analyses of phenotypically diverse clinical isolates and the impact of stochastic variation in catalase expression upon stress resistance in genetically homogeneous C . albicans populations . Accordingly , cat1Δ cells are more sensitive to neutrophil killing . However , we find that catalase inactivation does not attenuate C . albicans virulence in mouse or invertebrate models of systemic candidiasis . Furthermore , our direct comparisons of fitness in vitro using isogenic barcoded CAT1 , cat1Δ and tetON-CAT1 strains show that , while ectopic catalase expression confers a fitness advantage during peroxide stress , it confers a fitness defect in the absence of stress . This fitness defect is suppressed by iron supplementation . Also high basal catalase levels induce key iron assimilatory functions ( CFL5 , FET3 , FRP1 , FTR1 ) . We conclude that while high basal catalase levels enhance peroxide stress resistance , they place pressure on iron homeostasis through an elevated cellular demand for iron , thereby reducing the fitness of C . albicans in iron-limiting tissues within the host .
Of the circa three million fungal species that are thought to inhabit our planet [1] , only a relatively small number have been reported to cause infections in humans . ( About 400 species are described in the Atlas of Clinical Fungi [2] . ) Nevertheless , there is an increasing awareness that these fungal pathogens impose a major burden on human health worldwide [3] . These clinically important fungi generally share common features that promote colonization of their human host , such as the thermotolerance that permits growth at body temperatures . These common features include relatively robust stress responses , which mitigate against the stresses imposed by host immune defences [e . g . 4–6] . They also include the ability to scavenge essential micronutrients , such as iron , from their host [7–10] . Iron is an essential micronutrient that is required for the functionality of key ferroproteins and haem proteins . However , excess iron is toxic because ferrous ions promote the Fenton reaction which produces highly toxic hydroxyl radicals [11] , and therefore host and pathogen alike must tightly regulate their acquisition , storage and release of iron . Consequently , the levels of free ion are vanishingly low in some host niches [12] . Furthermore , following infection the host activates the process of nutritional immunity in an effort to limit iron availability for the invading microbe [10 , 12] . Fungal pathogens respond to this iron limitation by down-regulating genes encoding iron-containing proteins and upregulating efficient iron scavenging mechanisms [13–17] . In Candida albicans this response includes the induction of genes encoding ferric reductases ( e . g . CFL5 , FRP1 ) , high affinity iron permeases ( e . g . FTR1 ) and proteins involved in iron assimilation ( e . g . FET3 ) [15] . This response allows the fungus to counter the changes in iron homeostasis within the host that are triggered by systemic candidiasis [10] . Fungal pathogens activate oxidative stress responses when they come in contact with the host [18–22] , and these responses promote resistance to phagocytic attack and fungal virulence [5 , 23–26] . In an attempt to clear invading fungal pathogens , host neutrophils and macrophages phagocytose the fungal cells and subject them to a battery of reactive oxygen species ( ROS ) that damage proteins , DNA and lipids , and can induce programmed cell death [27] . The impact of ROS is augmented when combined with a cationic stress , and this synergistic impact of combinatorial oxidative and cationic stresses is thought to contribute to the potency of human neutrophils [28 , 29] . C . albicans cells respond to oxidative stress by inducing functions that detoxify the ROS , repair the oxidative damage , synthesize antioxidants and restore redox homeostasis . This includes the induction of genes encoding catalase ( CAT1 ) , superoxide dismutases ( SOD ) , glutathione peroxidases ( GPX ) and components of the glutathione/glutaredoxin ( GSH1 , TTR1 ) and thioredoxin ( TSA1 , TRX1 , TRR1 ) systems [6 , 30–32] . In particular , CAT1 mRNA levels are strongly induced by oxidative stress [30 , 33] . However , C . albicans cells are unable to activate a normal transcriptional response to oxidative stress when subjected to combinatorial oxidative plus cationic stress or acute peroxide stress , and this contributes to the lethality of these types of stress [28 , 29] . Catalase ( Cat1 ) plays a major role in protecting C . albicans against peroxide stress [28 , 29] . This iron-requiring enzyme , which has been well-characterised structurally [34] , belongs to a superfamily of heme peroxidases and catalases that are conserved across bacteria , plants , fungi and animals [35] . Catalase catalyses the conversion of hydrogen peroxide ( H2O2 ) to water . C . albicans cells rapidly detoxify extracellular H2O2 following exposure to an acute peroxide stress , and this detoxification is mainly dependent on catalase ( CAT1 ) [28] . We showed previously that ectopic expression of catalase using the ACT1 promoter ( ACT1p-CAT1 ) protected C . albicans from acute oxidative and combinatorial stresses [28] . More recently , Jesus Pla’s group has confirmed that catalase overexpression protects C . albicans against peroxide stress [36] . They also demonstrated that high catalase levels provide protection against antifungal drugs . These observations raise an interesting conundrum: if catalase overexpression confers effects that might be expected to promote host colonisation , why has C . albicans not evolved to express high basal levels of catalase ? We address this in this study . We show that while high basal catalase levels enhance the fitness of C . albicans in the presence of oxidative and combinatorial stresses , these high catalase levels reduce fitness in the absence of stress . We also reveal the molecular basis for this fitness defect . Our observations suggest a partial explanation for the lack of emergence of catalase overexpression during the evolution of this major fungal pathogen . We also show that , in contrast to the prevailing view [23] , the virulence of C . albicans is not compromised by catalase inactivation .
We demonstrated previously that ectopic expression of catalase from an ACT1 promoter-CAT1 fusion ( ACT1p-CAT1 ) reproducibly protected C . albicans against acute peroxide stress ( 5 mM H2O2 ) and combinatorial stress ( 5 mM H2O2 plus 1 M NaCl ) [28] . Subsequently we noted that the stress resistance of ACT1p-CAT1 cells declined over time ( S1 Fig ) . Therefore , we constructed new C . albicans strains in which catalase expression is regulated by the doxycycline conditional tetON promoter [37–39] . Control strains were made by transforming congenic wild-type ( CAT1 ) and catalase null strains ( cat1Δ ) with empty tetON vectors . Test strains were made by integrating a tetON-CAT1 plasmid into the genome of the cat1Δ null mutant . We refer to these strains , which all have the same genetic background ( Materials and Methods; S1 Table ) , as wild-type ( CAT1 ) , null ( cat1Δ ) and tetON-CAT1 strains , respectively . Three isolates were generated for each strain type . For each strain type , the isolates displayed similar stress phenotypes ( below ) . First we tested Cat1 expression levels in wild-type ( CAT1 ) , null ( cat1Δ ) and tetON-CAT1 cells . Catalase levels were induced in response to oxidative stress in wild-type ( CAT1 ) cells , and were undetectable in cat1Δ cells ( Fig 1 ) . Catalase levels in these strains were not affected by doxycycline addition . In contrast , catalase levels were strongly induced by doxycycline in tetON-CAT1 cells ( red bars , Fig 1 ) . Significantly , wild-type cells express significant basal levels of catalase in the absence of stress ( Fig 1 ) , as we reported previously [33] . Catalase levels in doxycycline-treated tetON-CAT1 cells were higher than these basal levels ( Fig 1 ) . We then compared the stress resistance of wild-type , null and tetON-CAT1 cells ( Fig 2A & S1 Fig ) . As expected [23 , 28 , 40] , wild-type ( CAT1 ) cells displayed modest resistance to an oxidative stress ( H2O2 ) and a combinatorial stress ( H2O2 plus NaCl ) , whereas the null mutant ( cat1Δ ) was sensitive to both types of stress . These phenotypes were not affected by the presence or absence of doxycycline ( Fig 2A ) . In the absence of doxycycline , the tetON-CAT1 strains were sensitive to both oxidative and combinatorial stress , reflecting their null cat1Δ background . When these strains were pre-grown with doxycycline , they displayed enhanced oxidative and combinatorial stress resistance ( Fig 2A ) . This reinforces our earlier conclusion [28] that elevated basal CAT1 expression levels protect C . albicans cells against a sudden and acute oxidative or combinatorial stress . Interestingly , the tetON-CAT1 strains were sensitive to both oxidative and combinatorial stress when pre-grown in the absence of doxycycline and the inducer was only provided when the stress was imposed ( Fig 2A ) . We then measured the impact of individual and combinatorial oxidative ( H2O2 ) and cationic ( NaCl ) stresses upon cell death by cytometric analysis of propidium iodide ( PI ) stained doxycycline-grown cell populations ( Fig 2B ) . Relative to the control wild-type strain , cat1Δ null cells were more sensitive , and tetON-CAT1 cells were more resistant to these oxidative and combinatorial stresses . Compared to the control wild type cells , doxycycline-treated tetON-CAT1 cells displayed 9-fold less cell death following exposure to the oxidative stress , and 2 . 5-fold less death after the combinatorial stress ( Fig 2B ) . This correlated with a reduction in internal ROS accumulation following stress imposition by tetON-CAT1 cells relative to the wild-type and cat1Δ cells ( Fig 2C ) . The accumulation of intracellular ROS was 2 . 6-fold lower in doxycycline-treated tetON-CAT1 cells after the peroxide stress , and 1 . 5-fold lower following the combinatorial stress , compared to the wild type control ( Fig 2C ) . Taken together , our data indicate that cells with low catalase levels at the point of stress imposition are more sensitive to peroxide than cells with high catalase levels . This suggests if catalase levels are low at the point of stress imposition , the dynamics of catalase induction are too slow to permit the normally rapid clearance of peroxide [28] and to prevent ROS-mediated cell death [27] . The data indicate that C . albicans cells require high basal levels of catalase at the time of stress imposition if they are to survive an acute oxidative or combinatorial stress . C . albicans clinical isolates display a high degree of natural variation [41 , 42] . We exploited this to select strains that display relatively low levels of oxidative stress resistance . A diverse range of C . albicans clinical isolates ( 65 in total ) from different epidemiological clades and from different patient colonisation sites were subjected to a robotic screen in which they were plated on YPD containing different peroxide concentrations ( Fig 3A ) . All of the isolates tested displayed a degree of resistance to this stress , showing some growth at 3 . 2 mM H2O2 . However , some isolates were more resistant to peroxide , displaying robust growth at 6 . 4 mM H2O2 , whereas sensitive strains were unable to grow at this H2O2 concentration . We selected a subset of four sensitive isolates and three resistant isolates ( which included SC5314 , the clinical isolate from which most laboratory strains are derived ) , and compared the basal CAT1 expression levels in these isolates to a standard laboratory strain ( CAI4 containing CIp10 ( URA3 ) ) . Basal CAT1 mRNA levels were lower in the oxidative stress sensitive isolates tested ( Fig 3B ) , and furthermore , the basal levels of the enzyme were also lower in these isolates ( Fig 3C ) . These data are consistent with the idea that elevated basal catalase levels promote oxidative stress resistance in C . albicans . Next we examined how a subset of cells within an apparently homogeneous population of C . albicans cells can survive an acute oxidative stress [28 , 36 , 43] . Based on the above observations , we reasoned that this might be partly explained by stochastic variation in basal catalase levels between individual cells in such a population . To test for potential population heterogeneity in basal catalase levels we generated a strain in which both CAT1 alleles were tagged with GFP ( CAT1-GFP/ CAT1-GFP ) to express a Cat1-GFP fusion protein . Western blotting revealed a Cat1-GFP protein of the expected mass in these cells ( approximately 80 kDa: Fig 4A ) , and the GFP fluorescence was located in punctate spots ( Fig 4B ) , consistent with the peroxisomal localisation of catalase in C . albicans [44] . We then compared the oxidative stress resistance of the CAT1-GFP strain with congenic control wild-type ( CAT1/CAT1 ) , heterozygous ( CAT1/ cat1Δ ) and null ( cat1Δ/cat1Δ ) strains . The CAT1-GFP strain was as resistant to oxidative stress as the wild-type control ( Fig 4C ) , indicating that the CAT1-GFP alleles are functional . We then examined the basal levels of GFP fluorescence in unstressed populations of exponentially growing C . albicans CAT1-GFP cells by flow cytometry . As predicted , these genetically homogeneous cell populations displayed heterogeneity with respect to their Cat1-GFP expression levels ( Fig 4D & S2 Fig ) . Using flow cytometry , we selected cells of similar size , sorted cells that display relatively low levels of Cat1-GFP from those expressing high levels ( Fig 4D & S2 Fig ) , and then plated them onto media containing a range of H2O2 concentrations . Cells expressing relatively high levels of Cat1-GFP were more resistant to peroxide stress ( Fig 4E ) . When an analogous experiment was performed with cells expressing a control gene ( ACT1-GFP ) , stochastic differences in ACT1-GFP expression did not affect oxidative stress resistance ( S3 Fig ) . These observations reinforce our conclusion that high basal levels of catalase promote oxidative stress resistance . Furthermore , this confirms that C . albicans cell populations display stochastic variation in their basal CAT1 expression levels , and that this contributes to the survival of a subset of C . albicans cells following an acute oxidative stress . We tested whether high basal catalase levels affect the ability of C . albicans to colonise different tissues during systemic infection . At first we reasoned that the elevated oxidative stress resistance conferred by high basal catalase levels ( above ) might enhance host colonisation . To test this we compared directly the three isolates for wild-type ( CAT1 ) , null ( cat1Δ ) and tetON-CAT1 strains ( nine in total ) using a barcode sequencing ( barseq ) strategy . The C . albicans strains were pre-grown separately in the presence or absence of doxycycline . Approximately equal amounts of the nine doxycycline-treated strains were mixed and used to induce disseminated candidiasis in doxycycline-treated mice . In parallel , the nine untreated control C . albicans strains were mixed and used to infect untreated mice . Mice from each group were culled after four days , and the fungal cells harvested from their kidneys , livers , spleens and brains . Barseq was then performed on genomic DNA from these fungal populations to determine the relative proportion of each C . albicans strain in each tissue . We observed significant differences between the wild-type ( CAT1 ) and tetON-CAT1 strains in their ability to colonise certain tissues ( Fig 5 ) . Doxycycline-treated tetON-CAT1 cells were less able to colonise the kidney and brain than the control untreated tetON-CAT1 cells , but this was not the case in the liver and spleen . This effect was observed for tetON-CAT1-1 cells , but not for the other two tetON-CAT1 isolates ( 4 and 10: S1 Table ) . This correlated with a reduction in catalase levels in these isolates ( S4A Fig ) and a corresponding loss of phenotype ( S4B Fig ) . Therefore , like ACT1p-CAT1 cells ( above; S1 Fig ) , isolates 4 and 10 appeared to have lost their phenotype over time . Taken together , our data indicate that , contrary to our initial prediction , high basal catalase expression levels appear to compromise , rather than enhance , the ability of C . albicans to colonise certain tissues . To our surprise , we did not observe any significant differences between the wild-type ( CAT1 ) and null ( cat1Δ ) strains in their ability to colonise the host ( Fig 5 ) . All of the wild-type and null isolates displayed similar levels of colonisation . This indicated that cells lacking catalase can infect the host—a conclusion that contrasts with the prevailing view that C . albicans cat1Δ null cells display attenuated virulence [23 , 40] . We reasoned that cat1Δ cells might be able to colonise host tissues if they are co-infected with CAT1 and tetON-CAT1 cells . For example , cat1Δ null cells might be rescued via a “cheater” or “bystander” effect [45 , 46] , whereby catalase expressing cells protect the null mutant against local peroxide stress . We tested this by comparing the virulence of our wild-type ( CAT1 ) and null ( cat1Δ ) strains separately in the three-day murine model of systemic candidiasis [47] . We observed no significant difference between the wild-type or mutant strains with respect to their fungal burdens in the kidneys , and the strains induced similar levels of weight loss in mice , yielding similar outcome scores that displayed no significant difference ( Fig 6A ) . This observation reinforced the idea that inactivating CAT1 does not attenuate the virulence of C . albicans . Wysong and co-workers observed a virulence defect for cat1Δ cells using a long-term mouse model of systemic candidiasis [23] . Therefore , it seemed possible that our short-term and their long-term model of systemic infection might yield differing outcomes for C . albicans cat1Δ cells . To test this we re-examined the virulence of our wild-type ( CAT1 ) and null ( cat1Δ ) strains in mice over 14 days . ( We were unable to access the strains used by Wysong and co-workers [23] . Hence we could not perform a direct comparison with their mutant . ) No major difference in the virulence of wild-type and cat1Δ cells was observed using a long term infection model ( p = 0 . 074: Fig 6B ) . We also compared our wild-type ( CAT1 ) , null ( cat1Δ ) and tetON-CAT1 strains in Galleria mellonella , observing no significant difference in their virulence in this invertebrate model of systemic candidiasis ( p = 0 . 68: Fig 6C ) . The cat1Δ mutant generated by Wysong and co-workers had the URA3 marker inserted at the cat1 locus ( cat1::URA3 ) [23] . In contrast , in our cat1Δ mutant URA3 was reintroduced at the RPS1 locus using the CIp10 plasmid backbone [48] . After the study of Wysong and co-workers was published [23] , URA3 position effects were found to influence C . albicans virulence , and reinsertion of URA3 at RPS1 using CIp10 was shown to overcome these effects [49] . We conclude that CAT1 inactivation does not significantly attenuate the virulence of C . albicans . It has been reported that catalase null mutants do not display significantly higher sensitivities to neutrophil killing [5] . Once again , these experiments were performed with a cat1Δ null mutant in which URA3 was integrated at the CAT1 locus ( cta1Δ::loxP-URA3-loxP: [5] ) . Therefore , in light of our findings ( above ) , we retested neutrophil killing using our new cat1Δ strain in which URA3 is integrated at the RPS1 locus . We tested the strains separately to exclude potential cheater effects [45 , 46] . We observed that , following exposure to human neutrophils , our new cat1Δ strain displayed significantly reduced survival compared to the congenic wild-type control ( Fig 7 ) . This strengthens the observation of Miramon and co-workers , who reported a slight difference between cat1Δ and CAT1 cells that was not statistically significant [5] . Furthermore , we also observed a statistically significant difference in neutrophil killing between tetON-CAT1 cells that were pre-grown in the presence or absence of doxycycline ( Fig 7 ) . These data indicate that catalase promotes the resistance of C . albicans against neutrophil attack . We note that elevated basal levels of catalase did not enhance the resistance of C . albicans to neutrophil killing in our hands ( Fig 7: compare wild type and doxycycline-treated tetON-CAT1 cells ) . C . albicans cat1Δ cells are clearly sensitive to oxidative stress ( Fig 2 ) . However , in mixed populations they could conceivably be rescued by catalase expressing cells . Therefore , we tested whether cat1Δ cells act as cheaters by examining their fitness in mixed cultures alongside wild-type ( CAT1 ) and tetON-CAT1 cells . The three barcoded for the null mutant , wild-type and tetON-CAT1 strains were pre-grown separately in the presence of doxycycline , mixed in approximately equal proportions , and then used to inoculate YPD cultures containing doxycycline . A parallel mixture of untreated barcoded strains was also prepared , and this untreated mixture used to inoculate YPD cultures without doxycycline . The relative fitness of each strain was then compared in the presence or absence of oxidative stress ( 5 mM H2O2 ) , by comparing the relative abundance of each barcode over time in each culture by barseq . With one notable exception ( discussed below ) , the three isolates for each strain type displayed similar behaviours ( Fig 8 ) . In the absence of doxycycline and stress , the relative abundance of the wild-type ( CAT1 ) , null ( cat1Δ ) and tetON-CAT1 strains did not change significantly over the twelve hour period examined ( Fig 8A ) . In contrast , in the absence of doxycycline but in the presence of stress , the abundance of cat1Δ and tetON-CAT1 cells rapidly declined in the population and these strains were rapidly outcompeted by the wild-type CAT1 strains ( Fig 8B ) . The comparable behaviour for the cat1Δ the tetON-CAT1 cells under these conditions was entirely consistent with the negligible catalase levels in tetON-CAT1 cells without doxycycline induction ( Fig 1 ) . These data strongly reinforce the view that catalase is vital for peroxide stress resistance in C . albicans [5 , 23 , 33 , 36 , 40] . Our data also show that cat1Δ cells do not act as cheaters: they are not rescued by catalase expressing cells under peroxide stress conditions ( Fig 8B ) . In the presence of doxycycline in the presence of stress , the tetON-CAT1 cells rapidly outcompeted the null ( cat1Δ ) cells ( Fig 8D ) . This again highlighted the peroxide sensitivity of cat1Δ cells . Significantly , the tetON-CAT1 cells also out-competed wild-type ( CAT1 ) cells ( Fig 8D ) , confirming directly that ectopic catalase expression enhances oxidative stress resistance ( Fig 2 ) [28 , 36] . Therefore , elevated basal catalase levels increase the fitness of C . albicans cells in the presence of peroxide stress . Interestingly , in the presence of doxycycline but in the absence of peroxide stress , there was a decrease in the abundance of tetON-CAT1-01 cells in the population over the twelve hour time-course , relative to the wild-type ( CAT1 ) and null ( cat1Δ ) cells ( Fig 8C ) . This suggested that ectopic CAT1 expression might render C . albicans cells less fit in the absence of stress . Doxycycline-treated C . albicans tetON-CAT1 cells appeared to display a fitness defect in the absence of stress ( Fig 8C ) . We tested this further by examining biomass formation on YPD ( final OD600 ) ( Fig 9A ) . All of the strains displayed similar growth in the absence of doxycycline , and the wild-type ( CAT1 ) controls remained unaffected by doxycycline . However , the growth of tetON-CAT1 cells decreased in the presence of doxycycline , reinforcing the view that elevated catalase levels reduce fitness in the absence of stress . Catalase is a ferroprotein [34] expressed at relatively high basal levels in C . albicans ( approximately 1 . 5 x 105 molecules per cell [33] ) . In bacteria , catalase overexpression has been reported to affect the requirement for iron [50] . Therefore , we reasoned that the fitness defect conferred by high basal catalase levels in C . albicans might be mediated by an elevated cellular demand for iron . Hence , we tested whether iron supplementation can suppress this fitness defect . Growth of tetON-CAT1 cells was measured in YPD containing doxycycline supplemented with different concentrations of ferric ions ( Fig 9B ) . These data indicate that the fitness defect caused by ectopic catalase expression can be completely suppressed by iron supplementation . This suppression was due to the improved growth of doxycycline-treated tetON-CAT1 cells in the presence of iron ( S5A Fig ) . We also showed that iron supplementation suppresses the reduced fitness of doxycycline-treated tetON-CAT1 cells in direct competition experiments with wild type ( CAT1 ) cells ( S5B Fig ) . These observations suggested that high basal catalase expression increases the cellular demand for iron in C . albicans . To test this further we examined the impact of ectopic catalase expression upon key genes involved in iron assimilation and homeostasis: CFL5 ( encoding a ferric reductase that is induced in low iron ) , FET3 ( encoding a copper oxidase that is required for growth in low iron ) , FRP1 ( encoding a ferric reductase that is induced by iron chelation ) and FTR1 ( encoding a high-affinity iron permease that is required for growth in low iron ) . All of these genes are targets of the iron-responsive transcriptional activator Sef1 [15] . CFL5 , FET3 , FRP1 and FTR1 transcript levels were measured relative to the ACT1 mRNA internal control in tetON-CAT1 cells grown in the presence and absence of doxycycline . Their levels were then normalised against those in doxycycline-treated wild type ( CAT1 ) cells to exclude any potential effects of this treatment on these transcripts [51] . All four iron-responsive transcripts were strongly induced following tetON-CAT1 induction ( Fig 9C ) . Taken together , the data indicate that high basal catalase levels increase the requirement for iron in C . albicans .
This study has important implications for the impact of the key peroxide detoxifying enzyme , catalase , upon the stress resistance and virulence of the major fungal pathogen , C . albicans . Firstly , our analyses of new cat1Δ null mutants , in which potential URA3 position effects have been circumvented [49] , have reinforced the view that catalase is essential for normal levels of oxidative and combinatorial stress resistance in C . albicans ( Figs 2 & 8 ) . They also show that catalase contributes to the resistance of this pathogenic fungus against neutrophil killing ( Fig 7 ) . However , our most surprising finding was that , in contrast to the generally held view [23 , 40] , catalase is not essential for the virulence of C . albicans , at least in models of disseminated candidiasis . This unexpected finding is supported by virulence assays in both short term and long term murine models of systemic infection , and in an accepted invertebrate model of systemic infection ( Fig 6 ) . This view is further reinforced by our in vivo competition experiments , in which the cat1Δ null mutant competed effectively against wild-type and catalase overexpressing strains for colonisation of the kidney , liver , spleen and brain ( Fig 5 ) . We suggest that the attenuated virulence of the cat1Δ mutants reported previously [23 , 40] might be explained by URA3 position effects in these strains [49] . Why might catalase be important for oxidative stress resistance and yet apparently not required for systemic infection ? The sensitivity of cat1Δ cells to neutrophil killing ( Fig 7 ) does indicate that protection against peroxide is required in certain contexts in vivo . Therefore , this lack of cat1Δ virulence defect probably reflects the multifactorial nature of virulence phenotypes , as well as the nature of the systemic infection models often used to examine virulence . In these models sufficient fungal doses are applied to overcome immediate clearance by circulating phagocytes [47] . Furthermore , few of the fungal cells colonising the kidney appear to be exposed to oxidative stress [31] . Secondly , our data indicate that high basal levels of catalase promote the resistance of C . albicans to peroxide and combinatorial stress ( Fig 2 ) . These data reaffirm previous reports that elevated catalase expression promotes peroxide resistance [28 , 36] . Significantly , our data indicate that this phenotype is dependent on high basal levels of catalase at the point of stress imposition , rather than CAT1 induction in response to stress . Three independent observations support this view . ( A ) tetON-CAT1 cells are only protected against peroxide or combinatorial stress if these cells are pre-treated with doxycycline , not if doxycycline is only provided at the same time as the stress ( Fig 2 ) . ( B ) Clinical isolates that are relatively resistant to oxidative stress tend to express catalase at relatively high levels ( Fig 3 ) . ( C ) Unstressed C . albicans cell populations display heterogeneity in Cat1-GFP levels , and those cells that express more Cat1-GFP are less susceptible to killing by oxidative stress ( Fig 4 ) . Hydrogen peroxide is normally rapidly detoxified by wild-type C . albicans cells ( within 60 minutes ) in a catalase-dependent fashion [28] . Elevated basal levels of catalase presumably enhance cellular protection by accelerating the clearance of this reactive oxygen species . The heterogeneity in catalase expression within C . albicans populations , which might arise via stochastic differences between cells [52–54] , appears to account , in large part , for the ability of a subset of C . albicans cells to survive an acute oxidative stress . This would appear to represent the first example in C . albicans of the kind of “bet-hedging” strategies that have been observed in bacterial and S . cerevisiae populations [55 , 56] . Furthermore , these observations are entirely consistent with the well-established observation that an adaptive response to a small dose of a particular stress can transiently endow yeasts with resistance to a subsequent acute dose of the same stress by inducing the expression of key stress protective functions . This observation has been reported for heat shock , osmotic and oxidative stress in S . cerevisiae for example [57 , 58] , and has been extended to other yeasts including C . albicans [43 , 59 , 60] . Thirdly , our data provide key insights into the impact of catalase levels on the virulence of C . albicans . In our hands , direct competition assays suggested that elevated catalase levels might affect C . albicans colonisation of the kidney and brain ( Fig 5 ) . This is consistent with a parallel study which reported that catalase overexpression attenuates the virulence of C . albicans [36] . Roman and co-workers described this as “a most unexpected result” given that catalase overexpression enhances oxidative stress resistance . They speculate that this might have arisen via some alteration in fitness , which they were unable to detect in vitro , but which might interfere with activation of the Hog1 and Mpk1 MAP kinases [36] . In this study we show clearly in direct competition assays that elevated basal catalase levels attenuate the fitness of C . albicans in the absence of stress ( Fig 8 ) . We conclude that catalase overexpression confers a selective disadvantage in C . albicans in the absence of stress . Fourthly , we have identified a major cause of this fitness defect . High basal catalase levels increase the cellular requirement for iron in C . albicans . We present two key observations that support this . ( i ) The fitness defect is suppressed by iron supplementation ( Fig 9B and S4 Fig ) . This effect , which has also been observed in bacteria [50] , is probably mediated by the depletion of intracellular iron through high level expression of an abundant heme-requiring enzyme . ( ii ) Ectopic catalase expression induces the expression of iron-responsive genes that play key roles in iron scavenging and homeostasis: e . g . CFL5 , FET3 , FRP1 and FTR1 ( Fig 9C ) . Therefore , the demand for iron and catalase expression are intimately linked in C . albicans . Both modulate the accumulation of intracellular ROS . Iron stimulates CAT1 expression in C . albicans [16 , 61] . This increase in catalase affects iron demand and homeostasis ( Fig 9B & 9C ) and also enhances the detoxification of hydrogen peroxide , thereby decreasing the production of highly toxic hydroxyl radicals via the iron-dependent Fenton reaction [11] . Parallels exist in S . cerevisiae , where heterogeneity in superoxide dismutase ( SOD1 ) gene expression affects the fitness of individual cells in the presence of copper [62] . The impact of catalase levels on the requirement for iron is likely to have a profound effect on C . albicans pathogenicity because iron homeostasis is tightly regulated during infection [10 , 15] and efficient iron assimilation is essential for colonisation of iron limiting niches in the mammalian host [7] . It would appear significant , therefore , that we observed reduced colonisation for catalase overexpressing cells in the kidney and brain , but not in the iron-rich liver and spleen ( Fig 5 ) . In conclusion , elevated basal catalase levels appear to be a double-edged sword whereby they protect C . albicans against oxidative and combinatorial stresses imposed by the host while increasing the pathogen’s demand for an essential , but limiting micronutrient in the host . This double-edged sword would appear to account for the apparently counterintuitive observation that catalase overexpression in C . albicans decreases host colonisation in some tissues [36] . It also helps to explain why C . albicans has not evolved to express the high levels of catalase that would protect it from phagocytic killing [28 , 36] .
The strains used in this study are listed in S1 Table . C . albicans was routinely grown at 30°C , 200 rpm in YPD ( 2% dextrose , 2% mycological peptone , 1% yeast extract ) containing 20 μg/ml doxycycline ( Dox ) when required . On the day of an experiment , overnight cultures were diluted into fresh YPD to an OD600 of 0 . 2 , and incubated at 30°C at 200 rpm until they reached an OD600 of 0 . 8 , whereupon they were subjected to the appropriate treatment and analysed . Plates were incubated for 48 h at 30°C . Osmotic stress was applied with 1 M NaCl and oxidative stress was applied with H2O2 at the specified concentration . Combinatorial stress was imposed using 1 M NaCl plus 5 mM H2O2 as described previously [28 , 63] . Robotic plating was performed using a Singer RoToR robot ( Singer Instruments , Watchet , UK ) . Fitness was assayed by monitoring growth in microtitre plates at OD600 every 20 min for 48 h , and data from independent triplicate experiments were analysed . The CAT1 locus was deleted from the C . albicans strain CEC2908 using the Clox system as previously described [64] ( S2 Table ) , thereby generating the homozygous cat1Δ null mutant Ca2037 ( S1 Table ) . Using published procedures [38] , the C . albicans CAT1 ORF was then cloned into barcoded CIp10-PTET-GTw plasmids and these plasmids were integrated at the RPS1 locus in C . albicans Ca2037 ( S1 Table ) to generate the strains Ca2038 , Ca2040 , Ca2041 , Ca2043 , Ca2044 , and Ca2046 ( S1 Table ) . Empty barcoded CIp10-PTET-GTw plasmids were transformed into C . albicans CEC2908 to create strains Ca2084 , Ca2085 and Ca2087 ( S1 Table ) . Empty barcoded CIp10-PTET-GTw plasmids were also transformed into C . albicans Ca2037 to generate strains Ca2089 , Ca2092 and Ca2130 ( S1 Table ) . This created an isogenic set of nine barcoded wild-type ( CAT1 ) , null ( cat1Δ ) and tetON-CAT1 strains . Their 25 bp barcodes are described in S3 Table . The CAT1-GFP/CAT1-GFP strain Ca2213 ( S1 Table ) was constructed by PCR amplifying CAT1-GFP-URA3 and CAT1-GFP-HIS1 cassettes ( S2 Table ) [65] and integrating these sequentially at the 3’-end of the CAT1 alleles in C . albicans RM1000 ( S3 Table ) . To quantify the relative concentration of each barcoded strain in mixed populations of tetON strains , genomic DNA was prepared from the populations by phenol: chloroform extraction method [66] . A 60 bp region carrying the barcodes ( S3 Table ) was amplified with common primers ( S2 Table ) using the KAPA HiFi HotStart ReadyMix PCR Kit ( KAPA Biosystems , London , UK ) and ethanol precipitated . These purified amplicons , which contained the Illumina overhang , were then indexed with Illumina Nextera XT v2 indices ( Illumina , Inc . , San Diego , CA , USA ) . Briefly , the dual indexed Illumina libraries were prepared with 5 μl of DNA , 5 μl each of i5 and i7 index primer , 25 μl KAPA HiFi HotStart ReadyMix , and 10 μl of PCR grade water and PCR amplified ( 95°C for 3 min; 8 cycles of 95°C for 30 sec , 55°C for 30 sec and 72°C for 30 sec; 72°C for 5 min; and a final hold at 4°C ) on a Life Technologies Veriti thermal cycler ( Thermo Fisher Scientific , Waltham , MA , USA ) . The libraries were purified and size selected using a double size selection with SPRIselect ( Beckman Coulter , Brea , CA , USA ) with a SPRIselect to sample ratio of 0 . 85x followed by 1 . 0x . Libraries were quantified using the Thermo Fisher Scientific Quant-iT dsDNA High Sensitivity Assay and the fluorescence measured on a BMG Labtech FLUOstar Omega microplate reader ( BMG Labtech GmbH , Ortenberg , DE ) . The quality and size ( bp ) of the libraries were analysed on an Agilent 2200 TapeStation with High Sensitivity D1000 ScreenTapes ( Agilent Technologies , Santa Clara , CA , USA ) . The libraries were pooled in equimolar amounts and sequenced on an Illumina MiSeq Sequencing System using MiSeq v3 chemistry with 76 bp paired-end reads . Base calling and fastq output files were generated with RTA v1 . 18 . 54 software on the MiSeq instrument . To analyse the barseq data , a wrapper script was coded over the open source BBDuk tool ( BBMap suite version 35 . 43 [67] ) . The wrapper visits each sample directory and runs the 3rd-party bbduk . sh script over each of the compressed read 1 and read 2 FASTQ files , generating corresponding FASTQ output files for the “matched” and “not-matched” reads for each barcode . The wrapper then computes the total number of reads for each barcode and its abundance relative to the total number of barcode reads . The barseq data are presented as the relative abundance of a barcode normalised to its starting concentration in the population . Means and standard deviations from three replicate measurements are presented . RNA was extracted from C . albicans cells using the Zymo Research YesStar RNA Kit ( Cambridge Bioscience , Cambridge , UK ) . cDNA was prepared using SuperScript II reverse transcriptase from Invitrogen ( Fisher Scientific , Loughborough , UK ) , and qRT-PCR was performed with a Roche Light Cycler 480 II using the primers described in S2 Table . Transcript levels were measured in triplicate , expressed relative to the internal ACT1 mRNA control [28] , and then normalised against the levels in doxycycline-treated wild type ( CAT1 ) cells to exclude potential effects of doxycycline on these transcripts . C . albicans cells grown in YPD containing 0 or 20 μg/μl Dox were subjected to no stress or one hour of 5 mM H2O2 , protein extracts prepared , and catalase activities measured using the BioAssay Systems EnzyChrom catalase assay kit ( Universal Biologicals Ltd . , Cambridge , UK ) , according to the manufacturer’s instructions [28] . Assays were performed in triplicate . CAT1-GFP and ACT1-GFP expression in C . albicans cell populations was examined and cell subsets isolated using the BD Influx cell sorter . Heterogeneity in C . albicans cell size was first analysed ( Forward Scatter ( FSC ) , Side Scatter ( SSC ) ) and cells of similar size selected ( S2 & S3 Figs ) . Cells were then sorted on the basis of their GFP expression level ( S2B Fig ) . Cells ( n = 200 ) that expressed GFP at relatively low levels and 200 cells expressing GFP at high levels were plated onto YPD containing various concentrations of H2O2 . These two populations sorted were separated to 99% purity . Control experiments were performed to confirm cell viability by propidium iodide staining ( 2 μg/ml ) . Data were analysed using BD FACS software and Flowjo software version 10 . 0 . 8 . CAT1-GFP cells were visualized using a DeltaVision Core microscope ( Applied Precision , Issaquah , WA ) . Western blotting was performed as described previously [68] . Cell viability was assayed by measuring colony forming units ( CFU ) on YPD plates and by propidium iodide ( PI ) staining and flow cytometry on a BD LSR II , as described previously [28 , 63] . Intracellular ROS accumulation was measured by staining the cells with 20 μM dihydroethidium for one hour in darkness , at 30°C and 200 rpm , and then analysed using a BD LSR II flow cytometer . Data were analysed using Flowjo software version 10 . 0 . 8 . Blood from healthy donors was obtained according to the protocol approved by the University of Aberdeen College Ethics Review Board ( Application number—CERB/2012/11/676 ) . Polymorphonuclear ( PMN ) cells , or neutrophils , were isolated from this blood using Histopaque-1119 and Histopaque-1077 ( Sigma Aldrich ) as described previously [28] . C . albicans cells pre-grown with 20 μg/ml Dox were incubated with PMNs ( 1:10 ratio of yeasts to neutrophils ) for 2 h in RPMI 1640 containing 10% heat inactivated foetal bovine serum . After incubation the PMNs were treated with 0 . 25% sodium docecyl sulphate and DNase I and yeast survival determined by assaying CFU . Data from eight healthy donors are presented with their means and standard deviation . The virulence of C . albicans wild type and cat1Δ cells were measured in a short term murine model of systemic candidiasis [47] . Strains were pre-grown in YPD and injected intravenously ( 4 x 104 CFU/g body weight ) into the lateral tail vein of 6–10 week old female BALB/c mice ( Envigo , UK ) . Mice were randomly assigned to cages ( n = 6 per group ) and inocula were randomly assigned to cages . Infections were allowed to proceed for 4 days whereupon the mice were humanely culled by cervical dislocation and fungal burdens ( CFU/g ) determined in the kidneys . Fungal burden and weight loss were used as measures of virulence [47] . The virulence of C . albicans wild type and cat1Δ strains were also tested in a longer term mouse infection model . Again , C . albicans cells were injected into the tail veins of 6–10 week old female BALB/c mice ( 3 x 104 CFU/g body weight ) . Once again , the mice were randomly assigned to cages ( n = 8 per group ) and inocula were assigned randomly to cages . The mice were monitored and weighed daily , and were humanely culled when they had lost 20% of their body weight and death recorded as having occurred on the following day . Experiments were continued for a maximum of 14 days , when all surviving mice were culled and analysed . The data are presented as Kaplan-Meier survival curves ( log rank tests ) . To directly compare the colonisation of C . albicans tetON strains in the mouse model of systemic candidiasis , the strains were pre-grown in YPD containing 0 or 20 μg/ml doxycycline and injected into the tail vein of 6–10 week old female BALB/c mice ( 4 x104 CFU/g body weight: n = 6 mice per group ) . Mice were gavaged with 100 μl of 0 or 40 mg/ml doxycycline . Infections were allowed to proceed for up to 4 days . Mice were culled , their kidneys , spleen , liver and brain removed and homogenized in 500 μl saline , and the entire sample from each organ plated onto YPD . The fungal colonies from each individual organ were then pooled , and genomic DNA prepared for barseq ( above ) . The virulence of C . albicans strains was also evaluated using the invertebrate Galleria mellonella infection model [69] . For each C . albicans strain , 105 cells were injected into 20 Galleria larvae ( 6th instar: BioSystems Technology , Exeter , UK ) . Sterile PBS was injected into control larvae . Survival was monitored for 5 days at 37°C , represented using Kaplan-Meier curves , and analysed using log rank tests . All animal experiments were conducted in compliance with United Kingdom Home Office licenses for research on animals , and were approved by the University of Aberdeen Ethical Review Committee ( project license number PPL 70/8583 ) . Animal experiments were minimised , and all animal experimentation was performed using approaches that minimised animal suffering and maximised our concordance with EU Directive 2010/63/EU . Power analyses based on data generated in previous experiments were applied to estimate the minimum number of animals per group required to achieve statistically robust differences ( P <0 . 05 ) . The power analyses to determine group size for the short term systemic infection model were based on the variation in fungal burdens between animals , whereas those for the long term model were based upon mean survival times . Animals were monitored at least twice daily for signs of distress , which was minimised by expert handling . Euthanasia was performed humanely by cervical dislocation when animals showed signs of progressive illness ( e . g . ruffled coat , hunched posture , unwillingness to move and 20% loss of initial body weight ) . During these studies there were no unexpected deaths . Analgesia and anaesthesia were not required in this study . Statistical analyses were performed in GraphPad Prism 5 and IBM SPSS Statistics ( v24 . 0 . 0 ) . Two tailed Mann-Whitney U analysis was used to test the statistical difference between two sets of data with a non-parametric distribution . Associations between growth parameters , such as doubling time , lag phase or propidium iodide staining , were determined by one-way and two-way ANOVA and Dunnett post-hoc t-tests . Unstressed samples were used as controls and the values of other samples were compared against these controls . The following p-values were considered: * p < 0 . 05; ** p <0 . 01; *** p < 0 . 001; **** p < 0 . 0001 .
|
The pathogenic yeast Candida albicans faces multiple challenges within its human host . These include the need to protect itself against the toxic oxidants used by the host to kill invading microbes , and the need to scavenge iron , an essential micronutrient that is limiting in certain tissues . The iron-containing enzyme , catalase , detoxifies hydrogen peroxide , thereby playing a major role in protecting C . albicans against reactive oxygen species and neutrophil killing . Indeed , we show that high basal catalase expression increases the resistance of this yeast to oxidative and combinatorial ( oxidative plus cationic ) stresses . Yet , rather than enhancing the virulence of C . albicans as had been predicted , high basal catalase expression decreases fungal colonisation in certain iron-limiting tissues . Furthermore , we demonstrate that catalase inactivation does not significantly perturb the virulence of C . albicans in models of systemic infection . We also show that ectopic catalase expression increases the demand for iron in C . albicans , thereby reducing the fitness of this pathogen in the absence of stress under iron-limiting conditions . Therefore , high basal catalase expression is a double-edged sword: it enhances the fitness of C . albicans in the presence of stress , but reduces fitness in the absence of stress . This explains why catalase overexpression reduces rather than enhances virulence .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"chemical",
"compounds",
"oxides",
"oxidative",
"stress",
"enzymes",
"pathogens",
"immunology",
"microbiology",
"enzymology",
"animal",
"models",
"fungi",
"model",
"organisms",
"experimental",
"organism",
"systems",
"kidneys",
"fungal",
"pathogens",
"neutrophils",
"research",
"and",
"analysis",
"methods",
"mycology",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"chemistry",
"mouse",
"models",
"catalases",
"yeast",
"biochemistry",
"candida",
"cell",
"biology",
"anatomy",
"biology",
"and",
"life",
"sciences",
"yeast",
"and",
"fungal",
"models",
"physical",
"sciences",
"cellular",
"types",
"renal",
"system",
"peroxides",
"organisms",
"candida",
"albicans"
] |
2017
|
Elevated catalase expression in a fungal pathogen is a double-edged sword of iron
|
Trachoma , caused by ocular infection with Chlamydia trachomatis , is hyperendemic on the Bijagós Archipelago of Guinea Bissau . An understanding of the risk factors associated with active trachoma and infection on these remote and isolated islands , which are atypical of trachoma-endemic environments described elsewhere , is crucial to the implementation of trachoma elimination strategies . A cross-sectional population-based trachoma prevalence survey was conducted on four islands . We conducted a questionnaire-based risk factor survey , examined participants for trachoma using the World Health Organization ( WHO ) simplified grading system and collected conjunctival swab samples for 1507 participants from 293 randomly selected households . DNA extracted from conjunctival swabs was tested using the Roche Amplicor CT/NG PCR assay . The prevalence of active ( follicular and/or inflammatory ) trachoma was 11% ( 167/1508 ) overall and 22% ( 136/618 ) in 1–9 year olds . The prevalence of C . trachomatis infection was 18% overall and 25% in 1–9 year olds . There were strong independent associations of active trachoma with ocular and nasal discharge , C . trachomatis infection , young age , male gender and type of household water source . C . trachomatis infection was independently associated with young age , ocular discharge , type of household water source and the presence of flies around a latrine . In this remote island environment , household-level risk factors relating to fly populations , hygiene behaviours and water usage are likely to be important in the transmission of ocular C . trachomatis infection and the prevalence of active trachoma . This may be important in the implementation of environmental measures in trachoma control .
Trachoma is caused by ocular infection with Chlamydia trachomatis and is the leading infectious cause of blindness worldwide . It manifests as distinct clinical syndromes beginning with an acute self-limiting keratoconjunctivitis , which following repeated episodes may progress to a more chronic inflammatory and immunofibrogenic process leading to conjunctival scarring and blinding sequelae . Trachoma is endemic in 50 countries , with 325 million people at risk of blinding disease [1] . Trachoma is responsible for visual impairment in 1 . 2 million people and 3% of blindness globally [1] . The highest prevalence of active trachoma ( trachomatous inflammation-follicular ( TF ) and/or trachomatous inflammation-intense ( TI ) ) is in sub-Saharan Africa and the distribution of disease is heterogeneous [2] . Ocular C . trachomatis is probably transmitted between individuals through direct spread from eye to eye during close contact , direct or indirect spread of infected nasal or ocular secretions on fingers or cloths ( fomites ) and indirect passive transmission by eye seeking flies . There is no known animal reservoir of C . trachomatis in endemic environments , the primary reservoir being young children . Blinding trachoma is usually found in hot , arid , dusty regions . A recent systematic review examined studies reporting higher trachoma prevalence in savannah areas and areas of lower rainfall , and found weak but consistent evidence supporting anecdotal findings that trachoma is associated with semi-arid environments [3] . This study was conducted on the Bijagós Archipelago , a remote group of islands off the coast of Guinea Bissau with a total population estimated at 24 , 000 [4] , where trachoma is hyperendemic . The climate and environment are not typical of trachoma-endemic areas . The islands are covered with subtropical forest and altitude does not exceed 50 m . The climate is tropical , hot and humid . The islands are surrounded by mangroves and mudflats . There is significant rainfall ( average 400 mm/month ) from May to November [5] . Many studies have suggested that the prevalence of trachoma is associated with environmental risk factors such as poor sanitation , access to water and latrine use [6] , [7] . Eye-seeking flies ( Musca sorbens ) have also been associated with trachoma as passive vectors [8] but significant disease exists in areas where fly populations are scarce and are therefore less likely to contribute to trachoma transmission [9] . M . sorbens preferentially breeds in human faeces and there may be association between fly populations and lack of latrine access or use [6] , [8] . Social risk factors such as migration events and crowded living conditions have also been shown to be important in transmission of C . trachomatis and the appearance of active trachoma [10] , [11] . Clustering of disease at the community , household and bedroom levels has been noted and is likely to reflect the dynamics of transmission between family members with prolonged close contact [6] , [10]–[12] . Most transmission events have been shown to occur at the household level with more gradual spread within the community [13] . The World Health Organization ( WHO ) advocates the implementation of the SAFE strategy ( Surgery for trichiasis , Antibiotics for active infection , Facial cleanliness to prevent disease transmission and Environmental improvement to increase access to water and sanitation ) for trachoma elimination . The WHO recommends annual mass treatment of entire communities with oral azithromycin for three years if the prevalence of TF in 1–9 year olds within a district or community exceeds 10% . Mass antibiotic treatment aims to clear infection from communities such that transmission ceases to be a public health concern [14] . Following this , an assessment is made of A , F and E interventions and a decision is taken to continue or cease treatment [15] . Despite their inclusion in the SAFE strategy , local environmental factors are not well understood , though many are potentially modifiable risk factors for infection and disease . The relative importance of these risk factors is not clear and may differ between communities . Fewer studies have investigated risk factors for disease and infection simultaneously [16]–[19] . Understanding risk factors associated with trachoma and C . trachomatis infection may increase our understanding of disease and transmission dynamics allowing for optimization of community-specific interventions . We examined household and individual-level risk factor associations with ocular C . trachomatis infection and active trachoma in this unique environment , where trachoma is a significant public health problem . Prior to these surveys , these communities were treatment-naïve and had not been exposed to any trachoma control interventions .
This study was conducted in accordance with the declaration of Helsinki . Ethical approval was obtained from the Comitê Nacional de Ética e Saúde ( Guinea Bissau ) , the LSHTM Ethics Committee ( UK ) and The Gambia Government/MRC Joint Ethics Committee ( The Gambia ) . Verbal consent was obtained from community leaders . Written informed consent was obtained from all study participants or their guardians on their behalf if participants were children . A signature or thumbprint is considered an appropriate record of consent in this setting by the above ethical bodies . We conducted a cross-sectional population-based trachoma prevalence survey on four islands of the Bijagós Archipelago of Guinea Bissau ( Bubaque , Canhabaque , Soga and Rubane ) in January 2012 . Trachoma survey methodology has been described previously [20]–[22] . We randomly sampled one in five households , representing a one stage probability sample design satisfying desired criteria for population-based prevalence surveys [20] , [21] . A sample size of 1500 ensured adequate power with conservative correction ( using a design effect of 4 ) to account for anticipated household clustering . The sample size provides over 90% power to detect an odds ratio ( OR ) of 2 associated with a risk factor found in 20% of subjects without disease or infection , or an OR of 3 for a risk factor present in 5% of subjects without disease or infection with 95% confidence . The sample size also provides good precision for an estimated TF prevalence of >25% in 1–9 year olds on the four islands of Bubaque and Canhabaque ( ±4% ) , Soga ( ±6% ) and Rubane ( ±10% ) , which is adequate to determine whether these communities require mass drug treatment with azithromycin in line with WHO policy . A census of persons resident in randomly selected households was conducted prior to the household survey . Residency was defined as living within the household for longer than the preceding month or intending to stay resident in the household for longer than one month . This was updated to reflect the de facto population ( those present in the household on the previous night ) to limit absenteeism . Demographic , socio-economic and environmental information was collected at household and individual levels . Household-level risk factor data were obtained using questionnaires administered to the household head or an appropriate responsible adult and included items on the level of education of the household head , their socio-economic status , whether the household had been exposed to any health education or promotion within the community , household access to and use of latrines , access and use of water and measures of sanitation , waste and presence of flies in the environment . The questionnaire was supported through observational data collected on water use , latrine use and environmental sanitation . Household size ( measured as number of members of all ages ) and number of children under the age of 10 years within the household was recorded . Researchers were masked to trachoma status of household members at the time of the household survey . Following the household risk factor survey all individuals from study households were invited to attend for clinical examination and conjunctival sampling . Individuals' age , sex and ethnic group and data on facial cleanliness ( the presence of ocular and/or nasal discharge and whether or not there were flies on the face ) were collected at the time of examination . A single trained examiner assessed each participant using the WHO simplified grading system where TF ( trachomatous inflammation – follicular ) and/or TI ( trachomatous inflammation – intense ) constitute active trachoma and TS ( trachomatous scarring ) , TT ( trachomatous trichiasis ) and CO ( corneal opacity ) are trachomatous sequelae which may lead to blindness [23] . A trachoma grade was assigned to the upper tarsal conjunctivae of each consenting participant using adequate light and a 2 . 5× binocular magnifying loupe . Two sequential samples were taken from the left upper tarsal conjunctiva of each participant with Dacron swabs ( Fisher Scientific , UK ) using a standardised procedure [24] , [25] . The first swab was collected into transport medium for other studies . The second dry swab was collected into a microcentrifuge tube ( Simport , Canada ) and used in this study . Previous work using the Roche Amplicor CT/NG assay ( Roche Molecular Systems , NJ USA ) in a population-based study has shown that there was good agreement between first and second swabs with respect to C . trachomatis DNA positivity by PCR [26] . Swabs were kept on ice in the field and frozen to −80°C within 8 hours of collection . Measures were taken to avoid cross-contamination in the field . Control swabs ( pre-marked swabs drawn at random from the swab dispenser and passed 10 cm in front of the eye ensuring no contact between the swab tip and participant ) were taken to ensure field and laboratory quality control . After survey completion all communities on the study islands were treated with a single height-based dose of oral azithromycin in accordance with WHO and national protocols . Each swab was suspended in 400 µl sterile phosphate buffered saline ( PBS ) after thawing at room temperature . DNA was extracted from the swab/PBS suspension using an adapted whole blood protocol on the QIAxtractor ( Qiagen , Crawley , UK ) automated instrument and eluted into a final volume of 50 µl DX Elution Buffer ( Qiagen ) . C . trachomatis DNA was detected using the Roche Amplicor CT/NG assay ( validated for use with ocular swabs [27] ) . Required reaction buffer conditions were obtained as described previously and used in the standard assay [28] . Positive and negative samples were assigned according to the manufacturer's instructions . In this study , C . trachomatis infection is defined as the presence of C . trachomatis DNA by Amplicor PCR . Data were double entered into a customised database ( MS Access 2007 ) . Discrepancies were resolved through reference to original data forms . Data were further cleaned prior to analysis in STATA 13 ( Stata Corporation , College Station , Texas USA ) . Random effects logistic regression models were used to assess the variability between villages and households assuming a three tier hierarchy to the data ( at village , household and individual levels ) . Null models were used to examine the effect of cluster variables on the outcome using the likelihood ratio test ( LRT ) , which if significant , provided strong evidence that between-village and household variance was non-zero . The log likelihood and the LRT were used to compare models . Univariable associations with active trachoma ( TF/TI ) and infection with C . trachomatis were examined using two-level hierarchical random effects logistic regression , accounting for between-household variation . Covariates associated with active trachoma or C . trachomatis infection with p<0 . 10 ( using the Wald test ) were sequentially added to the multivariable model after a priori adjustment for age and gender ( as categorical variables ) . Covariates were retained in the final model if the Wald p-value≤0 . 05 unless otherwise specified . Further exploration of environmental predictor variables was conducted using logistic and hierarchical random effects logistic regression models as appropriate using the same criteria . As C . trachomatis infection is on the causal pathway between several risk factors and active trachoma , models with and without C . trachomatis infection were fitted . The model including C . trachomatis infection provides estimates of independent associations of other risk factors with active trachoma which are not mediated through C . trachomatis infection . All statistical analyses were carried out using STATA 13 . Statistical significance was determined at the 5% level .
From an estimated total rural population of 5 , 613 inhabitants on the four study islands [4] , 1 , 511 individuals from 293 randomly selected households across 39 villages were enrolled . Of these , 1 , 508 had an ocular assessment and conjunctival swabs were obtained from 1 , 507 . The median age of participants was 13 years ( range 1 month–88 years ) and 57% were female . The majority of participants were of the Bijagós ethnic group ( Table 1 ) . The prevalence of active trachoma in 1–9 year olds was 22 . 0% ( 95% Confidence Interval ( CI ) 18 . 9–25 . 5% ) ( 136/618 ) . The prevalence of active trachoma was highest in children under the age of 5 years ( 27 . 3% ( 95% CI 23 . 1–31 . 9% ) ( 113/416 ) ) . Overall , 11 . 1% ( 95% CI 9 . 4–12 . 6% ) ( 167/1508 ) of the study population had active trachoma . The relationship between trachoma and infection is shown in Table 2 . C . trachomatis DNA was detected in 18 . 0% overall ( 269/1507 ) and 25 . 4% of 1–9 year olds ( 157/618 ) . All 15 ( ∼1% of total ) control swabs were negative for C . trachomatis DNA . Null models for both active trachoma and C . trachomatis infection adjusted for age and gender showed significant clustering at island , village and household levels . For active trachoma , the variance estimated due to between-household clustering was 1 . 11 ( standard error ( SE ) 0 . 17 , p<0 . 0001 ) . The between-village clustering variance was 0 . 75 ( SE 0 . 16 , p<0 . 0001 ) and between-island clustering variance was 0 . 50 ( SE 0 . 28 , p = 0 . 0100 ) . For C . trachomatis infection , the variance estimated due to between-household clustering was 1 . 37 ( SE 0 . 15 , p<0 . 0001 ) , between-village clustering was 0 . 89 ( SE 0 . 14 , p<0 . 0001 ) and between-island clustering was 0 . 40 ( SE 0 . 18 , p = 0 . 0005 ) . The clustering effect was strongest at household level and models adjusting for clustering at household level only were a better fit than those including adjustment for village and island clustering . Adjusting for clustering at household level significantly improved the model versus standard logistic regression analyses ( p<0 . 0001 ) . Two-level hierarchical regression models with adjustment for household level clustering are presented in this analysis . Univariable associations with active trachoma are presented in Table 3 . The final multivariable model showed that active trachoma was strongly independently associated with C . trachomatis infection ( OR = 11 . 2 ( 95% CI 6 . 9–18 . 1 ) ) , ocular ( OR = 2 . 0 ( 95% CI 1 . 0–4 . 0 ) ) and nasal ( OR = 2 . 5 ( 95% CI 1 . 5–4 . 3 ) ) discharge , male gender ( OR = 1 . 9 ( 95% CI 1 . 2–2 . 9 ) ) and being aged 0–5 years ( OR = 10 . 2 ( 95% CI 5 . 1–20 . 4 ) compared to being >15 years of age ) ( Model 2 , Table 4 ) . There was also a strong independent association between household water access and active trachoma , such that households with access only to a traditional natural spring as a water source had an increased risk of active trachoma compared to households with access to multiple water sources ( OR = 1 . 9 ( 95% CI 0 . 9–3 . 9 ) ) . The model without C . trachomatis infection shows stronger associations , indicating that some effect of these factors is mediated through C . trachomatis infection ( Table 4 ) . Comparison of the two models suggests that some of the effect of younger age and water source is partly mediated through C . trachomatis infection , but these remain independently associated with trachoma beyond this effect . Univariable associations with C . trachomatis infection are presented in Table 5 . In the final multivariable model C . trachomatis infection was strongly independently associated with being aged ≤10 years . The presence of ocular discharge ( OR = 2 . 3 ( 95% CI 1 . 3–4 . 4 ) ) and household access only to a traditional natural spring ( OR = 6 . 6 ( 95% CI 2 . 8–15 . 2 ) ) and or access to a single water source only ( OR = 3 . 9 ( 95% CI 1 . 9–8 . 0 ) ) ( rather than households who had access to multiple water sources ) were strongly associated with infection ( Table 6 ) . The presence of flies around a latrine was also independently associated with infection ( OR = 2 . 1 ( 95% CI 1 . 1–3 . 8 ) ) . The presence of flies around a latrine were strongly associated with the presence of flies in the environment surrounding the household ( OR = 8 . 3 ( 95% CI 5 . 4–12 . 7 ) , p<0 . 0001 ) and the presence of visible faeces within the latrine ( OR = 46 . 7 ( 95% CI 28 . 5–76 . 6 ) , p<0 . 0001 ) . There was no association between flies in the environment ( OR = 1 . 1 ( 95% CI 0 . 4–3 . 0 ) , p = 0 . 91 ) nor flies around the latrine ( OR = 0 . 5 ( 95% CI 0 . 1–2 . 6 ) , p = 0 . 43 ) with flies on the face at the time of examination .
We have described individual and household-level risk factor associations with active trachoma and ocular infection with C . trachomatis on the Bijagós Archipelago to improve our understanding of the relationship between disease and infection in this remote treatment-naïve trachoma-hyperendemic population . These data suggest that in this environment household-level risk factors relating to fly populations , hygiene behaviours and water usage are likely to be important in the transmission of ocular C . trachomatis infection . Education about cleanliness , sanitation and hygiene practices is likely to be important in reducing transmission of infection in these communities . Ensuring the provision of water sources which allow adequate water to be allocated for hygiene may assist this , and further studies examining specific hygiene practices may be useful . Reducing fly populations around the latrines where they exist may be of benefit . These findings may be important in the implementation of the F and E components of SAFE in this population . In order to fully understand the factors associated with active trachoma and ocular C . trachomatis infection in these communities , further epidemiological studies examining transmission and clustering of C . trachomatis infection are required . These studies should focus on pathogen factors such as the role of infection intensity and strain diversity , and socio-behavioural factors such as specific hygiene behaviours .
|
Trachoma , caused by ocular infection with Chlamydia trachomatis , is the leading infectious cause of blindness worldwide . The World Health Organization elimination strategy includes community mass treatment with oral antibiotics , education regarding hygiene and facial cleanliness and environmental improvements . Population-based trachoma prevalence surveys are essential to determine whether community interventions are required . Knowledge of risk factors associated with trachoma and C . trachomatis infection in a particular setting may help prioritise trachoma elimination activities . We conducted a trachoma prevalence survey to establish the prevalence of active ( follicular and/or inflammatory ) trachoma and C . trachomatis infection on the Bijagós Archipelago of Guinea Bissau . We also collected household risk factor data from survey participants . Active trachoma prevalence was 11% overall and 22% in children aged 1–9 years . C . trachomatis infection prevalence was 18% overall and 25% in children aged 1–9 years . Active trachoma and the presence of C . trachomatis infection were strongly correlated . Risk factors for disease and infection were similar . In this environment , measures of facial cleanliness ( ocular and nasal discharge ) and household-level risk factors relating to fly populations , hygiene behaviours and water usage are likely to be important in C . trachomatis transmission . This may have implications in the implementation of trachoma elimination activities .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"medicine",
"and",
"health",
"sciences",
"global",
"health",
"epidemiology"
] |
2014
|
Risk Factors for Active Trachoma and Ocular Chlamydia trachomatis Infection in Treatment-Naïve Trachoma-Hyperendemic Communities of the Bijagós Archipelago, Guinea Bissau
|
Miniature inverted-repeat transposable elements ( MITEs ) are numerically predominant transposable elements in the rice genome , and their activities have influenced the evolution of genes . Very little is known about how MITEs can rapidly amplify to thousands in the genome . The rice MITE mPing is quiescent in most cultivars under natural growth conditions , although it is activated by various stresses , such as tissue culture , gamma-ray irradiation , and high hydrostatic pressure . Exceptionally in the temperate japonica rice strain EG4 ( cultivar Gimbozu ) , mPing has reached over 1000 copies in the genome , and is amplifying owing to its active transposition even under natural growth conditions . Being the only active MITE , mPing in EG4 is an appropriate material to study how MITEs amplify in the genome . Here , we provide important findings regarding the transposition and amplification of mPing in EG4 . Transposon display of mPing using various tissues of a single EG4 plant revealed that most de novo mPing insertions arise in embryogenesis during the period from 3 to 5 days after pollination ( DAP ) , and a large majority of these insertions are transmissible to the next generation . Locus-specific PCR showed that mPing excisions and insertions arose at the same time ( 3 to 5 DAP ) . Moreover , expression analysis and in situ hybridization analysis revealed that Ping , an autonomous partner for mPing , was markedly up-regulated in the 3 DAP embryo of EG4 , whereas such up-regulation of Ping was not observed in the mPing-inactive cultivar Nipponbare . These results demonstrate that the early embryogenesis-specific expression of Ping is responsible for the successful amplification of mPing in EG4 . This study helps not only to elucidate the whole mechanism of mPing amplification but also to further understand the contribution of MITEs to genome evolution .
Transposable elements ( TEs ) are DNA sequences that are capable of jumping from one genomic locus to another and make up a large fraction of eukaryotic genomes . More than 80% of the maize ( Zea mays ) and barley ( Hordeum vulgare ) genomes are composed of TEs [1] , [2] , and they constitute 35% and 14% of the genomes of rice ( Oryza sativa ) and Arabidopsis ( Arabidopsis thaliana ) , respectively [3] , [4] . TEs are harmful to the host because their mobilities perturb genome stability , whereas they play greatly generative roles in genome evolution such as alternation of gene structure , change of expression pattern , and rearrangement of chromosome structure [5] , [6] . TEs are classified into two groups according to their transposition mechanisms: class I elements ( retrotransposons ) that transpose through a copy-and-paste mechanism via an RNA intermediate , and class II elements ( transposons ) that transpose through a cut-and-paste mechanism without undergoing an RNA intermediate . Class I elements easily attain tens of thousands of copies , whereas the majority of class II elements cannot amplify themselves to 50 copies at most . Unlike other class II elements , miniature inverted-repeat transposable elements ( MITEs ) have the capacity to amplify themselves to high copy numbers ( hundreds or thousands ) [7]–[9] . In the rice genome , MITEs are numerically predominant TEs [10] , constituting 8 . 6% of the genome [11] . Because MITEs are too short ( <600 bp ) to encode any protein , their transpositions must depend on the proteins encoded by the autonomous elements . Well-studied MITEs are classified into the Stowaway and Tourist families , which belong to the Tc1/mariner and PIF/Harbinger superfamilies , respectively . Because MITEs are mainly deployed in gene-rich regions [10] , [12] and affect adjacent gene expression [13] , they are considered to play an important role in genome evolution . However , little is known about how MITEs attain high copy numbers . Miniature Ping ( mPing ) is the first active MITE identified in the rice genome [14]–[16] . Although MITEs are deployed in the genome at a high copy number , the copy number of mPing exceptionally remains at a low level in most rice cultivars: indica and tropical japonica cultivars have fewer than 10 copies , and temperate japonica cultivars including Nipponbare have approximately 50 copies [14] . The transposition of mPing is suppressed in most rice cultivars , but , like other TEs , mPing is activated by exposure to various stress conditions such as gamma-ray irradiation [16] , hydrostatic pressurization [17] , cell culture [14] , anther culture [15] , and inhibition of topoisomerase II [18] . Introgression of distantly related genomes also causes mPing transposition [19] , [20] . However , mPing is actively transposing without such stresses in the temperate japonica rice strain EG4 ( cultivar Gimbozu ) under natural growth conditions , and its copy number is approximately 1000 copies [21] . This indicates that mPing has overcome the silencing mechanism or established a novel strategy for its amplification in the EG4 genome . In this sense , mPing in EG4 is an appropriate material to study the amplification of MITEs in plant genomes . The autonomous element Ping and its distantly related element Pong , which both belong to the PIF/Harbinger superfamily , provide two proteins required for mPing transposition . Both Ping and Pong have two open reading frames ( ORFs ) , ORF1 and ORF2 [22] , [23] . The former encodes a Myb-like DNA-binding protein , and the latter encodes a transposase lacking DNA binding domain . Transposase of most class II elements contains a conserved catalytic domain ( DDE motif ) and a DNA-binding domain [23] , [24] , whereas these domains are encoded separately by two ORFs in both Ping and Pong [22] , [23] . The study of other members of the PIF/Harbinger superfamily suggested that the Myb-like DNA-binding protein directly binds to the subterminal regions of the transposon in order to recruit the transposase [25] . Both Myb-like protein and transposase of either Ping or Pong or both elements are necessary for mPing transposition [22] , [23] . In this study , we demonstrate that mPing is actively transposing in the embryo of EG4 during the period from the regionalization of shoot apical meristem ( SAM ) and radicle to the formation of the first leaf primordium ( 3 to 5 days after pollination , DAP ) with the aid of developmental stage-specific expression of Ping . Our results provide important evidence for the amplification mechanism not only of mPing but also of other MITEs .
Plants have acquired the silencing mechanism of TEs in germ cells . In Arabidopsis , for example , TEs are activated specifically in the vegetative nucleus of the pollen , and siRNAs from the activated TEs accumulate in the sperm cells [26] . On the basis of these results , Slotkin and colleagues proposed that siRNAs derived from TEs activated in the vegetative nucleus silence TEs in the sperm cells [26] . We conceived that mPing might overcome such a silencing mechanism in EG4 . To confirm this hypothesis , we developed two F1 populations from reciprocal crosses between the mPing-active strain EG4 and the mPing-inactive cultivar Nipponbare , and investigated the transposition activity of mPing by transposon display ( TD ) analysis . Success of reciprocal crosses was confirmed by PCR analysis using locus-specific primers ( Figure S1A ) . One of the results of TD analysis using two selective bases is shown in Figure 1A; all 16 possible primer combinations were analyzed . The banding patterns of F1 plants were almost the same as those of EG4 . The bands that appeared in all F1 plants but not in the parental EG4 plant were derived from another parental Nipponbare plant ( Figure S1B ) . Furthermore , the bands that appeared in only one of eight F1 plants but not in the parental EG4 plant are herein referred as de novo insertions . These bands were confirmed not to be PCR artifacts by sequence and locus-specific PCR analysis ( Table S1 and Figure S2 ) . We detected 15 . 5 de novo insertions per plant in the selfed progenies of EG4 , whereas Nipponbare yielded no de novo insertions in the selfed progenies ( Figure 1B ) . This confirmed that mPing is active in EG4 under natural growth conditions but inactive in Nipponbare . If mPing was specifically activated in the pollen of EG4 , we could obtain de novo insertions only in the F1 plants from the Nipponbare/EG4 cross . However , we obtained de novo insertions in both Nipponbare/EG4 and EG4/Nipponbare populations ( Figure 1B ) . Moreover , there was no significant difference in the number of de novo insertions per plant between the two F1 populations . This indicates that the activating factor ( s ) for the mPing transposition is present in both male and female gametes of EG4 . We performed TD analysis of mPing using genomic DNA samples extracted from endosperm , radicle , and leaf blades of eight progenies ( S1 ) derived from a single parental EG4 plant ( S0 ) , and investigated the mPing transposition during ontogeny of rice plants ( Figure 2A ) . One of the results of TD analysis using two selective bases is shown in Figure S3; all 16 possible primer combinations were analyzed . We examined de novo insertions in the same way as described above . Consequently , a total of 228 de novo insertions were detected . These insertions were divided into five groups ( Figure 2B ) : ( 1 ) endosperm-specific insertions that appeared only in the endosperm sample , ( 2 ) radicle-specific insertions that appeared only in the radicle sample , ( 3 ) leaf-specific insertions that appeared only in one sample from the 1st to 3rd leaf blades , but not in the 4th and 5th leaf blades , ( 4 ) shoot-specific insertions that appeared in at least one sample of 1st , 2nd , and 3rd leaf blades , and in at least one sample of 4th and 5th leaf blades , and ( 5 ) radicle/shoot-specific insertions that appeared in both radicle and leaf blade samples . These de novo insertions were confirmed by sequence and locus-specific PCR analysis ( Table S2 and Figure S4 ) . Numbers of each insertion obtained in this study are summarized in Figure 2C . Plant development is divided roughly into three successive phases: embryogenesis , vegetative phase , and reproductive phase . If mPing transposed in the SAM of the S0 plant during vegetative and/or reproductive phases , the de novo insertions would segregate according to Mendel's law among the S1 progenies . We obtained no band that appeared in at least two S1 progenies and was not seen in the S0 plant . This indicates that the transmissible insertion of mPing to the next generation seldom ( or never ) arises during the vegetative and reproductive phases . Flowering plants have evolved a unique reproductive process called double fertilization . In this process , either of two sperm cells in pollen fuses with either an egg cell or a central cell in the ovule , and then the egg cell fertilized with the sperm cell initiates embryogenesis [27] . In rice , the SAM and radicle are regionalized in the embryo 3 DAP , and three leaves and the radicle are already present in the mature embryo [28] . We detected only three radicle/shoot-specific insertions ( Figure 2C ) , indicating that mPing scarcely transposes during the period from the onset of gametogenesis to the early stage ( until 3 DAP ) of embryogenesis . Among the 228 de novo insertions , 116 and 17 were shoot-specific and leaf-specific insertions , respectively ( Figure 2C ) . This indicates that mPing actively transposes in the embryo during the period from the regionalization of SAM and radicle ( at 3 DAP ) to the formation of the 3rd leaf primordia ( at 8 DAP ) . Of the 133 shoot- and leaf-specific insertions , 108 were of the 1st leaf blade ( Figure 2D ) . Since the 1st leaf primordium is formed at 5 DAP , the most active phase of the mPing transposition was considered to be from 3 to 5 DAP . We detected a large number of radicle-specific insertions as well as shoot-specific insertions , and the sum of these insertions accounted for 90% of all insertions detected in this study ( Figure 2C ) . Taken together , we concluded that mPing in EG4 most actively transposes in the 3 to 5 DAP embryo . Endosperm is a triploid tissue that is produced by fusing a central cell containing two polar nuclei with one of two sperm cells in no particular order . The endosperm formation occurs in parallel with embryogenesis . The endosperm-specific insertions result from the mPing transposition occurred in either gametogenesis or endosperm formation . We observed only two endosperm-specific insertions ( Figure 2C ) , supporting that mPing scarcely transposes during the period from the onset of gametogenesis to the early stage of embryogenesis . The relationship between the banding patterns obtained in TD analysis and the timing of mPing transposition is summarized in Figure S5 . In order for mPing to amplify , the de novo insertions must be transmitted to the next generation . We performed TD analysis using 12 progenies ( S2 ) derived from the main culm and the primary tiller of a single selfed parent ( S1 ) to investigate whether the de novo insertions detected in ontogenical analysis are inheritable ( Figure S6 ) . Both radicle-specific and leaf-specific insertions in the S1 plants were not detected in the S2 progenies ( 0 of 15 , 0 of 2 , respectively ) . In contrast , 85% ( 11 of 13 ) of the shoot-specific insertions that were detected in the S1 plants also appeared in the S2 progenies . This value ( 85% ) is consistent with the estimated number of inheritable de novo insertions in our previous report [21] . Thus most of the de novo insertions that arose in the 3 to 5 DAP embryo were successfully inherited to the next generation . We have already determined the sites of all mPing insertions ( 1163 in total ) in the EG4 genome [13] , and have investigated mPing excisions in a small EG4 population using locus-specific primer pairs [29] , [30] . Here we examined the timing of the mPing excision with locus-specific PCR using the genomic DNA samples that were used for the ontogenical analysis of the de novo insertion . We randomly chose 48 markers for this study ( Table S3 ) . We divided the mPing excisions into five types with the same criteria as those used for the de novo insertions: endosperm- , radicle- , leaf- , shoot- , and radicle/shoot-specific excisions ( Figure S7 ) . There were no endosperm-specific and radicle/shoot-specific excisions , indicating that no mPing transposition occurs during the period from the onset of gametogenesis to the early stage of embryogenesis . We detected seven radicle-specific , six leaf-specific , and three shoot-specific excisions . All shoot-specific excisions were detected from the 1st leaf blade sample . These results indicate that , like the de novo insertion , the mPing excision also occurs during the period from the regionalization of the SAM and radicle to the formation of the first leaf primordium , although we cannot completely rule out the possibility that these excisions occur also in somatic cells of mature tissues . Thus , in addition to the experimental results of the de novo insertion , we concluded that mPing of EG4 was most actively transposing in the 3 to 5 DAP embryo . Both Ping and Pong provide a Myb-like protein and a transposase , which are encoded by their ORF1 and ORF2 , respectively ( Figure 3A ) , and have been considered as autonomous elements responsible for the mPing transposition . We investigated the expression of Ping-ORF1 , Ping-ORF2 , Pong-ORF1 , and Pong-ORF2 during embryogenesis to evaluate which autonomous element plays a predominant role in driving the mPing transposition in EG4 . Reverse transcription-PCR analysis revealed that Ping-ORF1 and Ping-ORF2 constitutively expressed in the ovary during embryogenesis ( Figure 3B ) . On the other hand , no transcriptions of Pong-ORF1 and Pong-ORF2 ( Figure 3B ) were observed . This strongly suggests that Ping predominantly controls the mPing transposition in EG4 . We performed real-time quantitative PCR ( qPCR ) analysis to compare the expression level of Ping-ORF1 and -ORF2 between EG4 and Nipponbare during embryogenesis . In all developmental stages from 1 to 6 DAP , the expression levels of both Ping-ORF1 and -ORF2 were higher in EG4 than in Nipponbare ( Figure 3C , D ) . Since EG4 harbors seven copies of Ping , whereas Nipponbare has only one copy ( Table S4 ) , the difference in the expression levels between EG4 and Nipponbare is considered to be attributable to the different copy number of Ping . However , we found that Ping of EG4 showed different expression patterns from that of Nipponbare . In Nipponbare , the expression level of Ping-ORF1 and -ORF2 gradually declined until 3 DAP , and restored to the basal level at 6 DAP . In contrast , in EG4 , the expression levels of both Ping-ORF1 and -ORF2 rapidly increased , with a peak at 3 DAP ( Figure 3C , D ) . The ratio of relative expression level ( EG4/Nipponbare ) clearly demonstrated that Ping might be up-regulated in a developmental stage-specific manner in the ovary of EG4 ( Figure 3E ) . Since mPing transposed during the period from 3 to 5 DAP , the rapid increase in Ping expression most likely drive the mPing transposition . We investigated the spatial expression pattern of Ping by in situ hybridization using Ping-specific probes . The probe positions were indicated in Figure 3A . The Ping transcripts were detected in all tissues , viz . embryo , endosperm , and ovary wall , in both EG4 and Nipponbare ( Figure 4A–C , S8 ) . Among the tissues , the 3 DAP embryo of EG4 yielded an exceptionally strong signal , indicating a high accumulation of Ping transcripts ( Figure 4A ) , whereas the 5 DAP embryo showed a much lower accumulation of Ping transcripts in EG4 ( Figure 4D–F ) . Such a drastic change in accumulation quantity of Ping transcripts with the advance of embryogenesis was consistent with the change in the expression quantity of Ping with the advance of embryogenesis , which was investigated by real-time qPCR ( Figure 3C–E ) . These results suggest that the tissue- and developmental stage-specific accumulation of the Ping transcripts triggers mPing transposition at this stage in EG4 . To confirm this hypothesis , we evaluated the spatial expression pattern of Ping in the SAM during the vegetative phase . As described above , mPing hardly transposes in the SAM during this phase . The Ping transcripts were detected in all tissues including the SAM , and , as expected , there was no obvious difference in the signal intensity between EG4 and Nipponbare ( Figure 4G–I ) . Thus the Ping transcripts proved to accumulate developmental stage-specifically only in the tissue where mPing actively transposes . We therefore concluded that the high accumulation of Ping transcripts triggers the transposition of mPing in the 3 DAP embryo of EG4 . EG4 has seven Ping elements ( Ping-1 to -7 ) , whereas Nipponbare has only one ( Ping-N ) ( Table S4 ) . When we sequenced and compared all Ping elements , a single nucleotide polymorphism ( SNP ) in the first intronic region of Ping-ORF1 was detected between EG4 and Nipponbare ( Figure 5A ) . Ping-N has a ‘T’ nucleotide on the SNP region , whereas all Ping elements in EG4 have a ‘C’ nucleotide . We named the former ‘T-type Ping’ and the latter ‘C-type Ping’ . In addition to EG4 , several Aikoku and Gimbozu landraces ( hereafter AG strains ) are known to exhibit high mPing activity [21] . We investigated the SNP-type of Ping and the copy number of Ping and mPing in 93 AG strains , and evaluated the effect of C-type Ping on the mPing activity . These 93 AG strains were divided into three groups according to the SNP-type of the Ping allele ( Table S4 ) : strains harboring C-type Ping; strains harboring T-type Ping; and strains harboring no Ping . The strains with C-type Ping had more mPing copies than those with T-type Ping or no Ping ( Figure 5B , Steel–Dwass test , p<0 . 01 ) . This implies that the C-type Ping could drive the mPing transposition . We further investigated the expression patterns of Ping-ORF1 and -ORF2 in two mPing-active strains ( A119 and A123 ) and two mPing-inactive strains ( A105 and G190 ) during embryogenesis ( from 1 to 6 DAP ) . A119 and A123 have six and ten copies of C-type Ping , respectively , and both A105 and G190 have one copy of T-type Ping ( Table S4 ) . Expression analysis revealed that A105 and G190 kept low expression levels of Ping-ORF1 and -ORF2 , whereas A119 and A123 showed high expression levels with a peak around 3 DAP ( Figure 5C–F ) . This indicates that , in EG4 , A119 , and A123 , the developmental stage-specific expression of Ping is controlled by the same factor ( s ) described in the Discussion .
Chromosomal position and copy number of TE often affect the transposition activity . The former is known as ‘position effect’ and the latter as ‘dosage effect’ . Eight independent Tam3 copies residing in the Antirrhinum majus genome show different transposition activities from each other [36] . In Arabidopsis , germinal reversion frequency of Tag1 increases in proportion to its copy number [32] . The mPing-inactive strains Nipponbare , A105 , and G190 have only one Ping at the same locus , whereas the mPing-active strains EG4 , A119 , and A123 have respectively seven , six , and ten copies of Ping at different loci except for the Ping-1 locus . Furthermore , the expression pattern of Ping showed slight variation among the mPing-active strains harboring only C-type Ping . These results suggest that the developmental stage-specific up-regulation of Ping expression is probably regulated by the position-effect and/or the dosage-effect . Intronic SNPs are known to cause drastic effects on gene expression . In humans , an intronic SNP in SLC22A4 affects transcriptional efficiency in vitro , owing to an allelic difference in affinity to the transcriptional factor RUNX1 [37] . Furthermore , a SNP located in the intronic enhancer region of the thyroid hormone receptor β gene enhances pituitary cell-specific transcriptional activity [38] . In this study , we demonstrated that a SNP is present in the intronic region of Ping-ORF1 , and Ping elements in the AG strains were categorized into either T-type or C-type Ping according to the SNP-type . Since all strains that showed a peak in the expression analysis had only C-type Ping , the intronic SNP might influence the developmental stage-specific up-regulation of Ping expression . T-type Ping was present in 14 AG strains as one copy , and its chromosomal location did not differ among strains . In contrast , the copy number of C-type Ping varied from one to ten , and their chromosomal locations , except for Ping-1 , differed from each other . These results indicate that T-type Ping has lost its activity , whereas C-type Ping may be still active in the rice genome . Furthermore , we found that the copy number of mPing was significantly larger in strains harboring C-type Ping than in strains harboring T-type Ping . This strongly supports that C-type SNPs in the intronic region of Ping contribute to the amplification of mPing , presumably by the developmental stage-specific up-regulation of Ping expression . Since the transposition of TEs often damages the host genome , TEs with high transposition activity are targeted by the silencing mechanisms . Nevertheless , MITEs amplify to very high copy numbers not only in plant genomes but also in animal genomes . Very little is known about how MITEs attain their high copy numbers by escaping the silencing mechanism . The transposition of mPing is transiently induced by various stresses [14]–[18] , indicating that the activity of mPing is suppressed by the silencing mechanisms in many cultivars . Thus , mPing must overcome the silencing mechanism in order to maintain the transposition activity under natural growth conditions . Our results revealed that mPing in EG4 was mobilized by the sufficient supply of Ping transcripts produced only during the period of mPing transposition . This stage-specific activation is thought to be a strategy of the mPing family to amplify mPing by escaping from the silencing mechanism of the host genome . Since no active MITEs other than mPing so far have been identified , it is very difficult to elucidate if the other MITEs also attain their high copy numbers in the same way as mPing amplifies . Given that the other active MITEs are identified , however , our study will help to understand their amplification mechanisms . Our previous study documented the generation of new regulatory networks by a subset of mPing insertions that render adjacent genes stress inducible [13] . In addition to mPing , other MITEs also contribute to gene and genome evolution via providing new promoter regulatory sequences , transcriptional termination elements , and new alternative exons [39] , suggesting that the amplification of MITEs causes gene and genome evolution . Our results provide clues to further understand not only the amplification mechanism of MITEs but also the co-evolution of MITEs and the host genome .
EG4 ( cultivar Gimbozu ) , Nipponbare , and 94 Aikoku/Gimbozu landraces were used in this study ( Table S4 ) . Aikoku/Gimbozu landraces were provided from the GenBank project of the National Institute of Agrobiological Science , Ibaraki , Japan . Reciprocal crosses between EG4 and Nipponbare were made in a green house . Before pollination , all anthers were removed from the flowers of maternal plants . The pollinated flowers were covered with protective bags to prevent outcrossing until harvest . After harvesting , success of crosses was checked with the molecular markers . For ontogenical analysis , eight progenies of EG4 ( S1 ) derived from a single parental plant ( S0 ) were grown in a greenhouse , and all S2 seeds were harvested . For S1 plants , each seed was cut into two halves , and the half including the embryo was germinated and the other was sampled . After germination , the radicle and the 1st , 2nd , 3rd , 4th , and 5th leaf blades were sampled . The second leaf was collected from S0 and S2 plants . For estimation of Ping and mPing copy numbers , eight bulked plants were sampled . For RNA extraction , ovaries before pollination and ovaries at 1 , 2 , 3 , 4 , 5 , and 6 DAP were collected . All samples were immediately frozen in liquid nitrogen and stored at −80°C until use . DNA extraction and transposon display was performed following a published protocol [30] . For DNA extraction from endosperm , we used GM quicker 2 ( Nippon Gene ) . Sequencing of mPing-flanking fragments excised from transposon display gels and primer design were performed following a published protocol [30] . The genomic locations of the mPing insertion sites were forecasted by a BLAST search in the Rice Annotation Project Database ( RAP-DB; http://rapdb . dna . affrc . go . jp/ ) [40] , [41] using mPing flanking sequences as queries . To prepare enough templates for PCR , whole genome amplification was performed using an illustra GenomiPhi V2 Kit ( GE Healthcare ) . mPing excision was detected by PCR with mPing-sequence characterized amplified region ( SCAR ) markers [29] . PCR was performed in 10-µl reaction volumes containing 10 ng of the template DNA , 5 µl of GoTaq Green Master mix ( Promega ) , 5% DMSO , and 0 . 25 µM of each primer ( Table S3 ) . PCR conditions were as follows: 94°C for 3 min; 40 cycles of 98°C for 10 s , 57°C for 30 s , and 72°C for 45 s; and 72°C for 5 min . To detect the presence of Ping-N , -1 , -2 , -3 , -4 , -5 , -6 , and -7 , eight Ping-SCAR markers were used . The genomic locations of the Ping insertion sites were referred from a previous report [42] . For detection of the Ping-1 allele , PCR was performed in 10-µl reaction volumes containing 10 ng of template DNA , 0 . 2 U of KOD FX Neo ( Toyobo ) , 1×PCR buffer for KOD FX Neo ( Toyobo ) , and 0 . 2 µM of each primer ( Table S5 ) . PCR conditions were as follows: 94°C for 3 min; 35 cycles of 98°C for 10 s , 60°C for 30 s , and 68°C for 90 s; and 72°C for 1 min . For detection of Ping-N , -2 , -3 , -4 , -5 , -6 , and -7 alleles , PCR was performed in 10-µl reaction volumes containing 10 ng of template DNA , 5 µl of GoTaq Green Master mix ( Promega ) , 5% DMSO , and 0 . 25 µM of each primer ( Table S5 ) . PCR conditions were as follows: 94°C for 3 min; 35 cycles of 98°C for 10 s , 60°C for 30 s , and 72°C for 45 s; and 72°C for 1 min . Total RNA was isolated using TriPure isolation reagent ( Roche ) and digested using RNase-free DNase ( TaKaRa ) . First strand cDNA was synthesized using a Transcriptor first strand cDNA synthesis kit ( Roche ) . For reverse transcription PCR , PCR was performed in 10 µl reaction volumes containing cDNA generated from 4 ng total RNA , 0 . 2 U of KOD FX Neo ( Toyobo ) , 1×PCR buffer for KOD FX Neo ( Toyobo ) , and 0 . 5 µM of each primer . PCR conditions were as follows: 94°C for 3 min; 35 or 45 cycles of 98°C for 10 s , 60°C for 10 s , and 68°C for 10 s . Relative quantification of Ping-ORF1 and Ping-ORF2 were calculated by the 2−ΔΔCT method [43] using Light cycler 1 . 5 ( Roche ) . The UBQ5 gene was used as the calibrator gene . The thermal profile consisted of 10 min at 95°C; and 45 cycles of 4 s at 95°C , 10 s at 60°C , and 1 s at 72°C . Amplification data were collected at the end of each extension step . The primer pairs used in this study are listed in Table S6 . Plant samples were fixed with 4% ( w/v ) paraformaldehyde and 1% Triton X in 0 . 1M sodium phosphate buffer for 48 h at 4°C . They were then dehydrated in a graded ethanol series , substituted with 1-butanol , and embedded in Paraplast Plus . The samples were sectioned at 8-µm thickness using a rotary microtome . Fragments of Ping-ORF1 ( 1091 bp ) and Ping-ORF2 ( 1368 bp ) were cloned into pBlueScript SK+ ( Stratagene ) and sequenced . For digoxigenin-labeled antisense/sense RNA probe synthesis , in vitro transcription was performed using T7 RNA polymerase and T3 RNA polymerase . In situ hybridization and immunological detection with alkaline phosphatase were performed according to Kouchi and Hata [44] . PCR was performed in 10-µl reaction volumes containing 10 ng of template of DNA , 5 µl of GoTaq Green Master mix ( Promega ) , 5% DMSO , and 0 . 25 µM of each primer . PCR conditions were as follows: 94°C for 3 min; 35 cycles of 98°C for 10 s , 60°C for 30 s , and 72°C for 30 s; and 72°C for 1 min . PCR primers used in this study are listed in Table S6 . Because the original sequence contained an Afa I restriction site , one mutation was introduced into the reverse primer . The 5-µl PCR products were mixed with 5 µl restriction mixture containing 1 U Afa I ( TaKaRa ) , 33 mM Tris-acetate ( pH 7 . 9 ) , 10 mM Mg-acetate , 0 . 5 mM Dithiothreitol , 66 mM K-acetate , and 0 . 01% ( w/v ) bovine serum albumin . After 16 h incubation at 37°C , DNA gel electrophoresis was performed . PCR products ( 502 bp ) including +1261T SNP were not digested , whereas PCR products including +1261C SNP were digested into two fragments ( 352 bp and 150 bp ) . To determine the copy number of Ping by Southern blot analysis , genomic DNA samples were digested with Eco RI restriction enzyme . These samples were loaded onto an agarose gel , separated by electrophoresis , blotted onto a nylon membrane , and probed with the Ping fragment . The mPing copy number was determined by real-time quantitative PCR as described previously [45] with little modification . Quantitative PCR was performed using the LightCycler 480 system ( Roche ) . PCR was performed in 20 µl reaction volumes containing 5 µl genomic DNA ( 0 . 4 ng/µl ) , 1×LightCycler 480 SYBR Green I Master mix ( Roche ) , and 0 . 5 µM of each primer . Specificity of the amplified PCR product was assessed by performing a melting curve analysis on the LightCycler 480 system .
|
Transposable elements are major components of eukaryotic genomes , comprising a large portion of the genome in some species . Miniature inverted-repeat transposable elements ( MITEs ) , which belong to the class II DNA transposable elements , are abundant in gene-rich regions , and their copy numbers are very high; therefore , they have been considered to contribute to genome evolution . Because MITEs are short and have no coding capacity , they cannot transpose their positions without the aid of transposase , provided in trans by their autonomous element ( s ) . It has been unknown how MITEs amplify themselves to high copy numbers in the genome . Our results demonstrate that the rice active MITE mPing is mobilized in the embryo by the developmental stage-specific up-regulation of an autonomous element , Ping , and thereby successfully amplifies itself to a high copy number in the genome . The short-term expression of Ping is thought to be a strategy of the mPing family for amplifying mPing by escaping the silencing mechanism of the host genome .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"plant",
"breeding",
"mobile",
"genetic",
"elements",
"plant",
"science",
"agronomy",
"cell",
"biology",
"genetic",
"elements",
"genetics",
"transposable",
"elements",
"biology",
"and",
"life",
"sciences",
"dna",
"transposons",
"plant",
"genetics",
"molecular",
"cell",
"biology",
"agriculture"
] |
2014
|
Early Embryogenesis-Specific Expression of the Rice Transposon Ping Enhances Amplification of the MITE mPing
|
Flagella are assembled sequentially from the inside-out with morphogenetic checkpoints that enforce the temporal order of subunit addition . Here we show that flagellar basal bodies fail to proceed to hook assembly at high frequency in the absence of the monotopic protein SwrB of Bacillus subtilis . Genetic suppressor analysis indicates that SwrB activates the flagellar type III secretion export apparatus by the membrane protein FliP . Furthermore , mutants defective in the flagellar C-ring phenocopy the absence of SwrB for reduced hook frequency and C-ring defects may be bypassed either by SwrB overexpression or by a gain-of-function allele in the polymerization domain of FliG . We conclude that SwrB enhances the probability that the flagellar basal body adopts a conformation proficient for secretion to ensure that rod and hook subunits are not secreted in the absence of a suitable platform on which to polymerize .
Some bacteria swim through liquid and swarm over surfaces by synthesizing trans-envelope nanomachines called flagella . Flagella spontaneously self-assemble from over twenty separate proteins thought to be organized into three structural domains called the basal body , the hook and the filament [1–3] . The basal body is composed of a ring of a transmembrane protein called FliF that surrounds a membrane-embedded , dedicated type III secretion export apparatus [4–7] . Beneath the basal body sits the FliG rotor that interacts with the proton-conducting stators to generate torque , as well as the C-ring proteins FliM and FliN that control the direction of flagellar rotation [8 , 9] . The export apparatus secretes subunits of the drive-shaft rod that transits the peptidoglycan , the universal joint hook , and the long helical filament . When the basal body rotates , rotational energy is transmitted through the rod and the hook to turn the filament that generates propulsion like a propeller . Central to flagellar assembly is the control of the flagellar type III secretion export apparatus . Type III secretion export apparati are housed within both the flagellar basal body and the evolutionarily-related “injectisome” used by various pathogenic bacteria to secrete toxins directly into eukaryotic host cells [10 , 11] . Recent cytological evidence suggests that the export apparatus is likely the first substructure to form and serves as the nucleation point for subsequent flagellar/injectisome basal body assembly [7 , 12 , 13] . The export apparatus requires a conserved set of 5 transmembrane proteins: FliP , FliQ , FliR , FlhA , and FlhB [6 , 14] . FlhA is important for the recognition of secretion substrates and FlhB controls the substrate specificity of the export apparatus [15–18] . The roles of FliP , FliQ , and FliR are poorly understood , but presumably these proteins either transduce the proton motive force that powers protein export or serve as the export channel [19 , 20] . Flagellar export systems differ from injectisome export systems in that flagella require a sixth transmembrane protein FliO that appears to function as a regulator [21 , 22] . The flagellar export apparatus secretes two classes of proteins considered “early” and “late” depending on the temporal order in which they are secreted . Early class flagellar structural components like the rod and the hook subunits are secreted first , and are recognized by information encoded within the N-terminus of the secreted protein [23–25] . Late class subunits , which include the hook-associated proteins and flagellin , are ushered by chaperones and are secreted only after a substrate specificity switch occurs within the export apparatus in response to hook completion [17 , 26–28] . Furthermore , the expression of the genes encoding the late class secretion substrates is inhibited by the anti-sigma factor FlgM prior to hook completion . Once the hook is complete and the substrate specificity switch occurs , FlgM is exported , its cognate sigma factor is liberated , and late class flagellar genes are expressed [29 , 30] . Thus , the activity of the export apparatus is morphogenetically coupled to flagellar structure so as to govern subsequent flagellar gene expression and assembly . Finally , there is evidence for another morphogenetic coupling event that precedes hook completion as mutants defective in the cytoplasmic flagellar rotor FliG and C-ring components FliM and FliN have flagellar type III secretion defects [31–37] . Here we provide further evidence for an early morphogenetic checkpoint in flagellar assembly whereby completion of the flagellar basal body and the single pass transmembrane protein SwrB of Bacillus subtilis functionally activate the flagellar type III export apparatus to become secretion proficient . Mutants defective for SwrB assemble a reduced number of flagella due to a reduced frequency of hook assembly , and a secretion defect was implicated as many genetic suppressors increased the abundance of the secretion apparatus component FliP . We show that the proficiency of the export apparatus for hook subunit secretion was coupled to the conformation of the basal body as C-ring mutants phenocopied the absence of SwrB . Furthermore , overexpression of SwrB rescued hook assembly to cells defective for FliG , and FliG gain-of-function alleles suppressed the absence of SwrB . We conclude that SwrB chaperones the formation of a completed basal body conformation thereby activating the flagellar type III export apparatus for the secretion of the hook , and likely rod , structural subunits .
SwrB is a single-pass transmembrane protein of unknown function that is required for swarming motility in B . subtilis [38 , 39] . Swarming motility over a solid surface requires an increase in the number of flagella relative to swimming in liquid , and mutants defective in the master regulator of flagellar biosynthesis , SwrA , have reduced flagellar number and are unable to swarm [40–42] . To determine whether cells mutated for SwrB have a defect in flagellar filament number , a variant of the flagellar filament protein Hag that could be labeled with a fluorescent dye ( HagT209C ) was introduced to the wild type , swrB , and swrA mutant backgrounds [43] . Cells mutated for SwrB appeared to have fewer filaments than wild type and resembled cells mutated for SwrA ( Fig 1A ) . We conclude that SwrB is required for the assembly of wild type numbers of flagellar filaments and we infer that the reduction in filament number accounts for the swarming defect of a swrB mutant . Consistent with a reduction in flagellar filament assembly , cells lacking either SwrB or SwrA showed a reduced level of the flagellar filament protein , Hag , by Western blot analysis ( Fig 2A ) . The reduction in Hag protein was likely due to reduced transcription of the hag gene as cells mutated for either SwrB or SwrA showed lower expression from a reporter in which the hag promoter ( Phag ) was fused to the lacZ gene encoding β-galactosidase ( Phag-lacZ ) ( Fig 2B , white bars ) , and a reduced frequency of cells expressing a reporter in which the Phag promoter was fused to green fluorescent protein ( GFP ) ( Phag-gfp ) ( Fig 2C , white bars , and S1 Fig ) [39 , 41] . The Phag promoter is transcribed by RNA polymerase and the alternative sigma factor , σD [44] . Whereas cells mutated for SwrA exhibited a reduced level of σD protein , cells mutated for SwrB exhibited a level of σD protein comparable to the wild type ( Fig 2A ) . We conclude that the absence of SwrB resulted in reduced expression of the hag gene but unlike the absence of SwrA , the defect occurred downstream of σD protein levels . One reason that hag expression might be reduced in the SwrB mutant despite wild type levels of σD is due to enhanced inhibition of σD by its cognate anti-sigma factor FlgM [45–48] . Consistent with enhanced inhibition by FlgM , mutation of FlgM increased the magnitude of expression of the Phag-lacZ reporter ( Fig 2B , black bars ) , and increased the frequency of expression of the Phag-GFP reporter in the wild type and swrB mutant backgrounds ( Fig 2C , black bars , and S1 Fig ) . The SwrA protein controls the frequency of cells that have σD protein levels above a critical threshold upstream of FlgM regulation [49 , 50] and thus simultaneous mutation of both FlgM and SwrA increased expression magnitude in a subpopulation ( Fig 2B ) but did not increase the frequency of σD activity ( Fig 2C ) . Finally , mutation of FlgM , did not appear to increase flagellar filament number in the absence of either SwrB or SwrA ( Fig 1B ) despite a modest increase in flagellin protein levels ( Fig 2A ) . We conclude that enhanced inhibition of σD by FlgM was a consequence and not a cause of the reduction of flagellar number in the swrB mutant . FlgM inhibition of σD activity is enhanced when cells are defective in flagellar hook synthesis , and a defect in flagellar hook synthesis would also account for the reduction in flagellar filaments observed in the SwrB mutant [51 , 52] . To measure flagellar hook numbers , a variant of the flagellar hook protein FlgE that could be labeled with a fluorescent dye ( FlgET123C ) was introduced to the wild type , swrB , and swrA mutant backgrounds [51] . Cells mutated for swrB appeared to have fewer hooks than the wild type and instead resembled the reduced numbers of hooks in cells mutated for swrA ( Fig 1C ) . To count flagellar hooks , 3D structured illumination microscopy ( 3D-SIM ) was conducted on hook-stained cells of each genetic background and each individual cell was expressed as a point on a scatter plot representing the number of hooks versus cell length ( Fig 3 , green symbols ) . Whereas wild type cells had an average of 15 hooks , swrB and swrA mutants had an average of 5 and 4 hooks per cell , respectively . We conclude that the swrB mutant was defective in hook assembly and as a result exhibited enhanced inhibition of σD by FlgM . Because hook assembly depends on flagellar basal body synthesis , the reduction in hook numbers observed in a swrB mutant could be due to a commensurate reduction in basal body number . To measure flagellar basal body synthesis , a translational fusion of GFP to the C-ring protein FliM ( amyE::Pfla/che-fliM-GFP ) was introduced to the wild type , swrB , and swrA mutant backgrounds [53] . Cells mutated for swrB appeared to have wild type numbers of flagellar C-rings , whereas cells mutated for swrA appeared to have a reduction in flagellar basal body number , as previously reported ( Fig 1D ) [53] . To count flagellar C-rings 3D-SIM was conducted on cells of each genetic background and each individual cell was expressed as a point on a scatter plot representing the number of basal bodies versus cell length ( Fig 3 , red symbols ) . On average , wild type had 24 C-rings , the swrB mutant had 23 C-rings , and the swrA mutant had 9 C-rings per cell . We conclude that SwrB and SwrA increased flagellar numbers at two different steps in flagellar assembly by promoting hook and C-ring ( basal body ) assembly respectively . We further conclude that SwrB somehow promoted hook assembly at a step downstream of the incorporation of FliM into the basal body . To determine how SwrB regulates hook assembly , spontaneous suppressor mutations were isolated that restored swarming motility to a swrB mutant . When a swrB mutant was inoculated in the center of a swarm agar plate , cells initially grew as a tight central colony and , unlike the wild type , failed to spread from the inoculum origin ( Fig 4A ) . However , after 24 hours of incubation , flares of cells that had regained the ability to swarm emerged from the central colony and cells from these flares were clonally isolated . Twenty-four spontaneous sob ( suppressor of swrB ) mutants were independently isolated . Each suppressor was validated by PCR length polymorphism to confirm the presence of the swrB mutant allele . A combination of candidate gene sequencing , phage transduction linkage mapping , and Illumina whole-genome sequencing was used to identify the location of each of the sob suppressor mutations . Based on the swarm behavior and location of the suppressor mutations , the sob alleles were divided into six classes and analyzed separately ( Table 1 and Figs 4B–4G and 5A ) . Mutations in FliP and FliG that suppressed the absence of SwrB suggested that the mechanism by which SwrB enhanced hook assembly was related to flagellar secretion by way of basal body structure . Therefore , we further explored the relationship of SwrB to components of the basal body and C-ring for enhancing hook polymerization . To do so , strains were generated that contained the FlgET123C allele for fluorescent labeling of the flagellar hook in backgrounds mutated for SwrB , FliM , FliG , and FliF . Cells mutated for either FliM or FliG displayed a low frequency of flagellar hooks and resembled cells mutated for SwrB ( Fig 9 ) . By contrast , no hooks were detected in cells mutated for FliF ( Fig 9 ) . Next , the swrB gene was cloned downstream of an IPTG inducible Physpank promoter and integrated at an ectopic site in each of the strains tested ( amyE::Physpank-swrB ) . In the presence of IPTG , overexpression of SwrB increased the frequency of hooks for the swrB , fliM and fliG mutants but did not restore hook formation to the fliF mutant ( Fig 9 ) . We conclude that the activation of hook assembly by SwrB does not require FliG or FliM , but that FliGQ132R nonetheless compensates for the absence of SwrB . We further conclude that the basal body structural component FliF is required for SwrB to activate hook assembly . FliF could be required for SwrB activation of hook assembly because FliF also serves as the polymerization platform for the flagellar rod and hook . Thus , SwrB could stimulate flagellar secretion in the absence of FliF but the hook subunits would accumulate in the supernatant due to an inability to polymerize . To determine the effect SwrB has on hook protein secretion , strains were generated that were mutated for the extracellular hook chaperone protein FlgD such that all strains would fail to polymerize FlgE , and FlgE would thus be secreted into the supernatant [51 , 65] . Cells mutated for FliM and FliG reduced the amount of secreted hook protein whereas cells mutated for FliF , FliP , and FlhA abolished hook secretion ( Fig 10A ) . Next , the IPTG-inducible Physpank-swrB construct was added to each strain to determine the effect that over-expression of SwrB would have on hook secretion . When SwrB was over-expressed , the amount of hook protein in the supernatant appeared to increase in the fliM , fliG , and swrB mutants , but not in the fliF , fliP , or flhA mutants ( Fig 10B ) . Finally , addition of the Pfla/che-fliPsob22 complementation construct provided in trans appeared to increase FlgE secretion in fliM , fliG , swrB , and fliP mutants but not in the fliF or flhA mutant ( Fig 10C ) . We conclude FliF , FliP , and FlhA are downstream of SwrB for hook protein secretion . In sum , we conclude that SwrB activates an early flagellar morphogenetic checkpoint by catalyzing a FliF-basal body conformation that activates the type III secretion export apparatus for hook subunit secretion .
Flagellar regulation is morphogenetically coupled to flagellar structural intermediates to ensure that the structural subunits are assembled in the proper order . A classic example of morphogenetic coupling is illustrated by how the completion of the flagellar hook induces a change in the substrate specificity of the export apparatus and governs the expression of the flagellar filament protein via secretion of the anti-sigma factor FlgM [2 , 27] . Here we find evidence of an earlier morphogenetic coupling event in which SwrB and the completion of the flagellar basal body govern the secretion of the hook ( and likely rod ) structural subunits . Given that the type III secretion export apparatus appears to be the first flagellar subdomain assembled within the membrane , we suggest that if the export apparatus was immediately active upon assembly , structural components would be secreted in the absence of the basal body upon which they are polymerized . Instead , it appears that the export apparatus becomes functional only after the basal body is complete; an event indicated by a conformational change adopted by the flagellar base plate protein FliF . Thus , basal body completion is a discrete flagellar morphogenetic checkpoint and we argue that the B . subtilis membrane protein SwrB potentiates the ability of the basal body to adopt a “completed” conformation . SwrB ( YlxL ) was originally discovered as the product encoded by the last gene in the 32 gene fla/che operon and was shown to be required for the activation of σD-dependent gene expression and swarming motility [38 , 39 , 41] . Here we account for both previously reported phenotypes of the swrB mutant . We found that cells defective for SwrB were unable to swarm because they failed to synthesize wild type numbers of flagellar filaments and resembled the hypoflagellated state of cells defective for another swarming regulator , SwrA [53] . The SwrB defect in flagellar number was not due to a reduction in the number of flagellar basal bodies as seen in cells defective for SwrA however , but rather due to a reduction in the number of flagellar hooks . The swrB mutant defect in hook synthesis is consistent with the observed reduction in σD-dependent gene expression as hook completion is needed to antagonize the σD anti-sigma factor FlgM [39 , 51] . Thus , SwrB increased the probability that basal bodies became proficient for hook assembly . Spontaneous suppressors that restored swarming motility to swrB mutants indicated that SwrB increased the frequency of hook assembly by activating hook subunit secretion via the type III secretion component FliP . FliP is a transmembrane protein that is incorporated early in the nascent basal body and required for nucleation of other components of the flagellar type III secretion export apparatus FliO , FliQ , FliR , FlhA and FlhB [5 , 66 , 67] . Mutation of FliP abolishes secretion , and while the precise function of FliP is unknown , it appears to be regulated by at least two mechanisms . First , the FliP N-terminal signal sequence appears to be inhibitory in S . enterica and is processed presumably after assembly of the export apparatus [5 , 68] . B . subtilis FliP however lacks the N-terminal signal sequence altogether , perhaps an indication of the need for additional regulatory mechanisms [68 , 69] . Second , the levels of FliP appear to be important in S . enterica as the accessory protein FliO protects FliP from proteolytic degradation [21 , 22] . Here we further support the importance of FliP protein levels in B . subtilis as 22 out of 24 spontaneous suppressor-of-swrB ( sob ) mutations restored swarming to a swrB mutant by increasing FliP expression . FliP was directly implicated by a single suppressor-of-swrB ( sob ) mutation that mutated the FliP Shine-Dalgarno sequence closer to consensus , thus enhancing translation . Furthermore , enhanced transcription of the native fliP gene within the fla/che operon by another 21 sob alleles or an ectopically integrated fliP gene expressed from artificial IPTG-inducible promoter was sufficient to rescue swarming to a swrB mutant . FliP is hypothesized to sit within the confines of a ring of the basal body protein FliF , the stoichiometry of which along with the rest of the type III secretion export apparatus components is thought to be definite and precise [5] . Thus , how extra copies of wild type FliP would improve swarming and/or become incorporated into basal bodies is unclear . Perhaps extra FliP protein titrates an inhibitor of the type III secretion export apparatus . Alternatively , FliP is thought to be one of the earliest proteins assembled in the flagellum and overexpression may increase the population of FliP molecules in a secretion-active conformation that preferentially promotes basal body nucleation . A potential candidate for either the inhibitor and/or conformational regulator of the type III secretion system is found in the final class of SwrB suppressors that fall within the rotor protein FliG . FliG forms the gear-like rotor that docks to the cytoplasmic surface of the FliF basal body protein , interacts with the MotA/MotB proton channel stator , and serves as a scaffold for the assembly of the FliM/FliN ( FliY ) cytoplasmic C-ring [33 , 70–75] . Although FliG , FliM , and FliN are found in the cytoplasm , it has long been known that mutations in each protein cause extracytoplasmic flagellar assembly defects [31–37] . Here we show that a gain-of-function mutation found in the ARM motif that controls FliG polymerization bypasses the C-ring requirement for flagellar hook ( but not flagellar filament ) assembly [64] ( Fig 8 and S6 Fig ) . We posit that proper conformation of the FliG rotor in vivo , normally promoted by the completion of the C-ring and the presence of SwrB , activates the type III secretion export apparatus . Since FliG and the export apparatus are not known to directly interact , we infer that a regulatory conformational change is propagated from the rotor through basal body protein FliF [74] . FliF is a critical transmembrane structural protein that defines the flagellar basal body as it docks to the FliG rotor , surrounds the type III export apparatus , and forms a polymerization platform for the rod [4 , 76–78] . FliF has been hypothesized to exist in active and inactive conformations . In S . enterica , FliF has a large periplasmic domain that may regulate the export apparatus as electron microscopy of FliF rings show two conformations , one with a lumen that appears to be open and one in which the lumen appears to be closed [4 , 79–81] . Furthermore , a FliF closed conformation was hypothesized to be adopted when the poorly-understood periplasmic protein FliE was mutated [6 , 82 , 83] . In B . subtilis , FliF is required for secretion as cells defective in FliF fail to secrete both the early class substrate FlgE and the late class substrate FlgM ( Fig 10 ) [52] . Thus , FliF could be required for secretion by surrounding the type III export apparatus and regulating its function . Based on cytological and genetic suppressor data , we conclude that SwrB functions as an assembly chaperone to enhance the probability that the flagellar basal body adopts a conformation proficient for secretion ( Fig 11 ) . We propose that the flagellar type III secretion apparatus and FliF form first in the membrane as a “proto-basal body” that is inactive for export ( Fig 11A ) [7] . The proto-basal body is able to spontaneously mature to become proficient for hook secretion at a low frequency ( Fig 11B ) . Under normal conditions , the frequency of proto-basal body maturation is increased by both SwrB and assembly of the C-ring as the absence of either share a low hook-to-basal body ratio ( Fig 11C ) . Indeed , SwrB and FliG appear to be required at the same step as FliG gain-of-function alleles that enhance rotor stability bypass the requirement of SwrB for hook assembly , and artificial overexpression of SwrB bypasses the need for FliG . The most likely convergence point for SwrB and FliG is the membrane-basal body protein FliF , as FliG is a known interactor and SwrB , itself a membrane protein , could be adjacent . We hypothesize that the previously documented changes in FliF conformation are responsible for activating the export apparatus and do so by activating the membrane protein FliP . In sum , maturation of the flagellar basal body to a secretion proficient state is a morphogenetic checkpoint that must be passed prior to proceeding to the export and assembly of more distal components . SwrB is conserved only in members of the genus Bacillus but the notion that flagellar basal body completion acts as a conformational liscencing event to permit subsequent flagellar secretion is consistent with genetic and biochemical observations in a variety of systems . In Campylobacter jejuni , the conformation adopted by the completed basal body activates expression of the rod and hook genes as a separate regulatory “tier” , and similar “4-tier” flagellar regulatory hierarchies have been demonstrated in Caulobacter crescentus and Pseudomonas aeruginosa [84–91] . Thus instead of regulating the type III export apparatus functionally , completion of the basal body transcriptionally controls the availability of the rod and hook cargo . Between our results in B . subtilis and the “4-tier” flagellar systems , we suggest that the rod should be considered separate from the basal body and therefore the structural domains of the flagellum should be considered the “basal body” , the “rod-hook” , and the “filament” , at least for these organisms . Finally , our conformation control model is consistent with the notion that a conformational change propagates from the needle tip to the export apparatus at the base of the type III secretion injectosomes of pathogens to activate the secretion of effectors in response to direct contact with host cells [92–94] . Thus , we conclude that the activation of secretion is a regulated morphogenetic checkpoint , and disrupting functional regulators could be a strategy to attenuate both flagella and injectisome virulence factors .
B . subtilis strains were grown in Luria-Bertani ( LB ) ( 10 g tryptone , 5 g yeast extract , 5 g NaCl per L ) broth or on LB plates fortified with 1 . 5% Bacto agar at 37°C . When appropriate , antibiotics were included at the following concentrations: 10 μg/ml tetracycline , 100 μg/ml spectinomycin , 5 μg/ml chloramphenicol , 5 μg/ml kanamycin , and 1 μg/ml erythromycin plus 25 μg/ml lincomycin ( mls ) . Isopropyl β-D-thiogalactopyranoside ( IPTG , Sigma ) was added to the medium at the indicated concentration when appropriate . For the swarm expansion assay , swarm agar plates containing 25 ml LB fortified with 0 . 7% Bacto agar were prepared fresh and the following day were dried for a total of 20 minutes in a laminar flow hood ( see below ) . All constructs were first introduced into the competent ancestral strain DS2569 by natural competence and then transferred to the 3610 background using SPP1-mediated generalized phage transduction or by transformation in the competent ancestral strain DK1042 ( as indicated by the presence of the comIQ12L allele in the genotype ) [95 , 96] . All strains used in this study are listed in Table 3 . All plasmids used in this study are listed in S1 Table . All primers used in this study are listed in S2 Table . To 0 . 2 ml of dense culture grown in TY broth ( LB broth supplemented after autoclaving with 10 mM MgSO4 and 100 μM MnSO4 ) , serial dilutions of SPP1 phage stock were added and statically incubated for 15 minutes at 37°C . To each mixture , 3 ml TYSA ( molten TY supplemented with 0 . 5% agar ) was added , poured atop fresh TY plates , and incubated at 30°C overnight . Top agar from the plate containing near confluent plaques was harvested by scraping into a 15 ml conical tube , vortexed , and centrifuged at 5 , 000 x g for 5 minutes . The supernatant was treated with 25 μg/ml DNase before being passed through a 0 . 45 μm syringe filter and stored at 4°C . Recipient cells were grown to stationary phase in 3 ml TY broth at 37°C . 1 ml cells were mixed with 25 μl of SPP1 donor phage stock . 9 ml of TY broth was added to the mixture and allowed to stand at 37°C for 30 minutes . The transduction mixture was then centrifuged at 5 , 000 x g for 5 minutes , the supernatant was discarded and the pellet was resuspended in the remaining volume . 100 μl of cell suspension was then plated on LB fortified with 1 . 5% agar , the appropriate antibiotic , and 10 mM sodium citrate . Cells were grown to mid-log phase at 37°C in LB broth and resuspended to 10 OD600 in pH 8 . 0 PBS buffer ( 137 mM NaCl , 2 . 7 mM KCl , 10 mM Na2HPO4 , and 2 mM KH2PO4 ) containing 0 . 5% India ink ( Higgins ) . Freshly prepared LB containing 0 . 7% Bacto agar ( 25 ml/plate ) was dried for 10 minutes in a laminar flow hood , centrally inoculated with 10 μl of the cell suspension , dried for another 10 minutes , and incubated at 37°C [101] . The India ink demarks the origin of the colony and the swarm radius was measured relative to the origin . For consistency , an axis was drawn on the back of the plate and swarm radii measurements were taken along this transect . For experiments including IPTG , cells were propagated in broth in the presence of IPTG , and IPTG was included in the swarm agar plates . For the pellet fraction ( cytoplasmic and cell-associated proteins ) , B . subtilis strains were grown in 20 ml LB broth in the presence 1 mM IPTG when appropriate to an OD600 of 1 . 2–1 . 7 , and 10 ml samples of broth culture were harvested by centrifugation , resuspended to 10 OD600 units in lysis buffer ( 20 mM Tris [pH 7 . 0] , 10 mM EDTA , 1 mg/ml lysozyme , 10 μg/ml RNase I , 1 mM PMSF ) , and incubated 30 minutes at 37°C . For the supernatant fraction ( secreted extracellular proteins ) , 10 ml of supernatant was collected from the same cultures as those used to generate the pellet fractions . The supernatant was clarified by centrifugation at 5 , 000 × g for 30 min and treated with 1 ml of freshly prepared 0 . 015% sodium deoxycholate for 10 min at room temperature . Proteins from the supernatant were precipitated by adding 500 μl chilled trichloroacetic acid ( TCA ) and incubating the mixture for >2 h on ice at 4°C . Precipitated proteins were pelleted at 9 , 447 × g for 10 min at 4°C , washed twice with 1 ml ice-cold acetone , and resuspended to 10 OD600 units in 0 . 1 N sodium hydroxide . Ten microliters of cell pellet or supernatant sample was mixed with 2 μl 6× SDS loading buffer . Samples were separated in parallel by 15% SDS-PAGE . Proteins were electroblotted onto nitrocellulose for 1 hour at 400 mA and probed with a 1:20 , 000 dilution of anti-FlgE primary antibody and with a 1:10 , 000 dilution of secondary antibody ( horseradish peroxidase [HRP]-conjugated goat anti-rabbit immunoglobulin G ) . Immunoblots were developed using the Pierce ECL Western blotting substrate kit ( Thermo Scientific ) . Fluorescence microscopy was performed with a Nikon 80i microscope along with a phase contrast objective Nikon Plan Apo 100X and an Excite 120 metal halide lamp . FM4-64 was visualized with a C-FL HYQ Texas Red Filter Cube ( excitation filter 532–587 nm , barrier filter >590 nm ) . GFP was visualized using a C-FL HYQ FITC Filter Cube ( FITC , excitation filter 460–500 nm , barrier filter 515–550 nm ) . Images were captured with a Photometrics Coolsnap HQ2 camera in black and white , false colored and superimposed using Metamorph image software . For Phag-GFP microscopy , cells were grown at 37°C in LB broth to OD600 0 . 6–1 . 0 , resuspended in 30 μl PBS buffer containing 5 μg/ml FM 4–64 and incubated for 5 min at room temperature . The cells were pelleted , resuspeneded in 30μl PBS buffer , and were observed by spotting 4 μl of suspension on a cleaned microscope slide and immobilized with a poly-L-lysine-treated glass coverslip . For fluorescent microscopy of flagellar filaments , 1 . 0 ml of broth culture was harvested at 0 . 6–1 . 0 OD600 , resuspended in 50 μl of PBS buffer containing 5μg/ml Alexa Fluor 488 C5 maleimide ( Molecular Probes ) , incubated for 3 min at room temperature , and washed once in 1 . 0 ml of PBS buffer . The suspension was pelleted , resuspended in 30 μl of PBS buffer containing 5 μg/ml FM 4–64 , and incubated for 5 min at room temperature . The cells were pelleted , resuspeneded in 30 μl PBS buffer , and were observed by spotting 4 μl of suspension on a cleaned microscope slide and immobilized with a poly-L-lysine-treated glass coverslip . For fluorescent microscopy of flagellar hooks , 1 . 0 ml of broth culture was harvested at 0 . 6–1 . 0 OD600 , resuspended in 50 μl of PBS buffer containing 5μg/ml Alexa Fluor 488 C5 maleimide ( Molecular Probes ) , incubated for 3 min at room temperature , and washed once in 1 . 0 ml of PBS buffer . The suspension was pelleted , resuspended in 30 μl of PBS buffer containing 5 μg/ml FM 4–64 , and incubated for 5 min at room temperature . The cells were pelleted , resuspeneded in 30 μl PBS buffer , and were observed by spotting 4 μl of suspension on a cleaned microscope slide and immobilized with a poly-L-lysine-treated glass coverslip . For PfliM-fliM-GFP microscopy , cells were grown at 37°C in LB broth to OD600 0 . 6–1 . 0 , resuspended in 30 μl PBS buffer containing 5 μg/ml FM 4–64 and incubated for 5 min at room temperature . The cells were pelleted , resuspeneded in 30μl PBS buffer , and were observed by spotting 4 μl of suspension on a cleaned microscope slide and immobilized with a poly-L-lysine-treated glass coverslip . For super-resolution microscopy , the OMX 3D-SIM Super-Resolution system was used . Supper-resolution microscopy was performed by using a 1 . 4-numerical-aperture ( NA ) Olympus 100X oil objective . FM4-64 was observed using laser line 561 and emission filter 609 nm to 654 nm , and GFP ( along with Alexa Fluor 488 nm ) was observed using laser line 488 nm and emission filter 500 nm to 550 nm . Images were captured using a Photometrics Cascade II electron-multiplying charge-coupled-device camera , processed using SoftWorx imaging software , and analyzed using Imaris software . Images that were processed as described above were used in Imaris ( Bitplane ) to determine the number and location of FliM-GFP or FlgE labeled with a fluorescent dye ( FlgET123C ) . The spots feature within the software labelled each puncta by the search parameter of identifying spots of 1 μM in the 488 nm wavelength . Cell pole positions were determined by using the 561 nm wavelength using the slice feature . The x , y , and z coordinates of each puncta in each cell was exported from Imars . Scripts were developed in MATLAB ( The Mathworks ) for importing two-dimensional Cartesian coordinates for basal body , hook , and cell lengths from Imaris . Cells were harvested from cultures growing at 37°C in LB broth . Cells were collected in 1 . 0 ml aliquots and suspended in an equal volume of Z buffer ( 40 mM NaH2PO4 , 60 mM NaHPO4 , 1 . 0 mM MgSO4 , 10 mM KCl , and 38 mM 2-mercaptoethanol ) . Lysozyme was added to each sample to a final concentration of 0 . 2 mg/ml and incubated at 37°C for 30 min . Each sample was diluted in Z buffer to a final volume of 500 μl , and the reaction was started with 100 μl of 4 mg/ml 2-nitrophenyl β-galactopyranoside in Z buffer and stopped with 250 μl of 1M Na2CO3 . The OD420 of the reaction mixture was measured , and the β-galactosidase-specific activity was calculated according the equation [OD420/ ( time X OD600 ) ] X dilution factor X 1000 . For sob and soa mutations contained within the flache promoter , a PCR product containing the flache promoter was amplified from B . subtilis chromosomal DNA ( either from strain 3610 or the appropriate suppressor strain ) using the primer set 1921/1922 . The Pflache PCR product was then sequenced using primer 1921 and 1922 individually . Genomic samples were fragmented using the Corvaris S220 ultrasonicator and then assayed using the Agilent TapeStatin using D1K HS tapes . The fagmenented samples were processed into Illumina DNA-Seq libraries using Bechman SWHT chemistry in conjunction with BioScientific NextFlex Adaptors on the BIomekFx automated workstation . The large ( 350bp-750bp ) size selection option was selected , and 12 μl of preamplified library was used as template in a 10 cycle amplification reaction . Following amplification , the libraries were cleaned using a 1X AmpureXP ratio and eluted in EB buffer . After Illumina sequencing , MiSeq reads were trimmed using a quality cutoff of 20 and remaining sequencing adapters were removed using Cutadapt 1 . 2 . 1 . FLASH 1 . 2 . 2 was then used to merge read pairs in which the forward and reverse read overlapped . The assemblies were performed using Newbler 2 . 7 using the merged reads as singletons and the unmerged reads as paired end reads . For SNP prediction , reads were mapped against both the NCIB 3610 reference and the DS234 assembly . Mapping was performed with bowtie 2 . 0 . 2 using the default parameters . Samtools was used to convert the mapping data to a pileup format . VarScan 2 . 3 . 2 was used to call SNPs , using parameters to require a minimum read depth of 20 and a minimum variant frequency of 90% . To verify the point mutation in sob28 , A PCR product containing fliG was amplified from B . subtilis chromosomal DNA strain DS9155 using the primer set 1229/3632 . The fliG PCR product was then sequenced using primer 788 to verify the sob mutation . To verify the deletion boundaries within sob24 and sob25 , A PCR product containing the deletion region was amplified from B . subtilis chromosomal DNA strain DS9151 and DS9152 using the primer set 3628/3629 . The PCR product was then sequenced using primer 3476 to verify the deletion boundaries . To determine the frequency of co-transduction of the sob22 mutation to the swrB::tet allele , a lysate was generated on the suppressor strain ( DS9149 ) and the swrB::tet allele was transduced to the wildtype 3610 strain . Three hundred of the resulting colonies were then picked onto 0 . 7% LB swarm agar plates to enumerate the number of swarm proficient colonies . The percentage of colonies that were swarm proficient was inversely proportional to the distance between the swrB::tet and the suppressor mutation . The primer set 1692/2293 was used to generate a PCR product 13kb upstream of the swrB::tet allele within the flache operon . The flache PCR product was then sequenced using primer 1601 . 25 ml of cells were grown in triplicate in LB broth at 37°C until the cultures reached 1 . 0 OD600 . 10 ml of each culture was then harvested into 15 ml conical tubes containing 1 . 25 ml of stop solution ( -20°C—5% phenol diluted in 100% ethanol ) and centrifuged at 5 , 800 rpm for 10 min at 4°C . The resulting pellets were then transferred to 1 . 5 ml microfuge tubes containing 500 μl of -80°C methanol , centrifuged at 14 , 000 rpm for 1 min at 4°C , and then stored at -80°C . The pellets where then resuspended in 850 μl TE buffer and then transferred to microfuge tubes containing 10 mg/ml lysozyme , inverted 4–6 times and placed at 37°C for 45 min . After this incubation , 50 μl of 10% SDS ( Sodium dodecyl sulfate ) was as added to the samples , the samples were inverted 4–6 times , and then 50 μl of 3M sodium acetate , pH 5 . 2 was added and the tubes were again inverted 4–6 times . Each sample was then split into to 500 μl volumes into separate microfuge tubes . 500 μl of phenol was added to each sample , the samples were inverted 10 times , and then placed in a 64°C water bath for 6 min . The tubes were inverted every minute during this water bath incubation . After the water bath incubation , the samples were centrifuged at 14 , 000 rpm for 10 min at 4°C . The resulting aqueous layer was then transferred to a fresh 1 . 5 ml microfuge tube where an equal volume of chloroform was added . The tubes were inverted 6–10 times , and then centrifuged at 14 , 000 rpm for 5 min at 4°C . The resulting aqueous layer was then transferred to a fresh 1 . 5 ml microfuge tubes where 1/10 the volume of 3M sodium acetate , pH 5 . 2 and 2 volumes of -20°C 100% ethanol were added to the tubes . The resulting mixtures were incubated at -80°C for 30 min and then centrifuged at 14 , 000 rpm for 40 min at 4°C . The ethanol layer was then removed , and the resulting white pellet was washed with 1 ml of -20°C 75% ethanol , centrifuged at 14 , 000 rpm for 5 min at 4°C . The ethanol layer was removed and the pellet was resuspended in 50 μl of RNase free water , and each split pool was rejoined together an incubated at 50°C to encourage resuspension . Each primer pair was diluted to both 1 μM and 5 μM stocks and subsequently mixed into 25 separate reactions in duplicate . Each of the 25 reactions was a permutation of one forward primer at either 50 nM , 100 nM , 300 nM , 600 nM , or 900 nM concentration with its corresponding reverse primer at either 50 nM , 100 nM , 300 nM , 600 nM , or 900 nM concentration along with SYBR Green SuperMix reagent ( Quanta Biosciences ) , and 106 copies of template . Each primer optimization assay also contained two additional control reactions containing forward and reverse primers each at either 50nM or 900 nM , Green SuperMix reagent ( Quanta Biosciences ) and no template . Quantitative PCR was conducted on each reaction set for each primer used in downstream quantitative PCR assays on the Stratagene MX3500 Pro thermocycler . Data were analyzed using the MXPro Stratagene software package . Total RNA was isolated as described above . Isolated RNA was DNase digested used TURBO DNA free ket ( Ambion ) . cDNA was reverse transcribed from each DNase-digested RNA sample using random dT primers of the qScript cDNA superMix ( Quanta Biosciences ) . Quantitave PCR was performed with specific primer pairs whose concentrations were optimized as described above and either diluted cDNA templeate , DNase digested RNA , or no template using SYBR Green SuperMix ( Quanta Biosciences ) on the Stratagene MX3500 Pro thermocycler . Data were analyzed using the MXPro Stratagene software package . The following primers where used: 1448/1449 ( sigD ) , 1450/1451 ( sigA ) , 1560/1561 ( flgB ) , 1562/1563 ( cheD ) , 1564/1565 ( hag ) , 1596/1597 ( fliF ) , 1598/1599 ( flgE ) , 1602/1603 ( flhA ) , 1604/1605 ( cheB ) , 1652/1653 ( fliI ) , 1654/1655 ( fliK ) , 1656/1657 ( fliY ) , 1658/1659 ( fliR ) , 1662/1663 ( cheW ) , and 3758/3759 ( fliP )
|
Bacteria build needle-like injectsomes to secrete toxins into host cells and build propeller-like flagella to swim through their environment using a molecular machine called the type III secretion system ( T3SS ) . Both the injectisome and the flagellum are large self-assembling complexes and regulation of the T3SS ensures that proteins are secreted sequentially for proper structure and function . Here we report genetic and cytological data that the SwrB protein of Bacillus subtilis helps the base of the flagellum adopt a completed conformation which in turn activates the enclosed T3SS to export proteins for the next stage of flagellar assembly . Thus SwrB presents a novel mechanism to supervise an early structural checkpoint regulating machine assembly . Targeting functional regulators like SwrB could inhibit T3SS-based strategies of pathogens .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Functional Activation of the Flagellar Type III Secretion Export Apparatus
|
Orthobunyaviruses such as Cache Valley virus ( CVV ) and Kairi virus ( KRIV ) are important animal pathogens . Periodic outbreaks of CVV have resulted in the significant loss of lambs on North American farms , whilst KRIV has mainly been detected in South and Central America with little overlap in geographical range . Vaccines or treatments for these viruses are unavailable . One approach to develop novel vaccine candidates is based on the use of reverse genetics to produce attenuated viruses that elicit immune responses but cannot revert to full virulence . The full genomes of both viruses were sequenced to obtain up to date genome sequence information . Following sequencing , minigenome systems and reverse genetics systems for both CVV and KRIV were developed . Both CVV and KRIV showed a wide in vitro cell host range , with BHK-21 cells a suitable host cell line for virus propagation and titration . To develop attenuated viruses , the open reading frames of the NSs proteins were disrupted . The recombinant viruses with no NSs protein expression induced the production of type I interferon ( IFN ) , indicating that for both viruses NSs functions as an IFN antagonist and that such attenuated viruses could form the basis for attenuated viral vaccines . To assess the potential for reassortment between CVV and KRIV , which could be relevant during vaccination campaigns in areas of overlap , we attempted to produce M segment reassortants by reverse genetics . We were unable to obtain such viruses , suggesting that it is an unlikely event .
The bunyaviruses are a large grouping of animal and plant-infecting viruses with a segmented , negative-stranded genome . The order Bunyavirales was recently proposed to include ‘bunyavirus like’ viruses that could not be assigned to the previous 5 genera ( https://talk . ictvonline . org/taxonomy/ ) [1] . This has resulted in the following new families: Feraviridae , Fimoviridae , Hantaviridae , Jonviridae , Nairoviridae , Phasmaviridae , Phenuiviridae and Tospoviridae; the remaining Peribunyaviridae family , previously called Bunyaviridae , contains the previous Orthobunyavirus genus which includes Cache Valley and Kairi viruses ( CVV and KRIV , respectively ) . This genus contains several emerging and re-emerging members that have caused disease in farmed livestock including Akabane virus ( AKAV ) in Africa and Asia , and Schmallenberg virus ( SBV ) in Europe [2–5] . CVV was first isolated from Culiseta inornata mosquitoes in Utah , USA in 1956 and has been detected in serosurveys in farm animals throughout North and Central America [6–14] . It has also been isolated from various other culicine and anopheline mosquitoes , including Aedes ( Ae . ) sollicitans , Ae . vexans , Ae . cinereus , Ae . albopictus , Anopheles ( An . ) punctipennis , An . quadrimaculatus , Coquillettidia perturbans , Mansonia titillans , Culex ( Cx ) . salinarius , several Ochlerotatus species , and Psorophora columbiae in the U . S . , Canada and Mexico [9 , 15–21] . Sheep are particularly affected and CVV causes abortions or congenital malformations in pregnant ewes [22–26] as well as other disease symptoms [14] . CVV continues to increase its geographical range and was recently diagnosed in sheep in Ontario and Quebec although the virus was detected in Ontario much earlier in 1977 [27–29] . This virus has also been detected in serosurveys of humans and has been linked to several cases of sometimes fatal meningitis and encephalitis [15 , 30–33] . Moreover different lineages of CVV are beginning to emerge and a subtype , Maguari virus ( MAGV ) , is also associated with disease in humans [34 , 35] . KRIV , like CVV , belongs to the Bunyamwera serogroup . It is another example of a potentially emerging virus of the Americas and was first isolated from various mosquito species in Trinidad including Aedes , Wyeomia , Culex and Psorophora ssp . [36] . It has also been isolated from mosquitoes and vertebrates in Central and South America , including from a febrile horse in Argentina [37–39] . In one serosurvey , antibodies ( Abs ) were identified in 6–18% of humans and up to 48% of horses [7 , 40 , 41] . KRIV does not cause any documented clinical disease symptoms in humans or animals . Although the geographical ranges of CVV and KRIV , in North and South America respectively are mostly distinct ( though MAGV has been detected in South America ) , they have both been isolated from the Yucatan peninsula of Mexico along with the closely related virus , Cholul ( CHLV ) , which was suggested to be a reassortant of CVV and the related virus Potosi ( POTV ) [7 , 42] . POTV itself was suggested to be a reassortment of CVV and KRIV or a closely related virus [43] . Although this shared host range is limited , it may change and suggests the potential for reassortment between CVV and KRIV as well as other or yet unknown orthobunyaviruses to generate novel viruses . This is relevant to consider in vaccine design and vaccination studies , as vaccines could reassort with naturally circulating , related viruses . Among the unifying characteristics of the previous family Bunyaviridae and the vast majority of the new order Bunyavirales members is the possession of a tri-segmented single-stranded genome of negative or ambi-sense polarity that encodes four structural proteins . The three genome segments ( called L [large] , M [medium] and S [small] ) are encapsidated by the nucleocapsid ( N ) protein and are associated with the viral RNA-dependent RNA polymerase , the L protein , to form ribonucleoprotein complexes ( RNP ) termed nucleocapsids . RNPs are contained within a lipid envelope also containing the viral glycoproteins , Gn and Gc . Virus replication occurs in the cytoplasm of infected cells , and viruses mature primarily by budding from Golgi membranes . As well as the four structural proteins ( L , Gn , Gc and N ) many bunyaviruses encode two non-structural proteins , termed NSs and NSm . Genetic and biochemical analyses have shown that the S RNA segment encodes the N and NSs proteins; the M RNA segment encodes Gn , Gc and NSm as a polyprotein precursor; and the L RNA segment encodes the L protein [44–50] . The NSs proteins have been linked to viral virulence [51] . Reverse genetics systems for several orthobunyaviruses including Bunyamwera virus ( BUNV ) , SBV , AKAV and La Crosse virus ( LACV ) have been developed over the last few decades [52–56] and studies on these viruses have increased our understanding of these pathogens . More importantly the use of reverse genetics systems of orthobunyaviruses have been used extensively to show that NSs proteins are important antagonists of type I interferon ( IFN ) responses [54 , 55 , 57–65] . Engineered orthobunyaviruses that do not express NSs or NSm proteins have also been assessed as candidate live vaccines for SBV [66] . The original protocol for BUNV reverse genetics was significantly improved by use of the T7 RNA polymerase-expressing cell line BSR-T7/5 [67 , 68] . This allowed recovery of BUNV directly from the transfected cells without a subsequent insect cell passage . It was also found that the antigenomic plasmids provided low levels of support proteins which were sufficient for an efficient rescue even in absence of support plasmids expressing L and N proteins [68] . Thus , recombinant BUNV can be generated just from antigenomic RNA . Here we have established minigenome and reverse genetics systems for CVV and KRIV and showed that deletion of NSs leads to attenuation and viruses that potently induce type I interferon responses . As CVV and KRIV overlap geographically in some areas but are phylogenetically distant we also assessed whether these viruses can reassort , by reverse genetics . No reassortants at least for the M segment were obtained; and vaccination with attenuated CVV in areas of shared geographical range is unlikely to result in reassortment between a CVV vaccine candidate and KRIV at least for this segment . This is important information to assess the environmental risk associated with vaccination .
To develop reverse genetic systems for CVV and KRIV , viral segments had to be cloned and this required accurate sequence data . We used the CVV strain 6V633 , for which no sequence information was available at the time of cloning although an almost complete sequence was available for another strain ( MNZ-92011 ) . More recently sequence information was submitted to Genbank for the CVV strain 6V633; see Table 1 . For KRIV we used strain TR8900 , obtained by R . M . Elliott ( an older designation for prototype KRIV TRVL8900; to distinguish virus used in this study , we will refer to our isolate as TR8900 and TRVL8900 for other published KRIV strain data ) . GenBank contained partial or full sequence for the S , M and L segments of different strains ( Mex 07 and BeAr8226 ) though only partial information for KRIV TRVL8900 including a previously described S segment sequence for KRIV ( strain not being specified in the supporting paper [69] but from records is likely to be TR8900; GenBank , accession number X73467 . 1 ) which was generated by using consensus primers for the S segment termini . These sequences were used to design gene specific primers for RACE analysis to amplify the full 5’ and 3’ untranslated regions ( UTRs ) to obtain the precise sequence of the 5’ and 3’ ends for all 3 segments . This allowed the design of RT-PCR primers to amplify and clone cDNA of the antigenomic RNAs for all 3 segments of both viruses . All segments for L , M and S antigenomes were cloned into pTVT7R [70] , and sequenced . The coding sequences for L and N ( which include the NSs open reading frame [ORF] ) , were also subcloned into the expression plasmid pTM1 for both viruses [71] . To confirm sequence data obtained following cloning and sequencing , next generation sequencing ( NGS ) was performed on total cell RNA from BHK-21 cells infected with wild type ( wt ) CVV as described in the Methods . Results were obtained for CVV ( read depth averaging 337 to 1695 reads/nucleotide; and with > 99 . 4% coverage ) . For KRIV , two separate batches of viral RNA from supernatant of approximately 20 ml and 50 ml from infected BHK-21 cells were used for NGS analysis . Results were obtained for KRIV ( read depth >8836 reads/nucleotide; and >99 . 9% coverage ) , and we compiled consensus sequences for all three segments for both viruses with the information above ( NGS and cDNA sequences ) ; see GenBank accession below ( Table 1 ) . Full NGS data is available at the European Nucleotide Archive ( ENA ) with accession number PRJEB25770 . A comparison of sequencing data was carried out to assess our consensus sequences for all segments . For CVV three differences were noted between the recently available GenBank reference for strain 6V633 and our NGS and cDNA sequence data ( Table 2 ) ; occasional mismatches at the extreme 3’ and 5’ termini following NGS were discounted . Two differences resulted in amino acid changes; one each in the M segment polyprotein and L protein . In another instance our sequence data of cloned virus-derived cDNA differed from the NGS and the reference sequence for a nucleotide in the 3’ UTR region of the antigenome of the CVV M segment . This was likely an error introduced when cloning the CVV M segment and the NGS and reference strain sequence are in all likelihood the correct sequences . This mutation was then corrected in our CVV M segment cDNA clone for virus rescue . Comparing our consensus sequence data for KRIV ( strain TR8900 ) to the partial reference sequences available for the prototype TRVL8900 strain in GenBank , six nucleotide differences were found ( Table 2 ) . Two of the differences were in the non-coding regions whereas three other differences resulted in amino acid changes- two in nucleoprotein N and one amino acid change in the M segment polyprotein . The final nucleotide difference in KRIV M segment resulted in a synonymous codon in the polyprotein coding sequence . The different sequence data sets we generated by NGS and RACE analysis/sequencing of cloned segments were in complete agreement for KRIV ( again mismatches at the extreme 3’ and 5’ termini following NGS were discounted ) . Additional sequences for KRIV segments ( strain TRVL8900 ) , but mostly missing the extreme termini ( with the following accession numbers: MH484302 , MH484301 , MH484300; see Table 1 ) were , apart from the missing termini , identical to those described here . For CVV , pTVT7R-based plasmids containing the full length antigenomic cDNAs , were used in a three-plasmid rescue system , by transfection into the T7 RNA polymerase-expressing BSR-T7/5 cells without the use of additional expression plasmids for L and N proteins . The supernatant was harvested after 3–4 days , with highly visible cytopathic effect ( CPE ) and the virus was titrated by plaque assay . Similarly , for KRIV plasmids containing the full length antigenomic cDNAs , cloned into pTVT7R , were used in a five-plasmid rescue system . This included the use of expression plasmids ( pTM1-KRIVL and pTM1-KRIVN ) for the L and N proteins as described previously [68] . The supernatant media was harvested after 6 days , when CPE was visible , and the virus titrated in a plaque assay . The rescue of both viruses was repeated twice to assess robustness of the assay . The plaque morphology of the recombinant wild type viruses , designated rKRIV and rCVV , was indistinguishable from that of the authentic wt viruses ( Fig 1 ) . To investigate the function of the NSs proteins of both viruses , mutations were introduced into the NSs ORFs to produce recombinant viruses that no longer expressed this protein . These mutations were introduced in such a way as to prevent amino acid changes in the overlapping N protein ORF ( Fig 2 ) . Viruses denoted by rCVVdelNSs and rKRIVdelNSs were rescued as above by substituting the plasmids containing wt S segments with those containing mutations in the NSs genes . Nucleotide sequences of recovered viruses were determined by RT-PCR and sequencing except for the extreme termini , which were only sequenced by RACE for S of rCVV and rCVVdelNSs ( see below ) . All sequences obtained matched those of the parental plasmids and mutations in the S segments of recombinant rCVVdelNSs and rKRIVdelNSs were confirmed . The identities of the wt and recombinant viruses were also confirmed by western blotting of lysates of infected BHK-21 cells using polyclonal Abs raised against peptide regions of the N proteins ( Fig 1b ) . We observed N levels consistently higher in rCVVdelNSs-infected cells compared to wt CVV and rCVV . As we did not find any mutations ( including termini , as determined by RACE analysis of rCVVdelNSs , and for completeness also rCVV S segments ) that could explain these observations , we speculate that removal of the NSs ORF is sufficient to lead to higher levels of N . Growth of wt and recombinant CVV and KRIV were compared in BHK-21 and Vero-E6 cells infected at a multiplicity of infection ( MOI ) of 0 . 1 ( Figs 3 and 4 ) . No differences were observed in growth kinetics between wt and recombinant viruses in the two cell lines . Growth of the recombinant viruses rCVV and rKRIV and the NSs deletant rCVVdelNSs and rKRIVdelNSs were compared in the type I IFN competent A549 cell line with growth in A549 NPro cells , that express BVDV NPro protein which inhibits type 1 IFN production [72] . rCVVdelNSs displayed slower growth kinetics in A549 cells than rCVV , although in A549 NPro cells the two viruses displayed identical growth kinetics . Similar growth patterns were observed for rKRIV and rKRIVdelNSs in A549 and A549 NPro cells . Sheep are the main affected livestock species therefore the growth of rCVV and rCVVdelNSs were also compared in the ovine cell line SFT-R [66] . rCVVdelNSs displayed lower replication in SFT-R cells compared to rCVV . Due to the significant infection of the economically important horse species the growth of rKRIV and rKRIVdelNSs were also compared in the equine cell line E-Derm ( NBL-6 ) , following infection with a MOI of 3 . Both recombinant viruses showed replication , with lower growth for rKRIVdelNSs ( Fig 4 ) . Since CVV and KRIV are both transmitted by mosquitoes , we also studied replication of these viruses in a commonly used mosquito cell line . Growth of the recombinant viruses rCVV , rKRIV as well as rCVVdelNSs and rKRIVdelNSs were assessed in Ae . aegypti-derived Aag2 cells infected at a MOI of 0 . 1 . rCVV and rCVVdelNSs productively infected Aag2 cells ( Fig 3 ) ; both rKRIVdelNSs and rKRIV productively infected Aag2 cells though titres were low ( Fig 4 ) . A biological assay used previously to monitor type I IFN production in response to orthobunyavirus infection was employed here to investigate the induction of type I interferon by rCVV , rKRIV , rCVVdelNSs and rKRIVdelNSs [73 , 74] . A549 cells were infected with recombinant viruses ( MOI = 1 ) and following this , UV-inactivated medium from these cells was used to treat fresh A549 NPro cells which can respond to but cannot produce type I interferons . If present and active in the medium , type I IFN would induce an antiviral state and the cells would be protected from subsequent infection with a challenge virus , encephalomyocarditis virus ( EMCV ) . Recombinant BUNV ( rBUNV ) and BUNV with a deleted NSs protein ( rBUNVdelNSs2 ) were used as negative and positive controls , respectively [61] . The relative amounts of type I IFN produced were calculated according to the highest dilution of supernatant affording protection to the cells from EMCV infection as described in Methods . As shown in Fig 5 , medium from A549 cells infected with wt viruses contained much less IFN than medium infected with their NSs-lacking counterparts . Medium from rCVVdelNSs , rKRIVdelNSs and rBUNVdelNSs2-infected cells largely protected A549 NPro cells from EMCV infection , indicating induction of IFN in the initial infections . Thus , the NSs proteins of CVV and KRIV suppress the production of type I IFN . Phylogenetic analysis of CVV and KRIV open reading frames as well as comparisons of nucleotide and amino acid identities indicate that CVV and KRIV are not closely related ( Fig 6 , Table 3 ) . However as it is of interest to know whether such viruses could still reassort , for example post vaccination in areas of geographical overlap , we employed several genetic tools developed here . Previous work has shown the intracellular reconstitution of replication active BUNV nucleocapsids from transiently expressed components [75] . Here we developed similar minigenome systems for CVV and KRIV . The minimal components of this system are expression plasmids for L and N proteins and a minigenome; here an analogue of viral RNA supplied as a negative sense reporter gene ( Renilla luciferase , Ren ) cloned within the M genome segment UTRs and transcribed by T7 RNA polymerase from the corresponding promoter . The CVV-derived minigenome contained one change described above in Table 1 ( position 4448 of the antigenome or 16 of the genome; the minigenome was active , as shown below ) . Plasmids were transfected into BSR-T7/5 CL21 cells and the results showed strong increase in luciferase levels above negative control levels ( no expression of either N or L proteins , or both ) , demonstrating that we were able to reconstitute replication active CVV and KRIV nucleocapsids ( Fig 7 ) . In order to explore the reassortment potential of KRIV and CVV with each other , we also investigated if the minigenomes containing the M segment UTRs for both viruses could be swapped between the 2 minigenome systems . The results showed that for both CVV and KRIV , minigenomes are interchangeable and that functional nucleocapsids can be formed , at least with M-derived minigenomes ( Fig 7 ) . To assess if we could obtain hybrid viruses between CVV and KRIV , we used the 3 and 5 plasmid rescue system described above , and swapped the M segments because this is a frequent natural reassortment combination [76] , though other reassortment combinations may be possible and were not assessed here . However , several attempts to rescue such hybrids , failed to produce viruses ( Fig 8 ) .
In this study we developed reverse genetics system for two orthobunyaviruses , CVV and KRIV , and used these to produce recombinant viruses ( rCVVdelNSs and rKRIVdelNSs ) that no longer express NSs . These viruses were attenuated across several cell lines and IFN protection assays confirmed the role of the KRIV and CVV NSs proteins as type I interferon antagonists in mammalian cells , as expected from previous studies [51] . CVV and KRIV did not require NSs to support replication in the mosquito cell line Aag2 . This has also been observed for BUNV and SBV in the mosquito cell line C6/36 [53 , 77] . It would also be informative to test the growth of CVV and KRIV in other mosquito cell lines to further assess the role of species and immune status in replication . The production and characterisation of NSs-deletant , recombinant CVV is a first step towards developing an attenuated viral vaccine for this increasingly important livestock pathogen . Further attenuation could possibly be achieved by also deleting the NSm protein as described for SBV [66] . We also assessed the reassortment potential of CVV and KRIV . Indeed reassortment between orthobunyaviruses has been shown to be a key driver of orthobunyavirus evolution [43] and pathogenic viruses such as Ngari are reassortants of other known “parent” viruses , BUNV and Batai virus [43 , 78 , 79] . SBV is also a potential Shamonda/Sathuperi virus reassortant [80] . Indeed , POTV has been suggested to be a reassortant of CVV and KRIV , or a closely related virus [43] . In areas of overlap , vaccination with attenuated CVV could thus potentially lead to novel reassortants and should be considered in risk assessments . As CVV and KRIV L and N can replicate M segment derived minigenomes of either virus , we assessed compatibility of M segments between CVV and KRIV in virus rescues . This is the predominant segment that is acquired in novel hybrid orthobunyaviruses found in nature and evidence suggests that this is due to the interaction between N and L leading to linkage of L and S segments [76 , 81]; though viruses such as CHLV ( likely a combination of the M and L segments from POTV and the S segment from CVV [42] ) have been described . However , here we were unable to rescue the hybrid viruses containing alternate M segments . There could be several reasons for this; low efficiency of the rescue system , or that these hybrid viruses are unable to propagate . Several lines of evidence also suggest an interaction between both of the bunyavirus glycoprotein cytoplasmic tails and nucleocapsids , which could also influence the packaging of a hybrid virus [44 , 82] . In summary , we describe the generation of rescue systems for both CVV and KRIV and use these to demonstrate that the deletion of NSs leads to attenuated viruses no longer able to inhibit type I interferon responses . An attenuated CVV or KRIV containing a deleted NSs gene could serve as vaccine candidate . Additionally , reassortment by swapping CVV and KRIV M segments could not be demonstrated here by using reverse genetics . The poor conservation between CVV and KRIV M segments ( Table 3 and Fig 6 ) may explain this observation . However other reassortment combinations can now be explored using reverse genetics systems , including those with more or less closely related orthobunyaviruses . We emphasize that reassortment by reverse genetics may have technical limitations and co-infection experiments or co-infections in nature may generate reassortants that are not obtained by such approaches , or that different plasmid combinations and concentrations used for reverse genetics may still generate other types of viral reassortants . Indeed , as discussed above , CHLV virus ( M and L segments likely from POTV , S segment from CVV ) [42] suggests that different reassortment outcomes are possible . Reverse genetics systems for CVV and KRIV , as well as other orthobunyaviruses , will be useful to test to what extent reassortment between these viruses is possible . To conclude , the systems developed here are relevant to further studies on orthobunyaviruses that include reassortment and vaccine design .
BHK-21 cells ( provided by R . M . Elliott , University of Glasgow , UK ) were grown in Glasgow’s minimal essential medium ( GMEM ) supplemented with 10% tryptose phosphate broth ( TPB ) , 10% newborn calf serum ( NBCS ) , 1000 units/ml penicillin and 1 mg/ml streptomycin ( p/s ) . BSR-T7/5 cells , which stably express T7 RNA polymerase [67] were provided by K . -K . Conzelmann ( Max-von-Pettenkofer Institute , Munich , Germany ) and grown in GMEM supplemented with 10% TPB , 10% fetal calf serum ( FCS ) , p/s and 0 . 25 mg/ml G418 . These cells were used to rescue rCVV , rKRIV , rCVVdelNSs and rKRIVdelNSs . The BSR-T7/5 cells were also sub-cloned by dilution cloning to create a population of cells with a higher expression of T7 RNA polymerase; this new cell line developed by us was designated BSR-T7/5 CL21 [83] and used for minireplicon studies and rescues of virus reassortants . SFT-R cells ( CCLV-RIE 43 , Collection of Cell Lines in Veterinary Medicine [CCLV] ) , were obtained from the Friedrich-Loeffler-Institute , Greifswald-Insel Riems , Germany . These cells were grown in DMEM supplemented with 10% FCS and low glutamine ( 1g/L ) . E-Derm cells ( also called NBL-6 ) ( CCL-57 , ATCC ) were grown in DMEM containing 15% FBS and 1% NEAA . Vero-E6 cells were also provided by R . M . Elliott ( University of Glasgow , UK ) . The A549 and A549 NPro cell lines were a kind gift from R . E . Randall ( University of St . Andrews , UK ) . A549 , A549 NPro and Vero-E6 cells were maintained in Dulbecco’s modified Eagle’s medium ( DMEM ) supplemented with 10% FCS and p/s . The A549 NPro cell media also contained 10 μg/ml blasticidin or 2 μg/ml puromycin as a selection agent . All mammalian cell lines were grown at 37 °C with 5% CO2 . Aag2 cells were obtained from P . Eggleston ( Keele University , UK ) and grown in L-15 medium with 10% FBS , 10% tryptose phosphate broth and p/s . All media and reagents were purchased from Gibco , Life Technologies . BHK-21 cells were chosen to propagate and titre CVV and KRIV in plaque assays . CVV ( strain 6V633 ) and KRIV isolates ( strain TR8900 ) were made available by R . M Elliott . The isolates were used to infect Vero-E6 cells and propagated for 5–6 days at 33 °C . Both viruses were then titrated by plaque assay , for initial characterisation studies on BHK-21 and Vero-E6 cells , under an overlay comprising MEM supplemented with 2% NBCS and 0 . 6% Avicel ( FMC ) and incubated at 37 °C for 3–5 days . Cell monolayers were fixed with 4% formaldehyde and plaques were visualized by staining with crystal violet staining solution . A working stock for both viruses was made by growing these in BHK-21 cells for 5–6 days at 33 °C at a MOI of 0 . 05 . Recombinant viruses ( rCVV , rCVVdelNSs , rKRIV and rKRIVdelNSs ) were treated similarly , except rKRIVdelNSs which required 5–6 days to develop plaques on BHK-21 cells . Recombinant rBUNV and BUNV containing a deleted NSs protein ( rBUNVdelNSs2 ) were used as controls in the IFN bioassay [53] . EMCV used in the IFN protection assay was obtained from R . E . Randall , University of St . Andrews , UK . Antibodies were produced in collaboration with Cambridge Research Biochemicals . Peptide regions of the N proteins of CVV and KRIV , that were different in sequence from each other , were selected to raise polyclonal Abs to distinguish between both viruses ( http://www . discoveryantibodies . com/anti-cache-valley-virus-cvv-nucleocapsid-antibody ) ( catalogue number crb2005018 ) , ( http://www . discoveryantibodies . com/anti-kairi-virus-kriv-nucleocapsid-antibody ) ( catalogue number crb2005080 ) . Total cell RNA was isolated from BHK-21 cells infected with wt virus in 6 well plates or 25 cm2 flasks for recombinant viruses , and purified using Trizol according to the manufacturer’s instructions . Viral RNA was isolated from supernatant of infected BHK-21 cells grown in 150 cm2 flasks collected at 5 days post infection . Briefly , supernatant was first clarified by centrifugation at 4000 RPM for 10 min . Then virus was concentrated by ultracentrifugation at 26000 RPM for 90 min on 20% sucrose cushion in PBS . RNA from pelleted virus particles was isolated using Trizol as above . Samples were prepared using an Illumina TruSeq Stranded RNA kit . Paired end data was generated with 2x150bp on MiSeq for CVV . Initial quality assessment was done using FastQC ( https://www . bioinformatics . babraham . ac . uk/projects/fastqc/ ) . The sequencing adaptors and sequence reads with Phred quality score less than 33 were trimmed using Trim Galore ( https://www . bioinformatics . babraham . ac . uk/projects/trim_galore/ ) . Quality trimmed and cleaned sequences were mapped to the reference genome segments for CVV ( KC436106 . 1 , KC436107 . 1 and KC436108 . 1 ) using short read mapper Tanoti ( http://www . bioinformatics . cvr . ac . uk/tanoti . php ) and consensus sequences were generated using SAM2CONSENSUS ( https://github . com/vbsreenu/Sam2Consensus ) . Assembly statistics including number of mapped reads , depth and breadth of coverage were generated by using the weeSAMv1 . 1 software package ( https://github . com/josephhughes/Sequence-manipulation/blob/master/weeSAMv1 . 1 ) . Two samples of KRIV were also sequenced using the MiSeq protocol and paired end data for the samples were analysed using the bioinformatics methods described above . However , Bowtie2 [84] was deemed to be a better aligner for the KRIV data and was used for the reference mapping . Short reads from KRIV samples were mapped to all segments by using cDNA ( RACE/cloned segment ) sequences obtained previously for the TR8900 strain and consensus was called using SAM2CONSENSUS . Default parameters were applied to all software packages used in this analysis . Supernatant ( 1 ml ) from infected BHK-21 cells was concentrated in a Vivaspin 500 ( Sartorius ) and used to extract viral RNA ( QIAamp viral RNA extraction kit ) . The viral RNA was used for RACE with a poly-A tailing Kit ( Ambion ) and standard PCR methods . The RNA was also used for first strand synthesis using primers designed from sequences derived from RACE analysis ( Superscript III First Strand Synthesis System ) ( Thermofisher ) . All cDNAs were then amplified using Phusion high fidelity polymerase ( NEB ) purified , digested , and ligated into pTVT7R that had been linearized with BbsI [70] . The CVV S genomic segment was ligated into pTVT7R using BsmBI sites and the CVV M genomic segment was ligated into pTVT7R similarly using BfuAI sites also added to the 5’ and 3’ ends of the RT-PCR primers . The CVV L genomic segment was synthesized from a deep sequencing consensus , re-amplified and cloned into pTVT7R using the In-Fusion system . Resulting plasmids were named pTVT7R-CVVS , pTVT7R-CVVM and pTVT7R-CVVL . The KRIV S segment was cloned into pTVT7R using BsmBI sites that were added to both RT-PCR primers . The KRIV M and KRIV L genomic segments were ligated into pTVT7R using the In-Fusion system ( Clontech ) . The plasmids were named pTVT7R-KRIVS , pTVT7R-KRIVM and pTVT7R-KRIVL . A G residue was added to the 5’ end of all cDNAs to increase transcription efficiency of the antigenome transcripts by T7 RNA polymerase . Viral RNA polymerase L and nucleocapsid N ( containing the NSs ORF ) protein expression constructs were also constructed by subcloning the cDNAs into the expression vector pTM1 [71] . This was achieved using PCR and the In-Fusion system ( Clontech ) . The plasmids created were named pTM1-CVVL , pTM1-CVVN , pTM1-KRIVL and pTM1-KRIVN . All primers are available from the authors on request . To stop expression of NSs expression by CVV S and KRIV S segments , start codons were removed and stop codons introduced by PCR mutagenesis using a method based on the Quikchange site-directed mutagenesis system ( Stratagene ) . All mutations were designed to prevent amino acid mutation in the overlapping N proteins . For CVV NSs , 2 rounds of mutagenesis removed 2 start codons at amino acid positions 2 and 30 and introduced 2 stop codons at positions 3 and 28 in the NSs ORF to create the plasmid pTVT7R-CVVdelNSs ( Fig 2 ) . For KRIV NSs , 3 rounds of mutagenesis removed 3 start codons at amino acid positions 2 , 17 and 30 and introduced 2 stop codons at positions 3 and 28 in the NSs ORF to create the plasmid pTVT7R-KRIVdelNSs ( Fig 2 ) . BSRT-7/5 cells ( 2x105 per well ) were seeded in 6 well plates and the media changed to GMEM supplemented with 10% TPB and 2% FCS on the day of transfection . Plasmids containing viral cDNAs ( 0 . 5 μg per plasmid ) were transfected using Trans-IT LT-1 ( Mirus ) and the development of CPE monitored over 3 days . Supernatant was harvested after highly visible levels of CPE or after 6 days and titrated in a plaque assay to confirm the presence of rescued virus . For the attempted rescue of CVV and KRIV hybrid viruses , BSRT-7/5 CL21 cells ( 6x105 per well ) were seeded in small flasks ( 25 cm2 ) . The media was changed to GMEM supplemented with 10% TPB and 2% FCS on the day of transfection . Plasmids containing viral cDNAs ( 1 μg per plasmid ) were transfected using Trans-IT LT-1 . The development of CPE was monitored after 3 days . After 6 days , with no visible CPE , supernatant was harvested and added to small flasks ( containing BHK-21 cells ) before testing for the presence of virus . 5 plasmid rescues were attempted using expression plasmids for N and L . Plaque assays were performed ( staining with crystal violet ) , or western blotting of passaged virus using anti KRIV and anti CVV N antibodies . Minigenome reporter constructs were created by substituting the polyprotein sequence of the M genomic segment with the Renilla luciferase ( Ren ) gene . The constructs had in order the following components; T7 RNA polymerase promoter , 5’UTR of the genome , ( Ren ) luciferase ( -ve sense ) , 3’UTR of the genome , hepatitis delta ribozyme and T7 RNA polymerase terminator . The resulting plasmids , designated pUC57-T7-CVVMRen ( - ) and pUC57-T7-KRIVMRen ( - ) were synthesized by Genscript . BSR-T7/5 CL21 cells ( 1 . 5x105 per well ) in 12-well plates were co-transfected with 500 ng each of pTM1-based L and N protein expression plasmids , and M-based reporter plasmids , 10 ng of pTM1-FF-Luc as a transfection control , and appropriate amount of empty pTM1 vector to equalise total amounts of plasmid DNA; using 1 . 5 μl of Trans-IT LT1 per reaction . Renilla and Firefly luciferase activities were measured at 24 hrs post-transfection using a Dual-Luciferase Assay kit ( Promega ) according to the manufacturer’s protocol . This assay was carried out essentially as described [73] . In brief A549 cells ( 5x104 per well ) , grown in a 24 well plate , were infected with the different viruses as indicated at an MOI of 1 and incubated at 37 °C for 48 h . The supernatant fluid was clarified by centrifugation and residual virus inactivated by UV treatment . Thereafter twofold serial dilutions of the medium were applied to fresh A549 NPro cells grown in 96-well plates for 24 hrs . The cells were then infected with EMCV , which is sensitive to IFN , and incubated for 4 days at 37 °C . Cells were then fixed with formaldehyde and stained with crystal violet to monitor the development of CPE . The production of IFN was calculated according to the highest dilution of supernatant giving protection against EMCV infection and is expressed as relative IFN units . The newly sequenced full genomes of CVV ( 6V633 ) and Kairi virus ( TR8900 ) were collated with sequences from GenBank . The protein sequences were aligned using MAFFT [85] and the best substitution model was selected using ProtTest [86] using the Bayesian Information Criterion ( BIC ) . A maximum likelihood tree was reconstructed using the best substitution model with RAxML [87] using 1000 bootstrap replicates for node support . Data for Figs 3 , 4 and 7 are available under http://dx . doi . org/10 . 5525/gla . researchdata . 599 while database accession numbers for sequencing data are available as indicated above .
|
Cache Valley and Kairi viruses ( CVV and KRIV; Peribunyaviridae , Orthobunyavirus ) are important animal pathogens of the Americas . In this study we developed reverse genetics systems to study and manipulate viral genomes of both viruses . Viral genomes were mutated to prevent the expression of the NSs protein , a key virulence factor and antagonist of the type 1 interferon ( IFN ) system . Replication studies in IFN producing cell lines showed slower growth of the NSs-deletion carrying viruses compared to wild type virus . In contrast , in IFN-deficient cell lines growth of both viral types was comparable , highlighting the role of NSs as an IFN antagonist in both CVV and KRIV . We also demonstrated using genetic studies that CVV and KRIV are unlikely to combine by reassortment to form novel viruses at least for one combination tested here , and propose that such recombinant viruses would be suitable live attenuated vaccine candidates .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"sequencing",
"techniques",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"microbiology",
"plasmid",
"construction",
"viruses",
"next-generation",
"sequencing",
"rna",
"viruses",
"genome",
"analysis",
"dna",
"construction",
"molecular",
"biology",
"techniques",
"microbial",
"genetics",
"reverse",
"genetics",
"bunyaviruses",
"research",
"and",
"analysis",
"methods",
"genomics",
"proteins",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"recombinant",
"proteins",
"molecular",
"biology",
"nucleotide",
"sequencing",
"viral",
"genetics",
"biochemistry",
"virology",
"viral",
"pathogens",
"interferons",
"transcriptome",
"analysis",
"biology",
"and",
"life",
"sciences",
"genetics",
"viral",
"gene",
"expression",
"computational",
"biology",
"dna",
"sequencing",
"organisms"
] |
2018
|
Development of reverse genetics systems and investigation of host response antagonism and reassortment potential for Cache Valley and Kairi viruses, two emerging orthobunyaviruses of the Americas
|
Large-scale sequencing efforts have documented extensive genetic variation within the human genome . However , our understanding of the origins , global distribution , and functional consequences of this variation is far from complete . While regulatory variation influencing gene expression has been studied within a handful of populations , the breadth of transcriptome differences across diverse human populations has not been systematically analyzed . To better understand the spectrum of gene expression variation , alternative splicing , and the population genetics of regulatory variation in humans , we have sequenced the genomes , exomes , and transcriptomes of EBV transformed lymphoblastoid cell lines derived from 45 individuals in the Human Genome Diversity Panel ( HGDP ) . The populations sampled span the geographic breadth of human migration history and include Namibian San , Mbuti Pygmies of the Democratic Republic of Congo , Algerian Mozabites , Pathan of Pakistan , Cambodians of East Asia , Yakut of Siberia , and Mayans of Mexico . We discover that approximately 25 . 0% of the variation in gene expression found amongst individuals can be attributed to population differences . However , we find few genes that are systematically differentially expressed among populations . Of this population-specific variation , 75 . 5% is due to expression rather than splicing variability , and we find few genes with strong evidence for differential splicing across populations . Allelic expression analyses indicate that previously mapped common regulatory variants identified in eight populations from the International Haplotype Map Phase 3 project have similar effects in our seven sampled HGDP populations , suggesting that the cellular effects of common variants are shared across diverse populations . Together , these results provide a resource for studies analyzing functional differences across populations by estimating the degree of shared gene expression , alternative splicing , and regulatory genetics across populations from the broadest points of human migration history yet sampled .
A central challenge in modern medical and population genomics is identifying trait-disposing genetic variants , interpreting their molecular consequences , and determining the transferability of their functional roles across individuals and populations . While genome-wide association studies ( GWAS ) have correlated an abundance of common and ( increasingly ) rare variants with disease , far fewer studies have pinpointed causal variants , discovered the biological mechanism of the association , or replicated their findings in different populations . Here , we build upon previous work using transcript abundance and splicing as model systems for understanding how population substructure can impact the genetic architecture of biomedical traits [1]–[4] . In particular , we focus on a set of populations that span the “Out-of-Africa” migration of anatomically modern humans using CEPH Human Genome Diversity Panel cell lines , for which we have collected an extensive ‘omics profile described below . Genetic studies of microsatellites panels and single nucleotide polymorphisms ( SNPs ) have shown a decrease in genetic diversity as a function of a population's geographic distance from eastern or southern Africa [5]–[7] . This pattern fits a serial founder effect model , but it remains unclear whether transcriptome variation follows this pattern and how closely genetic effects on regulation mirror human migration history . Previous work has shown that population bottlenecks reduce heterozygosity and are associated with an accumulation of damaging and loss-of-function variation which can impact gene expression [8] , [9] . However , further molecular work is needed to settle the controversy regarding demography and its impacts on the distribution of functional genetic variation among populations . Gene expression studies within and between well-studied populations have been transformative in cataloging gene expression differences , expression quantitative trait loci ( eQTLs ) with different types of regulatory variants , as well as allele-specific expression ( ASE ) that underlie many disease associations [3] , [10]–[16] . Technological advances in RNA sequencing and transcript assembly have also enabled analysis of variation in transcript structure and regulation of alternative splicing . For example , splicing ratios can differ between distant populations even in the absence of expression differences , and some population-specific splicing differences are involved in known disease-susceptibility genes that correspond with differences in prevalence [4] , [17] . Additionally , thousands of unannotated transcripts have been identified within populations [18] , [19] , highlighting the difficulty in distinguishing population-specific transcripts that are functionally relevant versus those that simply arise from noisy splicing [20] . Elucidating how gene expression regulation and splicing are impacted by historical human migrations will aid functional interpretation of the genome and improve our understanding of the transferability and evolution of genetic regulation across populations . This study aims to characterize regulatory , splicing , and expression differences via RNA sequencing across a global sampling of seven populations from the HGDP . We have also performed medium pass genome ( ∼8X ) and high coverage ( ∼96X ) exome sequencing of these individuals , enabling us to characterize genetic effects on transcriptome variation . These integrated DNA and RNA sequencing datasets are generated from the broadest points of human migration history yet sampled , and serve as a resource for future studies analyzing functional differences across populations .
We randomized library preparations and sequencing across populations , including approximately one individual per population in each lane of sequencing in order to ensure that expression differences were due to biological rather than technical variation . We also sequenced technical replicates for each sample by sequencing each library preparation twice per individual . We assessed the correlation between replicates and identified problematic samples as previously described [24] . Briefly , we applied an optimal power space ( OPS ) transformation to expressed gene and transcript quantifications to ensure that all data points contributed equally to correlation measures , eliminating bias by low and high FPKM values . Pearson correlations between technical replicates were high ( r = 0 . 915±0 . 034 ( mean ± sd ) for genes ( Figure S1 ) , r = 0 . 641±0 . 167 for transcripts ) . Higher correlations between replicates for gene versus transcript quantifications likely reflect the greater uncertainty in the deconvolution of the relative abundance of transcripts within a gene . Because reproducibility between replicates was high , we pooled reads across replicates and reassessed gene and transcript quantifications with Cufflinks . For each sample , we determined the median Pearson correlations ( D-statistics ) with all other samples . D-statistics were high overall ( median D-statistic = 0 . 948 for genes , median D-statistic = 0 . 862 for transcripts , Figure S2 ) . We identified two outliers , both within the San population ( HGDP01029 and HGDP00992 ) , and we removed these samples as well as the two remaining San samples from all downstream analyses . To compare gene expression patterns across individuals , we first normalized our data . Exon and gene counts were quantified over regions annotated in UCSC known gene tables . Previous work has shown that the sample preparation protocol for RNA-seq introduces nonlinear , sample-specific effects that explain more than 50% of the variation in expression data [25] , [26] . These nonlinear effects can manifest as sequence-specific biases [13] , which we accounted for via conditional quantile normalization ( CQN ) [27] . This normalization strategy removed large distributional outliers ( Figure S4 ) by accounting for non-linear guanine-cytosine ( GC ) content and feature length effects . As previously observed , genetic variation clearly differentiates globally diverse populations [28] , [29] ( Figure 2A–D ) . A tree generated via hierarchical clustering of FST distances ( Figure 2A–B ) shows a clear separation of sub-Saharan African populations and out-of-Africa populations . Additionally , principal component analysis ( PCA ) of autosomal single nucleotide polymorphisms ( SNPs ) in the HGDP dataset ( Figure 2C–D ) shows population-specific clustering [29] with these seven global populations separating within the first four PCs . Despite clear clustering among the selected populations at the genetic level , PCA of gene expression levels assessed via Cufflinks reflects high individual expression variability and shows no clear population clustering ( Figure 2E–F ) . A formal test of this hypothesis is presented in the last subsection of the Results section , “Variability in expression and alternative splicing ratios , ” which also considers the impact of population labeling as a factor in gene expression differences among individuals . We next sought to identify individual exons and genes that show strong evidence of differential expression ( DE ) among populations . We used a negative binomial model for gene expression analyses ( Methods ) and incorporated a normalization offset term from CQN via edgeR [30] ( Figure S3 ) ; we find that our model provides a good fit to the data ( Methods , Figure S4 ) . We identified 251 DE exons via generalized linear model with a false discovery rate ( FDR ) of less than 5% when comparing all populations ( Table S1 ) . Two examples of genes containing highly DE exons are shown in Figure 3 ( expression of all individuals shown in Figure S5 ) , both of which are involved in immune function and have some previous evidence for population-specific effects [31] , [32] . Figure 3A shows the expression of MX1 colored by population ( FDR = 1 . 57% ) . MX1 is known to affect the immune response to influenza , the West Nile Virus , the avian flu , and other DNA and RNA viruses [33] , [34] . Additionally , LSP1 ( lymphocyte-specific protein 1 , Figure 3B , FDR = 0 . 87% ) has been associated with breast cancer risk in Europeans . Interestingly , this signal did not replicate using admixture mapping in Latina women , perhaps due to differences in allele frequency among the GWAS and attempted replication populations [32] . We also identified 44 differentially expressed transcripts at ≤5% FDR ( Table S2 ) . We used gene set enrichment analysis ( GSEA ) of ranked p-values to detect functional enrichment of differentially expressed transcripts [35] . The following categories were enriched with a FDR≤5%: RXR and RAR heterodimerization with other nuclear receptors ( q = 0 . 007 , canonical pathway ) , IL 2 signaling pathway ( q = 0 . 015 , BioCarta ) , and Top 40 genes from cluster 7 of acute myeloid leukemia ( AML ) expression profile ( q = 0 . 018 , chemical and genetic perturbations ) ( Figure S6 ) . Allele-specific expression ( ASE ) can be detected as a read imbalance at a given heterozygous site; it has previously been shown to tag regulatory variants [12] . To identify the degree to which allelic effects on expression vary , we compared ASE sharing among individuals for variants in the high coverage exomes . We define normalized ASE sharing as the number of shared significant ASE events ( p<0 . 05 ) with at least 30 reads , normalized by sharing of SNPs that are heterozygous with at least 30 reads , regardless of presence or absence of a significant allelic imbalance . Reads were sampled to have equal counts in order to account for expression variability . There is a rapid reduction in normalized ASE sharing as the number of individuals in the comparison set increases ( Figure S7A ) . That is , even when heterozygous sites are shared , most allelic imbalances are private to an individual . Some allelic imbalances are shared by pairs of individuals; rarely do three individuals in the set share an imbalance and very little sharing occurs across more than four individuals . We compared normalized ASE sharing across individual pairs and found similar levels of sharing within and between populations ( Figure S7B ) . A potential explanation for this lack of ASE sharing among individuals is that the allelic state of the underlying causal regulatory variant tagged by the ASE exome site is acting in cis but in weak linkage disequilibrium , potentially with a rare regulatory variant . In a previous study , Stranger et al . ( 2012 ) mapped eQTLs in eight populations from the HapMap3 dataset . To determine if the effects of these previously identified cis-regulatory variants can be captured in our more diverse HGDP populations , we compared ASE events in our dataset to previously discovered eQTLs [3] across populations . We hypothesized that if an individual is heterozygous for a previously discovered cis eQTL SNP ( eSNP ) , and a significant ASE signal exists in the associated gene , then the allelic imbalance is more likely to be driven by the eQTL ( see Figure S8 for a graphical representation of the model ) . We assessed the HGDP genotypes of eSNPs identified in HapMap3 and determined that there is a significant ASE enrichment within eQTLs associated with heterozygous versus homozygous eSNPs ( p<2 . 2×10−16 , Figure S9 ) . This finding is consistent with our model and previous studies [12] and indicates that our measures of ASE are tagging shared regulatory variation between these studies . We also calculated an enrichment score similar to an odds ratio to determine how often ASE events are found in heterozygous versus homozygous eQTLs compared to the number of measured sites ( Methods ) for each HGDP and HapMap3 population . We observe an enrichment of ASE events in heterozygous eQTLs versus homozygous eQTLs consistently in all populations , but we do not observe a signal showing stronger effects in HGDP populations that are more closely related to the eQTL discovery population ( Figure S10 ) . This supports the previous notion that the effects of common regulatory variation are largely shared across populations with taggability depending on patterns of shared LD [3] . We next sought to determine whether regulatory events discovered within populations replicate more consistently in more closely related populations . Because of the limited sample size and structured populations in this study , de novo eQTL discovery is infeasible . We therefore assessed cross-population regulatory sharing using previously discovered eQTLs [3] . We compared Spearman's rank correlation coefficient ρ2 values , a measure akin to variance explained , between our dataset and the HapMap3 study and find consistency between the associations ( r = 0 . 22 , p<2 . 2 * 10−16 ) . The −log10 ( p ) values across studies were also significantly correlated ( r = 0 . 14 , p<2 . 2 * 10−16 ) . We next measured the associations between eQTLs identified in each population . We find that the effect sizes of eQTLs are significantly associated across most pairwise populations ( Figure 4 ) , independent of genetic divergence . The reproducibility of eQTLs is similar across populations , indicating that previously discovered common eQTLs reflect either the true causal SNPs or tag the causal eQTL due to similar LD at the locus ( Figure S11 ) . We also assessed the impact of similarity in allele frequencies between studies on the ρ2 values and find that eQTLs with similar minor allele frequencies ( MAFs ) between studies replicate better than eQTLs with different minor allele frequencies . As expected , eQTLs with high MAFs in one study and low MAFs in another study replicate poorly ( Figure S12 ) . Using the genome , exome , and RNA-seq resource described above , we characterize the completeness of current gene annotations as previously described [13] . By pooling our dataset of 1 . 7 billion paired reads , we identify regions of novel transcription that lie outside of previously characterized gene structures . By calculating per-base global sequencing coverage and merging together continuous transcribed regions above our cutoff filters , we identified 445 , 091 total regions of putative transcription in our LCLs , 384 , 285 ( ∼86% ) of which corresponded to annotated exons in Refseq , Ensembl , UCSC , or Gencode databases ( Methods , Figure S13 ) . Conversely , 34 , 555 regions ( ∼7% ) meeting our minimum expression threshold did not overlap with known annotations ( Figure S14 ) . When we filter regions expressed in at least one individual per population at greater than or equal to 1 RPKM , there are only a few hundred of these 34 , 555 regions expressed across all individuals in that population ( Table S3 ) . Additionally , we see that every novel transcribed region expressed ubiquitously in one population is also present in at least one other individual of another population . This result suggests that the vast majority of novel transcribed regions are not population specific , but can be found across multiple diverse human groups . Previous work indicates that exonic splicing may vary significantly more than gene expression variability across species within the same tissue [36] , [37] . The majority of previous human transcriptome work has focused on expression and regulatory variability , leaving the degree of alternative splicing variation across diverse human populations relatively unexplored . To understand expression and splicing relationships within and between human populations , we measured the coefficient of variation , cv , in gene expression ( standard deviation divided by the mean ) and the variability in alternative splicing ratios ( Hellinger distance to the centroid of the splicing ratios of each gene across all individuals in the population , ) using methods developed previously [4] . We find that the cv and values for genes are highly correlated between pairwise populations ( cv correlations are , on average , within [0 . 44 , 0 . 67] between pairwise populations , p<2 . 2×10−16 for each comparison ( Figure S15 ) , and correlations are on average within [0 . 64 , 0 . 82] between pairwise populations . p<2 . 2×10−16 for each comparison ( Figure S16 ) ) . The relationships overall between cv and values do not reflect the genetic divergences seen between pairwise populations ( Mantel test with 1 , 000 Monte Carlo repetitions between cv Spearman rank correlation distance matrix and FST gives ρ = 0 . 38 , p = 0 . 16 , and the same test between and FST gives ρ = 0 . 44 , p = 0 . 14 ) . We next used established methods to assess the proportion of gene expression variation among individuals attributable to population identity [1] . We find that population label , on average , explains 25 . 0% of the variation in gene expression among individuals ( Figure 5A ) for all genes expressing at least two transcripts . To assess significance for each gene , we used a permutation test reshuffling population labels among individuals and find that the p-value distribution is heavily skewed towards low p-values compared to the expected uniform distribution ( Figure 5B ) . This genome-scale level of population stratification for gene expression is higher than previously seen by the GEUVADIS consortium [1] , which reported ∼3% of the variance attributable to population label as a factor when considering populations of mostly European descent in the 1 , 000 Genomes Project . These results are perhaps expected given that the populations in our study span a greater breadth of human genetic diversity . We repeated this analysis comparing each population to all other populations and find that a smaller proportion of the variation is due to population-specific differences and that these differences do not follow the pattern expected by population divergence ( Figure S17 ) . We also decomposed population-specific variability into variability in overall expression levels as opposed to splicing variability via multiplicative model , which , as previously demonstrated [1] , [4] , accounts for differences in scales and units between expression and splicing metrics . We find that on average , variation in gene expression explains the majority ( 75 . 5%±22 . 3% ( mean ± sd ) , Figure 5C ) of population-specific variation , indicating that alternative splicing generally makes up the minority of population-specific variation within humans . We repeated this analysis comparing each population to all other populations and find consistent results ( Figure S18 ) . We next assessed differential splicing between pairwise populations . In Figure 6 , we show a sashimi plot of a gene ( ENSG00000183291 . 11 , SEP15 ) with substantial differential splicing across all pairwise populations . Overall , we do not see evidence for differential splicing patterns consistent with population genetic divergence ( Figure S19 ) ; this result is consistent with a minority of population-specific variance in gene expression levels explained by splicing variability .
We have analyzed the transcriptome landscape from populations spanning the breadth of human genomic diversity . While other studies have characterized variation within and among populations [12] , [13] , this study provides a unique opportunity to discover regulatory drivers of expression diversity in serially bottlenecked populations throughout human migration history . The HGDP populations in this study were explicitly chosen to encompass a large geographic range that experienced varied demographic histories , and thus they provide unique insight into global variation in transcription . In addition to gaining an understanding of transcriptome variation in diverse populations , this study also enables the discovery of novel gene structures and provides a public resource for analyses of diverse human transcriptomes . In this study , we have assessed population-specific expression variability , alternative splicing , and regulatory variation . We account for technical artifacts in our analyses , including GC content and feature length effects , which otherwise add nonlinear systematic noise to expression data . We show that we substantially reduce technical sources of variation from these effects in our data and obtain high reproducibility between sequencing replicates . We detect few differentially expressed exons , which is likely affected by the fact that we analyze cultured cell lines grown in a highly homogenous environment . Further , given our sample size per population , we are only powered to detect very dramatic differences in expression among populations . Using variance decomposition methods developed previously , we find that 25 . 0% of transcription variability can be attributed to population differences among the six we study here . A previous study that sought to detect expression differences between the CEU and YRI estimated that ∼17% of genes were differentially expressed across these populations [38] . This estimate is quite comparable to ours . However , the estimates from both studies are substantially larger than those reported by the GEUVADIS consortium , which found that population labels accounted for only ∼3% of transcription differences among 462 individuals sampled from the European populations in the 1000 Genomes Project as well as Yorubans . One potential reason why our analysis produced estimates larger than GEUVADIS is that the European populations sampled there are more closely related to each other than the breadth of populations studied here . Immunity genes as a whole are overrepresented in the set of differentially expressed genes across populations . This is highly consistent with the immune role of LCLs we study here . This finding is also consistent with previous work showing that natural selection may have favored different alleles in certain immune genes across human populations and that differences in autoimmune disease risks may be a side consequence of differences in these evolutionary histories [39] , [40] . The increased expression of immune genes in LCLs also improves our power to detect differences with respect to most other gene functions . Potential mechanisms for differential expression across populations include variation in cis and trans eQTL allele frequencies , environmental differences , and epigenetic differences . We also measured the population-specific variance attributable to expression versus splicing and find that on average , 75 . 5% is due to gene expression differences . This result is consistent with previous findings in humans and indicates that , within tissues , splicing differentiates populations less than expression . While this finding is consistent with previous human studies [1] , [4] , it appears to be inconsistent with other cross-species work [36] , [37] . This suggests that splicing potentially plays a greater role on longer evolutionary time-scales . Additionally , the methodology used to assess splicing varies substantially between these studies; in this study , we have used variance decomposition methods relying on gene and transcript annotation data , which is more limited in many other species . In the cross-species studies , exonic splicing was measured via “percent spliced in” ( Ψ ) , which may be affected by expression variability or other forms of transcript differences , such as those arising from alternative start sites . Further work on the efficacy of alternative splicing quantification methodologies would benefit future studies . We also show that eQTLs that were previously identified across a wide range of human populations show allelic imbalances and replicate consistently across populations , but this replication is dependent on minor allele frequencies . Our results suggest that rare eQTLs within a population that are common in another population will likely have differing effect sizes . Given that the ∼1 . 2 million SNPs assayed in HapMap3 are common and therefore largely shared globally , we have only limited power to assess the effects of rare regulatory variants . As more transcriptomes are sequenced across diverse populations , we expect that rarer eQTLs identified in large population-based genome- and RNA-sequencing studies will identify more population-specific enrichment patterns . This study provides the first analyses of transcriptome diversity from serially bottlenecked populations spanning the breadth of human migration history . In this study , we integrated genome , exome , and transcriptome sequencing data from LCLs that are part of the HGDP . This enabled us to assess regulatory drivers of global expression variation in serially bottlenecked populations across a large geographic range and different demographic histories . We find that population of origin accounts for ∼25% of variation in transcription . While we are powered to detect only large differences in expression among populations , genes involved in immunity are overrepresented in this set . Of the 25% difference in transcription explained by population of origin , expression differences accounts for three-fold more of the effect than do splicing differences . Further , the common regulatory variants we replicate here impact expression across broad geographic groups relatively uniformly and do not correlate with the degree of genetic divergence among populations . We look forward to larger studies spanning the breadth of human diversity that are better powered to detect additional population-specific effects and cellular mechanisms of global expression variation . Here , we analyze the total variance in expression and splicing explained by global populations , which , together with other studies , suggests a complex genetic mechanism for population level variation in transcription .
Total RNA was extracted from lymphoblastoid cell lines in four San , seven Mbuti Pygmies , seven Mozabites , six Pathan , seven Cambodians , seven Yakut , and seven Mayans from the Human Genome Diversity Panel using an RNeasy Mini Kit ( Qiagen ) . mRNAs were purified using magnetic oligo-dT beads and randomly fragmented to 300–400 nucleotides in length . First-strand cDNA synthesis was performed using random hexamers and reverse transcriptase . This was followed by second-strand cDNA synthesis with dUTP via the dUTP strand-marking protocol [41] . Illumina TruSeq adaptors were ligated to the ends of the double-stranded cDNA fragments followed by digestion with uracil N-glycosylase ( UNG ) to remove second strand cDNA . A 300–400 bp size-selection of the final product was performed by gel-excision , following the Illumina-recommended protocol . Each individual was sequenced in a 7-plex library on an Illumina HiSeq 2000 producing 101-bp paired end reads . Lanes were assessed for multiple quality metrics including number of reads , read quality , and reads mapping to the human genome . Two San individuals failed sequencing quality control and so all four San individuals were excluded from further analysis . Sample genomic DNA was extracted from lymphoblastoid cell lines . Exonic regions were enriched using an Agilent SureSelect XT 44 Mb All-Exon Capture Kit ( v2 ) and sequenced on Illumina HiSeq machines . Illumina sequencing reads were mapped to the human reference genome ( hg19 ) using a standard pipeline informed by the 1000 Genomes Project [42] . Briefly , reads were mapped and paired using bwa v0 . 5 . 9 [43] . Duplicate read pairs were identified using Picard ( http://picard . sourceforge . net/ ) . Base qualities were empirically recalibrated , indels were realigned , and variants were called using the Genome Analysis Tool Kit ( GATK ) v1 . 6 [44] . SNP calls that failed the Variant Quality Score Recalibration ( VQSR ) step were filtered out . Exonic SNPs were annotated using the RefSeq database to identify synonymous coding variants . High confidence and high coverage synonymous variants were used to compute Weir & Cockerham FST values [45] for each pairwise population using vcftools ( v0 . 1 . 11 ) [46] . Reads were mapped to the human reference genome ( hg19 ) with bowtie-2 . 0 . 0 and tophat-2 . 0 . 4 split read mapping algorithms using the “-b2-very-sensitive” parameters [46] . Reads were subsequently filtered to include only properly paired reads . This yielded between 12 . 1 and 44 . 8 million reads per individual ( 29 . 3 mean±7 . 9 s . d . million reads ) , which corresponds to 62 . 17±13 . 79% of the total reads per individual . Exon and gene count estimates were created by using bedtools to count read overlap with known genes and exons from the UCSC “knownGene” table file downloaded on July 17th , 2012 for differential expression analysis . Raw exon and gene read counts were normalized through conditional quantile normalization , which reduces expression outliers by accounting for feature level GC nucleotide content and overall feature length [27] . UCSC knownGene tables were also used for novel transcript structure analysis because a larger collection of gene structures have been catalogued in this annotation set . For all other analyses , gencode v13 annotations were used , because they give one-to-one correspondence of transcript to gene annotation , enabling the Gonzalez-Porta methods to be used as they were developed . Transcript level quantification was performed with cufflinks-2 . 0 . 2 and produced FPKM ( fragments per kilobase of exon per million ) estimates per transcript . Cufflinks uses a generative statistical model of paired-end sequencing experiments to derive a likelihood for the abundances of a set of transcripts given a set of fragments . The likelihood function can be shown to have a unique maximum , which Cufflinks finds using a numerical optimization algorithm . The program then multiplies these probabilities to compute the overall likelihood that one would observe the fragments in the experiment , given the proposed abundances on the transcripts [23] . In order to compare expression levels in this dataset with those identified in Stranger et al [3] , we reran Cufflinks ( v2 . 1 . 1 ) using the Gencode v13 annotations to get both gene and transcript quantifications . These expression abundances were subsequently used to quantify the relative importance of variability in gene expression and variability in alternative splicing to individual transcript variability . Sequencing variants called from the differentially expressed and differentially spliced regions were annotated for a series of functional predictions , conversation scores , and RefSeq database annotations as described below . This was done in order to better assess the significance of genetic variants present in the data and their potential contribution or involvement in modulating gene expression , transcript splicing , and phenotypic variability . General annotations include information from: the NHLBI Exome Sequence Project allele frequencies; 1000 Genomes Project allele frequencies; publically available Complete Genomics sample allele frequencies; region and exonic annotations from both Ensembl and RefGene; and information about protein structure and function from the UNIPROT and INTERPRO databases . Conservation scores were also produced from the following algorithms: GERP++ , SLR , SIFT , LRT , PHYLOP , and SiPhy based on 29 mammalian genomes [47]–[51] . Lastly , functional prediction annotations were produced from the following sources: FATHMM , MutationTaster , Mutation Assessor , LRT , PolyPhen2 , and the RefSeq RefGene database [50] , [52] . Methods to characterize regions of previous unannotated transcription closely followed previously described work [13] ( Figure S14 ) . In brief , for each base of the genome we calculated global sequencing coverage and split the genome into continuous transcribed regions . Expression of a region was defined as the maximum per base coverage of bases in the region . As in previous studies , we chose a threshold of an average expression level of 5×10∧-8 ( or 0 . 05 reads/million ) to consider a region expressed and merged together regions separated by less than 15 bp [13] . Sample specific expression of these novel regions was then quantified by calculating RPKM of each region for each individual . For these analyses , we ran Cufflinks ( v2 . 0 . 2 ) using the UCSC KnownGene tables downloaded on July 16 , 2012 because there were fuller annotations than in Gencode v13 . ASE was determined as previously [12] . Briefly , variants were called for all HGDP individuals in this project using high coverage , high quality exome variant calls generated according to the GATK best practices . Samtools was used to determine the number of reads that matched the reference and non-reference allele . Imbalance reference allele mapping bias was compensated using the per individual overall reference ratio within the binomial test . We used conditional quantile normalization for all exons and genes with unique start and stop positions , accounting for GC content and length as covariates , and generated an offset term per gene or exon and individual . We filtered to exons or genes where the standard FPKM expression was > = 2 and the length was at least 100 bp , which left 207 , 180 of all UCSC knownGene annotated exons ( 29 . 7% ) and 72 , 931 of all annotated genes ( 26 . 8% ) . Then , we used the following negative binomial model to detect differential expression:Here , y is the count at gene g in individual i , β is the vector of population effects , x is the population label , o is the offset term from conditional quantile normalization , and ε is the error term . We perform an analysis of variance ( ANOVA ) comparing the null hypothesis of β = 0 to the alternative hypothesis of β≠0 . In pairwise population comparisons , we computed genewise exact tests for differences in the means between the two groups of negative-binomially distributed counts . eQTLs discovered in the HapMap3 populations were replicated in our HGDP dataset using genotypes derived from the exome sequencing variants and preliminary results for the full genomic variants ( Henn & Botigue et al , unpublished data ) for eQTLs outside the exome ( Data Access ) . The SRA accession number for the genome and exome sequence data reported in this paper is SRP036155 . The GEO accession number containing the RNA-Seq data and gene/transcript expression matrices reported in this paper is GSE54308 . Links to additional data ( exome variant files , eQTL SNP data , FST matrices , gene/transcript expression quantifications , ASE tables , and eQTL data ) and scripts are provided on an FTP site by the Stanford Center for Genomics and Personalized Medicine computing cluster located here: http://bustamantelab . stanford . edu/datasets . html .
|
Previous gene expression studies have identified factors influencing population-level variation in gene regulation . However , these efforts have been limited to a small set of well-studied populations . By leveraging the high resolution of RNA sequencing and broad population sampling , we survey the landscape of transcriptome variation across a globally distributed set of seven populations that span a breadth of human genetic variation and major dispersal events . We assess differences in gene expression , transcript structure , and regulatory variation . We find only 44 transcripts that show significant differences in expression , likely as a result of the small sample size , but we find that 25% of the variance in gene expression is due to population differences . This is a larger fraction than previously observed , and it is likely due to the greater breadth of human diversity assayed in this study . We also find that population-specific variance is mostly due to transcription variability rather than the configuration of expressed gene products . Additionally , known common regulatory variants have similar effects across populations including those we study here . These data and results serve as a resource cataloging the wide array of gene expression regulation affecting population variation among diverse groups , improving our understanding of transcriptional diversity .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biotechnology",
"genomics",
"genome",
"analysis",
"gene",
"expression",
"transcriptome",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"genetic",
"engineering"
] |
2014
|
Transcriptome Sequencing from Diverse Human Populations Reveals Differentiated Regulatory Architecture
|
The role of CpG island methylation in normal development and cell differentiation is of keen interest , but remains poorly understood . We performed comprehensive DNA methylation profiling of promoter regions in normal peripheral blood by methylated CpG island amplification in combination with microarrays . This technique allowed us to simultaneously determine the methylation status of 6 , 177 genes , 92% of which include dense CpG islands . Among these 5 , 549 autosomal genes with dense CpG island promoters , we have identified 4 . 0% genes that are nearly completely methylated in normal blood , providing another exception to the general rule that CpG island methylation in normal tissue is limited to X inactivation and imprinted genes . We examined seven genes in detail , including ANKRD30A , FLJ40201 , INSL6 , SOHLH2 , FTMT , C12orf12 , and DPPA5 . Dense promoter CpG island methylation and gene silencing were found in normal tissues studied except testis and sperm . In both tissues , bisulfite cloning and sequencing identified cells carrying unmethylated alleles . Interestingly , hypomethylation of several genes was associated with gene activation in cancer . Furthermore , reactivation of silenced genes could be induced after treatment with a DNA demethylating agent or in a cell line lacking DNMT1 and/or DNMT3b . Sequence analysis identified five motifs significantly enriched in this class of genes , suggesting that cis-regulatory elements may facilitate preferential methylation at these promoter CpG islands . We have identified a group of non-X–linked bona fide promoter CpG islands that are densely methylated in normal somatic tissues , escape methylation in germline cells , and for which DNA methylation is a primary mechanism of tissue-specific gene silencing .
CpG islands ( CGIs ) are discrete CpG-rich regions present in the promoters of 50%–70% of human genes [1] . DNA methylation within CGIs is associated with mitotically stable gene silencing ( an epigenetic process ) . Such CGI methylation is involved physiologically in genomic imprinting [2] and X inactivation [3] and pathologically in developmental diseases [4] and cancer [5] . The role of CGI methylation in normal development , stem cell physiology , and differentiation , however , remains poorly understood [6] . Inhibition of DNA methylation can transform fibroblasts into muscle cells and other differentiated cells , suggesting that gene methylation regulates the process of differentiation [7] . However , support for this idea was dampened when CGIs were generally found to be unmethylated regardless of tissue-specific expression [8] , and tissue-specific genes thought to be regulated by methylation were unaffected by demethylation in vivo [9] . It is now widely held that CGIs associated with both housekeeping and tissue-specific genes are unmethylated at any developmental stage , except when associated with certain imprinted genes and genes subject to X inactivation . While most CGIs are unmethylated in normal tissues , there is increasing evidence that a few CGIs are in fact methylated in normal tissues independent of imprinting and X inactivation , and may play a role in the differentiation process through programmed expression of tissue-specific genes . At several genes , CpG island methylation has been correlated with transcriptional inactivation in normal cells . Most such correlations , however , involve intragenic CGIs rather than promoter CGIs [10 , 11] . A few promoter CGIs were found at which methylation correlates with transcription [12 , 13] , but those contain a low density of CpG sites . For example , a recent study profiling genome-wide DNA methylation in three human chromosomes ( Chr 6 , 20 , and 22 ) demonstrated that a small subset of promoter-region CGIs are methylated in various normal tissues , but all have a CpG density less than 10% [14] . That study also identified a considerable number of tissue-differentially methylated regions in CGIs , but these were preferentially located several kilobases away from the transcription start site of associated genes . Thus , the dogma that promoter-associated dense CGIs are unmethylated in normal tissues persists . A full appreciation of the role of CGI methylation in normal development awaits a careful high-throughput analysis of the process . In this study , we compared differential methylation in normal tissues by genome-wide CGI methylation profiling and discovered non-X–linked bona fide promoter CGIs that are densely methylated in normal somatic tissues and escape methylation in germline cells . CGI hypermethylation at these gene promoters appears to be a primary mechanism of tissue-specific gene silencing .
We performed a comprehensive DNA methylation profiling of gene promoter regions in normal peripheral blood by methylated CpG island amplification ( MCA ) in combination with microarrays ( MCAM ) . This procedure is described in Materials and Methods and outlined in Figure S1 . Essentially , MCAM involves two steps . First , MCA [15] is used to enrich for methylated fragments . Second , labeling and cohybridization of MCA products to arrays enables comparison of locus-specific methylation between samples . We used promoter microarrays that contained 45- to 60-mer oligonucleotide probes covering from −1 . 0 kb to +0 . 3 kb relative to the transcription start sites of 18 , 000 human genes . Bioinformatic analysis predicted that 22 , 294 probes corresponding to 6 , 177 unique genes on the array would be informative when using SmaI/XmaI enzymes to generate methylated fragments up to 1 kb in size ( see Materials and Methods ) . We also annotated all the SmaI/XmaI sites for CGIs and repetitive sequences . Among these informative genes on the arrays , gene promoters associated with dense-CGIs , sparse-CGIs and non-CGIs were 5 , 692 ( 92% ) , 318 ( 5% ) , and 167 ( 3% ) respectively . In an initial validation of MCAM , we compared in vitro fully methylated genomic DNA with DNA isolated from normal peripheral blood leukocytes ( PBLs ) . As expected , the signal intensity of Cy5 ( fully methylated ) was high at most ( 87 . 1 % ) informative probes ( Figure S2A and S2B ) . The 12 . 9% of probes that did not show high signal in the positive control could be attributed to a PCR bias caused by the increased complexity of performing MCA on an artificially hypermethylated genome . Alternatively , nonsignificant signal could occur at probes with poor discriminative ability . Therefore , we estimated that MCAM technique has a false negative rate of less than 12 . 9% . We compared the signal intensity of fluorescent probes between two independent hybridizations using MCA products processed at different times from the same normal PBL DNA sample , and found that the correlation between the two duplicates was 0 . 94 , indicating that the technique is highly reproducible ( Figure S2C ) . In the cohybidization of MCA product from fully methylated DNA and normal PBL DNA , we observed probes showing high signal intensity in both channels ( Figure S2B ) , suggesting genes hypermethylated in normal PBLs . Surprisingly , a subset of genes showed significantly higher signal in PBLs than in fully methylated DNA . This is possible because the relatively low number of hypermethylated regions in PBLs will amplify with higher efficiency compared with fully methylated DNA in the MCA reaction . We randomly selected 38 genes showing such high signal intensity in PBLs and analyzed them by a quantitative bisulfite pyrosequencing method . Of these , 17 showed dense ( >70% ) methylation in normal PBLs , six showed moderate ( 15% to 70% ) methylation , and 15 showed low ( <15% ) methylation ( Figure S3 ) . Genes that were densely methylated in PBLs showed the highest signal intensity ratio ( PBLs versus fully methylated DNA ) , with median ratio of 2 . 2; the median ratio in genes with moderate and low methylation was 1 . 5 and 0 . 7 , respectively . Hence , a relatively high signal intensity ( >3-fold of background ) combined with a signal ratio ≥ 1 . 5 relative to in vitro-methylated DNA appears to identify hypermethylated loci in PBLs with 93% specificity and 74% sensitivity . Using these criteria , we identified 455 genes methylated in normal PBLs ( Table S1 ) . 258 of these gene promoters were associated with dense-CGIs , 129 were associated with sparse-CGIs , and 68 were associated with non-CGIs . Thus , we estimate that 4 . 5% ( 258/5 , 692 ) of promoter-associated dense-CGIs are methylated in normal PBLs , while 40 . 5% ( 129/318 ) and 40 . 7% ( 68/167 ) of sparse-CGI and non-CGI promoters show such methylation . Methylated promoter CGIs were distributed throughout the genome ( Figure S4 ) . Interestingly , most of the identified CGIs were autosomal; in these , the frequency of methylation was 4 . 0% ( 223/5 , 549 ) for dense-CGI promoters , 39 . 4% ( 121/307 ) for sparse-CGI promoters and 40 . 9% ( 65/159 ) for non-CGI promoters . Except for MEST , none of these was associated with known imprinted genes . Together , these data modify the notion that CGI methylation is limited to X chromosome and imprinted genes in normal tissues . Our results also indicate that both non-CGI and sparse-CGI promoters are frequently methylated in normal somatic tissues . Since most CGIs previously known to be methylated in normal tissues are located in intragenic regions or promoters of intermediate CpG density , we were surprised to find this exceptional class of autosomal gene promoters associated with dense-CGIs ( 4 . 0% of all such CGIs analyzed ) that are methylated in normal PBLs . We used bisulfite pyrosequencing to measure DNA methylation quantitatively at such promoter CGIs for 17 genes and confirmed the methylation level of each gene ranged from 68% to 93% in normal PBLs ( Table 1 ) . We also measured the methylation of all these genes in two cancer cell lines ( the leukemia cell line K562 and the colon cancer cell line RKO ) and normal testis . Relative to PBLs , we observed promoter hypomethylation in the cancer cells and testis . This was most striking in K562 , where hypomethylation was found in all 17 genes . As shown in Table 2 , analysis of chromosomal location , CpG density , and GC content of these genes revealed that all have typical CGIs in their promoters , with high CpG densities ranging from 13 . 2% to 23 . 2% , and all are located on autosomes with no apparent association with common repetitive sequences or pseudogenes . To determine if methylation affects only a few CpG sites or all CpG sites across the island , we carried out bisulfite cloning and sequencing for seven of these genes , ANKRD30A , FLJ40201 , INSL6 , SOHLH2 , FTMT , DPPA5 , and C12orf12 . All genes were selected on the basis of methylation levels greater than 80% in PBLs by bisulfite pyrosequencing for limited CpG sites . Bisulfite cloning and sequencing provided allele specific methylation data on a larger number of CpG sites , and again showed extensive methylation at all CpG sites within CGIs in normal PBLs ( Figure 1 , left ) , but little or no methylation in K562 ( Figure 1 , right ) . To assess the extent of tissue and cell-type specific DNA methylation at these CGI promoters , we used quantitative bisulfite pyrosequencing to analyze 33 normal samples derived from ten human tissues: blood , colon , liver , breast , brain , fibroblast , prostate , skeletal muscle , testis and sperm . Dense methylation at these CpG island promoters was found in all tissues except sperm and testis ( Figure 2A ) . Bisulfite cloning and sequencing of sperm DNA identified cells carrying completely unmethylated alleles ( Figure 2B ) . Among the sequences obtained from testis , some alleles were almost completely unmethylated , whereas others were heavily methylated ( Figure 2C ) . As adult testicular tissue contains a mixture of germ line and somatic cells , these results suggest that these unmethylated alleles are derived from germ line cells . We hypothesized that the methylation patterns we observed could predict tissue-specific silencing . To test this , we examined the expression of all genes except two intronless genes ( FTMT and C12orf12 ) in a cDNA panel from 20 normal tissues . Consistent with the DNA methylation status , all five genes analyzed were strongly expressed in testis ( Figure 3 ) . Except SOHLH2 , expression of four genes ( strong expression of three and weak expression of one ) was also detected in sperm . However , expression of SOHLH2 has been reported in oocytes [16] . In contrast , most normal somatic tissues showed no or weak expression of these genes with the exceptions of INSL6 and SOHLH2; INSL6 was expressed in kidney , placenta , prostate , and salivary gland , and SOHLH2 was expressed in placenta and prostate . Due to limited tissue availability , we were unable to examine methylation and expression in all tissues; however , we analyzed methylation of INSL6 and SOHLH2 in placenta and found promoter hypomethylation for both genes ( 17 . 6% for INSL6 and 20 . 9% for SOHLH2 ) . We conclude that these genes belong to a unique class of promoter CGI associated genes that are methylated and silenced in a tissue-specific manner . To explore potential shared functionality of this class of methylated genes with dense-CGI promoters , we used GOstat gene ontology analysis [17 , 18] and Benjamini-Hochberg multiple testing correction to identify gene ontology categories that are significantly over-represented . We found most of these genes to be involved in intracellular membrane bound organelle functions ( 34 . 1% , p = 0 . 05 ) , followed by metal ion binding ( 29 . 4% , p = 0 . 0006 ) and signalosome functions ( 1 . 6% , p = 0 . 04 ) ( Figure 4A ) . Using published microarray expression databases and applying Z-scores to assign equal weight to each gene , we compared the expression levels of 127 genes among 66 different normal tissues and/or cell-types ( see Materials and Methods for details ) . As shown in Figure 4B , expression level analyzed by Z-score for all genes was highest in testis , and high in testis-derived cells . Consistent with our methylation data , expression level was greatly decreased in whole blood , as well as various subtypes of blood cells . Indeed , 69% of genes analyzed showed negative Z-scores in whole blood relative to other tissue types , which is highly significant ( p < 2 × 10−7 , assuming a binomial distribution ) . These results again suggest that methylation at these promoter CGIs is associated with tissue-specific gene silencing . To identify sequence characteristics that could differentiate this class of methylated dense-CGI promoters from the bulk of unmethylated dense-CGI promoters , we first compared CGI size , GC content , and the ratio of observed to expected CpG frequency for these two group of genes identified by MCAM ( see Materials and Methods for details ) . There was no significant difference in CGI length or the ratio of Obs/Exp CpG between the two groups ( the average CGI length was 1 , 157 bp in methylated CGIs versus 1 , 248 bp in unmethylated CGIs , and the ratio of Obs/Exp CpG was 0 . 88 in methylated CGIs versus 0 . 88 in unmethylated CGIs ) . The methylated CGIs had a slightly higher GC content compared to unmethylated CGIs ( 66 . 5 versus 65 . 6 , respectively , p = 0 . 02 ) . Next we used a weight matrix finding algorithm ( MEME ) [19] and motif alignment and search tool ( MAST ) [20] to identify sequence motifs that predict methylation patterns . We generated two sets of sequences , one containing 138 sequences ( 2 kb window ) flanking the center of CGIs at methylated genes , and the other containing 2 , 125 sequences flanking the center of CGIs at unmethylated genes . MEME was used to identify the top 20 significant motifs in each set of sequences ( methylated or unmethylated group ) , and then MAST was used to identify motifs that occur differentially between the methylated and unmethylated groups . Of the top 20 motifs enriched in the methylated group , five showed a significantly higher occurrence in the methylated relative to the unmethylated group ( p < 0 . 02 by Fisher exact tests ) ( Figure 5 ) . In contrast , the top 20 motifs identified in the unmethylated group were present at the same frequency in both groups . Using the TRANSFAC database search , none of these five discriminating motifs was associated with any known transcription factor binding site . Interestingly , however , all were frequently located within sequences homologous to Alu sequences ( Figure 5 ) . Global genomic hypomethylation and aberrant promoter hypermethylation are epigenetic hallmarks of tumorigenesis [21–23] . We therefore wished to determine if the genes we identified have aberrant promoter methylation in tumors . Methylation analysis of a panel of 61 cancer cell lines from 13 major tumor types including leukemia , melanoma , teratocarcinoma , bladder , breast , brain , ovarian , colon , liver , lung , prostate , kidney , and skin revealed a considerable number of tumors with promoter hypomethylation at the seven promoter CGIs we analyzed . As shown in Figure 6A , the frequency of promoter hypomethylation ( defined as methylation level less than 70% ) was 9 . 8% ( 6/61 ) for ANKRD30A , 9 . 8% ( 6/61 ) for FLJ40201 , 8 . 2% ( 5/61 ) for INSL6 , 41 . 0% ( 25/61 ) for SOHLH2 , 44 . 2% ( 27/61 ) for FTMT , 49 . 2% ( 30/61 ) for C12orf12 , and 6 . 6% ( 4/61 ) for DPPA5 . Next , we examined gene expression in 12 of these cancer cell lines , and observed aberrant expression in association with promoter hypomethylation in several . For example , hypomethylation of ANKRD30A corresponded with expression in K562 , and relative hypomethylation of SOHLH2 corresponded with its expression in K562 , LNCAP , UC13 , and UC3 ( Figure 6A and 6B ) . We also observed a few cases in which hypomethylation does not correlate with gene activation , such as FLJ40201 in K562 . This could occur if , in addition to hypomethylation , gene activation requires specific transcription factors . Among 12 cancer cell lines for which both methylation and expression of these five genes were assessed ( Figures 6B and 7A ) , in only one case ( INSL6 in LNCAP cells ) did we observe both aberrant expression and promoter hypermethylation . This expression could possibly originate from a rare subpopulation of hypomethylated cells . Methylation profiles were also assessed in tumor tissue samples from ten primary colorectal cancer patients and ten myelodysplastic syndrome patients ( Figure 6C ) . As in the cancer cell lines , we observed promoter hypomethylation in the primary tumors; the frequency of hypomethylation was 5% ( 1/20 ) for ANKRD30A , 20% ( 4/20 ) for FLJ40201 , 0% ( 0/20 ) for INSL6 , 30% ( 6/20 ) for SOHLH2 , 20% ( 4/20 ) for FTMT , 25% ( 5/20 ) for C12orf12 , and 30% ( 6/20 ) for DPPA5 . To investigate the role of DNA methylation in transcriptional regulation of these genes , we examined the effects of treatment with DNA-methyltransferase inhibitor 5-aza-2′-deoxycytidine ( DAC ) or histone deacetylase inhibitor trichostatin A ( TSA ) on gene expression in six cancer cell lines and two primary cultures of normal cells ( Figure 7A ) . In most cancer cell lines , reactivation of the silenced genes was observed in response to the treatment with DAC or the combination of DAC and TSA; in contrast , TSA alone had no or little effect . We also observed gene reactivation in the primary cultures of normal cells after DAC and TSA combined . Relatively weak gene reactivation was observed in these normal cells after 1 or 5 μM DAC alone for 3 d; perhaps the result of slower cell division in normal cells , since DAC causes time- and cell division–dependent demethylation by trapping DNMT1 [24] . Bisulfite pyrosequencing demonstrated that low-dose DAC ( 1 μM ) alone and the combination of DAC with TSA significantly reduced methylation at all gene promoters ( Figures 7B and S5 ) , whereas TSA alone did not affect methylation . Interestingly , we observed less hypomethylation in cells treated with high-dose DAC ( 5 μM ) , consistent with the notion that low-dose DAC specifically inhibits DNA methylation , whereas high-dose DAC results in cytotoxicity . We next analyzed methylation of each gene in a colorectal cancer cell line ( HCT116 ) after partial knockout of DNMT1 , knockout of DNMT3b or knockout of both enzymes ( DKO ) [25 , 26] . All genes were heavily methylated in parental cells , and showed dramatic hypomethylation in DKO cells ( Figure 8A ) . For six of the genes ( ANKRD30A , FLJ40201 , INSL6 , FTMT , C12orf12 , and DPPA5 ) , slightly reduced methylation was observed in DNMT1 partial knockout cells with almost no changes in DNMT3b knockout cells . By contrast , the methylation level of SOHLH2 was significantly decreased in DNMT1 knockout cells ( from 88% to 26% ) , and slightly less but still significantly reduced in DNMT3b knockout cells ( to 40% ) . Consistent with methylation results , expression of all genes analyzed was observed in DKO cells , and expression of SOHLH2 gene was also found in DNMT1 knockout cells ( Figure 8B ) .
During development , a small but significant number of CGIs become methylated and stably silenced [27] . This process of developmentally programmed CGI methylation has been best characterized in genomic imprinting and X chromosome inactivation . Here , using a restriction enzyme–based MCAM approach , we found that non-CGI and sparse-CGI promoters were more susceptible to methylation than dense-CGI promoters , in agreement with a very recent report using an antibody approach to compare methylation profiles between three classes of promoters [28] . Although most dense-CGI promoters remain free of methylation , we found a small exceptional class of such promoters ( 4 . 0% ) that become methylated in normal somatic tissues and are not associated with X-chromosome or imprinted genes . By detailed characterization of a subset of such genes , we found dense promoter CGI methylation and gene silencing in most normal somatic tissues except germ-line cells . Using RT-PCR and data from published microarray experiments , we confirmed tissue-specific silencing for this class of genes . Further , we showed that inhibition of methylation reactivates expression in these genes . Our results suggest that DNA methylation plays an important role in silencing germ-cell specific genes in somatic tissues . A previous analysis of methylation using RLGS ( restriction landmark genome scanning ) identified 150 regions ( including promoters , exons , and introns ) as TDMs ( tissue-specific differentially methylated regions ) in C57BL/6J mice [29] . By comparing 14 of these mouse TDMs with the human genome , six showed human homologs , and five had conserved tissue-specific methylation and expression , being preferentially expressed in testis [30] . Our results indicate that this pattern affects a relatively large number of genes . On the other hand , other studies failed to identify many dense promoter CpG islands hypermethylated in normal tissues , suggesting that the class of genes we describe here is unique . It will be important to determine how methylation at these promoter-associated CGIs is established and maintained . One possibility is that methylation is instructed by local sequence features . By comparing DNA sequence flanking the center of CGIs , we identified five sequence motifs significantly enriched in methylated promoter CGIs relative to the bulk of unmethylated CGIs . Although these motifs do not match known transcription factor binding sites , all of them have significant overlap with Alu , a family of SINEs ( short interspersed elements ) . Alu repeats are rich in CpG dinucleotides and are common targets for DNA methylation . About one-third of methylated CpGs in the genome are located within Alu repeats . Alu repeats have been proposed as methylation centers for neighboring genes [31] and we hypothesize that Alu-related cis-regulatory elements may play a role in establishment and/or maintenance of tissue-specific methylation . Experimental approaches such as transfection and transgenic studies will be needed to test this model . Interestingly , while Alu repeats are almost completely methylated in most tissues , some , particularly young Alu repeats , are almost completely unmethylated in germ line cells [32 , 33] , similar to the genes described here . It remains unclear why tissue-specific gene silencing by DNA methylation is relatively rare , since many genes showing tissue-specific expression are not methylated . What is most exceptional about the genes we describe here is their restricted expression in germ cells . Considering that testis is an immune-privileged site , it is possible that some of these genes , if expressed in somatic tissues , could trigger autoimmune phenomena , justifying the need for a strong mechanism to maintain silencing . In this respect , it is also interesting that we observed hypomethylation of these genes in several cancers . Although the causes and biological effects of cancer-linked hypomethylation remain unclear , such hypomethylation can lead to gene expression that induces an immune response [34 , 35] . The patterns of methylation and gene expression we observed for this class of genes suggest that some may well be cancer-testis antigens . In summary , we have identified a group of non-X–linked bona fide promoter CpG islands that are densely methylated in normal somatic tissues , escape methylation in germ line cells , and for which DNA methylation is a primary mechanism of tissue-specific gene silencing .
Normal tissue samples were obtained from one of the following sources: normal peripheral blood samples from eight healthy donors ( three females and five males ) ; 12 normal colon mucosa ( five females and seven males ) , and three normal liver samples ( all males ) from the MDACC tissue bank; normal brain , breast , placenta , and testis tissues were purchased from BioChain Institute ( Hayward , CA ) ; the sperm sample was obtained from a healthy donor and human primary cells were obtained from Cambrex BioScience ( East Rutherford , NJ ) and American Type Culture Collection ( ATCC , Manassas , VA ) . Tumor samples examined in the present study constitute over 60 cell lines that cover 13 major tumor types ( bladder , breast , brain , colon , liver , lung , ovary , prostate , kidney , skin , teratocarcinoma , leukemia , and melanoma ) from ATCC . Genomic DNA was extracted using a standard phenol–chloroform method . DNA from the colon cancer cell line HCT116 with DNMT1 knockout , DNMT3b knockout and double knockout ( DKO ) were kindly provided by Dr . Bert Vogelstein at the Johns Hopkins Kimmel Cancer Center [25 , 26] . Fully methylated DNA was prepared by treating genomic DNA with M . SssI methylase ( New England Biolabs , Beverly , MA ) . Briefly , 5 μg DNA was incubated at 37 °C in 300 μl containing 20 U of SssI methylase , 320 μM S-adenosylmethionine ( SAM , New England Biolabs ) , and 1× NEB buffer 2 ( New England Biolabs ) . During the incubation , same amounts of SssI methylase and SAM were added one more time to ensure the complete reaction . To verify complete methylation , we performed bisulfite pyrosequencing analysis of seven randomly selected genes that were completely unmethylated before treatment , and found dense methylation at all CpG sites analyzed ( 41 CpG sties in total ) after treatment ( Table S2 ) . Primary colon cancer samples from ten colorectal cancer patients and bone marrow samples from ten MDS patients were collected at the Johns Hopkins Hospital and M . D . Anderson Cancer Center in accordance with institutional policies . All patients provided written informed consent . Tumors were selected solely on the basis of availability . Methylated CpG island amplification from fully methylated DNA and normal peripheral blood was performed as described [15] . A detailed protocol can be found in the document titled Methylated CpG Island Amplification ( in the “Protocols” section; see at http://rd . plos . org/10 . 1371_journal . pgen . 0030181_01 ) . Briefly , 5 μg of genomic DNA was digested with 100 U of methylation-sensitive restriction endonuclease SmaI ( New England Biolabs ) for 16 h at 20 °C , which cuts unmethylated DNA and leaves blunt ends ( CCC/GGG ) . Subsequently , the DNA was digested with 20 U of methylation-insensitive restriction endonuclease XmaI for 9 h at 37 °C , which leaves sticky ends ( C/CCGGG ) . Adaptors were ligated to the sticky ends . The adaptors were prepared by incubation of the oligonucleotide RMCA12 ( 5′-CCGGGCAGAAAG-3′ ) and RMCA24 ( 5′-CCACCGCCATCCGAGCCTTTCTGC-3′ ) at 65 °C for 2 min , followed by cooling to room temperature for 30–60 min . 500 ng of digested DNA was ligated to 5 nmol of adaptor using T4 DNA ligase ( Invitrogen , Carlsbad , CA ) . The PCR amplification of sequences flanked by adaptors was performed in a 100 μl reaction mixture comprising 67 mM Tris-HCl ( pH 8 . 8 ) , 4 mM MgCl2 , 16 mM NH4 ( SO4 ) 2 , 10 mM β-mercaptoethanol , 0 . 1 mg/ml bovine serum albumin , 5% DMSO , 300 μM dNTP mix , 100 pmol of RMCA24 primer , and 15 units of Taq polymerase ( New England Biolabs ) . The thermocycling conditions were 5 min at 72 °C to fill in the overhanging ends of the ligated DNA fragments , and at 95 °C for 3 min; this was then followed by 25 cycles of 1 min at 95 °C and 3 min at 77 °C , with a final extension of 10 min at 72 °C . Human promoter arrays were purchased from Agilent Technologies ( Agilent , Santa Clara , CA ) . Microarray protocols , including labeling , hybridization and post-hybridization washing procedures , can be found at http://www . agilent . com/ . Briefly , MCA products were labeled with Cy5 ( red ) for fully methylated DNA or Cy3 ( green ) for PBLs using a random primed Klenow polymerase reaction ( Invitrogen's BioPrime Array CGH Genomic Labeling Kit ) at 37 °C for 3 h . Labeled samples were then hybridized to arrays in the presence of human Cot-1 DNA for 40 h at 65 °C . After washing , arrays were scanned on an Agilent scanner and analyzed using Agilent Feature Extraction software at M . D . Anderson Microarray Core Facility . We built a database to simulate the performance of MCAM in detecting hypermethylated CpG islands using the SmaI/XmaI isoschizomers . Human genome sequences were downloaded from the UCSC Genome Database ( http://genome . ucsc . edu/; version hg17 , May 2004 ) . The SmaI/XmaI site “CCCGGG” was searched along each chromosome in a case insensitive fashion . Fragments between two SmaI/XmaI sites were extracted . If the fragment length was between 20 b and 10 kb , the fragment was saved in FASTA format with the first line indicating chromosome number , the starting point of the fragment along chromosome ( counting from CCCGGG ) , and the length of the fragment ( including starting and ending CCCGGG ) . CGIs were classified into three classes: dense-CGIs contain a 500 bp area with GC content above 55% and CpG ratio above 0 . 65; non-CGIs do not contain a 200 bp area with GC content above 50% and CpG ratio above 0 . 60; and sparse-CGIs are neither dense-CGI nor non-CGI , thus contain CGIs that are either small or have moderate CpG richness . GC content was calculated based on the number of C and G nucleotides within the sequences analyzed . We used the formula cited in Gardiner-Garden et al . [36] to calculate the CpG ratio ( Obs/Exp CpG ) : ( Number of CpG × total number of nucleotides in the sequences analyzed ) ÷ ( number of C × number of G ) . Repetitive regions were masked in the genome downloaded from UCSC genome database using RepeatMasker/RepBase ( versions: RepBase Update 9 . 11 , RM database version 20050112 ) . The databases for SmaI/XmaI fragments were in FASTA format with annotations for ( 1 ) chromosome , ( 2 ) start point of the fragment along the chromosome , ( 3 ) length of the fragment , ( 4 ) status of CGI in the starting site , ( 5 ) status of CGI in the ending site , ( 6 ) if the starting site was within repetitive region , and ( 7 ) if the ending site was within repetitive region . Probe sequences were downloaded from the Agilent website at http://www . agilent . com/ . Each probe was BLASTed against all sequences in the SmaI/XmaI database using BLAST v2 . 2 . 8 downloaded from NCBI ( http://www . genebee . msu . su/blast_new/blastform . php ? program=blastn ) . Probes with multiple BLAT hits were excluded from further study . Probes residing in SmaI/XmaI fragments were identified with the annotation for fragment length , status of CGI , and repetitive sequences . We used probes located outside of SmaI/XmaI fragments ( length up to 10 kb ) for normalization and background calculation . The signal intensity for the probes within the SmaI/XmaI fragments was adjusted for background and analyzed for the ratio between Cy3 and Cy5 signals . All data analysis ( sensitivity and reproducibility , correlation between methylation level and chromosome location , CGI , and repetitive sequences ) were carried out in Excel ( Microsoft ) . The resulting data sets are accessible in Table S1 . We used the following criteria to select hypermethylated probes in PBL ( Cy3 ) relative to fully methylated DNA ( Cy5 ) : signal intensity of Cy3 > 3 ×background and ratio of Cy3/Cy5 ≥ 1 . 5 × background . We performed bisulfite pyrosequencing on 38 randomly selected genes showing higher signal intensity in PBLs , and determined that these criteria most accurately identified hypermethylated loci . Six cancer cell lines from five tumor types: 786–0 ( renal ) , K562 ( leukemia ) , HCT116 ( colon ) , UC13 ( bladder ) , MCF7 , and HTB126 ( breast ) were purchased from ATCC . 786–0 , K562 , and MCF7 were grown in RPMI 1640 containing 10% fetal bovine serum . HCT116 and HTB126 were grown in high-glucose Dulbecco's modified Eagle's medium containing 10% fetal bovine serum . UC13 was grown in MEM Earle's Salts plus NEAA and 10% fetal bovine serum . Media were purchased from Invitrogen . Two human primary cells , PrEC ( prostate ) and HMEC ( breast ) , were obtained from Cambrex BioScience and cultured in the media according to the supplier's instructions up to a maximum of five passages . PrECs were grown in prostate epithelial cell growth medium ( Clonetics PrEGM bullet kit ) containing 0 . 4% bovine pituitary extract , 5 μg/ml hydrocortisone , 0 . 5 ng/ml recombinant human epithelial growth factor , 0 . 5 μg/ml epinephrine , 10 μg/ml transferrin , 5 μg/ml insulin , 0 . 1 ng/ml retinoic acid , and 6 . 5 ng/ml triiodothyronine . HMEC cells were grown in mammary epithelial cell basal medium ( Clonetics MEMG bullet kit ) containing 0 . 4% bovine pituitary extract , 5 μg/ml hydrocortisone , 0 . 5 ng/ml recombinant human epithelial growth factor , and 5 μg/ml insulin . Cells were split 12–24 h before treatment . Cells were then given one of the following treatments . ( 1 ) DAC ( 1 or 5 μM; Sigma , MO ) or phosphate-buffered saline for 72 h . Medium containing DAC or phosphate-buffered saline was changed every 24 h . ( 2 ) TSA ( 200 nM; MP Biomedicals , OH ) or an identical volume of ethanol for 24 h . ( 3 ) DAC ( 1 μM ) for 48 h followed by DAC ( 1 μM ) and TSA ( 200 nM ) for an additional 24 h . The timing and sequencing of DAC and/or TSA were based on our preliminary studies as well as published studies [37] . Bisulfite treatment was performed as reported previously [38] . Briefly , 2 μg of genomic DNA was denatured with 2 M NaOH for 10 min , followed by incubation with 3 M sodium bisulfite ( pH 5 . 0 ) for 16 h at 50 °C . After treatment , DNA was purified by using a Wizard Miniprep Column ( Promega , Madison , WI ) , precipitated with ethanol , and resuspended in 30 μl of distilled water . 2 μl of the aliquot were used as template for PCR . We used a quantitative bisulfite pyrosequencing method for all DNA methylation analyses [39 , 40] . Global DNA methylation was measured by the LINE1 methylation as previous report [41] . Primer sequences and PCR conditions for bisulfite pyrosequencing assays are summarized in Table S3 . The methylation levels at different C sites measured by pyrosequencing were averaged to represent the degree of methylation in each sample for each gene . For each assay , set-up included positive controls ( samples after SssI treatment ) and negative controls ( samples after whole genomic amplification ) , mixed experiments to rule out bias , and repeated experiments to assess reproducibility . Optimizing annealing temperature of PCR was used to overcome PCR bias as reported [40] . For selected genes , bisulfite sequencing ( performed at the M . D . Anderson Core Sequencing Facility ) of cloned PCR products was used to confirm methylation of CpG sites within the CGI promoters . For this analysis , we cloned the PCR products into the TA vector pCR2 . 1 ( Invitrogen ) and extracted plasmid DNA from the resulting clones with the use of a QIAprep Spin Miniprep kit ( Qiagen , Valencia , CA ) . A panel of RNA from 20 different normal human tissues was obtained from BD Biosciences ( multiple tissue cDNA panels ) that comprises cerebellum , whole brain , fetal brain , fetal liver , heart , kidney , lung , placenta , prostate , salivary gland , skeletal muscle , spleen , testis , thymus , trachea , uterus , colon , small intestine , spinal cord , and stomach . RNA from normal peripheral blood , sperm and cell lines was prepared by using TRIzol reagents ( Invitrogen ) . Total RNA ( 2 μg ) was used as a template to generate complementary DNA ( cDNA ) by random hexamers and M-MuLV reverse transcriptase ( Roche , Indianapolis , IN ) . Reverse-transcription samples without reverse transcriptase also were included as negative controls . One-thirtieth of the cDNA product was used to amplify a 306-bp product of glyceraldehyde-3-phosphate-dehydrogenase ( GAPDH ) gene as an RNA quality control and one-tenth of the cDNA was used to amplify individual genes . The primer sequences and exons analyzed for RT-PCR were listed in Table S4 . PCR conditions were as follows: the reaction volume was 50 μl; initial denaturation was 15 min at 95 °C , followed by 25 cycles ( for GAPDH ) or 35 cycles ( for other genes ) of 30 s at 95 °C , 30 s at 55 °C , and 30 s at 72 °C , with a final extension at 72 °C for 10 min . PCR products were visualized on 6% polyacrylamide gels stained with ethidium bromide . We used GOstat [18] ( http://gostat . wehi . edu . au/ ) from Gene Ontology Tools ( http://www . geneontology . org/GO . tools . shtml ) for gene ontology analysis and determined the statistical significance of the overlap with each gene ontology ( GO ) category using the Fisher exact test . The default multiple testing correction is the Benjamini-Hochberg procedure [42] to control false discovery rate . For gene expression pattern analysis , we downloaded the original profiles from GNF expression database ( http://expression . gnf . org/ ) using probes corresponding to genes identified as dense-CGI promoters methylated in normal blood . From this database , we were able to obtain gene expression data for 127 genes in a panel of 66 normal tissues/cells . To assign equal weight for expression of each gene , we substituted all raw expression values in each data set by their respective Z-scores , and the Z-score was calculated by ( X − μ ) /σ , where X stands for expression data of each gene in each sample; μ stands for mean of expression of each gene among all samples; and σ stands for standard deviation . To analyze expression for all genes , each tissue/cell was assigned a score by the sum of Z-scores for all genes . For sequence comparison analyses , we identified two groups of autosomal genes with dense-CGI promoters based on MCAM results: the methylated group , containing 138 genes with signal intensity of PBLs relative to fully methylated DNA greater than 1 . 5; and the unmethylated group , containing 2 , 125 genes with signal intensity of PBL relative to fully methylated DNA less than 0 . 1 . The general features of CGIs analyzed in this study include CGI length , GC content , and the ratio of observed to expected CpG frequency . Statistical differences were analyzed by unpaired two-tailed t test . Motif analysis was performed as previously reported [43 , 44] . We generated two sets of sequence databases by extracting 2 kb genomic segments ( from the CGI center ) for methylated and unmethylated CGIs . Using each dataset as input into the MEME algorithm ( http://meme . sdsc . edu/meme/intro . html ) [19] , we obtained the top 20 “best fit” motifs for both the methylated and unmethylated groups ( minLen = 6 , maxLen = 50 , with a position-specific goodness-of-fit p-value less than 10−6 after Bonferroni-correction for multiple testing ) . The 20 top motifs from each group were then individually aligned to the entire two datasets with MAST ( http://meme . sdsc . edu/meme/intro . html ) [20] to determine the number of occurrences of each motif in methylated and unmethylated promoters . The Fisher exact test was used to compare the frequency of each motif between the two groups . We used TRANSFAC ( http://www . gene-regulation . com/cgi-bin/pub/databases/transfac/ ) to search for matches between motifs with known transcription factor binding site . To evaluate if the overlap between motifs and Alu consensus sequence is significantly greater than expected by chance , we used MAST to search for matches and determine p-values .
The National Center for Biotechnology Information ( NCBI ) ( http://www . ncbi . nlm . nih . gov ) accession numbers for the genes studied in this paper are shown in Tables 1 and S1 . All microarray datasets in this paper are available at Gene Expression Omnibus ( http://www . ncbi . nlm . nih . gov/geo/ ) under the accession number GSE8810 .
|
About half of all human genes contain a CpG-rich region called a “CpG island” in the 5′ area , often encompassing the promoter and transcription start site of the associated gene . DNA methylation was initially suggested to control tissue-specific gene expression in mammalian cells , but most promoter region CpG islands were found to be unmethylated regardless of tissue specificity of expression . In this study , we discovered an exceptional subset of autosomal genes associated with dense promoter CpG islands that is methylated in normal tissues . We observed tissue-specific gene silencing correlated with hypermethylation in this class of genes , and provided evidence for a direct role of methylation in maintaining the silencing state . Furthermore , we identified five sequence motifs significantly enriched in this class of genes , suggesting the influence of cis-regulatory elements on the establishment and/or stability of DNA methylation . Together , these results provide important new insights into the role of CpG island methylation in normal development and differentiation .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] |
[
"molecular",
"biology",
"developmental",
"biology",
"homo",
"(human)"
] |
2007
|
Genome-Wide Profiling of DNA Methylation Reveals a Class of Normally Methylated CpG Island Promoters
|
Leishmania virulence factors responsible for the complicated epidemiology of the various leishmaniases remain mainly unidentified . This study is a characterization of a gene previously identified as upregulated in two of three overlapping datasets containing putative factors important for Leishmania’s ability to establish mammalian intracellular infection and to colonize the gut of an insect vector . The investigated gene encodes ATP/GTP binding motif-containing protein related to Leishmania development 1 ( ALD1 ) , a cytosolic protein that contains a cryptic ATP/GTP binding P-loop . We compared differentiation , growth rates , and infective abilities of wild-type and ALD1 null mutant cell lines of L . mexicana . Loss of ALD1 results in retarded growth kinetics but not defects in differentiation in axenic culture . Similarly , when mice and the sand fly vector were infected with the ALD1 null mutant , the primary difference in infection and colonization phenotype relative to wild type was an inability to achieve maximal host pathogenicity . While ability of the ALD1 null mutant cells to infect macrophages in vitro was not affected , replication within macrophages was clearly curtailed . L . mexicana ALD1 , encoding a protein with no assigned functional domains or motifs , was identified utilizing multiple comparative analyses with the related and often experimentally overlooked monoxenous flagellates . We found that it plays a role in Leishmania infection and colonization in vitro and in vivo . Results suggest that ALD1 functions in L . mexicana’s general metabolic network , rather than function in specific aspect of virulence as anticipated from the compared datasets . This result validates our comparative genomics approach for finding relevant factors , yet highlights the importance of quality laboratory-based analysis of genes tagged by these methods .
The genus Leishmania unites parasitic protozoa of the family Trypanosomatidae causing leishmaniases , diseases affecting human and animal populations mainly in tropical and subtropical regions . Clinical manifestations vary from spontaneously healing skin lesions to potentially fatal visceral organ failures . Leishmaniases are a global human health problem with over 350 million people at risk [1 , 2] . Approximately two dozen species of Leishmania pathogenic to humans have been described [3] . These are mainly transmitted by the bite of female phlebotomine sand flies [4 , 5] . Knowledge of factors affecting Leishmania growth and development in both sand fly and vertebrate hosts is fundamental to understanding the complicated epidemiology of leishmaniases and the development of new treatments and preventatives . The life cycle of Leishmania consists of two replicative developmental stages—extracellular promastigotes , which multiply and develop within the sand fly's alimentary tract , and intracellular amastigotes , multiplying within the phagolysosomal vacuoles of their vertebrate host's phagocytic cells [6 , 7] . All differentiation events are characterized by dramatic changes in parasite morphology , composition of cell surface glycolipids , gene expression , and other cellular features [8–10] . Gene products involved in either differentiation or infection maintenance in both hosts are potential virulence factors , i . e . entities determining pathogenicity . Comparative genomics and transcriptomics provide promising new tools to identify crucial factors affecting parasite development in its hosts . For example , analysis of ploidy among Leishmania spp . revealed that the tetrasomic chromosome 30 in L . mexicana is highly enriched for amastigote-specific genes , thus linking chromosome number and adaptation to the vertebrate host in Leishmania [11] . Several orthologs of the genes located on the chromosome 30 in L . mexicana , e . g . amino acid transporters ( LmxM . 30 . 0330 , LmxM . 30 . 0571 , LmxM . 30 . 0870 , LmxM . 30 . 1820 ) , tryparedoxin ( LmxM . 30 . 1960 ) , aquaglyceroporin 1 ( LmxM . 30 . 0020 ) , and one member of the ABC transporter superfamily ( LmxM . 30 . 1290 ) were already implicated in virulence [12–14] . Another example stemmed from the generally accepted view that dixenous ( shuttling between insect vector and vertebrate host ) Leishmania parasites emerged within the clade of monoxenous ( one host ) insect trypanosomatids of the subfamily Leishmaniinae [15–17] . Our previous OrthoMCL analysis on a dataset of 27 annotated trypanosomatid genomes delineated 99 orthologous groups ( OGs ) gained at the basal node of Leishmania [18] . The timing of their acquisition suggests that they may be critical for dixeny , and indeed , this set includes several known virulence factors . As is typical for the Trypanosomatidae , 87 of 99 OGs correspond to proteins of unknown function , highlighting the need for future gene-focused functional studies . In a search for new candidates for Leishmania virulence factors we examined gene expression in three different datasets [18]: i ) genes up-regulated at elevated temperatures in Leptomonas seymouri as a pre-adaptation to dixeny; ii ) genes up-regulated in a virulent isolate of Leishmania major LV561 compared to an attenuated isolate; and iii ) genes overexpressed in the virulent metacyclic promastigotes and amastigotes of L . mexicana M379 . Twenty OGs were found to be shared by at least two of these datasets , eleven of which had known virulence factor annotations , while the remaining nine represented proteins of unknown function that are potentially involved in Leishmania virulence . Of these nine , we selected the only gene located on the tetrasomic amastigote-gene enriched chromosome 30 of L . mexicana ( LmxM . 30 . 2090 ) . Its orthologs were up-regulated in Leptomonas seymouri at 35°C and in L . mexicana amastigotes [18] . Here we investigate the putative function ( s ) of this protein using in silico and in vitro approaches and analyzed its role in Leishmania development in insect vectors and mouse hosts .
Protein sequences encoded by LmxM . 30 . 2090 and its orthologs were extracted from the TriTrypDB v . 9 . 0 [19] and aligned using Muscle v . 3 . 8 . 31 with default parameters [20] . Alignment was visualized using Jalview v . 2 . 9 [21] . Leishmania mexicana ( isolate MNYC/BZ/62/M379 ) culture was maintained in M199 medium ( Sigma-Aldrich , St . Louis , USA ) supplemented with 2 μg/ml Biopterin ( Sigma-Aldrich ) , 2 μg/ml Hemin ( Jena Bioscience GmbH , Jena , Germany ) , 25mM HEPES , 50 units/ml of penicillin , 50 μg/ml of streptomycin and 10% Fetal Bovine Serum , FBS ( all from Life Technologies , Carlsbad , USA ) at 23°C . Both ALD1 knock-out ( KO ) and WT L . mexicana were passaged through insects and mice prior to in vitro analyses . Growth kinetics comparison was performed for 6 days from the starting density of 5 x 105 parasites per ml . Cell numbers were counted using a hemocytometer every 48 hours as described previously [22] in four biological replicates . In vitro differentiation was performed as described elsewhere by varying pH and temperature [23] with modifications . In particular , procyclic and metacyclic promastigotes were collected 8 hours apart for WT and KO lines on days 3 and 11 , respectively . WT amastigotes were collected on day 18 , while KO amastigotes were collected on day 21 of the experiment based on the assessment of cell morphology in culture . For normalization , expression values of LmxM . 07 . 0510 ( gene encoding a 60S ribosomal protein L7a ) and LmxM . 36 . 1140 ( gene encoding a short chain 3-hydroxyacyl-CoA dehydrogenase ) were used [24] . Quantitative PCR analysis ( RT-qPCR ) was performed as described previously [25] . Primer sequences for RT-qPCR are listed in the S1 Table . To ablate LmxM . 30 . 2090 in L . mexicana , both alleles were sequentially replaced with selectable markers for Nourseothricin ( Sat ) and Hygromycin ( Hyg ) . Targeting constructs were generated by fusion PCR [26] . In the first round of PCR , 5' and 3' arms of homology were amplified from the L . mexicana genomic DNA using primers A/B ( or C ) and D ( or E ) /F , respectively ( S1 Table ) . The ORFs of the Sat and Hyg-resistance genes were amplified from the plasmids pF4T7polNLS1 . 4sat and pF4TR1 . 4hyg [27] using primers SAT_5'f/ SAT_3'r and Hyg_5'f/ Hyg_3'r ( S1 Table ) . In the fusion PCR reaction , 5' and 3' arms of homology were combined with either Sat or Hyg-resistance gene and amplified with nested primers G and H ( S1 Table ) . L . mexicana promastigotes were transfected with 5 μg of the targeting constructs as described previously using BTX ECM 630 electroporator ( Harvard Apparatus , Inc , Holliston , USA ) [28] . The first allele knockout cell line ( LmxM . 30 . 2090+/- ) was isolated in complete M199 medium containing 100 μg/ml of Sat ( Jena Bioscience GmbH ) . The L . mexicana LmxM . 30 . 2090-/- ( knock-out , KO ) clones were selected on solid M199 medium supplemented as above with additional 100 μg/ml of Hyg . Correct integration was confirmed by PCR on genomic DNA with specific primers and by Southern blot [29] . In brief , total genomic DNA was isolated using DNeasy Blood & Tissue Kit ( Qiagen , Hilden , Germany ) , digested with Nco I overnight , separated on 0 . 75% agarose gel , and transferred to a Zeta-Probe blotting membrane ( Bio-Rad , Hercules , USA ) . Blots were blocked and hybridized with 32P-labeled PCR probes for Sat , Hyg , 5' UTR- , 3' UTR- , and ORF of LmxM . 30 . 2090 gene . The following primers ( S1 Table ) were used to amplify probes: SBp_SAT_f and SBp_SAT_r ( for Sat ) , SBp_Hyg_f and SBp_Hyg_r ( for Hyg ) , SBp_LmxM . 30 . 2090_5'f and SBp_LmxM . 30 . 2090_5'r ( for 5' UTR ) , SBp_LmxM . 30 . 2090_3'f and SBp_LmxM . 30 . 2090_3'r ( for 3' UTR ) , SBp_LmxM . 30 . 2090_f and SBp_LmxM . 30 . 2090_r ( for LmxM . 30 . 2090 ORF ) . Probes were labeled with radioactive 32P using the DecaLabel DNA Labeling kit ( ThermoFisher Scientific , Waltham , USA ) . PCR confirmation of the correct integration was performed using primers pairs SBp_Hyg_f—SBp_Hyg_r ( expected size KO 0 . 3 kb ) , Hyg190_f—Hyg_3’r ( expected size KO 0 . 8 kb ) , A—SBp_Hyg_r ( expected size KO 1 . 7 kb ) , SBp_SAT_f—SBp_SAT_r ( expected size KO 0 . 3 kb ) . The complete ablation of the LmxM . 30 . 2090 gene was confirmed with primers pairs A—SBp_LmxM . 30 . 2090_r ( expected size wild type , WT 2 . 0 kb ) and LmxM . 30 . 2090_f–LmxM . 30 . 2090_r ( expected size WT 0 . 25 kb ) . See S1 Table for primer sequences . For localization studies , the LmxM . 30 . 2090 gene was tagged with HA3 and expressed from the 18S rRNA locus of L . mexicana . The open reading frame was amplified from the genomic DNA using primers 30 . 2090_NcoI_f—30 . 2090_3xHA_NotI_r ( S1 Table ) . The HA3 tag was included in the reverse primer . The amplified fragment was cloned into pLEXSY-sat2 ( Jena Bioscience GmbH ) . Five μg of the resulting plasmid were linearized with SwaI and transfected as above . LmxM . 30 . 2090-HA3 cell line was isolated in complete M199 medium containing 100 μg/ml of Sat . PCR confirmation of the correct integration was performed using primers SSU_dir and SBp_LmxM . 30 . 2090_r ( expected size 1 . 8 kb ) . Leishmania mexicana expressing HA-tagged LmxM . 30 . 2090 promastigotes were incubated with 100 nM MitoTracker Red CMXRos ( ThermoFisher Scientific ) in M199 medium according to the manufacturer's instructions . Promastigotes were fixed in 1% paraformaldehyde , immobilized on the polylysine-coated coverslips , permeabilized with 0 . 1% Triton X100 in PEM buffer ( 100 mM PIPES pH 6 . 9; 1 mM EGTA; 100 μM MgSO4 ) and blocked in PEMBALG buffer ( PEM buffer; 1% BSA; 0 . 5% cold water fish skin gelatin; 100 mM lysine ) . To detected HA-tagged ALD1 , primary rat anti-HA antibodies ( Roche Diagnostics GmbH , Mannheim , Germany ) and Alexa Fluor 488—conjugated secondary anti-rat antibodies ( Life Technologies ) were used . The coverslips were mounted on slides using VECTASHIELD anti-fade mounting medium with DAPI ( Vector Laboratories , Burlingame , USA ) and observed with a Leica TCS SP8 WLL SMD-FLIM inverted confocal microscope ( Leica Microsystems , Wetzlar , Germany ) . Images were processed in Fiji Image J v . 2 . 0 . 0 [30] . Bone marrow-derived macrophages were differentiated for 7–9 days from the precursor cells of BALB/c mice in the presence of 20% L929 fibroblast cell culture supernatant as a source of macrophage-colony stimulating factor . Differentiated macrophages were cultivated in complete RPMI-1640 medium containing 10% FBS , 1 x PenStrep solution , 2 mM of L-glutamine ( all from Sigma-Aldrich ) , and 50 μM of β-mercaptoethanol at 37°C with 5% CO2 . To assess infection in macrophages , cells ( 5 x 104 per well ) were plated on Lab-Tek chamber slides ( ThermoFisher Scientific ) and infected with WT or KO L . mexicana promastigotes freshly passaged through the mice , at a parasite to macrophage ratio of 5:1 . Cells remained either unstimulated in complete RPMI 1640 medium or were stimulated 2 hours post infection ( p . i . ) with 50 U/ml of IFN-γ ( Bio-Rad ) and 500 ng/ml of LPS ( Sigma-Aldrich ) . Slides were stained with Giemsa at 4 hours , 72 hours , and 6 days p . i . and the percent of infected macrophages and parasite load were counted from four biological replicates with two technical replicates each ( 400 cells per condition ) . To measure NO production and arginase activity , 5 x 104 macrophages per well were plated in the flat bottom Costar 96-well plates ( Sigma-Aldrich ) and infected as above . After 72 h incubation , the supernatant and cell lysate ( in quadruplicates ) were used for nitrite and urea analysis , respectively . The accumulation of NO2- was determined by Griess reagents , the arginase activity was analyzed by measuring the conversion of L-arginine to urea , as previously described [31] . To compare the development of LmxM . 30 . 2090 null mutant ( KO ) and WT of L . mexicana in sand flies , three independent experiments were performed . Laboratory colony of Lutzomyia longipalpis ( Jacobina , Brazil ) was maintained at 26°C under standard conditions as described previously [32] . Sand fly females were fed through a chick skin membrane on a suspension of heat-inactivated rabbit blood containing 106 promastigotes per ml . Blood-fed females were separated and maintained at 26°C . On days 1–2 and 7 post infection ( d . p . i ) females were checked for localization and intensity of infection with light microscope . Infection was graded as light ( < 100 parasites/gut ) , medium ( 100–1 , 000 parasites/gut ) , and heavy ( > 1 , 000 parasites/gut ) as previously described [33] . Smears of dissected and examined guts 7 d . p . i . were air dried , fixed by methanol , stained by Giemsa ( Sigma-Aldrich ) , examined under the Olympus BX51 light microscope equipped with a DP72 CCD camera ( Olympus , Tokyo , Japan ) . Morphometric parameters ( length and width of the cell body , as well as length of the flagella ) of 540 randomly selected promastigotes from 9 females/smears per each group were measured using QuickPHOTO micro v . 3 . 0 ( Promicra , Prague , Czech Republic ) . Three main morphotypes were distinguished and categorized as described previously [34] with the slight modifications: i ) long nectomonads: body length ≥ 12 μm; ii ) short nectomonads ( = leptomonads ) : body length <12 μm; and iii ) metacyclic promastigotes: body length ≤ 8 μm and , simultaneously , flagella/body length ratio > 1 . 5 . Results were evaluated using Statistica v . 6 . 1 ( Quest , Aliso Viejo , USA ) . To quantify the numbers of Leishmania parasites in the guts of female sand flies on days 1 and 7 p . i . , RT-qPCR with Leishmania kinetoplast DNA-specific primers was performed as described before [33] . Total DNA from infected females was extracted using a High Pure PCR Template Preparation Kit ( Roche Diagnostics ) according to the manufacturer's protocol . Log-transformed data were evaluated using Statistica v . 6 . 1 . For mice infections , sand fly females were infected with L . mexicana WT or KO strains as described above and checked for the presence and localization of parasites 8–9 d . p . i . Pools of freshly dissected thoracic midguts ( TM ) , with colonized stomodeal valve and high parasite density , were homogenized in sterile saline solution . Immediately , 5 μl of the suspension ( from ~ 10 sand fly TMs ) was intra-dermally injected into the ear pinnae of a ketamin/xylazin anesthetized BALB/c mouse . Groups of four BALB/c mice of each studied group ( WT vs . KO ) were used in two independent experiments . Disease development was monitored weekly . Mice were sacrificed at the 15th week ( experiment 1–4 mice per each group ) or 13th week ( experiment 2–4 mice per each group ) p . i . and infected ears were used for both Leishmania re-isolation by cultivation and quantification of parasite infestation by RT-qPCR as described above .
TriTryp database annotation of the LmxM . 30 . 2090 [19] revealed no conserved motifs or domains for the predicted 382 amino acid ( aa ) long protein . However , we have previously shown that affiliation to protein families can be identified by manual visual analysis [35] , so we inspected the sequence further . DELTA-BLAST search with default parameters [36] produced no known protein hits , but gave high E-value ( range from 10−7 to 10−32 ) hits to smaller regions of annotated proteins with 23–32% identity . Many of these contained ATP-binding folds such as bacterial UvrB helicase [37] with its classical P-loop [38] . While the putative protein encoded by LmxM . 30 . 2090 does not resemble a helicase , visual inspection revealed sequences resembling the Walker A and B motifs of ATP/GTP binding P-loops [39 , 40] . We named this protein ALD1 ( ATP/GTP binding motif-containing protein related to Leishmania development 1 ) ( S1 Fig ) . Using the current canonical definitions of GxxxxGKT/S for Walker A and ΦΦΦΦD for Walker B ( Φ denotes hydrophobic aa ) , ALD1 contains no P-loop . Only by comparison of ALD1 with the original alignment , which included additional conserved sites and alternative site spacing [39] did we observe a P-loop in ALD1 ( Fig 1A ) . ALD1 still lacks an important Mg2+-binding D residue in the Walker B domain ( substituted by G ) , but this domain is less conserved and some P-loops lack it entirely [40] . The putative Walker B domain we identified in ALD1 may be vestigial and another D ( or E ) residue from the sequence may fulfill this role; there are 4 of them in the 20 aa following the Walker B hydrophobic patch in Fig 1 . A Plasmodium falciparum study demonstrates that significant sequence and spacing deviations are to be expected in the P-loop-containing proteins in parasitic protists [41] . In silico secondary structure inference of ALD1 by PredictProtein [42] revealed a region from aa 193 to 220 strongly predicted as alpha helical ( labeled H ( p ) in S1 Fig ) . A larger region ( dotted area in S1 Fig ) modeled to alpha-helical domains of a range of proteins in Swiss-Model [43] and was also predicted as alpha-helical by PredictProtein with less confidence . Protein families with alpha-helical domains preceding the Walker B domain include the ABC transporter/SMC family , AAA+ superfamily , and MCM proteins [44] . The transmembrane domain prediction programs indicated that ALD1 is unlikely to be a membrane protein . In agreement with this , direct analysis of the tagged protein localization by confocal microscopy revealed it to be cytoplasmic ( Fig 1B ) . If the cryptic ALD1 P-loop is important for Leishmania’s ability to infect multiple hosts , there may be differences in this domain between dixenous Leishmania and monoxenous Leishmaniinae . Indeed , this is the case ( Fig 1A ) . The intact Walker A motif ( GxxxxxGxxGKT/S ) was documented only in Leishmania spp . Similar sequence motifs were present in monoxenous Leptomonas and Crithidia spp . , but the most vital residue in the P-loop ( K117 of the motif A of L . mexicana ALD1 [40] ) was substituted by either E or R residues in these orthologs ( Fig 1A ) . In summary , ALD1 is a cytosolic protein that most likely requires nucleotide binding for a function in Leishmania not shared with its orthologs in related monoxenous flagellates . ALD1 was initially identified as potentially related to adaptation to a dixenous lifestyle [18] . We therefore hypothesized that ALD1 may be involved in parasites’ ability to infect insect or vertebrate hosts . To examine the role of ALD1 in Leishmania virulence , we established LmxM . 30 . 2090-/- L . mexicana strains by homologues recombination . Genes for Sat or Hyg resistance were sequentially transfected into promastigotes replacing all wild type alleles . Southern blot analysis ( Fig 2 ) confirmed replacement of the LmxM . 30 . 2090 alleles with resistance markers and generation of the complete null-mutant . Additional confirmations by PCR , qPCR and RT-qPCR are presented in S2 Fig . Of note , chromosome 30 of L . mexicana is tetraploid . Although only two selectable markers were used , all four alleles were successfully ablated . Molecular mechanisms behind this phenomenon need to be investigated further . The strain , lacking all copies of ALD1 , was designated “ALD1 KO” , and the parental cell line “WT” . We first investigated the effect of LmxM . 30 . 2090 ablation on L . mexicana growth independent of a host infection by comparing cell division kinetics of WT and ALD1 KO strains in vitro . To exclude the negative effect of continuous cultivation [45] , procyclic promastigote cultures were started from Leishmania cells that had previously been passaged first through insects and then mice . Cell growth monitored every 48 hours in a continuously-growing culture revealed that Leishmania lacking ALD1 grew significantly slower than their WT counterparts in vitro ( Fig 3A ) . While WT cells exhibited a typical culture growth profile where initial rapid growth slowed by day 4 and reached a plateau at approximately days 6–8 , ALD1 KO cells grew slowly and remained in the linear growth phase for the duration of the experiment . This result indicates that the role of ALD1 may not be confined to the ability to establish an infection in either an insect or mammalian host . Leishmania’s ability to complete the life cycle is key for survival and infectivity . Facilitating the study of Leishmania cell cycle progression , the transition from procyclic- through metacyclic promastigote stage to the amastigote stage can be reproduced in axenic culture [23] . To discover whether loss of ALD1 resulted in any gross defects in these transitions , both WT and ALD1 KO of L . mexicana were differentiated in vitro and compared for indications of incomplete or altered transitions between life stages . Morphological examination of Leishmania cells in vitro demonstrated normal development of both strains , albeit with differences in timing ( see Materials and methods for details ) that are likely due to the differences in relative replication rates as presented in Fig 3A . Also indicating normal ALD1 KO development in vitro is qRT-PCR expression analysis of previously identified stage-specific markers ( Pfr1d expressed in procyclic and metacyclic promastigotes , Sherp expressed in metacyclic promastigotes , and Amastin expressed in amastigotes ) . Expression of these genes was analyzed for each strain upon the culture reaching the indicated life stage ( Fig 3B , 3C and 3D ) and showed no major defect in ALD1 KO . We documented slight , yet statistically significant , increase in levels of Pfr1d and Sherp in ALD1 KO procyclic and metacyclic promastigotes , respectively , compared to the wild type L . mexicana . A possible explanation for this is compensatory: the slower dividing mutant cells must possess higher levels of these gene products to achieve the same stage of differentiation as their wild type counterparts . In conclusion , our axenic culture studies indicate that while ALD1 is essential for optimal growth of L . mexicana promastigotes , its absence does not present obvious barriers to the life stage transitions we can approximate in culture . Since axenic ALD1 KO promastigotes replicate more slowly than WT cells , it is important to determine whether the loss of ALD1 impacts L . mexicana’s ability to infect its insect host where they also replicate extracellularly . In sand flies , the development of L . mexicana WT and ALD1 KO was studied on days 1–2 and 7 post infections in three independent biological replicates . Both strains established infection well in Lu . longipalpis . Percentages of infected females were over 95% for the WT and about 80% for the ALD1 KO strain on day 7 p . i . ( Fig 4A ) . This difference was statistically significant ( p = 0 . 003 ) , the percentage of infected females remained high in both groups tested . As shown in Fig 4B , no differences were observed in the localization of parasites at late-stage infections ( 7 d . p . i . ) and stomodeal valve colonization was observed in the majority ( > 90% ) of these infected females in both tested groups . On the other hand , the quantitative PCR analysis revealed significantly lower numbers of parasites in ALD1 KO-infected sand flies compared to the WT infected group in both , 1 d . p . i . ( F ( 1; 38 ) = 16 . 56; p < 0 . 001 ) as well as 7 d . p . i . ( F ( 1; 134 ) = 56 . 11; p < 0 . 0001 ) ( Fig 4C ) . This trend can also be observed in Fig 4A , where the parasite burden was estimated directly by microscopic observation rather than indirectly by PCR . The slower growth of ALD1 KO strain in vitro ( Fig 3A ) , and lack of any differences in colonization suggest that the differences in parasite burden are most likely due to slower growth of ALD1 KO in the insect once the infection is started . Finally , morphological analysis of parasite cells obtained from infected female flies 7 d . p . i . ( 540 cells per strain ) revealed that metacyclic promastigotes , the sand fly life stage capable of infecting macrophages , represented ~19% of all forms in both ALD1 KO and WT infections . Remaining two forms , long nectomonads and the developmental precursor of metacyclic promastigotes , short nectomonads ( leptomonads ) , constituted 35% ( WT ) / 17% ( ALD1 KO ) and 46% ( WT ) / 64% ( ALD1 KO ) of flagellates , respectively . As short nectomonads develop from the long form , the higher proportion of short nectomonads in ALD1 KO may be due to enhanced transformation of long to short nectomonads . Interestingly , the stable percentage of metacyclic promastigotes in WT and KO ( despite the differing precursor form ratios ) implies the existence of a yet unknown regulatory mechanism for maintaining metacyclic promastigote abundance in sand flies . This morphological analysis represents the single clear relationship we found between ALD1 and Leishmania developmental transitions . To determine whether the ALD1 KO in vitro parasites' growth phenotypes would result in differences in parasites' infectivity in a mammalian host , BALB/c murine infections with ALD1 KO or WT strains were compared . Infection with both strains resulted in clinical symptoms in all inoculated mice . The mean size of nodular lesions at the end of the experiment in WT and ALD1 KO-infected mice was also similar: 8 . 2 ( ± 2 . 5 ) and 6 . 9 ( ± 1 . 1 ) mm , respectively . Quantitative PCR analysis on infected ears from mice sacrificed 13 and 15 weeks p . i . showed high numbers of Leishmania parasites in both groups ( S3 Fig ) . However , the parasite load was slightly but significantly ( F ( 1; 14 ) = 5 . 28; p = 0 . 037 ) higher in the WT control- compared to the ALD1 KO-infected mice . From this analysis in BALB/c mice , we conclude that L . mexicana lacking ALD1 is able to generate a symptomatic mammalian infection . Importantly , however , ALD1 is necessary for achieving maximal L . mexicana parasite load , a finding that may be related to its reduced replication rate in an extracellular environment . To parse the source of the compromised mammalian parasite loads in ALD1 KO infections , we asked whether defects in ALD1 KO virulence existed in infections of cultured primary murine macrophages . We first tested the ALD1 KO strain’s ability to infect unstimulated macrophages at three time points , early infections ( 4 h . p . i . ) , established infections ( 3 d . p . i . ) , and later in infection ( 6 d . p . i . ) . While we observed some differences in percentages of macrophages infected between ALD1 KO and WT , the magnitude of the differences were quite small ( ~10–15% ) and trended both up and down , suggesting that ALD1 KO did not possess a gross defect in the ability to infect macrophages or differentiate to the amastigote stage in the context of a cellular infection ( Fig 5A , left panel ) . Similarly , at the first two time points very little difference in number of amastigotes per cell was observed . However , at the 6 d . p . i . a robust increase in number of amastigotes per cell was observed in the WT , whereas numbers failed to go up in the ALD1 KO , pointing to a defect in replication once intracellular infection was established ( Fig 5A , right panel ) . We then examined whether results of unstimulated macrophage infections would be reflected in classically activated macrophages stimulated with LPS/IFN-γ . In this case , at later time points there are far fewer infected macrophages in both WT and ALD1 KO ( Fig 5B , left panel ) . As macrophages stimulated in this manner typically produce nitric oxide ( NO ) which is toxic for parasites , such a result is not unexpected . It did however prompt us to ask whether macrophage production of NO ( a marker of classically activated macrophages ) , or urea ( a marker of alternatively activated macrophages ) [46] differed in cells infected with ALD1 KO and WT L . mexicana strains . We found no significant differences between the ALD1 KO and WT infected non-stimulated or IFN-γ+LPS stimulated macrophages regarding the production of NO ( F ( 1 , 61 ) = 1 . 74 , p = 0 . 19non-stimulated; F ( 1 , 61 ) = 0 . 68 , p = 0 . 41IFN-γ+LPS ) or urea ( F ( 1 , 61 ) = 0 . 27 , p = 0 . 61non-stimulated; F ( 1 , 61 ) = 3 . 76 , p = 0 . 07IFN-γ+LPS ) . Because numbers of infected cells were so low , it was more difficult to arrive at firm conclusions regarding the effect of ALD1 ablation on stimulated macrophage infection . However , we noticed that at 4 h . p . i . similar numbers of macrophages were infected , suggesting that ALD1 KO harbored no defects in initial infection and transition to the amastigote stage . The only differences between WT and ALD1 KO infections are again observed 6 d . p . i . While the percentage of infected cells in WT at 6 d . p . i . was quite low ( 5% ) , there was almost a complete absence of infected cells in ALD1 KO , with only one cell containing a single amastigote observed ( Fig 5B , right panel ) . We interpret this result in the light of activated macrophages’ enhanced ability to kill Leishmania . In WT L . mexicana infections , at 6 d . p . i . percentages of infected cells are barely being maintained despite the amastigotes’ normal replication rate . It would then follow that in the ALD1 KO , a failure to replicate intracellularly would result in a loss of infection overall at this time point . In summary , we concluded that WT and ALD1 KO L . mexicana differ in their ability to survive in primary murine macrophages in vitro , and that this difference lies in amastigotes‘ ability to replicate intracellularly .
We have identified ALD1 as a novel factor affecting the fitness of L . mexicana . Our investigation utilized an ALD1 null mutant for in vitro growth and differentiation studies , and infections of insect vectors , vertebrate hosts , and macrophages . ALD1 ablation lead to diminishment of parasite loads . In host environments , only minor differences were observed in L . mexicana’s ability to progress normally through its life cycle . Thus , the simplest explanation for the reduced parasitaemia in sand flies and mice infected with ALD1 KO is attenuated fitness ( slower growth ) that was observed even in axenic culture promastigotes . Considering the strong reduced growth rate phenotype of ALD1 KO , it is plausible that this protein is deeply wired into the metabolic networks of Leishmania . This view is corroborated by its ancient acquisition by a common ancestor of Leishmaniinae [18] , presumably in the late Cretaceous period [15] . Since this ancestor was a monoxenous trypanosomatid [16] the primordial role of the ancestral LmxM . 30 . 2090 homologue may be a broad one relevant for survival in insects , and ALD1 may have retained a similar role . It is important to note that orthologs of LmxM . 30 . 2090 are restricted to Leishmaniinae ( monoxenous Crithidia , Leptomonas , Lotmaria , Novymonas , Zelonia , united with dixenous Leishmania ) and were not found in other trypanosomatids outside of this clade . This implies that ALD1 may have provided evolutionary advantages to the ancestor of Leishmaniinae and allowed it to colonize a wider range of insect hosts by outcompeting slower dividing ALD1-negative kins . At present it would be premature to discuss the mechanisms governing ALD1’s role in parasite growth and development in hosts . However , since ALD1 shows a global importance for L . mexicana , identifying the cellular pathway ( s ) to which it contributes could reveal proteins that are even more vital for basic L . mexicana function . To that end , biochemical assaying of recombinant ALD1 nucleotide binding and potentially ATP/GTPase activity as well are warranted in the future . Likewise , tagged ALD1 protein pull-down studies to identify binding and interacting proteins would be useful , as some of these may contain domains and motifs that may give a clearer idea of the cellular pathways to which ALD1 contributes . It is possible that future studies will also reveal an explanation for LmxM . 30 . 2090 upregulation in amastigote-stage cells [18] that is not evident from this initial characterization .
|
Leishmaniases are human parasitic diseases caused by several species of insect-transmitted Leishmania . The pathogenic genus Leishmania has close relatives that also colonize the guts of insects but are not able to infect mammalian cells . When the genes and gene expression profiles of pathogenic Leishmania are compared to those of their nonpathogenic relatives , some genes are either only present or expressed at higher levels in the pathogenic parasites . Protein products of these genes are likely to be involved in the parasites‘ ability to colonize two hosts . Identification of their function will lead to a better understanding of leishmaniases and may reveal potential new drug targets to combat disease . Experiments here are aimed at characterizing a gene identified in this manner . The sequence of the gene’s protein product was examined to identify motifs that gave clues to its function , and the protein itself was determined to be located in the cytosol of Leishmania mexicana cells . When the gene was deleted from L . mexicana , it compromised L . mexicana’s ability to grow quickly in culture , to achieve maximal levels of parasitemia in mice , and to achieve maximal parasite loads in insects .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"immunology",
"microbiology",
"parasitic",
"diseases",
"protozoan",
"life",
"cycles",
"parasitic",
"protozoans",
"developmental",
"biology",
"protozoans",
"leishmania",
"sequence",
"motif",
"analysis",
"molecular",
"biology",
"techniques",
"promastigotes",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"white",
"blood",
"cells",
"artificial",
"gene",
"amplification",
"and",
"extension",
"animal",
"cells",
"bioinformatics",
"life",
"cycles",
"pathogenesis",
"molecular",
"biology",
"amastigotes",
"cell",
"biology",
"host-pathogen",
"interactions",
"polymerase",
"chain",
"reaction",
"database",
"and",
"informatics",
"methods",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"protozoology",
"macrophages",
"organisms"
] |
2017
|
A putative ATP/GTP binding protein affects Leishmania mexicana growth in insect vectors and vertebrate hosts
|
Oocyte meiotic progression and maternal-to-zygote transition are accompanied by dynamic epigenetic changes . The functional significance of these changes and the key epigenetic regulators involved are largely unknown . Here we show that Setdb1 , a lysine methyltransferase , controls the global level of histone H3 lysine 9 di-methyl ( H3K9me2 ) mark in growing oocytes . Conditional deletion of Setdb1 in developing oocytes leads to meiotic arrest at the germinal vesicle and meiosis I stages , resulting in substantially fewer mature eggs . Embryos derived from these eggs exhibit severe defects in cell cycle progression , progressive delays in preimplantation development , and degeneration before reaching the blastocyst stage . Rescue experiments by expressing wild-type or inactive Setdb1 in Setdb1-deficient oocytes suggest that the catalytic activity of Setdb1 is essential for meiotic progression and early embryogenesis . Mechanistically , up-regulation of Cdc14b , a dual-specificity phosphatase that inhibits meiotic progression , greatly contributes to the meiotic arrest phenotype . Setdb1 deficiency also leads to derepression of transposons and increased DNA damage in oocytes , which likely also contribute to meiotic defects . Thus , Setdb1 is a maternal-effect gene that controls meiotic progression and is essential for early embryogenesis . Our results uncover an important link between the epigenetic machinery and the major signaling pathway governing meiotic progression .
Mammalian development begins with fertilization , when the haploid sperm and egg fuse to form the diploid zygote . Although both gametes have equal genetic contributions to the offspring , the early embryo is almost entirely dependent on the egg for the supply of subcellular organelles and macromolecules for initial survival and development [1] . These maternal components are encoded by maternal-effect genes , which are transcribed in oocytes and their products ( RNA or protein ) are present in early embryos before expression of zygotic genes is initiated . Since the identification of the first mammalian maternal-effect genes in 2000 [2 , 3] , multiple such genes have been reported [4] . Genetic studies in mice suggest important roles of maternal-effect genes in developmental processes , including epigenetic reprogramming , zygotic genome activation ( ZGA ) , and cell specification [4] . Despite the progress , the molecular machinery and regulatory mechanisms involved in meiotic progression and maternal-to-zygotic transition are not well understood . In females , meiosis is initiated during fetal development , and oocytes are arrested at prophase I around the time of birth . During subsequent folliculogenesis , the diameters of oocytes increase dramatically , even though prophase I arrest remains in effect . Transcription of the maternal genome occurs predominantly during oocyte growth . Some transcripts are translated immediately into proteins , and others are stored for later activation [1] . Prophase I arrest is sustained until puberty when luteinizing hormone ( LH ) induces resumption of meiosis . The first visible sign of meiotic resumption is nuclear envelope ( called germinal vesicle , GV ) breakdown ( GVBD ) . Following GVBD , a metaphase I spindle forms and stable microtubule-kinetochore interactions are established in all chromosome bivalents before proceeding to anaphase I and telophase I . After completion of meiosis I ( MI ) , as indicated by the extrusion of the first polar body , oocytes enter directly into meiosis II without an intervening S-phase and arrest again at metaphase II ( Met II ) . Fertilization triggers resumption and completion of meiosis II [5] . Meiotic progression is governed by the maturation-promoting factor ( MPF ) , which consists of cyclin-dependent kinase 1 ( Cdk1 , also known as Cdc2 ) and a regulatory subunit Cyclin B1 . In prophase I-arrested GV oocytes , Cdk1 is inactivated by Wee2-mediated phosphorylation on Thr14 and Tyr15 , and Cyclin B1 is constantly degraded by the anaphase-promoting complex/cyclosome ( APC/C ) , a multisubunit E3 ubiquitin ligase . The preovulatory LH surge triggers meiotic resumption by alleviating Cdk1 phosphorylation and inducing Cyclin B1 accumulation [6] . Other kinases and phosphatases also participate in meiotic progression . These include Cdc14b , a highly conserved dual-specificity phosphatase that counteracts the activity of Cdk1 [7] . In somatic cells , Cdc14b has been implicated in multiple cellular processes , including nuclear organization , spindle assembly , mitotic exit , and DNA damage response and repair [7] . In oocytes , Cdc14b is a negative regulator of meiotic progression . Oocytes overexpressing Cdc14b are significantly delayed in resuming meiosis and fail to progress to the Met II stage . Conversely , depletion of Cdc14b in GV oocytes leads to premature meiotic resumption [8] . Cdc14b is also present in preimplantation embryos . Overexpression of Cdc14b in 1-cell embryos has been shown to cause mitotic arrest and inhibit ZGA [9] . These findings suggest that proper regulation of Cdc14b expression is important for meiosis and early embryogenesis . However , little is known about how Cdc14b expression is regulated . Meiotic progression and early embryogenesis are accompanied by drastic chromatin remodeling and epigenetic reprogramming [10 , 11] . Epigenetic events , including posttranslational modifications of histones , are believed to play crucial roles during meiosis and embryogenesis . Indeed , progress has been made in documenting epigenetic states in these processes . For example , during meiotic maturation , histone H3 and H4 are globally deacetylated , whereas H3 lysine 9 di- and tri-methyl ( H3K9me2/me3 ) marks remain constantly high [12 , 13] . However , the functional relevance of epigenetic events and the key epigenetic regulators involved during oogenesis and early embryogenesis remain largely unknown . Setdb1 , also known as Eset and KMT1E , is a lysine methyltransferase ( KMT ) specific for the repressive histone H3 lysine 9 di- and tri-methyl ( H3K9me2/me3 ) marks [14 , 15] . It is associated with transcriptional repression of euchromatic genes and maintenance of heterochromatin structure [14 , 15 , 16] . Recent evidence suggests that Setdb1 also plays a critical role in silencing retrotransposons in undifferentiated embryonic stem ( ES ) cells , as well as in early embryos and primordial germ cells ( PGCs ) , where DNA methylation levels are low due to epigenetic reprogramming [17 , 18] . DNA methylation is required for retrotransposon silencing in somatic cells [19] . Setdb1 is an evolutionally conserved gene . Its Drosophila ortholog dSetdb1 ( also known as dEset and Eggless ) is involved in multiple developmental processes , including oogenesis [20 , 21 , 22] . Mouse embryos lacking Setdb1 die at the peri-implantation stage ( around 3 . 5–5 . 5 days post coitum ( dpc ) ) [23] , which is significantly earlier than the phenotypes of mice deficient for other H3K9 KMTs , such as Suv39h1/Suv39h2 ( developmental defects after ~12 . 5 dpc ) [24] and G9a ( lethality at ~9 . 5 dpc ) [25] . Setdb1 is present at high levels in oocytes and zygotes and persists through preimplantation development [26 , 27] . However , expression of zygotic Setdb1 is undetectable until the blastocyst stage [23 , 26] . These observations suggest that maternal Setdb1 may play important roles in oogenesis and/or early embryogenesis . Here , we show that maternal Setdb1 is essential for meiotic progression in oocytes and mitotic cell cycle progression in early embryos . Conditional deletion of Setdb1 in growing oocytes leads to severe defects in meiotic resumption and maturation , largely due to up-regulation of Cdc14b , resulting in the production of considerably fewer Met II oocytes . Although these Met II oocytes are fertilizable , the resulting embryos display impaired cell cycle progression , progressive delays in preimplantation development , and degeneration before reaching the blastocyst stage . The functions of Setdb1 in these processes require its catalytic activity . Our work identifies Setdb1 as a maternal-effect gene essential for fertility and uncovers a functional link between Setdb1 and the signaling pathway governing meiotic progression .
Although previous work detected Setdb1 transcript and protein in isolated oocytes [23 , 26 , 27] , its expression during oogenesis has not been characterized . We examined Setdb1 expression in the ovary , taking advantage of the availability of the Setdb13lox allele ( schematically shown in S1 Fig ) , which expresses the lacZ -galactosidase reporter under the control of the regulatory elements of endogenous Setdb1 [28] . X-gal ( 5-bromo-4-chloro-3-indoyl-D-galactoside ) staining of paraffin-embedded sections of ovaries from 4-week-old Setdb13lox/+ ( heterozygous ) mice detected strong lacZ signal in growing oocytes , with little staining in granulosa cells ( Fig 1A ) . The lacZ signal was specific , because no staining was observed in ovaries from wild-type ( WT ) mice ( Fig 1A ) . Quantitative RT-PCR ( qRT-PCR ) and Western blot analyses confirmed the presence of Setdb1 transcript and protein in fully-grown GV oocytes ( Fig 1B and 1C ) . These results demonstrated that Setdb1 is actively transcribed and translated during oocyte growth . Zygotic Setdb1 is essential for embryonic development [23] . To determine the role of maternal Setdb1 , we conditionally deleted exon 16 of Setdb1 in oocytes . Deletion of exon 16 would remove 209 amino acids in the catalytic bifurcated SET domain and create a stop codon , thus resulting in a functionally null allele [28] . To maximize the deletion efficiency , heterozygous mice bearing a null allele , Setdb11lox , were first crossed with Zp3-Cre transgenic mice , which express the Cre recombinase exclusively in growing oocytes [29] , and the resulting Setdb11lox/+/Zp3-Cre+ male mice were then crossed with female mice homozygous for the Setdb1 conditional allele , Setdb12lox ( see S1 Fig for breeding scheme ) . Setdb12lox/1lox/Zp3-Cre+ female mice were used as the experimental group and , for simplicity , will be referred to as Setdb1 knockout ( KO ) mice hereafter . Mice of the other genotypes ( Setdb12lox/+/Zp3-Cre- , Setdb12lox/1lox/Zp3-Cre- , and Setdb12lox/+/Zp3-Cre+ ) produced from the breeding scheme showed no defect in fertility and other phenotypic abnormalities , and Setdb12lox/+/Zp3-Cre- female mice were used as the control group . Genotypes were determined by PCR analysis of tail DNA samples ( see S1 Fig for examples ) . qRT-PCR and Western blot analyses confirmed the complete elimination of Setdb1 transcript and protein in Setdb1 KO GV oocytes ( Fig 1B and 1C ) . Consistent with previous findings that Setdb1 is the predominant H3K9 KMT in oocytes and it catalyzes di- and tri-methylation [14 , 27] , immunofluorescence ( IF ) and immunohistochemistry ( IHC ) analyses revealed that the global level of H3K9me2 significantly decreased and that of H3K9me3 slightly decreased in Setdb1 KO oocytes , whereas the levels of H3K9me1 and H3K4me2 showed no alterations ( Fig 1D and 1E and S2 Fig ) . These results indicated that , in oocytes , Setdb1 controls the global level of H3K9me2 mark and its effect on H3K9me3 could be loci-specific . To determine the impact of maternal Setdb1 depletion on fertility , six Setdb1 KO females were mated with WT males for 5 months . Although vaginal plugs were frequently observed , none of the mice produced pups , indicating that maternal Setdb1 is essential for fertility . The infertility phenotype could be due to defects in oogenesis , embryogenesis , or both . We first examined Setdb1 KO ovaries and found that they were morphologically and histologically indistinguishable from control ovaries , with the presence of follicles at various stages ( S3 Fig ) . Fully-grown GV oocytes isolated from Setdb1 KO mice also appeared normal in morphology and number ( S3 Fig ) . These observations indicated that Setdb1 depletion had no effect on folliculogenesis and oocyte growth . We then assessed whether Setdb1 deficiency affected meiotic resumption and maturation . After superovulation , the vast majority ( >90% ) of oocytes collected from the oviducts of control mice , as expected , were arrested at the Met II stage , judged by the presence of a polar body . Although similar numbers of oocytes were recovered from the oviducts of superovulated Setdb1 KO mice , much smaller fractions were at the Met II stage , varying from ~20% to ~60% in different litters . The rest was arrested at the GV stage ( ~15–40% ) , based on the presence of an intact GV , and at MI ( ~15–40% ) , as evidenced by the absence of both GV and polar body , or were abnormal ( ~10–20% ) ( Fig 2A and 2B ) . These observations suggested that a significant fraction of Setdb1 KO oocytes failed to develop to the Met II stage before being released at ovulation . To verify the meiotic arrest phenotype , we isolated GV oocytes and performed in vitro meiotic maturation assays . Fully-grown GV oocytes , when removed from their follicular environment , undergo spontaneous meiotic resumption , which can be reversibly inhibited by cyclic adenosine monophosphate ( cAMP ) phosphodiesterase inhibitors such as 3-isobutyl-1-methylxanthine ( IBMX ) . Control and Setdb1 KO GV oocytes were initially collected in IBMX-containing medium and then cultured in the absence of IBMX for various periods of time . Examination at 2 and 5 hours after IBMX removal revealed that Setdb1 KO oocytes underwent GVBD significantly more slowly than control oocytes ( Fig 2C ) . Following 20 hours of culture , ~90% of control oocytes resumed meiosis , and ~80% progressed to the Met II stage . In contrast , nearly 30% of Setdb1 KO oocytes remained arrested at the GV stage , ~25% arrested at MI , only less than 30% reached the Met II stage , and a considerable fraction ( ~20% ) was abnormal ( Fig 2D ) . These results were consistent with the in vivo data ( Fig 2A and 2B ) , thus confirming that Setdb1 depletion in growing oocytes led to severe defects in meiotic resumption and maturation . A substantial fraction of Setdb1 KO oocytes underwent GVBD but failed to progress to the Met II stage in vivo and in vitro ( Fig 2A–2D ) , suggesting that they were arrested at MI . We therefore assessed whether Setdb1 deficiency affected spindle formation and chromosome dynamics during MI . To this end , GV oocytes were cultured in maturation medium for 5 hours , and the spindle and chromosome structures were examined with α-tubulin and DAPI ( 4' , 6-diamidino-2-phenylindole ) staining . By the time of examination , the majority of control oocytes that had undergone GVBD were at the metaphase I stage , and a small fraction was at prometaphase I . Most of them exhibited normal spindle and chromosome structures ( Fig 2E and 2F ) . Consistent with the delay in meiotic resumption ( Fig 2C ) , ~50% of Setdb1 KO oocytes remained arrested at the GV stage , and the ones that had resumed meiosis were mostly at the prometaphase I stage . Nearly 60% of Setdb1 KO MI oocytes had obvious spindle abnormalities , including dispersed , tread-like , non-polar , and multiple spindles , and ~30% also exhibited defects in chromosome congression or alignment ( Fig 2E and 2F ) . These defects likely played an important part in meiotic arrest at MI . Taken together , our results provided genetic evidence that Setdb1 is critical for meiotic resumption and maturation of mouse oocytes . Given the important role of Setdb1 in transcriptional repression , it is likely that the observed defects in meiotic progression were due to aberrant expression of essential genes . Indeed , previous ChIP-Seq analysis in mouse embryonic stem ( ES ) cells [30] revealed Setdb1 binding , as well as H3K9me3 enrichment , in several genes involved in meiosis , including Cdc14b , Cdc25b , Bub1b , and Ppp2cb ( S4 Fig ) . We performed qRT-PCR analysis to compare the expression of these genes , as well as other important meiosis genes Cdk1 , Ccnb1 ( encoding Cyclin B1 ) , Wee2 , and Fzr1 ( encoding Cdh1 ) , in Setdb1 KO and control GV oocytes . The level of Cdc14b mRNA was substantially elevated in Setdb1 KO oocytes ( ~2 . 8 fold relative to control ) , whereas the expression of the other genes tested showed no alterations ( Fig 3A ) . Western blot and IF analyses also confirmed the increase in Cdc14b protein in Setdb1 KO oocytes ( Fig 3B–3D ) . Notably , the meiotic phenotypes of Setdb1 KO mice , including meiotic arrest , spindle and chromosome perturbations ( Fig 2 ) , were highly similar to the consequences of Cdc14b overexpression [8] . These observations led us to hypothesize that Cdc14b up-regulation may contribute to the meiotic defects associated with Setdb1 deficiency . Cdc14b has been shown to negatively regulate meiotic resumption by promoting APC/C-mediated degradation of Cyclin B1 [8] . We assessed whether Cyclin B1 level was altered in Setdb1 KO oocytes . Western blot analysis revealed that Cyclin B1 was present at low levels in control GV oocytes , but hardly detectable in Setdb1 KO GV oocytes . Following in vitro maturation , both control and KO oocytes that had undergone GVBD exhibited Cyclin B1 accumulation . However , Setdb1 KO GVBD oocytes had ~40% lower levels of Cyclin B1 compared to their control counterparts ( Fig 3E ) . Thus , Cdc14b up-regulation in Setdb1 KO oocytes correlated with low levels of Cyclin B1 . Indeed , the ability of Setdb1 KO GV oocytes to undergo GVBD was substantially restored when treated with the proteosome inhibitor MG132 ( S5 Fig ) , consistent with the notion that enhanced Cyclin B1 degradation contributed to the defect in meiotic resumption . ChIP-Seq analysis of mouse ES cells identified a major Setdb1-binding and H3K9me3 enrichment peak centered at the transcriptional start site ( TSS ) of Cdc14b [30] ( S4 Fig ) , raising the possibility that Setdb1 may directly repress Cdc14b transcription by depositing H3K9 methylation marks . We assessed the impact of Setdb1 depletion on H3K9me3 enrichment at the Cdc14b locus , as well as Cdc14b expression , in ES cells . Because Setdb1-null ES cells are not viable [23] , we used an inducible approach to deplete Setdb1 . Setdb12lox/1lox ES cells expressing tamoxifen-inducible Cre ( known as Cre-ERT2 , a fusion protein consisting of Cre and a mutant form of the estrogen receptor ( ERT2 ) ligand-binding domain ) were treated with 4-hydroxytamoxifen ( 4-OHT ) , which induces translocation of Cre-ERT2 to the nuclei , resulting in excision of exon 16 of the conditional Setdb12lox allele [28] . In agreement with our previous work [28] , Setdb1 mRNA and protein was hardly detectable after 4 days of 4-OHT treatment ( Fig 4A and 4B ) , while the cells ( referred to as Setdb1 KO after treatment ) were still viable and looked healthy . In Setdb1 KO ES cells , both Cdc14b mRNA and protein were considerably elevated ( Fig 4A and 4B ) , consistent with the effect of Setdb1 depletion in oocytes ( Fig 3A–3D ) . Chromatin immunoprecipitation coupled to quantitative real-time PCR ( ChIP-qPCR ) analysis confirmed Setdb1 binding and H3K9me3 enrichment at a region ( R1 ) spanning the Cdc14b TSS , but not at a region ( R2 , negative control ) in intron 1 ( Fig 4C ) . Setdb1 depletion led to a significant reduction in H3K9me3 at the R1 region ( Fig 4C ) , indicating that Setdb1 is responsible for H3K9me3 enrichment at the Cdc14b TSS . Collectively , these results suggest that Cdc14b is a direct transcriptional target of Setdb1 . To test the hypothesis that excess Cdc14b played a key role in inducing meiotic arrest of Setdb1 KO oocytes , we assessed the effect of Cdc14b depletion on meiotic resumption and maturation . Setdb1 KO GV oocytes were microinjected with either Cdc14b or control small interfering RNA ( siRNA ) , and the injected oocytes , as well as control GV oocytes , were incubated in IBMX-containing medium for 24 hours to allow Cdc14 depletion to occur while maintaining GV arrest . The oocytes were then released of GV arrest at the same time by washing out IBMX , followed by in vitro maturation for 20 hours . Analyses at 24 hours post-injection revealed that Cdc14b siRNA led to substantial decreases in Cdc14b mRNA ( by ~70% ) and protein ( by ~55% ) ( Fig 5A–5C ) . Before the initiation of in vitro maturation , almost all oocytes injected with either Cdc14b siRNA or control siRNA remained arrested at the GV stage . Following 20 hours of in vitro maturation , the majority of Cdc14b-depleted oocytes resumed meiosis , with over 50% reaching the Met II stage , whereas microinjection of control siRNA had no effect on the meiotic arrest phenotype ( Fig 5D and 5E , compare with Fig 2D ) . The improvement in meiotic progression appeared to be partially due to the amelioration of spindle defects during MI , as examination of Cdc14b-depeleted oocytes after 6 hours of in vitro maturation revealed a significant lower proportion of MI oocytes exhibiting abnormal spindles ( S6 Fig ) . Taken together , up-regulation of Cdc14b , to a large extent , contributed to the defects in meiotic resumption and maturation in Setdb1 KO oocytes . A subset of retrotransposons , including long terminal repeat ( LTR ) -containing endogenous retroviruses ( ERVs ) and non-LTR long interspersed nuclear element 1 ( Line1 ) , maintain the ability to retrotranspose and thus need to be actively suppressed [19] . Recent studies have demonstrated that Setdb1 is essential for retrotransposon silencing in undifferentiated ES cells , early embryos , and PGCs [17 , 18] . To determine whether Setdb1 is also required to silence retrotransposons in growing oocytes , we measured the transcript levels of retrotransposons in Setdb1 KO and control GV oocytes . As shown in Fig 6A , Setdb1 depletion led to marked up-regulation of Line1 and several ERV elements , including intracisternal A particles ( IAP ) , mouse transposon A ( MTA ) , and the type D murine LTR retrotransposon ( MusD ) . Derepression of retrotransposons could lead to genomic instability [31] . Changes in global H3K9me2 levels could also impair chromatin structure and genome stability [24] . We measured DNA double-strand breaks ( DSBs ) by phosphorylated H2AX ( γ-H2AX ) immunostaining and found that Setdb1 KO oocytes had substantially more γ-H2AX foci than control oocytes ( Fig 6B and 6C ) . DNA DSBs have been shown to adversely affect oocyte meiotic progression [32 , 33] . It is thus likely that DNA damage induced by Setdb1 depletion also played a role in the meiotic arrest phenotype . Despite the defects in meiotic resumption and maturation , a considerable fraction of Setdb1 KO oocytes was able to develop to the Met II stage ( Fig 2A , 2B and 2D ) . To assess the fertilizability and developmental competence of these oocytes , superovulated Setdb1 KO females were mated with WT males , and embryos ( referred to as Setdb1m-z+ for maternal deficient and zygotic wild-type ) were collected at various time points . Examination of the embryos/oocytes collected at 0 . 5 dpc ( E0 . 5 ) suggested that most Setdb1 KO Met II oocytes were fertilizable , as the number of zygotes ( ~44% ) recovered ( Fig 7A and 7B ) was similar to that of Met II oocytes collected from the oviducts of superovulated Setdb1 KO mice ( Fig 2B ) , and only a small number of unfertilized Met II oocytes ( ~4% ) were observed ( Fig 7B ) . Consistent with the meiotic arrest phenotype , considerable numbers of GV ( ~16% ) , MI ( ~10% ) , and abnormal ( ~26% ) oocytes were also present at E0 . 5 ( Fig 7A and 7B ) . At E2 . 5 , the vast majority of control embryos ( referred to as Setdb1m+z+ ) were at the 8-cell and morula stages ( ~48% and ~44% , respectively ) . In contrast , considerable fractions of Setdb1m-z+ embryos remained at the 1-cell ( ~20% ) , 2-cell ( ~15% ) and 4-cell ( ~28% ) stages , and only ~19% developed to the 8-cell stage and very few ( ~5% ) appeared to be morulae with abnormal morphologies ( Fig 7A and 7B ) . At E3 . 5 , Setdb1m+z+ embryos were predominantly at the blastocyst stage ( ~84% ) , as expected , whereas the small numbers of Setdb1m-z+ embryos recovered ( 3 . 2 per litter on average ) were all undergoing degeneration ( Fig 7A and 7B ) . These data revealed that , although Setdb1-depleted Met II oocytes were fertilizable , embryos lacking maternal Setdb1 exhibited progressive delays in development , with most of them undergoing degeneration prior to the morula stage and none of them reaching the blastocyst stage . To confirm the embryonic phenotype , we isolated morphologically “normal” Setdb1m+z+ and Setdb1m-z+ zygotes at E0 . 5 and cultured them for 24–72 hours in vitro . As shown in Fig 7C and 7D , the results were generally consistent with the in vivo data ( Fig 7A and 7B ) , thus strengthening the conclusion that the development of Setdb1m-z+ embryos was severely delayed and defective . The male and female pronuclei form shortly after fertilization and then expand and migrate toward each other before the first mitosis ( Fig 8A ) . In examining E0 . 5 embryos , we noticed that the pronuclear ( PN ) stages of Setdb1m-z+ zygotes were frequently less advanced , as compared to Setdb1m+z+ zygotes . To exclude the possibility that the delayed PN maturation displayed by Setdb1m-z+ zygotes were due to different timing of fertilization , we carried out in vitro fertilization experiments . Met II oocytes from Setdb1 KO and control mice were fertilized with WT sperm , and the PN stages were determined by DAPI staining at 5 hours post-fertilization ( hpf ) . Setdb1m-z+ zygotes were generally delayed in PN maturation . Whereas the vast majority ( ~85% ) of Setdb1m+z+ embryos reached the PN2-3 stages at 5 hpf , only less than 40% of Setdb1m-z+ embryos did , and the rest was mostly at the PN1 stage ( Fig 8B and 8C ) . Delayed PN maturation could reflect impaired cell cycle progression . We therefore measured M-phase entry of zygotes . Control and Setdb1 KO females were mated with WT males , and zygotes collected at E0 . 5 were cultured for 18 hours in the presence of colcemid , which depolymerizes microtubules and arrests zygotes at mitosis . Most Setdb1m+z+ zygotes ( ~85% ) arrested at the M phase with mitotic condensed chromosomes . In contrast , a much smaller fraction ( ~27% ) of Setdb1m-z+ zygotes reached the M phase , and the majority ( nearly 60% ) remained in the interphase ( Fig 8D and 8E ) . Collectively , these results indicated that Setdb1m-z+ embryos had severe defects in progressing through the first mitotic cell cycle . It is likely that subsequent cell cycles were also impaired , given the progressive developmental delays exhibited by these embryos . To determine whether the meiotic and embryonic phenotypes can be rescued by Setdb1 re-expression and whether its catalytic activity is required , Setdb1 KO GV oocytes were microinjected with mRNA encoding Flag-tagged WT Setdb1 ( Flag-Setdb1 ) or Setdb1 with a point mutation altering cysteine 1243 to alanine ( Flag-C1243A ) . The C1243A mutation is located in the bifurcated SET domain ( Fig 9A ) and abolishes the catalytic activity [14 , 15] . The mRNAs for microinjection were produced by in vitro transcription ( with Poly ( A ) tailing ) using plasmid constructs as templates ( S7 Fig ) . IF analysis , performed 2 hours post-injection , confirmed the expression of Flag-tagged Setdb1 proteins ( Fig 9B ) . Following 18 hours of in vitro maturation , the vast majority of Setdb1 KO oocytes expressing Flag-Setdb1 resumed meiosis , with over 50% reaching the Met II stage , albeit the meiotic defects were not completely prevented , as compared to control oocytes . In contrast , the expression of inactive Setdb1 ( Flag-C1243A ) had no effect on the meiotic arrest phenotype , with ~30% of oocytes remaining arrested at the GV stage and only ~20% reaching the Met II stage , similar to uninjected Setdb1 KO oocytes ( Fig 9C and 9D ) . To assess the effect of Setdb1 re-expression on embryonic defects , the Met II oocytes obtained from the in vitro maturation experiments ( described above ) were inseminated with WT sperm , and the embryos were cultured for 48 hours . In all four groups , the majority of oocytes were fertilized , although small fractions ( ~10% ) of unfertilized Met II oocytes were observed . After 48 hours of in vitro development , most embryos derived from control oocytes developed to the morula ( ~55% ) or 8-cell ( ~20% ) stages ( Fig 9E and 9F ) , similar to the results from in vitro development of Setdb1m+z+ embryos ( Fig 7C and 7D ) . Among the embryos derived from Setdb1 KO oocytes , injected or uninjected , considerable fractions ( ~30–35% ) were morphologically abnormal or undergoing degeneration , indicating that the developmental competence of mature eggs derived from Setdb1-expressing oocytes were still severely compromised . However , in the Flag-Setdb1 group , nearly 20% of the embryos reached the morula stage , and another ~10% developed to the 8-cell stage , albeit substantial fractions were arrested at the 1-cell ( 22% ) or 2-cell ( 16% ) stages . In contrast , in the uninjected group , none of the embryos developed beyond the 4-cell stage , and in the Flag-C1243A group , ~50% of embryos were arrested at the 1-cell stage and only a small fraction reached the 2-cell stage ( Fig 9E and 9F ) . It is also noteworthy that Setdb1m-z+ zygotes derived from natural mating failed to develop beyond the 8-cell stage in vitro , even with 72 hours of culture ( Fig 7C and 7D ) . Thus , restoration of Setdb1 activity in Setdb1 KO GV oocytes not only facilitated meiotic progression but also improved the ability of mature oocytes to support early embryogenesis . The partial effects on meiotic and embryonic phenotypes could be because the Setdb1 levels were not optimal or some genomic/chromatin defects that had already occurred could not be remedied by Setdb1 re-expression .
In summary , we demonstrated that , in mouse , maternal Setdb1 controls global H3K9me2 level in developing oocytes , plays crucial roles in meiotic progression , and is essential for preimplantation development . Conditional deletion of Setdb1 in growing oocytes resulted in inhibition of meiotic resumption and impairment of meiotic progression following GVBD , largely due to up-regulation of Cdc14b , a negative regulator of meiotic progression . Other consequences of Setdb1 depletion and altered H3K9 methylation , including derepression of retrotransposons , increased DNA damage , aberrant expression of additional genes , and chromatin defects , likely also contributed to the meiotic arrest phenotype ( Fig 10 ) . Although some Setdb1-deficient oocytes developed to fertilizable eggs , embryos derived from these eggs were severely defective in cell cycle progression and failed to reach the blastocyst stage . Importantly , re-expression of WT Setdb1 , but not catalytically inactive Setdb1 , in Setdb1 KO GV oocytes partially rescued the meiotic and embryonic defects , suggesting that the catalytic activity of maternal Setdb1 is essential for meiotic progression and early embryogenesis . Nevertheless , further work is required to determine how depletion of maternal Setdb1 leads to severe defects in preimplantation development . The consequences of Setdb1 deficiency , including decreased H3K9 methylation , altered gene expression , and genomic and chromatin defects , and/or the lack of maternal Setdb1 itself may affect essential cellular processes ( Fig 10 ) . Our work demonstrates that Setdb1 is a maternal-effect gene essential for fertility . Meiotic progression is accompanied by epigenetic changes . However , little is known about the functional significance of these changes , as well as the key epigenetic regulators involved . The finding that Setdb1 , the predominant histone H3K9 KMT in oocytes , regulates the expression of Cdc14b , a phosphatase that counteracts Cdk1 activity [7] , uncovers a functional link between the epigenetic machinery and the major signaling pathway that governs meiotic progression . While the roles of Cdc14b , as well as the mechanisms involved , in meiosis are not fully understood , there is evidence that Cdc14b promotes Cyclin B1 degradation and regulates meiotic spindle dynamics [8] . Setdb1-deficient oocytes had elevated Cdc14b levels , which correlated with Cyclin B1 reduction , meiotic arrest in GV and MI stages , and abnormal meiotic spindles . Importantly , siRNA-mediated knockdown of Cdc14b in Setdb1-deficient oocytes considerably alleviated the defects in meiotic resumption and maturation . These findings led us to conclude that excess Cdc14b was largely responsible for the meiotic defects in Setdb1-deficient oocytes . Our results suggest that Setdb1 , by keeping Cdc14b below a threshold level , plays important roles in controlling the timing of meiotic resumption and in regulating spindle formation and function following GVBD . The limited number of oocytes one can obtain makes it difficult to determine whether Setdb1 directly or indirectly regulates Cdc14b expression . We performed ChIP analysis in mouse ES cells instead , because Setdb1 depletion leads to identical effect on Cdc14b expression in oocytes and ES cells . Our results confirmed that Setdb1 binds to and deposits the repressive H3K9me3 mark at a region spanning the Cdc14b TSS , thus strongly suggesting that Setdb1 directly represses Cdc14b transcription . Because a subset of retrotransposons remains active , regulatory mechanisms have evolved to suppress their expression . DNA methylation and Setdb1-mediated H3K9 methylation have been shown to silence retrotransposons in somatic cells and undifferentiated cells ( ES cells , early embryos , and PGCs ) , respectively [17 , 18 , 19] . Our finding that Setdb1 is also required for suppressing retrotransposons in developing oocytes suggests a more general role for Setdb1-mediated H3K9 methylation in retrotransposon silencing . Derepression of retrotransposons is often detrimental to the genome . Indeed , Setdb1-deficient oocytes showed increased DNA damage , which likely contributed to the meiotic and embryonic defects . Due to meiotic arrest , Setdb1 KO female mice produced considerably fewer Met II oocytes , which were mostly fertilizable . However , embryos lacking maternal Setdb1 ( Setdb1m-z+ ) exhibited progressive developmental delays , with the vast majority being degenerated prior to the morula stage and none reaching the blastocyst stage . The significantly more severe phenotype and earlier lethality of Setdb1m-z+ embryos , compared to embryos deficient for zygotic Setdb1 ( Setdb1-/- embryos develop to the blastocyst stage and die at 3 . 5–5 . 5 dpc ) [23] , suggest that preimplantation development mainly ( if not entirely ) relies on maternal Setdb1 . We showed that the progression of the first mitotic cell cycle was severely impaired in Setdb1m-z+ embryos , suggesting that maternal Setdb1 may be important for the transitioning from meiotic to mitotic divisions . Based on the progressive developmental delays of these embryos , it is highly likely that subsequent cleavage divisions were also impaired . While the molecular mechanisms underlying these defects remain to be determined , our results suggest that the catalytic activity of maternal Setdb1 is essential for early embryogenesis . Because many cell cycle regulators regulate both meiosis and mitosis [5] , one possibility is that misregulated genes in Setdb1-deficient oocytes not only led to meiotic arrest , but also contributed to the defects in mitotic cell cycle progression in early embryos . Previous studies have shown that Cdc14b is a component of the G2 checkpoint that prevents entry into mitosis following DNA damage and that Cdc14b overexpression in zygotes causes mitotic arrest at the 1- and 2-cell stages and inhibits ZGA [9 , 34] . However , the phenotype of Setdb1m-z+ embryos was much less severe , as compared to Cdc14b-overexpressing embryos [9] . In mouse , ZGA occurs at late two-cell stage and is essential for further development . The fact that a significant fraction of Setdb1m-z+ embryos developed to the 4-cell stage and beyond argues against a general failure in ZGA as the major cause of preimplantation development defects . It is therefore unlikely that Cdc14b elevation played a major role in the developmental defects of Setdb1m-z+ embryos . It is also possible that Setdb1-deficient mature oocytes , albeit fertilizable , had chromatin and genomic defects that impair cellular processes in preimplantation embryos . Maternal Setdb1 persists through preimplantation development and exhibits dynamic localization patterns , implying multiple roles during early embryogenesis [35] . Thus , another possibility is that the lack of Setdb1 activity in early embryos , rather than changes in gene expression and chromatin in oocytes , was mainly responsible for the embryonic defects . These possibilities are not mutually exclusive , and they may all have contributed to the phenotypic abnormalities .
Experimental mice were maintained on a C57BL/6-129Sv hybrid background and used in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory animals , with Institutional Care and Use Committee-approved protocols at The University of Texas MD Anderson Cancer Center ( MDACC ) . The Setdb13lox , Setdb12lox ( conditional ) , and Setdb11lox ( null ) alleles ( schematically shown in S1 Fig ) were described previously [28] . Zp3-Cre transgenic mice were used to disrupt Setdb1 in growing oocytes ( the breeding scheme is shown in S1 Fig ) . Mice were genotyped by PCR , and the primers used are listed in S1 Table . Ovaries from 4-week-old Setdb13lox/+ and WT mice were fixed in 2% paraformaldehyde-0 . 1% glutaraldehyde in phosphate buffered saline ( PBS ) for 1 hour on ice and permeabilized in Rinse buffer ( 2mM MgCl2 , 0 . 01% sodium deoxycholate , 0 . 02% NP-40 in PBS ) three times ( 30 min each ) at room temperature . The tissues were then incubated in X-gal solution ( 1mg/ml X-gal , 5mM potassium ferricyanide and 5mM potassium ferrocyanide in Rinse buffer ) overnight at 37°C , post-fixed in 10% formalin at room temperature , and embedded in paraffin using standard protocols . Ovary sections were deparaffinized and counterstained with nuclear fast red ( Sigma ) and mounted . Fully-grown GV oocytes were obtained from the ovaries of 4–6 week-old female mice 48 hours after intra-peritoneal injection of 5 IU of pregnant mare’s serum gonadotrophin ( PMSG , Sigma ) . Ovaries were placed in a Petri dish with pre-warmed ( 37°C ) M2 medium ( Invitrogen ) supplemented with 200 μM of 3-isobutyl-1-methylxanthine ( IBMX , Sigma ) so as to prevent oocytes from undergoing GVBD . GV oocytes were released by puncturing antral follicles with a fine needle on the stage of a dissecting microscope . To obtain Met II oocytes , 5 IU of human chorionic gonadotrophin ( hCG , Sigma ) was administered 48 hours after PMSG injection . Mice were euthanized the following morning , and oocytes were collected from the oviducts and released into a hyaluronidase/M2 solution for removal of the cumulus cells . For in vitro maturation , oocytes were washed and cultured in IBMX-free M16 medium ( Millipore ) for various periods of time at 37°C in 5% CO2 atmosphere . Epididymis was dissected into pre-warmed ( 37°C ) Human Tubal Fluid ( HTF ) . 4 μl of fresh sperm were added to a 200 μl HTF drop covered with mineral oil and capacitated for 2 hours in the incubator . Then Met II oocytes , either obtained from superovulated mice or derived from in vitro maturation , were added directly to the sperm suspension . After incubating for a maximum of 5 hours at 37°C , 5% CO2 in HTF , eggs were washed with KSOM medium and incubated for various periods of time in KSOM medium at 37°C in 5% CO2 atmosphere . Setdb1 KO and control mice were superovulated and mated with WT males . Fertilized oocytes ( zygotes ) were collected from the oviducts at E0 . 5 and released into a hyaluronidase/M2 solution for dissociation . E2 . 5 embryos were flushed out the infundibulum of the oviducts , and E3 . 5 embryos were flushed out of the uterus . For in vitro embryo development , zygotes were cultured in KSOM medium at 37°C in 5% CO2 atmosphere for 24–72 hours . To knockdown Cdc14b or express Flag-tagged Setdb1 proteins in Setdb1 KO GV oocytes , siRNAs or mRNAs were introduced by microinjection . Briefly , fully-grown GV oocytes were isolated 48 hours after PMSG injection , kept in M2 medium containing IBMX ( 200 μM ) , and injected with either 4 μM of siRNA ( Cdc14b or control ) or 10 pl of mRNA ( Flag-Setdb1 or Flag-C1243A , 0 . 2 μg/μl ) with a FemtoJet microinjector . The injected oocytes were incubated in IBMX-containing medium either for 24 hours after siRNA injection or for 2 hours after mRNA injection , followed by in vitro maturation in IBMX-free medium . The Silencer Select Cdc14b siRNA ( s104254 ) and Silencer Select Negative Control No . 1 siRNA ( 4390843 ) were purchased from Life Technologies . ARCA ( Anti-Reserve Cap Anaolog ) capped and poly ( A ) tailed mRNAs encoding Flag-Setdb1 and Flag-C1243A were produced by in vitro transcription using HiScibe T7 ARCA mRNA kit ( with tailing ) from New England Biolabs ( E2060S ) . The templates for in vitro transcription were generated by cloning Flag-Setdb1 and Flag-C1243A cDNAs , respectively , in pBluescript KS ( see S7 Fig for cloning strategy ) . The C1243A mutation was introduced by PCR . Primers used for molecular cloning are listed in S1 Table . All plasmid constructs were confirmed by DNA sequencing . Ovaries were collected and fixed in formalin overnight , processed , and embedded in paraffin by the Pathology Core Services Facility at MDACC using standard protocols . Ovaries were serially sectioned at 5 μm and stained with hematoxylin and eosine ( H&E ) or with periodic acid-Schiff ( PAS ) -hematoxylin . For IHC analysis , paraffin sections were deparaffinized and hydrated in xylene followed by 100% and 95% ethanol . Endogenous peroxidase activity was blocked with 3% H2O2 in water for 10 min . Antigen retrieve was done with 10 mM Citrate Buffer pH 6 . 0 in a microwave oven for 3 min . After blocking slides with Biocare Blocking Reagent ( BS966M ) for 10 min , slides were incubated with respective primary antibodies ( listed in S2 Table ) for 1 hour at room temperature . After incubating with appropriate horseradish peroxidase ( HRP ) -conjugated secondary antibodies ( indicated in S2 Table ) for 30 minutes at room temperature , slides were incubated with DAB monitoring staining development for viewing . Isolated oocytes were washed in PBS containing 1% polyvinylpyrrolidine ( PVP ) , fixed in 3 . 7% paraformaldehyde in PBS for 30 min , permeabilized for 15 min in 0 . 1% Triton X-100 in PBS , and then stained with respective primary antibodies ( listed in S2 Table ) overnight at 4°C . After washing three times with PBS containing 1mg/ml BSA , the oocytes were incubated for 1 hour with appropriate secondary antibodies conjugated to Fluorescein Isothiocyanate ( FITC ) , Texas Red or Alexa Fluor 488 ( indicated in S2 Table ) , followed by incubation with DAPI . Western blot analysis of GV oocytes or ES cells was performed using standard procedures . GV oocytes were collected , washed in PBS containing 1% PVP , and boiled in sodium dodecyl sulfate ( SDS ) sample buffer . For comparisons , the same numbers of oocytes were used to assure equal loading . ES cells were lyzed in lysis buffer ( 20 mM Tris-HCl pH7 . 9 , 25% glycerol , 150 mM NaCl , 1 . 5 mM MgCl2 , 0 . 1% NP-40 , 0 . 2 mM EDTA , and 0 . 5 mM DTT ) supplemented with protease inhibitor cocktail ( 1861279 , Nalgene ) and phosphatase inhibitor cocktail ( 78427 , Nalgene ) . The cells were then sonicated , centrifuged , and the supernatants were measured for protein concentrations using a protein assay kit ( 500–0116 , Bio-Rad ) and boiled in SDS sample buffer . For comparisons , equal amount ( 25 μg ) of total proteins were loaded . The blots were probed with respective primary antibodies ( listed in S2 Table ) by overnight incubation at 4°C , followed by 1-hour incubation at room temperature with appropriate HRP-conjugated secondary antibodies ( indicated in S2 Table ) . Protein bands were detected by Western Lightning ECL Pro detection reagent ( NEL121001EA , PerkinElmer ) . Total RNA was extracted from 50–100 GV oocytes using the PicoPure RNA Isolation Kit ( Life Technologies ) according to the manufacturer's instruction , followed by reverse transcription ( RT ) using Superscript RT kit ( Bio-Rad ) to generate cDNA libraries . qRT-PCR was performed using iTaq Universal SYBR Green Supermix with ABI 7900 Real-Time PCR system ( Applied Biosystems ) using primers ( listed in S1 Table ) for the following genes and transposons: Setdb1 ( NM_1163641 ) , Cdc14b ( NM_172587 ) , Cdc25b ( NM_023117 ) , Bub1b ( NM_009773 ) , Ppp2cb ( NM_017374 ) , Cdk1 ( NM_007659 ) , Ccnb1 ( NM_172301 ) , Wee2 ( NM_201370 ) , Fzr1 ( NM_019757 ) , IAP , MTA , MusD , and Line1 . ChIP assay was performed as previously described [36] , using rabbit polyclonal antibodies to Setdb1 and H3K9me3 or normal rabbit IgG as negative control ( see S2 Table for information about the antibodies ) . Briefly , Setdb12lox/1lox mouse ES cells transfected with pCAG-Cre-ERT2 [28] , as well as WT ES cells , were treated with 2 μM of 4-OHT for 4 days , and the treated cells ( referred to as Setdb1 KO and WT ES cells , respectively ) were fixed with freshly prepared 1% paraformaldehyde for 10 min at room temperature . The cells were harvested and their nuclei extracted , lyzed , and sonicated . The samples were immunoprecipitated with 8 μg of Setdb1 , H3K9me3 , or normal IgG antibodies . The eluted protein:DNA complex was reverse-crosslinked at 65°C overnight . DNA was recovered after proteinase and RNase A treatment and then analyzed by real-time PCR using primers for the Cdc14b locus ( listed in S1 Table ) . Statistical comparisons between samples were made using unpaired t-test or one-way ANOVA , and P < 0 . 05 was considered statistically significant .
|
During oogenesis , oocytes accumulate transcripts and proteins that support meiotic maturation and early embryogenesis . Although a number of such maternal-effect factors have been identified , our knowledge about the molecular machinery that drives meiotic progression and maternal-to-zygotic transition is still limited . In particular , the functional significance of epigenetic changes , which accompany meiotic maturation and early embryogenesis , and the key epigenetic regulators involved are largely unknown . Here , we identify Setdb1 , a lysine methyltransferase specific for the repressive histone H3 lysine 9 ( H3K9 ) methylation , as a maternal-effect factor that is essential for meiotic progression in oocytes and mitotic cell divisions in early embryos in mouse . We show that Setdb1 is highly expressed in growing oocytes and directly represses the expression Cdc14b , a phosphatase that inhibits meiotic progression . Setdb1 is also required to repress retrotransposons and maintain genomic stability in oocytes . Embryos derived from Setdb1-depleted oocytes show severe defects in cell cycle progression , progressive delays in preimplantation development , and degeneration before reaching the blastocyst stage . The roles of Setdb1 in meiotic progression and preimplantation development require its catalytic activity . Our findings demonstrate that Setdb1 is an important regulator of Cdc14b , thus uncovering a molecular link between the epigenetic machinery and the major signaling pathway that drives meiotic progression .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"meiosis",
"retrotransposons",
"gene",
"regulation",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"germ",
"cells",
"zygotes",
"developmental",
"biology",
"oocytes",
"genetic",
"elements",
"embryos",
"small",
"interfering",
"rnas",
"embryology",
"mobile",
"genetic",
"elements",
"animal",
"cells",
"chromosome",
"biology",
"gene",
"expression",
"biochemistry",
"rna",
"cell",
"biology",
"ova",
"nucleic",
"acids",
"genetics",
"transposable",
"elements",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"genomics",
"non-coding",
"rna",
"cyclins"
] |
2016
|
Maternal Setdb1 Is Required for Meiotic Progression and Preimplantation Development in Mouse
|
Altered protein homeostasis underlies degenerative diseases triggered by misfolded proteins , including spinal and bulbar muscular atrophy ( SBMA ) , a neuromuscular disorder caused by a CAG/glutamine expansion in the androgen receptor . Here we show that the unfolded protein response ( UPR ) , an ER protein quality control pathway , is induced in skeletal muscle from SBMA patients , AR113Q knock-in male mice , and surgically denervated wild-type mice . To probe the consequence of UPR induction , we deleted CHOP ( C/EBP homologous protein ) , a transcription factor induced following ER stress . CHOP deficiency accentuated atrophy in both AR113Q and surgically denervated muscle through activation of macroautophagy , a lysosomal protein quality control pathway . Conversely , impaired autophagy due to Beclin-1 haploinsufficiency decreased muscle wasting and extended lifespan of AR113Q males , producing a significant and unexpected amelioration of the disease phenotype . Our findings highlight critical cross-talk between the UPR and macroautophagy , and they indicate that autophagy activation accentuates aspects of the SBMA phenotype .
Many adult onset neurodegenerative disorders are characterized by the accumulation of abnormally folded proteins that self-associate into soluble oligomeric species or coalesce into insoluble protein aggregates . Among these disorders are ones caused by expansions of CAG/glutamine tracts [1] , [2] . Spinal and bulbar muscular atrophy ( SBMA ) , a member of this group , is a progressive neuromuscular disorder caused by an expanded glutamine tract near the amino terminus of the androgen receptor ( AR ) [3] . This mutation leads to hormone-dependent AR unfolding , and to the predominant loss of lower motor neurons in the brainstem and spinal cord of affected males . Clinical onset occurs in adolescence to adulthood and is characterized initially by muscle cramps and elevated serum creatine kinase [4] , [5] . These myopathic features commonly precede muscle weakness , which inevitably develops as the disease progresses and is most severe in the proximal limb and bulbar muscles . Late in the course of disease , the pathologic features of SBMA include loss of motor neurons in the brainstem and spinal cord and the occurrence of myopathic and neurogenic changes in skeletal muscle 6 , 7 . Studies in mouse models have defined several general principles that guide our understanding of SBMA pathogenesis . Transgenic over-expression of the expanded glutamine AR leads to disease , consistent with the notion that toxicity is predominantly mediated through a gain-of-function mechanism [8] , [9] . This toxicity is androgen-dependent in mice and in SBMA patients , an observation that led to recent clinical trials with anti-androgens [10]–[12] . To model SBMA in mice , our laboratory used gene targeting to exchange 1340 bp of mouse Ar exon 1 with human sequence containing 21 or 113 CAG repeats [13] , [14] . Mice expressing the expanded glutamine AR ( AR113Q ) develop androgen-dependent neuromuscular and systemic pathology that models SBMA [14] , [15] , whereas AR21Q males are similar to wild type littermates [13] , [14] . In AR113Q mice , denervation and muscle pathology occur early in life , prior to detectable motor neuron loss , indicating that neuronal dysfunction or distal axonal degeneration and myopathy are early disease manifestations . The notion that pathology arising in muscle contributes to disease is consistent with findings in transgenic mice in which over-expression of the wild type AR in skeletal muscle leads to hormone-dependent myopathy and motor axon loss [16] , and with data showing a rescue of the disease phenotype in SBMA transgenic mice by over-expressing IGF-1 in skeletal muscle [17] . Taken together , these observations focused our attention on the role of skeletal muscle in disease pathogenesis . The cellular pathways by which the expanded glutamine AR mediates toxicity are complex and incompletely understood , with evidence in several model systems showing disruption of gene expression [18]–[23] , alterations in RNA splicing [24] , impairments in axonal transport [25]–[27] and defects in mitochondrial function [28] . Toxicity likely results from the cumulative effects of altering a diverse array of cellular processes , indicating that potential treatments targeting a single downstream pathway are likely to be unsuccessful . These findings prompted us to concentrate instead on understanding the proximal mechanisms that regulate degradation of the mutant protein . Work in cellular and mouse models has established that degradation and aggregation of the polyglutamine AR are regulated by the Hsp90-based chaperone machinery [29] , [30] , and that manipulating the expression or function of Hsp70-dependent E3 ubiquitin ligases markedly affects AR turnover through the ubiquitin-proteasome pathway [31]–[33] . In addition to the chaperone machinery , other pathways regulating protein quality control have been implicated in SBMA pathogenesis . Here we explored the role of the unfolded protein response ( UPR ) , an integrated signal transduction pathway that transmits information about protein folding within the ER lumen to the nucleus and cytosol to regulate protein synthesis and folding and to influence cell survival [34] , [35] . Prior studies showed that amino-terminal fragments of the polyglutamine AR activate the UPR in vitro [36] , but little is known about the role of this pathway in more complex models of disease . We now show that the UPR is activated in skeletal muscle from SBMA patients and AR113Q mice . Moreover , genetic disruption of the ER stress response by deletion of the gene encoding the transcription factor C/EBP homologous protein ( CHOP ) , a mediator of the UPR [34] , accentuates skeletal muscle atrophy in AR113Q mice . Further , we show that enhanced muscle wasting in the setting of CHOP deficiency is due to increased macroautophagy ( hereafter referred to as autophagy ) , a lysosomal protein quality control pathway implicated in the pathogenesis of polyglutamine and motor neuron diseases . In contrast , diminished autophagy due to Beclin-1 haploinsufficiency decreased muscle wasting and extended the lifespan of AR113Q males , unexpectedly ameliorating the disease phenotype . Our findings highlight cross-talk between the UPR and autophagy , and demonstrate that increased autophagy promotes atrophy of SBMA muscle .
To determine whether the ER stress response is activated in SBMA we obtained skeletal muscle from patients and male controls . Gene expression analysis by qPCR demonstrated that SBMA muscle contained significantly higher levels of several mRNAs that are induced in response to ER stress ( Figure 1A ) [34] , [35] . These encoded the ER chaperone immunoglobulin binding protein ( BiP ) , the transcription factors activating transcription factor-4 ( ATF4 ) and its target CHOP , and the ER folding enzyme protein disulfide isomerase ( PDI ) . Further , increased splicing of mRNA encoding X-box binding protein-1 ( XBP1 ) was detected ( Figure 1B ) , indicating that activation of the proximal UPR sensor inositol-requiring protein-1 ( IRE1 ) had occurred . Analysis of proximal hind limb muscle from adult AR113Q male mice similarly demonstrated the induction of mRNAs encoding BiP , ATF4 , CHOP and PDI ( Figure 1C ) . This was associated with increased expression of BiP and PDI proteins , as demonstrated by western blot ( Figure 1D ) . As the neuromuscular phenotype of these mice is both hormone and glutamine-length dependent [14] , we sought to determine whether the occurrence of ER stress was similarly regulated . Surgical castration at 5–6 wks ameliorated the induction of these transcripts in adult AR113Q males , demonstrating that UPR activation was responsive to levels of circulating androgens ( Figure 1C ) . Further , direct comparison with mice generated using the same gene targeting strategy but with only 21 CAG repeats in the Ar gene [13] confirmed that UPR activation was dependent upon the presence of an expanded glutamine tract ( Figure 1E ) . In contrast , we did not detect induction of ER stress-induced mRNAs such as BiP and CHOP in spinal cords of AR113Q males ( Figure 1F ) , nor did we detect increased expression of BiP or PDI proteins in spinal motor neurons ( not shown ) . We conclude that the UPR is activated in skeletal muscle from SBMA patients and knock-in mice . As the UPR plays a central role in protein homeostasis in the ER and influences survival in a cellular model of SBMA [36] , we sought to determine its role in disease pathogenesis in vivo . To accomplish this , we generated AR113Q males deficient in CHOP , a regulator of cell survival during ER stress that we found to be up-regulated in SBMA muscle . CHOP null mice exhibit impaired programmed cell death following pharmacological induction of ER stress [37] . Further , CHOP deficiency accentuates the phenotype of Pelizaeus-Merzbacher Disease mice [38] yet rescues the motor deficits of Charcot-Marie-Tooth 1B mice [39] , demonstrating that deletion of this transcription factor is an informative approach to probing the role of the UPR in model systems . Notably , CHOP null mice do not display neuromuscular pathology , thereby enabling us to assess the outcome of genetic disruption of the UPR on the SBMA phenotype . CHOP deficiency markedly affected AR113Q muscle , the site of UPR activation , by accentuating skeletal muscle atrophy ( Figure 2A , 2B ) . This unexpected effect on muscle fiber size yielded a significant shift in the distribution of fibers towards a smaller cross sectional area , resulting in a mean fiber size that was ∼1/3 smaller than that measured in AR113Q males . In contrast , CHOP null males expressing the wild type AR had muscle fibers that were similarly sized to age matched wild type males ( Figure 2C ) . Although CHOP deficiency did not alter AR113Q total body mass or survival ( not shown ) , our data show that disruption of the UPR by CHOP deletion increased muscle wasting in AR113Q male mice . To determine the mechanism by which CHOP deficiency increased skeletal muscle atrophy , we initially considered the possibility that motor neuron degeneration was more severe in AR113Q mice deficient in CHOP , resulting in enhanced neurogenic atrophy . However , we found no evidence of increased motor neuron loss in the spinal cords of these double mutants ( not shown ) . Furthermore , skeletal muscle expression of mRNAs induced following denervation [40] , including those encoding myogenin and MyoD , was similar in AR113Q and AR113Q , CHOP null males ( Figure S1 ) . These findings suggested that enhanced muscle atrophy in animals deficient in CHOP was not mediated by increased motor neuron loss , but rather reflected augmented activation of a pathway that mediates muscle wasting . To directly test this notion , we first examined the expression of muscle RING-finger protein 1 ( MuRF1 ) and Atrogin1/Muscle Atrophy F-box ( MAFbx ) ( Figure 3A ) , two E3 ubiquitin ligases that are induced in atrophying skeletal muscle and mediate enhanced protein degradation through the proteasome [41] . While modest induction of MuRF1 mRNA was observed in AR113Q muscle , its expression was not further increased by CHOP deficiency . No significant change in MAFbx expression was detected . Additionally , CHOP deficiency did not alter expression of the 20S proteasome subunit in skeletal muscle ( Figure S2 ) . We conclude that enhanced atrophy of hind limb muscle in AR113Q , CHOP null mice was not associated with a significant induction of E3 ligases that promote muscle protein degradation through the ubiquitin-proteasome pathway . These findings prompted us to consider the possibility that another protein degradation pathway underlies the increased atrophy triggered by CHOP deficiency . As recent studies demonstrate that autophagy contributes to skeletal muscle wasting [42] , we next examined the activity of the autophagic pathway following disruption of the UPR . Western blot demonstrated a ∼3-fold increase in the autophagosome marker LC3-II ( microtubule-associated protein 1 , light chain 3-II ) in skeletal muscle from AR113Q , CHOP null mice ( Figure 3B ) . No accumulation of p62 was detected ( Figure 3B ) consistent with the notion that flux through the autophagic pathway was intact following disruption of the UPR . Consistent with the notion that CHOP deficiency induced autophagy in AR113Q muscle , we detected increased expression of mRNAs encoding the autophagy regulators Atg5 , Atg9B , LC3B and UVRAG ( Figure 3C ) . Notably , induction of autophagy was not associated with altered levels of AR protein ( Figure 3D ) or the appearance of AR immunoreactive intranuclear inclusions in skeletal muscle nuclei ( Figure 3E ) . These observations are consistent with a prior report demonstrating that the androgen receptor largely escapes autophagic degradation following its translocation into the nucleus [43] , and indicate that enhanced muscle atrophy in CHOP null mice is independent of changes in AR protein levels . CHOP deficiency did not alter phosphorylation of eukaryotic translation initiation factor 2 alpha ( eIF2 alpha ) or splicing of XBP1 mRNA ( Figure 3F ) , signals generated by the proximal UPR sensors protein kinase RNA-like ER kinase ( PERK ) and IRE1 that have been linked to the regulation of autophagy [44] , [45] . In contrast , we observed a modest , but significant increase in the phosphorylation of c-Jun N-terminal kinases ( JNK ) ( Figure 3F ) , suggesting that signaling through JNK may contribute to enhanced activation of autophagy in AR113Q , CHOP null muscle , as observed in other systems [46] . Our observation of robust UPR activation in AR113Q skeletal muscle raised the possibility that muscle denervation induces ER stress , and that disruption of the UPR by CHOP deficiency enhances wasting by altering the cellular response to ER stress . To first test whether denervation is sufficient to activate the UPR in skeletal muscle , wild type male mice underwent unilateral sciatic nerve transection , and denervated and intact gastrocnemius muscles were harvested at 3 or 7 days post surgery . Denervation significantly increased phosphorylation of eIF2 alpha and splicing of XBP1 mRNA ( Figure 4A ) indicating that activation of the proximal UPR sensors PERK and IRE1 had occurred . Further , gene expression analysis by qPCR demonstrated a significant induction of BiP and CHOP mRNAs in denervated muscle , while ATF4 mRNA levels exhibited a similar trend that failed to reach statistical significance ( Figure 4B ) . We conclude that denervation activated the UPR in skeletal muscle . These results encouraged us to use this system to further explore the relationship between the UPR and autophagy , and to test the notion that CHOP deficiency enhances muscle wasting through the induction of autophagy . Surgical denervation of male mice expressing the wild type AR demonstrated that CHOP deficiency significantly increased activity of the autophagic pathway , similar to our findings in AR113Q muscle . Denervated CHOP null muscle harvested 7 days post surgery contained ∼2 . 5 fold more LC3-II than did wild type muscle ( Figure 4C ) . p62 did not accumulate in CHOP deficient muscle , indicating that flux through the autophagic pathway was intact . CHOP deficiency also accentuated skeletal muscle atrophy following denervation , producing a significant decrease in mean fiber size ( Figure 4D ) . Our findings demonstrate that CHOP deficiency enhances autophagy and increases muscle wasting following denervation . To confirm that autophagy contributes to muscle atrophy following surgical denervation , we transected the sciatic nerve of Beclin-1 haploinsufficient male mice [47] . Beclin-1 ( encoded by Becn1 ) is a critical regulator of autophagy that binds class III phosphoinositide 3-kinase and is both required for the initiation of autophagosome formation and contributes to autophagosome maturation [48] . Mice haploinsufficient for Beclin-1 form fewer autophagosomes in skeletal muscle [49] and therefore allowed us to probe the role of autophagy in the response of muscle to sciatic nerve transection . Muscle haploinsufficient for Beclin-1 exhibited significantly increased mean fiber size compared to either wild type or CHOP null muscle following surgical denervation ( Figure 4D ) supporting a role for autophagy in muscle wasting . To directly test the notion that CHOP deficiency enhanced muscle wasting by activating autophagy , we generated CHOP null mice haploinsufficient for Beclin-1 ( Figure 4E ) . Following denervation , these mice exhibited significantly less atrophy than CHOP null males , demonstrating that the effects of CHOP deficiency on muscle wasting were mediated through autophagy . Our finding that enhanced autophagy triggered by CHOP deficiency promoted muscle wasting in AR113Q mice prompted us to determine the consequences of limiting autophagy on the SBMA phenotype . To accomplish this , we generated AR113Q males haploinsufficient for Beclin-1 . Similar to effects following surgical denervation , Beclin-1 haploinsufficiency significantly increased AR113Q muscle fiber size , although in this case the effect was less robust ( Figure 5A ) . Limiting activity of the autophagic pathway did not alter levels of either the wild type or polyglutamine AR protein ( Figure 5B ) , consistent with the notion that other protein quality control pathways , such as the proteasome , degrade the receptor once localized to the nucleus . Despite the limited changes in AR113Q muscle , Beclin-1 haploinsufficiency had a striking effect on survival . The lifespan of AR113Q males haploinsufficient for Beclin-1 was extended on average by ∼10 wks compared to AR113Q , Beclin-1 wild type littermates ( Figure 6A ) . AR113Q males exhibited a mean survival of 21 . 6 wks; Beclin-1 haploinsufficiency extended mean lifespan by ∼44% to 31 . 1 wks . Lifespan extension was not associated with rescue to wild type levels of body mass or motor performance as measured by grip strength ( Figure 6B , 6C ) . However , AR113Q males haploinsufficient for Beclin-1 aged over 20 weeks maintained motor function while AR113Q , Beclin-1 wild type littermates exhibited a marked drop-off ( Figure 6C ) . Consistent with the notion that the effects of Beclin-1 haploinsufficiency on motor function were most manifest in older mice , we found no change in the age of disease onset ( defined as the point at which grip strength was 5% less than controls ) due to Beclin-1 haploinsufficiency ( Figure 6D ) . Our data indicate that Beclin-1 haploinsufficiency significantly extended the duration of disease by prolonging survival and maintaining motor function of SBMA mice .
The accumulation of misfolded , mutant proteins is a common basis for adult onset neurodegenerative diseases including those caused by CAG/glutamine tract expansions [1] , [2] , and pathways controlling protein homeostasis are central to the cellular response to these stressors . Here we investigated the role of the UPR , a regulator of ER protein quality control [34] , [35] , in the pathogenesis of SBMA , a neuromuscular disease caused by a glutamine tract expansion in the AR . Our findings demonstrate the occurrence of ER stress in skeletal muscle from SBMA patients , AR113Q mice and wild type mice following surgical denervation . To identify the functional consequence of this response , we generated AR113Q mice deficient in the UPR-mediator CHOP , a transcription factor induced downstream of ATF4 following ER stress . We show that CHOP deletion accentuates muscle atrophy in both AR113Q mice and in surgically denervated wild type males . Notably , in both cases , enhanced muscle wasting due to CHOP deficiency is mediated by increased autophagy , a lysosomal protein quality control pathway that has emerged as a central regulator of proteostasis in several protein aggregation neurodegenerative diseases . While CHOP deficiency activates autophagy and enhances muscle wasting in SBMA mice , limiting autophagy by Beclin-1 haploinsufficiency diminishes muscle atrophy , maintains motor function in aged animals and markedly extends lifespan . Our data highlight the central role of the UPR in remodeling the activity of the protein quality control machinery , and indicate that robust activation of autophagy accentuates the muscle atrophy of SBMA . Activation of the UPR has been reported previously in yeast and mammalian cell culture models of polyglutamine disease [36] , [50] , [51] , and the induction of ER stress responsive transcripts has been noted in Huntington disease mice [52] . The findings reported here extend these observations , demonstrating that the ER stress response is triggered in skeletal muscle from both SBMA patients and knock-in mice . Further , we define new aspects of the functional link between the UPR and autophagy . Several mechanisms by which the UPR regulates autophagy have been proposed based on studies in mammalian models , but a role for CHOP has not been identified previously . Data from a cellular model of polyglutamine disease indicate that phosphorylation of eIF2 alpha by PERK mediates the induction of LC3-II [45] , while a recent study in cellular and mouse models of superoxide dismutase 1 ( SOD1 ) -linked ALS show that XBP1 deletion activates autophagy [44] . As CHOP deficiency altered neither phosphorylation of eIF2 alpha nor splicing of XBP1 in AR113Q mice , we suggest that the effects identified here occur through a distinct mechanism . JNK , a downstream target of IRE1 [53] , can also stimulate LC3-II formation [46] , and the occurrence of increased JNK phosphorylation in AR113Q , CHOP null muscle raises the possibility that this signaling pathway contributes to autophagy activation . The functional consequences of altered autophagy in SBMA mice were unexpected and suggest that limiting activity of this pathway is beneficial for certain aspects of the disease phenotype . As the polyglutamine AR resides in the nucleus in the presence of ligand and largely escapes degradation through this pathway [43] , we found that soluble and aggregated species of the mutant AR do not change when mice are deficient in CHOP or haploinsufficient for Beclin-1 . We suggest that this reflects predominant degradation of the AR by the proteasome , a protein quality control pathway active in the nucleus . The extension of AR113Q lifespan by Beclin-1 haploinsufficiency contrasts with findings in Drosophila showing that disruption of autophagy exacerbates degeneration when the polyglutamine AR is expressed in the eye [54] . This difference may reflect variations in the extent to which autophagy is disrupted , as Beclin-1 haploinsufficiency decreases autophagosome number but does not completely block this pathway . Additionally , small molecule activators of autophagy reportedly promote survival of cultured motor neurons expressing the polyglutamine AR [43] , raising the possibility that the findings described here in AR113Q mice reflect predominant effects outside the CNS , such as in skeletal muscle . While activation of autophagy following UPR disruption exacerbates atrophy of SBMA muscle in mice , recent studies in SOD1 models of ALS show that autophagy induction following XBP1 deletion ameliorates the disease phenotype [44] . Mutant SOD1 , a cytosolic protein , is a target for autophagic degradation and stimulating this pathway clears aggregates of the mutant protein . Of the clinical symptoms experienced by SBMA patients , muscle wasting is a substantial contributor to morbidity . Here we show that activation of autophagy significantly enhances atrophy of surgically denervated and AR113Q muscle . In contrast , limiting autophagy prolongs lifespan and maintains motor function in SBMA mice . While the effects of Beclin-1 haploinsufficiency are relatively mild in AR113Q muscle , lifespan extension is striking , and likely reflects benefits of limited autophagy in cell types other than muscle fibers , perhaps including effects on metabolism . Defining the targets affected by Beclin-1 haploinsufficiency that mediate lifespan extension remains an important goal for future work . Notably , strategies to modulate the activity of the autophagic pathway have attracted considerable attention as studies in several polyglutamine disease models have documented degradation of cytoplasmic protein aggregates through autophagy [55] . Efforts are now underway to identify small molecules that activate the autophagic pathway in hopes of ameliorating the phenotypes of these diseases [56] , [57] . Our data suggest that autophagy activators are unlikely to be effective therapeutics for the subset of protein aggregation disorders where nuclear localization of the mutant protein is required for toxicity . Furthermore , in SBMA , the effects of disease on muscle may be accentuated by activation of autophagy . We suggest that alternative approaches to stimulate other components of the protein quality control machinery , such as the Hsp90-based chaperone machinery , are more likely to yield clinical benefits in SBMA and related protein aggregation disorders .
Derivation of mice with targeted Ar alleles containing 21 or 113 CAG repeats in exon 1 was described previously [14] , [15] . Briefly , mice were generated by recombining a portion of human exon 1 encompassing amino acids 31–484 with the mouse Ar gene in CJ7 embryonic stem cells . Male chimeras were mated with C57BL/6J females , and females heterozygous for the targeted Ar allele were backcrossed to C57BL/6J to generate mice used in this study . Surgical castration of 5–6 wk old males was as previously described [14] . Unless otherwise specified , skeletal muscles were harvested from adult AR113Q male mice at 3–5 months , except from castrated AR113Q males , in which case animals were 18 months of age . CHOP deficient mice ( B6 . 129S-Ddit3tm1Dron/J ) [37] were purchased from the Jackson Laboratory and backcrossed to C57BL/6J ten or more generations . Mice with a Becn1 null allele were previously reported [47] and backcrossed to C57BL/6J ten or more generations . All procedures involving mice were approved by the University of Michigan Committee on Use and Care of Animals , in accord with the NIH Guidelines for the Care and Use of Experimental Animals . 7 wk old C57BL/6J , CHOP deficient or Becn1 haploinsufficient male mice congenic to C57BL6/J were used for studies of denervated muscle . Under deep inhaled anesthesia with 2% isoflurane , the right sciatic nerve was exposed at the thigh just below the sciatic notch . Both the proximal and distal sides were ligated with monocryl 4-0 suture , and about 2 mm of sciatic nerve was cut between the ligations to prevent axonal regeneration . At 3 and 7 days after surgery , the right gastrocnemius and tibialis anterior muscle were dissected and frozen for histology or RNA and protein analysis . The contralateral side was used as control . Anonymized SBMA muscle and control biopsy samples were obtained from the University of Michigan Medical School in accordance with IRB procedures and in a manner that assured patient privacy . Additionally , anonymized skeletal muscle was harvested from SBMA patients at autopsy , as approved by the ethics committee of the Nagoya University Graduate School of Medicine and in accordance with the Declaration of Helsinki ( Hong Kong Amendment ) . Muscle was frozen in isopentane chilled by liquid nitrogen , cut in cross section at a thickness of 5 µm and stained by H&E . Digital images were captured using a Zeiss Axioplan 2 imaging system . The area of each muscle fiber was defined using Adobe Photoshop CS4 or ImageJ , and the pixel number was converted to µm2 according to scale . 100 adjacent fibers from each section were measured . Total RNA isolated from tissues with Trizol ( Invitrogen , Carlsbad , CA ) served as a template for cDNA synthesis using the high capacity cDNA archive kit from Applied Biosystems ( Foster City , CA ) . Gene-specific primers and FAM labeled probes ( Human: BiP , Hs99999174_m1; CHOP , Hs99999172_m1; ATF4 , Hs00909568_g1; PDI , Hs00168586_m1; Mouse: BiP , Mm00517691_m1; CHOP , Mm00492097_m1; ATF4 , Mm00515324_m1; PDI: Mm01243184_m1; MAFbx , Mm00499518_m1; MuRF1 , Mm01185221_m1; α-acetylcholine receptor , Mm00431627_m1; Myod1 , Mm00440387_m1; Myog , Mm00446194_m1; Atg5 , Mm00504340_m1; Atg9b , Mm01157883_g1; Maplc3b , Mm00782868_sH; Uvrag , Mm00724370_m1 ) were purchased from Applied Biosystems . TaqMan assays were performed in duplicate using 5 ng aliquots of cDNA on an ABI 7500 Real Time PCR system . Relative expression levels were calculated comparing with the expression of 18S rRNA . Semi-quantitative RT-PCR analysis of Xbp1 RNA splicing was performed using primers ( mouse: 5′-GAACCAGGAGTTAAGAAC-3′ and 5′-AGGCAACAGTGTCAGAGT-3′; human: 5′-GAATGAGTGAGCTGGAACAG-3′ and 5′-GAGTCAATACCGCCAGAATC-3′ ) to amplify 10 ng of cDNA through 22 cycles . One tenth of the total PCR products were resolved on 15% nondenaturing polyacrylamide gels and stained with SYBR Green 1 ( Invitrogen , Eugene , OR ) after electrophoresis . Bands were visualized on a Typhoon Trio+ scanner ( Amersham Biosciences , Pistcataway , NJ ) and analyzed with AlphaImager 2200 software ( Alpha Innotech Corporation , San Leandro , CA ) . Muscle tissue was homogenized in RIPA buffer containing complete protease inhibitor cocktail ( Roche , Indianapolis , IN ) and phosphatase inhibitor ( Thermo scientific , Rockford , IL ) using a motor homogenizer ( TH115 , OMNI International , Marietta , GA ) . Sample lysates were incubated on a rotator at 4°C for 1 hour and the pre-cleared by centrifugation at 15 , 000 g for 15 minutes at 4°C . Samples were resolved by 7 or 10% SDS-PAGE and transferred to nitrocellulose membranes ( Bio-Rad , Hercules , CA ) . Blots were probed with primary antibodies and proteins were visualized by chemiluminescence ( Thermo Scientific , Rockford , IL ) . The AR ( N-20 ) , HSP90 and eIF2α antibodies were from Santa Cruz Biotechnology ( Santa Cruz , CA ) , phospho-eIF2α ( Ser51 ) and phospho-JNK antibodies were from Cell Signaling Technology ( Danvers , MA ) , LC3B antibody was from Novus Biologicals ( Littleton , CO ) , GAPDH , BiP and PDI antibodies were from AbCam ( Cambridge , MA ) , 20S proteasome antibody was from Calbiochem ( Gibbstown , NJ ) and p62 antibody was from American Research Products ( Belmont , MA ) . Western blot quantification was performed using ImageJ . Frozen muscle tissue was sectioned at 5 µm with a cryostat and stained with H&E or NADH . For immunofluorescence , 5 µM frozen sections were stained with an antibody against AR and an Alexa Fluor 594 conjugated secondary antibody ( Invitrogen ) . Confocal images were captured with a Zeiss LSM 510 microscope and a water immersion lens ( ×63 ) . The grip strength meter ( Columbus Instruments ) was positioned horizontally and mice were lowered toward the apparatus . Mice were allowed to grasp the smooth metal triangular pull bar with their fore limbs only , and then were pulled backward in the horizontal plane . The force applied to the bar at the moment the grasp was released was recorded as the peak tension ( kg ) . The test was repeated 5 consecutive times within the same session , and the highest value from the 5 trials was recorded as the grip strength for that animal . Statistical significance was assessed by two-tailed Student's t-test or by ANOVA with the Newman-Keuls multiple comparison test . The distribution of muscle fiber size was analyzed by Mann-Whitney test . All statistics was performed by the Prism 5 ( GraphPad Software , San Diego , CA ) . P values less than 0 . 05 were considered significant .
|
In many age-dependent neurodegenerative diseases , the accumulation of misfolded or mutant proteins drives pathogenesis . Several protein quality control pathways have emerged as central regulators of the turnover of these toxic proteins and therefore impact phenotypic severity . In spinal and bulbar muscular atrophy ( SBMA ) , the mutant androgen receptor with an expanded glutamine tract undergoes hormone-dependent nuclear translocation , unfolding , and oligomerization—steps that are critical to the development of progressive proximal limb and bulbar muscle weakness in men . Here we show that the unfolded protein response ( UPR ) , an endoplasmic reticulum stress response , is triggered in skeletal muscle from SBMA patients and knock-in mice . We find that disruption of the UPR exacerbates skeletal muscle atrophy through the induction of macroautophagy , a lysosomal protein quality pathway . In contrast , impaired autophagy diminishes muscle wasting and prolongs lifespan of SBMA mice . Our findings highlight cross-talk between the UPR and autophagy , and they suggest that limited activation of the autophagic pathway may be beneficial in certain neuromuscular diseases such as SBMA where the nucleus is the essential site of toxicity .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"neurological",
"disorders",
"neurology",
"neurodegenerative",
"diseases"
] |
2011
|
Macroautophagy Is Regulated by the UPR–Mediator CHOP and Accentuates the Phenotype of SBMA Mice
|
Postembryonic development in Caenorhabditis elegans is a powerful model for the study of the temporal regulation of development and for the roles of microRNAs in controlling gene expression . Stable switch-like changes in gene expression occur during development as stage-specific microRNAs are expressed and subsequently down-regulate other stage-specific factors , driving developmental progression . Key genes in this regulatory network are phylogenetically conserved and include the post-transcriptional microRNA repressor LIN-28; the nuclear hormone receptor DAF-12; and the microRNAs LIN-4 , LET-7 , and the three LET-7 family miRNAs ( miR-48 , miR-84 , and miR-241 ) . DAF-12 is known to regulate transcription of miR-48 , miR-84 and miR-241 , but its contribution is insufficient to account for all of the transcriptional regulation implied by the mutant phenotypes . In this work , the GATA-family transcription factor ELT-1 is identified from a genetic enhancer screen as a regulator of developmental timing in parallel to DAF-12 , and is shown to do so by promoting the expression of the LET-7 , miR-48 , miR-84 , and miR-241 microRNAs . The role of ELT-1 in developmental timing is shown to be separate from its role in cell-fate maintenance during post-embryonic development . In addition , analysis of Chromatin Immnoprecipitation ( ChIP ) data from the modENCODE project and this work suggest that the contribution of ELT-1 to the control of let-7 family microRNA expression is likely through direct transcription regulation .
Extensive study of postembryonic development in the nematode Caenorhabditis elegans has advanced our understanding of the temporal regulation of development and the roles of microRNAs ( miRNAs ) in controlling gene expression [1–5] . In C . elegans , developmental timing is regulated by the heterochronic gene network , which directs the transitions among discrete developmental stages largely by initiating the stage-specific of miRNAs that down-regulate other stage-specific factors [6–8] . Many gene products of the C . elegans heterochronic regulatory network are conserved in metazoans , including the LET-7 family of miRNAs and LIN-28 , a post-transcriptional repressor of these miRNAs [9–11] . LET-7 family miRNAs regulate the expression of multiple targets , including LIN-41 , and the LIN-28-LET-7-LIN-41 pathway has been shown to regulate differentiated states of stem cells in both C . elegans and mammals [3 , 4 , 12–17] . The LIN-28-LET-7 axis is important in human physiology and disease , as it is involved in induced pluripotency [17–19] , adult intestinal stem cell function [20] , tissue repair [21] , and malignancy [22 , 23] . During normal development , dafachronic acid steroid hormones are synthesized by C . elegans in response to favorable growth conditions [24] . They stimulate the nuclear hormone receptor ( NHR ) DAF-12 , the vitamin D NHR ortholog , to promote progression from the 2nd larval stage ( L2 ) to the 3rd larval stage ( L3 ) [24–26] by , in part , initiating expression of the LET-7 family of miRNAs , miR-48 , miR-84 , and miR-241 , during or near the L2-to-L3 molt [27 , 28] . In this way , the nuclear-hormone receptor DAF-12 acts as a key switch in the regulation of developmental fate [27–29] . Expression of miRs has been proposed to drive transition from one larval stage to the next [8] . DAF-12/NHR is known to regulate miRNA expression in this system [7] , but cannot itself account for all of the upstream transcriptional regulation of the LET-7 family of miRNA , as its null phenotype is much weaker than that of the LET-7 family itself [25 , 26 , 30] . LET-7 is known to be under-expressed in both daf-12 mutants [28] and alg-1 mutants [31] , and a portion of its promoter region has been identified to be required for correct temporal expression [32] , but the factor ( s ) that directly regulate its transcription are not yet known . Additionally , the transcriptional regulation of LIN-4 remains largely unknown . A previous study determined that LIN-66 provides regulation of developmental timing in parallel to daf-12 , but the molecular function of the LIN-66 protein remains unknown [33] . Recently , the PERIOD homolog lin-42 has been found to negatively regulate the expression of multiple microRNAs , including LIN-4 and LET-7 [34] . The presence of other regulatory factors that act on the transcription of these miRs is implied , and the identification of these factors would significantly advance our understanding of developmental timing regulation as well of miRNA function in general . In this study , we performed a forward genetic screen to identify enhancers of the heterochronic phenotype of daf-12 ( null ) animals; the purpose was to identify new factors that act in parallel to it in the regulation of the heterochronic genetic network . A partial loss-of-function allele of the GATA transcription factor elt-1 was positionally cloned , and the role of ELT-1 in the heterochronic gene network is described .
An EMS-mutagenesis screen was performed to identify mutations that enhance the heterochronic phenotype of daf-12 ( rh61rh411 ) animals . One such enhancer allele was identified and mapped to the elt-1/GATA gene by genetic mapping techniques including genetic and SNP markers , whole-genome shotgun sequencing to identify candidate variations , and transgene-mediated phenotype complementation . As shown in Figs . 1 , 2A–2F , and S1 , and Table 1 , the elt-1/GATA ( ku491 ) mutation causes delayed heterochronic phenotypes when animals are double-mutant for daf-12 ( rh61rh411 ) , but not when animals are daf-12 ( + ) . Specifically , the number of seam cells in the lateral hypodermis is dramatically higher than normal during the 4th larval ( L4 ) and young adult ( Y . A . ) stages ( Fig . 1 ) . In addition , the double-mutants have an L4-stage bursting vulva phenotype ( Table 1 ) , similar to that caused by mutation of delayed heterochronic genes , including let-7 and the three other let-7 family miR genes ( mir-48 , mir-84 , and mir-241 ) [3 , 30] . These results indicate that elt-1/GATA has an important role in regulating developmental timing in parallel to daf-12 . daf-12 is known to regulate developmental timing by promoting the expression of the three LET-7 family miRNAs [27 , 28] , suggesting that elt-1 may regulate developmental timing by acting either in parallel to or upstream of the miRNAs . In addition to the heterochronic phenotypes that are only present when daf-12 is null , elt-1/GATA ( ku491 ) animals have daf-12 ( null ) -independent defects in adult alae formation at the L4 molt and in the maintenance of seam cells during post-embryonic development ( Table 1 ) . These data are consistent with previous studies that have used post-embryonic RNAi against elt-1/GATA to show that it is required for the maintenance of seam-cell cell identities during post-embryonic development and for the formation of adult alae at the L4 molt [35–37] . The timing of premature differentiation of seam-cells in elt-1 ( ku491 ) single mutants and of excessive seam-cell divisions in the elt-1 ( ku491 ) ;daf-12 ( rh61rh411 ) double-mutants is variable but primarily during the L4 stage ( Fig . 1L ) . Additionally , no supernumerary molts were observed , and male elt-1/GATA ( ku491 ) single-mutant animals are able to cross-fertilize hermaphrodites . Among seam-cell nuclei that formed after seam-cell fusion into the hypodermal syncytium , we did not determine the proportion due to duplicate nuclei within fused cells versus in cells with completed divisions . Proliferation of seam cells and the L4 bursting vulva phenotype in elt-1 ( ku491 ) ; daf-12 ( null ) double-mutant animals are novel phenotypes for elt-1/GATA and indicate that it is a heterochronic gene . As shown in Table 1 , the elt-1 ( ku491 ) mutation is fully recessive . Animals with the elt-1 ( ku491 ) mutation in trans to a null allele , elt-1 ( ok1002 ) , have an equivalent or stronger phenotype for alae formation defects and L4 burst vulva when compared to elt-1/GATA ( ku491 ) homozgotes . These data indicate that the elt-1 ( ku491 ) mutation is likely a partial loss-of-function mutation that compromises the role of the ELT-1 protein in post-embryonic developmental timing and adult alae formation , while leaving it competent to function in the embryonic specification of the hypodermis . Interestingly , there is reduced seam-cell proliferation in elt-1 ( ku491 ) -over-elt-1 ( null ) ; daf-12 ( rh61rh411 ) animals compared to elt-1 ( ku491 ) ; daf-12 ( rh61rh411 ) animals . This is likely due to allelic haploinsufficiency of elt-1 ( ku491 ) for the maintenance of post-embryonic seam cell fate as , when daf-12 is wild-type , elt-1 ( ku491 ) -over-elt-1 ( null ) animals also have a decrease in the numbers of seam-cells compared to elt-1 ( ku491 ) single-mutants animals , and elt-1 has previously been shown to be required for the post-embryonic maintenance of seam-cell cell fate [35–37] . The elt-1/GATA ( ku491 ) mutant allele contains a C-to-T substitution in the 48th base-pair of exon 5 , causing a proline-to-serine missense mutation at amino-acid residue 298 of the ELT-1 protein isoform A . ELT-1 has previously been shown to contain two conserved Zinc-finger DNA binding domains , each of which contains a single C-X2-C-X17-C-X2-C motif [38]; proline298 is 6 amino acid residues C-terminal to the second cysteine of the N-terminal Zn-finger DNA binding domain . This proline specifically , and the N-terminal Zinc-finger DNA binding domain overall , is conserved among worms , fish , mice , and humans ( S2 Fig ) . Examining the structure of the murine GATA1 Zn-finger DNA-binding domains ( Protein Databank accession number 3VD6 ) [39] , we found that the amino acid residue conserved with C . elegans ELT-1 proline298 is located within a hairpin fold that bring the four cysteine residues of the N-terminal Zn-finger domain near to the required Zinc molecule ( S3 Fig ) . This suggests that the elt-1 ( ku491 ) mutation could potentially alter the secondary structure of the N-terminal Zn-finger domain in the ELT-1 protein by interfering with the folding required for Zinc binding , causing its functional inactivation . This presumably does not have an effect on the ability of the C-terminal domain to recognize its target sequences as the elt-1 ( ku491 ) mutant has a non-null phenotype . Five alleles of elt-1 were obtained from the million mutation project and examined in the presence of the daf-12 ( rh61rh411 ) mutation ( S1 Table ) ; two showed very mild increases in the number of seam-cells when compared to daf-12 ( rh61rh411 ) animals , but none showed an L4 bursting vulva phenotype or defective alae formation . These alleles contain mutations with mild effects on the ELT-1 protein , as listed in S1 Table , and have no previously described phenotype , so they likely are minor mutations that do not substantially interfere with the normal function of ELT-1 , unlike the proline298-to-serine mutation present in elt-1 ( ku491 ) . In sum , the elt-1/GATA ( ku491 ) allele significantly reduces the normal function of ELT-1/GATA during post-embryonic development , likely by disrupting the DNA binding ability of its N-terminal Zn-finger domain . The well-described heterochronic gene network controls developmental timing in C . elegans [8] . To assess a possible genetic relationship between elt-1/GATA and key genes in the heterochronic gene network , the phenotype of elt-1 ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals was examined while the expression of several key heterochronic genes were each reduced by feeding RNAi , applied starting at the L1 stage . The results of this interaction analysis ( Table 2 ) shows that the heterochronic phenotypes of elt-1/GATA ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals requires normal activity of the products of the heterochronic genes lin-14 , lin-28 , hbl-1 , lin-41 , lin-42 , and mab-10 , as RNAi of these genes significantly reduced both the high seam cell number and busting vulva phenotypes . Additionally , these phenotypes were enhanced by knockdown of the lin-46 gene , and the seam-cell proliferation phenotype was not affected by knockdown of the lin-29 gene . Knockdown of ceh-16 , which is involved in the regulation of seam-cell fate during post-embryonic development [40 , 41] , suppressed both the high seam cell number and busting-vulva phenotypes , while RNAi of either kin-20 or dre-1 , which are both involved in the promotion of late larval fates [42 , 43] , each suppressed the seam-cell proliferation phenotype but not the bursting-vulva phenotype . Interestingly , the daf-12 ( null ) -independent defect in the formation of adult alae seen in elt-1/GATA ( ku491 ) single-mutant animals was not affected by knock-down of any of the genes examined . These results indicate that the defect in developmental timing in elt-1 ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals is likely to be within or upstream of the part of the heterochronic gene network that controls late larval stages , while the daf-12 ( null ) -independent defects of elt-1 ( ku491 ) animals are independent of the heterochronic gene network . As shown in Figs . 1 , 2 , S1 , and Table 1 , the heterochronic phenotypes of the elt-1/GATA ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals are during late stages of post-embryonic development , L4 and Young Adulthood . The LIN-28 and HBL-1 proteins are known to be down-regulated at the L2 molt and during the L3 stage , respectively [8] , which is mostly prior to the emergence of the heterochronic phenotypes of elt-1 ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals . The expression level of these genes’ mRNA at the L4 stage was found to be normally down-regulated in elt-1/GATA ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals ( S4 Fig ) . Therefore , altered expression of lin-28 and hbl-1 is unlikely to be responsible for the phenotypes of elt-1 ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutants . The LIN-41 protein is also highly expressed during larval development , but in contrast to LIN-28 and HBL-1 , it is down-regulated primarily during the L4 stage by the LET-7 miRNA [3 , 13] . The heterochronic phenotypes of elt-1/GATA ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals are at the same developmental stage as when LIN-41 is normally down-regulated , so the dynamics of lin-41 gene expression were examined in these mutant animals . As shown in Fig . 3A , elt-1/GATA ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals fail to down-regulate the level of the lin-41 mRNA , in contrast to animals that are wild-type or carry either mutation individually . Summary descriptive statistics and statistical analysis for all mRNA qPCR results are shown in S4 Fig . To further examine the regulation of lin-41 mRNA during L4 in elt-1 ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals , the integrated transgene pkIs2084 was obtained and crossed into the mutant strains; pkIs2084 contains the beta-galactosidase gene under the control of the pan-hypodermal col-10 promoter and the lin-41 3’ untranslated region ( 3’UTR ) . This reporter has previously been shown to be down regulated in wild-type animals from the L3 stage to the L4 and young adult stages , and that this down regulation requires both LET-7 miRNA and the cofactors needed for miRNA-induced gene silencing ( e . g . , Argonaute ) [3 , 44] . We found that elt-1/GATA ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals fail to down-regulate the reporter correctly ( Fig . 3B-J ) , indicating that those animals are defective in the negative regulation of the lin-41 3’UTR that normally occurs during the L4 stage . These results suggest that elt-1 normally contributes to developmental timing , at least in part , by promoting the down-regulation of lin-41 expression during L4 , and that it may do so by promoting the expression of LET-7 . The LET-7 family of miRNAs ( miR-48 , miR-84 , and miR-241 ) have previously been shown to be expressed during or near the L2 molt to promote developmental progression [27 , 28] , while LET-7 is expressed primarily during L4 and required for the L4-to-Adult transition , largely by down-regulating lin-41 mRNA [3 , 13] . The expression of the LET-7 , miR-48 , miR-84 , and miR-241 miRNAs was therefore examined during the L4 stage in elt-1 ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals using RT-qPCR . As shown in Fig . 4A , elt-1 ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals have deficient expression of LET-7 and each of the LET-7 family of miRNAs ( miR-48 , miR-84 , and Mir-241 ) . The heterochronic miRNA LIN-4 was also analyzed , although its expression is initiated at early larval stages to drive the L1-to-L2 transition , and was therefore expected to be normal in these mutants . Summary descriptive statistics and statistical analysis results are in S2 Table . A recent study examined the binding sites of a wide range of C . elegans transcription factors , including of ELT-1 , using Chromatin Immunoprecipitation followed by high-throughput sequencing [45] . Examining their peak-calls near selected heterochronic genes ( summarized in Table 3 ) , we identified that ELT-1 binds to sites in the likely promoter regions for the DNA sequences encoding the let-7 family miRNAs during the L3 stage , while binding was not detected during the L2 stage . L4 stage-specific ChIP of ELT-1 was not included in this report [44] . To partially replicate their genome-wide ChIP-seq data for ELT-1 , we performed qPCR following ChIP using L3-L4 stage-enriched worms . We designed qPCR primers based on 9 putative GATA transcription-factor binding sites within three kilobases 5’ of the transcription start site of let-7 gene . Among them , one primer set showed statistically-significant enrichment of ELT-1 binding ( Fig . 4B ) . The genomic region corresponding to this primer set ( Ch . X: 14 , 747 , 074 to 14 , 747 , 179; ~1 . 7 kb upstream of the transcription start site ) overlapped with a binding site found in the modENCODE ChIP-seq data ( Ch . X: 14 , 746 , 915 to 14 , 747 , 299; Table 3 ) , supporting their finding that ELT-1 directly regulates the transcription of the let-7 gene .
The study of postembryonic developmental timing in C . elegans has made important contributions to our understanding of the mechanisms of temporal developmental control in multicellular animals , including the initial discovery of miRNAs and of their role in the temporal regulation of key heterochronic genes’ expression [1–3 , 5] . Given that stage-specific expressions of these miRNAs controls the dynamic state of the heterochronic gene network , understanding the regulation of the expression of these miRNAs is an important problem with significant gaps in our current understanding [7] . In this study , we used a genetic enhancer screen to identify the GATA transcription factor ELT-1 as a new heterochronic gene and have shown that it contributes to developmental timing by providing positive regulation of the expression of the developmental timing miRNAs LET-7 , miR-48 , miR-84 , and miR-241 . In C . elegans , the GATA transcription factor elt-1 has previously been shown to be required for formation of the hypodermis during embryonic development [46] and for the maintenance of cell fate in the seam-cell lineage and adult alae formation during post-embryonic development [35–37] . In this paper , analysis of a non-null allele of elt-1 identified from a random mutagenesis screen demonstrates that elt-1 is heterochronic gene that acts in parallel to the nuclear-hormone receptor daf-12 to provide essential regulation of late-larval stage-specific cell fates . Therefore , a genetic screen in a sensitized background with isolation of a partial loss-of-function allele allowed us to genetically separate the post-embryonic roles for elt-1 from its role in embryonic development; the role of elt-1/GATA in developmental timing was previously masked due to both pleiotropism and genetic redundancy . The phenotype of elt-1 ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals is during late developmental stages , with seam cell proliferation during the L4 and Young Adult stages and an L4 bursting vulva phenotype ( Figs . 1–2 , Table 1 ) . Epistasis analysis ( Table 2 ) shows that the heterochronic phenotypes of elt-1 ( ku491 ) ; daf-12 ( rh61rh411 ) mutants require the function of the heterochronic gene network . In addition , the seam-cell proliferation defects , but not bursting-vulva phenotype , can be suppressed by knock-down of genes previously shown to regulate seam-cell maintenance or fate downstream of the heterochronic gene network [40–42] . For genes with partial suppression , such as mab-10 and kin-20 , this would seem most likely due to premature adoption of later cell fates ( suppressing the seam-cell phenotype ) but without precocious expression of LET-7 family miRNAs ( to suppress the bursting-vulva phenotype ) . These data suggest that the molecular defect in elt-1 ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals may be in the expression of an L4-specific regulatory factor . During the L4 stage , the major target of heterochronic miRs is the lin-41 mRNA [3 , 8] , and the down regulation of lin-41 mRNA that occurs during that stage has previously been shown to require LET-7 [13] . Indeed , the RT-qPCR analysis of lin-41 mRNA levels during L4 presented here ( Fig . 3A ) is consistent with a defect in elt-1 ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals at the level of an L4-stage-specific regulatory factor that negatively regulates the lin-41 mRNA . This L4-stage-specific regulatory factor may , in fact , be the LET-7 miRNA , as elt-1 ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals fail to down-regulate the lacZ::lin-41 3’UTR reporter during L4 ( Fig . 3B-J ) and have decreased expression of the LET-7 miR as measured by RT-qPCR ( Fig . 4A ) . The regulation of lin-41 mRNA by LET-7 has previously been shown to be essential for L4-stage-specific developmental progression [3] and the phenotype of elt-1 ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals is consistent with that expected from reduced expression of LET-7 . These data indicate that elt-1 promotes LET-7 expression during the L4 stage , a novel and unexpected finding . While the LET-7 miR qPCR data ( Fig . 4A ) are limited by noise likely intrinsic to the time of the measurement , all of the data , including the phenotypes , mRNA qPCR and lacZ staining , are consistent with the interpretation that elt-1 and daf-12 each provide redundant regulation of LET-7 that is required for its L4-stage-specific expression . However , the decreased expression of LET-7 alone is unlikely to be the sole cause of the developmental timing phenotypes ( Figs . 1 , 2A , S1 , and Table 1 ) or defective L4-stage down-regulation of lin-41 mRNA seen in the elt-1 ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals ( Fig . 3A-J ) , as LET-7 is expressed at a similarly-decreased level in the elt-1 ( ku491 ) ; daf-12 ( rh61rh411 ) double mutants as in each of the single-mutant strains ( Fig . 4A ) , which lack strong developmental timing phenotypes and correctly down-regulate lin-41 mRNA ( Fig . 3A-B ) . However , the three LET-7 family miRNAs ( miR-48 , miR-84 , and miR-241 ) all have a statistically-significant decrease in their expression during the L4 stage in the elt-1 ( ku491 ) ; daf-12 ( rh61rh411 ) double-mutant animals but not in either single-mutant , which likely accounts for the differences in the phenotypes and data . This suggests that both daf-12 and elt-1 promote the expression of miR-48 , miR-84 , and miR-241 , but that this regulation is highly redundant , so that either transcription factor alone is sufficient to promote sufficient expression of the miRs to prevent a gross phenotype in the single-mutant strains , despite defective expression of LET-7 in those single-mutant animals . Fig . 5 is a proposed model for the role of elt-1 in the regulation miRNA expression in the heterochronic gene network . These data are also consistent with previous studies showing that daf-12 regulates developmental progression [25 , 26] and the expression of miR-48 , miR-84 , miR-241 and LET-7 at the L2 molt [28] and L3 stages [27] . A recent study of transcription factor binding sites in C . elegans included ELT-1 [45] , and analysis of their data ( Table 3 ) shows that the ELT-1 protein binds the likely promoter region of the DNA sequences encoding all members of the LET-7 miRNA family ( mir-48 , mir-84 , mir-241 , let-7 ) , supporting the idea that ELT-1 directly regulates the transcription of the let-7 family miRNAs during late larval development . Our independent analysis confirmed ELT-1 binding in the let-7 promoter region ( Fig . 4B ) . This binding site is 1 . 6kb upstream of the previously known temporal regulatory element ( TRE ) [32] , so it would be of interest to determine the relationship between the ELT-1 binding site and the TRE in the regulation of let-7 expression . ELT-1 also has binding sites near the lin-41 , lin-28 , and hbl-1 genes during the L3 stage , but it is unclear whether these sites are functionally significant , as lin-28 and hbl-1 appear to be normally expressed in the elt-1 ( ku491 ) mutants , and the abnormal expression of lin-41 in elt-1 mutants appears to be due to defective down-regulation of its 3’UTR . Intriguingly , ELT-1 protein is expressed in the hypodermis during embryonic development [46] and seam cells during post-embryonic development [36 , 37] , but it remains unclear why it only promotes the expression of LET-7 family miRNAs during late larval stages , rather than throughout development . Perhaps it interacts with other stage-specific transcription factors , undergoes stage-specific post-translational modifications , or its binding sites near those genes are masked or unmasked in a stage-specific manner . In summary , the elt-1/GATA ( ku491 ) allele described in this study has uncovered a function for elt-1 in regulating miRNA expression and developmental timing that was previously masked by pleiotropism and genetic redundancy . This forcefully supports the idea that the robust expression of key developmental timing genes comes from regulation by parallel and redundant regulatory mechanisms . Similar mechanisms of robustness may also be important in regulating miRNA expression in other organisms during critical developmental transitions , such as in the differentiation of stem cells and in the maintenance of differentiated cell states .
Worms were maintained at 20°C and handled as previously described [47] . Additional information can be found in the S1 Text . Phenotypes were scored , and feeding RNAi was performed starting at the L1 stage , as previously described [1 , 2 , 33 , 48] . Statistical comparisons of seam cell phenotypes were performed with Prism 6 using one-way ANOVAs with p-values calculated with Bonferroni’s multiple comparisons method . For L4 bursting vulva rate , data was analyzed with Prism 6 and p-values were calculated by 2-tailed binomial t-test . Additional information can be found in the S1 Text . Stage-specific samples were prepared by picking individual worms from mixed-stage plates based on gross appearance and vulval morphology; 50–100 animals were collected per sample . RT-qPCR was performed as previously described [33 , 49 , 50] with normalization to eef-2 . Expression of miRNAs was measured from the same RNA samples using TaqMan miRNA assay kits ( Invitrogen Corp . ) with normalization to the snoRNA U18 , as recommended by the manufacturer . Statistical analysis was performed with Prism 6 ( GraphPad ) using 2-way ANOVA with p-values calculated using Sidak’s multiple comparisons test . Additional information can be found in the S1 Text . The pkIs2084 integrated reporter [3 , 44] was obtained and crossed into the indicated strains . Staining for lacZ activity was performed as described [51]; saturated staining at any point in the animal was scored as strong positive , visible but unsaturated staining as weak positive , and undetectable staining as negative . modENCODE data sets with stage-specific ChIP-sequencing of an ELT-1::GFP array in strain OP354 at the L2 and L3 stage ( modENCODE data coordinating center identifying numbers 4632 and 3843 , respectively ) were obtained from the website ( listed below ) and examined for transcription factor sites near selected genes [45 , 52 , 53] . The Blacklist Filtered Peak Calls file was used for analysis and is available online at https://www . encodeproject . org/comparative/regulation/#Wormset7 . OP354 strain ( unc-119 ( tm4063 ) ; wgIs354 [elt-1::TY1::EGFP::3xFLAG + unc-119 ( + ) ] ) was synchronized by bleaching and collected at the L3-L4 stages . The ChIP experiment was performed as described previously [54] with minor modifications . Briefly , paraformaldehyde-fixed chromatin was immunoprecipitated with either mouse IgG ( Jackson immunoresearch ) complexed with Protein G beads ( GE healthcare ) or TrapA GFP beads ( Chromotek ) . Following extraction of the immunoprecipitated DNA , qPCR was performed according to the manufacturer’s instruction ( Bioneer ) . Primers specific to let-7 promoter region are as follows ( 5’-3’ ) : forward primer , TCTCACTGTGTGTCAGCCG , and reverse primer , TGCTGACGTACTACCGGTGC5 . The result was normalized to the level of 3’Untranlsated Region of let-7 gene completed from the same immune complexes using the following primers ( 5’-TCGATCTCTGTCCGCTTTGAAAC-3’ , 5’-CAGGAGGTGAAGAACGAGCA-3’ ) .
|
In the nematode roundworm C . elegans , seam cells , a type of adult stem cell , divide in a completely predictable manner throughout post-embryonic development . Study of the control of the timing of these cells’ division and differentiation led to the discovery of the first microRNAs , which are small non-coding RNAs that regulate the expression of protein-coding mRNAs , but knowledge of the regulation of expression of microRNAs themselves within C . elegans stem cells remains incomplete . In this study , the GATA-family transcription factor elt-1 , known to be important for the formation and maintenance of tissues during embryonic and post-embryonic development , is found to regulate the expression of let-7 family microRNAs in stem cells during late developmental stages . It is found to do so redundantly with daf-12 , the only other transcription factor previously known to directly regulate microRNA expression in C . elegans . In addition , the presence of ELT-1 in vivo binding near microRNA coding DNA sequences suggests that its contribution to the regulation of microRNA expression is likely through direct regulation of transcription . Stem cells are important in development , tissue homeostasis , and malignancy , so additional knowledge of the mechanisms underlying their maintenance , renewal , and differentiation is of broad interest .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
The GATA Factor elt-1 Regulates C. elegans Developmental Timing by Promoting Expression of the let-7 Family MicroRNAs
|
The CsrRS ( or CovRS ) two component system controls expression of up to 15% of the genome of group A Streptococcus ( GAS ) . While some studies have suggested that the sensor histidine kinase CsrS responds to membrane perturbations as a result of various environmental stresses , other data have implicated the human antimicrobial peptide LL-37 and extracellular Mg2+ as specific signals . We now report that Mg2+ and LL-37 have opposite effects on expression of multiple genes that are activated or repressed by the transcriptional regulator CsrR . Using a GAS isolate representative of the recently emerged and widely disseminated M1T1 clone implicated in severe invasive disease , we found marked up-regulation by CsrRS of multiple virulence factors including pyrogenic exotoxin A , DNase Sda1 , streptolysin O , and the hyaluronic acid capsular polysaccharide , among others . Topology and surface protein labeling studies indicated that CsrS is associated with the bacterial cell membrane and has a surface-exposed extracellular domain accessible to environmental ligands . Replacement of a cluster of three acidic amino acids with uncharged residues in the extracellular domain of CsrS abrogated LL-37 signaling and conferred a hyporesponsive phenotype consistent with tonic activation of CsrS autokinase activity , an effect that could be overridden by mutation of the CsrS active site histidine . Both loss- and gain-of-function mutations of a conserved site in the receiver domain of CsrR established an essential role for lysine 102 in CsrS-to-CsrR signal transduction . These results provide strong evidence that Mg2+ and LL-37 are specific signals that function by altering CsrS autokinase activity and downstream phosphotransfer to CsrR to modulate its activity as a transcriptional regulator . The representation of multiple antiphagocytic and cytotoxic factors in the CsrRS regulon together with results of in vitro phagocytic killing assays support the hypothesis that CsrRS mediates conversion of GAS from a colonizing to an invasive phenotype in response to signaling by host LL-37 .
Human beings are thought to be the principal if not exclusive host for group A Streptococcus ( S . pyogenes , GAS ) . The organism's primary environmental niche is the human pharynx where GAS can colonize the epithelium without evoking any clinical symptoms , or it can produce local inflammation and symptomatic streptococcal pharyngitis [1] , [2] . GAS also causes impetigo , a superficial skin infection , and , less commonly , severe invasive infections such as necrotizing fasciitis , bacteremia , and streptococcal toxic shock [3] , [4] . The regulated expression of a variety of gene products enhances GAS survival in the human host through a dynamic process of adaptation to stresses that may change depending on the precise anatomic location of the bacteria in the body , environmental factors , and engagement of host defense mechanisms [5] , [6] . Two component regulatory systems ( TCS ) play an important role in such dynamic adaptation of many bacteria to changing environmental conditions [7] , [8] . CsrRS ( also called CovRS ) is the most extensively characterized TCS in GAS . First identified as a regulator of the has operon that encodes the enzymes required for synthesis of the hyaluronic acid capsular polysaccharide , CsrRS has since been shown to affect expression of as much as 15% of the GAS genome including genes encoding many virulence factors [9]–[12] . Genetic evidence and similarity to TCS in other species have suggested that CsrS is a sensor histidine kinase whose phosphorylation state is influenced by environmental signals , while CsrR is a transcriptional regulator whose activity at target promoters is controlled by phosphorylation . It is presumed , but not proven , that phosphorylation of CsrR results from phosphotransfer from CsrS . It has also been proposed that CsrS has a phosphatase activity and can dephosphorylate CsrR [13] . Transcriptional profiling of CsrR- or CsrRS-mutants has indicated that CsrR acts primarily , although not exclusively , as a repressor of gene expression , as mutants exhibit increased expression of most CsrRS-regulated genes , and phosphorylation of CsrR in vitro enhances its binding to regulated promoters [9] , [10] , [14] , [15] . While it is clear that CsrRS influences expression of many important GAS products , a unifying explanation of the adaptive role of the CsrRS system is still unproven . One proposal is that CsrRS represents a system to detect and respond to a variety of environmental stresses , such as elevated temperature , acidic pH , and high osmolarity , all of which might result in alterations in physical properties of the bacterial cell membrane and consequent signaling through CsrS [13] . An alternative model is that CsrS recognizes specific ligands , and that interaction of these ligands with its extracellular domain ( ECD ) results in changes in CsrS autokinase activity and/or phosphatase activity for CsrR . The latter model is based on the findings that increased concentrations of extracellular Mg2+ result in widespread down-regulation of CsrR-repressed genes , an effect dependent on a functional CsrS and not reproduced by other cations [11] , [16] . Thus , Mg2+ may serve as a specific stimulus for activation of CsrS kinase activity with downstream phosphorylation of CsrR . The human antimicrobial peptide LL-37 has been shown to have effects on CsrRS signaling opposite to those of elevated Mg2+ . Concentrations of LL-37 far below those that inhibit GAS growth were shown to stimulate increased expression of the has operon and three other CsrR-repressed genes in a CsrS-dependent fashion [17] . While these two models are not necessarily mutually exclusive , it is difficult to reconcile LL-37 signaling with a model of non-specific membrane perturbation since the effects of LL-37 on gene expression were not reproduced by a broad range of doses of other antimicrobial peptides , including other cathelicidins , of similar or greater antibacterial potency [17] . The highly specific effect of LL-37 to stimulate up-regulation of CsrR-repressed genes suggests that CsrRS functions to detect and counteract host immune effectors that mediate bacterial clearance from the infected host . Circumstantial evidence for such a role comes from isolation of spontaneous CsrRS mutants in the setting of invasive GAS infection , both in patients with severe invasive GAS infection and in experimental animals [18]–[21] . Exposure of wild type GAS to LL-37 or inactivation of CsrRS by mutation results in increased expression of factors that dramatically enhance GAS resistance to opsonophagocytic killing [12] , [17] . These observations suggest that a physiologic role of CsrRS is to detect relatively low concentrations of LL-37 as a signal of mobilization of host defenses including the recruitment of phagocytic leukocytes and to trigger a global transcriptional response that enhances GAS resistance to phagocytosis . We now report the results of further investigation that provides strong support for this hypothesis . LL-37 not only activates expression of the four previously identified loci , but also stimulates either activation or repression of multiple CsrRS-regulated genes . Signaling by LL-37 is dependent on CsrS , which is shown to have a surface-exposed domain on the bacterial cell . Transduction of the LL-37 signal requires specific domains of both CsrS and its cognate regulator CsrR to induce changes in gene expression . A critical consequence of LL-37-mediated CsrRS-signaling is enhanced resistance to phagocytic killing by human blood leukocytes , a bacterial phenotype that is central to both persistence of GAS in the human host and pathogenesis of invasive infection .
Earlier work by Gryllos et al . found that exposure of GAS to subinhibitory concentrations of LL-37 up-regulated expression of hasB , spyCEP/scpC/prtS , mac/IdeS , and SPy0170 , genes that were shown previously to be down-regulated by extracellular Mg2+ in a CsrRS-dependent manner [11] , [17] . Furthermore , the stimulatory effect of LL-37 on CsrRS-regulated gene expression could be blocked by high concentrations of Mg2+ . These findings suggested the hypothesis that Mg2+ and LL-37 act as opposing extracellular signals for the CsrS sensor histidine kinase . To test if other CsrRS-regulated genes also respond to both stimuli , we investigated ten additional genes for their responsiveness to LL-37 and Mg2+ in GAS strain 854 . This strain was chosen for further analyses because initial experimentation showed marked up-regulation of the four previously characterized CsrRS target genes by LL-37 , signaling that was completely abrogated in an isogenic csrS deficient mutant [17] . Furthermore , strain 854 is representative of the widely disseminated M1T1 clone associated with invasive GAS infections over the past three decades [22]–[25] . In the present study , we found that exposure of strain 854 to 100 nM LL-37 resulted in up-regulation of speA ( pyrogenic exotoxin A ) , sda1 ( DNase ) , ska ( streptokinase ) , slo ( streptolysin O ) , nga ( NAD-glycohydrolase ) , and SPy0136 ( hypothetical protein; N . B . , throughout this paper , unnamed open reading frames are designated by SPy numbers according to the SF370 or MGAS315 genome sequences [26] , [27] ) , as assessed by quantitative RT-PCR ( qRT-PCR ) analysis of RNA samples from LL-37-treated and untreated bacteria ( Figure 1 ) . Culture of strain 854 in 15 mM Mg2+ had the opposite effect from that evoked by LL-37 . That is , Mg2+ exposure resulted in down-regulation of these genes relative to their expression at baseline in unsupplemented medium ( Figure 1 ) . Conversely , expression of several genes in strain 854 was repressed by LL-37 and up-regulated by Mg2+ . Genes in the latter category included metB ( putative cystathionine beta-lyase ) , SPy1414 ( putative cation ( potassium ) transport protein ) , grab ( protein G-related α2-macroglobulin-binding protein ) , and speB ( cysteine protease ) ( Figure 1 ) . To verify that the changes in gene expression observed in response to LL-37 resulted in corresponding changes in production of the encoded proteins , we assayed four representative virulence determinants from this group of CsrRS-regulated genes . Growth of strain 854 in the presence of LL-37 resulted in marked increases in SLO and NADase and repression of SpeB , as assessed by western blot , and increased DNase activity ( Figure S1 ) . DNase activity associated with invasive M1T1 isolates such as strain 854 has been shown to be due predominantly to the enzyme encoded by the prophage-associated sda1 gene ( also called sdaD2 ) , a member of the CsrRS regulon [28] . These results corroborate the qRT-PCR data and , together , they extend earlier findings that LL-37 can up-regulate gene expression to include several additional CsrRS-controlled genes . Moreover , they show that expression of certain CsrRS-regulated genes is repressed , rather than stimulated , by LL-37 . For both categories of genes , the effect of Mg2+ is opposite to that of LL-37 , an observation that supports the hypothesis that the two molecules act as functionally antagonistic stimuli for signaling through CsrRS . The predicted histidine kinase CsrS is thought to represent a cell-surface sensor component of the CsrRS TCS that detects and responds to environmental signals . According to secondary structure and membrane protein model predictions , CsrS contains two membrane-spanning domains near the N-terminus that flank a predicted ECD of 151 amino acids [16] . To test these model predictions , membrane and cytoplasmic fractions of wild type GAS 854 and control csrS-deficient strain 854csrSΩ were isolated from whole cell lysates , fractionated by SDS-PAGE , and analyzed by western blot with anti-CsrS serum . CsrS was found exclusively in membranes of wild type bacteria and , as expected , was absent from csrS mutant preparations ( Figure 2A ) . Like CsrS , the unrelated membrane protein OpuABC [11] was also in wild type 854 membranes , but not in the cytoplasmic fractions ( Figure 2A ) . Consistent with its predicted cytosolic localization , the CsrR protein was mainly detected in the cytoplasmic fraction . These results localized CsrS to the GAS cell membrane . In order to test whether CsrS is accessible to signaling molecules in the extracellular environment , we labeled proteins exposed on the bacterial surface with biotin via a disulfide linker . Biotinylated proteins were isolated from bacterial cell lysates using NeutrAvidin resin affinity chromatography . Resin-bound proteins were released by reduction of the disulfide bond linking biotin to the GAS surface proteins , fractionated by SDS-PAGE , and analyzed by western blot with CsrS antiserum . CsrS was detected predominantly in this eluted fraction ( Figure 2B ) , a result that indicates CsrS was accessible to biotinylation , i . e . , that a portion of the protein is exposed to the extracellular environment . CsrR , used here as a control cytosolic protein , did not react with biotin , and was detected only in the unbound flow-through fraction ( Figure 2B ) . These data demonstrate that CsrS is a membrane-associated protein and includes a surface-exposed domain , conclusions consistent with our hypothesis that the ECD of CsrS functions as the sensor domain for environmental signals . We noted previously that the predicted ECD of CsrS includes a cluster of negatively charged amino acids that corresponds to a similar cluster in PhoQ of S . typhimurium and E . coli implicated in binding of cationic ligands [17] , [29] , [30] . Preliminary experiments using wild type or mutant forms of csrS to complement in trans a csrS mutant of M-type 3 GAS strain DLS003 suggested that three charged residues in the ECD were required for LL-37 signaling . However , these experiments were not definitive as the level of CsrS protein expressed from the mutant construct was higher than that observed in the wild type strain [17] . To examine more thoroughly the role of the predicted CsrS ECD in LL-37 sensing by CsrS , we introduced point mutations into the chromosomal csrS locus of GAS strain 854 by allelic replacement . Four independent mutant strains were constructed in which one or all three negatively charged amino acids localized in a small cluster of acidic residues ( 148DHIED152 , Figure 3A ) were substituted with similar uncharged residues ( D148N , E151Q or D152N ) . Mutation of these three amino acids did not affect expression levels or surface localization of mutant CsrS , as similar quantities of CsrS were detected in western blots of membrane fractions obtained from the csrS triple point mutant strain 854csrSTM and wild type 854 ( Figure 2A ) , and similar amounts of CsrS were labeled by biotinylation on the mutant strain cell surface ( Figure 2B ) . The four resulting isogenic csrS mutants 854csrSD148N , 854csrSE151Q , 854csrSD152N , and 854csrSTM were tested for LL-37-mediated up-regulation of hasB , spyCEP , mac , and SPy0170 expression . Wild type strain 854 and each of the mutant strains were grown to early exponential phase in the presence or absence of 100 nM LL-37 , and gene expression was assessed by qRT-PCR . In contrast to wild type , the isogenic csrS triple mutant showed little or no change in gene expression in response to LL-37 ( Figure 3B ) . The csrS mutants with single amino acid substitutions ( D148N , E151Q or D152N ) all showed moderate LL-37-mediated up-regulation of the four target genes , but less than that observed in wild type ( Figure 3B ) . Mutation of this region of the CsrS ECD also abrogated or severely blunted the effect of Mg2+ to repress , or in the case of grab , to activate , CsrRS-regulated gene expression ( Figure S2 ) . The results above provide evidence that the mutated cluster of acidic residues in the predicted ECD is critical for LL-37 and Mg2+ signaling through CsrS in strain 854 . To confirm these findings and to test their generality for other GAS strains , we constructed an analogous csrS triple point mutant of M-type 49 strain NZ131 and examined its response to LL-37 by qRT-PCR . Similar to the results in the 854 background , we observed almost complete loss of LL-37-stimulated up-regulation of hasB , mac , and SPy0170 in NZ131csrSTM , and a marked reduction in spyCEP up-regulation ( Figure 3C , left panel ) . In wild type NZ131 , expression of these four genes was repressed during growth in 15 mM Mg2+ , but no such repression was observed for hasB , spyCEP , or Spy0170 in the NZ131csrSTM ( Figure 3C , right panel ) . Thus , similar findings in two independent strain backgrounds highlight the importance of a small cluster of negatively charged amino acids in the predicted CsrS ECD in LL-37 and Mg2+ signaling through CsrRS . During characterization of the csrS triple mutant , we noted that mutant bacteria formed compact , glossy colonies similar to wild type 854 , but distinctly different from the mucoid colony appearance of 854csrSΩ lacking CsrS . To verify that the distinctive colony morphology reflected a difference in capsule gene expression , we compared relative expression of hasB ( from the hyaluronic acid capsule biosynthetic operon ) in the three strains . As expected , in the absence of supplemental Mg2+ or LL-37 , expression of hasB was increased more than 50-fold in strain 854csrSΩ relative to that in wild type 854 , whereas hasB expression in the csrS triple point mutant was actually reduced by 40% compared to wild type ( Table 1 ) . This finding of reduced capsule gene expression suggested that the ECD mutations in the triple mutant resulted not only in refractoriness to regulation by LL-37 , but also in increased activity of the CsrR response regulator , presumably by enhancing its phosphorylation in the absence of signaling from an external ligand . Such an effect could result from increased autokinase activity of CsrS or reduced phosphatase activity of CsrS for phospho-CsrR . To test this hypothesis , we compared expression of additional CsrRS-regulated genes in the csrS triple mutant relative to wild type 854 . As observed for hasB , in the absence of supplemental Mg2+ or LL-37 , expression of spyCEP , mac , and SPy0170 was down-regulated in the triple mutant compared to wild type , whereas expression of each of these genes was significantly up-regulated in 854csrSΩ relative to wild type expression levels ( Table 1 ) . Furthermore , expression of grab was increased in the triple mutant relative to wild type levels ( data not shown ) . As grab is activated by supplemental Mg2+ and repressed by LL-37 ( Figure 1 ) , this result is also consistent with the proposed model of increased CsrR activity in the triple mutant . To test directly whether the altered ECD of the triple mutant changed gene regulation by affecting autokinase activity of CsrS , we inactivated the kinase by replacing the active site histidine residue with alanine ( H280A ) . As expected , when introduced in strain 854 , this mutation resulted in a mucoid colony morphology , and the mutant strain 854csrSH280A displayed marked up-regulation of CsrRS-repressed genes in a pattern very similar to that observed in 854csrSΩ ( Table 1 ) . Similarly , introduction of the H280A mutation into the CsrS triple mutant resulted in mucoid colonies and a comparable derepression of CsrRS-repressed genes as in 854csrSΩ and in 854csrSH280A ( Table 1 ) . Since mutation of the active site histidine of CsrS abrogated the suppressive effect of the ECD triple mutant , the most parsimonious model is that these mutations in the ECD affect gene expression by altering the autokinase activity of CsrS . While an effect on phosphatase activity is not excluded by these experiments , the results suggest strongly that the ECD triple point mutant expresses a constitutively active CsrS histidine kinase that is relatively refractory to signaling induced by external stimuli . CsrRS regulates the expression of several genes that encode products implicated in GAS resistance to opsonophagocytic killing and cytotoxicity: hasABC , slo , nga , spyCEP , sda1 , mac , and speB . Upregulation of antiphagocytic factors by host LL-37 is expected to enhance virulence in vivo; however , testing this hypothesis directly in an animal model is not possible , currently , since cathelicidins of other mammalian species do not share the CsrRS-signaling activity of human LL-37 [17] . Because in vitro resistance to phagocytic killing by human blood leukocytes correlates with GAS virulence in vivo [31] , we used an in vitro assay to assess the effect of LL-37 on phagocytic resistance as a proxy for effects on in vivo virulence . As would be predicted by the effects of LL-37 on regulation of antiphagocytic factors , exposure of four unrelated wild type GAS strains to LL-37 increased resistance of all four strains to phagocytic killing in vitro [17] . Inactivation of CsrR in the M-type 3 strain DLS003 also resulted in increased resistance to phagocytic killing by human peripheral blood leukocytes , consistent with the marked up-regulation of CsrRS-regulated antiphagocytic factors in the mutant strain [12] . Because deletion of CsrS results in a similar , although less marked , up-regulation of CsrRS-controlled genes , we expected that deletion of CsrS or inactivation of its histidine kinase activity would also lead to increased resistance to phagocytic killing . In vitro opsonophagocytic assays of 854csrSΩ and 854csrSH280A confirmed these predictions: both mutant strains were highly resistant to phagocytic killing by human blood leukocytes in vitro similar to a ΔcsrR mutant ( Figure 4 ) . In marked contrast , the csrS triple mutant was as susceptible to killing as wild type 854 in the absence of LL-37 , but did not show any increase in phagocytic resistance in response to LL-37 unlike wild type 854 ( Figure 4 ) . These observations further support the proposed model that the csrS triple mutant exhibits constitutive activation of CsrS autokinase activity and tonic phosphorylation of CsrR . An important consequence is down-regulation of CsrRS-controlled antiphagocytic factors and hyporesponsiveness to the stimulatory effect of LL-37 . The current working model for the CsrRS TCS is that of a classical sensor histidine kinase linked by phosphotransfer to a response regulator whose activity is controlled by its phosphorylation state . The experiments described above provide new evidence to support the surface location and stimulus-regulated histidine kinase activity of CsrS . We investigated also the role of CsrR in this model by further characterizing strain 950771 , a GAS M-type 3 strain that exhibits a high level of capsular polysaccharide production that does not increase upon exposure to LL-37 . Sequencing the csrRS locus in 950771 revealed a point mutation in csrR ( K102R ) in the highly conserved CsrR receiver domain [17] . The lysine residue that is mutated in 950771 is conserved not only in the csrR locus of all sequenced GAS strains , but also in response regulators of many TCS in a wide variety of bacterial species where it occupies a location near the conserved aspartic acid residue that is the site of phosphorylation [32] . In E . coli CheY , substitution of arginine for lysine at the corresponding site ( K109R ) did not prevent phosphorylation , but abrogated induction of tumbling motility that normally results from CheY phosphorylation , a finding interpreted to mean that the conserved lysine is required for phosphorylation to produce the active conformation of the response regulator [33] . To test whether K102 is required for GAS CsrR to transmit a signal from CsrS , we replaced R102 in strain 950771 with the consensus lysine residue ( R102K ) . Whereas isolate 950771 ( R102 ) showed no response to LL-37 or to Mg2+ , strain 950771csrRR102K restored both up-regulation of hasB , spyCEP , mac , and SPy0170 in response to LL-37 and down-regulation in response to 15 mM Mg2+ ( Figure 5A ) . In addition , correction of CsrR to the consensus K102 sequence markedly reduced expression of all four genes during growth in unsupplemented medium ( Figure 5B ) , a result that implies that K102 is necessary for transduction of the signal mediated by tonic phosphorylation of CsrR by CsrS under standard laboratory growth conditions . To further test whether the K102R CsrR mutation is sufficient to prevent CsrRS-mediated modulation of target gene expression , we also introduced the K102R mutation in wild type strain NZ131 . In contrast to wild type NZ131 that exhibited a 2- to 25-fold increase in expression of hasB , spyCEP , mac , and SPy0170 in response to 100 nM LL-37 , mutant strain NZ131csrRK102R showed no response ( Figure 5C , left panel ) . Moreover , the mutant failed to repress expression of these genes in response to 15 mM Mg2+ ( Figure 5C , right panel ) . Together , these results indicate that the conserved lysine residue at position 102 in CsrR is required for signal transduction from CsrS to modulate target gene expression in response to extracellular LL-37 or Mg2+ .
Previous studies have demonstrated that CsrRS regulates expression of more than 100 GAS genes including those encoding many important virulence determinants [9]-[11] . In addition , evidence has been presented that elevated levels of extracellular Mg2+ result in down-regulated expression of CsrR-repressed genes , whereas exposure of GAS to subinhibitory concentrations of LL-37 has the opposite effect [11] , [16] , [17] . Results of the present study demonstrate the same reciprocal pattern of regulation by LL-37 and Mg2+ for an expanded repertoire of GAS genes . While the predominant pattern of regulation is one of up-regulation of gene expression by LL-37 and down-regulation by Mg2+ , we report several instances of the opposite pattern , that is , repression of gene expression by LL-37 and activation by Mg2+ . The current investigation provides new experimental evidence that supports a model of CsrRS as a classical TCS that responds to these environmental signals through modulation of CsrS autokinase activity , with downstream signaling that depends on phosphotransfer from CsrS to the CsrR transcriptional regulator . A critical consequence of CsrRS signaling by LL-37 is the coordinated modulation of expression of multiple genes in a fashion that dramatically increases GAS resistance to killing by phagocytes , a bacterial phenotype that enhances virulence and promotes invasive infection in vivo . In addition to the previously demonstrated up-regulation by LL-37 of the hyaluronic acid capsule synthesis operon , mac/IdeS ( Mac/IgG protease ) , and spyCEP ( IL-8 protease ) , we found that LL-37 activated and Mg2+ repressed expression of genes encoding several important virulence factors including ska ( streptokinase ) , slo ( streptolysin O ) , and nga ( NAD-glycohydrolase ) , as well as speA ( pyrogenic exotoxin A ) and sda1 ( DNase ) , two virulence determinants encoded by prophages associated with the invasive M1T1 GAS clonal group [34] . The opposite pattern of regulation was observed for speB ( cysteine protease ) and grab ( protein G-related α2-macroglobulin-binding protein ) . Repressed expression of speB in response to LL-37 may also contribute to an invasive phenotype , as the speB-encoded cysteine protease has been proposed to degrade the anti-phagocytic M1 protein and to inactivate the sda1 gene product , a DNase that itself enhances GAS virulence by degrading neutrophil extracellular traps ( NETs ) [21] , [25] , [35] , [36] . Protein modeling of CsrS indicates the presence of two membrane-spanning regions that flank a domain predicted to form an extracellular loop that represents a potential site for interaction with environmental stimuli [12] , [16] . In cell-fractionation experiments , we found that CsrS is physically associated with the bacterial cell membrane , as predicted by this model . Furthermore , CsrS on intact bacterial cells was accessible to biotin-labeling , a result that implies that a domain of the protein lies in the extracellular space . We also investigated the role in CsrS-mediated signal transduction of a small cluster of negatively charged amino acids in the CsrS ECD . Because a similar cluster of acidic residues has been implicated in binding of cationic ligands to E . coli and S . typhimurium PhoQ , we tested in GAS the effect of substituting uncharged amino acids for these three residues . While our intent had been to disrupt binding of Mg2+ and/or LL-37 to the CsrS ECD , we discovered that this relatively small alteration in the ECD not only abrogated ligand signaling , but also resulted in a global effect on the CsrRS regulon consistent with tonic activation of CsrS autokinase activity . Support for this hypothesis also came from the observation that the effects of the csrS ECD mutations were overridden by mutating the active site histidine of the CsrS kinase domain , a result that implies that the effects of the former mutations on target gene regulation are mediated through CsrS kinase activity . Thus , in the absence of increased extracellular Mg2+ or exposure to LL-37 , expression of CsrR-repressed genes was reduced in the csrS triple mutant . The finding that neutralizing the charge of three amino acids in the CsrS ECD leads to an apparent activation of CsrS kinase activity and hyporesponsiveness to ligand signaling suggests that the mutations result in a conformational change in the cytoplasmic domain of CsrS that mimics that induced by binding of Mg2+ to the ECD . It is tempting to speculate that binding of Mg2+ and/or LL-37 to the same region of the ECD also modulates kinase activity by this mechanism , although attempts to demonstrate specific binding of either ligand to the ECD have , so far , been unsuccessful . The data summarized above suggest strongly that LL-37 signaling depends on direct interaction of the peptide and/or Mg2+ with the extracellular domain of CsrS . However , we cannot exclude an alternative signaling mechanism such as membrane disruption by LL-37 that secondarily results in altered CsrS autokinase activity . We found that deletion of CsrS or inactivation of its kinase activity produced a similar pattern of altered gene expression as deletion of CsrR , although the magnitude of change in gene expression was somewhat smaller for some genes . These observations imply that , under laboratory growth conditions , CsrS activates CsrR , presumably by phosphorylation , increasing its activity as a transcriptional regulator . Activation of CsrR by CsrS can be increased by exposure to elevated extracellular Mg2+ or reduced by exposure to LL-37 . The results discussed above support a model in which expression of the CsrRS regulon depends on the equilibrium between the phosphorylated and unphosphorylated states of CsrR . Increased extracellular Mg2+ or mutation of critical residues in the CsrS ECD increases CsrS phosphorylation and enhances phosphotransfer to CsrR , shifting the equilibrium toward phospho-CsrR with consequent repression of CsrR-repressed genes and activation of CsrR-activated genes . Conversely , exposure to LL-37 or deletion of CsrS shifts the equilibrium toward unphosphorylated CsrR , which is less active in regulating target promoters . Transduction of these modulating signals to altered transcriptional regulation depends also on the presence of a conserved lysine residue at position 102 in the receiver domain of CsrR . A natural mutant with a conservative arginine substitution at this position was refractory to signaling by extracellular Mg2+ or exposure to LL-37 and exhibited a pattern of gene expression similar to that of a CsrR deletion mutant . These phenotypes were confirmed by repairing the natural mutant to the consensus K102 and by introducing the K102R mutation into an unrelated wild type strain . On the basis of these findings and work by others on the role of the corresponding lysine residue in bacterial TCS , we conclude that CsrR K102 is critical to transducing the signal of CsrR phosphorylation and to modulation of CsrR-mediated transcriptional regulation at target promoter sequences . Several studies have documented the emergence of GAS strains with spontaneous inactivating mutations in CsrS or CsrR in the setting of invasive infection [18]–[20] . Because such mutants have a gene expression profile that results in a multifactorial enhancement of resistance to clearance by host phagocytes , these mutant variants have a strong selective advantage for survival in microenvironments such as the bloodstream or deep tissue sites where they are exposed to attack by host phagocytes . However , analysis of a collection of GAS pharyngeal isolates indicated that CsrRS mutants are distinctly rare in this setting , in marked contrast to isolates from patients with severe systemic infection [19] . The predominance of strains with a functional CsrRS system in the pharynx implies that CsrRS-mediated dynamic regulation of gene expression in response to environmental cues contributes to adaptation of GAS to its preferred environmental niche . During initial colonization , the low concentration of LL-37 on the resting pharyngeal epithelium is predicted to result in an intermediate level of CsrRS activation and a corresponding moderate expression of CsrRS-regulated virulence factors . This “colonizing” phenotype , however , can change quickly in response to increased local concentrations of LL-37 . The striking up-regulation of an antiphagocytic phenotype upon exposure to LL-37 enables the organism to maintain the capacity to arm itself against host effectors and thus resist clearance . The coordinated program of altered gene expression induced by LL-37 signaling can tip the balance of pathogen-host interaction from one of asymptomatic colonization to uncontrolled invasive infection . Paradoxically , secretion of LL-37 from injured epithelial cells or from degranulation of recruited neutrophils as part of the host innate immune response may trigger local or systemic invasion by GAS as a result of CsrRS-mediated virulence factor expression .
The human subjects aspects of this study were approved by the institutional review board of Children's Hospital Boston . Written informed consent was provided by study participants . Wild type GAS strains used in this study and isogenic mutants derived from them are described in Table 2 . GAS M-type 1 strain 854 is a clinical isolate from a patient with a retroperitoneal abscess [17] . GAS M-type 49 strain NZ131 is a skin isolate from a patient with glomerulonephritis [37] . GAS strain 950771 is an M-type 3 clinical isolate from a child with necrotizing fasciitis and sepsis [38] . GAS strains were grown at 37°C in Todd-Hewitt broth ( Difco ) supplemented with 0 . 5% yeast extract ( THY ) or on THY agar or trypticase-soy agar ( BD Bioscience ) supplemented with 5% defibrinated sheep blood . Escherichia coli ( E . coli ) strains DH5α ( New England Biolabs ) and StrataClone ( Stratagene ) were used for cloning . Recombinant protein overexpression for antisera production was carried out using E . coli strain BL21 ( DE3 ) ( Novagen ) . Antibiotics were used when necessary at the following concentrations: for GAS , erythromycin 1 µg/ml; for E . coli , erythromycin 200 µg/ml , kanamycin 50 µg/ml , carbenicillin 100 µg/ml , and ampicillin 100 µg/ml . GAS cultures were grown in THY broth supplemented with or without 100 nM LL-37 or 15 mM MgCl2 , and cells were harvested either at early exponential ( A600 nm 0 . 25 ) , mid-exponential ( A600 nm 0 . 5 ) , late exponential ( A600 nm 0 . 8 ) or early stationary ( A600 nm ∼1 ) growth phase . Total RNA extraction from bacterial cells was performed as described [11] during the growth phase at which target gene expression was maximal . RNA concentration and purity were determined using a NanoDrop spectrophotometer ND-1000 ( Thermo Fisher Scientific ) . Quantitative RT-PCR was performed on an ABI PRISM 7300 Real-Time PCR system ( Applied Biosystems ) using the QuantiTect SYBR Green RT-PCR kit ( Qiagen ) . Primers used are listed in Table S1 . Expression level of each target gene was normalized to recA ( spyM3_1800/SPy2116 ) and analyzed using the ΔΔCt method as described [11] . Replicate experiments were performed from at least three independent RNA preparations in triplicate . Statistical analysis was performed using the paired Student's t-test for expression level comparison under different growth conditions in a single strain and the unpaired t-test for testing differences between strains . The human cathelicidin LL-37 ( a gift of Robert I . Lehrer , UCLA , CA , USA ) was synthesized as described previously and its purity was confirmed by high-performance liquid chromatography and mass spectrometry [39] . To introduce single point and triple point mutations in the CsrS ECD region , vector pORIcsrS containing the wild type csrS sequence [11] served as template to amplify the entire plasmid by PCR with primer pair HTW 13/14 for csrS ( D148N ) substitution , HTW 15/16 for csrS ( E151Q ) substitution , HTW 17/18 for csrS ( D152N ) substitution , or csrS418-F ( muNHIQN ) /csrS480-R ( muNHIQN ) for csrS triple point substitution ( D148N , E151Q , D152N ) as described in the Quikchange site-directed mutagenesis protocol ( Stratagene ) . From the resulting plasmids , csrS fragments used for allelic replacement were PCR-amplified with Phusion high-fidelity DNA polymerase ( Finnzymes , New England Biolabs ) by using primers csrS-F ( PshAI ) and rt0245-R and cloned into vector pSC-B ( StrataClone blunt PCR cloning kit , Stratagene ) . To introduce a csrS H280A mutation into GAS strain 854 and into the isogenic triple mutant strain 854csrSTM , a csrS fragment was amplified by PCR from wild type 854 chromosomal DNA with primer pair 5005_149F/5005_1204R and cloned into pGEM-T ( Promega ) . The resulting plasmid was used for Quikchange site-directed mutagenesis to csrS H280A with primer pair H280A-F/H280A-R . For generating a csrR deletion mutant in strain 854 ( 854ΔcsrR ) , an overlap PCR using Phusion DNA polymerase was performed of the region upstream of csrR with primer pair Reg1P-F/HTW52 and the region downstream of csrR including about 500 bp of csrS with primer pair HTW 53/54 . The hybridized strands of the two resulting PCR products were used as a template for the second PCR amplifying a 1 kb fragment encompassing a CsrR deletion of amino acids 3–223 . The final product was ligated into vector pSC-B . Subsequently , the csrS or ΔcsrR fragments described above were released from pSC-B or pGEM-T by SalI/BamHI digestion and were subcloned into the temperature-sensitive shuttle vector pJRS233 [40] . To introduce the consensus CsrR ( csrR ( K102 ) ) into GAS strain 950771 and non-consensus CsrR ( csrR ( R102 ) ) into GAS strain NZ131 , csrR R102 was amplified from 950771 chromosomal DNA was amplified by using Platinum Taq high fidelity DNA polymerase ( Invitrogen ) and primers CsrP-F and csrS176-R . The PCR product was cloned into pGEM-T and then subcloned into pJRS233 using PstI and XbaI restriction sites to generate pJRS-csrR ( R102 ) . Resulting plasmid pJRS-csrR ( R102 ) was then used for Quikchange site-directed mutagenesis to convert csrR ( R102 ) to csrR ( K102 ) by using primer pair HTW 71/72 , creating plasmid pJRS-csrR ( K102 ) . All primers are described in Table S1 . Recombinant pJRS233 shuttle plasmids were electroporated into GAS strains 854 , 854csrSTM , NZ131 , or 950771 and then subjected to allelic gene replacement as described [38] . To confirm the genotype of mutant strains , csrR and csrS loci were PCR-amplified with Easy-A high fidelity DNA polymerase ( Stratagene ) from chromosomal DNA and the sequences confirmed by DNA sequencing ( DNA Sequencing Core , Brigham and Women's Hospital , Boston , MA , USA ) . Cultures of 854 , 854csrSΩ , and 854csrSTM were grown in THY at 37°C to an A600 nm of ∼0 . 4 , cells were collected ( 1250 × g , 8 min ) and washed once with 10 mM Tris-HCl , pH 8 . 0 , and resuspended in 360 µl hypotonic TEG buffer ( 10 mM Tris-HCl , 1 mM EDTA , 20% glucose , pH 8 . 0 ) supplemented with protease inhibitor cocktail III ( Calbiochem ) . For peptidoglycan degradation , mutanolysin ( ∼500 units , Sigma ) and lysozyme ( ∼17 , 700 units , Sigma ) were added and samples were shaken at 1000 rpm at 37°C for 1 h in an Eppendorf thermomixer . Cells were washed once in 500 µl TEG buffer and resuspended in 500 µl TE buffer ( 10 mM Tris-HCl , 5 mM EDTA , pH 8 . 0 ) supplemented with protease inhibitor cocktail ( Roche ) . Cells were lysed by ultrasonication ( 5×3 sec bursts on level 5 , Sonic Dismembrator model 60 , Fisher Scientific ) on ice followed by centrifugation ( Eppendorf 5417C 10 , 000 × g ) for 20 min at 4°C to remove cell debris . Membranes were separated from the cytoplasmic fraction by ultracentrifugation of supernatants ( Beckman Coulter Ultima , TLA-100 . 3 rotor ) for 1 h at 90 , 000 × g at 4°C . Membranes and cytoplasmic fractions were resuspended in SDS-PAGE sample buffer and heated to boiling . GAS strain 854 was grown in liquid culture to A600nm 0 . 7 ( late exponential phase ) or A600nm 1 . 2 ( stationary phase ) in the absence or presence of 100 nM LL-37 . Bacteria were removed by centrifugation ( 21 , 000 × g , 5 min ) . Cell-free supernatants were used for assays of DNase activity or mixed with sample buffer and heated to boiling before SDS-PAGE and western blot analysis . Samples were fractionated on 10% ( membrane and cytoplasmic fractions ) or 4–12% gradient ( supernatant proteins ) NuPAGE Novex Bis-Tris gels and then transferred to nitrocellulose membranes for western blotting as previously described [11] . Blots were incubated with specific rabbit antiserum against GAS CsrS ECD [11] , CsrR , SLO [41] , NADase [41] , or SpeB ( Toxin Technology , Sarasota , FL ) at a 1∶1000 dilution , or with mouse antiserum against GAS membrane protein OpuABC ( courtesy of Giuliano Bensi , Novartis Vaccines ) at a 1∶3000 dilution , each followed by horseradish-peroxidase-linked secondary antibody [11] . Signal development was carried out using the SuperSignal West Pico chemiluminescence substrate ( Thermo Scientific Pierce ) . For surface biotinylation the Cell Surface Protein Isolation Kit ( Pierce ) was used according to the manufacturer's protocol with the following modifications . GAS were grown in 30 ml THY at 37°C to A600 nm ∼ 0 . 3 , cells were harvested ( Centra CL3 , Thermo IEC , 1250 × g , 8 min ) , washed twice in 1 . 5 ml PBS ( Eppendorf 5417C , 9800 × g , 1 min ) , and resuspended in 1 . 5 ml Sulfo-NHS-SS-Biotin labeling solution . The biotinylation is reversible by cleavage of the disulfide bond in Sulfo-NHS-SS-Biotin . As a negative control , wild type 854 cells were incubated with PBS instead of biotin labeling solution . After 30 min agitation at 4°C , 100 µl quenching solution was added to treated cells and the cells were centrifuged at 6800 × g for 4 min . Cells were washed twice in 1 . 5 ml TBS ( 25 mM Tris-HCl , 0 . 15 M NaCl , pH 7 . 2 ) and frozen at −20°C . Frozen cells were resuspended in 250 µl lysis buffer supplemented with 2 . 5 µl protease inhibitor cocktail III ( Calbiochem ) and lysed by two rounds of ultrasonication ( 5×1 s bursts , level 1 ) with an incubation on ice in between . Lysates were centrifuged at 20 , 800 × g for 4 min at 4°C . Supernatants were incubated with 250 µl immobilized NeutrAvidin resin for 60 min in spin-columns with end-over-end rotation . Flow-through samples were retained and mixed with SDS-PAGE sample buffer . Resin was washed four times with 500 µl wash buffer supplemented with protease inhibitor and incubated with 200 µl SDS-PAGE sample buffer with 50 mM DTT for 60 min with end-over-end rotation at RT . Eluates were collected by brief centrifugation of uncapped spin-columns . Samples were heated at 95°C for 5 min and stored at −20°C until needed for western blot analysis . A 1∶3 mixture of antiserum to CsrS ECD and antiserum to N-terminal truncated CsrS was used to detect CsrS protein on the blots . Full-length CsrR and N-terminal truncated CsrS ( CsrSΔ1-231 ) and were fused separately to a N-terminal His6 tag by cloning into overexpression vector pET-28a ( Novagen ) PCR-amplified DNA fragments obtained with primer pairs JL-48/JL-49 and HTW 37/46 , respectively . Following overexpression by IPTG induction , recombinant proteins were affinity purified using Ni2+-NTA resin ( Qiagen ) under native conditions ( His6-CsrR ) or under denaturing conditions ( His6-CsrSΔ1-231 ) according to the manufacturer's protocol . Purified proteins were used to immunize rabbits ( LAMPIRE Biological Laboratories , Inc . , Pipersville , Pennsylvania , USA ) . Reactivity of immune sera against CsrS or CsrR was evaluated by western blotting of GAS lysates . DNase activity in GAS culture supernatants was assayed as described by Aziz et al . with modifications [42] . Supernatants were diluted 1∶125 in sterile deionized water , and 10 µL samples were mixed with 1 µg plasmid DNA in 100 mM Tris , pH 7 . 5 , supplemented with 1 mM CaCl2 and 1 mM MgCl2 in a 15 µL total reaction volume . Samples were incubated at 37°C for 20 min and then were stopped by the addition of 20 mM EDTA . Samples were analyzed on 1% agarose gels and DNA was visualized with SYBR Safe DNA stain ( Invitrogen ) . GAS resistance to phagocytic killing was evaluated by an in vitro assay as described [43] . In brief , GAS strains grown to early exponential phase with or without 100 nM LL-37 were mixed with freshly isolated human peripheral blood leukocytes at a multiplicity of infection of 3 – 4 in the presence of 10% human serum as complement source . Aliquots were withdrawn for quantitative culture immediately after mixing and after 1 h end-over-end rotation at 37°C . Results were reported on a log scale as the fold-change in cfu defined as the total cfu after incubation divided by the total starting cfu . Statistical significance of differences in the capacity of GAS strains to resist opsonophagocytic killing were evaluated by one-way ANOVA with Bonferroni's post-test analysis .
|
Group A Streptococcus ( S . pyogenes or GAS ) is exclusively a human pathogen that can inhabit the human throat as a harmless commensal , cause localized , self-limited infection in the form of pharyngitis or strep throat , or invade local tissues or the bloodstream to produce life-threatening disease states such as necrotizing fasciitis or streptococcal toxic shock . We present evidence that the GAS CsrRS ( or CovRS ) two component system governs the transition from a colonizing to an invasive phenotype by transducing a specific signal from the antimicrobial peptide LL-37 that is secreted as part of the human innate immune response to GAS infection . We show that LL-37 signaling requires specific domains of both the CsrS sensor kinase and the CsrR response regulator , and that signaling results in a coordinated and marked increase in expression of multiple bacterial factors that confer resistance to phagocytic killing , a hallmark of GAS virulence .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"bacteriology",
"emerging",
"infectious",
"diseases",
"immunity",
"medical",
"microbiology",
"innate",
"immunity",
"immunology",
"biology",
"microbiology",
"host-pathogen",
"interaction",
"immune",
"response",
"pathogenesis"
] |
2011
|
Signal Transduction through CsrRS Confers an Invasive Phenotype in Group A Streptococcus
|
Biologists and environmental scientists now routinely solve computational problems that were unimaginable a generation ago . Examples include processing geospatial data , analyzing -omics data , and running large-scale simulations . Conventional desktop computing cannot handle these tasks when they are large , and high-performance computing is not always available nor the most appropriate solution for all computationally intense problems . High-throughput computing ( HTC ) is one method for handling computationally intense research . In contrast to high-performance computing , which uses a single "supercomputer , " HTC can distribute tasks over many computers ( e . g . , idle desktop computers , dedicated servers , or cloud-based resources ) . HTC facilities exist at many academic and government institutes and are relatively easy to create from commodity hardware . Additionally , consortia such as Open Science Grid facilitate HTC , and commercial entities sell cloud-based solutions for researchers who lack HTC at their institution . We provide an introduction to HTC for biologists and environmental scientists . Our examples from biology and the environmental sciences use HTCondor , an open source HTC system .
The life and environmental sciences are becoming computationally intense disciplines [1] . Individual studies can generate gigabytes and even petabytes of data [2 , 3] . Examples include -omics data [2 , 4] , high-resolution imagery [5] , lidar [6] , and real-time monitoring data [3] . Additionally , simulation has emerged as a third branch of science to complement empirical and theoretical approaches [7] . Examples include climate projections and assessments [8 , 9] , responses to climate change [10 , 11] , and hydrological models [12 , 13] . Last , scientists may need to evaluate models using computationally intensive Monte Carlo methods [14 , 15] . Scientists have traditionally processed "big" data sets and run large-scale simulations using either desktop computing ( and long wait times ) or high-performance computing ( HPC; with higher overhead costs in a central location ) . These "supercomputers" use parallel computing to locally run a simulation or process data on a single , large computer . Historically , only researchers at large institutes , national laboratories , and large universities had access to HPC systems and the funding to restructure code to run on such a computer . Many problems can be broken down into smaller , loosely coupled jobs ( defined in Box 1 ) and distributed across computer pools . Examples from the environmental sciences can include simulating study designs to compare model performance and study performance [14] , model calibrations [16] , and processing telemetry data [17] . This alternative paradigm is called high-throughput computing ( HTC ) [18–20] . Originally designed as a method to "scavenge" unused computational resources ( e . g . , idle desktop computers ) in the 1980s , HTC has emerged as a method for distributed computing that can be used with dedicated servers . HTC-based approaches require users to break a large problem into many small , independent jobs that are distributed across a pool by the HTC software . Computers then process the task in many small parts . Furthermore , this allows both the task from a specific user as well as the tasks from multiple users to be distributed and processed in parallel . HTC software automatically manages the individual jobs and is designed to be robust . Jobs may be stopped ( e . g . , a desktop computer is no longer idle because the owner has returned to work ) or lost ( e . g . , a connection is lost and a job needs to be restarted ) . This distributed computing paradigm has led to HTC being used to run other large tasks that would be too big even for one supercomputer . Currently , many different industries and scientific fields use HTC . One specific implementation of HTC is HTCondor . HTCondor , originally called Condor , coordinates computing resources on either dedicated machines or idle machines such as desktop computers [18] . When first developed in the 1980s , HTCondor could submit and manage thousands of jobs . Currently , HTCondor can handle more than 200 , 000 jobs [21] . Many different organizations use HTCondor: the Open Science Grid uses HTCondor to pool computational resources across institutions [22]; the European Organization for Nuclear Research ( commonly known by its French acrynomn CERN ) uses HTCondor to process high-energy physics data [23] , such as finding the "God particle" [24]; Disney and DreamWorks use HTCondor to create movies ( http://research . cs . wisc . edu/htcondor/images/WildLetter . pdf ) ; the National Aeronautics and Space Administration ( NASA ) uses HTCondor to process telescope , atmospheric , and space data as well as run simulations [25]; the National Oceanic and Atmospheric Administration ( NOAA ) uses HTCondor to process weather data[26]; the Search for Extraterrestrial Intelligence ( SETI ) uses HTCondor to search for extraterrestrial life ( http://research . cs . wisc . edu/htcondor/manual/v7 . 8/3_12Setting_Up . html ) ; and the United States Geological Survey ( USGS ) uses HTCondor for groundwater modeling [16] . Although HTC is currently used by some environmental sciences ( e . g . , genomics [20] ) , we see the opportunity for expanded use of HTC in the environmental sciences based upon our own experiences working with researchers and resource managers . HTC could help these scientists with their current computationally intensive computing tasks . Hence , we provide an introduction to HTC for these scientists .
HTCondor "has enabled ordinary users to do extraordinary computing" [19] . Broadly , HTCondor is a distributed-batch computing system , and our description of HTCondor applies to other HTC systems as well . Users submit a job to HTCondor . HTCondor chooses when and where to run the job based upon the job requirement and available worker machines . HTCondor monitors the jobs' progress , and HTCondor notifies the user upon completion . Practically , this means users submit their jobs from a submit machine ( see Box 1 for definitions ) . A job is typically a batch script on Windows or a shell script on Linux/Unix/macOS . The batch/shell script calls other programs in the workflow . A key aspect of HTCondor is that it requires no custom programming or linking to run in the cluster . This eases debugging and increases confidence of correctness , for the same job can usually be run interactively on a local machine . The submit machine submits jobs to different node machines . These node machines are computers with HTCondor installed . Node machines can be either "normal" computers ( e . g . , the administrative assistant's computer ) or dedicated computers/servers that only HTCondor uses . Nodes have ClassAds that are like "classified advertisements" in a newspaper and communicate a machine's resources ( e . g . , operating system , available memory , or central processing unit [CPUs] ) . ClassAds are a language developed by the HTCondor project . End users will not need to change ClassAds unless they are configuring machines in the HTCondor pool . Users send submit files to the submit machine , which matches jobs to machines . Once jobs are done , HTCondor gathers up and returns output files to the user . Ideally , a user will not need to setup their own HTCondor pool . Many universities , government agencies , and national laboratories offer HTC facilities . Additionally , many countries have science grids available that use HTCondor ( e . g . , Open Science Grid in the United States [22]; GridPP in the United Kingdom [https://www . gridpp . ac . uk/] ) . Commercial vendors also offer HTCondor ( e . g . , Amazon Web Service ) . Last , HTCondor is free and open source , and anybody may setup their own local pool . We have included documentation outlining our installation process ( AR1 ) , but otherwise , assume users have access to their own pool .
The HTCondor Manual provides a Quick Start Example ( https://research . cs . wisc . edu/htcondor/manual/quickstart . html ) , also in AR1 . This demonstrates HTCondor by submitting a job ( sleep . bat for Windows; sleep . sh for Linux ) that tells the computer to "sleep" for 6 s . The submit file ( sleep . sub ) instructs HTCondor where the job is located , what HTCondor should do with the executable , and what to do with the results from the executable . In this example , the "results" include a log file , an errors file , and a message file . The job is submitted using the command condor_submit sleep . sub . The job may be monitored with the condor_q command . When the job is finished running , HTCondor will update the log file and no longer show the job when condor_q is entered . This section describes our five steps for using HTCondor , which also apply to other HTC systems . Step 1: Define the computing problem . Where are the bottlenecks and what can be broken down into small tasks ? The answer may not be HTC . HPC or even code optimization or language changes may be an appropriate solution ( e . g . , using C++ rather than R or Python ) . An important consideration is if a problem is processor limited or memory limited [28] . Memory limited means that a program runs out of random access memory ( RAM ) memory , whereas processor limited means that calculations done by a processor constrains the program . A tradeoff exists between these two constraints . Additionally , large computing jobs can be network-intensive tasks and limited by data transfers . For these tasks , HTC may not be an ideal solution or may require special consideration to avoid network limitations . Also , some computing tasks may be able to run in parallel but cannot be tightly coupled . These jobs are ill suited for HTC and require HPC . Step 2: Discretize the problem into smaller jobs that may be run on HTCondor . These small discrete problems should be "pleasantly" parallel ( also known as "embarrassingly" parallel ) and run independently of one another . This involves defining all input files for each job and ensuring output files do not overwrite output files from other jobs . For simulations , this involves deciding how to run simulations ( e . g . , should one simulation be a job or multiple simulations that share the same inputs ? ) . Simulations are often processor limited . For data processing , breaking a task down may involve splitting data into smaller files . Data processing can often be memory limited but may be processor limited if it involves many calculations . Step 3: Preprocess the inputs . For large data , this may mean breaking the data up into smaller files and having a way to identify files by some naming convention . For large simulations , for instance , this may require creating an input table of parameters in which each row corresponds to a job . Step 4: Run the jobs on HTCondor . This step requires placement of a submit file and corresponding files on the central manager . Depending upon the size of the job being run , the processing might take a few minutes to run or possibly take weeks or even months . Step 5: Postprocess the data . This step consists of combining the output from the previously run HTCondor jobs . Often , this step is most easily done using a scripting language such as Python or R . Pre- and postprocessing can often be done on personal computers before or after the data have been moved to the HTCondor submit machine . In some cases , both the pre- and postprocessing steps might be done using their own HTCondor submit files for large projects ( i . e . , breaking files or jobs down into smaller jobs to run on many machines ) . DAGMan is a program that is included with HTCondor capable of submitting multiple HTCondor batches ( conceptually , it is a program to run and manage HTCondor submit files ) and is documented in the HTCondor Manual [27] .
An important and sometimes difficult step with HTCondor is "sandboxing" code . Sandboxing is necessary so that all programs will run independently and predictably across platforms . The easiest method for using HTCondor avoids sandboxing by having the same version of programs installed on local machines . For example , simple scripts that only use base R or Python might not require sandboxing . There are two downsides to this approach . First , the software must be installed on the local machines . Second , machines might not have the same version installed ( e . g . , Python 2 . x verses 3 . x ) . On Windows , sandboxing can sometimes be easier because Windows executables are often independent . Thus , sandboxing the code only requires using HTCondor to send all required programs ( e . g . , Python . exe ) with the script . All required dependencies are then locally installed with the program and deleted when HTCondor is finished running . We include examples of Sandboxing R script ( AR1 ) . On Linux/Unix systems ( including macOS ) , multiple options exist . In theory , programs could be sandboxed like Windows . However , Unix software architecture is built upon many small building blocks that are combined , creating intricate dependencies [29] . This can be side stepped with container programs such as Docker . Docker has the added bonus of having an HTCondor "Universe" setting that allows it to integrate well with HTCondor . Docker works by building images ( i . e . , a collection of software that makes programs able to run ) similar to a virtual box but with a smaller footprint ( i . e . , size , memory , and processor use ) and takes advantage of the shared kernel across Linux platforms . An image is based upon a specified version of Linux ( e . g , . Ubuntu 16 . 04 ) and has specific software installed in it ( e . g . , R 3 . 0 . 1 with ggplot2 2 . 0 . 1 ) . Users may build their own image either from a Dockerfile ( a script that tells Docker how to build an image ) or using the command line and an existing Docker image . Many images have already been developed for commonly used programs such as Python and R , and these can often meet the needs of HTCondor users or , at the very least , serve as starting places for building your own image . We prepared a brief tutorial on Docker Images ( AR1 ) and additional documentation may be found on the Docker homepage ( https://docs . docker . com/engine/tutorials/dockerimages/ ) .
We have prepared examples of how to use HTCondor for scientific computing ( AR1 ) . These examples include Quick Start ( "hello world"-like ) examples , R examples , Python examples , and Docker examples . The repository also includes tutorials and a suggested course of study for teaching oneself HTCondor . Additionally , we regularly use ( and document ) HTCondor . For example , we have simulated data sets to see how statistical trends can be recovered ( AR3 ) . This code demonstrates how to use HTCondor on Windows , including sandboxing code . We have also simulated study designs and examined how well parameters could be estimated ( AR2 ) . This example demonstrates how to use HTCondor in Linux , including Docker for sandboxing code . Briefly , we provide a walkthrough of the statistical trends example ( AR3 ) . First , we defined our computer problem as the amount of time necessary to fit models across a wide parameter space . Second , we discretized our problem by parameter combinations ( i . e . , each HTCondor "job" corresponded to a set of parameter combinations ) . Third , we presimulated all data sets for each parameter combination and also recorded these combinations in a comma-seperated values ( CSV ) file . Fourth , we used a submit file to run our jobs . Fifth , we used R to combine , summarize , and plot our results .
The previous examples all worked well with HTC because they could be discretized into smaller jobs . When jobs cannot be readily broken down , HTC may not always be the solution . However , HTCondor allows for jobs to request multiple CPUs and HTCondor can be used to manage multiple HPC jobs across HPC resources . As a concrete example , running a computationally intense individual-based model might require HPC resources because the model is closely connected and is not pleasantly parallel . But running multiple instances of these models could be managed through HTCondor . Last , we have observed cases in which simply optimizing code removed our need for HTC or HPC .
https://my . usgs . gov/bitbucket/projects/CDI/repos/hunting_invasive_species_with_htcondor/ This Git repository contains an example of how to use HTCondor to simulate and recover a study design using R , RStan , and Docker: https://my . usgs . gov/bitbucket/users/rerickson_usgs . gov/repos/edna-sampling-design/browse . Git repository contains an example of how to use HTCondor to compare trend estimation R , MARSS , and Docker: https://my . usgs . gov/bitbucket/projects/UMESC/repos/ltrm-trend_comparison/browse .
Biology and the environmental sciences have become computationally intense . HTC is an important framework for dealing with computationally intense problems . We provided an introduction to HTC and an overview of using HTCondor . Our specific five steps were 1 ) define the computational problem and confirm that HTC is the solution , 2 ) determine how a problem can be broken down into small discrete jobs and can be identified by a unique number , 3 ) preprocess inputs into small jobs that can be distributed , 4 ) run the job using HTC ( e . g . , HTCondor ) , and 5 ) postprocess the data and combine the results from the many distributed jobs .
|
Computational biology often requires processing large amounts of data , running many simulations , or other computationally intensive tasks . In this hybrid primer/tutorial , we describe how high-throughput computing ( HTC ) can be used to solve these problems . First , we present an overview of high-throughput computing . Second , we describe how to break jobs down so that they can run with HTC . Third , we describe how to use HTCondor software as a method for HTC . Fourth , we describe how HTCondor may be applied to other situations and a series of online tutorials .
|
[
"Abstract",
"Introduction",
"HTCondor",
"Materials",
"and",
"methods",
"Sandbox",
"code",
"Examples",
"When",
"not",
"to",
"use",
"HTC",
"Additional",
"resources",
"Conclusion"
] |
[
"employment",
"ecology",
"and",
"environmental",
"sciences",
"education",
"social",
"sciences",
"computers",
"neuroscience",
"learning",
"and",
"memory",
"scientists",
"cognition",
"memory",
"information",
"technology",
"data",
"processing",
"science",
"and",
"technology",
"workforce",
"computer",
"and",
"information",
"sciences",
"labor",
"economics",
"computing",
"methods",
"jobs",
"economics",
"people",
"and",
"places",
"professions",
"science",
"policy",
"careers",
"in",
"research",
"computer",
"software",
"population",
"groupings",
"biology",
"and",
"life",
"sciences",
"cognitive",
"science"
] |
2018
|
Wrangling distributed computing for high-throughput environmental science: An introduction to HTCondor
|
Little is known about the strains of Trypanosoma cruzi circulating in Central America and specifically in the most important vector in this region , Triatoma dimidiata . Approximately six million people are infected with T . cruzi , the causative agent of Chagas disease , which has the greatest negative economic impact and is responsible for ~12 , 000 deaths annually in Latin America . By international consensus , strains of T . cruzi are divided into six monophyletic clades called discrete typing units ( DTUs TcI-VI ) and a seventh DTU first identified in bats called TcBat . TcI shows the greatest geographic range and diversity . Identifying strains present and diversity within these strains is important as different strains and their genotypes may cause different pathologies and may circulate in different localities and transmission cycles , thus impacting control efforts , treatment and vaccine development . To determine parasite strains present in T . dimidiata across its geographic range from Mexico to Colombia , we isolated abdominal DNA from T . dimidiata and determined which specimens were infected with T . cruzi by PCR . Strains from infected insects were determined by comparing the sequence of the 18S rDNA and the spliced-leader intergenic region to typed strains in GenBank . Two DTUs were found: 94% of infected T . dimidiata contained TcI and 6% contained TcIV . TcI exhibited high genetic diversity . Geographic structure of TcI haplotypes was evident by Principal Component and Median-Joining Network analyses as well as a significant result in the Mantel test , indicating isolation by distance . There was little evidence of association with TcI haplotypes and host/vector or ecotope . This study provides new information about the strains circulating in the most important Chagas vector in Central America and reveals considerable variability within TcI as well as geographic structuring at this large geographic scale . The lack of association with particular vectors/hosts or ecotopes suggests the parasites are moving among vectors/hosts and ecotopes therefore a comprehensive approach , such as the Ecohealth approach that makes houses refractory to the vectors will be needed to successfully halt transmission of Chagas disease .
Trypanosoma cruzi is the causative agent of Chagas disease and infects approximately six million people in Latin America [1] as well as many other mammals . Although initially considered largely clonal with only rare genetic mixing during its evolution [2] , studies of the population structure and increasingly detailed studies of the nuclear and mitochondrial DNA of T . cruzi clones have suggested that this clonal propagation is overlaid with more frequent and recent hybridization and genetic exchange events than was previously appreciated [3] . Six discrete-typing units ( DTU TcI-VI ) are currently recognized , adopted by international consensus [4] , including at least two hybrid lineages ( TcV and TcVI ) , and one additional lineage found mostly in bats ( TcBat ) [5 , 6] , which is closely related to TcI . Among these DTUs , TcI is the most widespread and diverse lineage , with the smallest genome and the least amount of aneuploidy so is likely a parent of some hybrid lineages [4] . Knowledge of the T . cruzi strains present will be useful for understanding the epidemiology , and for treatment and control as different strains are roughly associated with different geographic locations and ecotopes , and hosts and vectors ( reviewed in [7] ) . Association of particular strains with the diverse disease spectrum observed is an area of intense investigation [7 , 8] . For example , the megasyndrome form of Chagas disease is found mostly in Southern Cone countries where TcV and TcVI are found in humans , whereas cardiomyopathy , rather than megasyndrome , is common in Central and North America where TcI is associated strongly with human infection [7] . Although direct association of particular disease spectrum and specific T . cruzi DTU remains elusive [7] , some recent studies are beginning to dissect the mechanisms and show that at least some of different disease spectra are likely due to different T . cruzi strains [8] . In addition , the particular strains involved in human infections should also be a consideration for treatment efficacy studies and drug design [9] . TcI is found from the southern U . S . to southern South America and was first associated with sylvan cycles ( marsupials and rodents ) in South America ( reviewed in [4] ) . Later studies showed that TcI is the most common strain identified in northern South America [10 , 11] , as well as Central and North America and is frequently associated with human disease from the Brazilian Amazon basin northwards [11–18] . TcI is considered more diverse ( and therefore originating ) in South America compared to North and Central America [19] . However , the higher diversity in the south as compared to the north could reflect the relative geographic range surveyed and/or the sampling effort as little T . cruzi strain typing has been reported other than South America [20] . In addition , although most strains found in humans are reflective of the strains found in nearby hosts and vectors [7] , there is an intriguing divergent , and fairly homogenous TcI subgroup associated with human infections ( now called TcIDOM ) [19 , 21 , 22] . Identified to date largely in South America and surprisingly distinct from strains found in nearby hosts or vectors; TcIDOM clusters with North and Central America strains by phylogenetic inference [19] . A recent inventory notes that of the DTUs published for T . cruzi isolates , 90 . 7% are from South America; little is known about strains present in Central and North America [20] . Even less is known about T . cruzi DTUs present in Triatoma dimidiata , the principal Chagas vector in Central America and a secondary vector in Mexico and northern South America . Of limited studies from this geographic region , most report the predominance of TcI and less frequent presence of TcIV [12 , 14 , 15 , 17 , 18 , 23–25] . A broader range of DTUs has also been reported in Mexico [26 , 27] . The purpose of this study was to determine the strains of T . cruzi present in T . dimidiata , across its broad geographic range from southern Mexico to northern South America [28] , by comparing the sequence of two nuclear markers: 18S rDNA and the spliced leader , also known as mini-exon , intergenic region ( SL-IR ) to that of strains of known DTU . In addition , we explore how adding the largest sample of Central American T . cruzi strains to date informs the diversity present within TcI . We investigate clustering of TcI haplotypes by geography , host and ecotope using Principle Component and Median-Joining Network analyses . This study provides new information describing the diversity of T . cruzi circulating in T . dimidiata from Mexico to Colombia , and relates this TcI diversity to that found elsewhere in the Americas .
A total of 334 adult T . dimidiata were collected from 19 sites in eight countries across the geographic range of the species , from Mexico to Ecuador , by professionals trained in safe handling of biohazardous materials ( S1 Table ) . Specimens were collected by the person/hr method inside homes ( domestic ) , in areas surrounding homes ( peridomestic ) or in sylvan areas . Specimens were stored at -20°C in a 95% ethanol / 5% glycerol until DNA was isolated . DNA from reference strains was kindly provided by Drs . Christian Barnabé and Frédérique Brenière ( IRD , France ) . A TcI reference strain , Silvio X10 was purchased from ATCC ( Manasses , VA ) . DNA was isolated from the distal two abdominal segments of T . dimidiata specimens exactly as specified in the DNeasy Blood and Tissue kit ( Qiagen , Inc . , Valencia , CA ) . Infection of the T . dimidiata specimens with T . cruzi was determined by PCR ( AmpliTaq DNA polymerase , Life Technologies , Grand Island , NY ) using the TCZ1 and TCZ2 primers [29] and these cycling conditions: an initial denaturation at 94°C for 10 min; 30 cycles of 94°C for 20 sec , cooling to 57°C for 10 sec , and heating to 72°C for 30 sec , and final extension for 7 min at 72°C . Amplified products were electrophoresed alongside a positive control of amplified T . cruzi DNA and a negative control of water on a 1 . 8% agarose gel containing DNA SafeStain ( Lambda Biotech , Inc . , St . Louis , MO ) and visualized on a UV transilluminator ( Bio Rad , Hercules , CA ) . The PCR was repeated if the controls did not give the expected results . T . cruzi strains present in T . cruzi-positive T . dimidiata specimens were determined by amplification and sequencing of 18S rDNA and the SL-IR; these two nuclear genomic regions can distinguish all six T . cruzi DTUs [4] . 18S rDNA was amplified using V1 and V2 primers [30] and these cycling conditions: initial denaturation at 94°C 2 min , followed by 30 cycles of: 94°C 1 min , 54°C 1 min , 72°C 1 min , and a final step of 72°C for 5 min . SL-IR was amplified using Tc , Tc1 , and Tc2 primers together [30] and these cycling conditions: initial denaturation 94°C 2 min , followed by 27 cycles of 94°C 30 sec , 55°C 30 sec , 72°C 30 sec , and a final step at 72°C 5 min . PCR products were electrophoresed on a 2% MetaPhor Agarose ( Cambrex Bio Science Rockland , Inc . , Rockland , ME , USA ) and visualized by transillumination . Eighty-two T . dimidiata abdominal DNA samples amplified at the expected band size for their respective marker [4] and were sequenced ( Beckman Coulter Genomics , Danvers , MA , USA ) . 18S sequence was determined for 44 specimens by direct sequencing or sequencing following cloning if overlapping peaks were observed in the chromatogram ( 12 specimens , p-GEM-T easy vector system , Promega , Madison , WI , USA , Table 1 ) . SL-IR sequence was determined for 22 specimens by direct sequencing . Sequence for both markers was obtained for 15 specimens , which allowed us to determine the T . cruzi strain in 51 individual specimens . T . cruzi strains were unambiguously determined for each sequence based on ≥97% query coverage and ≥98% identity to a published strain ( DTU ) in a Blast query ( http://blast . ncbi . nlm . nih . gov/Blast . cgi ) ) . Diversity measures ( S , h Hd , π , and Tajima’s D ) of 18S rDNA and SL-IR haplotypes were calculated in DNAsp ( v . 5 . 10 . 01 ) [31] . To test for associations of TcI haplotypes among geographic regions , ecotopes and host/vector associations , haplotypes were analyzed by Principle Component Analysis ( PCA ) and Median-Joining Network analysis . These analyses used our sequences ( Table 1 ) combined with those available on GenBank that were of sufficient length and had no ambiguous nucleotides ( S2 and S3 Tables ) . The GenBank sequences were a mix of isolates , molecular clones , and cellular clones obtained originally from humans , wild mammals , and triatomine vectors . “Rodent” includes both R . rattus and other rodent species . The majority of the 18S TcI sequences were from Colombia and Brazil , with one each from Venezuela and Panama . SL-IR sequences were from Argentina , Brazil , Bolivia , Chile , Colombia , French Guyana , Mexico , Panama , Paraguay , Venezuela , and the USA . Since Colombia was so heavily sampled we chose to identify those sequences separately from the other broad geographic regions . Sequences were aligned in MacVector ( v . 14 . 5 . 3 , Apex , North Carolina ) using Muscle [32] . The 18S TcI alignment contained 116 sequences including 27 sequences ( this study ) , 87 from GenBank , and two outgroups: TcII ( reference strain IVVcl4 [33] , sequenced in our lab ) , and TcBat ( JQ965548 ) . The total 18S rDNA alignment including gaps was 179 bp , and excluding gaps sequences ranged from 155 to 178 base pairs . For SL-IR , only the single nucleotide polymorphism ( SNP ) region was used for network analysis because studies have shown that alignments of the microsatellite region are ambiguous [34 , 35] . The alignment of the 185 SL-IR TcI sequences included 16 sequences ( this study ) , 171 from GenBank , and one TcBat outgroup ( TCC 203 , KT305859 . 1 ) The total SL-IR alignment including gaps was 231 bp with sequences ranging from 221–223 bp , excluding gaps . For the two markers , 18S and SL-IR , genetic differences among ecotopes and geographic regions were visualized using Principle Components Analysis in GenAlEx ver 6 . 502 [36] to plot genetic relationships among individuals and identify those from the same geographic region or ecotope . Individuals close to each other in the graph are more closely related to each other than to distant individuals . Nominal logistic regression was used to test for differences among groups based on the principle components ( JMP Pro , Version 12 . 0 . SAS Institute Inc . , Cary , NC ) . For the two markers , 18S and SL-IR , phylogenetic relationships among individuals were visualized using Median-Joining Networks and color-coded to identify those from the same geographic region , host/vector or ecotope . Median-Joining network analysis is preferred to phylogenetic inference for intraspecific analyses [37] . The sequences were analyzed using Network , DNA alignment and Network Publisher ( fluxus-engineering . com , version 5 . 0 . 0 . 0 ) . Median-Joining ( MJ ) Networks [38] were calculated and the post-processing maximum parsimony cleanup procedure [39] performed for both genes . Networks were then arranged by hand and nodes colored using Network Publisher . In addition , isolation-by-distance was tested ( GenAlEx ver . 6 . 5 , [36 , 40] ) using the number of differences between sequences as a measure of distance .
TcI was the predominant DTU found in T . dimidiata based on the DNA sequence of two markers , 18S rDNA and the SL-IR . Thirty-eight percent ( 126/334 ) of the T . dimidata specimens examined were infected with T . cruzi , showing the T . cruzi-specific band by PCR . We were able to strain type 51 of the infected T . dimidiata and TcI was present in 94% ( 48 ) of these specimens based on ≥97% query coverage and ≥98% identity to T . cruzi specimens identified as TcI . Eighty-six percent ( 44/51 ) were determined based on 18S rDNA sequence , 43% ( 22/51 ) by the SL-IR sequence , and 29% ( 15/51 ) by both sequences ( Table 1 and Fig 1 , GenBank accession numbers: MF099414-MF099427 ) . Where sequence was available from both markers , strain identifications were concordant . TcIV was found at a much lower prevalence , 6% ( 3/51 ) of the T . dimidiata tested , all determined by 18S rDNA sequence ( Table 1 and Fig 1 ) . The few TcIV identified were all found in the northern end of the range of T . dimidiata , in Yucatan , Mexico and Belize . Both genetic markers showed similar high haplotype diversity ( nearly 1 ) and a low nucleotide diversity ( 2–3% ) , which suggested that almost every individual presents a unique haplotype and that haplotypes differ by few nucleotides ( Table 2 ) . The negative Tajima’s D suggested a bottleneck occurred in the recent past of the population’s evolution . Significant differences in TcI haplotypes among geographic regions ( North and Central America / Colombia / South America ) were evident in the PCA for both markers , 18S and SL-IR ( Fig 2A and 2B ) . The variation among geographic regions in both 18S and SL-IR was statistically significant ( 18S: Chi-Square 44 . 8 , n = 109 , d . f . = 2 , < 0 . 0001; SL-IR: Chi-Square 133 . 8 , n = 179 , d . f . = 2 , P < 0 . 0001 ) . For the 18S marker the first two components explained 46 . 0% and 21 . 7% , or a total of 77 . 7% of the variance . For the SL-IR marker the first two components explained 31 . 9% and 19 . 4% , for a total of 51 . 3% of the variance ( Fig 2A and 2B ) . The PCA also showed significant differences in TcI haplotypes among ecotopes ( sylvatic compared to domestic / peridomestic ) for 18S but not SL-IR ( Fig 2C and 2D ) . Logistic regression indicated that the 18S sylvatic haplotypes were significantly different from domestic and peridomestic ( Chi-Square 38 . 3 , n = 66 , d . f . = 2 , P < 0 . 0001 ) , however , the haplotypes did not differ for the SL-IR marker ( Chi-Square 5 . 0 , n = 166 , d . f . = 2 , P > 0 . 05 ) . For the 18S marker the first two components accounted for 61 . 6% and 16 . 6% of the variance for a cumulative total of 78 . 2% and for the SL-IR marker 31 . 8% and 20 . 4% of the variance for a total of 52 . 2% ( Fig 2C and 2D ) . The 18S MJ Network of TcI haplotypes included 39 haplotypes ( 10 from this study and 29 from GenBank ) , 49% ( 19/39 ) of the haplotypes contained only one sequence and one predominant haplotype contained 18 sequences ( N1 , Fig 3A ) . Geographic association of haplotypes was also evident in the MJ Network analysis where the majority of the Brazilian TcI haplotypes appeared in group I . The only other location represented in group I was a tight cluster of four Colombian haplotypes; only one mutational step separated each of these four haplotypes and represented 19 sequences . Group II contained all of the North and Central American haplotypes and the remaining Colombian haplotypes including the predominant haplotype ( N1 ) . In general , haplotypes appeared to be spread across hosts and vectors by MJ Network analysis with the 18S marker ( Fig 3B ) . For example , N1 contained TcI isolates from five different taxa . However , some haplotypes were only identified in T . dimidiata ( two clusters within group II ) and the cluster of four haplotypes from Colombia were all from humans ( arrow , Fig 3B ) . Also by the 18S marker , TcI haplotypes from different ecotopes appeared to be spread across the network as was evident in Group 2 ( Fig 3C ) . Ecotope data was mostly lacking for specimens in Group 1 with the exception of the cluster containing domestic isolates from humans in Colombia ( Fig 3C ) . The MJ SL-IR network included 89 haplotypes ( four new from this study and 85 from GenBank ) , with 74% ( 66/89 ) of the nodes represented by a single sequence and one predominant haplotype containing 50 sequences ( N1 , Fig 4A ) ; the remaining nodes contained between one and five sequences . Because of the large number of SL-IR sequences , we grouped the data into geographic regions instead of individual countries with the exception of Colombia because of the extensive sampling in this country . The predominant haplotype , N1 , contained sequences from all three geographic regions ( Fig 4A ) and included 63% ( 10/16 ) of all countries represented in the data ( Table 1 and S3 Table ) . Clustered close to N1 were nearly all the remaining TcI haplotypes from North and Central America ( circle within Group II ) . Two branches extended off this cluster: one of just Colombian haplotypes ( Group III ) and one of South American and Colombian haplotypes ( Group I ) . This SL-IR network showed no association between haplotype and host; indeed , N1 included isolates from 75% ( 9/12 ) of vector/host taxa ( Fig 4B ) . In addition , there was no clear clustering of haplotypes and ecotopes . N1 haplotypes were from all ecotopes and haplotypes from sylvan and domestic ecotopes appeared to be spread throughout the network , although peridomestic haplotypes were lacking in group 1 ( Fig 4C ) . Genetic distance was significantly correlated with geographic distance for both markers ( S1 Fig ) thus supporting an isolation-by-distance mechanism of genetic differentiation .
This study shows that TcI is the predominant DTU in T . dimidiata collected across its geographic range from Mexico , through Central America and into Colombia . Indeed , ninety-four percent of the T . cruzi-infected T . dimidiata contained TcI ( Table 1 , Fig 1 ) . TcIV was the only other DTU found , at a much lower frequency , 6% of the T . cruzi positive T . dimidiata . As previously 90 . 7% of T . cruzi DTUs reported were from South America [20] , this study provides important information about strains circulating in Central America and Mexico and in the most important vector in this region , T . dimidiata . Furthermore , because only a subset of the strains known from South America ( 2/7 ) were found in Central America and Mexico , these results do not challenge the South American origin hypothesis for T . cruzi [41]; a broader sampling will be required to answer this question . Our results are in accordance with other studies that show that TcI is the predominant DTU across Latin America and more specifically in the geographic range of T . dimidiata: Mexico [12 , 17 , 42] , Central America [14 , 18 , 23 , 24] , and northern South America [11 , 16] . TcIV is also the most commonly reported secondary strain in this region and the ratio we found ( 94% TcI / 6% TcIV ) is nearly identical to previous reports from Central America ( 93 . 3% TcI / 6 . 7% TcIV [20] ) . We did not find any of the rarely reported other strains in our T . dimidiata specimens [26 , 27 , 42] . This may be because our study did not include specimens from central Mexico or the southern Yucatan peninsula where these strains were identified . In addition , the infection prevalence we observed ( 38% ) is quite comparable to what was previously reported in Guatemalan T . dimidiata ( 39% ) , also determined by PCR [43] . The predominance of TcI and the high prevalence in T . dimidiata ( nearly 40% are carrying the parasite ) mean that this strain , in the most important vector in Central America , is responsible for the majority of Chagas disease in this region . A broader sampling of T . cruzi strains in other hosts and vectors will better clarify the epidemiology of Chagas in this region . TcI is also genetically quite diverse across the geographic range of T . dimidiata . A high diversity across this continental scale was also observed by Llewellyn , et al . [21] . The high diversity found in Central American TcI isolates , if it holds with additional sampling , may challenge the South American origin hypothesis of TcI [21 , 35] . Moreover , the clustering the majority of the Central/North American TcI isolates in and around the predominant node in the MJ Networks ( Figs 3 & 4A ) suggests that Central/North America may actually be the origin of TcI . A broader sampling , especially of other vectors and hosts , is needed to resolve this question . Importantly , we amplified directly from T . dimidiata abdominal DNA , therefore avoiding the biases resulting from culturing isolates prior to sequencing [44] . We found evidence of geographic structuring of TcI haplotypes by two markers and three types of analysis . First , significant separation of haplotypes between North and Central America / Colombia / South America was shown by PCA using both markers , 18S and SL-IR ( Fig 2A and 2B ) . Second , MJ Network analysis with both markers shows geographic separation of TcI haplotypes between South America and North/Central America , with the exception of Colombian haplotypes , which were found in both groups ( Figs 3A and 4A ) . Third , a significant correlation between genetic and geographic distance by the Mantel test also supports geographic structure , suggesting isolation by distance . This result is consistent with other studies that also found genetic and geographic structure in TcI isolates [19 , 21 , 35] . Although there is strong support for geographic structuring among TcI isolates , there is weak support for structure among ecotopes . By PCA , only one marker ( 18S ) showed statistically significant differences between domestic/peridomestic and sylvan isolates ( Fig 2C and 2D ) . This significant difference could just reflect geographic structuring as ecotope information is largely absent for isolates from South America from which 18S sequence was determined , with the exception of the domestic cluster from humans in Colombia . The difference in results between the two markers may also reflect sampling of different T . cruzi populations: 18S sequences are nearly all from specimens from Colombia and Brazil , whereas SL-IR sequences are from a broader geographic range . Ecotope structuring is also not supported by MJ Network for the SL-IR marker . There is also no evidence of host/vector association by MJ networks with either marker , consistent with previous studies [35 , 45] . A notable exception is an interesting human cluster from Colombia , evident in the 18S MJ network ( Fig 3B , arrow ) . The high similarity between these human TcI isolates from Colombia is not due to geographic proximity . In fact , the four nodes include 19 sequences that originate from geographically distant localities within Columbia , including six departments . It is possible that these represent the TcI subgroup , TcIDOM . TcIDOM has been described using the SL—IR [19 , 21 , 22] , and later also identified using cyt b [22] . However , this TcIDOM subgroup has not previously been typed using the 18S marker , and in our SL—IR network the subgroup was not observed , limiting the ability to correlate with the previously published TcIDOM . Suggesting that it is a distinct subgroup is the observation that the subgroup we identified clusters with the South American isolates , not the North/Central American isolates as was reported for TcIDOM . It will be important to check these isolates with multiple markers to confirm or refute an association with TcIDOM . This study provides new information about T . cruzi strains circulating within T . dimidiata across its large geographic range . Our results indicate that TcI predominates from Mexico through Central America , extending into Colombia and TcIV is also present in T . dimidiata collected from Mexico and Belize . Central American TcI strains add to the tremendous diversity found within TcI and provide additional evidence for geographic structuring , and a lack of evidence of host/vector or strong ecotope association . The high diversity found within this T . cruzi strain may challenge vaccine development and treatment improvement , if the genetically different strains respond differently to particular medications . The lack of host/vector and ecotope association suggests the parasite ( via the vector ) is moving frequently between hosts and ecotopes . These results support previous studies showing that T . dimidiata is a quite mobile vector [46] so that reinfestation some months following pesticide treatment is common [47] . This lends further support to development approaches for Chagas control in Central America/southern Mexico such as the Ecohealth approach [48 , 49] , which uses local materials and community participation to improve houses . Unlike the temporary effects of pesticide application , the Ecohealth approach makes houses refractory to the vectors long-term , so is likely to be more effective for sustainable interruption of transmission .
|
Little is known about the strains of the Chagas parasite circulating in Central America . This parasite is responsible for the most serious parasitic disease in Latin America and is presently divided into seven different strains . In Central America , the Chagas parasite is spread mainly by one species of kissing bug but what strains are present in this species was largely unknown . We investigated which strains of the parasite are present in this species of kissing bug by examining the DNA extracted from the abdomens of kissing bugs collected across its geographic range . We matched the DNA sequence we obtained with what is available in databases to determine the strain . We found mostly strain TcI ( 94% ) and less of strain TcIV ( 6% ) . As particular strains are associated with particular habitats , hosts , and disease symptoms , this work will help us understand why particular symptoms occur in particular areas and help us to target control efforts more efficiently .
|
[
"Abstract",
"Introduction",
"Materials",
"&",
"methods",
"Results",
"Discussion"
] |
[
"geographical",
"locations",
"parasitic",
"protozoans",
"genetic",
"mapping",
"north",
"america",
"protozoans",
"research",
"and",
"analysis",
"methods",
"central",
"america",
"sequence",
"analysis",
"south",
"america",
"bioinformatics",
"sequence",
"alignment",
"people",
"and",
"places",
"trypanosoma",
"cruzi",
"mexico",
"haplotypes",
"trypanosoma",
"dna",
"sequence",
"analysis",
"colombia",
"heredity",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"organisms"
] |
2017
|
The diversity of the Chagas parasite, Trypanosoma cruzi, infecting the main Central American vector, Triatoma dimidiata, from Mexico to Colombia
|
Metabolic networks perform some of the most fundamental functions in living cells , including energy transduction and building block biosynthesis . While these are the best characterized networks in living systems , understanding their evolutionary history and complex wiring constitutes one of the most fascinating open questions in biology , intimately related to the enigma of life's origin itself . Is the evolution of metabolism subject to general principles , beyond the unpredictable accumulation of multiple historical accidents ? Here we search for such principles by applying to an artificial chemical universe some of the methodologies developed for the study of genome scale models of cellular metabolism . In particular , we use metabolic flux constraint-based models to exhaustively search for artificial chemistry pathways that can optimally perform an array of elementary metabolic functions . Despite the simplicity of the model employed , we find that the ensuing pathways display a surprisingly rich set of properties , including the existence of autocatalytic cycles and hierarchical modules , the appearance of universally preferable metabolites and reactions , and a logarithmic trend of pathway length as a function of input/output molecule size . Some of these properties can be derived analytically , borrowing methods previously used in cryptography . In addition , by mapping biochemical networks onto a simplified carbon atom reaction backbone , we find that properties similar to those predicted for the artificial chemistry hold also for real metabolic networks . These findings suggest that optimality principles and arithmetic simplicity might lie beneath some aspects of biochemical complexity .
The prominent role of metabolism in any biological process and the fact that a large portion of the environmental factors shaping living systems are ultimately metabolic in nature , suggest that strong selective forces have been acting on metabolic networks throughout the history of life . In laboratory evolution experiments [1]–[3] one can witness mostly short term metabolic adaptations , affecting metabolic enzyme regulation and fine tuning of kinetic parameters . However , especially during major transitions , such as the early stages of life's appearance or the rise of oxygen in the Earth's atmosphere , selective forces must have shaped the metabolic wiring itself [4] . Comparative genomics can provide top-down insight into some long-term evolution of metabolic pathways [5] , [6] . In addition , studies of prebiotic chemistry scenarios have suggested possible seeds of biochemical organization from a bottom-up perspective [7]–[10] . Yet , whether the long term evolution of metabolism was dominated by unpredictable frozen accidents , or by inevitable network optimization processes , remains a fundamental open question . In a 1961 review , Baldwin and Krebs suggested that biochemical network topologies may reflect the adaptation toward optimally efficient metabolic strategies , and that manifold use of certain molecules may be a crucial element of this adaptation , as “it is indeed a general principle of evolution that multiple use is made of given resources . ”[11] . Some computational studies have proposed that the topology of specific metabolic pathways may have evolved towards maximal efficiency [12] , minimal number of steps [13] , or that network properties may reflect optimal organization [14] . Here we seek to address this problem by exploring a system that can reach a level of complexity comparable to the one observed in the union of all known metabolic pathways , yet is simple enough to allow efficient computation and analytical calculations . In addition , we wish to explore at an ecosystem-level the potential role of “metabolic multi-tasking” , as suggested by Baldwin and Krebs . The increasing evidence of abundant horizontal gene transfer in the history of life suggests that this question may be indeed especially relevant at the ecosystem level , where the interchange of genetic information might have created a free economy of enzymes among simple organisms , allowing for the emergence of species that share common molecular tools [15] . Recent metagenomic studies of microbial consortia [16] also suggest the question of whether metabolic functions , more than individual species distributions , might be directly dependent on environmental conditions . Hence it is possible that hallmarks of metabolic optimality in metabolic network wiring may be observable at the level of global ( multi-species ) metabolic networks [17] , [18] , more than at the individual species level . Do specific molecules , reactions or pathway topologies appear to be universally useful in a biochemical network , i . e . , relevant for maximally efficient completion of several possible metabolic tasks , possibly across multiple organisms ? In the present work , we combine the study of an extremely simple artificial chemistry [19]–[22] with recent systems biology approaches [23] , [24] to systematically compute pathways that are optimal for an array of elementary metabolic tasks , converting an input molecule into an output one . Behind the apparent complexity of the ensuing pathways , we identify recurring , modularly organized categories of network topologies , and analytically predictable trends in pathway length . In addition , we observe the emergence of “universal metabolic tools” across all optimal pathways . Finally , despite the huge gap in the underlying chemical rules , we find that some properties of real metabolic pathways are consistent with the patterns detected in the model , suggesting that fundamental optimality principles may have played a role in shaping biochemical networks .
Our artificial chemistry consists of a set of N possible molecules {a1 , a2 , a3 , … , aN} , that can participate in reversible ligation/cleavage reactions of the form ai + aj ↔ ak , with i+j = k . This model could be viewed as the simplest possible string-based artificial chemistry [19] . The reaction network RN that includes all metabolites up to length N and all possible reactions ( of the order of N2/4 , see Methods ) between them ( Fig . 1 ) can be thought of as the underlying chemistry based on which specialized metabolic tasks could emerge . Here we were concerned with pathways , within the RN network , that can optimally perform a given metabolic task . In particular , we searched for optimal solutions to the problem of producing a specific end-product ( e . g . , aj , with output flux vout ) from a single available nutrient ( e . g . , ai , with input flux vin ) . We define an optimal pathway as one that satisfies the following conditions: ( i ) it allows a steady state solution , i . e . , a mass-conserving flow from input to output; ( ii ) it has maximal yield , and no waste [25] , such that vout = vin·j/i; and ( iii ) it has the fewest reaction steps possible . A pathway satisfying these conditions is termed a minimal balanced pathway ( MBP ) between ai and aj , and will be denoted ai ⇒ aj . MBPs ( also referred to below as optimal pathways ) can be thought of as the pathways that are most efficient for a specific metabolic task , in the sense that they require the smallest possible number of different enzymes for producing the maximal possible yield [12] , [26] , [27] . Despite the simplicity of our artificial chemistry , identifying the MBPs between all possible input-output pairs in a given artificial chemistry RN is a challenge for large N . We implement three algorithms to approach this problem: a mixed integer linear programming ( MILP ) akin to flux balance analysis ( FBA ) [28]; an algorithm that uses enumeration of elementary flux modes [23]; and finally an iterative algorithm that gradually assembles new MBPs from already identified simple ones ( see Methods ) . The three algorithms differ mainly in their scalability , and in their capacity to predict multiple degenerate solutions ( see Table S1 ) . A partial overview of the results of our calculations is shown in Fig . 2A and Fig . S1 ( see Tables S2 and S3 for a comprehensive list of MBPs ) . Behind the apparent complexity of the topologies encountered in each of the different pathways , it is possible to observe the recurrence of three fundamental categories: each MBP functions either as a pure “addition chain” [29] , where smaller metabolites are progressively added together to build the target molecule , or as an “addition-subtraction chain” , in which metabolites are both synthesized and degraded within the pathway . Addition and addition-subtraction chains are concepts borrowed from the field of cryptography , whose relevance to our question will become apparent later . There is also a third , smaller category of cyclical pathways that cannot proceed unless a certain intermediate molecule is already present in the system . These pathways are autocatalytic cycles ( Fig . 2B ) that very much resemble autocatalytic cycles found in real biochemistry , such as the reverse TCA cycle [7] , or the formose reaction [30] . Our results show that autocatalytic cycles can be simultaneously optimal for multiple tasks ( Fig . 2B ) , suggesting that such types of structure may have a fundamental evolutionary advantage in a biological context . In addition to the recurrence of these topological categories among MBPs , we find that some specific structures are used repeatedly , often in a modular fashion ( Fig . 2C ) . Specifically , many simple MBPs are used hierarchically as a toolkit for the construction of progressively more complex MBPs ( data not shown ) , similar to what has been observed in real metabolic networks [15] , [31] , [32] . This modular architecture of recurring graph types provides a topological signature of optimally efficient pathways in our idealized chemistry . Since these pathways are chosen based on their minimal length , one may expect that a systematic analysis of all MBP lengths will display additional distinctive properties . Indeed , pathway lengths increase roughly logarithmically with the size of the input ( or output ) molecule ( Fig . 3 and S3 ) , with superimposed sharp jumps . For example , the task a9 ⇒ a6 can be performed in 2 steps , but the neighbor task a9 ⇒ a7 requires a minimum of 6 steps . Moreover , while most MBPs have only one or a few optimal realizations , selected instances display a peak in possible redundant solutions ( Fig . S2 ) , usually due to interconversions between molecules of similar size ( e . g . , ax ⇒ ax+1 ) , or to the inherent complexity of a specific molecule ( e . g . , a7 ⇒ aj ) . These regular patterns suggest that it may be possible to reproduce the MBP length curves without having to actually compute the MBPs . A similar search for patterns associated with minimal steps had been previously encountered in the mathematics of addition-subtraction chains , of high importance in cryptography [29] . These are integer sequences , beginning with 1 , in which the i-th entry is either the sum or difference of any two previous entries in the sequence . These chains are often used in calculating large exponents of numbers [33] . For example , calculating n128 can either be performed in 127 multiplications ( n × n = n2 , n2 × n = n3 , … , n127 × n = n128 ) or in a chain of 7 exponent multiplications ( n × n = n2 , n2 × n2 = n4 , n4 × n4 = n8 , … , n64 × n64 = n128 ) . The latter can be further simplified by tracking the sums of the exponents in each calculation , which form an addition chain ( 1 , 2 , 4 , 8 , 16 , 32 , 64 , 128 ) . Shortest addition-subtraction chains are commonly used to calculate very large numbers in the fewest number of steps , thus speeding up computation time . These are often applied to methods in cryptography where the calculated exponents can have on the order of thousands to tens of thousands of bits [34] . The pathways explored in our model resemble optimal addition-subtraction chains . For example , the problem of obtaining a128 from a1 is formally equivalent to the addition chain example described above . However while typical addition-subtraction chains start with the number 1 , in our MBPs we explore minimal paths starting from any molecule ai ( i≥1 ) . As described in detail in the Methods , we extended previous work on addition-subtraction chains [29] , [35] to derive the following analytical estimate of the length of MBPs: ( 1 ) where L ( i , j ) is the number of reactions in the MBP with input ai and output aj , and gcd ( i , j ) is the greatest common divisor of i and j . As seen in Fig . 3 , Eq . ( 1 ) reproduces the corresponding pathway lengths obtained by computing individual MBPs . This agreement implies that the number of reaction steps needed to construct an efficient metabolic pathway between two metabolites in our artificial chemistry can be roughly estimated from Eq . ( 1 ) . The only feature that determines the pathway lengths is the complexity of the input and output molecules . We can now ask whether similar minimal pathway length signatures are discernible in real metabolic networks . To cope with the gap in complexity between our model and real chemistry , we mapped real metabolic networks onto a single atom backbone [13] , [14] . For example , the aldolase reaction , which cleaves fructose-1 , 6-bisphosphate ( C6H14O12P2 ) into dihydroxyacetone phosphate ( C3H7O6P ) and glyceraldehyde-3-phosphate ( C3H7O6P ) , can be mapped onto a carbon atom backbone , becoming simply C6 ↔ C3 + C3 ( see Methods ) . This reaction is now formally analogous to the a6 ↔ a3 + a3 reaction in the idealized chemistry . Upon performing this mapping onto a carbon atom backbone , we ask whether the structure of real metabolic networks allows interconversions that use the minimal , logarithmic number of steps found for the artificial chemistry ( Fig . 3 and Eq . ( 1 ) ) . Specifically , we identified all shortest pathways between any two carbon compounds in Escherichia coli's metabolic network . This was performed using two methods . The first was an explicit use of elementary flux modes as done in the artificial chemistry . As with the artificial chemistry method , this has the advantage of finding all of the shortest pathways that connect any two carbon compounds , but is limited in computational scope to a smaller network . Because of this limitation , we used the network of E . coli's central carbon metabolism [36] , [37] , modified to remove cofactors and reactions that do not affect carbon transfer ( see Methods ) . After finding all minimal elementary flux modes that connect every pair of carbon compounds in the network , we reduced those compounds to their carbon content alone , as described above . We determined , for each input compound , the length of the shortest elementary flux mode that reaches its closest molecule with j carbons; then , for each value of i , we averaged these path lengths over all input molecules with i carbons . The results show that the lengths of the E . coli elementary flux modes correlate with the lengths of the corresponding artificial chemistry MBPs and with the analytical predictions , though the actual E . coli values are overall larger than the artificial chemistry ones ( Fig . 4 ) . This last fact , as discussed later , may be due , for example , to energetic constraints , or to the higher complexity of real organic chemistry . The second method is aimed at identifying all shortest pathways between any two carbon compounds in the whole genome-scale metabolic network of Escherichia coli [38] , for which it is still infeasible to apply the elementary flux mode analysis . For this , we implemented a heuristic approach to analyze the set of shortest pathways between every pair of metabolites . We first determined , for each input compound , the minimal path length to reach its closest molecule with j carbons; then , for each value of i , we averaged these path lengths over all input molecules with i carbons . The results ( Fig . 3 and S3 ) show that these E . coli minimal path lengths approximately follow the predicted logarithmic trend . For some curves ( e . g . the one with C5 as an input ) , the specific peaks and valleys of the predicted function are closely followed by the E . coli network . While this does not prove that MBPs are indeed used in real metabolic networks , it suggests that the logarithmic strategy of MBPs is embedded in their architecture . However , because this second method focuses on shortest paths across a metabolite-to-metabolite network rather than on flow-conserving MBPs , the predicted values are likely an underestimate of the number of reactions necessary to construct one metabolite from another in a mass-conserved manner . So far , we have analyzed the properties of individual MBPs in our idealized chemistry , as well as analogous minimal length pathways in E . coli metabolism . However , some fundamental aspects of the architecture of metabolism may be visible only at the ecosystem-level , namely by collectively analyzing the metabolic network obtained as the union of the metabolic maps of known individual species ( sometimes called the “meta-metabolome” ) . For example , previous work using an algorithm of network expansion applied to this meta-metabolome has identified potential signatures of major evolutionary events [39] , including the metabolic transition that took place upon the great oxidation event , about two billion years ago [4] . Here , we build a meta-metabolome for our idealized chemistry by considering the collection of MBPs . One could imagine that each task ai ⇒ aj corresponds to a different organism , which has filled a specific metabolic niche ( availability of ai ) , and found an optimal solution ( the MBP ) for its main metabolic task ( produce aj ) . The question we ask next is whether , in this ecosystem of MBPs , all metabolites and reactions are used in roughly the same number of pathways , or if specific metabolites or reactions seem to be essential for many optimal tasks , hence representing “universal tools” . For this analysis we used the set of MBPs calculated on the R19 network using the MILP method . One first result of this analysis is that every metabolite of an even length is used in many more MBP reactions than their odd length neighbors , compared to the underlying chemistry ( Fig . 5B ) . Thus even-length metabolites are more important in that they can be used for more tasks . A possible explanation for this enhanced importance comes from the logarithmic nature of the MBP path lengths . For example , producing a8 from a1 requires only three doubling reactions ( a1 + a1 → a2 , a2 + a2 → a4 , and a4 + a4 → a8 ) . In addition , this same pathway , with one additional reaction , can also be used to optimally produce a9 and a10 ( see Tables S2 and S3 ) , overall increasing the number of pathways in which each of those even-length intermediates is used . Indeed , because similar logarithmic pathways can be used as a backbone connecting distant inputs and outputs , we expect metabolites of an even length to appear more often . Similarly , one can address the relevance of each possible reaction across different MBPs . The existence of ubiquitous reactions is visible in Fig . 2A and Fig . S1 , and can be more systematically assessed by plotting a usage distribution ( Fig . S2 ) . The most abundant reactions – the “universal tools” in this model chemistry – are the ones that ligate two identical molecules ( e . g . a2 + a2 ↔ a4 , see Tables 1 , S4 , and S5 ) . The distribution of reaction utilization follows a long-tailed distribution ( Fig . 5A ) , whose fit to a power law gives an exponent of approximately −1 . 1 ( R2 = 0 . 99 ) . This value is close to our theoretically predicted value of −1 ( See Methods ) . As in the case of the artificial chemistry network , we can now search for patterns of metabolites and reactions usage in the collective set of all metabolic reactions known in living systems , obtained from the KEGG database [17] . The presence of such signatures would suggest a long-term selective advantage of molecules and reactions that are useful for multiple tasks across different organisms and environments . By counting how many times each possible carbon backbone reaction is used across this biosphere-level metabolism we obtained a broad distribution , and a fit to a power-law gives the exponent of −0 . 89 , comparable with the analytically predicted value , and with that in the artificial chemistry model ( Fig . 5A and Fig . S4 ) . We found that several reactions that are top ranking in their count across MBPs in the artificial network , are also at the top of the list in the KEGG-derived reactions ( Spearman correlation p-value<10−6; see also Tables 1 and S6 , S7 , S8 , S9 , S10 , and S11 ) . This suggests that the RN network model , despite its simplified chemical rules , captures some fundamental features of the role of the carbon reaction backbone of real metabolic networks . In addition to a preference for specific reactions , we can ask whether the spectrum of metabolite usage across the whole KEGG metabolism reflects the possible optimality criteria encountered in the model ( Fig . 5C ) . The metabolite usage in the hydrogen backbone network ( see Methods ) is similar to that in the artificial chemistry: each even-length hydrogen metabolite is used more often than its odd-length neighbors ( Fig . S4A ) . For the carbon backbone distribution , we see a similar descending periodic behavior , but with a periodicity of approximately 5 ( Fig . 5C and Fig . S5B ) . Hence , molecules containing carbons in a number that is a multiple of 5 are used more abundantly than other molecules across different metabolic reactions . One possible explanation for this C5 periodicity is the profuse usage of adenine and nicotinamide adenine dinuculeotide compounds as energy carriers and redox balance molecules , although the removal of such compounds has little effect on the observed periodicity ( Fig . S6 ) . Hence , the prominent usage of compounds with specific numbers of carbons might reflect global network optimization principles for the efficiency of multiple pathways , as observed in the artificial chemistry model . The periodicity of 5 that we observe , together with the evidence displayed in Fig . 3C , may suggest that the evolutionary optimization of metabolism has been partially taking place around building blocks of five carbons , compatible with previous observations of prebiotic abundance of terpenoids [40] and pentoses [41] . It is also interesting to note that an unexplained periodicity of two had been previously observed in the distribution of the number of carbons among known organic compounds [42]–[44] . While our analysis is based on the distribution of usage of carbon compounds in different reactions , rather than the total count of molecules , future analyses may investigate possible connections between these trends .
We have explored the potential existence of general principles underlying the evolution of metabolic network architecture . Specifically , we studied the properties of pathways ( the MBPs ) that perform elementary metabolic tasks with maximal yield and minimal length in an idealized chemistry . Using the results from the model chemistry , we asked whether similar signatures of optimally efficient organization could be found in real metabolic networks . In computing possible MBPs , we have focused mostly on identifying modular features , on predicting their lengths , and on the statistics of usage of metabolites and reactions . In the future , it may be interesting to characterize the full spectrum of degenerate MBPs for large artificial chemistries . This would allow us to assess , for example , the density of specific topologies ( such as autocatalytic cycles ) , or the dependence of degeneracy on the numerical properties of input/output pairs . One of our algorithms ( the elementary modes one ) can find a large number of degenerate solutions , including autocatalytic cycles . This algorithm is currently not scalable , because of the difficulty of computing elementary flux modes , especially in the highly connected artificial chemistry network we have used , though very recent improvements in elementary flux mode calculations [45] might be useful for this enumeration . In addition , while intuitively this approach seems to capture the full degeneracy of MBPs , this still remains to be formally proven . Another intriguing possibility might be to modify our flux balance MILP approach to identifying degenerate solutions by employing integer cuts , as described in [46] . Alternative avenues for optimization using Linear Programming rather than MILP could be in principle devised to reduce the complexity of calculations . For example minimizing the sum of absolute values of fluxes allows for rapid calculation of pathways up to the R100 network , though in this case the ensuing pathways are not of minimal length ( see Fig . S7 ) . In any case , for the purpose of the current work , we verified that degeneracy does not affect the statistics of usage of different reactions ( Fig . S8 ) . Among the recurrent MBP topologies identified , we encountered numerous autocatalytic cycles . The properties of autocatalytic cycles have been studied previously [8] , [11] , and their self-replication potential has been theorized to be important in the early evolution of carbon fixation [7] , [9] . Autocatalytic cycles have also been shown to be kinetically stable , even in the absence of regulatory control [47] . We found that some autocatalytic pathways ( e . g . the pathway from a7 to a8 ) are simultaneously optimal for multiple metabolic tasks . In this specific case , the MBP a7 ⇒ a8 is also an MBP for the production of each intermediate in the cycle ( Fig . 2B ) . This special property of autocatalytic cycles in our artificial chemistry may have a parallel in real metabolism . For example , many metabolites in the TCA cycle ( which is autocatalytic when run in reverse [7] ) are precursors for fundamental anabolic processes [48] , [49] . Similar properties can be observed in the fundamental autocatalytic cycle known as the formose reaction [30] . Along with the structural details of MBPs , we also used analytical methods to estimate the length of MBPs as a function of the length of input and output molecules . This estimate closely matched the lengths of the artificial chemistry pathways computed with numerical algorithms . These calculations establish a new link between two apparently unrelated disciplines , namely the mathematics of addition-subtraction chains and biochemistry . It will be interesting to explore in the future whether extensions to more realistic artificial chemistries can be formalized in a similar fashion . Conversely , the MBP length estimate obtained for biochemical pathways may suggest new avenues in applied mathematics . To determine whether predictions of minimal MBP length in our idealized chemistry could have implications in real biochemistry , we searched for pathways of minimal length between compounds with different counts of carbon atoms in the E . coli metabolic network . The complexity of real chemistry relative to our idealized system made this comparison difficult to interpret . MBPs in E . coli were found to be composed of many more reactions than in the artificial chemistry . This might be due to the additional requirement of energy gradients ( e . g . the phosphorylation steps ) , to the complex interdependence of multiple elements , and to the properties of stability of intermediates . Previous work had addressed the question of optimality in specific metabolic pathways . For example , Meléndez-Hevia and Torres [50] , used optimality criteria to infer that some metabolic pathways , most notably the pentose-phosphate pathway , can be traversed using the fewest number of reactions . Heinrich and Schuster [51] , conversely , describe the identification of a series of phosphorylation/dephosphorylation and ATP consumption/production steps that maximize the flux of ATP production in central carbon metabolism pathways . In contrast , our search for MBPs in the artificial chemistry model corresponds to a systematic search for maximal yield , minimal length pathways for all possible input-output relationships ( something not yet feasible for real metabolic pathways ) . This allows us to infer analytical relationships and potential principles that may hold ( perhaps in an approximate way ) for virtually any evolved metabolism . In real systems , various compounds are often produced as waste byproducts and simply excreted into the environment , leading to suboptimal yields of target metabolites . For example , one could go in one step from a C6 compound to C5 , with a yield of 5/6 ∼83% and an additional C1 byproduct; yet , obtaining maximal yield in this transformation , would cost at least 3 more reactions . One could argue that the combination of all of these criteria , including the maximization of ATP production and optimization of enzymatic catalysis may have played a key role in the evolution of modern metabolism , leading to compromise solutions . Exploring pathways that produce multiple compounds from multiple inputs with the addition of thermodynamic constraints might constitute an interesting model extension for further investigation . We found that the statistical properties of the usage of reactions across MBPs recapitulate the statistics of reaction usage in the union of all known metabolic pathways ( represented by the KEGG metabolic database ) . Both across the set of all MBPs for the idealized chemistry , and in the KEGG metabolic map , we observed that a few reactions are used far more often than many of the others in the set . Another way of determining the importance of individual reactions in the context of the global functionalities of a meta-metabolome would be to perform perturbation experiments . We implemented such an experiment in our idealized chemistry , by progressively removing reactions and checking how many metabolites can still be produced . Depending on whether reactions are removed in random order or in the order determined by their usage across MBPs , the outcome is quite different ( Fig . S9 ) . This analysis sheds light on the importance of reactions in terms of the capacity to produce a certain output . Determining to what extent real metabolic networks obey optimality principles like the ones described here will take additional effort . Even if an underlying arithmetic simplicity governs idealized optimal pathways , deviations from ideal behavior should be expected . For example , parallel selection pressures for energy production and biochemical stability would likely sacrifice pathway minimality . However , guiding principles as the ones we are proposing could serve as reference points for future research , including circumstances in which metabolism can be different from what we are used to . Using synthetic biology techniques , for example , it might be possible to redesign metabolic pathways so as to approach predicted ideal efficiencies and minimal enzyme cost [52] , [53] . From a totally different perspective , in the field of astrobiology , having a prediction of possible signatures of an evolved metabolism might help select , among the molecular spectra of extrasolar planets , those possibly indicative of biogenic processes [54]–[56] .
We define an artificial chemistry inspired by previous string-based artificial chemistries ( see also main text and Fig . 1 ) . One may think of molecules in this artificial chemistry as polymers ( up to a given length N ) of a monomeric unit a . Since no specific assumption is made in the model about the nature of these molecules , they could equally represent aggregates or branched polymers of different sizes , as well as molecules with different counts of a specific atom . A network RN = {MN , CN} is defined by the set of N molecules MN = {ai | ∀ i = 1 , … , N} and the set of all possible uni-bi ligation/lysis reactions between them , CN = {ai + aj ↔ ak | ∀ i , j , and k , such that i≤j , i+j = k , and k = 2 , … , N} . The number of reactions in CN can be shown to grow quadratically with N . Since each reaction describes how to reversibly combine two molecules to make a larger one , there is a fixed number of ways to produce any given metabolite ai , equal to ⌊i/2⌋ . The number of reactions in CN is then given by , which can be approximated by N2/4 . Flux Balance Analysis ( FBA ) is a steady state constraint-based approach to study the flow of mass through metabolic networks [28] , [57] , [58] . Briefly , FBA represents the metabolic network of interest as an n×m stoichiometric matrix S , whose element Sij indicates the number of molecules of metabolite i ( i = 1 , … , m ) that participate in reaction j ( j = 1 , … , n ) ( with a positive sign if the metabolite is produced , negative if it is consumed ) . Each reaction can be associated with a flux , vj . Under the assumption of a steady state the following set of mass conservation constraints on the fluxes is generated: ( 2 ) Additional constraints ( such as availability of nutrients , experimentally observed irreversibility , maximal or minimal rates , etc . ) can be imposed on the fluxes as inequalities of the form ( 3 ) where αj is the minimal allowed rate of a reaction and βj is its maximal rate . Taken together , the above constraints define a convex polyhedron ( the “feasible space” ) in the n-dimensional space of fluxes . Linear programming ( LP ) can be used to identify , within the feasible space , flux vectors that maximize or minimize a given linear objective function . In microbial systems it has been often hypothesized that a biologically meaningful objective is the maximization of the flux through the reaction that represents cellular growth , or biomass production [2] , [59] . Hence , LP applied to FBA provides a prediction of all metabolic fluxes in a cell . FBA can be applied at genome scale , and corresponding stoichiometric models are available for a number of organisms . FBA predictions have been experimentally validated most thoroughly in Saccharomyces cerevisiae SC288 [60] and E . coli K-12 [38] . Minimal Balanced Pathways ( MBPs ) are defined as sets of reactions in the RN network that can optimally perform a given metabolic task . A task is defined as the production of a specific end product ( e . g . , aj , with output flux vout ) from a single available nutrient ( e . g . , ai , with input flux vin ) . A pathway between two molecules is a MBP if ( i ) it satisfies a steady state solution , analogous to Eq . ( 1 ) ; ( ii ) it produces the final product with maximal yield , i . e . , vout = vin·j/i; and ( iii ) it contains the smallest possible number of reaction steps . The MBP between ai and aj will be indicated as ai ⇒ aj . We have developed three different algorithms for computing MBPs , as described below: We use a modified FBA approach to formulate the MBP problem in a constrained optimization framework . Specifically , we impose the same constraints used in an FBA problem , and further require that the maximal yield condition vout = vin·y/x for the MBP ax ⇒ ay be satisfied . We then search for a solution that minimizes the number of active ( nonzero ) fluxes . Towards this goal , we use a modification of the LP problem described above to introduce binary variables ( bj ) that represent flux activity: bj = 0 if vj = 0 , and bj = 1 otherwise . To identify a minimal path , we can then search for the set of fluxes that minimize Σibi . Because of the nature of the variables involved – the fluxes are continuous , and the number of active fluxes is an integer – this problem must be solved using a mixed integer linear programming ( MILP ) algorithm . Our MILP problem for the optimal MBP satisfying ax ⇒ ay is formulated as follows: ( 4 ) The optimal solution for this problem will give the flux distribution v that uses the fewest nonzero values to maximize the objective . In our MBP computations , the only flux constraints used were those that limit the uptake of the single nutrient to an arbitrary value of 10 mmol·gDW−1·h−1 , and the production of the target metabolite to the known maximal yield vout/vin = j/i . Given a metabolic network defined by a stoichiometric matrix S ( as described in the above FBA section ) , a vector of fluxes v is said to correspond to an elementary flux mode ( EFM ) if it satisfies the following three conditions [23] . Because of these constraints , those EFMs that use the minimal number of reactions satisfy the requirements for being an MBP . We used the METATOOL software package [61] to find all EFMs in the R10 network , and then identified all of those EFMs that are also MBPs . We designed and implemented an algorithm to produce most MBPs de novo , without relying on prior steady state stoichiometric modeling methods . The algorithm works in an iterative manner , producing longer pathways from shorter ones . For example , we can start from two trivial MBPs: a1 ⇒ a1 ( which requires no reactions ) , and a1 ⇒ a2 ( requiring one reaction , a1 + a1 → a2 ) . To compute a1 ⇒ a3 , we identify all the ways in which we can decompose 3 into two smaller addends ( in this case , only one: 3 = 2+1 ) . Next we combine together the previously computed MBPs that progress from a1 to each of these two addends , giving a new putative MBP for the desired new task ( a1 + a1 → a2 , and a1 + a2 → a3 ) . This procedure can be then iterated to give a prediction of MBP ai ⇒ aj ( i , j≤N ) . This algorithm is fast and efficient compared to the previous methods , allowing us to apply it to even the R100 network . However , it has two main drawbacks . First , it will miss pathways that “overshoot” the target value then subtract down to it . Second , it may miss MBPs that are not built modularly from smaller ones . From a comparison of the MBPs predicted by the different algorithms , one can see that the approximations introduced in this algorithm cause 18 out 361 MBPs ( 5% ) in R19 to overestimate pathway length by one reaction . Also , this algorithm correctly identifies 204 of the 384 degenerate MBPs that the EFM algorithm finds in R10 . The reaction usage using this method is highly correlated with that of the MILP method applied to R19 ( Pearson correlation = 0 . 96 , p-val = 10−51 ) , and the EFM method applied to R10 ( Pearson correlation = 0 . 98 , p-val = 2·10−17 ) . Data used for the comparison between the RN metabolic network and real metabolism was gathered from the KEGG LIGAND database ( July 26 , 2009 release ) [17] . This database was parsed to convert its compounds and reactions into a single-atom form , as described in the text . Compounds that carried any uncertainty in their atomic makeup , including non-specific side-chains or variable chain length were removed from the current analysis . We also removed from the analysis reactions with no associated formula , as well as reactions involving non-specific molecules ( such as glycans and non-specific nucleotide or peptide chains ) . Finally , a number of reactions were found to leave the atomic composition of the compounds essentially unchanged on either side of the reaction ( e . g . , C3 ↔ C3 ) . These reactions were ignored as well , without consequences on the results ( data not shown ) . We counted how often each metabolite and reaction was used in the artificial chemistry pathways as well as in the KEGG-derived single-atom networks . In the model pathways , reaction usage was calculated by counting how many times each reaction was used across all pathways . Metabolite usage was similarly calculated by counting the occurrence of reactions in which each metabolite participates . For example , in the pathways that convert a9 to a10 in Fig . 2C , a9 participates in only one reaction , but a10 participates in two . In the KEGG-derived networks , a similar counting scheme was used . The reaction usage was calculated by counting how many times each reduced reaction appears , and the metabolite usage was calculated by counting how many times each metabolite appears across all reactions . The first method used EFMs to find all shortest pathways in the central carbon metabolism of Escherichia coli [36] , [37] . Because we are interested in the pathways that alter the carbon content of different molecular species , we removed those common cofactors that do not alter the carbon content of other metabolites ( ATP , ADP , AMP , NADH , NADPH , NAD+ , NADP+ , H+ , and PO4 ) . Also , we effectively ignored reactions involving transport and exchange . We then used the EFM method described above to find the number of reactions in each of the MBPs for this reduced network . For each input compound , we listed the lengths of the MBPs for all output compounds containing a number j of carbons . For each j , we select among these paths the shortest one , giving an estimate of the shortest path between any individual compound and the closest j-carbon compound . Finally , for each value of i , we averaged these path lengths over all input molecules with i carbons . The end result is a matrix that provides the average of the shortest paths from any i-carbon compound to its nearest j-carbon compound . The larger , genome-scale metabolic network of E . coli [38] has 761 metabolites and 1075 reactions , and is currently infeasible to study with the EFM method described above , limiting us to a graph theoretical approach . Because there are both metabolites and reactions , metabolic networks are inherently bipartite: metabolites connect to each other only through reactions . Pathway lengths were computed by transforming the metabolic reaction network stoichiometric matrix into an adjacency matrix where metabolites and reactions are all represented by the same type of node . As described above , we are only interested in the connections between carbon compounds , so we removed any non-carbonaceous metabolites ( water , phosphate , ammonia , etc . ) . Also , we removed the following cofactors that are used in many reactions , but do not participate in the transformation of carbons: ATP , ADP , AMP , NAD+ , NADH , NADP+ , NADPH , coenzyme A , acetyl-CoA , and the acyl carrier protein . Next , we used Johnson's all-pairs shortest paths algorithm ( available as a Matlab function ) to find shortest pathways between any two carbon compounds in E . coli's metabolic network . This set of pathways was collated as in the previous section , to produce a matrix of the average of the shortest paths from any i-carbon compound to its nearest j-carbon compound . We developed an analytical approximation for the expected numbers of reactions to be found in any MBP ai ⇒ aj . We begin with a simplified version of the artificial chemistry model in which only irreversible addition reactions of the form ( 5 ) are allowed . Under these restrictions , we first ask what is the smallest number of reactions necessary to produce any aj from a1 . We shall denote by l ( j ) the smallest possible number of such reactions ( we count the use of each reaction ( 5 ) once ) . This problem is equivalent to the problem of addition chains [29] , in which one attempts to compute a positive integer by generating a sequence of integers such that each term in the sequence is the sum of two previous terms . Addition chains have been studied extensively , mainly because of their applications in computer science and cryptography [29] . For addition chains , l ( j ) grows logarithmically with j: ( 6 ) Our artificial chemistry represents a generalization , in which a metabolite of any length i can be used to produce an output metabolite of any length j . If we still assume that only addition reactions are possible ( i . e . molecules cannot be broken down ) , a chain from ai to aj will exist only when i is a divisor of j . The problem can then be reduced to the case with a1 input and aj/i output . Therefore , in the irreversible case , we can assume that inputs consist of monomers without loss of generality . Let L ( j ) be the length of the shortest reaction chain in this case . Because not every reactant exists when dividing by the input length i , we have the obvious inequality ( 7 ) Sometimes the shortest chain can be found easily . For instance , {20 , 21 , … , 2k} is obviously the shortest chain from 1 to j = 2k whose length is k+1 . This suggests the general lower bound on the shortest length L ( j ) of the addition chain: ( 8 ) where ⌈x⌉ represents the ceiling of x , or the smallest integer not less than x . Likewise , as seen below , ⌊x⌋ represents the floor of x , or the largest integer not greater than x ( for example , ⌈3 . 14⌉ = 4 , and ⌊3 . 14⌋ = 3 ) . The longest minimal addition chain arises when the output length is j = 2m−1 . From this fact , we have the upper bound [29] ( 9 ) where υ ( j ) is the number of 1s in the binary representation of j . Since , the bound in equation ( 9 ) implies a simpler ( but weaker ) upper bound ( 10 ) The above bounds give precise values in some cases and act as bounds in others . For instance , L ( 16 ) = 5 , and L ( 17 ) = L ( 18 ) = 6 for both the lower and upper bounds , while L ( 31 ) = 8 is between the lower bound and the upper bound ( 5 and 9 , respectively ) . There are various conjectures regarding L ( j ) ; one of the most famous [35] asserts that computing L ( j ) is NP-hard . Nonetheless , the computation of L ( j ) has been pushed up to n≤225 . Two other conjectures [33] predict the general lower bound ( 11 ) and the upper bound ( 12 ) While algorithms for generating the shortest addition chains are discussed by Thurber [33] , these all hold for the specific case of pure addition where the input is always a1 . We are interested in the general case involving both addition and subtraction , and specifically the lengths l ( i , j ) of the shortest reaction chains ( MBPs ) with ai input and aj output . Addition-subtraction chains have also been studied previously as an expansion of addition chains , although these correspond to MBPs with only a1 as an input . Sometimes , in these cases , l ( j ) is readily computable , e . g . ( 13 ) while remains unknown for sufficiently large k . Both lengths can also be equal , i . e . l ( j ) = L ( j ) . For example , ( 14 ) ( 15 ) Note also an inequality: ( 16 ) All of these features explain the growth law in equation ( 6 ) . The quantity l ( i , j ) has a rich behavior , e . g . , there is only a trivial lower bound since l ( j , j ) = 1 . To ignore this non-interesting effect , let us divide i and j by their greatest common divisor as it never affects the length of the MBP: ( 17 ) then we can use an obvious inequality ( 18 ) Recalling ( 6 ) we finally arrive at an approximation for the number of reactions in an MBP that uses ai to produce aj: ( 19 ) The approximation in ( 19 ) can also be used to estimate the rank distribution of reaction usage . Consider all possible MBPs producing aj from ai . For each ( i , j ) pair , take an MBP and mark all reactions . Let the reaction in ( 5 ) occur Epq times: that is , there are Epq MBPs that use ( 5 ) . We now divide Epq by the total number of MBPs and call the reaction frequency . It is better to order reactions not according to ( p , q ) but to their ranking j , so that the reaction of rank j = 1 is the most frequent , that of rank j = 2 is the second in frequency , etc . This gives ej . How does ej decrease with rank ? To infer the answer we note that ( 20 ) From ( 19 ) it is clear that the average length 〈l〉 of the shortest reaction chain scales as log N . This is consistent with ( 20 ) if and only if we have rj ∼ j−1 . Thus we predict the power-law decay in ( 21 ) . ( 21 )
|
Metabolism is the network of biochemical reactions that transforms available resources ( “inputs” ) into energy currency and building blocks ( “outputs” ) . Different organisms have different assortments of metabolic pathways and input/output requirements , reflecting their adaptation to specific environments , and to specific strategies for reproduction and survival . Here we ask whether , beneath the intricate wiring of these networks , it is possible to discern signatures of optimal ( i . e . , shortest and maximally efficient ) pathway architectures . A systematic search for such optimal pathways between all possible pairs of input and output molecules in real organic chemistry is computationally intractable . However , we can implement such a search in a simple artificial chemistry , which roughly resembles a single atom ( e . g . , carbon ) version of real biochemistry . We find that optimal pathways in our idealized chemistry display a logarithmic dependence of pathway length on input/output molecule size . They also display recurring topologies , including autocatalytic cycles reminiscent of ancient and highly conserved cores of real biochemistry . Finally , across all optimal pathways , we identify universally important metabolites and reactions , as well as a characteristic distribution of reaction utilization . Similar features can be observed in real metabolic networks , suggesting that arithmetic simplicity may lie beneath some aspects of biochemical complexity .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"computational",
"biology/metabolic",
"networks",
"computational",
"biology/systems",
"biology",
"physics/interdisciplinary",
"physics"
] |
2010
|
Signatures of Arithmetic Simplicity in Metabolic Network Architecture
|
Defining the correlates of immune protection conferred by SIVΔnef , the most effective vaccine against SIV challenge , could enable the design of a protective vaccine against HIV infection . Here we provide a comprehensive assessment of immune responses that protect against SIV infection through detailed analyses of cellular and humoral immune responses in the blood and tissues of rhesus macaques vaccinated with SIVΔnef and then vaginally challenged with wild-type SIV . Despite the presence of robust cellular immune responses , animals at 5 weeks after vaccination displayed only transient viral suppression of challenge virus , whereas all macaques challenged at weeks 20 and 40 post-SIVΔnef vaccination were protected , as defined by either apparent sterile protection or significant suppression of viremia in infected animals . Multiple parameters of CD8 T cell function temporally correlated with maturation of protection , including polyfunctionality , phenotypic differentiation , and redistribution to gut and lymphoid tissues . Importantly , we also demonstrate the induction of a tissue-resident memory population of SIV-specific CD8 T cells in the vaginal mucosa , which was dependent on ongoing low-level antigenic stimulation . Moreover , we show that vaginal and serum antibody titers inversely correlated with post-challenge peak viral load , and we correlate the accumulation and affinity maturation of the antibody response to the duration of the vaccination period as well as to the SIVΔnef antigenic load . In conclusion , maturation of SIVΔnef-induced CD8 T cell and antibody responses , both propelled by viral persistence in the gut mucosa and secondary lymphoid tissues , results in protective immune responses that are able to interrupt viral transmission at mucosal portals of entry as well as potential sites of viral dissemination .
Despite the considerable resources committed to developing an effective HIV vaccine over the past three decades , this objective remains elusive . Recent failures of HIV vaccine trials to demonstrate protection against infection [1 , 2] and the only marginal apparent efficacy demonstrated in another , in which the observed limited protection was associated with unanticipated immune correlates [3] , have refocused the field on comprehensive efforts to identify the fundamental determinants of a protective vaccine-induced immune response . Although few models of spontaneous lentiviral control exist , long-term nonprogressors and live attenuated SIV ( LASIV ) vaccinated animals have both proved to be highly illuminating models regarding the requisite immune correlates of viral control . The most effective lentiviral vaccine to date , SIVΔnef , has demonstrated durable protection against both systemic [4 , 5] and mucosal challenge routes [6–8] , and against heterologous challenge virus [7 , 8] . However , safety issues identified first in SIVΔnef infection in infant macaques [9–11] , and subsequently in some adults [10] , preclude attenuated lentivirus vaccination as a viable vaccine strategy for HIV . Nonetheless , studies to identify correlates of protection in this premier model of successful vaccine-induced protection against lentiviral challenge could certainly shed light on critical attributes of a vaccine to protect against HIV . A hallmark of live attenuated SIV vaccines , including SIVΔnef , is that protection against wild-type SIV pathogenic challenge typically increases with time , plateauing at 15 to 20 weeks post vaccination [5 , 12 , 13] . However , the immune mechanisms that underlie this protection , and its maturation , have been hotly debated . SIVΔnef vaccination generates strong antibody responses [14 , 15] and depletion of CD8 T cells does not abrogate protection induced by SIVΔ3 ( which contains deletions in nef , vpr and the U3 region ) in Mamu-A*01- macaques [16] , implicating antibody responses as a potential mechanism of protection . Furthermore , recent studies of anti-Env antibody responses have described SIV-specific antibody-dependent cell-mediated cytotoxicity ( ADCC ) [15 , 17] and antibody responses to trimeric gp41 concentrated by the neonatal Fc receptor ( FcRn ) in the vaginal and cervical epithelium [17] as temporal and anatomic correlates of the maturation of protection in SIVΔnef-vaccinated animals . Similarly , SIV-specific CD4 and CD8 T cell responses have also been implicated in SIVΔnef-induced protection [7 , 8 , 18–20] . Fukazawa and colleagues have correlated the magnitude of the T cell response in lymph nodes during live attenuated SIV vaccination , which in turn correlated with persistent residual replication of the vaccine virus in this site , with protection against wild-type challenge [19] . We have since shown that increased anentropic specificity , i . e . CD8 T cell responses targeting highly conserved epitopes , correlates with maturation of protection during SIVΔnef vaccination [20] . Most recently , we have demonstrated that low-level persistent SIVΔnef stimulation generates a unique signature of expression of transcription factors in SIV-specific CD8 T cells [21] . Finally , studies using a related model involving immunization with an attenuated SHIV vaccine have correlated protection against wild-type challenge with CD8 T cell responses in the female reproductive tract [22] . In this study we sought to delineate the temporal immune correlates of protection associated with maturation of immunity induced by SIVΔnef vaccination by employing a comprehensive approach involving both longitudinal studies and cross-sectional , intensive tissue sampling studies . We conducted two large vaccine trials using identical viral stocks and the same vaginal challenge route to determine the immune responses that correlate with protection against mucosal wild-type SIV challenge . The first study was a longitudinal vaccine trial designed to ascertain the temporal kinetics of SIVΔnef-induced protection against wild-type SIV vaginal challenge , and the second was a cross-sectional study to assay for immune responses in mucosal and lymphoid tissues during the SIVΔnef vaccination period and after SIV challenge . To resolve the contributions to protection of varying immune components , we have quantitated SIVmac239Δnef and wild-type SIVmac251 viral loads using a discriminating set of primers and simultaneously assayed for serum and vaginal antibody responses , determined the number of infecting viral variants after challenge using single genome analysis , and analyzed the functionality , phenotype , specificity and tissue localization of CD8 T cell responses . In these complementary vaccine studies , we demonstrate that the extent of SIVΔnef replication in gut and lymphoid tissues drives CD8 T cell polyfunctionality , localization of CD8 T cell responses to the tissues , antibody accumulation and affinity maturation , all of which correlate with protection against wild-type SIVmac251 vaginal challenge .
To study the temporal immune correlates of SIVΔnef-induced protection against wild-type , SIVmac251 challenge , we intravenously vaccinated 18 female macaques with SIVmac239Δnef . MHC class I alleles associated with the control of SIV infection such as Mamu A*01 [23] and Mamu B*17 [24] were evenly distributed among experimental groups ( S1 Table ) . Given that previous reports have documented increased protection during the first 20 weeks of SIVΔnef vaccination following intravenous challenge [5 , 12 , 13] , the animals were challenged intravaginally with SIVmac251 at 5 , 20 and 40 weeks post-vaccination to establish the kinetics of protection for this route of infection . As a control group , a total of nine unvaccinated animals were also intravaginally challenged with SIVmac251; three control animals were inoculated simultaneously with each of the three experimental groups ( Fig 1A ) . To discern between the viral loads of the SIVmac239Δnef vaccine and the SIVmac251 challenge virus , discriminating primers were used for the real-time quantitative PCR assay . Although all animals challenged 5 weeks after vaccination ( Group 5 animals ) were productively infected with wild-type SIV after intravaginal challenge with SIVmac251 , nearly half of the animals in week 20 and week 40 groups manifested apparent sterile protection against this stringent high dose wild-type challenge , as assessed by plasma viremia ( Fig 1B ) . Three of the animals vaccinated for 20 weeks and two of the animals vaccinated for 40 weeks with SIVΔnef displayed apparent sterile protection based on the lack of detectable wild-type plasma viremia after high-dose vaginal challenge . No such sterile protection was observed in the Group 5 animals , and all had measurable viral RNA as detected by RT-qPCR using SIVmac251-specific primers . Of the nine control animals challenged with SIVmac251 , eight were productively infected . Despite significantly lower peak viremia for infected animals in all vaccinated groups compared to infected animals in the control group ( p≤0 . 0061 ) ( Fig 2A ) , set-point plasma viral load , measured at week 8 post-challenge , was similar for the eight infected , unvaccinated control animals and the six animals challenged at week 5 post-vaccination , at 1 . 3x106 and 2 . 6x105 RNA copies per ml of plasma , respectively ( Fig 2A ) . In contrast , animals that were productively infected with SIVmac251 at week 20 or 40 post-vaccination had significantly lower set-point wild-type viral loads than control animals . At week 8 post-challenge , SIVmac251-infected animals in Groups 20 and 40 had an average set-point viremia of 5 . 9 X 103 and 8 . 9 X 102 RNA copies per ml of plasma , respectively ( p = 0 . 024 and 0 . 002 ) ( Fig 2A ) . Similarly , chronic viral loads in infected Group 20 and 40 animals , measured at weeks 16 and 20 , were significantly lower compared to the Control Group ( p≤0 . 0083 and p≤0 . 024 , respectively ) ( Fig 2A ) . A linear mixed model of post-challenge viral loads was used to determine the statistical significance of SIVΔnef vaccination for the three experimental groups compared to the control group , encompassing all animals whether infected or uninfected by the challenge virus . The linear mixed model allows for comparison between different vaccination groups based on the fixed effect of the duration of the vaccination period prior to pathogenic SIVmac251 challenge ( 5 , 20 or 40 weeks ) , while taking effects such as animal genetic background into account . Statistical analysis was performed for three phases over the course of viral infection: peak viremia , viral set-point ( weeks 5 to 12 ) and chronic infection ( weeks 13 to 22 ) . All three vaccinated groups had significantly lower peak viremia than the unvaccinated controls ( p≤ 0 . 013 ) ( Fig 2B ) . However , the Group 5 animals demonstrated only a transient reduction in viremia compared to unvaccinated animals and had comparable levels of viremia during the viral set-point and chronic phases of infection ( Fig 2B ) . In contrast , Group 20 and Group 40 animals , including both partially and sterilely protected animals , had a significantly lower plasma viremia than the control group during acute infection ( p<0 . 008 ) , viral set-point ( p<0 . 0001 ) and chronic infection ( p≤0 . 004 ) . We defined protection from SIVmac251 challenge in a given group as a greater than two-log viral load decrease in the first 20 weeks post-challenge compared to the control group . The protection of Group 20 and 40 animals but not Group 5 vaccinees demonstrates that SIVΔnef-induced protection is temporally determined and that a longer duration of vaccination correlates with protection from wild-type pathogenic mucosal challenge . Having demonstrated disparate impact on plasma viremia among the different vaccination groups , we next asked whether the different levels of protection were also manifest as sieving effects on viral variants established in the disseminated systemic infection . Taking advantage of the fact that our challenge virus , SIVmac251 , is a quasispecies and not a clone , we used single genome amplification [25] to estimate the number of transmitted viral variants that established the initial disseminated infection manifested as plasma viremia . Plasma viral RNA was reverse transcribed and the env gene in its entirety was amplified and sequenced for every animal . The sequences were then plotted as phylogenetic trees to determine the minimum number of viral variants that were transmitted and established in each animal . SIVΔnef vaccination had a significant effect on the number of SIVmac251 viral variants established after challenge . Whereas unvaccinated control animals had an average of six viral variants established after challenge , vaccinated animals with detectable wild-type SIV plasma viremia had an average of three established variants , a significant narrowing of the genetic bottleneck between unvaccinated and vaccinated animals ( p = 0 . 0097 ) ( Fig 3A ) . However , although SIVΔnef vaccination reduced the number of established variants compared to unvaccinated controls , the number of established variants was not inversely correlated with time from vaccination , and thus the reduction in established viral variants was not correlated with the temporal maturation of protection . Rather , the average number of variants establishing productive viral replication in infected animals was strikingly similar among the three vaccine groups , an average of 3 . 3 viral variants for Group 5 animals , 3 . 3 viral variants for Group 20 animals and 3 . 5 viral variants for Group 40 animals ( Fig 3B ) . The number of founder variants in the vaccinated animals across all groups did inversely correlate with the time to peak viremia ( p<0 . 0001 ) ( Fig 3C ) . This result , and the significant reduction in peak viremia even at 5 weeks , is consistent with protective mechanisms in vaccinated animals with some immediate impact on constraining systemic infection ( and thus delaying peak viremia ) that , nonetheless , require maturation for sustained effects on established systemic infection seen in set-point reductions at 20 and 40 weeks . SIVΔnef induced significant IgG antibody responses in vaccinated animals as early as week 5 after vaccination . Serum IgG binding antibodies against Env and Gag-Pol rose quickly in the first 5 weeks after vaccination and continued to rise , albeit gradually , into week 40 ( Fig 4A ) . Similarly , vaginal IgG antibody responses to Env and Gag-Pol rose precipitously in the first 5 weeks after vaccination and continued a slow increase into week 40 ( Fig 4B ) . In both plasma and vaginal secretions , the SIV-specific antibody response was dominated by IgG responses , which were 2–3 logs higher than the SIV-specific IgA responses . The avidity of Env-specific serum IgG and IgA responses ( Fig 4C ) , as well as the concentrations of anti-Env serum IgG responses ( S1A and S1B Fig ) , were lower in Mamu-A*01+ animals compared to animals that did not express Mamu-A*01 , probably due to the overall lower viral load levels achieved in Mamu-A*01+ animals during the vaccination period ( S1C Fig ) . In addition to binding antibodies and avidity , we also analyzed the ability of serum antibodies to inhibit infection of CEM cells by a neutralization-sensitive , T cell line adapted ( TCLA ) stock of SIVmac251 . The titers of TCLA neutralizing antibodies in the serum of SIVΔnef-vaccinated animals doubled between weeks 5 and 20 and quadrupled between weeks 5 and 40 post-vaccination ( Fig 4D ) . Serum TCLA neutralizing antibody titers also directly correlated with the cumulative antigenic load , calculated as the area under curve ( AUC ) of the plasma viral load ( S2 Fig ) . The TCLA neutralizing antibody titers in vaccinated macaques at time of challenge inversely correlated with peak SIVmac251 viremia after vaginal challenge ( p = 0 . 011 ) ( Fig 5A ) . While TCLA neutralizing antibody titers were significantly different between unprotected animals and those with apparent sterile protection , neutralizing antibody titers on the day of challenge were not significantly different between unprotected animals and partially protected animals , nor were they significantly different between partially protected animals and sterilely protected animals ( Fig 5B ) . However , there was a trend toward higher TCLA neutralizing antibody titers between unprotected and partially protected animals and likewise between partially and sterilely protected animals . Although levels of anti-Env IgA binding antibodies in serum and vaginal secretions did not correlate with peak viral load post-challenge ( S3 Fig ) , post-challenge peak SIVmac251 viremia was inversely correlated to serum anti-Env IgG concentrations ( p = 0 . 017 ) ( Fig 5C ) and demonstrated an even stronger inverse relationship to anti-Env IgG specific activity in vaginal secretions ( p = 0 . 0027 ) ( Fig 5D ) . The avidity of anti-Env serum antibodies on the day of challenge was also significantly greater in sterilely protected animals ( Fig 5E & 5F ) , and it inversely correlated with peak viremia ( Fig 5G & 5H ) , indicating that higher quality antibody responses were generated in both partially and sterilely protected animals . The inverse correlation between peak viremia and multiple parameters of SIV-specific antibody titers in both serum and the vaginal mucosa strongly support a role for humoral responses in SIVΔnef-induced protection . Along with robust binding and TCLA neutralizing antibody responses , SIVΔnef also induced robust SIV-specific CD4 and CD8 T cell responses , as measured in blood . Following vaccination with SIVΔnef , plasma viremia peaked at week 2 post-inoculation and the total CD8 T cell response peaked at week 5 as measured by the IFN-γ ELISpot assay and declined following control of viremia ( Fig 6A ) . However , although total SIV-specific CD8 T cell responses decreased between weeks 5 and 20 , responses against the highly conserved Gag protein were maintained over time ( Fig 6A ) , highlighting the importance of antigen specificity for CD8 T cell response persistence . Gag-specific CD8 T cell responses were also shown to be maintained between weeks 5 and weeks 20 and 40 as measured by IFN-γ and TNF-α expression using intracellular cytokine staining ( ICS ) assays , in addition to a significant increase in IL-2 production ( p = 0 . 016 ) ( Fig 6C ) . In contrast , IFN-γ- , TNF-α- and IL-2-producing Gag-specific CD4 T cell responses increased significantly between weeks 5 and 20 ( p≤0 . 0067 ) ( Fig 6B ) , indicating a significant upsurge in the functionality and magnitude of Gag-specific CD4 T cell responses at week 20 . Similar trends of increased cytokine production by Gag-specific CD4 T cells between weeks 5 and 40 were also observed but were not statistically significant due to the small number of animals at week 40 . To ascertain the polyfunctionality of the T cell response at weeks 5 and 20 , we stimulated peripheral blood mononuclear cells ( PBMC ) with overlapping Gag peptides and analyzed IFN-γ , TNF-α and IL-2 production . Using Boolean gating , we evaluated the functionality of Gag-specific CD4 T cells by determining the percentage that concurrently expressed all three cytokines or two of the three cytokines at weeks 5 , 20 and 40 post-vaccination . The percentage of Gag-specific CD4 T cells expressing all three cytokines measured increased significantly between weeks 5 and 20 from 1 . 5 to 10 and between weeks 5 and 40 from 1 . 5 to 8 . 3 and , likewise , the percentage of Gag-specific CD4 T cells expressing 2 of the 3 cytokines assayed rose significantly from 12% at week 5 to 31% at week 20 and to 29% at week 40 ( p≤0 . 043 ) ( Fig 6D ) . Increased functionality in SIV-specific responses was also observed in CD8 T cells where the tri-functional Gag-specific CD8 T cells , coordinately expressing TNF-α , IFN-γ and IL-2 , increased from 0 . 6% to 3 . 4% of total responses between weeks 5 and 20 ( p = 0 . 023 ) and remained high at 3 . 8% up to week 40 ( p = 0 . 0053 ) . Similarly , Gag-specific CD8 T cell responses expressing 2 of the 3 cytokines comprised 11% of the response at week 5 and rose to 29% and 31% at weeks 20 ( p = 0 . 0072 ) and 40 ( p = 0 . 0020 ) , respectively ( Fig 6E ) . Thus , in contrast to the declining magnitude of the total SIV-specific CD8 T cell response as detected by ELISpot assays to the entire SIV proteome , the magnitude of Gag-specific CD4 and CD8 T cell responses was maintained throughout the vaccination period . In addition , both Gag-specific CD4 and CD8 T cell responses increased in functionality over the forty-week SIVΔnef vaccination phase . Given that the magnitude of Gag-specific CD8 T cell responses as measured by cytokine production was maintained during the first 40 weeks of SIVΔnef vaccination , we sought to locate SIV-specific CD8 T cell responses within the female reproductive tract , the route of challenge; the gut mucosa , the anatomical site most enriched for activated CD4 memory T cells that are the most permissive target for SIV; and the secondary lymphoid tissues , where antigen-processing and presentation are most likely to occur after infection . In order to quantitate the localization , the magnitude and the specificity of SIV-specific CD8 T cells in the tissues during the SIVΔnef vaccination phase and the subsequent post-challenge phase , we conducted a follow-up cross-sectional serial sacrifice study . In the follow-up study , a total of 26 animals were vaccinated with SIVΔnef , of which 1 to 4 animals were sacrificed at days 4 , 7 , 11 , 14 , 35 and 140 post-vaccination . The remaining animals were challenged with SIVmac251 intravaginally and 2 to 4 macaques were sacrificed at days 4 , 5 , 7 , 11 and 14 post-challenge ( Fig 7 ) . Having demonstrated that animals vaccinated for 5 weeks with SIVΔnef were not protected compared to animals vaccinated for 20 or 40 weeks in the longitudinal study , we comprehensively examined animals at weeks 5 and 20 post-vaccination to determine the tissue localization of SIV-specific CD8 T cell responses . Tissues from secondary lymphoid tissues , including lymph nodes and the spleen; the gut mucosa , including the jejunum , the ileum and the colon; and the female reproductive tract , including the cervix and the vagina , were processed and stained for CM9-MHC tetramer+ and SL8-MHC tetramer+ CD8 T cells . CM9 is a highly conserved epitope within the Gag protein , while Tat SL8 is a highly variable epitope that escapes CD8 T cell responses mounted against it frequently and rapidly [20 , 26] . These two epitopes on opposite extremes of the immune evasion spectrum were chosen to determine if epitope entropy , a measure of an epitope’s mutational flexibility , plays a role in the maintenance of its CD8 T cell response in the tissues . Overall , the frequency of CM9-specific CD8 T cell responses was maintained between weeks 5 and 20 in most tissues processed ( Fig 8A ) . On the other hand , SL8-specific CD8 T cell response frequencies declined in every tissue processed ( Fig 8B ) , confirming that epitope escape led to the waning of responding CD8 T cells in the tissues and that ongoing antigenic stimulation is necessary to maintain SIV-specific T cells in both lymphoid and non-lymphoid tissues . Looking at fold change of tetramer+ CD8 T cells at week 20 over week 5 , SL8-specific CD8 T cell responses ranged between 0 . 1 and 0 . 65 in all tissues , signifying a 35–90% decrease between weeks 5 and 20 , whereas CM9-specific CD8 T cell responses persisted at similar levels throughout most sites , with the notable exception of peripheral blood , where CM9-specific CD8 T cell responses experienced a similar drop to that of SL8-specific CD8 T cells ( Fig 8C ) . We also noted that the ileum , the jejunum , the colon and the mesenteric lymph nodes had increased frequencies of CM9-specific CD8 T cells from week 5 to week 20 post-vaccination . While CM9-specific CD8 T cell responses decreased significantly in peripheral blood ( p = 0 . 014 ) , their frequencies were maintained in secondary lymphoid tissues ( p = 0 . 51 ) and the female genital tract ( p = 0 . 81 ) and increased significantly in the gut mucosa ( p = 0 . 001 ) ( Fig 8D ) . Notably , the frequency of CM9-specific CD8 T cells was significantly higher in the female genital tract than it was in secondary lymphoid tissues at both week 5 ( p = 0 . 0087 ) and week 20 ( p<0 . 0001 ) . To determine if the maintenance of CM9-specific CD8 T cell responses in secondary lymphoid tissues , the gut mucosa and the female genital tract was due to SIVΔnef replication in these tissues , we measured the cell-associated SIV RNA in these tissues at week 20 . As has been previously shown , SIVΔnef replication persisted at relatively high levels in secondary lymphoid tissues ( Fig 8E ) . Interestingly , the jejunum and the colon had levels of cell-associated viral RNA comparable to secondary lymphoid tissues . In contrast , peripheral tissues like the female genital tract , the lungs , liver and the rectum had very low cell-associated viral loads ( Fig 8E ) . The presence of high frequencies of SIV-specific CD8 T cell responses in the gut , approaching 3% of total CD8 T cells for CM9-specific cells alone , highlighted the potential vaccination-induced protection of this important immune compartment and site of massive CD4 T cell depletion during wild-type SIV infection in unvaccinated animals . We phenotyped the CD4 T cell population in the jejunum , colon and mesenteric lymph nodes , all sites previously demonstrated to undergo massive depletion of CCR5+ , central memory CD4 T cells during acute SIV infection [27] . In SIVΔnef-vaccinated animals subsequently challenged with wild-type SIVmac251 at 20 weeks post-vaccination , the frequency of CCR5+ CD28+ CD95+ CD4+ T cells did not decrease in the jejunum , the colon or the mesenteric lymph nodes between days 0 and 14 post-challenge ( Fig 9 ) . Nor was there a decrease of total memory CD4 T cells ( CD95+ CD4+ ) in these tissues ( S4A Fig ) during the first 2 weeks post-challenge . Finally , the number of memory CD4 T cells as a percentage of CD3+ T cells in peripheral blood during the first two weeks was not different between uninfected and infected animals with the post-challenge SIVmac251 virus ( S4B Fig ) , highlighting the prevention of CD4 T cell depletion in the gut after challenge even in animals that are partially , but not sterilely , protected . Having demonstrated that SIV-specific CD8 T cell responses were maintained in the tissues , we phenotyped these responses using memory and functional markers over the 20-week vaccination period . Secondary lymphoid tissues and female genital tract CM9-specific CD8 T cells expressed higher levels of CD28 on the cell surface at week 20 than at week 5 ( p = 0 . 0005 , p = 0 . 001 ) , denoting a shift to a more quiescent memory phenotype at week 20 ( Fig 10A ) . Although CD28 expression levels in gut mucosa CM9-specific CD8 T cells were already high at week 20 , there was no significant increase in the levels of CD28 expression in gut mucosa CM9-specific CD8 T cells between weeks 5 and 20 , suggesting differences in the extent of antigenic stimulation in the different tissues ( Fig 10A ) . In contrast , CCR7 expression levels in CM9-specific CD8 T cells remained relatively constant across all tissues and peripheral blood between weeks 5 and 20 ( Fig 10B ) . Furthermore , in line with a shift from an effector memory cell to a more quiescent memory CD8 T cell response from week 5 to week 20 , intracellular perforin levels in CM9-specific CD8 T cells decreased significantly between weeks 5 and 20 across all tissue compartments and peripheral blood ( p<0 . 0001 ) ( Fig 10C ) . Increased localization of CM9-specific CD8 T cells from the periphery to the tissues between weeks 5 and 20 ( Fig 8D ) was correlated with significantly increased expression of CXCR3 in CM9-specific CD8 T cells in all tissues and peripheral blood ( p≤0 . 0124 ) ( Fig 10D ) . To determine if the sustained levels of CM9-specific CD8 T cells in the tissues were due to cellular proliferation , we assayed the intracellular expression of the proliferation marker Ki-67 . While the expression of Ki-67 fell significantly in CM9-specific CD8 T cells in peripheral blood between weeks 5 and 20 ( p = 0 . 0033 ) , it declined less significantly in secondary lymphoid tissues ( p = 0 . 0146 ) and did not decrease in the gut mucosa and the female genital tract ( Fig 10E ) . Comparing Ki-67 expression levels at week 20 for the escaped SL8-specific CD8 T cell response and the CD8 T cell response specific to the highly conserved CM9 , we observed similar percentages of cells expressing Ki-67 in peripheral blood . However , Ki-67 expression was significantly higher in CM9-specific CD8 T cells than in SL8-specific CD8 T cells in the secondary lymphoid tissue ( p = 0 . 0058 ) and the gut mucosa ( p = 0 . 02 ) ( Fig 10F ) . The higher level of Ki-67 in tissue-resident CM9-specific CD8 T cells , but not CD8 T cells targeting SL8 , likely reflects continued viral replication in the gut mucosa and the secondary lymphoid tissues that may explain both the maintenance of CM9-specific CD8 T cells . A number of recent reports have identified a distinct population of antigen-specific tissue resident CD8 T memory cells that do not recirculate and provide long-lived immunity against a variety of pathogens [28–30] . Although definitive phenotypic markers for tissue-resident memory have not yet been identified , expression of CD69 has been consistently observed in multiple studies [31 , 32] . Upregulation of the activation marker CD69 has been demonstrated to block tissue egress of tissue-resident memory CD8 T cells by interfering with sphingosine-1-phosphate receptor ( S1PR1 ) [33] , and tissue-resident memory CD8 T cells express high levels of CD69 in the absence of recent activation [34–36] . To determine if SIV-specific CD8 T cells in nonlymphoid tissues displayed characteristics of stable populations of CD8 TRM , we phenotyped CM9- and SL8-specific CD8 T cells for CD69 expression . By week 20 after SIVΔnef vaccination , almost all CM9- and SL8-specific CD8 T cells in the jejunum , the colon , the ileum and the vagina upregulated CD69 , in contrast to CD8 T cell responses in secondary lymphoid tissues and peripheral blood which were largely CD69- ( Fig 11 ) . Importantly , CD69 upregulation was not dependent on antigenic exposure , as the SL8 epitope is generally entirely escaped by week 5 [20 , 26] nor was it a sign of CD8 T cell proliferation , as SIV-specific CD8 T cells in vaginal tissue demonstrated low Ki-67 expression at weeks 5 and 20 ( Fig 10E , S5 Fig ) . Although SL8-specific CD8 T cells in gut and vaginal tissues also upregulated CD69 expression , the overall magnitude of the SL8 response dropped significantly in these tissues between weeks 5 and 20 ( Fig 8C ) .
The combined data from these two SIVΔnef vaccine studies offer a number of novel findings that demonstrate the complex interplay of adaptive immune responses that ultimately result in the protection of SIVΔnef-vaccinated animals from wild-type vaginal challenge . The longitudinal vaccine study illustrates the protective effect of SIVΔnef vaccination against a minimally heterologous , high dose vaginal challenge with SIVmac251 . SIVΔnef vaccine-induced protection against vaginal challenge , whether sterile or partial , was temporally dependent , increasing during the first 20 weeks of vaccination and sustained to 40 weeks post-vaccination . This time course is in contrast to SIVΔnef-induced protection against homologous intravenous challenge with SIVmac239 , in which apparent sterile protection was observed as early as 5 weeks after vaccination [37] . SIVΔnef vaccination created a genetic bottleneck by reducing the number of established wild-type viral variants , which inversely correlated with a longer time-to-peak viremia . Both neutralizing and binding antibody titers increased during the vaccination period , and titers at time of challenge inversely correlated with peak viral load , while Gag-specific CD4 and CD8 T cell functionality increased in peripheral blood . The cross-sectional vaccine study reveals the preservation of CM9-specific CD8 T cell responses in high viral-load tissues , primarily in secondary lymphoid tissues and the gut mucosa , even as SL8-specific CD8 T cell responses waned , highlighting the importance of ongoing antigenic stimulation to maintain tissue-resident memory cells . Moreover , the persistent expression of Ki-67 in CM9-specific CD8 T cells in the secondary lymphoid tissues and the gut mucosa correlated with the maintenance of responses in these tissues , reinforcing the link between persistent local replication of the vaccine virus , associated antigenic stimulation and persistence of immune responses . In vaginal tissues , high CD69 expression coupled with low Ki-67 expression by SIV-specific CD8 T cells in the context of low viral replication strongly suggests the maintenance of a stable tissue-resident memory CD8 T cell population . Taken together , these results document induction of a highly functional immune response by SIVΔnef , employing each arm of the adaptive immune response and localizing to key sites of potential CD4 T cell infection . Given the high efficiency of SIV infection of naïve controls following vaginal challenge using the SIVmac251 stock and challenge conditions employed in this study , which is consistent with previous studies with this stock [38 , 39] , the apparent sterile protection observed in the Group 20 and 40 animals is likely to represent a vaccine-mediated effect reflecting the maturation of protective cellular and humoral immune responses . Other potentially confounding factors , such as the menstrual cycle of the animals , appear unlikely to affect our results . Although there is evidence that the timing of the menstrual cycle may affect the susceptibility of macaques to low-dose SIV or SHIV challenge [40 , 41] , this effect has not been demonstrated in the setting of high dose SIV vaginal challenge . Given these considerations , as well as the fact that our animals were likely to be cycling asynchronously , it appears improbable that the differences in protection among the different experimental groups could be attributed to variations in the menstrual cycle of individual animals . The effects of SIVΔnef vaccination on post-challenge virological outcome may be categorized as acute virological suppression , sustained virological suppression , and apparent sterilizing immunity . Acute virological suppression was characterized by a lower peak viral load , fewer established viral variants and a prolonged time-to-peak viremia , but was not temporally dependent . The reduction of viral variants was not sufficient to induce protection from challenge , since all groups of vaccinated animals had similar levels of established viral variants . By significantly decreasing peak viremia , perhaps a result of the decreased established number of viral variants , SIVΔnef limited the initial impact of wild-type SIV infection even at early time points after vaccination . At later time points , protection against wild-type SIV vaginal challenge , whether partial or sterile , correlated with immune maturation of humoral and cellular immune responses . First , neutralizing antibody titers against TCLA SIV at day of challenge inversely correlated with peak wild-type SIV viral load . Second , in Mamu A*01+ animals the magnitude of the CM9-specific CD8 T cell response was largely maintained across most tissues during the vaccination period , even as SL8-specific responses declined to the limit of detection in all tissues assayed . This shift in specificity of the CD8 T cell response corroborates our previous results demonstrating increased anentropic specificity , characterized by the accumulation of CD8 T cell responses to highly conserved epitopes between weeks 5 and 20 post-SIVΔnef vaccination [20] . Furthermore , the difference in the persistence of CM9- and SL8-specific CD8 T cell responses can be attributed in part to increased proliferation of CM9-specific CD8 T cells . These results reinforce the concept that ongoing viral replication of SIVΔnef maintains the antigenic stimulation of CD8 T cell responses directed at conserved epitopes . In addition to the shift in specificity refocusing the immune response on more conserved epitopes during the vaccination period , the CD8 T cell response was redistributed among tissues by week 20 . During this time , the magnitude of CM9-specific CD8 T cells significantly increased in the gut mucosa and was maintained in secondary lymphoid tissues and the female genital tract , while CM9-specific CD8 T cell responses in peripheral blood significantly declined . At week 20 , Ki-67 expression was significantly higher in CM9-specific CD8 T cells than in SL8-specific CD8 T cells in the gut mucosa and in secondary lymphoid tissues , suggesting that the observed viral replication in these tissues stimulated CM9-specific CD8 T cell proliferation . Meanwhile , tissue-resident memory CD8 T cells , which expressed high levels of CD69 and low levels of Ki-67 , populated the vaginal tissues at week 5 and 20 post-SIVΔnef vaccination . Residual viral replication in gut and lymphoid tissues is likely to account for the transitional memory phenotype of CM9-specific CD8 T cells compared to SL8-specific CD8 T cells . Fukazawa and colleagues have previously reported that the magnitude of T cell responses in secondary lymphoid tissues in animals vaccinated with different LASIV strains predicts protection from wild-type SIV challenge [19] . Our current work extends these findings to the gut mucosa . Comprehensive SIVΔnef RNA quantitation in tissues demonstrated that the gut mucosa , particularly the jejunum and colon , were important sites of SIVΔnef replication at 20 weeks after vaccination , producing comparable levels of viral RNA to secondary lymphoid tissues . The same factors that contribute to increased SIV replication in lymph nodes , namely the presence of B cell follicles , which contain easily infectable follicular CD4 T helper cells [19] and represent an immuneprivileged site impeding entry of SIV-specific CD8 T cells [42] , are also found in Peyer’s patches in the lamina propia of the gut mucosa [43] . Sites of viral replication such as B cell follicles that are relatively protected from immune surveillance may be responsible for the ongoing antigenic stimulation required for protection induced by SIVΔnef . In addition to shifting the specificity and tissue localization of the CD8 T cell response , persistent replication of SIVΔnef is likely to promote increased production and affinity maturation of antiviral antibodies , resulting in higher antibody affinity and higher neutralizing antibody titers . Antibody responses also directly correlated with SIVΔnef viral load within a given vaccination group , indicating that antigenic load and viral persistence play a key role in driving the induction and the affinity maturation of the antibody response . Several observations indicate antibodies play an important role in preventing or suppressing replication of wild-type challenge virus in SIVΔnef-vaccinated animals . SIVΔnef-vaccinated animals with sterile protection had higher avidity Env-specific antibodies in serum on the day of challenge , and in all vaccinated animals , antibody avidity was inversely associated with peak post-challenge viremia . Serum anti-Env IgG binding antibody levels and TCLA neutralizing titers also demonstrated a strong inverse correlation with peak plasma viremia , consistent with prior observations of SIV-specific ADCC activity in this cohort of SIVΔnef-vaccinated animals [15] . Moreover , anti-Env binding antibody titers in the vaginal mucosa showed an even stronger inverse correlation with peak viral load post-challenge , suggesting that SIV-specific antibody responses inhibit viral transmission and spread in the female reproductive tract . The importance of SIV-specific responses localized to the female reproductive tract in mediating protection against vaginal challenge induced by SIVΔnef is also reinforced by our observations of a temporal maturation of antibodies to trimeric gp41 that are concentrated by the neonatal Fc receptor ( FcRn ) in the vaginal mucosa and cervical epithelium prior to [17] and following challenge [44] . Our results , suggesting a role for antibody responses during the early stages of wild-type SIV vaginal challenge , corroborate our previous findings [15] that ADCC activity in SIVΔnef-vaccinated animals exhibited a trend towards a negative correlation with post-challenge peak viral load . While our previous study did show a significant relationship between ADCC and apparent sterile protection , we did not detect any neutralizing antibody responses against the difficult-to-neutralize SIVmac251 stock in SIVΔnef-vaccinated animals on the day of challenge . In the present study , however , neutralizing antibodies against TCLA SIVmac251 did significantly correlate with early viral control post-SIV vaginal challenge , in addition to being significantly higher in sterilely protected animals than unprotected animals . Collectively the results of our longitudinal and cross-sectional analysis of the maturation of protection are consistent with mechanisms of immune control that operate at different stages of infection . In the early stages of infection following wild-type SIV vaginal challenge , the immediate effects of antibody responses and CD8 T cell responses are likely to account for the reduction of viral variants and the delay in time-to-peak viremia . However , sterilizing immunity or significant reduction in set point plasma viremia require an immunologically matured response that impacts systemic infection . At subsequent stages when virus has disseminated to the gut and secondary lymphoid tissues , control of systemic infection requires a functionally mature CD8 T cell response focused on conserved epitopes . This pattern of immune protection induced by SIVΔnef differs significantly from that induced by recombinant CMV vectors , which induce early control and subsequent clearance of SIV infection in about 50% of animals but do not result in reduction in set-point viremia in vaccinated animals with established infection [45 , 46] . Studies on SIVΔnef , the most effective vaccine to date against lentiviral challenge , continue to provide important insights as to how to design a successful HIV vaccine . SIVΔnef persistence drives the response specificity to conserved epitopes , redistributes the CD8 T cell responses to CD4-rich gut and lymphoid tissues , replenishes the SIV-specific resident memory CD8 T cell population in vaginal tissues , and induces increased antibody production and affinity maturation . Future studies on SIVΔnef will be required to tease out the relative contributions of CD8 T cell responses and neutralizing antibodies to apparent sterilizing immunity and long-term viral control . Taken together , these findings highlight the importance of developing clinically applicable vaccine strategies to provide ongoing antigenic stimulation that induces the maturation and localization of both humoral and cellular responses able to prevent lentivirus infection .
The animals included in this study were all female Indian-origin rhesus macaques ( Macaca mulatta ) , housed in a biocontainment facility at the New England Primate Research Center ( NEPRC ) . MHC class I genotypes were determined by sequence-specific PCR [47] . These experiments and procedures were approved by the Harvard Medical Area Institutional Animal Care and Use Committee ( the Harvard Medical Area Standing Committee on Animals ) , protocol 04383 . The Harvard Medical School animal management program is accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care , International ( AAALAC ) , and meets National Institutes of Health standards as set forth in the 8th edition of the Guide for the Care and Use of Laboratory Animals [48] . The institution also accepts as mandatory the PHS Policy on Humane Care and Use of Laboratory Animals by Awardee Institutions and NIH Principles for the Utilization and Care of Vertebrate Animals Used in Testing , Research , and Training . There is on file with the Office of Laboratory Animal Welfare ( OLAW ) an approved Assurance of Compliance ( A3431-01 ) . All animals were housed indoors in an SOP-driven , AAALAC-accredited facility . Husbandry and care met the guidance of the Animal Welfare Regulations , OLAW reporting and the standards set forth in The Guide for the Care and Use of Laboratory Animals . All research animals were enrolled in the NEPRC behavioral management program , including an IACUC-approved plan for Environmental Enrichment for research primates . This program included regular behavioral assessments , and provision of species appropriate manipulanda , and foraging opportunities . This protocol had an IACUC-approved exemption from social housing based on scientific justification . Primary enclosures consisted of stainless steel primate caging provided by a commercial vendor . Animal body weights and cage dimensions were regularly monitored . Overall dimensions of primary enclosures ( floor area and height ) met the specifications of The Guide for the Care and Use of Laboratory Animals , and the Animal Welfare Regulations ( AWR's ) . Further , all primary enclosures were sanitized every 14 days at a minimum , in compliance with AWRs . Secondary enclosures ( room level ) met specifications of The Guide with respect to temperature , humidity , lighting and noise level . The animals were provided ad lib access to municipal source water , offered commercial monkey chow twice daily , and offered fresh produce a minimum of three times weekly . Light cycle was controlled at 12/12 hours daily . The animals were subject to twice daily documented observations by trained animal care and veterinary staff , and enrolled in the facility's environmental enrichment , and preventative health care programs . Euthanasia took place at defined experimental endpoints using protocols consistent with the American Veterinary Medical Association ( AVMA ) guidelines . Animals were first sedated with intramuscular ketamine hydrochloride at 20 mg/kg body followed by sodium pentobarbital ( ≥100 mg/kg ) intravenously to achieve euthanasia . Animals were intravenously vaccinated with SIVmac239Δnef using either 5 ng or 25 ng of SIVp27 . Vaginal challenges were carried out using 1 . 0 ml of SIVmac251 ( kindly provided by Chris Miller , University of California-Davis ) inoculated twice in a single day , with a 4 hour interval between inoculations . This stock of SIVmac251 ( SIVmac251-CM ) contains approximately 2 × 109 viral RNA copies/ml and approximately 5 × 104 50% tissue culture infectious dose 50% ( TCID50 ) /ml [49] . This vaginal challenge regimen and stock has previously resulted in highly efficient infection of naïve controls [38 , 39] . At the time of vaginal challenge , animals were singly housed and assumed to be cycling asynchronously . Due to the requirement of performing vaginal challenges at defined times after SIVΔnef vaccination , it was not possible to perform vaginal challenges in relation to the menstrual cycle of individual animals . For quantitation of plasma viral loads in SIVΔnef-infected animals , highly specific , real-time RT-PCR assays were performed as described previously [50] . The assay specific for SIVmac239Δnef was developed by designing a reverse primer that recognizes the new sequence generated by the deletion of the nef gene . Viral RNA loads for animals infected with SIVmac251 were determined through highly specific , real-time RT-PCR assays using a specific reverse primer , which binds to the nef sequence in SIVmac251 but not SIVmac239Δnef . The nominal threshold for these assays was 30 viral RNA copy equivalents/ml plasma . Tissue viral load analysis was conducted as previously described [45] . Single genome amplification to determine the number of transmitted viruses was conducted as described previously [25 , 51] . Briefly , RNA was isolated using viral RNA extraction kit ( Qiagen ) and cDNA synthesized using gene specific priming and super script III ( Life Technology ) . The cDNA was diluted in 96-well plates such that fewer than 29 PCRs yielded an amplification product . First-round PCR was carried out in 1× High Fidelity platinum PCR buffer , 2 mM MgSO4 , 0 . 2 mM of each deoxynucleoside triphosphate , 0 . 2 μM of primers in vif and nef , and 0 . 025 U/μl platinum Taq High Fidelity polymerase ( Invitrogen , Carlsbad , CA ) in a 20-μl reaction mixture . The PCR mixtures were set up in MicroAmp optical 96-well reaction plates ( Applied Biosystems , Foster City , CA ) and sealed with ABI MicroAmp adhesive film . The following PCR conditions were used: 94°C for 2 min followed by 35 cycles of 94°C for 15 s , 55°C for 30 s , and 68°C for 4 min , with a final extension of 68°C for 10 min . Second-round PCR was carried out using 1 μl of the first-round product and 0 . 2 μM of a primer set spanning the env gene with the same PCR mixture as the first round . The PCR conditions included: 94°C for 2 min followed by 45 cycles of 94°C for 15 s , 55°C for 30 s , and 68°C for 4 min , with a final extension at 68°C for 10 min . The amplicons were sized on precast 1% agarose E-gel 96 ( Invitrogen Life Technologies , Carlsbad , CA ) . All products derived from cDNA dilutions yielding less than 30% PCR positivity were sequenced . Viral env genes were sequenced by using BigDye Terminator chemistry and the protocols recommended by the manufacturer ( Applied Biosystems , Foster City , CA ) . The sequences were determined by using an ABI 3730xl genetic analyzer ( Applied Biosystems , Foster City , CA ) and edited by using the Sequencher program , version 4 . 7 ( Gene Codes , Ann Arbor , MI ) . Both strands of DNA were sequenced . All chromatograms were carefully inspected for sites of ambiguous sequence ( double peaks ) , and those that contained one or more positions of mixed bases were excluded from further analysis . ELISA was used to measure antibodies in serum and in vaginal secretions collected with Weck-Cel sponges as previously described [52 , 53] . All samples intended for IgA analysis were first depleted of IgG using Protein G sepharose ( GE Healthcare ) as described [53] . Briefly , microtiter plates were coated with either recombinant SIVmac239 gp130 envelope ( Env ) protein ( ImmunoDiagnostics , Woburn , MA ) or SIVmac239 viral lysate ( Advanced Biotechnologies Inc , Columbia , MD ) . Because the lysate lacks detectable Env protein at the 1/400 coating dilution used , antibodies against it are referred to as being Gag , Pol-specific . Serial dilutions of samples and previously described macaque serum standards [52] were reacted overnight at 4°C with coated/blocked plates . Plates were developed by treatment with biotinylated polyclonal goat anti-human IgG ( SouthernBiotech , Birmingham , AL ) or–monkey IgA ( Open Biosystems , Huntsville , AL ) , followed by avidin peroxidase , and tetramethylbenzidine ( Sigma ) . Total IgA and IgG concentrations in secretions were measured by ELISA as previously described [54] . Concentrations of SIV Env- or SIV Gag , Pol-specific IgA and IgG were divided by the concentration of total IgA and IgG , respectively , to obtain the specific activity ( ng IgA or IgG antibody per μg total IgA or IgG ) . The avidity of anti-Env antibodies in serum was measured using plates coated with gp130 and a NaSCN displacement ELISA modeled after that described by Vermont et al . [55] . The avidity index was calculated by dividing the concentration of anti-gp140 IgG or IgA measured in 1 . 5M NaSCN-treated wells by that in untreated wells on the same plate . Antibody-mediated neutralization of T-cell-line-adapted SIVmac251 ( TCLA SIV251 ) was assessed in a CEMx174 cell killing assay as previously described [16] . Cell-free stocks of TCLA SIV251 prepared in H9 cells were added in triplicate to multiple dilutions of test serum in 100 μl of RPMI 1640–12% fetal bovine serum containing 50 μg gentamicin in 96-well culture plates . After incubation for 1 h at 37°C , CEMx174 cells ( 5 × 104 cells in 100 μl ) were added to each well . Infection led to extensive syncytium formation and virus-induced cell killing in approximately 4 to 6 days in the absence of antibodies . Neutralization was measured by staining viable cells with Finter's neutral red in poly-l-lysine-coated plates . The percent protection was determined by calculating the difference in absorption ( A540 ) between test wells ( cells plus serum sample plus virus ) and virus control wells ( cells plus virus ) , dividing this result by the difference in absorption between cell control wells ( cells only ) and virus control wells , and multiplying the result by 100 . Neutralization was measured at a time when virus-induced cell killing in virus control wells was >70% but <100% . Neutralizing titers are given as the reciprocal dilutions required to protect 50% of cells from virus-induced killing . Surface staining was carried out by standard procedures for our laboratory as described [56] . Except where noted , all reagents were obtained from BD Biosciences ( San Diego , CA ) and included monoclonal antibodies to the following molecules: CD3 ( clone SP34-2 , APC-Cy7 conjugate ) CD4 ( clone SK3 , PerCP-Cy5 . 5 conjugate ) , CD8α ( clone RPA-T8 , Alexa700 conjugate ) CD28 ( clone CD28 . 2 , PE-Texas Red conjugate , Beckman-Coulter , Fullerton , CA ) , CCR7 ( clone 150503 , Pacific Blue conjugate , custom ) CXCR3 ( clone 1C6 , PE-Cy5 conjugate ) , Ki-67 ( Clone B56 , PE conjugate ) , CD127 ( clone R34 . 34 , PE conjugate , Beckman-Coulter ) , perforin ( clone Pf-344 , FITC conjugate , Mabtech , Mariemont , OH ) . Intracellular staining for perforin expression was performed using Caltag Fix & Perm ( Invitrogen , Camarillo , CA ) according to the manufacturer’s suggested protocol . Enumeration of SIV-specific cells using PE- or APC-conjugated pentamers to Mamu-A*01 Gag181-189CM9 and Tat28-35SL8 ( Proimmune , Oxford , UK ) was performed as described previously [57] . All samples were analyzed using an LSR II ( BD Biosciences ) , and analyses were performed using FlowJo software ( Tree Star Inc . , Ashland , OR ) . Isotype-matched controls and/or fluorescence-minus-one ( FMO ) controls were included in all assays [58] . Intracellular cytokine staining ( ICS ) was performed using methods optimized for detection of SIV-specific responses in rhesus macaques [59–61] . Briefly , thawed cryopreserved PBMC were stimulated with SIVmac239 Gag peptides ( 2 μg/ml ) for 12 hours at 37°C in the presence of the co-stimulatory antibodies anti-CD28 and anti-CD49d . GolgiPlug ( 5 μg/ml ) , GolgiStop ( 0 . 7 μg/ml ) and FITC-conjugated antibodies ( 5x concentration ) to the lysosomal degranulation marker , CD107a ( clone H4A3 ) , were also added for the duration of stimulation . Media-only and SEB-stimulated cultures served as negative and positive controls , respectively . After culture , cells were surface-stained with fluorochrome-conjugated antibodies to CD4 and CD8 as described above . Cells were subsequently fixed and permeabilized using Caltag Fix and Perm , then incubated for 15 minutes at room temperature in the dark with APC-Cy7-conjugated anti-CD3 , PE-conjugated anti-IL-2 ( clone MQ1-17H12 ) , PE-Texas Red-conjugated anti-CD69 ( clone TP1 . 55 . 3 , Beckman-Coulter ) , PE-Cy7-conjugated anti-IFN-γ ( clone 4SB3 ) , APC-conjugated anti-TNF-α ( clone MAb11 ) . Samples were washed and fixed with 1% freshly prepared paraformaldehyde for at least 1 hour and then analyzed using an LSR II within 24 hours . Lymphocytes were gated based on forward-versus-side scatter characteristics , and the proportions of CD107a- and cytokine-expressing cells were determined by coexpression of CD69 on both CD3+CD4+ and CD3+CD8+ lymphocytes using FlowJo v8 . 8 . 6 software . All reported values are Gag-specific responses background subtracted from media controls . Multifunctional analyses were performed using Spice v5 . 32 [62] . All statistical analyses were performed using GraphPad Prism software ( GraphPad Software v6 . 0b , Inc . , La Jolla , CA , USA ) . Spearman’s rank correlation coefficients were used for assessing all correlations . Nonparametric Wilcoxon and Mann–Whitney tests were used for statistical analysis where the sample size was less than or equal to 6 . Otherwise , parametric t tests were conducted for quantitative outcomes that are approximately normally distributed; p values less than 0 . 05 were assumed to be significant in all analyses . Linear mixed effects models were used to evaluate longitudinal post-challenge viral loads of the three experiment groups compared to the control group . The linear mixed model allows for comparison between different vaccination groups based on the fixed effect of the vaccination duration ( 5 , 20 or 40 weeks ) as well as random effects ( animal-specific effects ) , while properly handling serial correlations among repeated viremia measures on the same experimental animal . In this analysis , post-challenge period was classified into three phases over the course of viral infection: peak viremia , viral set-point ( weeks 5 to 12 ) and chronic infection ( weeks 13 to 22 ) . Between-group differences in viremia in these three phases were evaluated .
|
Annually , more than two million people worldwide are infected with HIV , the virus that causes AIDS . Rhesus macaques can be infected with SIV , a close relative and ancestor of HIV , resulting in simian AIDS , recapitulating key aspects of human HIV infection . SIVΔnef , a live attenuated form of SIV , protects rhesus macaques from subsequent challenge with pathogenic SIV and is widely viewed as the most effective SIV vaccine . Here we demonstrate that SIVΔnef persistence during the vaccination period drives both cell-mediated and humoral immune response maturation . During the vaccination period , cell-mediated immune responses elicited by SIVΔnef target more conserved regions of the virus rendering immune escape more difficult . Furthermore , the localization of the cell-mediated immune responses is shifted over time from peripheral blood to sites of viral production that are rich in uninfected SIV target cells , thereby positioning cell-mediated immune responses where they are most needed after wild-type SIV challenge . Similarly , SIVΔnef persistence during the vaccination period also leads to the accumulation and maturation of the humoral immune response . Our findings highlight the unique capacity of persistent vaccines to elicit durable and effective immune responses against wild-type SIV challenge .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"enzyme-linked",
"immunoassays",
"pathogens",
"immunology",
"microbiology",
"retroviruses",
"viruses",
"preventive",
"medicine",
"immunodeficiency",
"viruses",
"rna",
"viruses",
"cytotoxic",
"t",
"cells",
"antibodies",
"vaccination",
"and",
"immunization",
"immunologic",
"techniques",
"antibody",
"response",
"research",
"and",
"analysis",
"methods",
"public",
"and",
"occupational",
"health",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"medical",
"microbiology",
"t",
"cells",
"microbial",
"pathogens",
"immunoassays",
"immune",
"response",
"siv",
"biochemistry",
"viremia",
"cell",
"biology",
"physiology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"viral",
"diseases",
"lentivirus",
"organisms"
] |
2016
|
Persistent Low-Level Replication of SIVΔnef Drives Maturation of Antibody and CD8 T Cell Responses to Induce Protective Immunity against Vaginal SIV Infection
|
Cells assemble numerous types of actomyosin bundles that generate contractile forces for biological processes , such as cytokinesis and cell migration . One example of contractile bundles is a transverse arc that forms via actomyosin-driven condensation of actin filaments in the lamellipodia of migrating cells and exerts significant forces on the surrounding environments . Structural reorganization of a network into a bundle facilitated by actomyosin contractility is a physiologically relevant and biophysically interesting process . Nevertheless , it remains elusive how actin filaments are reoriented , buckled , and bundled as well as undergo tension buildup during the structural reorganization . In this study , using an agent-based computational model , we demonstrated how the interplay between the density of myosin motors and cross-linking proteins and the rigidity , initial orientation , and turnover of actin filaments regulates the morphological transformation of a cross-linked actomyosin network into a bundle and the buildup of tension occurring during the transformation .
The actin cytoskeleton plays an important role in various cellular processes , such as changes in cell shape , cytokinesis , and cell migration [1] . Much of the mechanical forces required for these processes are generated by interactions between actin filaments ( F-actin ) and myosin II motors [2] . Actomyosin contractility regulates structural organization of the actin cytoskeleton and its rheological properties by interacting and competing with the dynamics of actin cross-linking proteins ( ACPs ) and actin filaments . For example , during Dictyostelium furrow ingression , interactions between myosin and ACP dynamics control cytokinesis contractility dynamics and mechanics [3] . In addition , during fission yeast cytokinetic ring assembly , an increase in ACP density prevents clump formation [4 , 5] . Representative cytoskeletal structures that are regulated by actomyosin contractility are various types of bundles , such as stress fibers , random polarity bundles , cytokinetic rings , and transverse arcs [6] . Despite similarity in their structural organization , these bundles are formed via very distinct mechanisms . Dorsal stress fibers are assembled via formin-driven polymerization of actin filaments occurring outside adhesion sites . Transverse arcs , that are located at the interface between lamellipodia and lamella , form via actomyosin-driven condensation of actin filaments within the lamellipodia [7] . During the condensation , actin filaments whose barbed ends are initially biased toward the cell margin are reoriented and thus become parallel to the margin . Transverse arcs move away from the cell margin and eventually coalesce with dorsal stress fibers , to transmit contractile forces to surrounding environments , without direct attachment to focal adhesions [8] . Several aspects regarding structural reorganization of a network into a bundle have been investigated in previous numerical studies . It was shown that an increase in myosin density induces a structural transition from networks into bundles through a series of hierarchical steps [9] with enhancement of forces generated by the actomyosin structures [10] . In addition , a recent study demonstrated that an increase in ACP density above a threshold value leads to a switch-like transition from random networks to ordered , bundled structures [11] . However , owing to the highly simplified models and limited scopes of the previous studies , it still remains inconclusive how a network is transformed into a bundle , how force is generated , and what happens on actin filaments during the structural reorganization . Several biophysical factors are likely to impact network transformation into a bundle . For example , an extent to which actin filaments are cross-linked will play an important role . If filaments are loosely cross-linked , they may be reoriented relatively easily to form a bundle , but low network connectivity could be antagonistic to the stability of formed bundles and generated forces . By contrast , if actin filaments are heavily cross-linked , they may not easily rotate without significant deformation . Because of the low bending rigidity of actin filaments , myosin motor activity could result in buckling during reorientation and compaction of cross-linked actin filaments . As suggested by a previous theoretical study [12] , filament buckling may play a critical role in either force generation or bundle formation or in both . In addition , fast turnover of actin filaments occurring via diverse actin binding proteins within cells has potential to modulate the morphological transformation and force generation . Using only experiments , it is challenging to accurately evaluate relative importance of each of these factors and isolate their effects . In this work , using an agent-based computational model , we systematically investigated morphological transformation of an actomyosin network into a bundle and force generation during the transformation . We investigated effects of diverse biophysical parameters on network compaction into a bundle , which were not systematically studied in previous computational works . Specifically , we focused on the impacts of the densities of ACPs and motors and of the rigidity , initial orientation , and turnover of actin . Results from the study were discussed in the context of the assembly of transverse arcs observed in migrating cells [7] . This study provides new insights into mechanistic understanding of a role of the interplay between various biophysical factors in bundle formation and force generation .
We employed our previous coarse-grained Brownian dynamics model for actomyosin structures [13] . In the model , actin filaments , actin cross-linking proteins ( ACPs ) , and motors are simplified into interconnected cylindrical segments ( Fig 1A ) . Actin filaments consist of serially-connected cylindrical segments with polarity ( barbed and pointed ends ) . ACPs are composed of a pair of cylindrical segments . Each motor has a backbone structure with 8 arms , each of which represents 8 myosin heads . Displacement of the segments is governed by the Langevin equation . Harmonic potentials with bending ( κb ) and extensional stiffnesses ( κs ) maintain equilibrium angles and lengths , respectively , formed by the segments . Repulsive forces account for volume-exclusion effects between actin filaments . Stochastic forces satisfying the fluctuation-dissipation theorem are applied to induce thermal fluctuation [14] . Positions of the segments are updated at each time step using the Euler integration scheme . ACPs bind to actin filaments at a constant rate and also unbind from actin filaments in a force-dependent manner following Bell’s equation [15] . A motor arm binds to an actin filament and walks toward the barbed end of the actin filament , generating tensile forces . Actin undergoes nucleation , polymerization , and depolymerization , staying in either monomeric or filamentous state . We simulate treadmilling of actin filaments by imposing equal polymerization and depolymerization rates at barbed and pointed ends , respectively . To alter the treadmilling rate without a large change in average length of actin filaments , a nucleation rate is dynamically adjusted . Monomeric actin and free ACP and motor that are not bound to any actin filament are considered implicitly by their local concentrations . Self-assembly of actins , ACPs , and motors in a 3D rectangular computational domain ( 4×8×0 . 5 μm ) results in a homogenous actomyosin network ( Fig 1B ) . A periodic boundary condition is imposed in the y-direction , whereas boundaries in the x- and z-directions exert repulsive forces on the segments to keep them within the domain . After network assembly , walking of motors on actin filaments is initiated , facilitating transformation of the network to a bundle . We measured a macroscopic force generated by a bundle and also microscopic forces acting on ACPs and motors . Definitions of terms are listed in S1 Table , and detailed values of parameters are listed in S2 Table . Consistent with previous theoretical and experimental studies [16–18] , densities of ACPs ( RACP ) and motors ( RM ) critically affect bundle formation and tension generation . With RM = 0 . 08 and RACP = 0 . 01 , a homogeneous network compacted into a bundle spanning the computational domain in the y-direction within ~10 s ( Fig 2A ) . However , the bundle was heterogeneous at 10 s in terms of actin concentration , showing a few regions with higher actin density . In addition , the bundle was highly unstable , resulting in a few separate aggregates over time . Tension measured in the bundle increased up to ~0 . 8 nN and then decreased to nearly zero ( Fig 2C ) . By contrast , with RM = 0 . 08 and RACP = 0 . 1 , a more compact , uniform bundle was formed within 15 s , and the bundle remained intact for the duration of the simulation ( Fig 2B ) . Tension increased up to ~4 nN , and then decreased slowly . Microscopic forces exerted on each motor ( fMmax ) and ACP ( fACPmax ) measured at maximum tension can explain the magnitude and sustainability of the generated tension ( Fig 2D ) . Note that fMmax and fACPmax are positive when they are exerted toward barbed ends of actin filaments . With a large number of ACPs ( RM = 0 . 08 and RACP = 0 . 1 ) , fMmax was higher , and fACPmax was smaller . If there are many ACPs , they share loads exerted by motors , leading to smaller force on each ACP . Since ACPs are assumed to exhibit slip-bond behavior , the smaller force on ACPs leads to less frequent unbinding events of ACPs . Thus , stable ACPs can help motors to generate higher force close to their stall force and support the force for a longer time . By contrast , with fewer ACPs ( RM = 0 . 08 and RACP = 0 . 01 ) , most motors failed to attain their stall force , and each ACP supported a larger force , leading to instability of the bundle and reduction in generated tension ( Fig 2D ) . We systematically varied RACP and RM to probe their effects on bundle formation and tension generation . Maximum tension was positively correlated with both densities ( Fig 2E ) , whereas sustainability was proportional to RACP but inversely proportional to RM ( Fig 2F ) . We measured time evolution of standard deviation of x positions of actins ( σx ) to quantify compaction of networks ( S1 Fig ) . σx tends to initially decrease , indicating compaction of networks . After reaching its minimum value , σx remained constant in most cases . However , in some cases , σx increased over time , which may indicate disintegration of a bundle into aggregates . Indeed , the increase in σx occurred in cases with higher RM and lower RACP where tension is not sustained well , and bundles are likely to form aggregates . In cases with very low RM , σx continuously decreased , indicating very slow compaction of networks . To quantify how fast networks compact , we defined compaction time as time at which the rate of change in σx over time becomes larger than 0 . 01 × ( the average rate of change in σx during first 5s ) . The compaction time was shorter at higher RM and lower RACP ( Fig 2G ) . We used the standard deviation at compaction time ( σxc ) as an indicator of how tightly a network is compacted in the x-direciton ( Fig 2H ) . A tighter bundle was formed with higher RM and RACP . A sufficient amount of ACPs can tighten bundles by helping force generation of motors and increasing connectivity of bundles . However , ACPs slow down formation of bundles because a network becomes much more stiffer with more ACPs . In sum , a network with more motors compacted faster into a tighter bundle exerting larger tension because there are more force generators . However , the bundle and the tension are likely to be unstable , leading to bundle disintegration into aggregates and significant tension relaxation . A network with more ACPs compacted more slowly into a tighter bundle generating larger and more sustained tension . In our previous studies , it was shown that buckling of actin filaments is necessary for contraction of a network and for force generation in a preformed bundle [16 , 19] . We quantified buckling events occurring in the simulations shown in Fig 2E–2H , by tracking the ratio of end-to-end distance to contour length of actin filaments . Since most actin filaments have multiple , transiently bound motors and ACPs , buckling takes place in various ways; some of the actin filaments experienced subsequent buckling events at multiple locations over time , and buckled filaments , at times , became straight again ( S2 Fig ) . We determined the number of actin filaments that underwent buckling at least once in each simulation by assuming that actin filaments with a ratio of end-to-end distance to contour length smaller than 0 . 6 are buckled . We found that buckling occurred less frequently with higher RACP because the critical force above which buckling occurs becomes larger with higher RACP ( Fig 3A ) ; this is associated with a decrease in distance between adjacent cross-linking points on an actin filament . Although motors generate larger forces with higher RACP ( Fig 2D ) , the increase in the critical force required for buckling is greater , leading to less frequent buckling events . With higher RM , buckling took place more frequently since more motors generate larger contractile forces that can induce buckling . These buckling events mostly occurred during the transformation to a bundle before tension reached its maximum , rather than after the peak tension ( Fig 3D ) . We tested whether buckling is required for the transformation of a network into a bundle by suppressing the filament buckling via a 100-fold increase in the bending stiffness of actin filaments ( κb , A = 100×κb , A* ) , where κb , A* is the reference bending stiffness . At both high and low levels of RACP , a bundle rarely formed although some of the actin filaments formed a pseudo bundle at the center ( Fig 3B and 3C ) . At RM = 0 . 08 and RACP = 0 . 1 , the developed tension in a network with 100×κb , A* was much smaller than that in a network with κb , A* , and buckling rarely occurred ( Fig 3D ) . Smaller tension for the case with 100×κb , A* can be attributed to low values of fMmax; although some values reached stall force , there was a general tendency for the forces to be smaller overall than those in the case with κb , A* ( Fig 3E ) . Negative values of fACPmax were also slightly smaller in magnitude for the case with 100×κb , A* since ACPs sustain lower positive fMmax in this case . Note that negative or positive fACPmax sustain positive or negative fMmax , respectively . Positive fACPmax showed higher value for the case with 100×κb , A* , since this case exhibits a significant amount of negative fMmax while the case with κb , A* does not . Due to the catch-bond nature of motors , the lower positive fMmax makes motors stay for a shorter time on actin filaments , which corresponds to a lower duty ratio of motors . Then , motors are less able to stably generate a large amount of forces . Suppression of bundle formation and generation of lower tension observed in Fig 3B–3D might originate largely from a decrease in the duty ratio rather than an increase in κb , A . To confirm the importance of κb , A , we ran a simulation using motors with a much higher unbinding rate ( i . e . lower duty ratio ) than the motors used in the case shown in Fig 2B where a stable bundle was formed . We varied one of the mechanochemical rates in the parallel cluster model [20 , 21] , which leads to a decrease in the stall force from 5 . 7 pN to 5 . 3 pN and an increase in the unbinding rate from 0 . 049 s-1 to 0 . 49 s-1 . As shown in S3 Fig , a bundle still formed well , and tension inside the bundle and sustainability were similar to those of the reference case shown in Fig 2B and 2C . Thus , the inhibition of bundle formation and the decrease in tension result mostly from the change in the κb , A , not the change in the duty ratio of motors . Maximum tension measured under various values of RM and RACP with 100×κb , A* ( Fig 3F ) was much lower than that measured with κb , A* ( Fig 2E ) . Dependences of sustainability and compaction time on RM and RACP ( Fig 3G and 3H ) were similar to those in the cases with κb , A* ( Fig 2F and 2G ) . We also measured time evolution of σx for quantification of network compaction ( S4 Fig ) . Interestingly , in cases with lower RACP and higher RM , σx increased beyond its initial value after reaching the minimum . σxc was overall higher in the cases with 100×κb , A* ( Fig 3I ) than that in the cases with κb , A* ( Fig 2H ) , quantitatively showing suppression of bundle formation with stiffer actin filaments . Interestingly , with more ACPs , σxc was larger , which is opposite to the observation in Fig 2H . As shown in Fig 3A , buckling occurred less frequently at higher RACP even with κb , A* . However , since a fraction of actin filaments were still buckled , the number of buckled actin filaments is not a critical factor determining the extent of network compaction . By contrast , with 100×κb , A* , most of actin filaments cannot be buckled due to a significant increase in the critical buckling force . Then , network compaction becomes very sensitive to the number of buckled actin filaments because buckling is necessary for network compaction , resulting in less network compaction with higher RACP . In sum , these results demonstrate that even with a sufficient number of ACPs that sustain tension and help motors reach their stall force , buckling of actin filaments is required for formation of tight bundles and generation of large tension . Myosin II motors compact actin filaments in lamellipodia into transverse arcs that generate contractile forces [22] . Since the barbed ends of all actin filaments in lamellipodia are directed toward the cell margin , the lamellipodia is not an isotropic actin network . We probed the effects of anisotropic initial orientations of actin filaments on bundle formation and tension generation with RM = 0 . 08 and RACP = 0 . 01 by creating three networks consisting of actin filaments with biased initial orientations ( Fig 4A–4C ) . Note that the case shown in Fig 4B where actin filaments are initially oriented toward the +x direction mimics filament orientation in lamellipodia . Compared to the reference case with isotropic orientation of filaments ( Fig 2A and 2C ) , the networks with biased orientations showed lower maximum tension and slower bundle formation ( Fig 4A–4D ) because there were a smaller number of antiparallel pairs of actin filaments that are in configuration suitable for motors to produce force ( Fig 4F ) . Interestingly , a network with barbed ends directed toward +y was effectively transformed to a bundle with significant tension despite the fact that it initially had no antiparallel pairs of actin filaments in the y-direction . We found that some of the actin filaments changed their orientations ( S5 Fig and Fig 4C , right column ) during network contraction ( Fig 4F ) . Even in the network with barbed ends oriented toward +x/+y , a bundle could form slowly and generate tension due to changes in filament orientation ( Fig 4A , 4D and 4F ) . In all cases , bundles eventually collapsed into a few aggregates; this occurred at a rate proportional to the maximum tension because larger tension accelerates destabilization of ACPs , leading to faster disintegration of bundles . We also tested the influences of initial orientation of actin filaments ( diagonal or horizontal/vertical ) on bundle formation and tension generated in networks , and the results overall showed similar tendencies ( S6 and S7 Figs ) . At higher ACP density ( RM = 0 . 08 and RACP = 0 . 1 ) , actin filaments tend to rotate less than those at lower RACP because the filaments are confined more by a larger number of ACPs ( S8A , S8B and S8C Fig ) . However , some of the actin filaments were still able to change their orientations , contributing to tension generation ( S8D and S8E Fig ) . Note that unlike the case with lower ACP density , the bundles were not disintegrated into aggregates , regardless of initial filament orientation . This can explain a discrepancy between the unstable bundle shown in Fig 4B formed from a network mimicking the geometry of lamellipodia and a stable bundle observed at the interface between lamellipodia and lamella . It is expected that actin filaments with numerous branching points in lamellipodia have very high connectivity between actin filaments , preventing a bundle from being disintegrated . Taken together , these results demonstrate that networks with biased filament orientations can still be transformed to bundles owing to changes in filament orientation occurring during contraction . However , if orientations are biased , bundles are loose , and generated tension tends to be lower but is sustained for a longer time . We have observed that buckling is necessary for bundle formation in networks with isotropic filament orientation since contraction of antiparallel pairs of actin filaments requires buckling . We tested whether buckling is still necessary for bundle formation in networks with a much smaller number of antiparallel pairs by increasing the bending stiffness of actin filaments 100-fold as before ( κb , A = 100×κb , A* ) . We found that networks with barbed ends directed toward +x/+y or +y were still transformed to bundles because contraction in the y-direction does not need to occur in such configurations ( Fig 5A and 5C ) . Filaments in the network with barbed ends directed toward +x/+y initially form only parallel pairs of actin filaments , so they can be aligned in the y-direction ( S9A Fig ) . Filaments forming antiparallel pairs in the x-direction in the network with barbed ends directed toward +y can be aligned in the y-direction via polarity sorting due to the absence of a periodic boundary condition in the x-direction ( S9C Fig ) . Some of the filaments changed their orientation during bundle formation , resulting in antiparallel pairs in the y-direction that were also connected to other actin filaments in a bundle ( Fig 5E ) . Due to suppression of buckling , these pairs cannot contract , so the bundles remained curved rather than straight . Accordingly , forces generated on bundles remained close to zero and even became compressive ( i . e . negative ) ( Fig 5D ) . By contrast , a network with barbed ends directed toward +x/±y could not form a bundle since the antiparallel pairs of filaments that existed from the beginning were not able to contract ( Fig 5B and S9B Fig ) . Tension generated in these networks was similar to that in networks with isotropic orientations ( Fig 5D ) . Therefore , buckling is not always necessary for the transformation of a network to a bundle . If orientation of actin filaments is highly anisotropic , the transformation can still take place via polarity sorting of filaments by motors . However , tensile forces are not developed on the formed bundles . In our previous study , we demonstrated that actin turnover modulates the buildup and sustainability of tension generated by actomyosin networks [13] . We tested effects of actin turnover on bundle formation and tension generation by imposing actin treadmilling at various rates ( kt , A ) under a condition where bundles generate unsustainable tension and eventually form aggregates in the absence of any turnover ( RM = 0 . 08 and RACP = 0 . 01 ) . We additionally assumed that depolymerization of actin filaments can be inhibited by bound ACPs or motors to a different extent [2] . We defined the inhibition factor ( ξd , A ) to represent this effect; with ξd , A = 0 , depolymerization is not inhibited at all , whereas inhibition is complete with ξd , A = 1 . In a control case without turnover ( kt , A = 0 ) and a case with kt , A = 60 s-1 and ξd , A = 1 , bundles became aggregates within 100 s ( Fig 6A and 6D ) , and generated tension fell to nearly zero ( Fig 6E ) . With kt , A = 60 s-1 and ξd , A = 0 , some of the actin filaments in the network formed a thin bundle that was converted into aggregates over time ( Fig 6B ) , and tension ultimately relaxed to zero ( Fig 6E ) . By contrast , with kt , A = 60 s-1 and ξd , A = 0 . 6 , the bundle was maintained much longer , showing highly sustainable tension ( Fig 6C and 6E ) . We systematically probed the effects of kt , A and ξd , A on the maximum and sustainability of tension ( Fig 6F and 6G ) . While maximum tension showed no correlation with kt , A and ξd , A , sustainability tended to be higher at intermediate levels of ξd , A because too large ξd , A completely inhibits actin turnover , whereas too small ξd , A precludes bundle formation and destabilizes the bundle by ACP unbinding induced by actin turnover . The region with higher sustainability is wider with lower kt , A , since less turnover occurs at lower kt , A at the same level of ξd , A . Networks compacted faster with more turnover ( i . e . higher kt , A and lower ξd , A ) , but formed bundles were loose ( Fig 6H and 6I ) . This agrees with the observation that compaction occurred faster , and more loose bundles formed at lower RACP ( Fig 2G and 2H ) , because more frequent turnover facilitates unbinding of ACPs , leading to a decrease in the number of active ACPs bound on two actin filaments at dynamic equilibrium . Also , with low ξd , A , σx increased after reaching its minimum ( S10 Fig ) , which corresponds to disintegration of a bundle into a network . However , the increase in σx significantly slowed down after some time in several cases , which implies a steady state with coexistence of bundle and network structures as shown in Fig 6C . At high RACP shown in S11 Fig ( RM = 0 . 08 and RACP = 0 . 1 ) , bundle formation and the maximum tension were both enhanced with slower actin turnover ( i . e . lower kt , A and higher ξd , A ) . Compaction time , σxc , and σx showed similar trends with those in Fig 6 and S10 Fig ( S11 and S12 Figs ) . In this case , the bundle and generated tension are already stable without turnover owing to numerous ACPs . Actin turnover decreases the number of actin filaments involved with bundle formation as can be seen in a change in the diameter of bundles ( S11B , S11C and S11D Fig ) . Thus , the connectivity of filaments in the bundle is deteriorated , resulting in less sustainable tension . In addition , since turnover induces unbinding of ACPs which leads to instability , more motors failed to reach their stall force , leading to smaller maximum tension ( S11E Fig ) . Indeed , fMmax was lower with increasing turnover ( S11F Fig ) . fACPmax also decreased with increasing turnover , owing to lower tension and facilitated ACP unbinding by actin turnover . Note that the case with ξd , A = 1 showed more sustained tension than the case without actin turnover . With ξd , A = 1 , depolymerization occurs in regions of an actin filament which are not bound to ACPs or motors , thus unnecessary for tension generation . Depolymerized actin can be polymerized at barbed ends of actin filaments , helping sustain tension by increasing a walking distance of motors toward a barbed end . In sum , with an insufficient number of ACPs , actin turnover with intermediate values of ξd , A enhances the stability of bundles and generated tension , whereas with more ACPs , actin turnover plays only a negative role for the stability of bundles and tension .
Structural reorganization of a cross-linked actin network into a bundle occurs in several cellular phenomena , such as formation of transverse arcs at the interface between lamellipodia and lamella . Recent experiments have shown that in the absence of stress fibers , cells can still exert large tensions on surrounding environments due to contractile lamella that contain transverse arcs , implying the significance of transverse arcs in cells as a force generator [23] . To illuminate mechanisms of formation and force generation of transverse arcs , we here presented a computational study regarding transformation of actomyosin networks into bundles under diverse conditions . Results from this study demonstrate that formation of contractile bundles and force generation in the bundles are tightly regulated by the interplay between concentrations of cytoskeletal elements and the deformability , dynamics , and initial orientation of actin filaments that have not been tested systematically in previous studies . This study is significantly different from our previous study that employed actomyosin bundles preassembled by stacking straight actin filaments in parallel [16] since actin filaments are not stacked merely without any deformation during the morphological transformation . We found that during the transition from a network into a bundle , actin filaments undergo buckling and reorientation in various ways , and a large portion of tension is built during the structural reorganization rather than after bundle formation . In addition , we incorporated systematic variations of initial filament orientation that have not been included in our previous studies [13 , 16 , 24–27] , motivated by observation that transverse arcs located at the interface between lamellipodia and lamella are formed by compaction and realignment of actin filaments with biased orientations within the lamellipodia [28] . We investigated how the density of ACPs and motors and the buckling of actin filaments govern the bundle formation and tension generation . It was found that maximum bundle tension is proportional to motor and ACP densities , whereas sustainability of tension is proportional to ACP density but inversely proportional to motor density . A key factor for determining tension sustainability is how much force is exerted on each ACP because large force can make ACPs unstable by increasing their force-dependent unbinding rate . This is consistent with our previous studies where forces are generated by cortex-like actomyosin networks [19] and preformed bundles [16] . We observed that time required for bundle formation is inversely proportional to motor density but proportional to ACP density . Previous experimental studies showed that condensation of networks into transverse arcs occurs within 20 s [29] , which is comparable with the compaction time measured in this study . We also observed that buckling of actin filaments plays an important role in bundle formation , and most of the tension is generated during a transition from a network to a bundle . This is different from our previous study where we found the importance of filament buckling and force generation during contraction of the preformed bundles [16] . In addition , using networks consisting of filaments with biased orientations , we found that buckling should take place in antiparallel pairs of actin filaments initially aligned in the y-direction in order to induce transformation of networks into bundles . If there is not such an antiparallel pair in the y-direction , the transformation is possible without filament buckling . However , development of large tension on a formed bundle is possible only when filament buckling is allowed . In addition , we showed that networks with isotropic filament orientations result in the best bundle formation and the largest tension . Interestingly , even if orientations of actin filaments are too biased to initially have antiparallel pairs of actin filaments , some of the actin filaments change their orientations during network contraction , resulting in antiparallel pairs and formation of bundles . However , compared to the network with isotropic orientations of actin filaments , bundles are loosely formed , and tension is smaller . Since the smaller tension leads to lower force on each ACP , tension is sustained for a longer time . Also , we probed influences of actin turnover via treadmilling on bundle formation and tension generation as in our previous study . However , we made a new assumption that actin depolymerization rate can be varied by cross-linking points based on previous experimental observations [30] . We observed that actin turnover with moderate inhibition of actin depolymerization by motors and ACPs increases the sustainability of tension and confers structural stability to the bundles at low ACP density . If there is a selective inhibition of depolymerization , the region of a filament that contributes least to the connectivity of bundles ( from a pointed end to the first cross-linking point ) is depolymerized faster . Depolymerized actin can be polymerized at a barbed end of the same filament or other actin filaments . Since motors walk toward barbed ends , the newly polymerized actin can enable motors to walk further . By contrast , at high ACP density , actin turnover decreases tension sustainability and the stability of formed bundles because the connectivity of the bundles is already maximized by numerous ACPs . Loss of connectivity caused by actin turnover seems more critical than gain of stability from the turnover . Results from this study support observations from previous studies regarding actomyosin bundles and rings . A recent study showed the importance of architecture and connectivity for the contractility of actomyosin rings [17] . This study showed that each of polarity sorting , sarcomeric contractility , and filament buckling plays an important role at low , intermediate , and high connectivity , respectively . Significant ring contraction was observed only at regimes where sarcomeric contractility or filament buckling becomes important . Too high connectivity or too rigid filaments caused inhibition of filament buckling and ring contraction . Although we did not explore effects of very low connectivity in this study ( RACP ≥ 0 . 01 ) , we observed that buckling takes place less frequently at higher ACP density ( Fig 3A ) , and that suppression of buckling via an increase in filament bending stiffness results in inhibition of contraction ( Fig 3B and 3C ) . All of these are consistent with [17] and other studies showing significance of filament buckling for contraction [31 , 32] . Our study also predicted that compaction of an actomyosin network into a bundle is more significant with higher ACP and motor densities . This is in agreement with a recent computational study showing that an actomyosin network exhibits greater contraction and filament alignment with higher densities of motors and ACPs [11] . In addition , another recent computational study found that contraction of random actomyosin arrays mimicking cytokinetic rings is slower with more cross-linkers [33] , which is also consistent with our study ( Fig 2G ) . To summarize , in this study , we systematically investigated how the transformation of the thin actomyosin networks to bundles is regulated by various biophysical factors . We recently demonstrated impacts of severing of actin filaments induced by buckling on rheological behaviors of passive cross-linked actin networks [34] . In future studies , we will include buckling-induced filament severing to test its effects on bundle formation and tension generation .
Displacements of the segments constituting actin filaments , motors , and ACPs are governed by the Langevin equation with inertia neglected: Fi−ζidridt+FiT=0 ( 1 ) where ri is a position vector of the ith element , ζi is a drag coefficient , t is time , Fi is a deterministic force , and FiT is a stochastic force satisfying the fluctuation-dissipation theorem [14]: ⟨FiT ( t ) FjT ( t ) ⟩=2kBTζiδijΔtδ ( 2 ) where δij is the Kronecker delta , δ is a second-order tensor , and Δt = 1 . 5×10−5 s is a time step . Drag coefficients are computed using an approximated form [35]: ζi=3πμrc , i3+2r0 , i/rc , i5 ( 3 ) where μ is the viscosity of medium , and r0 , i and rc , i are length and diameter of a segment , respectively . Positions of the various elements are updated using the Euler integration scheme: ri ( t+Δt ) =ri ( t ) +dridtΔt=ri ( t ) +1ζi ( Fi+FiT ) Δt ( 4 ) Deterministic forces include extensional forces maintaining equilibrium lengths , bending forces maintaining equilibrium angles , and repulsive force between actin segments . Extension and bending of actin , ACP , and motor are governed by harmonic potentials: Us=12κs ( r−r0 ) 2 ( 5 ) Ub=12κb ( θ−θ0 ) 2 ( 6 ) where κs and κb are extensional and bending stiffness , respectively , r is the length of a segment , θ is an angle formed by adjacent segments , and the subscript 0 indicates an equilibrium value . An equilibrium length of actin segments ( r0 , A = 140 nm ) and an equilibrium angle formed by two adjacent actin segments ( θ0 , A = 0 rad ) are maintained by extensional ( κs , A ) and bending stiffness of actin ( κb , A ) , respectively . The reference value of κb , A results in persistence length of 9 μm [36] . An equilibrium length of ACP arms ( r0 , ACP = 23 . 5 nm ) and an equilibrium angle between two arms of each ACP ( θ0 , ACP = 0 rad ) are maintained by extensional ( κs , ACP ) and bending stiffness of ACPs ( κb , ACP ) , respectively . It is assumed that the values of extensional stiffness of a motor backbone ( κs , M1 and κs , M2 ) keeping an equilibrium length ( rs , M1 = rs , M2 = 42 nm ) are equal to the value of κs , A . The bending stiffness of a motor backbone ( κb , M ) keeping the backbone straight ( θ0 , M = 0 rad ) is assumed to be larger than κb , A . Extension of each motor arm is regulated by stiffness of transverse ( κs , M3 ) and longitudinal springs ( κs , M4 ) . The transverse spring maintains an equilibrium distance ( r0 , M3 = 13 . 5 nm ) between an endpoint of a motor backbone and the actin segment where the arm of the motor binds , whereas the longitudinal spring helps maintaining a right angle between the motor arm and the actin segment ( r0 , M4 = 0 nm ) . Forces exerted on actin segments by bound ACPs and motors are distributed onto the barbed and pointed ends of the actin segments as described in our previous work [16] . A repulsive force accounting for volume-exclusion effects between actin segments is represented by following harmonic potential [26]: Ur={12κr ( r12−rc , A ) 2ifr12<rc , A0ifr12≥rc , A ( 7 ) where κr is strength of repulsive force , and r12 is the minimum distance between two actin segments . ACPs can bind to binding sites located every 7 nm on actin segments with no preferential angle for binding . ACPs can also unbind from actin filaments in a force-dependent manner following Bell’s equation [15] . ku , ACP={ku , ACP0exp ( λu , ACP|F→s , ACP|kBT ) ifr≥r0 , ACPku , ACP0ifr<r0 , ACP ( 8 ) where ku , ACP0 is the zero-force unbinding rate , λu , ACP represents a sensitivity to applied force , and kBT is thermal energy . F→s , ACP is a vector representing an extensional force acting on an arm of ACP ( F→s , ACP=−∇Us , ACP ) . The references values of ku , ACP0 ( = 0 . 115 s-1 ) and λu , ACP ( = 1 . 04×10−10 m ) are determined to mimic the unbinding behavior of filamin A [37] . Motor arms can bind to binding sites on actin segments with a rate of 40Nh s-1 , where Nh is the number of myosin heads represented by each arm . Walking ( kw , M ) and unbinding rates ( ku , M ) of the motor arms are determined by the “parallel cluster model” ( PCM ) [20 , 21] to mimic mechanochemical cycle of non-muscle myosin II . kw , M and ku , M decrease with higher applied load since motors exhibit catch-bond behavior . It was assumed that kw , M and ku , M are governed by forces exerted on the longitudinal spring of a motor arm that is regulated by κs , M4 ( F→s , M4=−∇Us , M4 ) . Unloaded walking velocity of motors is set to ~140 nm/s and stall force ( fMstall ) is set to ~5 . 7 pN . In the model , actin experiences nucleation , polymerization , and depolymerization . Nucleation corresponds to de novo appearance of one actin segment . Polymerization and depolymerization are implemented by adding and removing one actin segment on filaments , respectively . We simulated treadmilling of actin filaments by imposing equal polymerization and ( reference ) depolymerization rate at barbed and pointed ends , respectively . A turnover rate indicates how fast an actin filament turns over , which is equal to either polymerization or depolymerization rate . We chose physiologically relevant turnover rates ( 30–120 s-1 ) . A nucleation rate is also adjusted to maintain a relatively constant actin filament length . We assumed that actin nucleation takes place in the y-direction within a bundle . It is assumed that depolymerization can be inhibited by bound ACPs or motors [30]; an inhibition factor ranging between 0 and 1 ( ξd , A ) determines the extent of inhibition: kd , A=kd , A0 ( 1−ξd , A ) ( 9 ) where k0d , A and kd , A are reference and adjusted depolymerization rates at a barbed end or a pointed end . Thus , ξd , A = 0 corresponds to no depolymerization inhibition , whereas ξd , A = 1 means complete inhibition . We used a 3D rectangular computational domain ( 4 × 8 × 0 . 5 μm ) with a periodic boundary condition in the y-direction . Self-assembly of actin filaments , ACPs , and motors in the domain results in a homogenous actomyosin network . During the network assembly , actin monomers are nucleated and polymerized into filaments . When creating anisotropic networks , direction of nucleation is controlled so that actin filaments lie along desired directions after network assembly . Motors are assembled into thick filaments , and motor arms bind to actin filaments without walking motion . ACPs bind to actin filaments to form functional cross-links between pairs of actin filaments . Due to the fixed ratio of nucleation rate to turnover rate , the average length of actin filaments is maintained at ~1 . 56 μm . After the network assembly , motors start walking on actin filaments , and the nucleation rate is dynamically controlled to maintain the average filament length at a constant level . Actin monomer concentration ( CA ) is 40 μM for all cases . To measure tension generated by a bundle , we consider 10 cross-sections that are regularly located in the computational domain in the y-direction . Tension is calculated by summing the normal component of extensional forces of all constituents crossing a cross-section . We repeat this calculation on 10 cross-sections and compute the average . Sustainability of the tension is calculated in the same manner as in [13] . Microscopic forces acting on ACPs ( fACPmax ) and motors ( fMmax ) are evaluated when tension reaches a maximum: fACPmax=F→s , ACP⋅u→ ( 10 ) fMmax=F→s , M4⋅u→/Nh ( 11 ) where u→ is a unit vector directed toward a barbed end of actin filaments . Note that F→s , ACP and F→s , M4 are directed from the center of ACP or the endpoint on a motor backbone to a binding point on an actin filament where ACP or motor is currently bound , and that fACPmax and fMmax are positive when the force vectors are directed toward barbed ends . Most of fMmax values are positive because motor arms walk toward barbed ends , and because the unbinding rate of the motor arm defined by the PCM model is assumed to be very large when F→s , M4 is directed toward a pointed end . By contrast , values of fACPmax show largely symmetric distribution due to the absence of walking motion and unbinding rate independent of the direction of F→s , ACP . However , there is slightly higher population on negative values of fACPmax since ACPs sustain forces exerted by motors which are mostly positive . We measured time evolution of standard deviation of x positions of actins ( σx ) . σx decreases as a bundle forms and then either remains relatively constant until the end of the simulations or increases slowly over time if the bundle is disintegrated . As a measure of how fast a network compacts into a bundle , we define the compaction time as time when the rate of change in σx becomes larger than 0 . 01 × ( the average rate of change in σx for the first 5s ) . In addition , we used the magnitude of the standard deviation at the same time point ( σxc ) as a measure of how tightly the bundle is formed in the x-direction .
|
Contractile networks and bundles generate mechanical forces required for various cellular processes , particularly cell division and migration . In many of these processes , networks are structurally reorganized into bundles by the activity of molecular motors . During this morphological transformation , filaments constituting networks are reoriented and undergo deformation and turnover , and large tensile forces are generated and sustained in bundles . However , it remains inconclusive how the morphological transformation and force generation are regulated . Here , using a rigorous computational model , we quantitatively demonstrated that the interplay between several factors determines the characteristics of generated tensile force and regulates the transformation from networks to bundles . Thus , results in this study provide insights into the physical and mechanistic basis of the complex transition from networks to bundles observed in cells .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"stiffness",
"cell",
"motility",
"mechanical",
"properties",
"actin",
"filaments",
"classical",
"mechanics",
"molecular",
"motors",
"actin",
"motors",
"materials",
"science",
"damage",
"mechanics",
"motor",
"proteins",
"dynamic",
"actin",
"filaments",
"buckling",
"polymer",
"chemistry",
"proteins",
"deformation",
"chemistry",
"physics",
"biochemistry",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"material",
"properties",
"depolymerization"
] |
2017
|
Morphological Transformation and Force Generation of Active Cytoskeletal Networks
|
To replicate in mammalian hosts , bacterial pathogens must acquire iron . The majority of iron is coordinated to the protoporphyrin ring of heme , which is further bound to hemoglobin . Pathogenic bacteria utilize secreted hemophores to acquire heme from heme sources such as hemoglobin . Bacillus anthracis , the causative agent of anthrax disease , secretes two hemophores , IsdX1 and IsdX2 , to acquire heme from host hemoglobin and enhance bacterial replication in iron-starved environments . Both proteins contain NEAr-iron Transporter ( NEAT ) domains , a conserved protein module that functions in heme acquisition in Gram-positive pathogens . Here , we report the structure of IsdX1 , the first of a Gram-positive hemophore , with and without bound heme . Overall , IsdX1 forms an immunoglobin-like fold that contains , similar to other NEAT proteins , a 310-helix near the heme-binding site . Because the mechanistic function of this helix in NEAT proteins is not yet defined , we focused on the contribution of this region to hemophore and NEAT protein activity , both biochemically and biologically in cultured cells . Site-directed mutagenesis of amino acids in and adjacent to the helix identified residues important for heme and hemoglobin association , with some mutations affecting both properties and other mutations affecting only heme stabilization . IsdX1 with mutations that reduced the ability to associate with hemoglobin and bind heme failed to restore the growth of a hemophore-deficient strain of B . anthracis on hemoglobin as the sole iron source . These data indicate that not only is the 310-helix important for NEAT protein biology , but also that the processes of hemoglobin and heme binding can be both separate as well as coupled , the latter function being necessary for maximal heme-scavenging activity . These studies enhance our understanding of NEAT domain and hemophore function and set the stage for structure-based inhibitor design to block NEAT domain interaction with upstream ligands .
An important determinant in the outcome of a bacterial infection is how well the invading pathogen can acquire host iron . Hosts with high levels of free iron are more susceptible to infection , and deletion of iron acquisition systems in a wide range of bacterial species generally attenuates virulence [1] . The low free iron concentration in host tissues ( 10−18–24 M ) likely acts as a barrier to efficient bacterial replication [2] . However , pathogenic bacteria have evolved at least two distinct uptake systems to attain iron . One such mechanism is to secrete siderophores , small molecules that chelate ferric iron with very high affinity [3] . The iron-bound siderophore binds to the bacterial surface and specific ferric iron receptors next deliver the iron or iron-siderophore complex into the cell [4] . The genetic deletion of biosynthetic systems that make siderophores , or the surface receptors that recognize siderophores , decreases the virulence of several pathogens , including the Gram-positive bacteria B . anthracis [5] and S . aureus [6] . The second system bacteria employ to attain host iron targets iron-protoporphyrin IX , or heme . Although heme constitutes up to 80% of the bodily iron reserves , free heme is rare . Most heme is tightly bound to hemoproteins such as hemoglobin . Hemoglobin's important role as an oxygen carrier protein means it is in high abundance and thus a target for bacterial iron uptake [7]–[9] . Bacterial proteins that acquire heme from hemoglobin are called hemophores [10]–[13] . Hemophores are generally secreted into the external milieu where they extract heme , via an undefined mechanism , from heme sources such as hemoglobin [14] . The heme-bound ( holo ) hemophore then delivers its bound iron-porphyrin to a cognate receptor on the bacteria surface , which leads to heme import into the bacterial cell [15] , [16] . Heme from hemoglobin can also be attained at the bacterial surface through similar mechanisms involving receptors on the cell wall ( Gram positive ) or outer membrane ( Gram negative ) . Heme import by bacterial pathogens is important for the establishment or maintenance of infections caused by Bordetella [17] , Haemophilus [18] , Brucella [19] , Vibrio [20] , Streptococcus [21] , and Staphylococcus [22] species . Further , more recent studies suggest heme is a major determinant in the promotion and severity of bacterial sepsis [23] . Collectively , these studies highlight the important role of heme acquisition during infection of mammalian hosts and support the contention that the inhibition of iron uptake systems is a promising direction for the development of new therapeutics . However , despite this knowledge , no clinically-used antibiotics have been made that directly target bacterial iron import . An understanding of the molecular mechanism by which heme acquisition systems extract , bind , and transfer heme into bacterial cells would be an important first step in creating new drugs to treat deadly infections . The first discovered hemophore was HasA from the Gram-negative pathogen S . marcescens [10] . HasA is a small ∼15 kDa protein that seems to passively acquire heme from hemoglobin by virtue of its high affinity for the iron-porphyrin [24] . HasA delivers its bound heme to HasR , an outer membrane surface receptor that delivers the heme into the cell [25] . Whereas the discovery of HasA set the precedence for the study of hemophores , recent studies indicate that Gram-positive bacteria also utilize hemophores . B . anthracis , a Gram-positive pathogen that is a potential weapon for bioterrorism , secretes two hemophores ( IsdX1 and IsdX2 ) , that acquire heme from hemoglobin and promote bacterial growth in low-iron environments [26]–[28] . IsdX1 transfers its bound heme to IsdC , a surface protein covalently attached to the peptidoglycan of the cell wall [29] , [30] . The heme acquisition functions of IsdX1 , IsdX2 , and IsdC are dependent on the activity of their NEAT ( near-iron transporter ) domain [26] , [29] . This approximately 125 amino acid domain is conserved in several Gram-positive bacteria , and the collective action of several NEATs on the bacterial surface is hypothesized to lead to transfer of heme through the cell wall [27] , [31]–[33] . The deletion of genes encoding for NEAT proteins leads to reduction in virulence in both B . anthracis and S . aureus [34]–[36] . Further , vaccination of mice with recombinant proteins containing NEAT domains induces adaptive immunity to staphylococcal infection . [37]–[39] . However , the development of a safe vaccine or a small molecule inhibitor of NEATs will require a detailed understanding of how NEAT proteins perform their various heme transport functions , including a structural appreciation of the residues required for activity . At the heart of this activity is an understanding of how bacterial hemophores like IsdX1 mediate the specific extraction of heme from host hemoglobin , and whether the ability to take up heme is coupled to the process of heme removal from the globin donor . The NEAT domain of IsdX1 is the only known NEAT domain to bind both heme and hemoglobin , as well as acquire heme from hemoglobin , thus making IsdX1 a useful model protein to study the relationship of these processes to one another . To better understand the molecular basis of hemophore activity and provide mechanistic insights into NEAT protein function , we solved the structures of heme free and heme bound IsdX1 and used these structures to understand the functional roles of residues in a small 310-helix in the vicinity of the heme binding site .
To gain an appreciation of the residues in IsdX1 necessary for the hemophore activity of this protein , we solved the crystal structure of apo-IsdX1 to 1 . 8 Å resolution ( Table 1 ) . Overall , the backbone structure of IsdX1 is similar to the solved structures of NEAT domains from S . aureus , albeit with a unique surface charge distribution close to the heme binding pocket ( Figure S1 . ) [40]–[44] . The structure consists of an immunoglobulin-like fold with eight β-strands arranged in two antiparallel β-sheets of a β-sandwich ( Figure 1A ) . The heme-binding pocket is enclosed primarily by a 310-helix ( residues Arg-54 to Tyr-58 ) on one side of the heme and a long β-hairpin ( β7–β8 ) on the other . The 310-helix is sometimes referred to as the “lip” because it seems to protrude over the heme-binding pocket [42] . The backbone of the 310-helix in IsdX1 is fairly well-ordered even in the absence of heme . The integrity of the 310-helix without heme could be due to the hydrogen bonding network between Ser-52 , Ser-53 , Arg-54 and Asn-56 and Met-55 from the helix making van der Waals contacts with residues in β4 , β7 , and β8 . The heme bound form of IsdX1 was solved to 2 . 15 Å by molecular replacement using the structure of apo-IsdX1 as the search model . Heme-iron coordination is achieved through Tyr-136 , which is conserved among all heme-binding NEAT domain proteins ( Figure S2 , upper panel ) . The distance between the iron atom and the tyrosine oxygen ligand is 2 . 3 Å , which is typical for an Fe-O bond of NEAT domains ( Figures 1B and 1C ) [40]–[45] . The coordination bond to iron is stabilized by the conserved residue Tyr-140 , which forms a hydrogen bond with the phenolate oxygen of Tyr-136 . The aromatic ring of Tyr-140 further stabilizes the pyrrole ring through π-stacking . Both of these residues lie on a β-hairpin region , a conserved region in bacterial NEAT proteins that provides a structural platform for the heme [40] , [42] . The least solvent exposed heme propionate forms a hydrogen bond with the hydroxyl group of Ser-53 , as well as the backbone nitrogen of Arg-54 . Additionally , the side chain NH1 group of Arg-54 from chain A also hydrogen bonds to the least solvent exposed heme propionate , and also to the hydroxyl group of Ser-53 ( Figure 2A , B ) . However , this hydrogen-bonding network is not observed for the side chain of Arg-54 from chain B . The heme molecule is further stabilized by Tyr-58 , that weakly π-stacks with the buried heme pyrrole ring . The other residues lining the heme-binding pocket are Arg-54 , Met-55 , Phe-59 , Ile-84 , Val-127 , Ile-129 , Ile-131 , Ile-142 and Phe-144 , which mainly form an aliphatic environment to accommodate the hydrophobic regions of the heme and its side chains . The location of the 310-helix in the structures of the NEAT domains suggests this region is important for NEAT protein function , including heme binding , hemoglobin association , heme extraction and NEAT-NEAT heme transfer [41]–[43] , [45] . There is some conservation in this region , as noted by Pilpa et al , with aromatic residues common in the equivalent positions of amino acids 54 and 58 for IsdX1 [46] . Interestingly , a serine residue ( Ser-53 in IsdX1 ) , which is immediately adjacent to the first residue of the helix ( Arg-54 ) , is well conserved in these NEATs , including every NEAT domain from B . anthracis ( Figure 3 , bold ) [47] . To determine the role of this and adjacent residues in the ability of IsdX1 to bind heme , each amino acid in 52-SSRM-55 was changed to alanine and mutant proteins purified from E . coli . Whereas wild-type IsdX1 co-purified with a significant amount of endogenous heme from E . coli , IsdX1 ( SSRM→AAAA ) bound approximately 10-fold less heme after purification ( Figure 4A , quantitated in 4C ) . Removal of the bound heme by organic extraction ( Figure 4B ) and quantitation of the Soret band intensity after titration of the apo protein with hemin confirmed the heme-binding defect of the mutant protein ( Figure 4D ) . To determine the residues responsible for this defect , single substitution changes in 52-SSRM-55 were generated and each purified mutant assessed for heme binding activity . As demonstrated in Figure 4C and D , mutation of Ser-52 , Ser-53 , or Met-55 decreased the ability of each of these mutant proteins to either co-purify with heme ( Figure 4C ) or bind exogenously added hemin ( Figure 4D ) . The raw spectra for these mutants are illustrated in Figure S3 . These effects are not due to gross disruption of IsdX1 secondary structure since the mutant proteins retained a similar overall β-sheet content as the wild-type protein when assessed by far-UV circular dichroism ( Figure S4 ) . Further , the heme binding site seems to be somewhat intolerant of even small structural changes , since substitution of Ser-53 to a threonine , which differs only in the length of the side chain ( extra methyl group ) , also abolished the interaction with heme ( Figure 4 C , D ) . Interestingly , mutation of Arg-54 led to a purified IsdX1 preparation with a Soret band approximately three times greater than wild-type protein ( Figure 4C ) . The high heme content in this sample was confirmed using the pyridine hemochrome method , which demonstrated the molar amounts of heme , on average , were 75–90% the molar concentration of protein , suggesting binding was stoichiometric for this preparation ( data not shown ) . However , upon removal of the heme and incubation of R54A with hemin , very little heme bound the protein , despite a far-UV spectrum indistinguishable from wild-type protein ( Figure 4D , Figure S4 ) . In the structure of apo and holo-IsdX1 , the side chain of Arg-54 shifts 2 . 5–2 . 8 Å to accommodate the heme ( Figure 2B ) . This suggests placement of a less bulky residue ( alanine ) in place of Arg-54 may allow heme access to the heme-binding site , but potentially only during translation of the protein when partially unfolded . Regardless of the exact reason for this , these results demonstrate residues in and around the 310-helix are involved in the binding of heme to IsdX1 . In an attempt to provide a more quantitative assessment of the importance of each residue in the 310-helix to the stabilization of bound heme , we measured the rates of heme dissociation from wild-type and mutant IsdX1 proteins . Each protein was purified from E . coli as described in the Materials and Methods , reconstituted with hemin , and holo-protein purified away from unbound hemin by gel filtration chromatography . The rate of heme dissociation was then assessed by mixing holo-IsdX1 preparations with excess H64Y/V68Y apo-myoglobin ( Mb ) , a mutant globin with a high heme affinity ( Kd∼10−12 M ) and very low rate of heme dissociation [30] , [48] , [49] . The dissociation rate constant of heme loss from IsdX1 can be determined by measuring the spectral changes that occur with time as released heme is scavenged passively by the apo-Mb reagent . As observed in Figure 5 , IsdX1 containing mutations in Ser-53 , Arg-54 , and Met-55 all lose heme significantly faster than wild-type IsdX1 , with S53A showing rates of heme loss that were greater than 400 times faster than the wild-type IsdX1 ( see Table 2 for rates ) . Interestingly , S52A showed comparable rates of heme dissociation to that of wild-type , despite its apparent poor heme binding ability at equilibrium ( Figure 4D ) . The best explanation for this is that while its rate of heme loss may be unaffected , its rate of heme association in the absence of hemoglobin may be poor . Taken together , the data are consistent with Ser-53 and Arg-54 of the 310-helix playing a substantial role in stabilizing the bound heme in IsdX1 . The position of 52-SSRM-55 extending over the heme-binding site implies these residues may also be involved in the direct interaction with hemoglobin . To test this hypothesis , the association of each mutant with holo-hemoglobin was investigated by surface plasmon resonance spectroscopy . Apo forms of wild-type or mutant IsdX1 were infused over holo-hemoglobin covalently coupled to a carboxy-methyl chip and the kinetics of binding recorded by quantifying the change in response units ( RUs ) with time . Whereas similar responses were observed for wild-type , S52A , and M55A IsdX1 , a significantly lower response was observed when S53A or R54A were infused over holo-hemoglobin ( Figure 6 , compare panels A , B , E to panels C , D ) . Indeed , estimations of the dissociation constants ( KD – Figure 6 legend ) from the association and dissociation phases of the response curves indicates that the S53A and R54A mutants , when compared to wild-type IsdX1 , display an approximately 10 and 386-fold increase , respectively , in KD , signifying a lower affinity for hemoglobin . There was no interaction of the wild-type or mutant IsdX1 proteins with apo-hemoglobin , which infers binding is dependent on a heme-bound conformation of hemoglobin or is facilitated by interactions with the solvent exposed heme propionates in holo-hemoglobin ( Figure 6F ) . These results indicate Ser-53 and Arg-54 are important in the interaction of the IsdX1 hemophore with holo-hemoglobin and collectively suggest the processes of heme binding and hemoglobin association in NEATs are coupled for some residues but not others . IsdX1 and IsdX2 promote the growth of B . anthracis on hemoglobin as the sole source of iron [26] . To determine the functional contribution of the 52-SSRM-55 helix towards heme scavenging activity , we tested the ability of wild-type and mutant proteins to rescue a hemophore-dependent growth defect of a B . anthracis strain ( ΔisdX1 , ΔisdX2 ) lacking both hemophores grown in iron-deficient media with or without hemoglobin . Although little growth is observed in the absence of hemophore , the addition of wild-type IsdX1 to cultures with hemoglobin led to a 3 to 8-fold increase in growth ( Figure 7 , compare 7 to 8 , from 4 to 8 hrs ) . This enhancement of growth is not due to heme or iron contamination of the protein preparation since only a marginal increase in growth was observed in the absence of hemoglobin ( Figure 7 , compare 8 to 1 , 2 ) . Interestingly , whereas ΔisdX1 , ΔisdX2 B . anthracis supplemented with S52A or M55A IsdX1 provided intermediate growth ( Figure 7 , compare 9 , 12 to 3 , 6 at 8 hrs ) , the level of replication in the S53A and R54A IsdX1 supplemented cultures was similar to the S53A/R54A-only controls , suggesting these proteins are unable to rescue a ΔisdX1 , ΔisdX2-dependent growth defect on hemoglobin as the sole iron source ( Figure 7 , compare 10 , 11 to 4 , 5 at 8 hrs ) . Taken together , these results indicate Ser-53 and Arg-54-mediated association of IsdX1 with hemoglobin is important for heme scavenging in iron-limiting environments and provides the first experimental demonstration of the biologic function of this dynamic region in NEAT proteins in growing cells .
Here , we report ( i ) the crystal structure of the apo and holo forms of a Gram-positive hemophore , ( ii ) the structure of a non-staphylococcal NEAT protein , ( iii ) that residues in and adjacent to the 310-helix contribute to the binding of the heme-iron in this hemophore , ( iv ) that the processes of heme binding and hemoglobin association can be delineated , and ( v ) that both optimal heme and hemoglobin binding are necessary for full hemophore activity for growing bacilli . Thus , these studies extend our knowledge of the molecular mechanism of NEAT protein function and provide evidence that abolishing a functional interface between the NEAT domain and hemoglobin can slow bacterial growth in iron-limiting environments . Research into bacteria hemophores is a growing field , and several hemophores have been discovered [10] , [16] , [50] , [51] . The most well-documented secreted hemophore is that of HasA from the Gram-negative pathogen S . marscescens [52] . Although functionally similar , it is likely that IsdX1 and HasA resulted from convergent evolution , as there is little structural similarity between the two proteins . Whereas HasA obtains heme from hemoglobin through a passive mechanism that does not seem to require a physical interaction , IsdX1 binds hemoglobin directly , an event that likely facilitates heme transfer [15] , [24] , [26] . Both proteins transfer the bound heme to their cognate cell surface receptors ( IsdC for IsdX1 and HasR for HasA ) through direct engagement . However , the IsdX1-IsdC interaction is dependent on the hemophore being heme loaded and is transient [30] . This contrasts with the HasA-HasR interaction , where both the apo and holo forms of HasA bind with similar affinities and to the same site on HasR [15] , [53] . More recently , several secreted hemophores have been reported . HmuY , from P . gingivalis , binds heme and may deliver its heme to the surface receptor HmuR [54] , [55] . Also in P . gingivalis , the recently described HusA is a heme-binding protein that is needed for growth under conditions of heme limitation [12] . A putative secreted hemophore from Mycobacterium tuberculosis has been characterized with a proposed heme-binding site consisting of one Tyr and two His , which has a similar heme-binding structural motif to that of HasA [56] . However , the overall fold of the Mtb hemophore is structurally diverse in comparison with both HasA and IsdX1 folds , and it was postulated that the Mtb hemophore may also be a product of convergent evolution [56] . Recent work has shed new insights into how Gram-positive bacteria acquire heme from mammalian hosts [57] , [58] . The central structural unit is the NEAT domain , a protein module that mediates heme acquisition and import at the bacterial surface [26] , [29] , [31] , [41] , [57]–[61] . If NEAT proteins are to be targets for the development of anti-infectives , the molecular determinants of their mechanism of action need to be defined . In this context , the study of IsdX1 is particularly appropriate because this protein contains the only known NEAT domain that contains all the activities associated with hemophore activity ( heme and hemoglobin association , heme extraction from hemoglobin , and heme transfer to receptors ) [26] . Thus , the study of IsdX1 allows for insights into the relationship between heme coordination , hemoglobin binding , and heme extraction from mammalian globins . To gain insights into these activities , we solved the 3-dimensional structure of apo and holo IsdX1 . The overall structure is similar to the structures of the staphylococcal NEAT domains of IsdH ( NEAT 1/3 ) [40] , [44] , IsdA [41] , IsdC [42] , [43] , and IsdB ( NEAT 2 ) [62] , with three common features: ( i ) an immunoglobulin-like fold arranged into eight β-sheets , ( ii ) a small 4–5 residue 310-helix extending over the heme-binding pocket , and ( iii ) two anti-parallel β-sheets that house a conserved tyrosine ( Tyr-136 in IsdX1 ) which coordinates the heme-iron ( Figure S2A ) . However , surface charge calculations indicate a charge distribution on IsdX1 quite distinct from the other NEAT domains , including a net positively charged region near the 310-helix of the heme pocket ( Figure S1 ) . It is not known if this difference relates to the fact that IsdX1 acts extracellularly after secretion , as opposed to the staphylococcal NEAT proteins that are covalently anchored to the cell wall . Of note is the finding that the staphylococcal homolog of the putative receptor for IsdX1 ( IsdC ) contains overall anionic character in this region , a feature which leads one to postulate that opposing charges may partially dictate the association between these two proteins upon heme transfer . The structures of apo-IsdX1 and holo-IsdX1 are highly similar with RMSD of 0 . 51 Å . However , superimposition of both apo and holo structures highlight subtle residue sidechain conformational differences surrounding the heme binding site . This finding suggests residues in the 310-helix may partially stabilize the heme-free form ( discussed below ) , a requirement of this hemophore immediately after secretion into host tissues . To accommodate the heme molecule , there is a shift in the residue sidechains of Arg-54 ( 2 . 5–2 . 8 Å ) , Arg-57 ( 7 . 3 Å ) , Tyr-58 ( 2 . 1 Å ) , Tyr-136 ( 1 . 6 Å ) and Tyr-140 ( 2 . 6 Å ) , away from the center of the heme binding pocket . Additionally , there is observable electron density for two side chain conformations for Met-55 within the apo structure , where one conformation is pointing toward the center of the heme binding pocket and the other away into the interior of IsdX1 . Met-55 in the holo structure has its side chain pointing into the interior of the protein and makes hydrophobic contacts with the heme pyrrole ring . In contrast , there is no conformational change observed for either Ser-52 or Ser-53 between the holo and apo structures . It is interesting to note that within the holo-IsdX1 structure , crystal lattice packing interactions occur through a heme-mediated protein interface between two holo-IsdX1 molecules from adjacent asymmetric units ( Figure S5A , B ) . Two heme molecules from adjacent subunits planar stack upon one another ( inter-iron distance of 5 . 7 Å ) and the crystallographic interface is further stabilized by two hydrogen-bonds; one from the NH2 group of Arg-54 from Chain A to the most solvent exposed heme propionate from Chain B and the second from the NH1 group from Arg-57 from Chain B to the hydroxyl group of Asn-135 on the adjacent molecule . The apo-IsdX1 structure also has a similar protein interface near the vicinity of the heme-binding sites of adjacent monomers; however due to the lack of heme , there are more protein interactions between loop regions surrounding the active site ( Figure S5A , B ) . We believe the heme stacking interaction , while caused by crystal packing , may reflect how IsdX1 transfers heme to downstream NEAT domain proteins such as B . anthracis IsdC . Indeed , Grigg et al ( published during review of this paper ) proposed that inter-NEAT domain interactions between S . aureus IsdB and IsdA may occur along the heme-mediated crystal symmetry interface based on in silico docking predictions [63] . Moreover , an NMR analysis by Villareal et al ( also published during review ) of the interaction between S . aureus IsdC and IsdA demonstrated that in the IsdC-IsdA complex the proteins are pseudosymmetrically arranged through a 180° rotation that is coplanar with the heme plane , which is identical to the symmetry relationship between IsdX1 monomers from adjacent asymmetric units [64] . Thus , our data also lend experimental support to these models . With respect to the residues in and around the 310-helix of IsdX1 , several interesting properties can be gleaned from this study . First , amino acids in this region contribute to the stabilization of heme , as demonstrated by the fact that the substitution of Ser-52 , Ser-53 , and Met-55 to alanine abrogated the ability of IsdX1 to attain heme from E . coli lysates as well as bind pure hemin when incubated with the apo protein . Mutation of Arg-54 led to significantly higher amounts of heme co-purifying with IsdX1 . However , removal of this heme and assessment of hemin binding yielded a protein unable to subsequently coordinate heme . We cannot definitively explain this result , other than to propose that it is possible that while being synthesized in E . coli , the partially unfolded IsdX1 binds the heme and then folds thereby enclosing upon the iron-porphyrin . Upon heme removal , the resulting apo protein , potentially highly ordered , now becomes restricted and does not allow heme to access the binding pocket . Interestingly , this was not observed in an IsdA variant harboring an alanine substitution in the equivalent position ( His-83 ) [41] . It was proposed this position in IsdA and IsdC ( Ile-48 ) sterically hindered access to the sixth coordination position of the heme-iron . Indeed , cyanide and azide , two ligands often used to probe accessibility to the sixth position , did not bind IsdA or IsdC [42] , [65] . The ability of the E . coli form of R54A to associate with more heme while expressed in E . coli may provide experimental support for this hypothesis , with Arg-54 sterically blocking access to Try-136 in the wild-type protein . Second , mutation of this region decreases hemoglobin association; however , differential effects are observed . Substitution of Ser-53 and Arg-54 with alanine significantly reduced binding to hemoglobin , with R54A yielding the largest effect ( greater than 300-fold ) . However , mutation of Ser-52 and Met-55 , the two residues flanking Ser-53 and Met-54 , produced hemoglobin-binding affinities similar to the wild-type protein . These findings support a model by which Arg-54 , being rather forward in its location over the heme-binding site , provides initial contact with hemoglobin , perhaps stabilizing the initial interaction . The engagement of Arg-54 with hemoglobin may “peel” the side chain of this residue away from the heme-binding site , thus removing the steric block observed for residues in this position for other NEAT proteins . Further support for this hypothesis is that the hemoglobin-binding NEAT domain , IsdB-N2 , has a Met in the same position as Arg-54 , and the crystal structure of IsdB showed an alternate conformation of Met coordinated with heme-iron , whereby the authors suggest that this Met might be involved in heme transfer . If alternate conformations of Arg-54 are sampled , one conformation has the NH1 group of Arg-54 coordinated to heme-iron similar to IsdB ( Figure S2 , lower panel ) . Thus , Arg-54 , by virtue of its unique orientation in the heme-binding pocket , may mediate several processes related to heme extraction from hemoglobin . In essence , this residue may function as a “molecular placeholder” by initially binding the heme during the first step of transfer . In support of this hypothesis , there is evidence that arginine can assist in heme binding in other systems [66] . Ser-53 , whose side chain is directed towards the heme-binding site , would next be free to hydrogen bond with a heme propionate , thereby strengthening the IsdX1-hemoglobin interaction . Once Ser-53 engages the heme , Tyr-136 of IsdX1 , from the opposite side of the heme , can now displace hemoglobin's His-iron coordination to become the fifth axial ligand of IsdX1 , with Tyr-140 providing additional strength for this coordination by H-bonding to the phenolate of Tyr-136 . In addition to Tyr-136 , the heme is also secured in the binding pocket by Ser-53 and Arg-54 , since these residues show the highest rates of heme loss when mutated . Ser-52 , whose mutation does not lead to greater heme dissociation , may also assist in drawing the heme into the heme-binding pocket but once the heme is in , does not contribute much to its stabilization . This hypothesis aligns well with the observation that the side chain of this residue sticks out into the solvent while the side chain of its neighbor , Ser-53 , points into the heme-binding pocket . During review of our manuscript , Kumar et al described the first ever co-crystal of a NEAT protein ( the first NEAT domain of IsdH from S . aureus ) in complex with the alpha-chain of hemoglobin [67] . The structure reveals several stabilizing contacts between the 310-helix of IsdH and hemoglobin , thereby confirming our predictions for the role of this helix in direct association with hemoglobin and heme scavenging . Perhaps most important is a hydrogen bond between a serine in IsdH NEAT 1 ( Ser-130 ) with Lys-11 on hemoglobin . This serine is just 5 residues downstream of what would be the position of Ser-53 in IsdX1 , leading us to speculate that the lack of functional interaction with hemoglobin observed upon mutation of this residue is because a key hydrogen bond is severed in the IsdX1-Hb interaction . Whether or not the function of the 310-helix is confined only to heme acquisition from hemoglobin remains to be determined . For example , Grigg et al found the conserved coordinating tyrosine in the first NEAT domain of IsdA ( the equivalent of Tyr-136 in IsdX1 ) , and not residues in the 310-helix , played a role in NEAT to NEAT heme transfer [63] . In contrast , Villareal et al found that mutation of a single 310-helix residue in S . aureus IsdA did affect NEAT to NEAT heme transfer; however , this mutation was coupled to another mutation elsewhere in the protein that was also believed to mediate NEAT-NEAT binding [64] . Thus , more work is required to determine if the 310-helix also functions in downstream heme transfer processes . Although this model of hemoglobin and heme binding is consistent with available data , we cannot rule out that there are additional residues outside of the 310-helix that participate in this process . Indeed , as demonstrated by Pilpa et al [46] for one of the staphylococcal hemoglobin receptors ( NEAT 1 of IsdH ) [68] , [69] , amino acids distal to this position ( on the β3–β4 loop ) , also seem to be important for hemoglobin association . However , NEAT 1 of IsdH does not bind heme , and instead requires a third NEAT domain ( NEAT 3 ) , to acquire and stably bind the heme from the IsdH NEAT 1-hemoglobin complex . Thus , the data suggests residues in and around the 310-helix in IsdX1 evolved to bind both heme and hemoglobin , as well as those that are not interdependent , in order to provide several functionalities ( hemoglobin binding coupled to heme extraction ) within a single NEAT domain . Finally , the exogenous addition of recombinant IsdX1 to culture restored the growth of a hemophore-deficient strain of B . anthracis on hemoglobin , suggesting all the information necessary for hemophore activity is contained within its amino acid sequence . The partial restoration of growth observed by S52A and M55A may be due to the ability of these mutants to still bind heme , albeit poorly , after association with hemoglobin . Although S53A and R54A bind heme at similar levels as S52A and M55A , it would seem their much lower affinity for hemoglobin precludes them from acquiring any heme in this assay ( or alternatively , does not promote the release of enough heme from hemoglobin ) , meaning no or little free heme is available for transfer ( or sequestration ) at the bacterial surface . Because mutation of Ser-52 , Ser-53 , Arg-54 , and Met-55 results in poor heme-binding activity , it is difficult to biochemically determine the exact role of these residues in the ability of IsdX1 to extract heme from hemoglobin . However , it is clear from the SPR and growth studies that mutations that effect the association with hemoglobin significantly compromises hemophore activity . Performing an identical experiment with a single mutant ( ΔisdX1 ) strain did not yield a significant enough phenotype to evaluate these mutants for function ( data not shown ) , likely because IsdX1 and IsdX2 are functionally redundant [26] . Taken as a whole , these experiments highlight the biological role of the 310-helix in NEAT protein function and hemophore biology and provide direct evidence that the mechanistic extraction and binding of heme from hemoglobin are important for B . anthracis replication in low-iron environments . Although attempts to develop a universal inhibitor of heme uptake is likely to be challenging , the seemingly conserved functional properties shared by distinct NEAT domains does offer the prospect of targeting these processes in select Gram-positive bacteria . As we make advances in our understanding of NEAT mechanism of action , the elucidation of additional structures , an understanding of side chain chemistry in transfer reactions , and the identification of factors that drive ligand binding specificity , will all be required for the creation of novel anti-infectives that prevent iron-porphyrin uptake in these pathogenic bacteria .
DNA encoding for amino acids 26–146 of IsdX1 were amplified from the genome of B . anthracis strain Sterne 34F2 using PCR and primers containing EcoRI and BamHI restriction sites [70] . Amplified DNA was digested , ligated into pGEX2TK , and pGEX-IsdX1 transformed into E . coli BL21 for the expression of IsdX1 as a glutathione-S-transferase ( GST ) fusion protein as described [26] , [29] . For the purification of IsdX1 used to generate the holo-protein crystals , DNA encoding residues 27–152 was cloned into pET28a ( Novagen ) encoding a fusion protein of IsdX1 with a His ( 6 ) -tag using NheI and NotI restriction enzyme sites . The 26–146 ( apo ) and 27–152 ( holo ) protein constructs were chosen because they yielded crystals that diffracted with the highest resolution . Site-directed mutagenesis of IsdX1 was performed on the 26–146 IsdX1 construct using QuikChange ( Stratagene , Santa Clara , CA ) according to the manufacturer's instructions . After DpnI digestion of reaction mixtures , DNA was transformed into E . coli BL21 and the resultant plasmid clones sequenced to confirm the presence of the mutation [71] , [72] . All E . coli strains were grown in Luria-broth ( LB ) supplemented with 50 µg/mL ampicillin ( Fisher Scientific , Waltham , MA ) except for the pET28a construct that was supplemented with 30 µg/mL kanamycin ( Fisher ) . Wild-type and mutant IsdX1 proteins were purified by GST-affinity chromatography as previously described [26] , [29] . Briefly , 50-mL of overnight cultures of E . coli BL21 containing wild-type or mutant pgst-isdX1 were inoculated into 2-L of LB with ampicillin and rotated at 250 rpm at 37°C . After 2 hours , isopropyl- β-D-thiogalactopyranoside ( IPTG - 1 . 5 mM ) was added and cultures grown for an additional 2 hours . Bacteria were centrifuged at 6 , 000×g for 8 min , resuspended in phosphate buffered saline ( PBS - 137 mM NaCl , 2 . 7 mM KCl , 10 mM sodium phosphate dibasic , 2 mM potassium phosphate monobasic , pH 7 . 4 ) , and cells lysed by French press . The supernatant was obtained by centrifugation at 30 , 000×g for 15 min and filtered through a 0 . 45-µm pore-size cellulose filter . Lysates were next applied to 1-mL of glutathione-sepharose ( GE Healthcare , Piscataway , NJ ) , washed with 40-mL of PBS , and bound protein incubated with thrombin ( 100 units , GE Healthcare ) for 3 hours at 25°C . GST-free IsdX1 preparations were next incubated with 200-µL of aminobenzamidine sepharose ( Sigma , St . Louis , MO ) to remove thrombin . To remove endogenous heme , preparations were treated with HCl ( final pH of 2 . 0 ) and methyl ethyl ketone was added to separate heme ( organic layer ) from IsdX1 ( aqueous layer ) as described [73] . Protein concentrations were either determined by UV/vis spectroscopy , the bicinchoninic acid method ( Pierce , Rockford , IL ) , or by SDS-PAGE [74] . All protein preparations were stored at −20°C . apo-IsdX1 - To obtain an IsdX1 preparation for crystal seeding , IsdX1 was expressed and bound to glutathione-sepharose as described above [26] , [29] . Bound protein was then cleaved from the column with Factor-XA ( 10 units , Amersham Biosciences ) in 1 mM CaCl2 , 100 mM NaCl , 50 mM Tris pH 7 . 9 for 16 hours . Heme was removed as described above and IsdX1 further purified by cation exchange chromatography using a Mono S column ( GE Healthcare ) and gel filtration chromatography using a Superdex 200 column ( GE Healthcare ) . Selenomethionine-substituted protein was produced by inhibiting methionine biosynthesis and purified as above . Holo-IsdX1 - IsdX1 was transformed into BL21-Gold ( DE3 ) cells and grown at 37°C in LB medium containing 30 µg/mL kanamycin . Protein expression was induced when cells reached OD600 nm of 0 . 8 by the addition of 1 mM IPTG and cells harvested after 4 hours by centrifugation at 5 , 100 rpm for 20 minutes , followed by resuspension in 50 mM Tris , pH 7 . 4 , and 350 mM NaCl . Cells were next lysed by sonication after addition of egg hen lysozyme ( 5 mg , Sigma ) with phenylmethylsulfonyl fluoride ( 40 µM , Sigma ) and the cell lysate centrifuged at 14 , 000 rpm for 20 minutes . After addition of 400 µL Proteoblock protease inhibitor cocktail ( Fermentas ) , the supernatant was loaded onto a Ni2+-charged HisTrap column ( GE Healthcare ) and eluted with a linear imidazole gradient ( between 100–250 mM imidazole ) . Fractions containing IsdX1 were identified by SDS-PAGE , pooled and concentrated using a Centricon centrifugal concentrator ( Millipore ) . Further purification of IsdX1 was achieved by running the protein over an S75 gel filtration column ( GE Healthcare ) equilibrated with 50 mM Tris pH 7 . 4 , 150 mM NaCl , which yielded nearly 100% homogeneous protein . Cleavage of the His ( 6 ) -tag was conducted in cleavage buffer ( 50 mM Tris pH 7 . 4 , 150 mM NaCl , 10 mM CaCl2 ) by adding 1 mL of thrombin-agarose suspension ( Sigma ) to the protein , followed by removal of thrombin-agarose on a glass frit . IsdX1 was then run over an S75 gel filtration column equilibrated with 50 mM Tris pH 7 . 4 , 150 mM NaCl to separate IsdX1 from the His ( 6 ) -tag . apo-IsdX1 - Purified IsdX1 in buffer A [10 mM MES pH 6 . 6 , 200 mM NaCl] crystallized at room temperature using the hanging drop vapor diffusion method and a reservoir solution of 0 . 1 M citric acid pH 3 . 5 and 2 M NaCl . Crystals were frozen in N2 ( l ) following cryoprotection with the reservoir solution containing 16% glycerol . Data were collected to 1 . 8 Å at the Structural Biology Center beamline 19-BM at the Advanced Photon Source ( APS ) , Argonne National Laboratory ( ANL ) , and processed using HKL2000 [75] . SeMet-IsdX1 crystallized with a reservoir solution of 0 . 1 M citric acid pH 3 . 5 and 3 . 3 M NaCl and was frozen as above . Data were collected to 2 . 1 Å at the Life Sciences Collaborative Access Team ( LS-CAT ) beamline 21-ID-D at the APS , and processed with XDS and scaled using SCALA ( Table 1 ) [76] . The structure was determined by the single-wavelength anomalous dispersion method using the anomalous scattering from two selenium atoms in the asymmetric unit cell with the program PHENIX [77] . The model was built automatically with PHENIX and manually with Coot[78] and refined with PHENIX . Holo-IsdX1 - Approximately 4 mg of heme were dissolved in 0 . 4 mL of ice cold 0 . 1 M NaOH and vortexed periodically . After 30 minutes , 0 . 4 mL of 1 M Tris , pH 7 . 4 was added to the solution . The solution was subsequently centrifuged for 10 minutes at 4°C at 13 , 000 rpm . The heme solution was then diluted with 50 mM Tris , pH 7 . 4 , 150 mM NaCl and centrifuged again at 5 , 100 rpm to remove any heme aggregates . Final concentrations were determined using ε385 = 58 . 44 mM−1 cm−1 . Heme solutions were used within 12 hours . Holo-protein was reconstituted by slowly adding 1 . 5-fold excess heme to IsdX1 preparations in small increments . The UV/vis absorption spectrum was recorded after each addition of heme and saturation with heme was evident by the shift in the Soret maximum from 399 nm to a shorter wavelength . After 1-hour incubation at room temperature , excess heme was removed using an S200 gel filtration column ( GE Healthcare ) and the protein collected in 1-mL fractions . The UV/vis absorbance spectrum for each fraction was recorded and those with a Soret peak maximum at 399 nm and Abs399/Abs280 ratios greater than 3 were pooled . Protein concentrations were measured using the modified Lowry assay ( Pierce , Rockford , IL ) . The extinction coefficient for the Soret peak at 399 nm was determined to be equal to 100 mM−1 cm−1 by the pyridine hemochrome assay and used for hemoprotein concentration determination [79] . Holo-IsdX1 was concentrated to 100 mg/mL in 50 mM Tris pH 7 . 4 , 150 mM NaCl for crystallization trials . Crystals of holo-IsdX1 were grown by the hanging drop , vapor diffusion method against a reservoir containing 0 . 1 M SPG buffer , pH 9 . 0 and 25% polyethylene glycol 1500 with crystallization drops containing 0 . 25 µL of protein to 0 . 25 µL of reservoir solution . Crystal growth was observed within 24 hours . The crystal was passed through a 1∶1 v/v solution containing the reservoir solution and glycerol for cryoprotection and the crystals harvested under cryoconditions . The diffraction data was collected at 77 K on beamline 9-2 at Stanford Synchrotron Radiation Lightsource . Two data sets were collected , a native set at 1 . 0 Å and a Fe-SAD set at the iron absorption edge ( 1 . 738 Å ) . Images were indexed , integrated and reduced using the HKL2000 suite resulting in a 99 . 1% complete dataset to 2 . 15 Å resolution for the native set and a 98 . 2% complete dataset to 2 . 3 Å resolution for the Fe-SAD set . The unit cell dimensions are equal to 65 . 1 Å×65 . 1 Å×74 . 4 Å in the space group P43 with two molecules of holo-IsdX1 per asymmetric unit ( Table 1 ) . The initial phases were calculated using Phaser [80] with apo-IsdX1 as the search model and the resulting electron density map was subjected to automated and manual model building procedures through a combination of PHENIX . autobuild [77] , PHENIX . refine [77] and Coot [78] . Subsequently , two heme-iron sites per asymmetric unit were detected from the anomalous iron signal using SHELXC/D/E [81] . This additional phase information was combined with the molecular replacement model phases through SIGMAA and DM in CCP4i [82] , which improved the resulting electron density map for a final round of refinement . Programs from the CCP4 package [82] as well as Phenix [77] , Pymol [83] , and Coot [84] were used to analyze the stereochemistry and geometry of the models and were found to be acceptable . Data collection and refinement statistics are presented in Table 1 . Apo forms of wild-type or mutant recombinant IsdX1 ( 5 µM ) were incubated with hemin chloride ( 5 µM – Sigma ) in PBS , pH 7 . 4 for 10 minutes at 25°C . Reactions were then subjected to a wavelength scan from 250–560 nm using a DU800 spectrophotometer ( Beckman-Coulter , London , UK ) [29] . The relative amount of bound heme was calculated as: ( T399 nm−C399 nm ) /T280 nm , where T399 nm = the absorbance maximum at 399 nm for samples containing IsdX1 with heme , C399 nm = the absorbance maximum at 399 nm for samples containing heme only , and T280 nm = the absorbance maximum at 280 nm for samples containing IsdX1 with heme . To measure the rates of heme dissociation , wild-type or mutant IsdX1 were re-constituted with heme , excess heme removed by gel filtration chromatography , and holo proteins ( 1 µM ) mixed with apo-Mb ( 26 µM ) at 25°C in PBS , pH 7 . 4 . Dissociation rates are determined by measuring the increase in absorbance at 419 nm ( apo-Mb ) versus a reference , control wavelength ( 380 nm ) . The interaction of wild-type and mutant IsdX1 proteins with hemoglobin was measured using a BIAcore 3000 biosensor ( Amersham Biosciences , Piscataway , NJ ) [30] . Briefly , 100 µL of holo- or apo-hemoglobin ( Sigma-H2500 , 10 µM in 50 mM Tris-HCl , pH 7 . 0 ) was covalently coupled to a CM5 sensor chip at 25°C to a density of 4000 response units ( RUs ) using amine chemistry as previously described [85] , [86] . Wild-type or mutant IsdX1 proteins ( 100 , 200 , 300 , 350 , or 500 nM ) in HBS-N ( 0 . 01 M HEPES , 0 . 15 M NaCl , pH 7 . 4 ) were injected at 20 µL/min for 300 s at 25°C and response curves followed for a total of 800 seconds . A parallel injection of IsdX1 over a blank CM5 surface ( no hemoglobin ) was used to control for non-specific binding . Injections at each concentration were performed in triplicate , and the data from the 300 nM injection used to calculate the dissociation constants , which were determined using BIAevaluation 4 . 1 software ( Amersham Biosciences ) after fitting the data to a 1∶1 Langmuir binding model with dR/dt = kaC ( Rmax−R ) −kdR , where R is the SPR signal ( in response units ) , ka is the association rate constant ( in M−1 s−1 ) , kd is the dissociation rate constant ( in s−1 ) , C is the concentration of holo-IsdX1 ( in M ) , Rmax is the maximum holo-IsdX1 binding capacity ( in response units ) , and dR/dt is the rate of change of the SPR signal [87] . Purified IsdX1 ( wild-type or mutant proteins ) were treated with HCl and methyl ethyl ketone to remove any endogenous heme [88] . Preparations were then dialyzed against 2-L of PBS ( pH 7 . 4 ) and treated with 100 mg/mL Chelex-100 ( Sigma ) to remove any contaminating free iron . B . anthracis Sterne strain ( 34F2 ) harboring deletions in both isdX1 and isdX2 [26] were subcultured into 3-mL of LB plus kanamycin at 30°C . After 12-hours , cells were washed 2× with PBS and 5-µL of washed , normalized cells inoculated into 500-µL of RPMI ( iron chelated by treatment with 100 mg/mL Chelex-100 for 12 hours ) with or without hemoglobin ( 10 µM ) in a 48-well Costar tissue-culture plate . Wild-type or mutant proteins ( 1 µM ) were next added to each well and OD600 recorded at 2 , 4 , 6 , and 8 hours at 37°C using a Tecan 200 Pro microplater reader . The results represent the mean and standard deviation of three independent experiments . Circular dicroism spectra of apo forms of wild-type and mutant IsdX1 were obtained using a JASCO-815 CD spectropolarimeter at 25°C [89] , [90] . Protein samples were resuspended in PBS buffer , pH 7 . 4 at a concentration of approximately 50 µM . Far-UV spectra were recorded from 200–260 nm using a 1-mm path length at a scanning speed of 50 nm/min , with a bandwidth of 2 nm . Raw spectra are shown and represent the average accumulation of six scans . The NCBI accession number for isdX1 is NC_005945 . 1 ( B . anthracis Sterne gene ID:2851614 ) .
|
Pathogenic bacteria need to acquire host iron to replicate during infection . Approximately 80% of mammalian iron is associated with a small molecule termed heme , most of which is bound to circulating hemoglobin and involved in O2 transport in red cells . Bacteria secrete proteins , termed hemophores , to acquire the heme from hemoglobin , a process thought to accelerate delivery of the heme to the bacterial surface for iron import into the cell . The mechanisms by which hemophores extract host heme from hemoglobin are not known . Here , we report that the IsdX1 hemophore from B . anthracis , the causative agent of anthrax disease , uses a conserved structural feature to link hemoglobin association with heme binding and extraction , thereby facilitating bacterial growth in low-iron environments . Such “molecular coupling” suggests that specific inhibition of the hemophore-hemoglobin interaction for this class of proteins may serve as a starting point for new anti-infective therapeutics aimed at short-circuiting iron uptake networks in bacterial pathogens .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biology",
"microbiology",
"bacterial",
"pathogens"
] |
2012
|
Differential Function of Lip Residues in the Mechanism and Biology of an Anthrax Hemophore
|
High tropical species diversity is often attributed to evolutionary dynamics over long timescales . It is possible , however , that latitudinal variation in diversification begins when divergence occurs within species . Phylogeographic data capture this initial stage of diversification in which populations become geographically isolated and begin to differentiate genetically . There is limited understanding of the broader implications of intraspecific diversification because comparative analyses have focused on species inhabiting and evolving in restricted regions and environments . Here , we scale comparative phylogeography up to the hemisphere level and examine whether the processes driving latitudinal differences in species diversity are also evident within species . We collected genetic data for 210 New World bird species distributed across a broad latitudinal gradient and estimated a suite of metrics characterizing phylogeographic history . We found that lower latitude species had , on average , greater phylogeographic diversity than higher latitude species and that intraspecific diversity showed evidence of greater persistence in the tropics . Factors associated with species ecologies , life histories , and habitats explained little of the variation in phylogeographic structure across the latitudinal gradient . Our results suggest that the latitudinal gradient in species richness originates , at least partly , from population-level processes within species and are consistent with hypotheses implicating age and environmental stability in the formation of diversity gradients . Comparative phylogeographic analyses scaled up to large geographic regions and hundreds of species can show connections between population-level processes and broad-scale species-richness patterns .
Phylogeographic studies leverage spatial and genetic data to examine the earliest stages of speciation , illuminating how populations differentiate across a landscape [1] . Comparisons across species show that the level of genetic structuring varies from deep phylogeographic breaks to unstructured panmictic populations [2] . This among-species variation in the amount and depth of phylogeographic structuring has been attributed to various factors , including differences in dispersal ability [3 , 4] , habitat preferences [5] , breeding phenology [6] , life history traits [7] , and the amount of evolutionary time in the landscape [8 , 9] . Because comparative phylogeographic studies usually examine species that occur within the same geographic region and in similar environmental and historical settings , the generality of associations between species traits and phylogeographic variation is largely unknown . Scaling phylogeography beyond the analysis of codistributed species and expanding comparative tests to multiple geographic assemblages of species would provide insight into whether the origins of genetic diversity link to large-scale biodiversity patterns . The latitudinal gradient in species richness is one of the most ubiquitous ecological patterns in nature [10] . Phylogenetic data suggest that higher tropical species richness is attributable to a multitude of factors , including higher long-term diversification rates [11 , 12] , niche conservatism [13] , and more time for speciation [e . g . , 14] . Latitudinal variation in phylogeographic structure is poorly understood , even though differences in diversification patterns among temperate and tropical clades could begin accumulating within species . Genetic divergence among populations has been shown to be higher in the tropics [15] , but divergence patterns in subspecies [16] and sister species [17 , 18] suggest that there is faster diversification in the temperate zone . Processes that result in greater population differentiation and/or less extinction over phylogeographic timescales would result in greater intraspecific diversity within a region , the effects of which could persist to deeper phylogenetic scales . Assuming intraspecific genetic diversity varies among regions [e . g . , 19] , comparative analysis of phylogeographic structure between the temperate and tropical zones should provide insight into the formation of a more general latitudinal biodiversity gradient . Habitat , landscape heterogeneity , and species dispersal ability are expected to influence phylogeographic structuring by determining how fragmented species distributions are and levels of gene flow between populations . These interactions among populations , however , may function differently across habitat types and regions . In tropical forests , higher available energy [20] and increased niche specialization along elevational gradients [21] and vertical habitat strata [22] are expected to lead to higher speciation rates . Greater climatic instability in temperate habitats during Pleistocene glacial—interglacial cycles has been linked to dynamic species histories of population expansion and contraction , higher extinction rates [23] , and in some cases higher speciation rates [e . g . , 17] . Species ecologies that differ between habitats and regions can also lead to differences in phylogeographic diversity . For example , genetic subdivision is often deeper in tropical species with lower dispersal abilities [4 , 7] . It is unclear if these traits have the same phylogeographic effect on temperate species , for which movements may be determined by changing climatic conditions more than intrinsic dispersal ability . Another possibility is that phylogeographic structuring may be random with respect to both the environment and ecology , such that the accumulation of genetic variation within a species may be due to its evolutionary persistence in the landscape [8] . The effects of environment , ecology , and persistence on species phylogeography are not mutually exclusive and may mediate broad patterns of intraspecific diversity among regions . Here , we used hemisphere-scale intraspecific data from New World birds to test whether phylogeographic diversity varies from the temperate region to the tropics . We compiled a large multispecies dataset of mitochondrial DNA ( mtDNA ) with environmental , trait , and morphological data for each species . The bird species included in this study inhabit various ecoregions , including tropical lowland and montane rainforests , deserts , temperate deciduous forests , and montane pine forests . We used a Bayesian coalescent model to quantify the degree of phylogeographic structure from each species’ time-calibrated mtDNA gene tree . To test for latitudinal biodiversity gradients below the species level , we asked whether or not the degree of phylogeographic structure declined with increasing latitude . We additionally assessed and accounted for the relative influence of a range of variables characterizing environmental preferences , contemporary and historical environmental conditions , life history , morphology , and range sizes ( S1 Table ) on phylogeographic diversity . Finally , we evaluated possible population-level mechanisms for an intraspecific diversity gradient by assessing whether metrics of the rates of formation and loss of phylogeographic structure in species are tied to latitude . By characterizing broad-scale patterns of phylogeographic diversity and investigating their environmental , ecological , historical , and population-level causes , we provide powerful insight into how biodiversity patterns form at the early stages of diversification .
We assembled mtDNA data from 17 , 573 individuals , representing 210 species , for which we collected environmental data from 67 , 779 observational records and morphological data from 1 , 139 museum specimens . The proportion of species sampled compared to the total diversity was higher in temperate North America versus tropical South America ( Fig 1A ) , but the number of species sampled was higher in the tropics ( Fig 1B ) . Species occurring in the highest latitude areas , such as temperate South America , were underrepresented in our dataset . Genetic sampling within species was poorest in areas shown in red ( Fig 1C ) ; for example , the northern limits of species occurring in Canada are undersampled . Our sampling of foraging guilds ( Fig 1D ) and body sizes ( Fig 1E ) qualitatively reflected their relative diversities with respect to unsampled species , except that very large birds are underrepresented in phylogeographic studies . To estimate phylogeographic structure , we required a metric that was comparable across species with varying degrees of sampling . The multispecies coalescent provides a framework for measuring genetic structure within species , and we used a Bayesian implementation demonstrated to perform well with variable sampling [24] . We found , on average , that our focal species ( n = 210; S2 Table ) included multiple genetic clusters ( mean = 2 . 710; standard deviation [SD] = 2 . 310; range: 1:17 ) that were geographically structured . Using phylogenetic generalized least-squares ( PGLS ) analysis , we evaluated the association between phylogeographic structure and latitude while accounting for phylogenetic nonindependence . We used multivariate models to account for alternative explanatory variables characterizing species ecologies , life histories , and habitats . The explanatory powers of the multivariate models were compared using a version of the Akaike information criterion ( AICc ) . The latitudinal midpoint of species’ distributions was a significant effect in the model , after accounting for all other variables ( S3 Table ) . Because the model also included the age of each species , the latitudinal trend in phylogeographic structure is not merely attributable to latitudinal variation in species ages ( S3 Table ) . Decreasing phylogeographic structure with latitude was more pronounced when we included species age based on stem age ( ΔAICc = 16 . 059 ) , the timing of divergence from the last common ancestor , as a covariate versus crown age ( ΔAICc = 3 . 163 ) , the time when all haplotypes coalesce within a species . The ΔAICc shows the change in the AICc score between the full model and a model excluding the predictor , with a ΔAICc > 2 considered a significant factor in the model . The models also accounted for sampling bias linked to the proportion of each species range sampled , which had a stronger effect in the model using stem age ( ΔAICc = 2 . 862 ) versus crown age ( ΔAICc = 1 . 353 ) . Heat maps that reflect the phylogeographic structure across species occurring in each pixel provide a visual evaluation of trends across regions ( Fig 2A–2C ) . The mean ( Fig 2B ) and SD ( Fig 2C ) of phylogeographic structure are higher in species occurring at lower latitudes in the tropics . We assessed whether higher phylogeographic structure in tropical species was an artifact of different sources of bias ( Fig 3 ) . Our PGLS analyses were robust to using an alternative taxonomy , examining passerines only ( n = 178 ) , excluding species that were outliers with respect to phylogeographic structure ( n = 199; excluding structure estimates greater than the 95% quantile ) , and uncertainty in estimates of population structure or species age ( Fig 3 bottom; S3 Table ) . To account for potential taxonomic biases , we compared our results using species currently recognized by the Checklist committees of the American Ornithological Society ( AOS; n = 210; S2 Table ) to a second treatment consisting of more inclusive monophyletic groups representing either single species or species complexes ( hereafter “lumped” species; n = 179; S2 Table ) . Taxonomic uncertainty arises where two currently recognized species are allopatric; thus , combining species complexes into single species should alleviate artifacts caused by uneven taxonomic splitting across species . The inclusion of both taxonomic treatments allowed us to assess if any of our results were caused by latitudinal biases in the frequency of paraphyletic species [26] or in species delimitation [27] . We found similar results in both taxonomies ( Fig 3 ) . An alternative way to account for taxonomic bias , assuming that taxa that contain multiple species are older , is to condition on species age in the multivariate models and assess whether the correlation remains significant . We found that latitudinal midpoint still showed a negative ( β = −0 . 001 ) and significant ( p < 0 . 00003 ) correlation with phylogeographic structure , even when we corrected for species stem age . Additionally , we randomized the phylogeographic structure value in each species to produce a null distribution and compared univariate models examining latitude using randomized versus observed values . The association between latitude and phylogeographic structure was significantly different than the null expectation ( p = 0; S2 Fig; S3 Table ) . Even though our sensitivity analyses cannot exhaustively account for all potential biases in species limits , we show that latitudinal variation in phylogeographic structure was robust to alternative taxonomies and species ages and significantly different from a null model . Despite sampling birds in diverse environments and with varied ecologies , these traits were often poor or inconsistent predictors of phylogeographic structure . From our measurements of museum specimens , we compiled data on species with a wide range of wing shapes ( hand-wing index range: 4 . 873–64 . 327 ) , body sizes ( tarsus length range: 4 . 25–64 . 48 mm ) , and life history strategies , ranging from sedentary taxa to long-distance seasonal migrants ( max = 7 , 833 km; S2 Table ) . We found that proxies for dispersal ability ( hand-wing index ) , body sizes ( tarsus length ) , and migratory distances were not significantly correlated with phylogeographic structure ( S3 Table ) . Mean elevational preference , overall landscape ruggedness , and changes in climate since the Last Glacial Maximum were also not generally significant effects in our models , but many of the contemporary environmental variables that covary with latitude were ( S3 Table ) . At the macroscale , phylogeographic diversity primarily varied along latitude , irrespective of the inclusion of other variables . Three primary processes might be responsible for higher phylogeographic diversity in the tropics: a higher rate of splitting of phylogeographic clusters , lower rate of extinction of phylogeographic clusters , or more time for phylogeographic structure to accrue in tropical species . Although most phylogeographic trees contain too few lineages to jointly estimate splitting and extinction rates within each species , they do contain information on the relative rate at which phylogeographic structure was formed and lost [28] that can be used to compare broad patterns across species . Diversification processes are more challenging to estimate from phylogeographic data of extant species than are the indices of phylogeographic diversity examined above , but latitudinal variation in these processes is of broad interest , and we address each process here .
Our comparative phylogeographic analysis of hundreds of New World bird species demonstrated that intraspecific genetic variation exhibits a pronounced latitudinal gradient . In comparison to temperate species , we found that tropical species are older and have accumulated and maintained more phylogeographic structure . These patterns are remarkably consistent with studies based on species-level data that show higher species richness [10] , older taxa [30] , and lower extinction in the tropics [11 , 31 , 32] . Although phylogeographic structure may not persist into the deeper evolutionary timescales examined in phylogenetic studies of species richness [33 , 34] , the concordant patterns across timescales suggest that similar processes may be responsible in both cases . Overall , our results demonstrate that latitudinal diversity gradients are evident at shallow evolutionary timescales and that comparative phylogeographic data are useful for examining patterns of diversity at large geographic scales . Higher climatic and environmental stability in the tropics has been implicated as a mechanism producing the latitudinal biodiversity gradient [10 , 35] . Pleistocene glacial—interglacial cycles had a global effect on species distributions and habitats [23] , but these environmental effects were particularly profound in northern latitudes , where large ice sheets covered much of the terrain [36] . Our estimate of lineage loss , a standardized metric of stem branch length , showed the predicted latitudinal pattern of higher extinction in the temperate zone , where climatic conditions were most volatile . These findings reiterate that the lower genetic diversity in temperate species is due to long-term historical processes and not to human-modified changes to the landscape , as suggested by recent work [19] . Although long stem branches do not provide information on the magnitude of the number of lineages lost over time , their lengths are nevertheless indicative of relative levels of extinction and pruning [37 , 38] . Our results are consistent with studies directly estimating extinction at phylogenetic scales and among sister species , which found higher extinction rates in temperate birds [17 , 31] . Long stem branches in high-latitude species could alternatively reflect a failure of populations within northern species to diversify until recently , but this explanation seems unlikely , given the lack of an obvious biological reason ( e . g . , long-term environmental stability ) for historical evolutionary stasis . A potential means by which temperate species could respond to both historical and seasonal climatic change is through long-distance migration , which is predicted to facilitate diversification [39] . However , our measurement of migration distance was not correlated with levels of phylogeographic structure . Low phylogeographic structure and high lineage loss , irrespective of dispersal ability or migratory behavior , are consistent with climatic instability , leaving a strong signature of lowered evolutionary persistence within temperate species . We identified species age as another important mediator of the latitudinal gradient in phylogeographic structure . Previous work on birds occurring in lowland Neotropical rainforests suggested that this age—diversity association is linked to species ecologies that influence evolutionary persistence in the landscape , with more sedentary and poorly-dispersing species being older and containing deeper phylogeographic structure across their ranges [8] . Reduced dispersal ability has been linked more directly to divergence in regional studies focused on lowland Neotropical [4] and South Pacific bird faunas [40] . Our study includes a more exhaustive survey of Neotropical birds in terms of range-wide sampling and number of species , and we did indeed identify many tropical species with low dispersal abilities and high phylogeographic structuring . However , there was substantial noise around this relationship , and dispersal ability did not emerge as an important predictor of phylogeographic structure , after accounting for other factors . In addition , our novel comparisons of species across biomes show uniformly lower phylogeographic structure in temperate species , irrespective of dispersal ability . The lack of an association between dispersal ability as estimated from hand-wing index and phylogeographic structuring may be an indication that species ecologies do not have the same influence on structuring populations in the temperate region as in the tropics , that temperate species have more generalized niches that predispose them to dispersing [41] , or that long-term climatic instability at high latitudes is the predominant factor that shapes levels of intraspecific genetic variation . The strong association between species age and phylogeographic structure in our dataset reflects a clear historical influence on intraspecific variation , but other aspects of species history ( e . g . , ancestral origin ) were not accounted for in our analysis . The historical biogeography of New World birds has focused on particular regions [e . g . , 42] and macroscales [e . g . , 43] , but the colonization times in each region are not known for many lineages . Resolving when species colonized regions is necessary to clarify if diversity is attributable to how long a lineage has been in a region [e . g . , 14] or if diversification is linked to expansion into novel environments [e . g . , 44 , 45] . The impacts of long-term interactions among species on phylogeography are not known , but the evolution of a species’ phylogeographic history may be influenced by codistributed species through ecological interactions . In the tropical mountains of the Andes , elevational replacement of closely related taxa has been linked to intraspecific competition during the history of populations [46] . If the evolutionary outcome of ecological processes varies between the temperate zone and the tropics , then species interactions could play a role in shaping the latitudinal phylogeographic diversity gradient . Although existing studies have found broadly concordant latitudinal patterns in diversity and evidence for greater evolutionary stability in the tropics , a general explanation for the link between diversification processes and latitudinal patterns has proven elusive . Studies have variously found that speciation or origination rates in lineages inhabiting the tropics are higher , lower , or similar to those of lineages in the temperate zone [17 , 33 , 47 , 48] . Differences obtained across these studies may be partially attributable to sampling different phylogenetic depths and scales [11] , which range from sister species [e . g . , 17] , to taxonomic clades [e . g . , 31] , and complete phylogenies [48] . Differences in the temporal scale of studies may highlight the different processes that are at work over different timescales ( e . g . , divergence versus persistence ) but may also reflect differences in our ability to model processes , such as extinction , in different types of data . At shallow timescales , we did not find evidence for accelerated intraspecific diversification in high latitudes , as observed in rates estimated from sister species divergence times [17] and subspecific diversity [16] . Reconciling these different stories will require not only improved data and models but also integrating the insights gained from different types of data collected over varying timescales . Comparative analyses , such as ours , are reliant on accurately delimited species . Avian taxonomy is complicated by several sources of bias , including greater efforts in describing diversity in the temperate zone , a lack of sampling to clearly delimit taxa in the tropics , and/or inconsistent criteria for species delimitation . A major perceived bias is that tropical species harbor multiple species , whereas temperate species have been finely split [27 , 49] . We evaluated this issue by accounting for species age in our analyses and by using an alternative lumped taxonomy . We found the latitudinal phylogeographic diversity gradient to be robust to taxonomic treatment and the age of recognized species . However , if all phylogenetic species were elevated to full species , as some have proposed [49] , species may no longer contain sufficient variation for comparative studies of intraspecific diversity . Under such a treatment , the patterns we observed in our data would reflect trends in recent speciation rather than intraspecific diversity , but they would still attest to the fact that broad patterns in diversity can form over very shallow timescales . We suggest that , when done carefully , comparative studies such as ours capture important biological patterns and processes , irrespective of taxonomic considerations . The accuracy of phylogeographic metrics at capturing genetic differentiation depends on the genetic markers employed and the density of sampling . We focused on a coarse-scale metric that reflected deep phylogeographic structure ( sensu [2] ) because the degree of population-level sampling varied across species , and our objective was to assess broad comparative patterns . Ecological effects on genetic differentiation may be more pronounced at the allelic level and may require large SNP-based datasets with dense sampling of individuals to detect associations . Alternatively , our proxies for species ecology may not accurately reflect the ecological processes that influence genetic variation . Another confounding factor in our estimates was the impact of coalescent stochasticity on mtDNA gene trees . However , it is unlikely that the significant relationships we recovered were artifacts of coalescent stochasticity because we sampled a large number of species , our estimates of phylogeographic structure were conservative , and we evaluated our results against null models . Our species ages are presumably overestimates because the method we employed did not take into account species tree—gene tree discordance and ancestral effective population sizes [50] . We do not expect this estimation error to be highly biased across species differing in phylogeographic structure . These limitations can be improved with genome-scale data and dense sampling , but obtaining large and comparable multilocus datasets for hundreds of species will likely not be possible for years to come . Furthermore , although genome-scale estimates of phylogeographic metrics will improve parameter estimation [51] , explaining the causes of among-species phylogeographic variation will remain a challenge . Our study focused on data from a relatively well-known and well-sampled group of organisms , New World birds . Despite that , we found assembling a dataset of this size and completeness to be challenging . Among the least straightforward aspects of assembling these data were filtering erroneous data ( e . g . , taxon misidentification ) and successfully extracting data and metadata from previous publications . As datasets grow , it will become particularly challenging to identify high-quality data . We opted for both manually checking data and trees with published results , along with automated processing that produced plots for additional data inspection . There may well be more signal in our data than we could extract in the broad analyses presented here , and it could provide further insight into the processes responsible for the observed latitudinal phylogeographic diversity gradient . For example , population expansions have been important in northern temperate species [52] , but we did not examine evidence of genetic bottlenecks , extirpation , and selective sweeps . Mechanisms directly responsible for the older age of tropical species will also require further investigation to determine if population persistence is due to more stable climates , tropical species colonizing areas earlier than the temperate zone , and/or additional taxonomic bias . More sophisticated models , the addition of spatially explicit statistics of genetic diversity , and larger multilocus datasets could provide higher resolution to these processes and perhaps reveal additional predictors of broad-scale patterns of diversity . In conclusion , phylogeographic data play a central role in elucidating the spatial and temporal dynamics of shallow evolutionary processes , and our study demonstrates that these processes are linked to broader biodiversity patterns . Species vary considerably in intraspecific diversity and the accumulation of this variability is time-dependent . Tropical species are older and harbor more phylogeographic structure , whereas temperate species are younger and have signatures of lineage loss , suggestive of pervasive impacts of environmental instability at high latitudes . Differential accumulation and persistence between tropical and temperate taxa may be correlated across phylogenetic scales , and this may produce gradients in both latitudinal phylogeographic structure and species diversity .
New World bird species are among the best-studied faunas at the phylogeographic level in sampling and diversity , and they occur across a broad environmental gradient that allows for comparative analysis at large spatial scales . We identified candidate phylogeographic datasets ( S2 Table ) and downloaded mtDNA sequence data from GenBank ( S1 Data ) . We chose mtDNA because nuclear DNA was not available for the majority of species in our study . To standardize our collection approach and optimize comparability among species , we selected species for which geographical sampling was available and omitted populations/species that occurred on Caribbean islands or in the Old World , to focus our analysis on mainland North and South America . To obtain units for comparative analyses , we delineated species using the taxonomy of the North American ( NACC ) and South American ( SACC ) Checklist committees of the AOS [53 , 54] . We used two taxonomic delimitation treatments to account for biases caused by lumping and splitting of species . Our first treatment consisted of single species recognized by the AOS ( n = 210; mean number individuals per species = 83 . 681; SD = 88 . 004 ) . Some species had paraphyletic mtDNA gene trees , which is most likely attributed to taxonomic error [26] . These currently recognized biological species may not represent natural groups , particularly in tropical species [27] , so we used a second taxonomic treatment in which closely related species were combined into species complexes and analyzed together . We combined all allopatric or parapatric populations and species that formed a monophyletic group for which we had range-wide genetic sampling . This alternative taxonomic treatment is referred to as the lumped dataset ( n = 179; mean number individuals per species = 100 . 201; SD = 95 . 553 ) . We estimated mtDNA gene trees for each species using BEAST v . 1 . 7 . 5 [55] . Because the AOS species were nested within the lumped species , we built gene trees using the lumped data and extracted relevant values for each taxonomic treatment . We used published substitution rates to calibrate the mtDNA gene trees because there were no appropriate fossils . For the cytochrome b ( cyt b ) and cytochrome oxidase subunits I and II ( COI and COII ) genes , we used 0 . 0105 substitutions/site/million yr ( s/s/my ) , a rate estimated for cyt b [56] . For the NADH dehydrogenase subunits II , III , and VI ( ND2 , ND3 , ND6 ) and for ATPase6&8 genes , we used 0 . 0125 s/s/my , as estimated for ND2 [42] . Comparative analysis of whole mitochondrial genomes show that COI and COII evolve at a similar rate as cyt b , and that ND3 and ND6 evolve at a similar rate as ND2 [57] . Estimated rates of evolution for the control region are highly variable , ranging from 2% to 20% sequence divergence per million yr [58 , 59] . Because of the uncertainty surrounding the substitution rate of the control region , we opted for a conservative rate ( 0 . 0125 s/s/my ) that was similar to that of the other , faster evolving mtDNA loci . For the uncorrelated lognormal relaxed clock mean ( ucld . mean ) parameter , we specified a lognormal distribution on the prior with the mean set to the above-mentioned mutation rates and a SD of 0 . 1 . This dating approach allowed us to account for rate heterogeneity among genes and branches and for uncertainty around mean estimates . We used a coalescent-constant-size tree prior and the best-fit nucleotide substitution model as determined in MEGA6 [60] , and we specified lognormal distributions on substitution model prior distributions . We ran each analysis for 50 million generations , sampling every 2 , 500 generations , performed multiple independent runs for validation , and assessed MCMC convergence and burn-in by examining ESS values and likelihood plots in Tracer v . 1 . 5 [61] . For some datasets that did not achieve Effective Sample Size ( ESS ) values > 200 after 50 million generations , we added up to 50 million additional generations to ensure that the results were stable . For each focal species , we included the sister taxon ( based on prior phylogenetic work ) and extracted the mean and the 95% highest posterior density for stem ( the time when the mtDNA haplotypes coalesce with the species’ last common ancestor ) and crown ( the time when all haplotypes in the species coalesce ) age estimates for each species in units of millions of years ago ( Mya ) . Some sister taxa did not share the same loci , and the resulting age estimates were incongruent , particularly for the stem age . To account for this discrepancy , when the data were available , we built multispecies alignments and ran BEAST analyses to obtain stem ages for multiple closely related taxa from a single tree . Mean stem ( t = −2 . 72; SD = 2 . 70; p = 0 . 007 ) and crown ages ( t = −2 . 51; SD = 1 . 66; p = 0 . 012 ) were older in the lumped dataset ( stem age: mean = 4 . 81 Mya; SD = 2 . 95; crown age: mean = 2 . 23 Mya; SD = 1 . 78 ) than the dataset using currently recognized species ( stem age: mean = 4 . 07 Mya; SD = 2 . 57; crown age: mean = 1 . 802 Mya; SD = 1 . 550 ) . Phylogeographic structure can be estimated via numerous means , including population genetic summary statistics [e . g . , 62] , assignment tests [e . g . , 63] and tree-based approaches [e . g . , 64] . For our study , we required a phylogeographic metric that could be estimated and compared across all species in our dataset that vary in terms of sampling . The multispecies coalescent provides an appropriate framework for delimiting genetic structure within species [65] . We used a Bayesian implementation of the General Mixed Yule Coalescent model ( bGMYC ) [24] . This model calculates the number of putative genetic species in a phylogeny by estimating the number of clusters in which the gene tree reflects intraspecific coalescent processes rather than interspecific processes . The bGMYC model provides a posterior probability that two tips in the phylogeny belong to the same genetic cluster , which can be used with a probability threshold to determine the number of clusters . We used the maximum clade credibility tree for each lumped species from BEAST for the bGMYC runs . We ran bGMYC for 250 , 000 generations using the single . phy function in R [66] and discarded the first 15 , 000 generations as burn-in . We ran each analysis multiple times for validation , and we assessed MCMC diagnostics by examining likelihood plots . We recorded the number of clusters per species using three different posterior probability thresholds ( 0 . 9 , 0 . 8 , and 0 . 7 ) to account for the uncertainty in delimited clusters . We recognize that finer-scale phylogeographic structure ( e . g . , significant FST values ) was present in some species and that the approach we implemented cannot accommodate this level of genetic variation . Given this limitation , there may be interactions between species traits and genetic structuring that we lack sufficient resolution to infer . We found the mean number of phylogeographic units in a species was , as expected , significantly ( t = −2 . 42; SD = 2 . 61; p = 0 . 016 ) higher in the lumped dataset ( mean = 3 . 35; SD = 2 . 92 ) than the dataset using currently recognized AOS species ( mean = 2 . 71; SD = 2 . 31 ) . We calculated a phylogeographic splitting rate under a pure-birth model , using the formula for stem age ( equation 6 ) from Magallón and Sanderson [67] and the code in the R package laser [68] . We elected to not use a more complex model that estimates speciation and extinction rates from branching times because of the overall low number of nodes in the gene trees and their shallow depths . For example , in the dataset using currently recognized species ( AOS species ) , the average number of phylogeographic clusters was less than three , with a mean age of less than 2 Mya . The Magallón and Sanderson [67] formula for diversification rates estimated from stem age assumes a starting diversity of one lineage , and the crown age formula assumes a starting diversity of two . There was no a priori reason to assume the starting number of phylogeographic units was two in a species , so we specified a starting diversity of one for the phylogeographic splitting rates estimated from both stem and crown ages . A comparison between the lumped dataset and the dataset using currently recognized AOS species showed that the splitting rates estimated from stem age were more similar ( t = −1 . 23; SD = 0 . 250; p = 0 . 22; lumped stem rate: mean = 0 . 253; SD = 0 . 230; AOS stem rate: mean = 0 . 221; SD = 0 . 267 ) than the rates estimated from crown age ( t = −1 . 82; SD = 0 . 641; p = 0 . 069; lumped crown rate: mean = 0 . 615; SD = 0 . 708; AOS crown rate: mean = 0 . 496; SD = 0 . 578 ) . In addition to calculating the degree of phylogeographic structure and splitting rate , we also calculated a stem branch index that served as a proxy for lineage loss in each species . We estimated this index by taking the difference between stem and crown age and standardizing the value by the stem age ( Lineage Loss = [Stem Age−Crown Age]/Stem Age ) . We used mean and the 95% high and low values from highest posterior density to independently calculate lineage loss . The pruning of lineages by extinction will increase the stem branch index , but a failure to diversify could also leave a similar signature . To distinguish between these two processes , we used a mean branch length index ( number of phylogeographic clusters/crown age ) within the crown clade of each species as a reference for how long it takes diversification to begin , assuming that speciation occurred at a constant rate . Crown group branch length indices that are longer than stem branches could indicate that there has not been enough time for diversification to occur . We found that 15 . 2% of the species had crown group branch length indices longer than the stem branch lengths , which suggests these species may not have had enough time to diversity . However , phylogenetic generalized least-squares analysis ( method details discussed below ) recovered no strong latitudinal trends in the difference between these branch lengths ( AOS species dataset: adjusted R2: −0 . 004; F-statistic = 0 . 269; p = 0 . 604; Lumped dataset: adjusted R2: −0 . 004; F-statistic = 0 . 242; p = 0 . 6234 ) . The lack of a strong correlation between crown group branch length indices and stem branch lengths suggests that there is not a latitudinal bias in tropical or temperate species having less time to diversify . We visualized phylogeographic data by projecting various metrics obtained from genetic data and our sampling strategy onto the 2-D plots of the New World . We downloaded digital range maps [69 , 70] and in ArcGIS 10 . 3 ( ESRI Inc . , Redlands , CA ) converted breeding ranges to rasters with a cell size of 0 . 1 and then reclassified each raster cell in which the species was present to its number of phylogeographic clusters , splitting rate based on crown and stem age , lineage loss , species age , and to one to represent where the species occurred for the species richness map . All functions used to make heat maps were done in ArcGIS , and the processes described below were automated by using Python scripts with ArcGIS functions . We used the Cell Statistics function in Spatial Analysis Tools to summarize across the species' range maps and to generate per-cell values for mean , SD , and/or sum for the above listed variables . For the Cell Statistics function settings , we set the geographical extent and mask to mainland North and South America . We also produced maps that visualize the proportion of total species sampled and the extent of sampling across each species range . We downloaded a global species-richness map of breeding birds constructed from the same digital range maps [69 , 70] used in our analyses , and estimated the proportion of species sampled per cell ( Fig 1A ) by dividing our species sampling map ( Fig 1B ) by a global species-richness layer [71] in Raster Calculator . To visualize sampling bias by producing sampling polygons ( Fig 1C ) , we built sampling polygons for each species by obtaining latitude/longitude coordinates , converting coordinates to a shapefile ( Split by Attribute add-in ) , converting the shapefile into a rectangle ( Minimum Bounding Geometry function ) , and then clipping the rectangle to fall within each species range map ( Clip function ) . We then produced polygons of unsampled areas by using the Erase function to identify the area in each species range not included in the sampling polygons . Sampling polygons were based on ten latitude-longitude coordinates that were compiled for each species from published records or georeferenced using descriptions of the sampling localities . For the lumped dataset , some lineages within the lumped species had fewer than ten samples . All points are plotted in S6 Fig Despite the large uncertainty surrounding some of these points , the coordinates , overall , provide coarse-scale resolution to how much of each species range was sampled . We summed the unsampled polygons ( Cell Statistics function ) and produced a heat map that shows areas that were undersampled ( Fig 1C ) . We performed additional diagnostics on the association between the proportion of each species ranged sampled and phylogeographic structure ( 0 . 9 posterior probability threshold ) , range size , absolute latitudinal midpoint , and elevational preference ( S7 Fig ) . Regression analyses indicated that there was only a significant association between the proportion of each species range sampled and phylogeographic structure ( Adjusted R2: 0 . 059 , p = 0 . 0002 ) , but based on the other three plots , this bias was not associated with range size , absolute latitudinal midpoint , or elevational preference ( S7 Fig ) . Finally , all maps were scaled to 110 , 000 km cells , to account for uncertainty in species range maps [72] . We measured the environmental space each species inhabits by extracting precipitation , climate , and net primary productivity data from observational records . We also extracted data from layers measuring the difference between present-day and Last Glacial Maximum climatic conditions . Our objective was to compare phylogeographic metrics in species that inhabit more seasonal environments in the temperate regions with those that occur in less seasonal tropical environments . To do this , we used climatic layers that averaged across the annual cycle . These climatic data were not used to characterize the niches of species because some of the taxa ( n = 52 ) in our dataset were migratory . Instead , the climatic data capture broad-scale habitat preferences ( e . g . , temperate broadleaf forests ) . We gathered 67 , 779 georeferenced observational records , representing all study taxa ( mean = 83 . 2 records/species’ SD = 41 . 48; min/max = 1:147; S4 Table ) . We obtained records from eBird ( May 2013 release ) , a real-time record of species distributions and abundances collected by amateur and professional ornithologists [73] . Prior to incorporation into the eBird database , all submitted observations are peer-reviewed by regional experts . Each record includes the start time , duration of data collection , and geographic distance covered . To minimize georeferencing inaccuracy while maximizing the number of localities per species , we included observations from all checklists that were less than 6 hr in duration and less than 5 km in distance traveled . For each species , we removed all duplicated localities and randomly selected 1000 records to which we applied a thinning algorithm such that no localities occurred within 1 km of each other , approximately the resolution of the climatic data grid cells . We further verified observational records against distributional maps [69 , 70] . For each locality record , we extracted elevation and 19 current climatic variables from the WorldClim database at a spatial resolution of 2 . 5 arc-seconds [74] . We also extracted net primary productivity for each record [75] . For each species , we estimated the range of climatic conditions inhabited by calculating the difference between the 95% high and low quantiles of each layer . The 95% range of climatic conditions acts as a proxy for breadth of habitat across a species range . We also incorporated climatic stability since the Last Glacial Maximum by measuring the per-cell difference between the 19 contemporary climatic layers and the corresponding paleoclimatic layers ( MIROC: Model for Interdisciplinary Research on Climate ) using the cell statistics function in Spatial Analysis Tools in ArcGIS . Using the eBird observational records , we extracted the cell values from each of these climatic stability layers . To reduce the dimensionality of the climatic niche estimates , we conducted a principal components analysis of the contemporary climatic variables , climatic stability variables , and elevation using the prccomp function in R [66] . We used the Kaiser Criterion ( Eigenvalues greater than one ) to reduce the number of components , and we retained principal components one through four . We calculated mean and standard error for the principal components for each species . For downstream analyses , we used either the first principal component ( PC1 ) , which explained 49% of the climatic variation across species ( S5 Table ) , or a combination of annual mean temperature ( BIO1 ) , seasonality temperature ( BIO4 ) , annual mean precipitation ( BIO12 ) , and precipitation seasonality ( BIO15 ) . For each currently recognized species , we determined the range size , maximum and minimum latitude , latitudinal range , landscape ruggedness , and midpoint of occurrence using digital range maps [69 , 70] . We projected the range map for each species with a lambert azimuthal equal area projection . All of the spatial variables we collected were from the resident distribution of each species or , in the case of migratory species , the breeding distribution . We estimated the area of each range and sampling polygon in km2 and calculated the proportion of range sampled by dividing the sampling polygon area by the range size . For migratory species , we calculated migratory distance as the difference between the breeding and wintering latitudinal midpoints of each species . For sedentary species , we specified migratory distance as zero . For species in the lumped dataset that consisted of more than one currently recognized species , we merged the range maps of these species and calculated the same metrics as above . We performed projections and calculations using the R packages maptools [76] , raster [77] , and rgdal [78] . To measure the topographic variability across species ranges , we used a modified Melton index [79]— ( Elevationmax−Elevationmin ) /log ( range size ) —that included a log-converted range size instead of a square root conversion , in order to account for the large variance in range sizes across species . We generated 250 random points per polygon in each species distribution , and we extracted the elevation at each point to estimate maximum and minimum elevation . We recorded wing length ( WL ) , secondary length ( SL ) , and tarsus length from vouchered specimens deposited at the American Museum of Natural History and the Museum of Natural Science at Louisiana State University ( S6 Table ) . We measured five male specimens in adult plumage per species , and for migratory species , we only included individuals collected during breeding months . We selected males because females may show greater variation in mass during the breeding season than males [80] . The shape of a bird’s wing influences its flight capabilities and serves as a proxy for dispersal ability . Birds with long , narrow wings are more capable of long-distance flight than species with rounder , short wings [81] . We calculated a proxy for dispersal ability using the wing measurements ( hand-wing index = 100 x ( WL − SL ) /SL ) , a metric that is positively correlated with dispersal ability [82] . We used tarsus length as a proxy for body size [83] , as these are positively correlated for most species in our dataset with the exception of parrots ( Order: Psittaciformes ) , which have relatively small tarsi given their body size . The biogeographical distributions of birds , including the species in our dataset , are nonrandom , with entire clades distributed only in the temperate or the tropical region . To account for this potential phylogenetic effect on patterns of latitudinal variation , as well as uneven geographical sampling among different groups of birds and the nonindependence of species trait data , we used PGLS [83] analysis . We tested whether variables were significantly correlated with different metrics of phylogeographic variation ( phylogeographic structure , splitting rates , and lineage loss ) by fitting data to a condensed set of multivariate models . The purpose of the multivariate modeling was to determine how much of the variation in the phylogeographic metrics could be explained by the predictor variables and to determine the relative importance of each of the variables . We independently examined four classes of response variables reflecting the phylogeographic history of species from the AOS ( n = 210 ) and lumped ( n = 179 ) datasets: ( 1 ) the degree of phylogeographic structuring within species , as determined by the number of bGMYC species clusters; ( 2 ) species age , as determined by crown and stem ages; ( 3 ) the rate at which diversification occurs within species or splitting rate , as determined by the phylogeographic diversification rate estimated from stem and crown species ages; and ( 4 ) lineage loss , as determined by the standardized length of time between the stem and crown nodes . To account for uncertainty in parameter estimates , we independently modeled metrics ( phylogeographic structure and splitting rates ) using three clustering threshold values ( 0 . 9 , 0 . 8 , 0 . 7 ) . We treated predictor variables as fixed effects in the models . To reduce the residual variance in the models we square root converted the following variables in the PGLS analysis—number of phylogeographic units , species age , range size , migratory distance , hand-wing index , and sample size . To account for the phylogenetic nonindependence of the species trait data , we used the Jetz et al . [48] tree , built using the Hackett et al . [85] phylogeny as a backbone . We downloaded 1 , 000 trees from birdtree . org ( Hackett All Species option; January 2017 ) , and we built a maximum clade credibility ( MCC ) tree using the pseudoposterior distribution of trees in Tree Annotator [55] . Four species in our dataset represent recent taxonomic changes and were not included in the Jetz et al . [48] tree . For these taxa , we grafted species onto the phylogeny to their sister taxon using the add . tip function in the ape package [86] in R [66] . All models are available in S3 Table . We used 100 subsampled trees for the multivariate models and the MCC tree for univariate tests . Using the PGLS function in the R [66] caper package [84] , we fit data to multivariate models . This function models phylogenetic signal in the data using the parameters lambda , kappa , and delta . We optimized the value for lambda using maximum likelihood , and we kept the default values for kappa ( 1 . 0 ) and delta ( 1 . 0 ) . We assessed whether the empirical response variables were significantly different from a random sample of values generated using the same mean , SD , and distribution type of the empirical data . For all response variables , we used truncated lognormal distributions , except for lineage loss , for which we used a truncated normal distribution instead . Both functions are part of the EnvStats R package [87] . From these distributions , we produced random values of response variables ( phylogeographic structure , species age , splitting rates , and lineage loss ) for each species 100 times . We then ran univariate PGLS models , recorded the adjusted R2 values for each model , and then compared the proportion of simulated R2 values above the empirical R2 value . If <5% of the simulated values were greater than the empirical R2 value , we concluded that the empirical species trait values were not generated by this random process . We plotted null models for phylogeographic structure ( S2 Fig ) , splitting rate using stem age ( S3 Fig ) , splitting rate using crown age ( S4 Fig ) , and the lineage loss index ( S5 Fig ) . For multivariate models , we estimated AICc scores with a correction for sample size for a full model with all variables and the AICc score for each model without each of the predictor variables . We assessed the relative importance of each variable by calculating ΔAICc = AICca − AICcf , where ΔAICc is the change in AICc between the model without a particular predictor variable ( AICca ) and the full model ( AICcf ) . Models with a ΔAICc > 2 are deemed to be significantly less likely than the full model , and the removed variable is considered important . We report model output from the median AICc score of the 100 full models , based on each of the 100 different trees . We then used the tree that produced the median AICc score for the full model to report the output for the alternative models . Because the environmental variables in our dataset are correlated , we ran different sets of multivariate models with uncorrelated variables . Each set of models assessing variable importance included only one of following: ( 1 ) latitudinal midpoint; ( 2 ) net primary productivity; ( 3 ) mean temperature and precipitation , and temperature and precipitation seasonality; ( 4 ) bioclimatic PC1; ( 5 ) climatic instability since the Last Glacial Maximum; and ( 6 ) range in mean temperature and precipitation , temperature and precipitation seasonality , and elevational preference . We report AICc weights in order to show the relative stability of similar models using different treatments . We did not perform model averaging because the large numbers of variables and models would lead to data dredging , and interpreting the biological significance of these models would be difficult . Our predictions for the influence of variables on phylogeographic metrics are shown in S1 Table . To briefly summarize , we expect that phylogeographic structure will be higher in older species that persist in the landscape [e . g . , 8] , inhabit areas that were more climatically stable between glacial—interglacial periods [e . g . , 88] , are distributed in areas with more energy [e . g . , 28] , inhabit more topographically complex areas [e . g . , 89] or broader environmental conditions , have lower dispersal abilities [e . g . , 4 , 41] , and have larger geographical ranges . We expect similar associations with splitting rates and species ages . The climatic stability of an area has been suggested to both increase and decrease diversification rates [16 , 17 , 28] . Splitting rates may asymptotically increase with range size as species fill up geographical space . Alternatively , species ranges may be dynamic and decoupled from their rate of diversification . The geographic distance species migrate may facilitate diversification by allowing species to rapidly colonize new environments [39] , or migratory behavior may alternatively inhibit diversification by limiting isolation among populations via gene flow [e . g . , 90] . We predict that lineage loss will be higher in temperate latitudes , species with smaller habitat breadth , areas with greater historical climatic instability , and species with higher dispersal abilities , longer migratory distances , and smaller ranges . We also mined foraging guilds ( Fig 1D ) and body sizes ( Fig 1E ) for sampled and unsampled New World bird species from the EltonTraits database [25] . Because our dataset , while large for a comparative phylogeographic study , includes only up to ~10% ( Fig 1A; South American tropics ) to ~30% ( Fig 1A; temperate North America ) of the total diversity , we compared the two above mentioned traits to unsampled species .
|
The causes of high tropical species diversity remain contentious and disputed . Recent studies have shown that latitudinal differences in speciation and extinction rates give rise to high tropical diversity . However , it is unclear if this gradient is the product of population-level , species-level , or clade-level processes . Here , we used genetic , environmental , and morphological data from hundreds of bird species occurring in the Western Hemisphere to evaluate patterns of within-species diversity . We found that tropical species have greater intraspecific genetic variation , and this diversity persists longer than in temperate species . These results indicate that biodiversity gradients can arise more rapidly than previously thought and that the processes governing these gradients operate on multiple evolutionary timescales .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"biogeography",
"taxonomy",
"ecology",
"and",
"environmental",
"sciences",
"mitochondrial",
"dna",
"atmospheric",
"science",
"population",
"genetics",
"species",
"delimitation",
"data",
"management",
"speciation",
"paleontology",
"forms",
"of",
"dna",
"dna",
"population",
"biology",
"ecological",
"metrics",
"geography",
"computer",
"and",
"information",
"sciences",
"paleoclimatology",
"cartography",
"phylogeography",
"species",
"diversity",
"biochemistry",
"climatology",
"ecology",
"nucleic",
"acids",
"earth",
"sciences",
"genetics",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"biology",
"evolutionary",
"processes",
"latitude"
] |
2017
|
A latitudinal phylogeographic diversity gradient in birds
|
High-order epistasis—where the effect of a mutation is determined by interactions with two or more other mutations—makes small , but detectable , contributions to genotype-fitness maps . While epistasis between pairs of mutations is known to be an important determinant of evolutionary trajectories , the evolutionary consequences of high-order epistasis remain poorly understood . To determine the effect of high-order epistasis on evolutionary trajectories , we computationally removed high-order epistasis from experimental genotype-fitness maps containing all binary combinations of five mutations . We then compared trajectories through maps both with and without high-order epistasis . We found that high-order epistasis strongly shapes the accessibility and probability of evolutionary trajectories . A closer analysis revealed that the magnitude of epistasis , not its order , predicts is effects on evolutionary trajectories . We further find that high-order epistasis makes it impossible to predict evolutionary trajectories from the individual and paired effects of mutations . We therefore conclude that high-order epistasis profoundly shapes evolutionary trajectories through genotype-fitness maps .
Epistasis creates historical contingency , as it means that the effect of a mutation depends on previous substitutions [1–6] . Interactions between pairs of mutations can cause mutations to accumulate in a specific order [1 , 4] , stochastically open and close pathways [3 , 6] , and make evolution irreversible [7 , 8] . The effects of high-order epistasis—interactions between three or more mutations—on evolution are less well understood than the effects of pairwise epistasis . Statistically-significant high-order epistasis has been observed in multiple genotype-phenotype maps [6 , 9–16] , even when steps are taken to minimize its contribution to epistasis models [15] . Its magnitude is generally lower than the individual and pairwise epistatic effects of mutations [15] . Several studies have suggested that it can alter evolutionary outcomes [6 , 10 , 16] , but its overall importance for evolution is not well understood . Does high-order epistasis alter evolutionary outcomes ? Or are trajectories primarily shaped by the additive and pairwise epistatic effects of mutations ? We set out to assess the effect of high-order epistasis on evolutionary trajectories through experimentally measured genotype-fitness maps . We decomposed these maps into contributions from nonlinear scale , additive effects , and epistasis at different orders ranging from second to fifth . We then calculated “truncated” maps with different orders of epistasis deleted . By comparing the fitness values and probabilities of individual evolutionary trajectories through the truncated maps , we can reveal the extent to which high-order epistasis determines evolutionary outcomes .
Our first goal was to determine the contributions of each order of epistasis to fitness in six experimentally measured genotype-fitness maps ( Table 1 ) . Each map consisted of all possible combinations of 5 mutations ( 25 = 32 genotypes ) in a haploid genome . The mutations in datasets I and IV arose during adaptive , experimental evolution of E . coli , and occur throughout the genome [19 , 28] . The mutations in datasets II and VI each occur in single genes that confer drug resistance in E . coli and HIV , respectively [1 , 29] . The mutations in datasets III and V were introduced randomly into the A . niger genome [30] . Previous workers characterized components of fitness for each genotype under defined experimental conditions . For two of the datasets ( I , and IV ) , the authors measured relative fitness using competition assays . In two of the datasets ( III and V ) , the authors measured growth rate of each strain . In dataset II , the authors measured minimum inhibitory concentration in the presence of an antibiotic , and from this estimated relative fitness [1] . In dataset VI , the authors measured HIV infectivity in an ex vivo assay , then treated this activity as a proxy for fitness [29] . All genotypes and phenotypes for each dataset are shown in S1 JSON . We previously analyzed four of these datasets , finding small magnitude , but statistically-significant , high-order epistasis in each map [15] . We used this same approach to characterize epistasis in the remaining two maps ( S1 Fig , Materials & methods ) . We sought to account for confounding effects that could lead to spurious epistasis , which would , in turn , lead to spurious effects on evolutionary trajectories . The most important confounding effect is the scale of the map . Models of high-order epistasis sum the effects of mutations and then account for deviation from this expectation by epistasis [12 , 17] . But there is no a priori reason to assume mutational effects should add: they may multiply or combine on some other nonlinear scale [5 , 12 , 15 , 31] . To account for this , we empirically determined a nonlinear scale for each map using a power-transform , and then used this to linearize each map [15] . We then decomposed the linearized maps into epistatic coefficients using Walsh polynomials [10 , 12 , 17] . This approach uses the geometric center of the genotype-fitness map as reference state and reveals global correlations in the effects of mutations across the map . Each order of epistasis accounts for variation that is not explained by the sum of all lower-order contributions . For example , third-order coefficients account for any “leftover” variation in the fitness of triple mutants after the first-order ( additive ) and second-order ( pairwise ) effects of those mutations are taken into account . We determined the contribution of each order of epistasis to the total variation in fitness for each dataset by sequentially setting fifth- , fourth- , third- , and second-order epistatic coefficients to zero . We recalculated the fitness of each genotype using each “truncated” model . This is directly analogous to decomposing a sound wave into a sum of frequencies using a Fourier transform [12] . After decomposition , the original sound wave can be approximated by a sum of principal frequencies , followed by a reverse Fourier transform . By selectively including frequencies , one can identify those that contribute most to the final sound wave . Our analysis follows the same logic , approximating fitness ( the sound wave ) using a collection of epistatic coefficients ( sound frequencies ) . We quantified the contribution of each epistatic order by measuring the change in fitness when the ith order of epistasis was included in the model . As a metric , we used ϕ = ρ i 2 − ρ i − 1 2 , where ρ x 2 is the squared Pearson’s coefficient between the measured fitness of each genotype and its fitness calculated for a model truncated to the xth order . ϕ ranges from -1 to 1 . Fig 1A shows this calculation for dataset I . As epistatic orders are added , ρ2 between the truncated model and measured fitness values improves . This allows determination of ϕ for each order: first-order coefficients ( additive effects ) account for 94 . 0% of variation in fitness; second-order ( pairwise epistasis ) for 3 . 8%; third for 1 . 2% , fourth for 0 . 9% , and fifth for 0 . 1% . We then applied this analysis to all six datasets . Fig 1B summarizes these results . The total contribution of epistasis to variation in fitness ranged from 6 . 0% ( dataset I ) to 32 . 2% ( dataset VI ) . Other datasets exhibited intermediate levels of epistasis , comparable in magnitude to high-order epistasis observed in similar datasets [10 , 15 , 16] . In all datasets , the first-order ( additive ) effects of mutations made the largest contribution to variation in fitness . Outside of this , there was no simple pattern in the relative contributions of the different orders . In dataset I , II and IV , the contribution of epistasis to variation decayed with increasing order . In dataset V , epistasis does not decay . In dataset VI , the addition third-order epistasis ( without fourth-order epistasis ) actually does a worse job of predicting fitness than second-order alone . The quantitative and qualitative differences in the contribution of epistasis across datasets allow us to study how altering epistasis alters evolutionary trajectories . Our next question was how each order of epistasis altered evolutionary trajectories . We first back-transformed our truncated , linearized maps onto the original scale . This creates a genotype-fitness map without specific epistatic interactions , but on the original , possibly nonlinear , scale of the map . We calculated the relative probabilities of all L ! forward trajectories through these maps , starting from the ancestral state and ending at the derived state [1 , 19 , 30] . Because the maps describe fitnesses of asexual organisms with large population sizes , we modeled trajectories as a series of sequential fixation events captured by a Gillespie model for haploid organisms with large population size ( Materials & methods ) [1 , 18 , 19] . In this scheme , the probability of a trajectory is the product of the probabilities of its individual fixation events , normalized across all trajectories . We visualized these trajectories by overlaying them on the genotype-fitness map weighted by their relative probabilities . Higher probability mutations have thicker lines connecting them . Fig 2 shows this analysis for dataset I . We started with a purely additive map ( top left ) . All trajectories are accessible with similar probabilities because , in this map , all mutations are individually favorable . We then added successive orders of epistasis and recalculated trajectories through each new map . The addition of second-order epistasis altered the availabilities of trajectories . The changes are most readily evident in the lower row in Fig 2 , which shows the change in the probability of each edge and node in the map . The left side of the map is red ( indicating loss of probability ) , while the right side of the map is blue ( indicating gain of probability ) . Addition of each new order , moving left to right across Fig 2 , alters the probability of trajectories through the map . To quantify differences in the sets of trajectories with increasing epistasis , we calculated the change in the probabilities of all 120 forward trajectories through maps with different amounts of epistasis included ( θ ) . A θ of 0 . 0 indicates that the set of trajectories through the spaces are identical , while a θ of 1 . 0 means the sets of trajectories do not overlap at all ( Materials & methods ) . Intermediate values indicate that some fraction of the trajectory probability density is shared between the maps . In dataset I , trajectories through the additive and second-order epistatic maps have θ = 0 . 390 . Put another way , the addition of pairwise epistasis to the additive map shifts 39 . 0% of the trajectory probability density . Addition of each new order of epistasis has a smaller effect on trajectory probability: θ2→3 = 0 . 340 , θ3→4 = 0 . 292 , and θ4→5 = 0 . 122 . To determine confidence intervals on our estimates of ϕ and θ , we sampled from the fitness measurement uncertainty for each genotype , generating a collection of pseudoreplicate genotype-fitness maps ( S2A Fig ) . We then decomposed each pseudoreplicate map into epistatic coefficients ( including refitting the scale ) and remeasured ϕ and θ for each epistatic order . Fig 3A shows this calculation for dataset I . From these distributions , we can determine 95% confidence intervals for ϕ and θ ( shown as gray , bracketed values in Fig 2 ) . We next asked whether the observed epistasis and its effect on trajectories could be the result of uncertainty in the fitness values . An epistasis model accounts for random noise as leftover variation , and thus as apparent epistasis [15] . We posed the following question: if the epistasis at a given order resulted only from noise , what effect would it have on ϕ and θ ? To ask this question , we constructed “null” maps with truncated epistasis , but noisy fitness values ( S2B Fig ) . We took our truncated maps at each order and then assigned each fitness the same variance that was measured for the original , un-truncated fitness values . We sampled from this uncertainty to generate pseudoreplicates , extracted apparent epistasis—in this case , arising from noise—and then calculated ϕ and θ for the pseudoreplicate . This allows us to construct distributions of ϕ and θ for epistasis arising purely from experimental noise . We show this calculation for dataset I in Fig 3B . Unlike the experimental distributions , which spread out in ϕ , the distributions arising from random noise cluster at low values of ϕ . The ϕ/θ distributions of second- , third- , and fourth-order epistasis minimally overlap in Fig 3A versus 3B . This indicates that the signal for epistasis in the datasets is greater than expected from noise in the measured fitness values . In contrast , the ϕ/θ distribution for fifth-order epistasis overlaps between Fig 3A and 3B: the effect of fifth-order epistasis cannot be distinguished from noise . Because we are interested in the effect of epistasis on trajectories ( θ ) , we determined a p-value for each θ . We took the mode of θ at each order from Fig 3A , and determined its percentile on the corresponding null distribution in Fig 3B . For second- , third- , and fourth-order epistasis , this yields a p-value < 0 . 05 . In contrast , the p-value for fifth order was 0 . 12 . With these quantification tools in hand , we next studied the relationship between epistasis and evolutionary trajectories for the increasing levels of epistasis exhibited by the remaining five datasets . S3–S8 Figs summarize our analyses for all six datasets . It is helpful to compare dataset I ( Fig 2 ) and dataset V ( Fig 4 ) . While epistasis accounts for 6 . 0% of variation in fitness for dataset I , it accounts for 32% of the variation in fitness for dataset V . The large amount of epistasis in dataset V means that epistasis at all orders has a massive effect on evolutionary trajectories through this space . The addition of fourth-order epistasis is particularly striking . With only third-order epistasis and down , there are multiple paths through the space . With the addition of fourth-order epistasis , all paths but two become inaccessible . The addition of fifth-order epistasis opens the space up again , but to a different set of trajectories than what existed in the third-order space . We next asked whether magnitude of epistasis or the order of epistasis was a stronger predictor of its effect on evolutionary trajectories . We plotted ϕ versus θ for each order for each dataset on a single plot ( Fig 5A ) . This reveals a correlation between the magnitude of the epistasis and its effect on trajectories . In contrast , we see no correlation between the order of epistasis and its effect on evolutionary trajectories ( Fig 5B ) . When epistasis contributes more than ≈5% of the variation in fitness , regardless of order , the divergence in trajectory probabilities with and without the epistasis is 40% or greater . The magnitude of epistasis—not its order—predicts its effect . Our next question was more practical: how important is epistasis for predicting evolutionary trajectories in these datasets ? We imagined an experiment in which we measured the effects of all mutations in the ancestral genotype . We then asked if we could take these individually measured mutational effects and predict evolutionary trajectories . To ask this question , we re-analyzed the epistasis present in all six datasets , this time using the ancestral genotype as the reference state . In this formulation , the first-order coefficients are the effect of each mutation by itself in the ancestral background , the second-order coefficients are the difference in the effects of mutations introduced in pairs versus separately , and third-order coefficients are the difference in the fitness of genotypes combining three mutations versus two mutations that cannot be explained by the first- and second-order coefficients . ( This has been called the “biochemical” or “local” model of high-order epistasis [12] . ) We describe this further in the Materials & method section . To characterize the effect of epistasis on our ability to predict evolutionary trajectories of increasing length , we calculated the probability of all possible forward trajectories of a defined number of steps starting from the ancestral genotype , and then repeated this probability calculation using maps truncated to various orders of epistasis . The difference in the actual and truncated map trajectory probability distributions measures our predictive power for evolutionary trajectories . We show these results in Fig 6 for all six datasets . In each panel , we plot inclusion of increasing orders of epistasis left-to-right ( starting from additive and going to fifth-order ) and increasing trajectory length bottom-to-top ( starting from one-step and going to five-step ) . The overlap between the trajectory distribution for the truncated and real map for each epistasis/trajectory-length is shown as a color ranging from white ( perfect prediction ) to red ( poor prediction ) . We found that all orders of epistasis were important for predicting evolutionary trajectories . Dataset IV ( panel D ) illustrates behavior seen across all datasets , so we will use it as a specific example . In this dataset , additive coefficients are inadequate to capture even two-step trajectories: the trajectory probability distribution for two-step mutations only overlaps by 53 . 0% for the truncated and real maps . The prediction gets worse for longer trajectories , dropping to 39 . 4% for three steps , 30 . 5% for four steps , and 0 . 0% for five steps . The overlap for the final step is 0 . 0% because the additive model does not predict that the five-mutation genotype will be more fit than the four-mutation genotype . Trajectories in the additive map therefore do not proceed to this final genotype . Adding pairwise epistasis to the model allows perfect “prediction” of the two-step trajectories , as we have perfect knowledge of the fitness values of all possible single and double mutants . But the three-step and four-step trajectories are predicted worse with pairwise epistasis included than with the additive map . The three-step overlap is 25 . 9% , while the four-step and five-step trajectory overlap is 0 . 0 . The four-mutation and five-mutation genotypes are predicted to have low fitness . Adding third-order epistasis—now imagining that we characterized all possible single , double , and triple mutants in the ancestral genotype—allows us to “predict” trajectories up to three steps long; however , it fails for four- and five-step trajectories . The overlap is 25 . 1% and 45 . 2% respectively . Even the addition of fourth-order epistasis is insufficient to capture the five-step trajectories: the overlap for five-step trajectories is 0 . 0% . Dataset IV is a particularly clean example , but all six datasets exhibit similar behavior ( Fig 6 ) . Neglecting epistasis leads to poor predictions of trajectories starting from the ancestral genotype . The lower-order the truncation , the worse the prediction as more mutations accumulate . Third- and fourth-order epistasis had an appreciable effect on all datasets . Fifth-order epistasis had an effect in four of the six datasets . Like the analysis using the global model above , high-order epistasis relative to the ancestral genotype potently alters evolutionary trajectories .
Our analysis reveals that high-order epistasis can strongly shape evolutionary trajectories . Removal of three- , four- , and five-way interactions between mutations significantly alters the probabilities of trajectories through genotype-fitness maps ( Figs 2 and 4 ) . This result is robust to uncertainty in the measured fitness values ( Fig 3 ) and appears to be a general pattern in many maps ( Fig 5 ) . Finally , neglecting high-order epistasis leads to poor predictions of evolutionary trajectories through these maps ( Fig 6 ) . In the majority of datasets , low-order models provide useful estimates of fitness . For datasets I-IV , ignoring three-way and higher-interactions yields fitness values within 15% of the actual map ( Fig 1B ) . Dataset I would be particularly close , yielding fitness values within 2 . 5% of the actual map . This is consistent with other analyses of high-order epistasis in other datasets , which suggest that additive and pairwise epistatic effects can often provide sufficient information to predict multi-mutation fitness values to within 5-10% [6 , 9–15] . While low-order models can often describe fitness with some degree of precision , low-order models are inadequate to describe evolutionary trajectories in any of the datasets . Even in dataset I , third- and fourth-order interactions potently shape evolutionary trajectories . The probability distributions of trajectories with and without fourth-order epistasis differ by 29 . 2% . And , as the magnitude of epistasis increases , its effect on trajectories grows ( Fig 5A ) . In some instances , addition of high-order interactions completely shifts the set of trajectories available ( Fig 4 ) . The effect of high-order epistasis on evolutionary trajectories is profound . We can build this intuition by imagining predicting evolutionary trajectories . If we start with knowledge of the individual effects of mutations in the ancestral background we can predict the first move perfectly , but not the second move . Pairwise epistasis means the effect of the second mutation is modulated by the presence of the first . We might try to overcome this difficulty by measuring the effect of each mutation and each pair of mutations , thereby accounting for pairwise epistasis . But our results reveal this is still insufficient to predict trajectories past the second step . There are three-way interactions that alter the effect of the third mutation , even after accounting for the first- and second-order effects of mutations . This continues all the way to fifth-order in these five-site datasets . This has two implications . First , this adds to the growing recognition of extensive contingency in evolution [3 , 4 , 8 , 32] . The effect of an event today is contingent on a whole collection of previous events . Remarkably , we found that this contingency is mediated by epistasis at all orders , including up to five-way interactions between mutations . Second , this work implies that measuring the individual effects of many mutations in a single genetic background , despite revealing a local fitness landscape [6 , 33 , 34] , will be of limited utility for understanding evolution past the first few moves . We expect the effect of high-order epistasis on trajectories will be amplified in larger maps that have more mutations . In a larger map , more mutations compete for fixation—each modulated by high-order interactions with previous substitutions—leading to even greater contingency on specific substitutions that occurred in the past . Further , the small maps we studied artificially limit the effects of high-order epistasis , as larger maps could , potentially , have even higher-order interactions . But even if no epistasis above fifth-order is present , trajectories will have more steps in a larger map; therefore , a fifth-order interaction could alter the relative probabilities of many more future moves in a larger space . One open question is the effect of recombination on this radical contingency . We studied trajectories in which mutations fixed sequentially . This means our results are directly applicable to asexual organisms and loci in tight linkage , such as mutations to individual genes . Once recombination comes into play , other dynamics become possible . While recombination can completely overcome pairwise epistasis [35] , it is unclear whether this result will apply to higher-order interactions . High-order epistasis appears to be a ubiquitous feature of experimental genotype-fitness ( and genotype-phenotype ) maps [6 , 9–15] . The origins of this epistasis remain unknown . Further , epistasis may go to much higher-order than yet observed , leading to extremely long-term memory in evolution . The observation of cryptic epistasis between genetic backgrounds that appear similar , but in which mutations have radically different effects , may point to high-order epistasis between mutations in diverging backgrounds [33 , 36] . Whatever the origins or order may be , our work reveals that combinations of early substitutions continue to have an effect as future mutations accumulate: the past continues to press upon the present .
We used the following protocol to remove specific orders of epistasis from genotype-fitness maps . The steps correspond directly to the pipeline shown in S1 Fig , which is described in detail in Sailer et al . [15] . We quantified the contribution of epistasis to each map ( ϕ ) by determining the difference in the variation explained by the ith and ( i − 1 ) th orders . ϕ = ρ i 2 − ρ i − 1 2 , where ρ x 2 is the squared Pearson coefficient between linear fitness values in a model truncated to order x ( F→ l i n e a r , t r u n c − t o − x ) and linear fitness values determined from the original map ( F→ l i n e a r ) . We calculated the probability of a given evolutionary trajectory as series of independent , sequential fixation events . We assumed that the time to fixation for each mutation was much less than the time between mutations ( the so-called strong selection/weak mutation regime ) [1 , 18 , 19] . The relative probability of an evolutionary trajectory i is the product of its required fixation events relative to all possible trajectories: p i = ∏ x ∈ S i π x → x + 1 ∑ j ∈ T ∏ x ∈ S j π x → x + 1 , ( 6 ) where πx→x+1 is the fixation probability for genotype x + 1 in the x background , Si is the set of steps that compose trajectory i , and T is the set of all forward trajectories . The model assumes the mutation rate is the same for all sites , and that population size and mutation rates are fixed over the evolutionary trajectory [1 , 19–22] . We calculated πx→x+1 for each step using the Gillespie model [23] π x → x + 1 = 1 - e - s x → x + 1 1 - e - N s x → x + 1 = 1 - e - ( 1 - w x + 1 / w x ) 1 - e - N ( 1 - w x + 1 / w x ) , ( 7 ) where N is population size , s is the selection coefficient and wx and wx+1 are the relative fitnesses of the x and x + 1 genotypes visited over the trajectory . To determine the difference between sets of trajectories in maps with and without high-order epistasis , we measured the magnitude of the difference in probability for all L ! forward trajectories through each space . We did so by: θ = ∑ i = 1 i = L ! p i e x p e r i m e n t a l - p i t r u n c 2 , ( 8 ) where p i e x p e r i m e n t a l is the probability of the ith trajectory within the experimental map and p i t r u n c is the probability of that same trajectory in a truncated map , with high-order epistasis removed . We implemented the epistasis and trajectory models using Python 3 extended with the numpy and scipy packages [24] . We used the python package scikit-learn to perform linear regression with truncated forms of these models [25] . Plots were generated using matplotlib and jupyter notebooks [26 , 27] . Our full software package is available in the epistasis package via github ( https://harmslab . github . com/epistasis ) .
|
A key goal for evolutionary biologists is understanding why one evolutionary trajectory is taken rather than others . This requires understanding how individual mutations , as well as interactions between them , determine the accessibility of evolutionary pathways . We used a robust statistical analysis to reveal interactions between up to five mutations in published datasets , meaning that the effect of a mutation can depend on the presence or absence of four other mutations . Simulations reveal that these interactions strongly shape evolutionary trajectories . These interactions lead to profound unpredictability in evolution , as one cannot use the effect of a mutation in the ancestor to predict its effect later in the trajectory .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"epistasis",
"physics",
"sound",
"waves",
"mutation",
"substitution",
"mutation",
"point",
"mutation",
"fitness",
"epistasis",
"acoustics",
"mathematics",
"statistics",
"(mathematics)",
"confidence",
"intervals",
"heredity",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"evolutionary",
"biology",
"evolutionary",
"genetics"
] |
2017
|
High-order epistasis shapes evolutionary trajectories
|
X-linked myotubular myopathy ( XLMTM ) is a congenital disorder caused by mutations of the myotubularin gene , MTM1 . Myotubularin belongs to a large family of conserved lipid phosphatases that include both catalytically active and inactive myotubularin-related proteins ( i . e . , “MTMRs” ) . Biochemically , catalytically inactive MTMRs have been shown to form heteroligomers with active members within the myotubularin family through protein-protein interactions . However , the pathophysiological significance of catalytically inactive MTMRs remains unknown in muscle . By in vitro as well as in vivo studies , we have identified that catalytically inactive myotubularin-related protein 12 ( MTMR12 ) binds to myotubularin in skeletal muscle . Knockdown of the mtmr12 gene in zebrafish resulted in skeletal muscle defects and impaired motor function . Analysis of mtmr12 morphant fish showed pathological changes with central nucleation , disorganized Triads , myofiber hypotrophy and whorled membrane structures similar to those seen in X-linked myotubular myopathy . Biochemical studies showed that deficiency of MTMR12 results in reduced levels of myotubularin protein in zebrafish and mammalian C2C12 cells . Loss of myotubularin also resulted in reduction of MTMR12 protein in C2C12 cells , mice and humans . Moreover , XLMTM mutations within the myotubularin interaction domain disrupted binding to MTMR12 in cell culture . Analysis of human XLMTM patient myotubes showed that mutations that disrupt the interaction between myotubularin and MTMR12 proteins result in reduction of both myotubularin and MTMR12 . These studies strongly support the concept that interactions between myotubularin and MTMR12 are required for the stability of their functional protein complex in normal skeletal muscles . This work highlights an important physiological function of catalytically inactive phosphatases in the pathophysiology of myotubular myopathy and suggests a novel therapeutic approach through identification of drugs that could stabilize the myotubularin-MTMR12 complex and hence ameliorate this disorder .
X-linked myotubular myopathy ( XLMTM ) is a congenital disorder caused by mutations of the MTM1 gene that encodes myotubularin [1] , [2] . Affected males are born with severe generalized hypotonia and weakness of skeletal muscles with respiratory insufficiency . In majority of cases the disease is fatal within the first months of life , but a proportion of affected males survive into their teens or beyond yet are non ambulant and require ventilatory support . Histopathologically , affected muscle fibers exhibit hypotrophy with a large number of centrally placed nuclei in a high proportion of myofibers . Thus , XLMTM is considered a subtype of centronuclear myopathy ( CNM ) [3] . MTM1 encodes a 3′-phosphoinositide ( PtdIns3P ) lipid phosphatase that catalyzes the dephosphorylation of phosphatidylinositol-3-phosphate ( PtdIns3P ) and phosphatidylinositol-3 , 5-bisphosphate PtdIns ( 3 , 5P ) 2 [4] , [5] , [6] , [7] . Phosphoinositides ( PIs ) are critical for a variety of physiological processes , including cell proliferation , cell death , motility , cytoskeletal regulation , intracellular vesicle trafficking , autophagy and cell metabolism [8] . PtdIns3P and PtdIns ( 3 , 5 ) P2 are both important regulators of membrane trafficking [9] . Similar to humans , loss of myotubularin function in animal models also results in clinico-pathological symptoms similar to myotubular myopathy signifying evolutionary conservation of mechanisms involving myotubularin [10] , [11] , [12] , [13] . Myotubularin is the prototypic member of one of the largest and most conserved protein lipid phosphatase subfamilies in eukaryotes , the myotubularin-related proteins ( MTMRs ) [14] , . There are nine catalytically active members of the MTMR family , and six members whose catalytic sites have been inactivated through missense changes ( designated myotubularin and MTMR1-14 ) . While several active members are known to play crucial roles in physiology and human diseases presumably related to loss of their enzymatic activities , the cellular functions of catalytically inactive members are still poorly understood . Catalytically inactive MTMR family members are highly conserved and retained in vertebrate genome from teleosts to humans during evolution [16] suggesting that these inactive members have important roles in development and diseases independent of their enzymatic activities , as functionally redundant family members are often lost during evolution [17] . This is further evident by a recent study that showed that catalytically inactive form of myotubularin could ameliorate many of the structural abnormalities seen in Mtm1 knockout mice [18] . Myotubularins form homo- and heteroligomers with themselves and other members of the MTMR family , and the catalytically inactive MTMRs are thought to largely form oligomers with active family members [14] , [19] . These interactions appear to regulate the sub-cellular localization or catalytic activity of the active members of the myotubularin family . In vitro , catalytically inactive MTMR5 and MTMR9 modulate the enzymatic activity of their interacting catalytically active partners , MTMR2 , and MTMR6 and MTMR8 , respectively [20] , [21] , [22] . Functional evidence for in vivo significance of these interactions comes from genetic studies in humans that show that mutations of either the catalytically active MTMR2 or its catalytically inactive binding partner MTMR13 result in similar forms of Charcot-Marie-Tooth disease [23] , [24] , [25] , [26] . Similarly , mutations in either of the binding partners MTMR2 ( catalytically active ) or MTMR5 ( catalytically inactive ) lead to defective spermatogenesis in mice , suggesting the biological importance of these interactions [27] , [28] . Myotubularin-related protein 12 ( MTMR12 ) , previously sometimes referred to as 3-phosphatase adaptor protein ( 3-PAP ) , is a catalytically inactive member of myotubularin family that interacts with MTM1 and MTMR2 to regulate their sub-cellular localization in in vitro studies of non-muscle cells [29] , [30] . The in vivo significance of such interactions remains unknown , especially with reference to disease process in myotubular myopathy that primarily affects the skeletal muscles due to lack of myotubularin function [10] . Protein-protein interactions are critical components of almost every biological process and any disruption of these networks leads to pathological conditions resulting in related diseases [31] , [32] . Therefore , a proper understanding of myotubular myopathy necessitates a comprehensive knowledge of the molecular mechanisms that govern such processes . We have employed zebrafish ( Danio rerio ) as a vertebrate animal model to understand the role of MTMR12 as a molecular regulator of myotubularin . Zebrafish are an excellent vertebrate model to study skeletal muscle development and disease due to high genomic synteny and similar clinico-pathological changes as seen in the human myopathies . Our studies revealed that deficiency of MTMR12 in zebrafish results in myopathy and impaired motor function similar to that caused by loss of myotubularin . Our data suggest MTMR12 affects muscle function primarily by regulating the stability of myotubularin in vivo . Studies in cells , mice and humans further implicate MTMR12 as an important player in the pathophysiology of XLMTM .
MTMR12 has been shown to interact with myotubularin and form oligomers in K562 and Cos7 cells [30] . To investigate if the binding between these proteins is a direct consequence of protein-protein interactions or is an indirect interaction mediated through another binding partner , GST-pull down assay was performed in vitro ( Figure 1A ) . Human myotubularin full-length protein was expressed as a GST fusion protein in E . coli ( MTM1-GST ) . Equivalent amounts MTM1-GST or control GST protein bound to glutathione beads were incubated with in vitro synthesized MTMR12 protein with a B10 tag ( MTMR12-B10 ) . MTM1-GST protein pulled down MTMR12-B10 protein whereas no interaction was observed with the GST alone , suggesting myotubularin and MTMR12 interact with each other by direct protein-protein interactions ( Figure 1A ) . The interaction between myotubularin and MTMR12 was also examined in the Cos1 cell line . MTM1 and MTMR12-B10 were over-expressed in Cos1 cells and cell extracts were immunoprecipitated by an antibody against myotubularin ( monoclonal , 1G1 ) . This resulted in co-immunoprecipitation of MTMR12-B10 with myotubularin but not with control IgG or empty beads confirming that MTMR12 also interacts with myotubularin in the cellular context as reported earlier ( Figure 1B ) [30] . Mutations in MTM1 result in XLMTM , which is a congenital myopathy that primarily affects the skeletal muscles in human patients and animal models . Therefore , the interaction between myotubularin and MTMR12 was investigated in skeletal muscles . Co-immunoprecipitation was performed using protein extracts from murine skeletal muscle ( tibialis anterior ) using an anti-myotubularin monoclonal antibody ( 1G1 ) ( Figure 1C ) . Antibody against myotubularin co-immunoprecipitated endogenous MTMR12 protein showing that myotubularin and MTMR12 interact in skeletal muscles . Control IgG or empty beads failed to immunoprecipitate MTMR12 suggesting the specificity of interactions between myotubularin and MTMR12 . In vivo interaction between these proteins was also investigated by immuno-colocalization of myotubularin and MTMR12 proteins in murine skeletal muscles . Immunofluorescence showed a striated expression of endogenous MTMR12 in mouse tibialis anterior skeletal muscle similar to the pattern of myotubularin staining , and double immunolabeling revealed that these colocalize at the resolution of confocal microscopy ( Figure 1D ) . In addition , partial co-localization of MTMR12 was observed with ryanodine receptors , a sarcoplasmic reticulum marker . The expression of MTMR12 was restricted to triads as no co-localization was observed with α-actinin at Z-lines ( Figure 1D ) . Therefore , a combination of GST pull down , co-immunoprecipitation and co-localization studies support the hypothesis that myotubularin and MTMR12 interact with each other in skeletal muscle . To understand the function of MTMR12 in vivo , zebrafish was used as a vertebrate animal model . The zebrafish mtmr12 gene encodes a protein of 736 amino acids that is 69% similar and 52% identical with the human MTMR12 protein . The expression of mtmr12 was analyzed by whole mount in-situ hybridization and RT-PCR in developing zebrafish embryos . Whole mount in-situ hybridization showed that mtmr12 transcripts were expressed ubiquitously in developing eyes , brain , heart and skeletal muscles at 1 day post fertilization ( dpf ) ( Figure 2A ) . This ubiquitous expression of mtmr12 was similar to the ubiquitous expression of mtm1 transcripts at 1 dpf . Similar to the zebrafish gene , human MTMR12 transcripts have also been found to express in all organs [33] . RT-PCR analysis further showed similar temporal expression of mtm1 and mtmr12 during zebrafish development ( Figure 2A ) . RT-PCR revealed that mtmr12 and mtm1 were first detected as maternal transcripts at the 1 cell stage . Expression of these maternal transcripts decreased during gastrulation and zygotic expression appeared around 8 hours post fertilization ( hpf ) . This zygotic expression persisted in all developmental stages tested ( until 5 dpf ) . In murine muscle cells , Mtmr12 transcripts were detected during proliferation as well as differentiation stages in C2C12 cells ( Figure 2B ) . The expression of Mtmr12 transcripts increased steadily during differentiation followed by a decrease in late differentiation . Comparison to Mtm1 expression revealed a similar expression pattern with a continuous increase with the progression of differentiation . Similar to the mRNA expression , MTMR12 protein expression was also highest during differentiation ( Figure 2B ) . Similar expression of mtmr12 and mtm1 during zebrafish development and in muscle cell differentiation suggests that they may be involved in similar physiological processes . To investigate in vivo functions of the mtmr12 gene in zebrafish , anti-sense morpholino technology was employed to achieve functional gene knockdown ( Figure 2C–E ) . Morpholinos were designed to disrupt either the translation or splicing of mtmr12 transcripts . Knockdown using either a splice site-morpholino targeting the exon3-intron3 junction or the translational ( ATG ) morpholino resulted in similar phenotypes at relatively low morpholino concentrations ( 3 . 5–5 . 0 ng ) , suggesting the specificity of their action . Microinjection of 1cell embryos with splice site morpholinos resulted in mis-splicing and exclusion of exon 3 from the mature mRNA as detected by RT-PCR assay ( Figure 2D ) . Mtmr12 knockdown fish ( mtmr12 morphants ) are smaller in size compared to controls and exhibited a dorsal curvature through the back and tail , instead of the normal flat dorsum , similar to mtm1 knockdown fish ( Figure 2C ) . Axial skeletal muscles of zebrafish embryos were examined using a birefringence assay that involves examination of axial skeletal muscles of live zebrafish embryos using polarized filter microscopy . Skeletal muscles of mtmr12 morphant embryos showed a reduced birefringence in comparison to the control morpholino injected fish suggesting a defect in skeletal muscle organization . Several mtmr12 morphants also displayed pericardial edema . Similar phenotypes were obtained with both translational as well as spice-site morpholinos suggesting the specificity of morpholino knockdown . None of the commercially available antibodies showed reactivity to zebrafish MTMR12 protein , therefore , the rest of the studies were performed with the splice-site blocking morpholino against exon3-intron3 junction as mRNA knockdown could be assayed . To further validate the specificity of these phenotypes and to rule out any off-site targeting effect of morpholinos , mRNA rescues were performed . Human MTMR12 mRNA was co-injected with mtmr12 morpholino in zebrafish embryos . Over-expression of MTMR12 mRNA in mtmr12 morphant fish resulted in a rescue of phenotypes seen in mtmr12 morphant fish indicating the specificity of morpholino targeting ( Figure 2E ) . To understand if the interacting partners myotubularin and MTMR12 are involved in similar biological processes or have functions that are independent of each other , double knockdowns were performed . Double knockdown fish were smaller in size then either mtm1 or mtmr12 morphant fish . Polarized live microscopy of morphant fish also showed that birefringence of mtmr12-mtm1 double morphant zebrafish muscle was lower than in either mtm1 or mtmr12 morphant fish suggesting a severe muscle phenotype . The exacerbated phenotypes of double knockdown fish suggests that in addition to regulating similar biological processes these genes may be involved in different processes independent of each other . To understand the consequences of MTMR12 deficiency on motor function , behavioral analysis of zebrafish embryos was performed . During early development , zebrafish embryos hatch out of their chorions by regular contractions of their skeletal muscles . Typically , approximately 86±7 . 4% of control embryos hatch by 60 hpf . In contrast , only 25±9 . 3% of mtmr12 morpholino-injected embryos hatched by this time , consistent with a continued decrease in motor activity early in development ( P<0 . 01 , n = 100–170 ) ( Figure 2F ) . Mtm1 morphant embryos displayed similar behavior as mtmr12 morphants ( 19 . 3±3 . 39 , n = 100–150 , P<0 . 001 ) . Knockdown of both MTMR12 and myotubularin resulted in a significant decrease in hatching behavior in comparison to the mtmr12 alone morphant fish . ( 9 . 33%±4 . 02 , P<0 . 001 , n = 100–130 ) . Mtmr12 morphant fish were largely immotile and their touch evoked response was blunted; instead of rapidly swimming out of the field of view like control fish ( 6 . 44±0 . 712 cm/0 . 1 sec ) , they twitched and only moved several lengths when stimulated with a needle ( 2 . 16± ( 2 . 832 cm/0 . 1 sec ) , suggesting a significant degree of overall muscle weakness ( Movie S1 , Figure 2G ) . This decrease in touch-evoked response was similar to that of mtm1 morphant fish ( 2 . 32±0 . 855 cm/0 . 1 sec ) ( Movie S1 , Figure 2G ) . The touch-evoked response was also evaluated in mtm1-mtmr12 double morphant fish . In comparison to either mtm1 or mtmr12 knockdown embryos , double morphant embryos showed a greater degree of reduction in touch-evoked escape response ( 0 . 98±0 . 230 cm/0 . 1 sec ) ( Movie S1 , Figure 2G ) . The delayed chorion hatching and diminished touch-evoked escape behaviors showed that mtmr12 is required for normal motor function in zebrafish . To study the effect of MTMR12 deficiency on skeletal muscle structure , ultrathin toluidine blue-stained longitudinal sections of control and mtmr12 knockdown fish were examined at 3 dpf . Histological examination of control muscle showed well-organized myofibers with elongated nuclei that were localized to the periphery of muscle fibers ( Figure 3A ) . In mtmr12 knockdown fish , areas lacking sarcomeric organization were observed ( Figure 3A , arrowhead ) . Moreover , occasional rounded , central nuclei were seen in skeletal muscles of the morphant fish but were absent in the control fish ( Figure 3A , arrow ) . Remarkably , the sarcomeric disorganization with central nucleation was very similar to the histological changes observed in mtm1 knockdown fish ( Figure 3A , arrow ) . The proportion of fibers with central nuclei was almost similar in mtmr12 morphant ( 54 . 5±7 . 1% ) and mtm1 morphant ( 57 . 8±7 . 1% ) fish and significantly higher than normal controls ( 2 . 2±1 . 1% ) . Mtm1-mtmr12 double knockdown fish exhibited severe muscle abnormalities with larger number of myofibers displaying sarcomeric disorganization and central nucleation ( 69 . 2±6 . 4% ) than single morphants ( Figure 3A ) . To evaluate if the sarcomeric defects observed in mtmr12 morphant fish are developmental or due to degenerative changes in muscle , skeletal muscle histology was evaluated at different time points during zebrafish development . A comparison of Hematoxylin and Eosin stained skeletal muscle sections at 2 dpf and 3 dpf showed an increase in sarcomeric disorganization at 3 dpf in mtmr12 morphant fish ( Figure 3C ) . This suggests the sarcomeric defects seen in mtmr12 morphant fish are due to degenerative processes in disease state . The number of central nuclei per myofiber showed no significant change during development in mtmr12 morphants ( Figure 3D ) . A comprehensive histological analysis of Hematoxylin and Eosin stained sections of wild-type and mtmr12 morphant showed no other histological abnormalities in other organs . To identify ultrastructural defects in sub-cellular compartments of skeletal muscle , transmission electron microscopy was performed at 3 dpf ( Figure 4 ) . Longitudinal views of the skeletal muscle in mtmr12 morphant fish revealed significant myofibrillar disarray in comparison to highly organized myofibrillar structures with peripheral elongated nuclei in control fish ( Figure 4A , C ) . Notably , skeletal muscle of mtmr12 morphants showed an increase in absent or disorganized Triads in the myofibers in comparison to the control fish ( Figure 4A , C ) . In addition , similar to the whorled membrane structures reported in myotubularin deficiency [13] , mtmr12 fish also exhibited whorled membrane structures in skeletal muscle ( Figure 4E ) . Mtm1 morphant fish displayed sarcomeric disorganization , with central nucleation and triad disorganization as previously reported for myotubularin deficiency in zebrafish , mouse and humans ( Figure 4B ) . The ultrastructural defects in double knockdown fish were more severe than either mtm1 or mtmr12 knockdown fish ( Figure 4D ) . Myofibers lacked the sarcomeric organization with absence of Z-lines in many myofibers . The numbers of disorganized triads also showed a small but significant increase in double knockdown fish ( 68 . 8±10 . 08% ) compared to mtm1 ( 63 . 5±8 . 00% ) or mtmr12 ( 57±12 . 12% ) fish ( p<0 . 005 , n = 5 embryos , 15 myofibers in each embryo ) ( Figure 4G ) . Like in myotubularin and MTMR12 deficient muscles , whorled membrane structures were also observed in mtmr12-mtm1 double knockdown zebrafish ( 4F ) . These abnormal membrane structures are seen in several types of myopathies , however , their functional role in disease pathology is not known . In a previous study on double mtm1-mtmr14 knockdown , the severe phenotype was a result of an increase in autophagy in the absence of both proteins [34] . Ultrastructural examination of mtm1-mtmr12 double knockdown exhibited no increase in autophagic vacuoles suggesting different pathological mechanism in two different disease states . Catalytically inactive myotubularin family members have been shown to regulate the activity and/or sub-cellular localization of their catalytically active interacting partners . Previous co-transfection experiments in cell culture have suggested that MTMR12 is required to regulate the subcellular localization of myotubularin [30] . Therefore , to investigate if MTMR12 controls the subcellular localization of myotubularin in zebrafish skeletal muscle , immunofluorescence studies were performed . Immunostaining of skeletal muscle of control fish with antibody against myotubularin detected a striated expression pattern of myotubularin protein corresponding to the triad compartment ( Figure 5 ) . Strikingly , highly reduced levels of myotubularin labeling were seen in morphant muscle as compared to control skeletal muscle . However , the residual myotubularin protein showed similar localization as the control fish . These data suggest that loss of MTMR12 does not affect subcellular localization of myotubularin in skeletal muscle in vivo . These observations suggest that MTMR12 is likely involved in regulating the stability of myotubularin . To address this point , we quantified myotubularin protein levels in mtmr12 morphant fish . Western blot analysis , performed in three independent groups of embryos injected with mtmr12 morpholino , revealed a ∼90% reduction in myotubularin levels in mtmr12 morphant fish in comparison to the controls ( Figure 6A ) . This suggests that myotubularin-MTMR12 interactions result in stabilization of myotubularin in zebrafish skeletal muscle . To investigate whether MTMR12 regulates myotubularin levels in a mammalian system , Mtmr12 siRNA C2C12 cell lines were created . SiRNA-mediated knockdown of Mtmr12 in C2C12 myoblasts or differentiated myotubes resulted in decreased levels of myotubularin protein compared to scrambled control siRNAs ( Figure 6B and 6C ) . MTMR12 deficient C2C12 myotubes also mimicked cellular changes previously seen in Mtm1 knockdown C2C12 cells [35] . Mtmr12 knockdown in C2C12 cells resulted in an increase in levels of the intermediate filament protein desmin which were associated with abnormal filament shape ( arrow ) in both myoblasts and myotubes ( Figure 6C and 6D ) . However , no change in differentiation markers such as myogenin or myotube formation was observed in Mtmr12 knockdown cells ( Figure 6C and 6E ) suggesting that Mtmr12 deficiency does not affect the differentiation program in C2C12 cells . To determine whether MTMR12 protein stability is regulated by myotubularin , MTMR12 levels were assessed in Mtm1 knockout mice and Mtm1 knockdown C2C12 cells ( Figure 6F–G ) . In Mtm1 knockout mice , expression of MTMR12 was reduced in skeletal muscle in pre-symptomatic ( 2 weeks ) as well as in symptomatic phases of disease progression ( 5 weeks ) ( Figure 6F ) . Similarly , in Mtm1 knockdown cell lines , a significant decrease in MTMR12 expression levels was also observed ( Figure 6G ) . Combined , these data support the notion that the MTMR12-myotubularin interaction enhances stability of the complex in muscle cells in vitro and in skeletal muscle in vivo . The only known and well-characterized biochemical functions of myotubularin related proteins is the dephosphrylation of phospholipids . Previous data have shown that Mtm1 mice and mtm1 zebrafish morphants display an increase in PtdIns3P levels ( the substrate of myotubularin ) [11] , [13] To investigate if mtmr12 knockdown also affects PItdIns3P levels in skeletal muscle as seen in myotubular myopathy , PtdIns3P staining was performed on control and mtmr12 morphant skeletal muscle . Indirect immunofluorescence on MTMR12-deficient muscle showed an increase in PtdIns3P staining compared to control suggesting that the reduction of myotubularin levels is paralleled by an overall decrease in its enzymatic activity ( Figure 5 ) . To quantify the levels of PtdIns3P in the absence of either myotubularin or MTMR12 or both , a lipid-protein overlay PtdIns3P ELISA was performed on lipid extracts from 3 dpf zebrafish morphant fish . As shown previously [11] , [13] , absence of myotubularin resulted in an increase in PtdIns3P levels in mtm1 morphant embryos ( Figure 5B ) . In the absence of MTMR12 or both myotubularin and MTMR12 , a significant increase in PtdIns3P levels were seen over levels in both control and myotubularin morphant fish . This suggests that in addition to regulating myotubularin activity , MTMR12 may also be regulating the functions of other PtdIns3P phosphatases . Mtmr12 morphant fish displayed reduced levels of myotubularin protein and similar pathological changes to those seen in myotubularin deficiency , suggesting that the ability of MTMR12 to stabilize endogenous myotubularin levels in vivo may be a primary physiological role . These findings gave rise to the intriguing possibility that overexpression of myotubularin may be able to reverse the phenotypes associated with MTMR12 deficiency in zebrafish . To test this hypothesis , we overexpressed exogenous human MTM1 mRNA to investigate its ability to rescue the muscle phenotype in mtmr12 morphant fish and evaluated the resulting skeletal muscle phenotypes by birefringence assay and electron microscopy ( Figure 7 ) . Overexpression of MTM1 mRNA in mtmr12 morphant fish restored the birefringence to almost normal levels ( Figure 7A ) . While the skeletal muscle birefringence was drastically improved , the length of zebrafish embryos was smaller ( 75±5 . 4% of control fish ) suggesting MTMR12 may have additional functions independent of myotubularin . The ultra-structure of skeletal muscle was also significantly improved with a decrease in fraction of abnormal triads . This result strengthens the finding that an important role of MTMR12 is to provide stability to myotubularin protein and hence regulate its function . The mtm1-mtmr12 double knockdown fish exhibited severe muscle defects consistent with the idea that MTMR12 may have independent functional roles in addition to providing stability to myotubularin . This was investigated by overexpressing MTM1 RNA in mtm1-mtmr12 double knockdown fish . Overexpression of MTM1 resulted in a moderate improvement in birefringence as well as skeletal muscle pathology in double knockdowns as quantified by number of normal triads in skeletal muscle of rescued fish . Nevertheless , rescued fish were still smaller in size ( 0 . 64±0 . 029 ) than wild-type controls ( 1 . 00±0 ) , consistent with the notion that MTMR12 may have functions independent of MTM1 in vivo ( Figure 7 C and E ) . To test the ability of MTMR12 to rescue the phenotypes associated with myotubularin deficiency in skeletal muscle , human MTMR12 RNA was overexpressed in mtm1 morphant fish . This resulted in partial rescue of the muscle phenotype in these fish as seen by an increase in birefringence and body length ( 71±4 . 9% of normal controls ) of mtm1 morphant fish rescued with MTMR12 mRNA versus mtm1 morphant ( 64±3 . 68% of normal controls ) , however , without any significant reduction in disorganized triads ( Figure 7B , C and E ) . These observations suggest that it is the missing catalytic activity or other gene-specific function of myotubularin that is primarily responsible for the pathology of XLMTM . The significance of MTM1-MTMR12 interactions was investigated in the human neuromuscular disease XLMTM . In XLMTM , more than 200 mutations have been reported that are distributed on different domains of the myotubularin protein ( http://www . dmd . nl/ ) . In addition to nonsense mutations , a large number of missense mutations in various myotubularin domains have been shown to be pathogenic . To test , if disease causing mutations in myotubularin affect its interaction with MTMR12 , interactions between mutant myotubularins and wild-type MTMR12 were examined . A series of human myotubularin proteins modeling human missense mutations in different domains were constructed ( Figure 8A ) . Exogenous wild type or mutant myotubularins were co-expressed with MTMR12-GFP protein in Cos7 cells . Immunoprecipitation of protein extracts with a myotubularin specific antibody revealed that missense mutations in the GRAM or RID domains abolished the interaction of myotubularin with MTMR12 ( Figure 8B ) . As over-expression in cell culture may not represent the physiological milieu , myotubes from XLMTM patients were also used to test if the pathogenic mutations on myotubularin protein affect myotubularin-MTMR12 interactions . Examination of XLMTM patient myotubes showed all mutations tested resulted in highly reduced levels of myotubularin compared to control samples . In addition , MTMR12 levels were also reduced in these patients suggesting an overall perturbation of myotubularin-MTMR12 complexes in XLMTM ( Figure 8C ) .
Studies presented in this work were aimed at gaining insights into the molecular regulatory mechanism ( s ) of myotubularin function in vivo . Previous studies have shown that MTMR12 is an interacting partner of myotubularin in vitro [30] , [33] . Here , we show that absence of catalytically inactive phosphatase MTMR12 protein resulted in skeletal muscle myopathy and pathological changes similar to XLMTM due to abrogation of protein-protein interactions between myotubularin and MTMR12 resulting in reduced stability and loss of myotubularin protein function . The interaction between myotubularin and MTMR12 in skeletal muscle and co-localization at the triad certainly suggests that they might be functioning together in similar biological processes in muscle cells . This hypothesis is supported by knockdown studies on the mtmr12 gene in zebrafish that resulted in myopathic muscle in the affected fish similar to the myotubularin deficient zebrafish model [13] . To understand if these proteins function in similar biological processes or play roles in other processes independent of each other , mtm1-mtmr12 double knockdown zebrafish were created . The phenotype of double knockdown zebrafish was more severe than fish deficient in either myotubularin or MTMR12 alone . Further , inability of MTM1 to rescue all the defects observed in double knock-down fish suggests that these proteins may play additional functions that are independent of each other either by interacting with other proteins within or outside of the myotubularin family . Previous studies have shown that in addition to myotubularin , MTMR12 also interacts with another catalytically active member , MTMR2 , by co-immunoprecipitation and yeast two hybrid interactions [29] , [30] . Therefore , future studies on identification of protein complexes of myotubularin and MTMR12 proteins may help in identifying other pathways that are regulated by these proteins . The presence of similar pathological changes in mtmr12 knockdown and XLMTM muscles , such as myofibrillar disarray , excessive central nucleation , triad disorganization and presence of whorled membranous structures , suggests that mtmr12 is a crucial regulator of disease pathology in XLMTM . Moreover , these pathological changes in zebrafish manifest early in zebrafish development ( 2–3 dpf ) , similar to pathological changes seen in human patients . As loss of mtmr12 resulted in clinical symptoms similar to those associated with centronuclear myopathies , MTMR12 represents an excellent candidate gene for patients with centronuclear myopathy but unknown genetic diagnosis . However , sequencing of 108 such cases failed to identify any pathogenic mutations in MTMR12 , suggesting MTMR12 may account for a small subset of patients or is mutated in a clinically different disease , perhaps with a skeletal muscle component , but related also to other functions of MTMR12 ( V . A . Gupta , unpublished data ) . In our cellular models of MTMR12 deficiency , highly reduced levels of MTM1 were observed . Therefore , any genetically unknown cases exhibiting low levels of myotubularin without any MTM1 mutations may also be good candidates for testing for MTMR12 mutations . Next generation sequencing technologies have been exhibited great promise in identifying rare gene variants and may yet identify MTMR12 mutations in the future . Regardless , these studies show a crucial role for MTMR12 function in XLMTM disease pathology . Complex formation between some catalytically inactive and active partners in the myotubularin family has been shown to increase the activity of the catalytically active binding partner either by recruitment to specific membrane subdomains rich in lipid substrate or by increasing the allosteric activity . In the absence of MTMR12 an increase in PtdIns3P was observed suggesting a decrease in myotubularin and/or another partner's enzymatic activity . However , unlike other catalytically active-inactive pairs , MTMR12 primarily regulates the function of myotubularin protein by affecting protein levels instead of modulating the enzymatic activity [22] , [36] . Interestingly , PtdIns3P levels were higher in mtmr12 morphant fish in comparison to mtm1 morphant fish , suggesting that MTMR12 may function in regulating the enzymatic activity/protein stability of other phospholipid phosphatases in cells . In the absence of MTMR12 , highly reduced levels of myotubularin protein was observed suggesting that protein-protein interactions between these proteins are required for maintaining myotubularin stability . Similarly , decreased levels of MTMR12 were seen in myotubularin deficient cells and mice suggesting that stability of MTMR12 is also dependent on the interaction with myotubularin . The abnormalities observed in MTMR12 deficient fish appear mainly due to loss of function of myotubularin as overexpression of MTM1 mRNA dramatically improved the skeletal muscle defects observed in mtmr12 morphant embryos while overexpression of catalytically inactive MTMR12 mildly improved muscle pathology in mtm1 morphant embryos . This suggests that enzymatic activity of myotubularin is required for its protein function , which was also seen , in a previous study where myotubularin deficiency could be rescued by catalytically active MTMR1 and MTMR2 [13] . Recent work also suggests that several but not all structural abnormalities observed in myotubularin deficiency can be rescued by over-expression of a catalytically inactive form of myotubularin [18] . As the catalytically inactive MTMR12 only partially but significantly rescues myotubularin function , it supports these recent findings [18] and suggests that MTMR proteins partially compensate for the lack of MTM1 through functions independent of their catalytic activity . This implies that the inability of catalytically inactive MTMR12 to rescue myotubularin function is not only due to the lack of phosphatase activity but could also be due to other properties of myotubularin not compensated by all MTMRs . MTMR12 also interacts with another catalytically active member , MTMR2 , by co-immunoprecipitation and yeast two hybrid [29] , [30] . Further MTMR2 and MTMR13 also interact by direct protein-protein binding . Mutations in genes encoding MTMR2 and MTMR13 have both been associated with Charcot-Marie-Tooth ( CMT ) disease raising the possibility that MTMR12 deficiency may result in neurological defects such as those seen in CMT . Mtmr2 as well as Mtmr13 knockout mice exhibit a progressive neuropathy that becomes evident much later in the life span of these mice ( ∼6 months ) [27] , [37] . In comparison , no such neurological defects were seen in mtmr12 morphant fish , which could be due to early mortality of these embryos ( between 3–5 dpf ) , before such changes become evident . Finally , it is worth noting that the ability of human MTM1 mRNA to rescue zebrafish mtm1 morphants reinforces the notion that these orthologues have been functionally conserved despite their considerable evolutionary distance , validating the relevance of the zebrafish model to studies of human XLMTM . Studying the interaction of MTMR12 with myotubularins modeling various human mutations illuminates the role of myotubularin-MTMR12 interactions in XLMTM . Missense mutations in the N-terminal GRAM and RID domains resulted in abolishment of interaction between myotubularin and MTMR12 . This was a surprising finding as previous studies have shown the myotubularin family members interact with each other through their SID or coiled-coil domains . It was also shown that an isoform of MTMR12 lacking the SID domain showed no interaction with myotubularin protein [30] . One reason for this discrepancy is that many of the previous studies have been performed using deletion constructs of different domains that may have a different effect on protein confirmation and thus its interactions than the missense changes we studied . Nonetheless a study of interactions between MTMR6 and KCa3 . 1 proteins has shown that along with the coiled-coil domain , the PH-GRAM domain is also required for protein-protein interactions [38] . Similarly , oligomerization of MTMR2-MTMR13 in a complex occurs independent of coiled-coil domains [39] . Analysis of XLMTM patient myotubes showed that many mutations of myotubularin that result in low levels of myotubularin protein also lead to decreases in MTMR12 levels . Many patients with these types of myotubularin mutations are described with severe phenotypes and a further reduction of MTMR12 may exacerbate the clinical severity as seen in myotubularin-MTMR12 double knockdown zebrafish . Apart from the type III intermediate filament desmin [35] , our study identifies a second interactor of myotubularin in skeletal muscle and underlines the concept that protein-interactions between myotubularin and MTMR12 are crucial for disease pathology in XLMTM . We propose a model whereby disruption of interactions between MTM1-MTMM12 results in destabilization of both partners in the complex , leading to centronuclear myopathy ( Figure 9 ) . As loss of MTMR12 results in reduction of myotubularin , we suspect that primary mutations of MTMR12 may result in centronuclear or related myopathies . As protein interactions play critical roles in almost all biological processes , efforts are currently focused on identifying drugs that can stabilize protein-protein interactions [40] . Stabilizing protein-protein interaction between myotubularin and MTMR12 may result in restoration of normal skeletal muscle function in XLMTM patients with certain missense mutations of myotubularin .
Fish were bred and maintained as described previously [41] . Control embryos were obtained from the Oregon AB line and were staged by hours ( hpf ) or days ( dpf ) post fertilization at 28 . 5°C . All animal work was performed with approval from the Boston Children's Hospital Animal Care and Use Committee . Animals were housed in a temperature-controlled room ( 19–22°C ) with a 12:12 hr light/dark cycle . Mice were humanely killed by CO2 inhalation followed by cervical dislocation , according to national and European legislations on animal experimentation . Isoform-specific riboprobes were constructed from the 3′UTRs of mtm1 and mtmr12 using adult zebrafish RNA . Total RNA was extracted from adult zebrafish muscle tissue using Trizol ( Invitrogen , Carlsbad , CA , USA ) . cDNAs were synthesized using superscript RT-PCR system ( Invitrogen ) and cloned in pZEM7Z ( + ) using XhoI and BamHI sites respectively . Sense or antisense digoxigenin-labeled riboprobes were synthesized by in vitro transcription using dig-labeling kits ( Roche Applied Sciences , Indianapolis , IN , USA ) . Whole mount in situ hybridization was performed as described [42] . Imaging was performed using a Nikon SMZ1500 microscope with a Spot camera system . Two splice site blocking morpholinos targeting different exon-intron boundaries , and a translational blocking morpholino , were designed to knockdown zebrafish mtm1 or mtmr12 transcripts ( Genetools , Philomath , OR , USA ) . The morpholino sequences are mtm1 ( translational ) : AGCCAGACCCTCGTCGAAAAGTCAT , mtmr12 ( translational ) : CTCCTCCGCTCCCCAAACTCAACAT , mtmr12 ( exon3-intron3 ) : GCCCGGTCAACTGTCCTTACCATCT . Morpholino against human β-globin was used as a negative control for all injections . Morpholinos were dissolved in 1X Danieau buffer and 1–2 nl ( 1–10 ng ) were injected into 1cell embryos . For rescue experiments , full-length human MTM1 and MTMR12 cDNAs were cloned in to a PCSDest destination vector ( a gift from Nathan Lawson ) using Gateway technology ( Invitrogen , Carlsbad , CA , USA ) . mRNA was synthesized in vitro using mMessage kits ( Ambion , Austin , TX , USA ) . 50–200 pg of mRNA was injected into embryos at the 1 cell stage . Indirect immunofluorescence staining was performed on zebrafish frozen sections as described previously [43] . Primary antibodies used for zebrafish experiments were rabbit anti myotubularin HPA010008 ( Sigma-Aldrich , St . Louis , MO , USA ) and anti PtdIns3P , Z-P345b ( Echelon Bioscience , Salt Lake City , UT , USA ) . For mouse muscle staining , tibialis anterior sections ( 8 µM ) were labeled successively with the anti-MTM1 polyclonal antibody ( 2827 ) [35] and the anti-MTMR12 antibody ( GTX119163 , GeneTex Inc . , Irvine , CA , USA ) . Briefly , after permeabilization and blocking , the anti-MTM1 antibody was applied ( diluted at 1/500 ) on section for 2 h at RT . After washing steps , sections were incubated with the secondary antibody coupled to Alexa fluor 488 ( Invitrogen ) for 45 min followed by a second step of washing cycles and a blocking step . Then , the anti-MTMR12 antibody was applied on section for 2 h at RT and revealed by the incubation with a secondary antibody coupled to Alexa fluor 594 ( Invitrogen ) . After final washing , sections were fixed , mounted and observed under confocal microscope ( Leica SP2 MP confocal microscope ) . The monoclonal antibody against RyR1 ( 1C3 ) was a gift from Dr Isabelle Marty ( Grenoble , France ) . The mouse anti α-actinin antibody ( clone EA-53 ) was from Sigma ( Sigma-Aldrich , St . Louis , MO , USA ) . Zebrafish embryos were homogenized in a buffer containing Tris-Cl ( 20 mM , pH 7 . 6 ) , NaCl ( 50 mM ) , EDTA ( 1 mM ) , NP-40 ( 0 . 1% ) and complete protease inhibitor cocktail ( Roche Applied Sciences ) . Western blotting was performed as described previously [43] . Primary antibodies used were mouse monoclonal anti-myotubularin 1G1 [19] , rabbit polyclonal anti-myotubularin R2827 [44] and mouse anti-β-actin clone AC-15 ( Sigma-Aldrich , A5441 at 1∶2000 ) . Mouse monoclonal antibodies for Desmin and the sarcomeric α-actinin were purchased from Sigma-Aldrich ( clone D33 and Clone EA-53 , respectively ) . The mouse monoclonal antibody for myogenin was from R&D system ( Clone 671037 ) . Protein bands were quantified using Quantity One software ( Biorad , Hercules , CA , USA ) . C2C12 cells were cultured in proliferation medium ( Dulbecco medium supplemented with 20% FCS and 400 U/ml of Gentamycin ) and differentiation was enhanced by decreasing the FCS in the media to 2% for 1 day and then accelerated by replacing the FCS by 5% Horse serum . Cells were kept in differentiation medium for 9 days with medium replacement every 2 days . For siRNA experiments the Accel Smart Pool siRNA ( Thermo Scientific , Dharmacon ) were used to knockdown Mtmr12 in C2C12 . Myoblasts ( at 30–40% of confluence ) were washed with PBS and incubated with siRNA medium containing permeable Accel siRNA pool against Mtmr12 ( CCAGCAGUAUAGAGGAAUA , GCGCUAUUUACGUUGGAUU , CCCGUGGGUUUAUAUAUUG , GGAUUAAGCUAUUAGACUG ) or scrambled control siRNAs diluted to the appropriate concentration . After 72 hours , cells were washed with PBS and incubated with proliferation medium or differentiation medium and left for 9 days in order to obtain myotubes . Myotubes ( containing min 2 nuclei ) were counted blindly under the microscope ( bright field ) at different stages and minimums of 100 cells per well were counted . MTMR12 constructs were transferred to Gateway destination vectors for eukaryotic expression ( pSG5 from Agilent Technologies ( Santa Clara , CA , USA ) , with tag corresponding to the B10 epitope of estrogen receptor or GFP ) and pSG5-MTM1-B10 and pcDNA3 . 1-MTM1 were used for co-immunoprecipitation assays in Cos1 cells . For GST-pull down experiment , MTM1 cDNA was inserted into the prokaryotic expression vector ( pGex4T3 , Invitrogen , Carlsbad , CA , USA ) . Myotubularin recombinant proteins were produced in the BL21-Rosetta 2 strain ( Novagen , Billerica , MA , USA ) ; GST fusion proteins were purified and coupled to glutathione sepharose beads as described before [35] . pSG5-MTMR12-B10 was translated in vitro according to the manufacture protocol ( TNT coupled reticulocyte lysate System , Promega , Madison , WI , USA ) . Resulted translated protein was diluted in Co-IP buffer: ( 50 mM Tris-Cl pH 7 . 5 , 100 mM NaCl , 5 mM EDTA , 5 mM EGTA , 1 mM DTT , 0 , 5% Triton X-100 , 2 mM PMSF ) supplemented with complete protease inhibitor tablet ( Roche Applied Sciences , Indianapolis , IN , USA ) and 1 mM Leupeptin and 1 mM pepstatin ( Sigma , St . Louis , MO , USA ) . Homogenates were centrifuged at 14 . 000×g and pull down was performed as previously described [35] . GST coupled beads was used as negative control . For Co-IP Cos7 cells were transiently transfected with pcDNA3 . 1-MTM1 and pSG5-MTMR12-B10 constructs or with pSG5-MTM1-B10 and pSG5-MTMR12-GFP for 24 hours and homogenized in ice-cold lysis buffer ( 10 mM Tris-Cl , pH 7 . 6; 140 mM NaCl; 5 mM EDTA; 5 mM EGTA; 0 . 5% [v/v] Triton X-100; and 2 mM PMSF ) . Homogenates were incubated with mouse monoclonal anti-myotubularin 1G1 [19] , or B10 epitope ( Mab anti-B10 , IGBMC , Illkirch , France ) [35] . Interacting proteins were analysed by Western blot as mentioned before . The panel of amino acid changes was engineered by PCR-based mutagenesis from the cDNA encoding the wild-type protein ( MTM1 ) using PFU DNA polymerase ( Agilent technologies ) . All constructs were verified by sequencing . Co-IP experiments in muscle were performed from fresh murine tibialis anterior muscles that were dissected and homogenized with a dounce homogenizer in ice-cold co-IP buffer ( 50 mM Tris-Cl , pH 7 . 5; 100 mM NaCl; 5 mM EDTA; 5 mM EGTA; 1 mM DTT; 0 . 5% Triton X-100; and 2 mM PMSF ) supplemented with 0 . 05% ( w/v ) SDS . Lysates were centrifuged at 14 . 000×g at 4°C and precleared with 50 µl of G-sepharose beads ( GE Healthcare ) and subsequently incubated with antibodies of interest for 12–24 hours at 4°C . Protein G-sepharose beads ( 50 ml ) were then added for 4 hours at 4°C to capture the immune complexes . Beads were washed 4 times with co-IP buffer and 1 time with high stringency co-IP buffer ( with 300 mM NaCl ) . For all experiments , two negative controls consisted of a sample lacking the primary antibody ( Beads ) and a sample incubated with IgG . Resulting immune-bound complexes were eluted in Laemmli buffer and analysed by SDS-PAGE and Western blotting . PtdIns3P mass ELISA was performed on lipid extracts from whole zebrafish embryos from 3 independently injected clutches ( number of embryos in each clutch = 50 ) following manufacturer's recommendations ( Echelon Biosciences , Salt Lake City , UT , USA ) . Extracted lipids were resuspended in PBS-T with 3% protein stabilizer and then spotted on PtdIns3P Mass ELISA plate . Following ELISA , PtdIns3P levels were detected by measuring absorbance at 450 nm on a plate reader . Specific amounts were determined by comparison of values to a standard curve generated with simultaneous readings of known amounts of PtdIns3P . To detect mtmr12 knockdown in 3 dpf zebrafish embryos , RNA was prepared using RNeasy fibrous tissue mini kits ( Qiagen , Valencia , CA , USA ) . cDNA was prepared using high capacity RNA-to-cDNA mastermix kits ( Ambion , Austin , TX , USA ) . RT- PCR was performed using equal amounts of RNA from control and mtmr12 morphant zebrafish . To measure Mtmr12 and Mtm1 expression levels in C2C12 cells , total RNA was purified from C2C12 ( from 0 to 9 days of differentiation ) cells using Trizol reagent ( Invitrogen ) according to manufacturer's instructions . cDNAs were synthesized from 2 to 5 µg of total RNA using Superscript II reverse transcriptase ( Invitrogen ) and random hexamers . Quantitative PCR amplification of cDNAs was performed on Light-Cycler 480 and Light-Cycler 24 instruments ( Roche Applied Sciences ) using 58°C as melting temperature . Gapdh gene expression was used as control as expression of this gene varies little during C2C12 cell differentiation . Primers used were: Mtm1 ( F: catgcgtcacttggaactgtgg , R : gcaattcctcgagcctcttt ) , Mtmr12 ( F: tgtctgaggtacacaaaggag , R: agccttcattcacactcactg ) and Gapdh ( F: agctttccagaggggccatccaca , R : ccagtatgactccactcacggcaa ) . Zebrafish embryos were fixed in formaldehyde-glutaraldehyde-picric acid in cacodylate buffer overnight at 4°C followed by osmication and uranyl acetate staining . Subsequently , embryos were dehydrated in a series of ethanol washes and finally embedded in TAAB epon ( Marivac Ltd . , Nova Scotia , Canada ) . 95 nm sections were cut with a Leica Ultracut microtome , picked up on 100 m Formvar coated Cu grids and stained with 0 . 2% Lead Citrate . Sections were viewed and imaged under a Philips Tecnai BioTwin Spirit Electron Microscope ( Electron Microscopy Core , Harvard Medical School ) . Centrally nucleated myofibers were quantified by analyzing serial sections from 6 different embryos . The relative number of centrally nucleated fibers in the middle somites ( 10–13 ) were counted . Total number of triads were counted in at least 15 myofibers within each embryo ( n = 5 embryos ) . Body length was measured in 10–15 embryos in each group . Data were statistically analyzed by parametric Student t-test ( two tailed ) and were considered significant when P<0 . 05 . All data analyses were performed using XLSTAT software .
|
Congenital myopathies are a group of heredity diseases characterized by muscle weakness and impaired locomotion that manifest in both children and adults . X-linked myotubular myopathy ( XLMTM ) is a subtype of congenital myopathy that predominantly affects males and is caused by mutations in the myotubularin ( MTM1 ) gene . To date , more than 200 pathogenic mutations have been identified in MTM1 . However , no effective therapy is available to treat patients presenting with XLMTM . This is largely due to a lack of understanding of molecular processes perturbed in the XLMTM disease state , thereby limiting the availability of suitable therapeutic targets . In this study , we show that catalytically inactive MTMR12 interacts with myotubularin in skeletal muscle . This complex formation is required to provide stability to myotubularin in the normal functioning of skeletal muscle and these interactions appear to be disrupted in XLMTM . This work therefore offers a novel direction for therapy development , both in XLMTM and other genetic diseases , by identifying crucial protein interactors of disease-causing proteins whose complexes might be stabilized in the disease state to restore normal function .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"biology"
] |
2013
|
Loss of Catalytically Inactive Lipid Phosphatase Myotubularin-related Protein 12 Impairs Myotubularin Stability and Promotes Centronuclear Myopathy in Zebrafish
|
Continuing efforts from large international consortia have made genome-wide epigenomic and transcriptomic annotation data publicly available for a variety of cell and tissue types . However , synthesis of these datasets into effective summary metrics to characterize the functional non-coding genome remains a challenge . Here , we present GenoSkyline-Plus , an extension of our previous work through integration of an expanded set of epigenomic and transcriptomic annotations to produce high-resolution , single tissue annotations . After validating our annotations with a catalog of tissue-specific non-coding elements previously identified in the literature , we apply our method using data from 127 different cell and tissue types to present an atlas of heritability enrichment across 45 different GWAS traits . We show that broader organ system categories ( e . g . immune system ) increase statistical power in identifying biologically relevant tissue types for complex diseases while annotations of individual cell types ( e . g . monocytes or B-cells ) provide deeper insights into disease etiology . Additionally , we use our GenoSkyline-Plus annotations in an in-depth case study of late-onset Alzheimer’s disease ( LOAD ) . Our analyses suggest a strong connection between LOAD heritability and genetic variants contained in regions of the genome functional in monocytes . Furthermore , we show that LOAD shares a similar localization of SNPs to monocyte-functional regions with Parkinson’s disease . Overall , we demonstrate that integrated genome annotations at the single tissue level provide a valuable tool for understanding the etiology of complex human diseases . Our GenoSkyline-Plus annotations are freely available at http://genocanyon . med . yale . edu/GenoSkyline .
Large consortia such as ENCODE [1] and Epigenomics Roadmap Project [2] have generated a rich collection of high-throughput genomic and epigenomic data , providing unprecedented opportunities to delineate functional structures in the human genome . As complex disease research rapidly advances , evidence has emerged that disease-associated variants are enriched in regulatory DNA elements [3 , 4] . Therefore , functional annotation of the non-coding genome is critical for understanding the genetic basis of human complex diseases . Unfortunately , categorizing the complex regulatory machinery of the genome requires integration of diverse types of annotation data as no single annotation captures all types of functional elements [5] . Recently , we have developed GenoSkyline [6] , a principled framework to identify tissue-specific functional regions in the human genome through integrative analysis of various chromatin modifications . In this work , we introduce GenoSkyline-Plus , a comprehensive update of GenoSkyline that incorporates RNA sequencing and DNA methylation data into the framework and extends to 127 integrated annotation tracks covering a spectrum of human tissue and cell types . To demonstrate the ability of GenoSkyline-Plus to systematically provide novel insights into complex disease etiology , we jointly analyzed summary statistics from 45 genome-wide association studies ( GWAS; Ntotal≈3 . 8M ) and identified biologically relevant tissues for a broad spectrum of complex traits . We next performed an in-depth , annotation-driven investigation of Alzheimer’s disease ( AD ) , a neurodegenerative disease characterized by deposition of amyloid-β ( Aβ ) plaques and neurofibrillary tangles in the brain . Late-onset AD ( LOAD ) includes patients with onset after 65 years of age and has a complex mode of inheritance [7] . Around 20 risk-associated genetic loci have been identified in LOAD GWAS [8] . However , our understanding of LOAD’s genetic architecture and disease etiology is still far from complete . Through integrative analysis of GWAS summary data and GenoSkyline-Plus annotations , we identified strong enrichment for LOAD associations in immune cell-related DNA elements , consistent with other data suggesting a crucial role for the immune system in AD etiology [9–11] . Jointly analyzing GWAS summary data for LOAD and Parkinson’s disease ( PD ) , we identified substantial enrichment for pleiotropic associations in the monocyte functional genome . Our findings provide support for the critical involvement of the immune system in the etiology of neurodegenerative diseases , and suggest a previously unsuspected role for an immune-mediated pleiotropic effect between LOAD and PD .
We use our previously established statistical framework to calculate the posterior probability of functionality for each nucleotide in the human genome [12] . Integrating tissue and cell-specific genomic functional data available through Epigenomics Roadmap Project [2] , we make available GenoSkyline-Plus scores for 127 individual tissue annotation tracks ( Methods; S1 Table ) . H3K4me3 and H3K9ac , known markers of open chromatin and active transcription [13] , are shown to have the largest odds ratios of predicting functionality across the genome ( Fig 1A ) . Identifying H3K4me3 and H3K9ac as strong indicators of genomic functionality is a finding consistent with previous studies of gene regulation through chromatin marks [14] . In contrast , H3K9me3 , a well established repressive mark [13] , has a reversed effect on genome functionality . The bimodal pattern of GenoSkyline scores [6] allows us to impose a score cutoff to robustly define the functional genome . Using a cutoff of 0 . 5 , 3% of the genome is considered functional on average across all annotation tracks ( Fig 1B ) . This functionality percentage varies from 1% in pancreatic islet cells to 8% in PMA-I stimulated T-helper cells . Our findings on functionality across all tracks are consistent with previous findings [12]; 34% of the intergenic human genome is predicted to be functional in at least one annotation track ( Fig 1C ) . Additionally , coding regions of the genome are predicted to have much greater proportions of functionality in multiple tissues than intronic and intergenic regions . To assess the ability of GenoSkyline-Plus to capture tissue and cell-specific , non-coding functionality in the human genome , we consider a diverse set of known non-coding regulatory elements studied across the genome . To start , we examined microRNAs ( miRNA ) , which are known to regulate a variety of cellular processes through the translational repression and degradation signaling of transcripts [15] . Recent work by Ludwig et al . profiled miRNA expression in 61 different human tissues and identified miRNAs with functionality unique to single tissues through a tissue specific index [16 , 17] ( TSI; Methods ) . We applied GenoSkyline-Plus scores to miRNA with tissue-specific functionality by calculating the total proportion of nucleotides predicted to be functional in each tissue . We next looked for which annotation tracks are able to predict the highest proportion of functionality for these known functional regions . The best predictors of high functionality for the three tissues with the largest sample sizes ( i . e . brain , liver , and muscle ) are tracks for brain structures , the liver track , and the muscle track , respectively ( Fig 2A ) . We next examined long non-coding RNAs ( lncRNA ) , another non-coding element known for its tissue-specific regulatory action [18] . Using a custom-designed microarray targeting GENCODE lncRNA , Derrien et al . profiled the activity of 9 , 747 lncRNA transcripts [19] . In order to reidentify and validate the set of lncRNA transcripts that are specific to their respective tissues , we calculated the previously described TSI and selected lncRNAs with expression specific to only a few cell types . Physiologically matching tracks show a higher proportion of predicted functionality than unmatched tracks in complex , heterogeneous tissue structures like the midbrain . More functionally uniform tissues , such as the thymus or placenta , show the highest functional proportion in matching annotation tracks ( Fig 2B ) . We also assessed enhancers , non-coding elements that can remotely regulate transcription of an associated promoter elsewhere on the genome with important roles in cell-type specificity [20] . We extracted tissue and cell type-specific enhancer facets identified through the FANTOM5 cap analysis of gene expression ( CAGE ) atlas and positive differential expression when compared against other defined facets [21] . To determine the utility of the large library of immune cells available in the Epigenomics Roadmap Project for which we developed annotation tracks , we focused on enhancer facets with differential CAGE expression in immune cells . While the method by which enhancers are defined to be differential in a facet is liberal ( Methods ) and does not imply facet-specific expression , GenoSkyline-Plus still showed outstanding ability to identify matching cell types . Indeed , matched annotation tracks for T-cells , natural killer cells , and monocytes show consistently higher functional proportions than other , non-matched immune cell annotation tracks ( Fig 2C ) . Finally , we present a case study of the IL17A-IL17F locus control region ( LCR ) in humans , a ~200kb regulatory region surrounding the IL17A gene locus . IL17A encodes the primary secreted cytokine effector molecule IL-17 of T helper 17 ( Th17 ) cells [22] . The LCR has been studied in mouse models and is found to contain many potential human-conserved intergenic regulatory elements that bind transcription factors that are essential for Th17 cell differentiation and effector function [23 , 24] . Experimentally , these conserved noncoding sequences ( CNS ) acquire functionally permissive H3 acetylation marks at much greater magnitudes under Th17-inducing conditions than naïve or combined Th1 and Th2 populations [25] . Comparing annotation tracks for naïve CD4+ T-cells , differentiated Th17 cells , and differentiated Th1/Th2 cell populations , we identified highly Th17-specific functionality in the conserved regions of the human genome corresponding to known murine CNS regions ( Fig 2D and 2E ) . CNS sites and their flanking regions showed substantially higher functional proportion in Th17 cells than in naïve CD4+ T-cells or Th1/Th2 cell subsets . We jointly analyzed three tiers of annotation tracks that respectively represent the overall functional genome , 7 broad tissue clusters , and 66 tissue and cell types ( Methods; S2 Table ) , with summary statistics from 45 GWAS covering a variety of human complex traits ( S3 Table ) . We applied LD score regression [26] to stratify trait heritability by tissue and cell type , and identified a total of 226 significantly enriched annotation tracks for 34 traits after correcting for multiple testing ( S4–S7 Tables ) . In general , GWAS with a large number of significant SNP-level associations showed stronger heritability enrichment in the predicted functional genome ( Fig 3A and 3B ) . Tissue and cell tracks refined the resolution of heritability stratification and provided additional insights into the genetic basis of complex traits ( Fig 3C and 3D ) . The immune annotation track was significantly enriched for 7 immune diseases , namely celiac disease ( CEL ) , Crohn’s disease ( CD ) , ulcerative colitis ( UC ) , primary biliary cirrhosis ( PBC ) , rheumatoid arthritis ( RA ) , systemic lupus erythematosus ( SLE ) , and multiple sclerosis ( MS ) . Using tracks for cell types , we identified several significant enrichments , including monocytes for CD ( p = 2 . 9e-11 ) and B cells for PBC ( p = 2 . 3e-6 ) , RA ( p = 1 . 2e-5 ) , and MS ( p = 2 . 2e-6 ) . Inflammatory bowel diseases showed significant enrichment in the gastrointestinal ( GI ) annotation track ( CD: p = 1 . 4e-4; UC: p = 5 . 6e-5 ) . Another autoimmune disease with a well-established GI component , CEL , also showed nominal enrichment in the GI annotation track ( p = 3 . 7e-4 ) . Several brain annotation tracks were significantly enriched for associations of schizophrenia ( SCZ ) , education years ( EDU ) , and cognitive performance ( IQ ) . Bipolar disorder ( BIP ) , neuroticism ( NEU ) , and chronotype ( CHT ) all showed nominally significant enrichment in the anterior caudate annotation track . Body mass index ( BMI ) and age at menarche ( AAM ) were significantly enriched in multiple brain annotation tracks . Compared to other brain regions , the substantia nigra annotation track showed weaker enrichment for these brain-based traits , which is consistent with its primary function of controlling movement . Hundreds of height-associated loci have been identified in GWAS [27] . Such a highly polygenic genetic architecture is also reflected in our analysis . 59 of 66 tier-3 tissue and cell annotation tracks were significantly enriched for height associations , with breast myoepithelial cell ( p = 6 . 2e-14 ) and osteoblast ( p = 8 . 5e-14 ) being the most significant . Waist-hip ratio ( WHR ) , birth weight ( BW ) , and three blood pressure traits showed significant enrichment in the adipose annotation track . Overall , cardiovascular ( CV ) annotation tracks showed strong enrichment for blood pressure and coronary artery disease ( CAD ) . Interestingly , the aorta annotation track is significantly enriched for pulse pressure ( PP ) but not systolic or diastolic blood pressure ( SBP and DBP ) . CAD and 4 lipid traits , i . e . high and low density lipoprotein ( HDL and LDL ) , total cholesterol ( TC ) , and triglycerides ( TG ) , shared a similar enrichment pattern in liver , adipose , and monocyte annotation tracks , which is consistent with the causal relationship among these traits [28] . Our results demonstrated that annotations with refined specificity could provide insights into disease etiology while broader annotations have greater statistical power . Age-related macular degeneration ( AMD ) was significantly enriched in broadly defined annotation tracks including immune , brain , CV , and GI , despite the non-significant enrichment results using tier-3 annotation tracks . Analyses based on all three tiers of annotations could systematically provide the most interpretable results for most traits . Importantly , we note that greater GWAS sample sizes will effectively increase statistical power in the enrichment analysis while leaving the overall enrichment pattern stable ( S1 Fig ) . Therefore , many more suggestive enrichment results are likely to become significant as GWAS sample sizes grow . Finally , some traits , e . g . type-II diabetes ( T2D ) and age at natural menopause ( AANM ) , showed strong enrichment in the general functional genome but not in specific tissues , suggesting that we may be able to gain a better understanding of these traits when annotation data for tissues or cell types more relevant to these traits are made available . Next , we performed an integrative analysis of stage-I GWAS summary statistics from the International Genomics of Alzheimer’s Project [8] ( IGAP; n = 54 , 162 ) with GenoSkyline-Plus annotations ( Methods ) . SNPs located in the broadly defined immune annotation track , which account for 24 . 4% of the variants in the IGAP data , could explain 98 . 7% of the LOAD heritability estimated using LD score regression ( enrichment = 4 . 0; p = 1 . 5e-4 ) . Somewhat surprisingly , the signal enrichment in DNA elements functional in immune cells was substantially stronger than the enrichment in brain and other tissue types ( Fig 4A ) . To investigate if immune-related DNA elements are also enriched for associations of other neurodegenerative diseases , we analyzed a publicly accessible GWAS summary dataset for PD [29] ( n = 5 , 691; Methods ) . Again , the immune annotation track was the most significantly enriched annotation ( enrichment = 6 . 3; p = 7 . 5e-6 ) , followed by epithelium and CV ( Fig 4A ) . Analysis based on 66 tissue and cell tracks further refined the resolution of our enrichment study . Monocyte ( enrichment = 10 . 9; p = 2 . 0e-5 ) and liver ( enrichment = 16 . 6; p = 4 . 1e-4 ) annotation tracks were significantly enriched for LOAD associations ( Fig 4B ) . In fact , the combined functional regions in monocyte and liver covered 8 . 8% of the SNPs in the IGAP data , but could account for 99 . 6% of the LOAD heritability currently captured in the IGAP stage-I GWAS ( Fig 4C ) . In PD GWAS , signal enrichment in liver was absent , but monocyte-functional regions remained strongly enriched ( enrichment = 16 . 3; p = 8 . 5e-7 ) . Our findings support the critical role of innate immunity in neurodegenerative diseases [10] . Significant enrichment for LOAD associations in liver-specific DNA elements also provides additional support for the possible involvement of cholesterol metabolism in LOAD etiology [30 , 31] . LOAD signal enrichment in liver remained significant after removing the APOE region ( chr19: 45 , 147 , 340–45 , 594 , 595; hg19 ) from the analysis ( S2 Fig ) , suggesting a polygenic architecture in this pathway . Finally , some adaptive immune cells also showed enrichment for AD and PD associations . LOAD signal enrichment in the B cell annotation track was nominally significant , while multiple T cell annotation tracks were significantly enriched for PD associations . These results not only suggest the involvement of adaptive immunity in neurodegenerative diseases , but also hint at distinct mechanisms of such involvement between AD and PD . Finally , for comparison , we applied several other annotations including CADD [32] , GWAVA [33] , and EIGEN [34] to the LOAD GWAS data . GenoCanyon and GenoSkyline annotations for seven tissues were also included in the comparison . Our annotations outperformed these methods , showing stronger fold enrichment and more significant p-values ( S8 Table ) . Our results showed strong enrichment for both AD and PD in the monocyte functional genome . Next , we investigate if the enrichment for both diseases is through shared or distinct genetic components . Recent studies have failed to identify statistically significant genome-wide pleiotropic effects between AD and PD [35] . We instead hypothesize that the same set of immune-related genetic components are involved in both diseases . Therefore , we aim to identify enrichment for pleiotropic effects in the genome localized to regions of monocyte functionality . We first partitioned AD and PD heritability by chromosome . Chromosome-wide heritability showed moderate correlation between the two diseases ( correlation = 0 . 65; Fig 5A ) . When focusing on monocyte functional elements , chromosome-wide heritability showed high concordance between AD and PD ( correlation = 0 . 96; Fig 5B ) . Interestingly , such high concordance cannot be fully explained by chromosome size . In fact , the correlation between chromosome size and per-chromosome heritability estimates is 0 . 56 for AD and 0 . 59 for PD , both lower than the correlation between AD and PD’s per-chromosome heritability estimates , especially in the monocyte functional genome . The percentage of explained LOAD heritability on chromosome 19 is lower than previous estimation [36] due to removal of SNPs with large effects in the APOE region ( Methods ) . Next , to quantify the shared genetics between AD and PD , we identified significant enrichment for pleiotropic effects in monocyte functional regions ( enrichment = 1 . 8; p = 9 . 4e-4 ) using a window-based approach ( Methods ) . To account for potential bias due to the moderate sample overlap between the two GWAS as well as other confounding factors , we applied a permutation-based testing approach ( Methods ) . Enrichment for pleiotropic effects in the monocyte functional genome remained significant ( p = 4 . 6e-3 ) . In addition , these results were robust with the choice of window size . We identified 15 candidate loci for pleiotropic effects ( Methods; S9 Table ) , among which signals at SLC9A9 and AIM1 are the clearest ( Fig 5C and 5D ) . SLC9A9 , whose encoded protein localizes to the late recycling endosomes and plays an important role in maintaining cation homeostasis ( RefSeq , Mar 2012 ) , is associated with multiple pharmacogenomic traits related to neurological diseases , including response to cholinesterase inhibitor in AD [37] , response to interferon beta in MS [38] , response to angiotensin II receptor blockade therapy [39] , and multiple complex diseases including attention-deficit/hyperactivity disorder [40] , autism [41] , and non-alcoholic fatty liver [42] . Gene AIM1 is associated with stroke [43] , human longevity [44] , and immune diseases including RA [45] and SLE [46] . A few candidate loci pointed to clear gene candidates but showed unclear or distinct peaks of association ( S3 Fig ) . These include an inflammatory bowel disease risk gene ANKRD33B [47] . PRUNE2 is a gene associated with response to amphetamine [48] and hippocampal atrophy which is a quantitative trait for AD [49] . HBEGF is associated with AD in APOE ε4- population [50] and involved in Aβ clearance [51] . PROK2 is a gene involved in Aβ-induced neurotoxicity [52] . Additionally , the protein product of AXIN1 negatively affects phosphorylation of tau protein [53] . Other gene candidates include CCDC158 , PRSS16 , and ZNF615 , which are previously identified risk genes for PD , SCZ , and BIP , respectively [54–56] . Some other windows showed complex structures of linkage disequilibrium ( LD ) and contained large association peaks spanning a number of genes ( S4 Fig ) , which include the region near PD risk gene PRSS8 [54] and the HLA region . Interestingly , we also identified the surrounding region of MAPT , a gene that encodes the tau protein which is a critical component of both AD and PD pathologies [50 , 54 , 57 , 58] . Pathway enrichment analysis for genes in 15 pleiotropic candidate loci identified significant enrichment in immune-related pathways staphylococcus aureus infection ( KEGG:05150; p = 1 . 9e-5 ) and systemic lupus erythematosus ( KEGG:05322; p = 3 . 7e-04; Methods ) . Both pathways remained significant after removing two HLA loci from our analysis . Finally , we reprioritize AD risk loci using monocyte and liver annotation tracks . We integrated IGAP stage-I summary statistics with GenoSkyline-Plus using genome-wide association prioritizer ( GenoWAP [59] ) , and ranked all SNPs based on their GenoWAP posterior scores ( Methods ) . Under a posterior cutoff of 0 . 95 , we identified 8 loci that were not reported in the IGAP GWAS meta-analysis using monocyte annotation and 4 loci using the liver annotation track ( S10 Table ) . We then sought replication for SNPs with the highest posterior score at each of these loci using inferred IGAP stage-II z-scores ( Methods ) . After removing shared SNPs between monocyte- and liver-based analyses , 10 SNPs remained in the analysis , 7 of which showed consistent effect directions between the discovery and the replication cohorts ( Fig 6A ) . One SNP was successfully replicated in the inferred IGAP stage-II dataset , i . e . rs4456560 ( p = 0 . 013 ) . SNP rs4456560 is located in SCIMP ( Fig 6B ) , a gene that encodes a lipid tetraspanin-associated transmembrane adaptor protein that is expressed in antigen-presenting cells and localized in the immunological synapse [60] . A moderate replication rate in the IGAP stage-II cohort was expected since we focused on loci that did not reach genome-wide significance in the IGAP meta-analysis and the IGAP stage-II cohort is relatively small ( n = 19 , 884 ) compared to the data in the discovery stage . Furthermore , data from IGAP stage-II cohort are not publicly available and we were limited to the inverse inference approach shown here . It is possible additional loci will replicate when IGAP stage-II summary or individual-level data are made available . However , all identified loci have been linked to AD or relevant phenotypes in the literature . RPN1 was linked to AD through a network-based technique [61] . Association between ECHDC3 and AD risk was established through a joint analysis of AD and lipid traits [62] . Association between DLST and AD has also been previously reported [63] . BZRAP1 and MINK1 were shown to be associated with cognitive function and blood metabolites , respectively [64 , 65] . A pleiotropic effect candidate gene HBEGF showed up again in the SNP reprioritization analysis . Multiple genes in the sorting nexin family have been found to participate in APP metabolism and Aβ generation [66] . Association between SNX1 and AD has also been previously identified using gene-based tests [67] . Finally , during the peer review process of this paper , three new genome-wide significant loci ( i . e . PFDN1/HBEGF , USP6NL/ECHDC3 , and BZRAP1-AS1 ) were reported in a trans-ethnic GWAS meta-analysis for AD [68] , all of which were among our reprioritized list of risk loci . Further , the most significant SNPs at loci PFDN1/HBEGF ( rs11168036 , p = 7 . 1e-9 ) and BZRAP1-AS1 ( rs2632516 , p = 4 . 4e-8 ) matched with our top reprioritized SNPs ( Fig 6A ) .
Increasing evidence suggests that non-coding regulatory DNA elements may be the primary regions harboring risk variants in human complex diseases . In this work , we have substantially expanded our previously established GenoSkyline annotation by incorporating RNA-seq and DNA methylation into its framework , imputing incomplete epigenomic and transcriptomic annotation tracks , and extending it to more than 100 human tissue and cell types . With the help of integrative functional annotations , we identified strong enrichment for LOAD heritability in functional DNA elements related to innate immunity and liver tissue using hypothesis-free tissue-specific enrichment analysis . This enrichment was also found in immune-related DNA elements using PD data . Our analysis also clearly indicated that monocyte functional elements in particular appear to be highly relevant in explaining AD and PD heritability . Of note , we analyzed 45 complex diseases and traits in addition to AD and PD . The substantial enrichment for multiple psychiatric and neurological traits in the brain functional genome shows that the lack of brain enrichment in neurodegeneration is not due to poor quality of brain annotations . Further , the monocytes annotation track was the most significantly enriched for Crohn’s disease among the 45 GWAS , and was not ubiquitously enriched for a large number of traits . Consistent and biologically interpretable enrichment results on a large collection of complex traits demonstrate the effectiveness of our approach and increase the validity of novel findings . It is worth noting that multiple studies have highlighted the role of myeloid cells in the genetic susceptibility of neurodegenerative diseases [11] . Several genes expressed in myeloid cells ( e . g . ABCA7 , CD33 , and TREM2 ) have been identified in GWAS and sequencing-based association studies for AD [8 , 69 , 70] . Further , AD risk alleles identified in GWASs have been shown to enrich for cis-eQTLs in monocytes [9] . In addition , two recent papers identified enrichment for AD heritability in active genome regions in myeloid cells [71 , 72] , which suggested a polygenic genetic architecture for immune-related DNA elements in AD etiology and hinted at a large number of unidentified , immune-related genes for AD . Compared to the aforementioned work , our study utilizes a better set of tissue-specific genome annotations and explicitly accounts for the similarity between different cell types through a multiple regression model . One major limitation in our analysis is lack of data for other potentially AD-relevant cell types such as microglia . Whether our findings correctly reflected the direct involvement of peripheral immune cells in neurodegenerative diseases rather than the detection of epigenomic similarities between monocytes and microglia remains to be carefully investigated in the future . Furthermore , we successfully identified enrichment for shared genetic components between AD and PD in the monocyte functional genome , which hints at a shared neuroinflammation pathway between these two neurodegenerative diseases . We note that several candidate loci with potential pleiotropic effects showed fairly marginal associations with AD and PD , which explains why they have been missed in traditional SNP-based association analysis . Importantly , SNPs in immune-related DNA elements explain a large proportion of AD and PD heritability in total . These results suggest that weak but pervasive associations related with immunity still remain unidentified . Further evaluations of these relationships using GWAS with larger sample sizes may provide insights into the shared biology of these neurodegenerative conditions . Through multi-tier enrichment analyses on 45 GWAS , an in-depth case study of neurodegenerative diseases , and validation of known non-coding tissue-specific regulatory machinery , we have demonstrated the ability of GenoSkyline-Plus to provide unbiased , genome-wide insights into the genetic basis of human complex diseases . The analyzed GWAS represent a variety of human complex diseases and traits , highlighting the effectiveness of our method in different contexts and genetic architecture . However , while our non-coding validation study demonstrated that GenoSkyline-Plus annotations indeed captured tissue-specific activity in a variety of intergenic machinery , there is a need to develop a more statistically robust framework to identify new non-coding elements rather than validate existing ones . Our approach of identifying the functionally active proportion of all elements in aggregate is only able to identify tissue specificity while considering large groups of highly specific non-coding elements . The availability of over 100 different annotation tracks introduces many multiple-testing issues that should be addressed in the case of a statistically sound analysis for tissue-specificity . We have also demonstrated how GenoSkyline-Plus and its explanatory power improve with the addition of more data . Currently , functionality in 28% of exonic regions still remains to be identified . As the quantity and quality of high-throughput epigenomic data continue to grow , GenoSkyline-Plus has the potential to further evolve and provide even more comprehensive annotations of tissue-specific functionality in the human genome . We will update our annotations when data for new tissue and cell types from the Roadmap consortium become available . Finally , several recent papers have introduced novel models to integrate functional annotations in tissue-specific enrichment analysis [73 , 74] . Many models that do not explicitly incorporate functional annotation information have also emerged in transcriptome-wide association studies and other closely-related applications in human genetics research [75–78] . Our annotations , in conjunction with rapidly advancing statistical techniques and steadily increasing sample sizes in genetics studies , may potentially benefit a variety of human genetics applications and promise a bright future for complex disease genetics research .
Chromatin data were extracted from the Epigenomics Roadmap Project’s consolidated reference epigenomes database ( http://egg2 . wustl . edu/roadmap/ ) . Specifically , ChIP-seq peak calls were collected for each epigenetic mark ( H3k4me1 , H3k4me3 , H3k36me3 , H3k27me3 , H3k9me3 , H3k27ac , H3k9ac , and DNase I Hypersensitivity ) in each Roadmap consolidated epigenome where available . Peak calls imputed using ChromImpute [79] were used in place of missing data . Next , peak files were reduced to a per-nucleotide binary encoding of presence or absence of contiguous regions of strong ChIP-seq signal enrichment compared to input ( Poisson p-value threshold of 0 . 01 ) . DNA methylation data were also collected from the Roadmap’s reference epigenomes database . CpG islands were identified in each sample using the CpG Islands Track of the UCSC Genome Browser ( http://genome . ucsc . edu/ ) , and unmethylated islands were those CpG islands with less than 0 . 5 fractionated methylation based on imputed methylation signal tracks in the Roadmap reference epigenomes database . Presence of an unmethylated CpG island was then encoded for each nucleotide as a binary variable . Finally , Roadmap’s RNA-seq data were dichotomized using an rpkm cutoff of 0 . 5 at 25-bp resolution and included in our annotations . We adapt the existing framework established by Lu et al . to a broader set of genomic data [12] . Briefly , given a set of Annotations A and a binary indicator of genomic functionality Z , the joint distribution of A along the genome is assumed to be a mixture of annotations at functional nucleotides and non-functional nucleotides . Assuming that each of the annotations in A is conditionally independent given Z , we factorize the conditional joint density of A given Z as: f ( A|Z=c ) =∏i=110fi ( Ai|Z=c ) , c=0 , 1 ( 1 ) All annotations have been preprocessed into binary classifiers , and the marginal functional likelihood given each individual annotation can be modeled with a Bernoulli distribution fi ( Ai|Z=c ) =picAi ( 1−pic ) 1−Ai , i=1 , … , 10; c=0 , 1 ( 2 ) With an assumed prior probability π of functionality , the parameter pic of each individual annotation can be estimated with the Expectation-Maximization ( EM ) algorithm . The posterior probability of functionality at a nucleotide , known as the GenoSkyline-Plus score , is then: P ( Z=1|A ) =π∏i=110fi ( Ai|Z=1 ) π∏i=110fi ( Ai|Z=1 ) + ( 1−π ) ∏i=110fi ( Ai|Z=0 ) ( 3 ) Giving us with 21 parameters for each annotation track: Θ= ( π , p1 , 0 , p2 , 0 , … , p10 , 0 , p1 , 1 , p2 , 1 , … , p10 , 1 ) ( 4 ) These parameters were estimated using the GWAS Catalog , downloaded from the NHGRI website ( http://www . genome . gov/gwastudies/ ) . 13 , 070 unique SNPs found to be significant in at least one published GWAS were expanded into 1kb bp intervals and formed a sampling covering 12 , 801 , 840 bp of the genome . This sampling method has been shown to be a robust representation of functional and non-functional regions along the genome [6] . Notably , other models have been recently developed to predict functional non-coding SNPs [34] . Quantile-normalized expression values were downloaded for all mature miRNAs profiled in Ludwig et al [17] . Due to inconsistent levels of miRNA specificity in the two donors in this study and to avoid diluting miRNA specificity , we used miRNA data from only body 1 , which had a higher fraction of tissue specific miRNAs . TSI values were calculated as described in the study: TSIj= ∑i=1N ( 1−xj , i ) N−1 ( 5 ) Where N is the total number of tissues measured , xj , i is the expression intensity of tissue i divided by the maximum expression across all tissues for miRNA j . We extract any miRNAs with a TSI score greater than the median value of 0 . 75 to produce a sufficiently large collection of miRNAs with expression highly specific to only a few tissues that we can then attempt re-identify using GenoSkyline-Plus . We next download genomic positions and identify the highest expressed tissue for each TSI-filtered miRNA . miRNA coordinates were extracted from miRbase ( http://mirbase . org/ ) and mapped to hg19 using the UCSC liftover tool ( http://genome . ucsc . edu/ ) . lncRNA data was prepared similarly to miRNA . Expression data of 9 , 747 lncRNA transcripts based on GENCODE v3c annotation across 31 human tissues [19] ( GEO accession: GSE34894 ) was downloaded . As above , the TSI of each lncRNA transcript was calculated , and transcripts with a TSI greater than 0 . 75 were labeled for genomic position and maximally expressed tissue . Pre-defined enhancer differentially expressed cell facets [21] were downloaded from PrESSto database ( http://enhancer . binf . ku . dk/presets/ ) . Andersson et al . define their enhancer sets via bi-directional CAGE expression collected by the FANTOM consortium [80] . Cell facets were manually constructed using hierarchical FANTOM5 cell ontology term mappings to create mutually exclusive and broadly covered histological and functional annotations . Enhancers were considered differentially expressed in a facet using Kruskal-Wallis rank sum test and subsequent pair-wise post-hoc tests to identify enhancers with significantly differential expression between pairs of facets . Based on this method , an enhancer is considered differentially expressed in a facet if it is significantly differentially expressed compared to any other facet and has overall positive standard linear statistics . For each of the three data validation sets , functional specificity is assessed by calculating the per-nucleotide functional proportion of all non-coding elements across a tissue . Functionality is defined by a Genoskyline-Plus score greater than 0 . 5 at that nucleotide . For Roadmap samples with multiple donors ( e . g . skeletal muscle and rectal mucosa ) we took the average GenoSkyline-Plus score at each nucleotide across the samples . For each set of non-coding elements we selected the top three tissues with the largest sample size that had matching annotations in Genoskyline-Plus . For example , we did not calculate scores for enhancers with maximal expressions in human testis because there is no corresponding Roadmap sample in which we would detect tissue-specific functionality . To examine cell-specific functionality of the IL17A LCR in T-cell subsets , we extracted GenoSkyline-Plus scores for each nucleotide along the ~200 kilobase region between the genes PKHD1 and MCM3 [23] . While scores for Th17 and Th1/Th2 subsets ( i . e . ‘CD4+ CD25- IL17+ PMA-Ionomycin stimulated Th17 Primary Cells’ and ‘CD4+ CD25- IL17- PMA-Ionomycin stimulated MACS purified Th Primary Cells’; S1 Table ) were extracted as-is , we took the average score of the two available CD4+ naïve T-cell subsets ( i . e . ‘CD4 Naïve Primary Cells’ and ‘CD4+ CD25- CD45RA+ Naïve Primary Cells’ ) . We identified the analogous human regions of previously identified functional murine CNS regions [25] by taking the top 20 most conserved intergenic sites between mouse and human in the LCR region using the VISTA browser ( http://pipeline . lbl . gov/cgi-bin/gateway2 ) . GenoSkyline-Plus scores in the 20 CNS sites and their 200-bp flanking regions were compared across different cell types . Summary statistics for 45 GWAS are publicly accessible . Details for these studies are summarized in S3 Table . IGAP is a large two-stage study based upon genome-wide association studies ( GWAS ) on individuals of European ancestry . In stage-I , IGAP used genotyped and imputed data on 7 , 055 , 881 SNPs to meta-analyze four previously-published GWAS datasets consisting of 17 , 008 Alzheimer's disease cases and 37 , 154 controls ( The European Alzheimer's disease Initiative–EADI , the Alzheimer Disease Genetics Consortium–ADGC , The Cohorts for Heart and Aging Research in Genomic Epidemiology consortium–CHARGE , and The Genetic and Environmental Risk in AD consortium–GERAD ) . In stage-II , 11 , 632 SNPs were genotyped and tested for association in an independent set of 8 , 572 AD cases and 11 , 312 controls . Finally , a meta-analysis was performed combining results from stages I and II . IGAP stage-I GWAS summary data is publicly accessible from IGAP consortium website ( http://web . pasteur-lille . fr/en/recherche/u744/igap/igap_download . php ) . GWAS summary statistics for PD was acquired from dbGap ( accession: pha002868 . 1 ) . Details for AD and PD studies have been previously reported [8 , 29] . Heritability stratification and enrichment analyses were performed using LD score regression implemented in the LDSC software ( https://github . com/bulik/ldsc/ ) . Annotation-stratified LD scores were estimated using dichotomized annotations , 1000 Genomes ( 1KG ) samples with European ancestry [81] , and a default 1-centiMorgan window . Enrichment was defined as the ratio between the percentage of heritability explained by variants in each annotated category and the percentage of variants covered by that category . A resampling-based approach was used to assess standard error estimates [26] . Three tiers of annotations of different resolutions were used in enrichment analyses: The smoothing strategy for GenoCanyon improves its ability to identify general functionality in the human genome [59] . GenoSkyline-Plus and smoothed GenoCanyon annotations were dichotomized using a cutoff of 0 . 5 . Such dichotomization is robust to the cutoff choice due to the bimodal nature of annotation scores [6] . We selected 66 annotation tracks in the tier-3 analysis by removing all the fetal and embryonic cells , and taking the union of different Roadmap epigenomes for the same cell type ( S2 Table ) . The 53 baseline annotations of LD score regression were always included in the model across all analyses as suggested in the LDSC user manual . Smoothed GenoCanyon annotation track was also included in tier-2 and tier-3 analyses to account for unobserved tissue and cell types . Of note , the proposed multiple regression model explicitly takes the overlapped functional regions across biologically-related cell types into account . Further , the linear mixed-effects model in LDSC does not assume linkage equilibrium , and therefore LD will most likely not introduce bias into heritability estimation and enrichment calculation . We removed the MHC region from our analysis due to its unique LD patterns . A slightly different strategy was adopted when comparing the performance of different computation annotation tools . To make fair comparison , we dichotomized all annotation tracks using each score’s top 90% quantile calculated from SNPs with minor allele count greater than five in 1000 Genomes samples with European ancestry . We then followed the suggested protocol of LDSC and kept baseline annotations in the model while adding each annotation track one at a time . We calculated chromosome-by-chromosome heritability percentage through summing up and normalizing per-SNP heritability estimated using LD score regression and tier-3 annotation tracks . Of note , since only GWAS summary statistics were used as the input , popular heritability estimation tools such as GCTA [82] could not be applied . The sums over complete chromosomes are compared with the sums over monocyte functional regions only . Notably , LDSC is conceptually different from some other tools ( e . g . GCTA [82] ) in its estimation of trait heritability . GCTA estimates the proportion of phenotypic variability that can be explained by SNPs in the GWAS dataset while LDSC aims to estimate the proportion of phenotypic variability explained by all the SNPs in samples from the 1KG Project . In practice , LDSC only uses HAPMAP SNPs to fit the LD score regression model and assumes that HAPMAP SNPs are sufficient for tagging all 1KG SNPs through LD [26] . Additionally , LDSC applies a few stringent SNP filtering steps for quality control reasons , e . g . removing SNPs with very large effect sizes ( i . e . χ2 > 80 ) , which leads to the removal of some SNPs in the APOE region in our analysis . Finally , we note that a recent method may potentially improve the heritability estimates based on LDSC [75] . To evaluate enrichment of pleiotropic sites in the monocyte functional genome , we partition the genome into windows with length of 1M bases . Sex chromosomes and windows without SNPs are removed in our datasets . For each disease ( i . e . AD and PD ) , we label a window 1 if the following criteria are met . Otherwise , the window is labeled 0 . This labeling results in two binary vectors , one for each disease . A window marked as 1 for both AD and PD is a window of interest that suggests a possible association in monocytes-related DNA for both diseases in that region . We use a hypergeometric test to assess if such a pattern of local association appears more often than by chance . Windows marked as 1 for both diseases are subsequently curated to identify the association peaks that potentially have pleiotropic effects for AD and PD . There is a moderate overlap of control samples between IGAP AD GWAS and the PD GWAS ( KORA controls , N~480 ) . To account for the bias introduced by sample overlap and other confounding factors , we designed a permutation-based approach . In each permutation step , we shuffle the annotation status while keeping the total proportion of annotated regions , and then pick out windows that meet condition 2 . We calculate the p-value through comparing the observed number of windows that meet conditions 1 and 2 for both diseases with the empirical distribution acquired in permutations . Of note , we also applied this approach using a window size of 500K bases . Results in all related tests remained similar . We briefly describe the SNP reprioritization approach implemented in the GenoWAP software available on our server ( http://genocanyon . med . yale . edu/GenoSkyline ) . First , we identify three disjoint cases for SNPs in a given GWAS dataset . A useful metric for prioritizing SNPs is the conditional probability that the SNP is classified under case-1 given its p-value in the GWAS study , i . e . P ( ZD = 1 , ZT = 1 | p ) . We can denote this probability using Bayes formula as follows: P ( ZD=1 , ZT=1 | p ) = P ( Case 1 | p ) = f ( p|Case 1 ) ×P ( Case 1 ) ∑k=13f ( p|Case k ) ×P ( Case k ) ( 7 ) First , P ( Case 3 ) = 1 − P ( ZT = 1 ) can be directly identified using GenoSkyline-Plus scores . We partition all the SNPs into two subgroups based on a mean GenoSkyline-Plus score threshold of 0 . 1 . Notably , these probabilities are not sensitive to changing threshold [6] . In this way , we can directly estimate f ( p|Case 3 ) = f ( p|ZT = 0 ) by applying a histogram approach on the SNP subgroup with low GenoSkyline-Plus scores . Next , we assume that SNPs that are functional in a tissue but not relevant to the phenotype will have the same p-value distribution to all other SNPs that are not relevant to the phenotype , which in turn behave similarly to SNPs that are not functional at all . We have previously demonstrated that this assumption is backed by empirical evidence [6] . More formally , this relationship is denoted as follows: f ( p|Case 2 ) =f ( p|ZD=0 , ZT=1 ) =f ( p|ZD=0 ) =f ( p|Z=0 ) ( 8 ) We estimate the distribution f ( p|Z = 0 ) by using a similar approach to estimating f ( p|ZT = 0 ) , but partitioning SNPs using the general functionality GenoCanyon score instead of tissue-specific GenoSkyline-Plus score . Finally , all remaining terms in Formula 6 can be estimated using the EM algorithm . The p-value distribution of the subset of SNPs located in tissue-specific functional regions ( i . e . ZT = 1 ) is the following mixture: f ( p|ZT=1 ) =P ( ZD=1|ZT=1 ) ×f ( p|Case 1 ) +P ( ZD=0|ZT=1 ) ×f ( p|Case 2 ) ( 9 ) Density function f ( p|Case 2 ) has been estimated in Formula ( 8 ) and f ( p|Case 2 ) is assumed to follow a beta distribution , which guarantees a closed-form expression in the EM algorithm . Notably , the APOE region was removed in the SNP reprioritization analysis for LOAD . Summary statistics from both IGAP stage-I GWAS and stage-I+II meta-analysis are publicly available ( http://web . pasteur-lille . fr/en/recherche/u744/igap/igap_download . php ) . We inferred z-scores from IGAP stage-II replication cohort using the following formula . In this formula , Z1 and Z1+2 indicate z-scores from the stage-I GWAS and the combined meta-analysis , respectively . Ni indicates the sample size from the ith stage . This formula was derived from the sample size based meta-analysis model , an approach known to be asymptotically equivalent to inverse variance based meta-analysis [83] . GenoSkyline-Plus annotation tracks , tiers 1–3 LD score files , and scripts for generating GenoSkyline-Plus scores are freely available on the GenoSkyline server ( http://genocanyon . med . yale . edu/GenoSkyline ) . All annotation tracks can be visualized using UCSC genome browser . Web server g:Profiler was used to perform pathway enrichment analysis [84] . The g:SCS threshold implemented in g:Profiler was applied to account for multiple testing . Locus plots were generated using LocusZoom [85] . Gene plots were generated using R package “Gviz” .
|
After years of community efforts , many experimental and computational approaches have been developed and applied for functional annotation of the human genome , yet proper annotation still remains challenging , especially in non-coding regions . As complex disease research rapidly advances , increasing evidence suggests that non-coding regulatory DNA elements may be the primary regions harboring risk variants in human complex diseases . In this paper , we introduce GenoSkyline-Plus , a principled annotation framework to identify tissue and cell type-specific functional regions in the human genome through integration of diverse high-throughput epigenomic and transcriptomic data . Through validation of known non-coding tissue-specific regulatory regions , enrichment analyses on 45 complex traits , and an in-depth case study of neurodegenerative diseases , we demonstrate the ability of GenoSkyline-Plus to accurately identify tissue-specific functionality in the human genome and provide unbiased , genome-wide insights into the genetic basis of human complex diseases .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"blood",
"cells",
"genome-wide",
"association",
"studies",
"medicine",
"and",
"health",
"sciences",
"functional",
"genomics",
"immune",
"cells",
"neurodegenerative",
"diseases",
"immunology",
"human",
"genomics",
"genome",
"analysis",
"genome",
"annotation",
"alzheimer",
"disease",
"white",
"blood",
"cells",
"animal",
"cells",
"dementia",
"mental",
"health",
"and",
"psychiatry",
"genetic",
"loci",
"cell",
"biology",
"monocytes",
"neurology",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"genomics",
"computational",
"biology",
"human",
"genetics"
] |
2017
|
Systematic tissue-specific functional annotation of the human genome highlights immune-related DNA elements for late-onset Alzheimer’s disease
|
All-trans retinoic acid ( ATRA ) is instrumental to male germ cell differentiation , but its mechanism of action remains elusive . To address this question , we have analyzed the phenotypes of mice lacking , in spermatogonia , all rexinoid receptors ( RXRA , RXRB and RXRG ) or all ATRA receptors ( RARA , RARB and RARG ) . We demonstrate that the combined ablation of RXRA and RXRB in spermatogonia recapitulates the set of defects observed both upon ablation of RAR in spermatogonia . We also show that ATRA activates RAR and RXR bound to a conserved regulatory region to increase expression of the SALL4A transcription factor in spermatogonia . Our results reveal that this major pluripotency gene is a target of ATRA signaling and that RAR/RXR heterodimers are the functional units driving its expression in spermatogonia . They add to the mechanisms through which ATRA promote expression of the KIT tyrosine kinase receptor to trigger a critical step in spermatogonia differentiation . Importantly , they indicate also that meiosis eventually occurs in the absence of a RAR/RXR pathway within germ cells and suggest that instructing this process is either ATRA-independent or requires an ATRA signal originating from Sertoli cells .
Spermatogenesis is a tightly regulated , cyclical , cell differentiation process , taking place in the seminiferous epithelium of the testis and yielding mature spermatozoa from stem cells . Spermatogonia in the single cell state , known as A single ( As ) spermatogonia , have traditionally be considered as the main spermatogonia stem cells in the mouse . Upon division , As spermatogonia give rise either to two new single cells or to a pair of daughter cells called A paired ( Apr ) spermatogonia that do not complete cytokinesis and remain connected through an intercellular bridge . The Apr spermatogonia divide further to form syncytial chains of 4 to 16 A aligned ( Aal ) spermatogonia [1] . Collectively , As , Apr and Aal ( referred to as “undifferentiated spermatogonia” ) are present throughout the seminiferous epithelial cycle and retain stem cell properties . Subsequently , Aal cells differentiate without mitotic division into A1 spermatogonia . Five cell divisions follow A1 formation , forming successively A2 , A3 A4 , In ( intermediate ) and B spermatogonia . Collectively , A1 to B spermatogonia ( referred to as “differentiating spermatogonia” ) express the prototypic marker KIT [2 , 3] and differentiate at given stages of the seminiferous epithelium cycle , each step of differentiation being associated with a mitotic division [4] . In rodent , all-trans retinoic acid ( ATRA ) , the biologically active form of vitamin A ( retinol ) is instrumental to spermatogonia differentiation as assessed from vitamin A deficiency studies . In mice fed a vitamin A-deficient ( VAD ) diet from weaning onwards , all spermatogonia progressively arrest at the Aal-A1 transition , yielding seminiferous tubules that contain only Aal spermatogonia and Sertoli cells . Systemic administration of ATRA to VAD mice reinitiates spermatogenesis from mitotically-arrested Aal spermatogonia , resulting in their massive differentiation into spermatogonia expressing KIT , the marker of the Aal-A1 transition [2] , and resuming their proliferation/differentiation [5 , 6] . The molecular mechanism through which ATRA controls Kit expression is however not yet fully elucidated . Characterizing this mechanism is important not only in the field of reproduction , but also for a better understanding of the biology of testicular germ cell tumors as KIT is also frequently deregulated in seminomas [7] . In cells , ATRA binds to and activates nuclear receptors ( RARA , RARB and RARG ) , which are ligand-dependent transcriptional regulators . They usually function in the form of heterodimers with rexinoid receptors ( RXRA , RXRB and RXRG ) to control expression of ATRA-target genes through binding to specific sites located in genomic regulatory regions and called retinoic acid response elements ( RARE ) [8] . In the adult mouse testis , RARG cell-autonomously transduces an ATRA signal required for spermatogonia differentiation . Accordingly , the testes abnormalities observed upon deletion of Rarg either in the whole organism or specifically in spermatogonia in sexually mature males are similar to those present in VAD males [9] . As to RXR isotypes , the situation is contrasted . Our in situ hybridization ( ISH ) analyses failed to detect any of them in spermatogonia in the normal mouse testis [10] , while another study evidenced RXRA in spermatogonia by immunohistochemistry ( IHC ) [11] . Thus , either RXR is absent and therefore dispensable for RAR functioning in spermatogonia , as it is the case in Sertoli cells [12] , or RXRA is required but its expression level in spermatogonia too low to be detected by ISH . To discriminate between these two possibilities , we have generated mice lacking all RXR isotypes specifically in spermatogonia from PN5 onwards and analyzed their phenotype . We demonstrate that ablation of all Rxr genes arrests differentiation of some spermatogonia at the Aal-A1 transition and recapitulates the full set of defects characteristic of the vitamin A deficiency-induced testis degeneration . We further show that efficient ablation of the 3 Rar genes in spermatogonia using the same genetic approach resembles ablation of the 3 Rxr genes . Importantly , some A1 spermatogonia still differentiate in these mutants , indicating the existence of a mechanism allowing the Aal to A1 transition independently of RAR/RXR in germ cells . Along these lines , both meiotic and post-meiotic cells devoid of RAR or RXR are also produced , in contrast to the situation when ATRA synthesis is impaired [13] . We propose that paracrine signals emanating from and transduced in Sertoli cells by ATRA-activated RARA stimulate some Aal spermatogonia to become A1 and trigger entry into meiosis . We finally provide evidence that RXR and RAR bind to the same Sall4 regulatory region to control ATRA-dependent expression of SALL4A in the RAR/RXR-dependent spermatogonia . As SALL4A is known to impair ZBTB16-mediated Kit repression [14] , our study provides novel insights into the molecular mechanism by which ATRA could control KIT expression , and thereby the differentiation of Aal into A1 spermatogonia in vivo .
The different generations of germ cells form cellular associations of fixed composition called epithelial stages . In control testes only the twelve normal epithelial stages ( I–XII ) [17] were identified ( Fig 1A ) . In contrast , analysis of 12-week-old Rxra;b;gSpg–/– mutant testes ( n = 5 ) , revealed that , aside from normal epithelial stages ( Fig 1C ) , 36 . 3 ± 9 . 6% of the tubule sections exhibited a degenerated seminiferous epithelium ( Fig 1B ) either lacking a large proportion of germ cells ( T3 ) or containing only spermatogonia and Sertoli cells ( T4 ) . In addition , 17 . 8 ± 3 . 4% of the tubule sections lacked , around their entire circumference , either one or two generations of germ cells , yielding abnormal variants of the epithelial stages ( T2 ) . The missing germ cell layers included: preleptotene spermatocytes ( Fig 1E and 1G ) , pachytene spermatocytes ( Fig 1D and 1H ) , and/or round spermatids ( Fig 1F and 1H ) . Thus , germ cell differentiation appeared altered in Rxra;b;gSpg–/– mutants . Analysis of other combinations of compound mutants at the age of 12 weeks revealed that the pathological phenotype was generated solely upon the simultaneous ablation of Rxra and Rxrb ( Fig 2 ) . This indicates that both RXRA and RXRB exert redundant functions in spermatogonia , while RXRG is dispensable . One year-old controls ( n = 4 ) displayed only normal germ cell associations , whereas mutants ( n = 4 ) displayed tubule sections containing only Sertoli cells and spermatogonia ( Fig 1I and 1J ) . The latter expressed molecular markers of undifferentiated spermatogonia such as Gfra1 and Zbtb16 [3 , 18] , but not of differentiating spermatogonia such as Kit and Stra8 [2 , 19] ( Fig 3A–3H ) . The histological defects displayed by Rxra;b;gSpg–/– mutants appeared to be much more severe than those observed when Rarg and Rara genes are deleted using the Tg ( Neurog3-cre ) 24Syos transgene [9] . This raised the possibility that RXR isotypes could be instrumental to some aspects of spermatogonia differentiation , independently of RARG and RARA . To test for this hypothesis , we analyzed the outcome of deleting Rar genes in spermatogonia by means of the same Tg ( Stra8-cre ) 1Reb transgene . Accordingly , mice carrying loxP-flanked alleles of Rara , Rarb and Rarg were crossed with Tg ( Stra8-cre ) 1Reb mice to generate Rara;b;gSpg–/– mutants and their controls . In 12 week-old Rara;b;gSpg–/– mutants ( n = 5 ) , 24 . 5 ± 10 . 1% of the seminiferous tubule sections were abnormal , amongst which 11 . 5 ± 5% , identified as variants of the normal epithelial stages , lacked one or two generations of germ cells and 12 . 9 ± 7 . 3% exhibited a seminiferous epithelium either with a complete disorganization of the germ cell layers or with spermatogonia and Sertoli cells only ( S2 Fig ) . In one-year-old mutants ( n = 3 ) , the seminiferous epithelium consisted only in Sertoli cells and spermatogonia , which expressed genes that are typical of undifferentiated spermatogonia ( i . e . , Gfra1 and Zbtb16 ) , but not of differentiating spermatogonia ( i . e . , Kit and Stra8 ) ( Fig 3I–3L ) . Altogether , these data indicate that age-matched Rara;b;gSpg–/– and Rxra;b;gSpg–/– mutants display similar , if not identical , phenotypes , including a slow and progressive loss of differentiating germ cells and the presence of spermatogonia blocked at an undifferentiated , Aal , stage ( i . e . , ZBTB16-positive , KIT-negative [2 , 3] ) in aged mutants , both of which are features of the VAD testis [20] . To further document the similarities between the phenotypes induced by Rxr and Rar loss-of-functions , we examined the effect of Rxr ablation on germ cell apoptosis . Terminal deoxynucleotidyl-transferase dUTP nick end-labeling ( TUNEL ) assays indicated that apoptosis of preleptotene spermatocytes was not increased in testes of 8 week-old Rxra;b;gSpg–/– mutants , relative to age-matched controls ( Fig 4A and 4B ) . Actually , we did not detect a single TUNEL-positive preleptotene spermatocyte in controls and in Rxra;b;gSpg–/– mutants ( n = 3 males for each genotype; n > 200 preleptotene spermatocytes per testis ) . Therefore , similarly to the situation in mice lacking Rara and Rarg in spermatogonia [9] , cell-death cannot account for the missing germ cell layers observed in Rxra;b;gSpg–/– mutant testes . We next examined the effect of Rxr ablation on the pace of preleptotene spermatocyte differentiation , because any delay or an arrest of this process may lead: ( i ) to the disappearance of pachytene spermatocytes through their normal differentiation into round , step 7 , spermatids after one cycle of the seminiferous epithelium ( i . e . , 8 . 6 days ) , then ( ii ) to the disappearance of step 7 spermatids through their normal transformation into mature , step 16 , spermatids after completion of a second cycle . Thus , we evaluated the duration of meiotic phase of spermatogenesis after 5-bromo-2'-deoxyuridine ( BrdU ) incorporation into S-phase nuclei . In adult testis , BrdU is mainly incorporated into B spermatogonia and preleptotene spermatocytes [20] . We fate-mapped the BrdU-labeled descendants of these cells 9 and 17 days after injection of the tracer . At the latter time-point , the most advanced , BrdU-positive , cell-type was step 7 spermatids in both control and Rxra;b;gSpg–/– testes ( n = 3 for each genotype; 8 week-old ) and there was no retained labeling in any spermatocyte ( Fig 4C and 4D ) . Thus , similarly to the situation in mice lacking Rara and Rarg in spermatogonia [9] , the duration of meiosis is not altered in Rxra;b;gSpg–/– mutant testes . In this context , it seems logical to assume that ablation of Rxr genes induces germ cell depletion solely through altering the spermatogonia proliferation/differentiation process . Accordingly , BrdU was detected in Rxra;b;gSpg–/– testes , 9 days after its incorporation , in cells displaying histological features of spermatogonia ( Fig 4E and 4F ) . BrdU was never detected in spermatogonia of control testes at this time-point after injection because its amount becomes diminished by half in each daughter cell upon cell-division , yielding a progressive decrease of the signal over time and its absence 9 days after BrdU incorporation . This observation indicates that some of the spermatogonia that had incorporated BrdU in the Rxra;b;gSpg–/– testes did not divide further or divided more slowly than in control testes , as it is the case for spermatogonia lacking Rar [9] and in VAD testis [20] . Our study shows therefore that the outcomes of ablating RAR or RXR in spermatogonia are identical at the histological level . Given the central role assigned to ATRA in spermatogonia differentiation and in male meiosis [13 , 21] , it was surprising that differentiation of only some spermatogonia was impaired , and that meiosis always proceeded normally in the absence of either Rar or Rxr . To exclude the possibility that Cre-mediated excision was mosaic , thereby resulting in the absence of RAR or RXR in some , but not all , differentiating germ cells , we analyzed their expression in Rara;b;gSpg–/– and Rxra;b;gSpg–/– testes . RT-qPCR analysis of whole testis RNA showed that the amount of Rarg transcripts was markedly reduced ( 4-fold ) in Rara;b;gSpg–/– mutants , as early as PN5 ( S3 Fig ) . Consistent with this finding , IHC analyses indicated a total absence of RARG in germ cells of Rara;b;gSpg–/– mutants , at PN5 and at PN60 ( S3 Fig ) . In fact 100% of spermatogonia and of preleptotene spermatocytes were devoid of RARG in Rara;b;gSpg–/– mutants . Efficient loss of RARA and RARB in these mutants can be assessed neither by RT-qPCR because they are not , or only weakly , expressed in spermatogonia nor by IHC analyses because reliable antibodies are not available [9 , 10] . However , efficient ablation of loxP-flanked Rara and Rarb alleles was assessed at the genomic level , using FACS-purified germ cell populations [22] . PCR analysis demonstrated that excised ( null , L– ) , but not conditional ( L2 ) alleles , were detected in genomic DNA isolated from spermatogonia , spermatocytes and spermatids of PN60 mice bearing the Tg ( Stra8-cre ) 1Reb transgene ( S3 Fig ) . Together , these data indicate that ablation of all 3 Rar genes was efficient in all germ cells , as early as PN5 . As the outcome of ablating all RAR in spermatogonia appears very close to that induced upon Rarg knockout ( Rarg–/– mutants ) [9] , our data suggest that RARG is the major functional RAR isotype in spermatogonia . Similar results were obtained in Rxra;b;gSpg–/– mutants , with Rxra transcript amounts markedly reduced ( 4- to 5-fold ) as early as PN5 ( S3 Fig ) . Assuming that the excision of Rar and Rxr genes was complete from PN5 onwards , the impact of RAR or RXR loss-of-functions during the pubertal development of the testis was evaluated at PN20 , i . e . , when the first post-meiotic cells appear . At this developmental stage , control , Rxra;b;gSpg–/– and Rara;b;gSpg–/– mutant testes ( n = 4 for each genotype ) were indistinguishable: in both situations , late pachytene and diplotene spermatocytes represented the most advanced germ cell in the vast majority of tubule sections ( Fig 5A ) . These results indicate that the spermatocytes present at PN20 in the mutants testes , appeared in due time , likely because the spermatogonia from which they derived started to differentiate before PN3 , at a time when Rar or Rxr genes were not yet knocked out . More importantly , they also indicate that all preleptotene spermatocytes initiated meiosis normally ( around PN8 ) , at a time when they were devoid of RAR or RXR since ablation was obvious from PN5 in their precursors ( see above ) . Analyzing the seminiferous epithelium later during pubertal development revealed the occurrence of abnormal cellular associations at PN25: few tubule sections in mutant testes displayed spermatogonia associated with round spermatids but without the intervening layers of preleptotene and pachytene spermatocytes ( Fig 5B and 5C ) . This observation confirms the initial wave of A1 spermatogonia differentiation was not affected ( yielding step 7 spermatids at PN25 ) , and suggests the second wave was arrested ( or delayed ) in few tubules at some point before meiosis , leading to the absence of spermatocytes at PN25 . However , the presence of normal cellular associations in the majority of tubule sections indicates that both A1 spermatogonia differentiation and meiosis occurred , despite absence of RAR or RXR in germ cells ( see above ) . In keeping with this , KIT-positive A1 spermatogonia were found at stages VII-VIII of the seminiferous epithelium cycle in Rara;b;gSpg–/– mutants at PN60 , similarly to the situation in control mice ( Fig 5E and 5F ) . The reason why ablation of RAR or RXR in germ cells affects only a fraction of A1 spermatogonia is unclear . In some cases the A1 transition appears to take place in due time , yielding normal cell associations , while in other instances no A1 are formed and Aal spermatogonia have to wait one ( or several ) epithelial cycles to become A1 , yielding seminiferous tubule segments with missing generations of spermatocytes and spermatids . Moreover , spermatogonia do not transition at random because , as a result of stage-dependent cell divisions [4] , extensive rows of differentiating , KIT-positive , RARG-negative , spermatogonia were observed at stages I-VI of the seminiferous epithelium cycle in Rara;b;gSpg–/– mutants , as it is the case in control males ( Fig 5G–5I ) . An unknown signal , distinct from ATRA , may operate as a backup only in the context of Rara;b;gSpg–/– and Rxra;b;gSpg–/– mutants to promote spermatogonia differentiation . Alternatively , two populations of Aal spermatogonia may normally exist in the seminiferous epithelium at stages VII-VIII , one requiring RAR/RXR to transition , the other one being committed to become A1 , independently of RAR/RXR heterodimers . As A1 spermatogonia were less and less often generated with aging and no longer observed in old Rara;b;gSpg–/– and Rxra;b;gSpg–/– mutants , the second population may become depleted with time resulting in a complete arrest of spermatogenesis . Given the pivotal role of ATRA and the precise timing in spermatogonia differentiation at stage VII-VIII of the seminiferous epithelium cycle [4] , instructing the transition in this second population may also require ATRA . We propose that an ATRA signal transduced by RARA in Sertoli cells stimulates some Aal spermatogonia to become A1 even though no RAR/RXR pathway is functional within these latter cells . In keeping with this proposal , it worth noting that the seminiferous epithelium of mice lacking both RARA in Sertoli cells and RARG in spermatogonia ( RaraSer–/–;Rarg–/– mutants ) was found to consist only in Sertoli cells and undifferentiated spermatogonia ( Fig 5D ) , as it is the case when ATRA synthesis is specifically abolished in Sertoli cells [13] . More surprisingly , following the Aal to A1 transition , germ cell differentiation progressed at a normal pace despite the lack of RAR or RXR inside these cells . This indicates that initiation ( and progression ) of meiosis can proceed even in the absence of a functional ATRA signaling pathway in spermatocytes . If one considers that ATRA is as a mandatory meiosis-inducing substance in vertebrates [23] , then our finding necessarily implies that the ATRA-dependent pathway instructing preleptotene spermatocytes is not autocrine in nature , as previously proposed [13] , but instead operates in Sertoli cells . In this context , ATRA would control the synthesis by Sertoli cells of yet unknown , intermediate , secreted factor ( s ) acting on spermatocytes to trigger meiosis . Alternatively , the possibility exists that meiotic initiation does not require ATRA in male germ cells , as it was shown to be the case in female germ cells [24] . Discriminating between these two possibilities awaits further investigations . Regardless of the scenario , our findings indicate that expression of Stra8 does not require a RAR/RXR-dependent signaling pathway in preleptotene spermatocytes , even though this receptor heterodimer can efficiently bind to RARE in the Stra8 promoter [13 , 24] . The similarities between Rar and Rxr gene ablations indicate that RAR and RXR exert convergent functions in spermatogonia , and support the possibility that ATRA signaling in these cells involves RAR/RXR heterodimers . They also indicate that RXR in spermatogonia are unlikely to play a role other than controlling differentiation in conjunction with RAR . To test whether RAR and RXR were actually recruited to an endogenous gene promoter in spermatogonia in vivo , we performed immunoprecipitation ( IP ) of these nuclear receptors using chromatin from PN5 wild-type testes , followed by qPCR analysis of the recovered DNA fragments , and assessed binding to Stra8 , which is proposed as a RAR target-gene in spermatogonia [21] . Both anti-RAR and anti-RXR antibodies were able to precipitate , with similar efficiencies , the DNA sequences containing the RAR binding sites of the Stra8 promoter ( S4 Fig ) . These data further support the notion that RAR/RXR heterodimers can be the functional units transducing the ATRA-signal in spermatogonia . However , although STRA8 promotes spermatogonia differentiation , it is not strictly required for this process [19 , 25 , 26] and its expression does not appears to be dependent upon RAR/RXR-signaling ( see above ) . Thus , effectors acting downstream of ATRA and distinct from STRA8 likely account for the Aal to A1 spermatogonia transition . To gain insights into the genetic cascade controlled by RAR/RXR heterodimers and aside from STRA8 , we set up an experiment aimed at identifying ATRA-controlled genes in spermatogonia . To this purpose , we used Aldh1a1-3Ser−/− mutants as a model , in which all retinaldehyde dehydrogenase activity is ablated in Sertoli cells . These mutants were chosen because ( i ) their spermatogonia differentiation is blocked at the Aal stage , ( ii ) Aal spermatogonia express RARG [9] and ( iii ) Aal spermatogonia are poised to differentiate into A1 spermatogonia upon activation of ATRA signaling [13] . We treated organotypic cultures of Aldh1a1-3Ser−/− testes with the RARG-selective agonist BMS961 ( n = 5 ) or with its vehicle ( n = 5 ) for 6 hours and extracted mRNA . Microarray expression profiling identified only a few transcripts that were differentially expressed upon activation of RARG , amongst which Sall4 . This gene encodes two isoforms named SALL4A and SALL4B [27] . They are zinc-finger transcription factors , which participate in regulatory networks and are critical for cell fate decisions and lineage specification [28 , 29] . In the mouse testis , their expression is restricted to spermatogonia [30 , 31] and mice deficient for Sall4 in these cell-type display testis defects that resemble those observed in Rxra;b;gSpg–/– and Rara;b;gSpg–/– mutants , namely loss of differentiating , KIT-positive , spermatogonia and of meiotic cells [14] . Thus Sall4 gene appears particularly relevant to ATRA-induced spermatogonia differentiation . We confirmed by RT-qPCR that Sall4a mRNA steady state level was increased upon BMS961 administration in Aldh1a1-3Ser−/− testes , without the need for intermediate protein synthesis as this increase occurred in the presence of cycloheximide ( Fig 6A ) . Western-blot analysis of protein extracts from Aldh1a1-3Ser−/− testes revealed that SALL4A protein level was increased by BMS961-activated RARG ( Fig 6B , compare lane 1 to 2 ) ; this increase was prevented in mice additionally carrying a Rarg knock-out ( Fig 6B , compare lane 3 to 4 ) and was not observed in BMS961-treated Rara;b;gSpg–/– mutants ( Fig 6B , compare lane 5 to 6 ) . In addition , Sall4a mRNA levels were significantly decreased in whole testis of Rara;b;gSpg–/– and Rxra;b;gSpg–/– mutants at PN60 , while Sall4b and Zbtb16 mRNA levels were unchanged ( Fig 6C ) . The finding that Sall4b mRNA level was not altered is in keeping with previous reports showing that SALL4B is expressed at a constant level in spermatogonia [31 , 32] . Altogether our results indicate that ( i ) Sall4a expression is decreased in testes of mice lacking RAR or RXR in spermatogonia ( Rara;b;gSpg–/– and Rxra;b;gSpg–/– testes ) ; ( ii ) SALL4A is detected at a low level in the seminiferous epithelium of mice deficient in ATRA ( Aldh1a1-3Ser−/− testes ) , but at a high level when RARG is activated by BMS961 in these mice ( Aldh1a1-3Ser−/− testes treated with BMS961 ) , except when RARG is lacking ( Aldh1a1-3Ser−/−;Rarg−/− testes , treated with BMS961 ) ; and ( iii ) SALL4A is not detected in testes of adult mice lacking RAR in spermatogonia even in the presence of the RARG agonist ( Rara;b;gSpg–/– testes , treated with BMS961 ) . Altogether , these data indicate that Sall4a expression is controlled by ATRA-activated RARG in spermatogonia . Using data sets locating RAR-occupied sites genome-wide in several cell-types [33 , 34] , we identified a 700 bp-long RAR-binding region located in the first intron of Sall4 ( RARE , Fig 7A ) . This DNA fragment contained a RAR binding sequence called IR1 , consisting of inverted repeats ( two core motifs 5’-RGKTSA-3’ oriented head-to-tail ) separated by 1 bp ( Fig 7B ) , as well as two additional sites called DR1 and DR0 ( direct repeats of the core motif separated by 1 and 0 bp , respectively ) . We performed triplicate IP experiments with anti-RAR and anti-RXR antibodies using chromatin extracted from PN5 wild-type mouse testes as substrate . At this developmental stage , Sall4a expression was dependent upon RARG ( Fig 7C , left panel ) . We analyzed the immuno-precipitated chromatin fragments by qPCR and evidenced robust binding of both RAR and RXR in vivo , in a 106bp-long region restricted to chr2:168 , 591 , 142–168 , 591 , 247 ( NCBI37/mm9 ) in Sall4 ( RARE , Fig 7C , right panel ) . To further confirm interaction of IR1 with RAR/RXR heterodimers , we performed electrophoretic mobility shift assays ( EMSA ) ( Fig 7D ) . They revealed that RARG isotype in combination with RXRA isotype ( lane 4 ) , but neither RARG nor RXRA alone ( lanes 2 and 3 , respectively ) , bound the radiolabelled IR1 sequence . Binding was competed by increasing amounts of unlabeled IR1 ( lanes 5–7 ) , but not by IR1m bearing point-mutations in the first core motif ( lanes 8–10 ) . They also showed that unlabeled IR1 efficiently competed binding of RARG/RXRA heterodimers to the radiolabeled , canonical , RAR binding site of Rarb gene ( called DR5 , Fig 7E ) . The data suggested therefore that RARG/RXRA heterodimers could enhance expression of SALL4A through binding to an IR1 motif located in Sall4 intron . This motif appeared moderately well-conserved in mouse , rat , human and primate genomes ( Fig 7B ) , but single mismatches do not necessarily abrogate RAR/RXR binding , even when located at highly conserved positions [33] . The DR1 and DR0 were also able to bind RARG/RXRA heterodimers ( S5 Fig ) . Their sequences were even well-conserved across the species than that of IR1 ( Fig 7B ) . Thus the RAR binding region in Sall4 belongs to the category of “composite elements” , the functionality of which has already been demonstrated [34] . Interestingly , SALL4A is also expressed in human spermatogonia [35] . Moreover , Fertilysin ( N , N’-1 , 8-octanediylbis[2 , 2-dichloro-acetamide] , also called WIN 18 , 446 ) , which acts by inhibiting ATRA synthesis [36] , reversibly inhibits spermatogenesis in men by inducing an arrest of germ cell differentiation at the spermatogonia stage [37 , 38] , which resembles the phenotype we describe here in the mouse . Thus it is possible that RAR/RXR heterodimers also drive SALL4A expression in human spermatogonia . A hallmark of the transition to a differentiating state in spermatogonia is the expression of KIT receptor at the surface of A1 spermatogonia [5] . However , it is unclear what regulatory steps control the expression of this crucial cell surface receptor . Several studies have shown that undifferentiated spermatogonia are primed to turns on KIT and initiate differentiation upon activation of an ATRA signal [2] . This ATRA signal acts indirectly on Kit because its mRNA is not induced by BMS961 , as assessed from our microarray expression profiling ( see above ) and no RAR binding site is found in Kit [39] . From our present study , we propose that ATRA enhances the level of Sall4a mRNA , allowing thereby an increase of the amount of SALL4A in spermatogonia . In agreement with this proposal , SALL4A appears at PN3-PN4 in wild-type spermatogonia [31 , 32] , coinciding with the onset of endogenous ATRA signaling [40] and differentiation of the first KIT-positive spermatogonia [41] . SALL4A in high amount could ( i ) sequester ZBTB16 , resulting in the release of the ZBTB16-mediated repression of Kit expression [14]; ( ii ) interact with DNMT3A and/or DNMT3B [42] , allowing the epigenetic shift which is instrumental to A1 transition to take place properly [43]; and ( iii ) act on yet unknown other components of the differentiation program required in spermatogonia to transition to the A1 state . In agreement with our proposal , Hobbs et al . [14] reported that the total amount of SALL4 protein detected in spermatogonia is higher at the A1 stage ( i . e . , in the ZBTB16-high , KIT-positive cell population ) than at the Aal stage ( i . e . , in the ZBTB16-high , KIT-negative cell population ) . It is however not possible to show by IHC that Aal spermatogonia expressing SALL4A upon ATRA signaling activation differentiate into A1 spermatogonia and express KIT because , contrary to what has been stated in a previous report [31] , antibodies to SALL4 do not distinguish between SALL4A and SALL4B . As SALL4B is expressed in spermatogonia and in their precursors from embryonic day 17 . 5 onwards [31 , 32] , these antibodies are unsuitable to detect a specific increase in the expression of the sole SALL4Aisoform . A comprehensive model summarizing the combination of transcriptional , post-transcriptional and non-genomic effects of ATRA pathways possibly controlling KIT expression and the commitment of Aal spermatogonia towards the A1 fate is proposed ( Fig 8 ) . The interest of better understanding the control of KIT expression in spermatogonia is not restricted to gametogenesis , but extends to testicular cancer . In fact , seminoma cells frequently bear somatic mutations activating KIT , or overexpress KIT or SALL4 [7 , 44 , 45] . Therefore , pharmacological modulation of mechanisms that regulate KIT expression in spermatogonia , such as antagonizing ATRA action , might have important applications for future therapeutic strategies .
Mice were on a mixed C57BL/6-129/Sv ( 50–50% ) genetic background . They were housed in a licensed animal facility ( agreement #A67-218-37 ) . All experiments were approved by the local ethical committee ( Com’Eth , accreditations #2012–080 and #2012–081 ) , and were supervised by N . B . G . or M . M . who are qualified in compliance with the European Community guidelines for laboratory animal care and use ( 2010/63/UE ) . To inactivate Rar- or Rxr-coding genes in spermatogonia , mice carrying loxP-flanked alleles ( L2 ) of Rara , Rarb and Rarg or of Rxra , Rxrb and Rxrg [12 , and references therein] were crossed with mice bearing the Tg ( Stra8-cre ) 1Reb transgene [16] . In F1 , RxraL2/L2;RxrbL2/L2;RxrgL2/L2 females were crossed with males bearing one copy of the transgene ( Stra8tg/0 ) . The resulting males ( Stra8-Cretg/0;Rxra+/L2;Rxrb+/L2;Rxrg+/L2 ) were backcrossed on RxraL2/L2;RxrbL2/L2;RxrgL2/L2 females to generate mutant males in F2 ( Stra8-Cretg/0;RxraL2/L2;RxrbL2/L2;RargL2/L2 ) , and their control littermates ( Rxra+/L2;Rxrb+/L2;Rxrg+/L2 and RxraL2/L2;RxrbL2/L2;RxrgL2/L2 males ) . The same approach was used to inactivate Rara , Rarb and Rarg in spermatogonia . Aldh1a1-3ser-/- mutants were generated as described previously [13] . Mice lacking RARA in Sertoli cells in a RARG-null genetic background were obtained by crossing RaraSer–/–[12] and Rarg–/– mice [9] together . BMS961 ( 50 mg/kg body weight , Tocris Bioscience ) dissolved in dimethylsulfoxide was administered to the mice by intra peritoneal injections . BrdU ( Sigma-Aldrich ) was dissolved in phosphate buffered saline and injected by intra peritoneal at 50 mg/kg body weight . Germ cell populations were purified from testes of Aldh1a1-3ser-/- , Rara;b;gSpg–/– and Rxra;b;gSpg–/– mice by FACS and characterized as described previously [13 , 22] . Organotypic cultures of testes from Aldh1a1-3ser-/- mice were also as described previously , except that the RARG-selective agonist BMS961 at 10–7M ( Tocris Bioscience ) was used to activate RAR signaling instead of BMS753 [13] . For histology , testis samples were fixed in Bouin’s fluid for 16 hours and embedded in paraffin . Histological sections ( 5 μm-thick ) were stained with hematoxylin and eosin or with periodic acid-Shiff ( PAS ) . The percentage of affected seminiferous tubules was established on PAS-stained histological sections by counting cross-sections of tubules ( n > 400 per testis ) . For all other methods , testes were fixed for 16 hours in 4% ( wt/vol ) buffered paraformaldehyde ( PFA ) . For detection of apoptotic cells , TUNEL assays were performed using the In Situ Cell-Death Detection kit , Fluorescein ( Roche Diagnostics ) . BrdU incorporation was detected by using an anti-BrdU antibody ( Roche Molecular Biochemicals ) and immunofluorescence labeling as described [9] . For IHC , 10 μm-thick frozen sections were incubated overnight at 4°C with rabbit anti-STRA8 ( Ab49602 , Abcam ) , rabbit anti-RARG1 ( D3A4 #8965 , Cell Signaling Technology ) , goat anti-ZBTB16 ( AF2944 , R&D Systems ) and rabbit anti-KIT ( D13A2 #3074 , Cell Signaling Technology ) antibodies diluted 1:200 to 1:500 . Detection of bound primary antibodies was achieved by incubating the section with Cy3-conjugated goat anti-rabbit IgG ( Jackson ImmunoResearch ) or Alexa Fluor 488-conjugated donkey anti-goat IgG ( Life Technologies ) . ISH using digoxigenin-labeled probes for detection of Gfra1 , Kit , Stra8 and Zbtb16 expression was performed as described [9 , 10 , 20] . The sections were all counterstained with 0 . 001% ( vol/vol ) 4 , 6-diamidino-2-phenylindole dihydrochloride ( DAPI ) and mounted in Vectashield ( Vector Laboratories ) . The pattern of Cre expression driven by Tg ( Stra8-cre ) 1Reb [16] was assessed through testing excision in mice carrying the Gt ( ROSA ) 26Sortm1Sor reporter transgene [46] . In these mice , E . coli beta-galactosidase is synthesized only in cells that have experienced Cre-mediated deletion of an intervening stop sequence . Analysis of beta-galactosidase activity was as described [10] . Total RNA was prepared using TRIzol reagent ( Life Technologies ) . Reverse transcription of total RNA followed by PCR amplification of cDNA was performed using QuantiTect Reverse Transcription ( Qiagen ) and LightCycler 480 SYBR Green I Master ( Roche Diagnostics ) kits , respectively . Primers were as indicated in Table 1 . Triplicates of at least three samples were used in each experimental condition . The transcript levels were normalized relative to that of Rplp0 or Gapdh transcripts , whose expressions are not changed by retinoid administration . Data were expressed as fold induction relative to vehicle or control conditions . To prepare chromatin , PN5 testes were fixed with 0 . 4% PFA ( wt/vol ) for 15 minutes , before being sonicated to shear DNA to an average size of 500 bp . For each reaction , 100 μg of chromatin was first incubated with 18 μg of ChIP grade anti-RAR ( sc-773; Santa Cruz biotechnology ) , anti-RXR ( sc-774; Santa Cruz biotechnology ) or anti-RNA polymerase II ( RNApol2; sc-9001; Santa Cruz biotechnology ) antibodies and then with protein G-Sepharose . Beads were washed , and eluted DNA–protein complexes were reverse cross-linked and purified . ChIP was performed in triplicate , using distinct chromatin extracts . The recovered immuno-precipitated DNA was analyzed by triplicate qPCR and was compared with input DNA . Quantitation was determined by the enrichment of the binding site compared with a site located upstream the TSS ( –11kb ) , and were expressed as mean fold-enrichment ( n = 3 ) . The sequences of the oligonucleotides used are indicated in Table 2 . Statistical significance was assessed by Student t tests or by one-way ANOVA followed by the post hoc Newman-Keuls test for comparison by pairs . They were performed as described previously [47] . Briefly , the oligonucleotides were annealed and labeled with [γ-32P]ATP ( Amersham Bioscience ) . For competition assays , unlabeled oligonucleotides were added in the incubation mixture in 1- to 1000-fold molar excess . The sequences of the oligonucleotides used are described in Table 2 . Protein extracts were prepared in 50 mM Tris-HCl ( pH7 . 5 ) buffer containing 150 mM NaCl , 0 . 5% ( wt/vol ) sodium deoxycholate , 1% ( vol/vol ) NP40 , 0 . 2% ( vol/vol ) sodium dodecyl sulfate ( SDS ) , and protease inhibitor mixture ( Roche diagnostics ) . They were resolved by 4–16% ( wt/vol ) gradient SDS polyacrylamide gel electrophoresis ( Expedeon ) and transferred to nitrocellulose membranes ( Protran ) using standard protocols . The membranes were incubated with anti-SALL4 antibodies ( Ab29112; Abcam ) diluted 1:500 and IgG were detected using goat anti-mouse coupled to horseradish peroxidase as secondary antibodies ( diluted 1:5000 ) followed by chemiluminescence according to the manufacturer’s protocol ( GE Healthcare ) . The blots were subsequently incubated 2 times for 10 minutes at room temperature in 0 . 2M glycine pH 2 . 2 containing 0 . 1% ( wt/vol ) SDS and 0 . 1% ( wt/vol ) Tween 20 , and were further probed with anti-actin antibodies ( sc-58673; Santa Cruz biotechnology ) diluted 1:500 to verify for equivalent loading in all the lanes .
|
Differentiation of spermatozoa from immature germ cells , called spermatogonia , critically depends on retinoic acid ( ATRA ) , the active metabolite of vitamin A that acts though binding to nuclear receptors called RXR and RAR . To understand the mechanism by which ATRA control germ cell differentiation , we generated mice simultaneously lacking all RXR or all RAR specifically in spermatogonia . From their phenotypic analysis , we demonstrate that meiosis does not require a RAR/RXR-dependent pathway in germ cells and propose that this process is either ATRA-independent or requires an ATRA signal originating from somatic cells . We also show that RXR , in the form of dimers with RAR , can drive spermatogonia differentiation through binding to a regulatory region located in the Sall4 gene . This finding is significant , as the transcription factor encoded by Sall4 is known to regulate the expression of KIT , a key tyrosine kinase receptor which is frequently deregulated in testicular cancer .
|
[
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Retinoic Acid Receptors Control Spermatogonia Cell-Fate and Induce Expression of the SALL4A Transcription Factor
|
It has been recently hypothesized that many of the signals detected in genome-wide association studies ( GWAS ) to T2D and other diseases , despite being observed to common variants , might in fact result from causal mutations that are rare . One prediction of this hypothesis is that the allelic associations should be population-specific , as the causal mutations arose after the migrations that established different populations around the world . We selected 19 common variants found to be reproducibly associated to T2D risk in European populations and studied them in a large multiethnic case-control study ( 6 , 142 cases and 7 , 403 controls ) among men and women from 5 racial/ethnic groups ( European Americans , African Americans , Latinos , Japanese Americans , and Native Hawaiians ) . In analysis pooled across ethnic groups , the allelic associations were in the same direction as the original report for all 19 variants , and 14 of the 19 were significantly associated with risk . In summing the number of risk alleles for each individual , the per-allele associations were highly statistically significant ( P<10−4 ) and similar in all populations ( odds ratios 1 . 09–1 . 12 ) except in Japanese Americans the estimated effect per allele was larger than in the other populations ( 1 . 20; Phet = 3 . 8×10−4 ) . We did not observe ethnic differences in the distribution of risk that would explain the increased prevalence of type 2 diabetes in these groups as compared to European Americans . The consistency of allelic associations in diverse racial/ethnic groups is not predicted under the hypothesis of Goldstein regarding “synthetic associations” of rare mutations in T2D .
Multiple common risk alleles have been identified as reproducibly associated with risk of type 2 diabetes ( T2D ) [1]–[13] . With the exception of the KCNQ1 locus which was identified in the Japanese population [1] , [2] , all of the well-replicated risk variants were first identified in populations of Northern European ancestry [3]–[13] . T2D morbidity varies widely across racial/ethnic groups; the prevalence is more than twice as high among African Americans , Japanese Americans , Latinos and Native Hawaiians as European Americans [14] , [15] . It is important to evaluate whether and how genetic variation may contribute to health disparities between populations . For example , genetic variation at 8q24 may contribute to population differences in risk of prostate cancer [16] , [17] , and genetic variation at MYH9 contributes substantially to the higher rates of kidney disease in African Americans [18] . It has recently been argued that single rare causal variants and/or collections of multiple different rare variants on unrelated haplotypes may create “synthetic associations” of common variants with disease risk [19]–[21] . One prediction of this model is that the associations with common variants will not be consistent across populations ( since many of the mutations will be young in age , and post-date the migrations that led to the founding of modern continental populations ) . Type 2 diabetes has been specifically discussed as a possible case in which synthetic associations might be operative , based on the lack of statistical significance in very small studies that examined allelic associations for T2D in multi-ethnic samples . Testing the association of each validated risk allele for T2D in multiple populations is an important step to determine ( a ) whether these genetic markers can be used to better understand population risk in non-European populations , ( b ) to measure their association with racial/ethnic variation in disease risk , and ( c ) to test a prediction of the Goldstein “common SNP , rare mutation” hypothesis [19]–[21] . To allow for comparability of estimates of genetic risk among racial/ethnic groups requires large studies comprised of cases and controls defined using identical criteria and sampled ideally from the same study population . In the present study , we , as part of the Population Architecture using Genomics and Epidemiology ( PAGE ) Study , examined genetic associations with 19 validated risk alleles for T2D in European American , African American , Latino , Japanese American , and Native Hawaiian T2D cases ( n = 6 , 142 ) and controls ( n = 7 , 403 ) from the population-based Multiethnic Cohort study ( MEC ) . We also evaluated whether these variants can be utilized to model the genetic risk of T2D in each population and their association to disparities in risk .
We next calculated a summary risk score comprised of an unweighted count of the 19 risk-associated alleles . The average increment in risk per allele was generally similar in all populations , except Japanese Americans , where the effect of each allele was nearly double that observed in Europeans ( ( odds ratio , 95% confidence interval ) : African Americans , 1 . 09 , 1 . 05–1 . 12; ( P = 3 . 0×10−6 ) ; Native Hawaiians , 1 . 10 , 1 . 06–1 . 15 ( P = 1 . 2×10−5 ) ; European Americans , 1 . 11 , 1 . 06–1 . 17 ( P = 1 . 2×10−5 ) ; Latinos , 1 . 12 , 1 . 09–1 . 14 ( P = 7 . 5×10−19 ) ; and , Japanese , 1 . 20 , 1 . 17–1 . 24; ( P = 7 . 0×10−32 ) ; Phet = 3 . 8×10−4 ) . Results were similar when limiting the analysis to individuals with complete genotype data for all variants and when including only those markers associated with risk ( at P<0 . 10 ) ( Table S5 ) . Individuals in the top quartile of the risk allele distribution were at 1 . 6 ( African Americans , P = 5 . 3×10−4 ) to 3 . 1-fold ( Japanese Americans , P = 7 . 9×10−26 ) greater risk of diabetes compared to those in the lowest quartile ( Table 3 ) . Using these ethnic-specific per allele odds ratio estimates and the aggregate risk allele counts , we built a quantitative risk model to compare the distribution of genetic risks between populations associated with these marker alleles . The higher average number of risk alleles in African Americans caused their distribution to be slightly right shifted ( towards higher log ORs ) compared to European Americans , however their relatively smaller per allele odds ratio resulted in wide overlap with the European American distribution ( Figure 2 ) . The Japanese Americans had a wider distribution of risk because of the large per allele odds ratio , but the low average risk allele counts caused the Japanese distribution to be left-shifted ( towards lower log ORs ) compared to European Americans . The distributions for Latinos and Native Hawaiians were very similar to the European Americans .
We tested 19 common genetic risk markers that were discovered in European populations . We found that association with all 19 of these SNPs trended in the same direction in this large multiethnic study , and the majority of these variants were nominally significant in their association with diabetes risk . A risk score comprised of these alleles was significantly associated with diabetes risk in all five racial/ethnic groups , with the only significant heterogeneity being larger effect sizes in Japanese Americans . However , in comparing the distribution of risk conferred by these alleles between populations we found that they explain little , if any , of known differences in the prevalence of diabetes between these populations . These observations indicate that most , if not all , of these alleles show directionally similar association to T2D across many populations . Such a pattern indicates that the causal alleles at these validated risk loci ( which have yet to be found ) likely predate the migrations that separated these populations now residing in Europe , Africa , East Asia , the Pacific Islands and the Americas . We note that this pattern is unexpected under the recently described “common SNP , rare mutation” model of Goldstein that suggests that GWAS signals with common alleles for T2D and other diseases may be “synthetic associations” created by one or more rare alleles [19]–[21] . Under the Goldstein Hypothesis the consistent associations that we noted at these loci across populations would only be observed if , in each population , one or more distinct rare alleles arose at each locus , and they happened to arise each time on the same haplotype background . Although possible , this scenario seems unlikely , and a more parsimonious explanation would be the “synthetic association” hypothesis of Goldstein does not apply to a majority of these T2D SNPs . The modest number of cases and controls in this study ( as compared to the initial discovery studies ) likely underlies the lack of statistically significant associations in some groups . Weaker associations in some racial/ethnic groups may also be due to differences in allele frequencies , linkage disequilibrium , and environmental and genetic modifiers . In two cases ( WFS1 and CDKAL1 ) , significant heterogeneity by race/ethnicity reflected a lack of association in African Americans , perhaps because of lower linkage disequilibrium between the marker and the biologically relevant allele . It is interesting that the odds ratios observed for these marker SNPs were larger in Japanese Americans than in the original discovery cohorts , and in the other ethnic groups in our study . A meta-analysis of 7 association studies in Japanese populations replicated associations from studies in European populations for 7 loci under study ( TCF7L2 , CDKAL1 , CDKN2B , IGF2BP2 , SLC30A8 , KCNJ11 , and HHEX ) [24] . A recent GWAS in Japanese observed significant associations in KCNQ1 as well as these same 7 loci and , similar to our observations , noted magnitudes of effect that were generally stronger than previously observed in European populations [25] . Additional studies in other Asian populations have replicated associations with many of these loci as well [24] , [26]–[28] . In the Multiethnic Cohort , we have found the prevalence of T2D to be at least 2-fold higher in African Americans , Latinos , Japanese and Native Hawaiians compared to European Americans , with these differences being independent of body weight [14] . We examined the extent to which the known genetic risk alleles for diabetes could explain these disparities by quantifying and comparing the relative risk distributions between populations . Compared to European Americans , we did not observe evidence of greater genetic risk in any population . Our findings therefore indicate that these risk markers explain little , if any , of racial/ethnic disparities in T2D prevalence . It remains possible that the actual causal alleles in these regions may be more common in frequency and/or have larger effects than the index signals in non-European populations . As seen with KCNQ1 [1] , [2] , GWAS in non-European populations are effective in discovering risk loci that are important in multiple populations but difficult to identify in European populations where the alleles are rare . This study had a number of limitations . First , a self-report of diabetes and use of medication for diabetes was used to define cases and controls . We observed that approximately 1% of a random sample of the controls in this study had HbA1C levels above 7 . 0% , which suggests that only a small portion of controls had undiagnosed diabetes ( see Materials and Methods ) . Also , our case definition did not differentiate between T1D and T2D , however we expect this misclassification to be minor as <3% of T2D cases had a previous diagnosis of T1D based on other sources ( see Materials and Methods ) . The highly consistent findings of this study , as compared to the discovery GWAS reports , argue that our phenotypic characterization is adequate to observe the association to T2D . Some caution should also be given to the interpretation of the risk modeling conducted in each ethnic group , as the genetic markers included are unlikely to be the causal alleles . Future fine-mapping and sequencing studies to identify the functional variants ( common and/or rare ) and large-scale testing of each allele will be required to more precisely model risk as well as assess differences in the distribution of genetic risk across populations . Another limitation is that we did not account for the potential confounding effects of population stratification . However , odds ratios were essentially unchanged after adjusting for global European ancestry in a subset of African Americans ( 336 cases 397 controls ) for whom ancestry markers were available , suggesting that effects due to population substructure were not substantial , at least in this group . We also noted that controlling for education , a proxy for SES which has been shown to be significantly associated with Native American ancestry in Latinos [23] , had little effect on the associations with these risk alleles . Furthermore , the risk alleles were not generally more frequent in Latinos than in European Americans which would be likely if these alleles were proxies for more general ancestry differences . While population stratification is unlikely to fully explain these findings , it remains possible that at some loci , the causal alleles may be more correlated with ancestry than the index SNPs . In summary , our data provide strong support for common genetic variation contributing to T2D risk in multiple populations . Our findings in T2D do not support the theory that GWAS signals are due to rare alleles . Nonetheless , GWAS and sequencing studies in these and other racial/ethnic populations are needed to reveal a more complete spectrum of risk alleles that are important globally as well as those that may contribute to risk disparities .
The Institutional Review Boards at the University of Southern California and University of Hawaii approved the study protocol . The MEC consists of 215 , 251 men and women , and comprises mainly five self-reported racial/ethnic populations: European Americans , African Americans , Latinos , Japanese Americans and Native Hawaiians [29] . Between 1993 and 1996 , adults between 45 and 75 years old were enrolled by completing a 26-page , self-administered questionnaire asking detailed information about dietary habits , demographic factors , level of education , personal behaviors , and history of prior medical conditions ( e . g . diabetes ) . Potential cohort members were identified through Department of Motor Vehicles drivers' license files , voter registration files and Health Care Financing Administration data files . In 2001 , a short follow-up questionnaire was sent to update information on dietary habits , as well as to obtain information about new diagnoses of medical conditions since recruitment . Between 2003 and 2007 , we re-administered a modified version of the baseline questionnaire . All questionnaires inquired about history of diabetes , without specification as to type ( 1 vs . 2 ) . Between 1995 and 2004 , blood specimens were collected from ∼67 , 000 MEC participants at which time a short questionnaire was administered to update certain exposures , and collect current information about medication use . Cohort members in California are linked each year to the California Office of Statewide Health Planning and Development ( OSHPD ) hospitalization discharge database which consists of mandatory records of all in-patient hospitalizations at most acute-care facilities in California . Records include information on the principal diagnosis plus up to 24 other diagnoses ( coded according to ICD-9 ) , including T1D and T2D . In Hawaii cohort members have been linked with the diabetes care registries for subjects with Hawaii Medical Service Association ( HMSA ) and Kaiser Permanente Hawaii ( KPH ) health plans ( ∼90% of the Hawaiian population has one of these two plans ) [15] . Information from these additional databases have been utilized to assess the percentage of T2D controls ( as defined below ) with undiagnosed T2D , as well as the percentage of identified diabetes cases with T1D rather than T2D . Based on the OSHPD database <3% of T2D cases had a previous diagnosis of T1D . We did not use these sources to identify T2D cases because they did not include information on diabetes medications , one of our inclusion criteria for cases ( see below ) . In this study , diabetic cases were defined using the following criteria: ( a ) a self-report of diabetes on the baseline questionnaire , 2nd questionnaire or 3rd questionnaire; and ( b ) self-report of taking medication for T2D at the time of blood draw; and ( c ) no diagnosis of T1D in the absence of a T2D diagnosis from the OSHPD ( California Residents ) . Controls were defined as: ( a ) no self-report of diabetes on any of the questionnaires while having completed a minimum of 2 of the 3 ( 79% of controls returned all 3 questionnaires ) ; and ( b ) no use of medications for T2D at the time of blood draw; and ( c ) no diabetes diagnosis ( type 1 or 2 ) from the OSHPD , HMSA or KPH registries . To preserve DNA for genetic studies of cancer in the MEC , subjects with an incident cancer diagnosis at time of selection for this study were excluded . Controls were frequency matched to cases on age at entry into the cohort ( 5-year age groups ) and for Latinos , place of birth ( U . S . vs . Mexico , South or Central America ) , oversampling African American , Native Hawaiian and European American controls to increase statistical power . Fasting glucose ( FG ) and HbA1C measurements were used to validate the case-control selection criteria . Among 185 T2D cases and 1 , 048 controls who met the T2D case-control definitions above and with FG measurements available from ongoing studies in the MEC , 57% of cases ( ranging from 43% in European Americans to 63% in Japanese Americans ) and 3% of controls ( ranging from 1% in African Americans to 6% in Latinos ) had a FG value >125 mg/dl . We also measured HbA1C ( ARUP Laboratories , Salt Lake City , Utah ) in 50 cases and 50 controls per each sex-ethnic group . Just over 1% ( 6/500 ) of controls were likely to have unreported T2D ( HbA1C value ≥7% ) . In contrast , ∼47% ( 234/500 ) of T2D cases had HbA1C ≥7% ( ranging from 41% in European Americans to 57% of Native Hawaiians ) . Since hypoglycemic medication use was part of the case selection criteria , some cases were expected to have FG and HbA1C levels in the normal range . Altogether , this study included 6 , 142 T2D cases and 7 , 403 controls ( European American ( 533/1 , 006 ) , African American ( 1 , 077/1 , 469 ) , Latino ( 2 , 220/2 , 184 ) , Japanese American ( 1 , 736/1 , 761 ) and Native Hawaiian ( 576/983 ) ) . Genotyping was conducted by the TaqMan allelic discrimination assay ( Applied Biosystems , Foster City , CA ) [30] . For all SNPs , genotype call rates were >95% among case and control groups in each population and HWE p-values among controls were >0 . 05 in at least 4 of the 5 ethnic groups and none of the values were <0 . 01 ( Table S6 ) . Subjects missing data for >5 SNPs ( n = 82 ) were removed from the analysis . Odds ratios and 95% confidence intervals were calculated for each allele in unconditional logistic regression models while adjusting for age at cohort entry ( quartiles ) , body mass index ( BMI , kg/m2 , quartiles ) , sex , and race/ethnicity ( pooled analysis ) in ethnic-stratified and pooled analyses . Associations with the two variants at KCNQ1 were examined adjusting for the other allele . Potential confounding factors including , smoking history , education , physical activity , and history of hypertension were evaluated but did not influence the results . Potential confounding by percent European ancestry was examined in a subset of African American men ( 336 cases , 397 controls ) with available genetic ancestry information [16] , [31] , [32] . We also modeled the cumulative genetic risk of T2D using these markers . We summed the number of risk alleles for each individual and estimated the odds ratio per allele for this aggregate unweighted allele count variable as an approximate risk score appropriate for unlinked variants with independent effects of approximately the same magnitude for each allele . We also examined a second model where each allele was weighted and multiplied by the log of the published odds ratio prior to summing all alleles . The results of the more parsimonious unweighted risk score is presented as the two risk scores were highly correlated in each ethnic group ( Pearson r≥0 . 92 ) and similar associations with T2D risk were observed for each score . For individuals missing genotypes for a given SNP , we assigned the average number of risk alleles within each ethnic group ( 2× risk allele frequency ) to replace the missing value for that SNP . We used these ethnic-specific per allele summary odds ratios and the total number of risk alleles among control subjects to estimate the distribution of relative risks conveyed by all risk alleles . To avoid making the reference group carriers of zero risk alleles ( a group which does not exist ) we centered the distribution on the mean number of risk alleles observed in the control population ( 18 . 5 ) . The log relative risk for each subject was calculated as logRR = ( RA−18 . 5 ) ×log ( ORi ) ( where RA is equal to the subject's total risk alleles and log ( ORi ) is the log of the ethnic specific per allele odds ratio . A spline function was used to capture the shape of the distributions of log OR for display purposes . Two variants in KCNQ1 were included in the risk modeling because both were significantly associated with T2D when co-modeled ( results were similar when only the most significant of the two , rs2237897 , was included ) . The variant in FTO was excluded from risk modeling procedures , as we found ( as have others ) that it is not a risk factor for diabetes independent of its effect on obesity .
|
Single rare causal alleles and/or collections of multiple rare alleles have been suggested to create “synthetic associations” with common variants in genome-wide association studies ( GWAS ) . This model predicts that associations with common variants will not be consistent across populations . In this study , we examined 19 T2D variants for association with T2D risk in 6 , 142 cases and 7 , 403 controls from five racial/ethnic populations in the Multiethnic Cohort ( European Americans , African Americans , Latinos , Japanese Americans , and Native Hawaiians ) . In racial/ethnic pooled analysis , all 19 variants were associated with T2D risk in the same direction as previous reports in Europeans , and the sum total of risk variants was significantly associated with T2D risk in each racial/ethnic group . The consistent associations across populations do not support the Goldstein hypothesis that rare causal alleles underlie GWAS signals . We also did not find evidence that these markers underlie racial/ethnic disparities in T2D prevalence . Large-scale GWAS and sequencing studies in these populations are necessary in order to both improve the current set of markers at these risk loci and identify new risk variants for T2D that may be difficult , or impossible , to detect in European populations .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"public",
"health",
"and",
"epidemiology/epidemiology",
"genetics",
"and",
"genomics",
"diabetes",
"and",
"endocrinology/type",
"2",
"diabetes"
] |
2010
|
Consistent Association of Type 2 Diabetes Risk Variants Found in Europeans in Diverse Racial and Ethnic Groups
|
During oogenesis , the egg prepares for fertilization and early embryogenesis . As a consequence , vesicle transport is very active during vitellogenesis , and oocytes are an outstanding system to study regulators of membrane trafficking . Here , we combine zebrafish genetics and the oocyte model to identify the molecular lesion underlying the zebrafish souffle ( suf ) mutation . We demonstrate that suf encodes the homolog of the Hereditary Spastic Paraplegia ( HSP ) gene SPASTIZIN ( SPG15 ) . We show that in zebrafish oocytes suf mutants accumulate Rab11b-positive vesicles , but trafficking of recycling endosomes is not affected . Instead , we detect Suf/Spastizin on cortical granules , which undergo regulated secretion . We demonstrate genetically that Suf is essential for granule maturation into secretion competent dense-core vesicles describing a novel role for Suf in vesicle maturation . Interestingly , in suf mutants immature , secretory precursors accumulate , because they fail to pinch-off Clathrin-coated buds . Moreover , pharmacological inhibition of the abscission regulator Dynamin leads to an accumulation of immature secretory granules and mimics the suf phenotype . Our results identify a novel regulator of secretory vesicle formation in the zebrafish oocyte . In addition , we describe an uncharacterized cellular mechanism for Suf/Spastizin activity during secretion , which raises the possibility of novel therapeutic avenues for HSP research .
Oogenesis prepares the egg to start the development of a new organism . During their development , oocytes actively import and secrete proteins using the basic cellular mechanism of vesicle transport [1]–[3] . As a consequence , oocytes contributed substantially to our understanding of vesicle trafficking e . g . Clathrin-coated vesicles were first described in mosquito oocytes [4] . Moreover , a plethora of novel regulators were discovered by exploiting the genetics of the Caenorhabditis elegans oocyte [5]–[9] . In zebrafish , oogenesis starts with a burst of secretory vesicle formation , which are called cortical granules [10] , [11] . Shortly afterwards , oocytes endocytose enormous amounts of the yolk precursor protein Vitellogenin ( Vtg ) by receptor-mediated endocytosis increasing its volume about 3000-fold within ten days [12] , [reviewed in 13] . Hence , the zebrafish oocyte provides the opportunity to integrate vertebrate genetics to visualize active trafficking of abundant and large vesicles in one big cell . In humans , defects in vesicle trafficking lead to neurodegenerative diseases [14]–[16] . The disorder Hereditary Spastic Paraplegia ( HSP ) is characterized by progressive loss of lower limb motility [17]–[19] . At the cellular level , this spasticity is caused by the axonal degeneration of neurons in the corticospinal tracts , which are considered the longest axons in the human body . However , the cellular analysis of the dying neurons is hampered by the adult onset of the disease and the complexity of the nervous system . Therefore , the precise cellular mechanism for most HSP genes is currently controversial . One of the HSP genes with multiple cellular roles is SPG15 , which encodes SPASTIZIN aka ZFYVE26 or FYVE-CENT [20] , [21] . In a cell culture RNAi screen , SPASTIZIN was shown to be necessary for cytokinesis , but this defect is not observed in human and murine mutants [21] , [22] . This cell culture study found that the endogenous protein localizes to centrosomes and the midbody during cytokinesis of human fibroblasts . In neuronal tissue culture , SPASTIZIN was shown to localize to microtubules , ER , mitochondria and endosomal vesicles [22] , [23] . In additional tissue culture reports , SPASTIZIN interacted with the UVRAG complex during DNA repair [24] and autophagosome maturation [25] . Furthermore , SPASTIZIN binds to the recently discovered adaptor protein complex 5 ( AP5 ) regulating multivesicular body ( MVB ) formation [26] , [27] . Although these localization data in different tissue culture cells are not mutually exclusive , it is difficult to reconcile the described cellular functions of SPASTIZIN in fibroblasts into an underlying cellular defect causing neurodegeneration in neuronal cells . Mutants provide an essential tool to determine the endogenous role of a gene . Previously , we used a mutagenesis screen in zebrafish to discover vertebrate regulators of egg development and early embryogenesis , which isolated the souffle ( suf ) mutation named after its defect during oogenesis [28] . Here , we positionally cloned suf and show that it encodes the homolog of SPASTIZIN . In suf mutants , we show that oocytes expand a Rab11b-positive compartment , but correctly transport the recycling cargo Transferrin . Moreover , Suf colocalizes with Rab11b on secretory vesicles called cortical granules in the oocyte . We demonstrate genetically that Suf/Spastizin is essential for the formation of cortical granules . Importantly , our subcellular analysis indicates that loss of Suf/Spastizin inhibits vesicle maturation , probably during sorting , which is necessary to complete the formation of Clathrin-coated buds and eventually to pinch-off vesicles . Finally , blocking vesicle scission with the pharmacological Dynamin-inhibitor Dynasore mimics the mutant phenotype supporting the hypothesis that Suf is required for vesicle fission , which is critical for the maturation of secretory granules in the egg . Collectively , these results identify Suf/Spastizin as a novel key gene controlling the maturation of cortical granules in zebrafish oocytes , which may also bring us closer to understand the cellular etiology of HSP .
Contrary to transparent eggs of wild-type ( wt ) mothers , maternal suf mutants spawn opaque eggs , which fail to proteolytically cleave yolk proteins [28] . In addition , mutant stage V eggs did not elevate their chorion after activation as observed in wt ( Figure 1A , B ) . With this early defect the suf mutation rather causes a female-sterile than a maternal-effect phenotype . To characterize the molecular mechanism of the defect , we positionally cloned the disrupted gene in suf mutants . We previously located the suf mutation on chromosome 13 of the zebrafish genome . Genotyping 1183 females identified 5 recombinants with the SSLP marker z25580/G47633 and another 5 recombinants with z21403/G41743 restricting the critical interval with the suf mutation to 1 . 22 Mb ( Figure 1C ) . Generating novel SSLP markers AL13-10 and AL13-13 , we reduced the interval to 270 kb . Within this genomic region , we sequenced the cDNAs of arginase , vti1b ( t-snare ) , rdh12 ( retinol dehydrogenase 12 ) , zfyve26 ( spastizin ) , galectin and pleckstrin2 ( Figure 1D ) . The zfyve26 gene consists of 41 exons and encodes a predicted mRNA of 7798 bp ( Figure 1E ) . Comparing the zfyve26 cDNA-sequence between wild type and the p96re allele of suf mutants showed a 25 bp deletion at the 3′-end of exon 35 ( Figure 1G ) . However , the genomic sequence revealed a single point mutation in a splice donor , which probably results in cryptic splice donor selection 25 bp upstream of the wt splice site ( Figure 1F ) . The deletion of 25 bp results in a frameshift , which creates a termination codon after six aberrant amino acids ( Figure 1G ) . The premature STOP deletes 282 of the 2552 amino acids in the predicted Zfyve26 protein resulting in a shortened protein of 2270 amino acids ( Figures 1H ) . Searching for protein motifs with the Prosite database [29] and MyHits [30] revealed a bipartite nuclear localization signal ( amino acid 714–730 ) , a serine-rich domain ( aa 1251–1342 ) and a zinc finger FYVE domain ( aa 1807–1865 ) . FYVE domains bind to the lipid phospatidylinositol-3-phosphate ( PI3P ) predominantly present on endosomes [31]–[34] . Phylogenetic analysis detected one homolog in most metazoan genomes and a similar gene in Drosophila melanogaster ( CG5270 ) , albeit no homolog in the C . elegans genome ( Figure 1I ) . Remarkably , although teleosts underwent an additional genome duplication compared to tetrapods [35] , all ten teleost genomes available at the Ensembl-database ( http://www . ensembl . org ) contain a single paralog of Suf/Spastizin , which we confirmed for zebrafish by BLAST searches with full-length Suf ( data not shown ) . Alignment of the vertebrate protein sequences discovered a highly conserved C-terminus , which we termed Suf-domain . Scanning the zebrafish genome with the Suf-domain did not detect other proteins with a similar amino acid motif . This motif is also conserved in plants , but the protein does not contain a FYVE-domain ( data not shown ) . In human tissue culture cells , this domain in SPASTIZIN interacts with BECLIN1 , KIF13A and TTC19 [21] , [36] and is predicted to form alpha-helical solenoids as found in Clathrin heavy chain [27] . Since the Suf domain is deleted in the p96re allele , it is essential for Suf function during zebrafish oogenesis . Together , these data identified souffle as the zebrafish homolog of SPASTIZIN [20] . HSP patients require SPASTIZIN in the nervous system [20] and similarly , a recent mouse k . o . shows an adult onset neurodegeneration but not an embryological defect [22] . In zebrafish , inhibiting Spastizin by injection of morpholino-oligonucleotides ( MO ) leads to embryogenesis defects such as a twisted tails [37] . A fraction of these embryos without morphological changes show motor axon outgrowth failure leading to reduced motility , but are not paralyzed . Unexpectedly , we did not detect a neurological phenotype in zygotic zebrafish suf −/− embryos such as a touch response defect or a failure in motor axon outgrowth ( data not shown ) . However , upon MO-injection , we confirmed the previously reported severe phenotypes ( Figure S1 ) . Hence , we investigated whether the suf/spastizin gene is expressed at the appropriate time to control oogenesis in zebrafish . Real-time PCR analysis of isolated follicles at selected stages showed that suf/spastizin mRNA is expressed during oogenesis with a slight increase at the onset of vitellogenesis ( stage II , for staging see Figure 1A ) and another increase after ovulation ( stage V ) ( Figure 2A ) . This expression profile is consistent with a role of suf/spastizin during zebrafish oogenesis . During embryogenesis , suf/spastizin mRNA decreased after 4 hpf ( hours post fertilization ) similar to other maternal genes ( Figure 2B ) . At 30 hpf suf/spastizin showed a small peak of expression . Later during larval stages , the expression steadily increases consistent with microarray data in the Espresso database ( http://zf-espresso . tuebingen . mpg . de; Unigene ID: Dr . 21642 ) . To analyze sex-specific expression of suf/spastizin , we compared mRNA levels in whole females , females without ovaries and males . Contrary to its specific phenotype in the oocyte , suf/spastizin was strongly expressed in males and even outside the female germline suggesting that it also acts in somatic cells ( Figure 2C ) . The higher expression of Suf in males can be explained by two non-exclusive reasons . A simple explanation might be technical , i . e . the genes gapdh and odc1 , which we used to normalize mRNA levels , may be differentially expressed between males and females as reported for ef1α [38] . Alternatively , Suf is indeed higher expressed in males , possibly in one of the non-reproductive organs with sex-specific gene expression such as the brain or the liver [39] . However , the p96re allele clearly demonstrates that Suf is required in the oocyte , but does not exclude that it has a critical role in other organs , which we did not notice . Since we did not observe a mutant phenotype outside the germline , we addressed whether the p96re mutation causes a complete loss-of-function null-allele or forms a hypomorph . Comparison of suf/spastizin mRNA levels between +/+ , +/− and −/− ovaries showed a strong reduction after loss of one suf/spastizin copy in heterozygous adults , but no phenotype ( Figure 2D ) . By contrast , −/− mutant females showed only a minor reduction of mRNA compared to +/− heterozygotes , but eggs displayed the mutant phenotype . To examine whether an alternatively spliced suf/spastizin mRNA hides a potential zygotic mutant phenotype in other tissues or in males , we analyzed the expression of exon 35 carrying the mutation during zebrafish embryogenesis and oogenesis . We did not observe a shorter transcript lacking exon 35 , which would generate a 413 bp product ( Figures 2E and S2A ) . However , in hetero- or homozygotes we detected the predicted 25 bp shorter transcript consistent with the mutant splice donor . Since suf/spastizin expression at the mRNA level was not completely eliminated , the residual protein might have sufficient activity to compensate for Suf/Spastizin requirement in somatic cells . To analyze Suf/Spastizin protein expression , we generated an antibody against the zebrafish protein ( Figure S2B ) and compared mutant and wt oocytes in whole-mount immuno-stainings . Although mRNA levels were strongly reduced in mutant ovaries , Suf/Spastizin protein still sufficiently accumulated during oogenesis and localized to the membrane and cytoplasmic vesicles in wt as well as mutant oocytes ( Figure 2F ) . Taken together , Suf/Spastizin was expressed during zebrafish oogenesis , but was also present outside of the germline . In tissue culture cells , the FYVE domain of Suf/Spastizin interacts with the endosomal lipid PI3P indicating a role in endosomal trafficking [21] . Moreover , in human and mouse cells Spastizin binds to the novel AP5 complex regulating endosomal transport [22] , [26] , [27] , [40] . To examine genetically in zebrafish oocytes , whether Suf/Spastizin is involved in endocytosis during oogenesis , we compared endosomal compartments between wt and mutant . The gross morphology of oocyte vesicles showed no difference in early endosomes ( Rab5 ) or late endosomes ( Rab7 ) ( Figure 3A ) [41] , [42] . In contrast , Rab11b-positive recycling endosomes showed a remarkable transformation of their tubular shape in wt to patches accumulating below the nuclei of the surrounding follicle cells in mutant oocytes ( Figure 3A ) [43] . To quantify the defect in suf/spastizin mutants , we counted Rab positive foci in optical sections in deeper layers of the oocyte cytoplasm about 50 µm below the cortex , where single , small vesicles are easier to discriminate than the large compartments at the cortex ( Figure 3B , C ) . Rab5 ( early endosomes ) showed no significant change ( 1 . 09 fold; p = 0 . 35; Figures 3C and S3A ) , whereas Rab7 positive endosomes increased moderately ( 1 . 78 fold; p = 0 . 0001; Figures 3C and S3B ) . However , Rab11 staining increased dramatically ( 3 . 82 fold; p = 0 . 0001; Figures 3C and S3C ) confirming the initial observation that Suf/Spastizin controls trafficking of recycling endosomes in zebrafish oocytes . Consistent with previous reports in cell culture and mouse mutants [22] , [25] , we also discovered in zebrafish that smaller lysosomes accumulated in suf/spastizin oocytes , possibly preventing yolk proteolysis in the egg and causing the opaque cytoplasm phenotype ( Figure 3B ) . To functionally analyze during oogenesis whether the observed accumulation of the compartment-specific Rab proteins reflects a defect in the corresponding transport route as described for tissue culture cells , we established cargo trafficking assays in the zebrafish oocyte . LDL follows the degradative transport route to lysosomes and the yolk-receptor belongs to the LDL-receptor superfamily [44] , [45] . Adding fluorescent LDL to the culture medium labeled yolk globules in zebrafish oocytes , which correspond to lysosomes of somatic cells [46] . However , we observed no difference in the LDL transport to wt or mutant lysosomes , which were also identified by a characteristic black halo after fixation ( Figure 3D , also visible in Figure 3B ) suggesting that transport along the degradative pathway is not disrupted . Transferrin follows the recycling route [47]–[49] . The zebrafish Transferrin-receptor is maternally expressed and the Transferrin recycling assay was previously also applied to Xenopus oocytes [50] , [51] . Notably , contrary to the accumulation of Rab11 vesicles , fluorescent Transferrin did not accumulate in suf/spastizin oocytes ( Figure 3D ) . More interestingly , we found rare overlap of Transferrin with Rab11 suggesting that the Rab11b antibody and Transferrin label different compartments in zebrafish oocytes . To analyze whether Suf/Spastizin directly controls the trafficking of endosomes or whether the accumulation of Rab11b-positive vesicles is a secondary defect , we investigated Suf/Spastizin and Rab colocalization in the oocyte . Although early endosomes did not show a defect in suf/spastizin mutants , we also discovered colocalization of Suf/Spastizin with Rab5 as shown in the mouse [22] , but not with Rab7 ( Figures 3E , S4A and S4B ) . Although Rab11b and Suf/Spastizin mostly overlapped in their localization , some Rab11b vesicles were negative for Suf/Spastizin ( Figures 3E and S4C ) . This finding suggests that Suf/Spastizin might not be involved in all processes regulated by Rab11b in the zebrafish oocyte . In summary , the loss of Suf leads to an accumulation of Rab11b positive vesicles , which in zebrafish oocytes are not involved in Transferrin recycling . Since recycling was not disrupted in suf oocytes , we examined additional processes requiring recycling endosomes during oogenesis . Endosomal recycling is involved in meiotic maturation in C . elegans [52] and Xenopus laevis [50] , which in teleost oocytes is induced by the progestin 17α , 20β-dihydroxy-4-pregnen-3-one ( DHP ) [53]–[55] . Staining of the germinal vesicle with fluorescent phalloidin revealed that suf/spastizin oocytes underwent GVBD ( germinal vesicle breakdown ) in response to DHP ( 93% , n = 28 ) similar to wild-type oocytes ( 82% , n = 17 ) , whereas ethanol-treated control-oocytes retained their germinal vesicle ( suf: 64% , n = 11; wt: 57% , n = 28 ) ( Figure 4A , B ) . This finding was also confirmed by following the dynamics of GVBD in vivo with a transgenic H2A-GFP ( Histone2A-GFP ) reporter line [56] ( Figure 4C ) . We also examined polar body extrusion ( Figure 4D ) and spindle formation ( Figure 4E ) during oocyte maturation , but observed no difference between wt and mutant . These results show that Suf is not required for meiotic maturation in the zebrafish oocyte . In eukaryotes , recycling vesicles and Rab11 are involved in cytokinesis during mitosis [reviewed in 57] , [58]–[60] . Consistently , after RNAi depletion of the Suf-homolog FYVE-Cent in HeLa cells , a cytokinesis defect was observed [21] . To analyze cytokinesis in embryonic cells with defective maternal Suf protein , we examined embryos from suf mutant mothers . We labeled the cell cortex of 32-cell embryos with Phalloidin and their nuclei with DAPI ( Figure 4F ) . Whereas embryos from wt mothers showed one nucleus per cell , age-matched embryos from suf mutants exhibited multinucleated cells . This result suggests that the maternally controlled cell cycles require Suf/Spastizin similar to the cytokinesis defect discovered in HeLa cells . However , these results did not conclusively provide evidence for a role of Suf in endosomal recycling of zebrafish oocytes . Cell culture studies implicate Rab11 endosomes in additional transport processes besides recycling [reviewed in 61] e . g . Rab11 localizes on secretory vesicles of mammalian cells [62] . Yeast , tissue culture cells and C . elegans oocytes require Rab11 for exocytosis of secretory vesicles [63]–[67] , which are also labeled by Rab11b [43] . Moreover in epithelial cells , Rab11a and −11b localize to distinct compartments [68] . In oocytes of many organisms including humans , secretory vesicles are also designated cortical granules [reviewed in 69] , [70] . They are most similar to large , dense-core vesicles found in secretory cells in humans , e . g . neurons or pancreatic β-cells [71]–[73] . The most prominent cargoes of cortical granules are carbohydrates , which after secretion increase their volume by hydration leading to chorion elevation and thereby create the perivitelline space between oocyte and chorion . As in mammals , cortical granules are secreted after fertilization during a process termed “cortical reaction” and the induced chorion elevation is important to inhibit lethal polyspermy and mechanical damage to the embryo [reviewed in 69] , [70] . To determine in fish oocytes , whether Rab11b marks secretory granules in fish oocytes , we double-labeled them with the cortical granule marker MPA-lectin [74] . Rab11b colocalized with MPA-lectin identifying the Rab11 vesicles of the zebrafish oocyte as secretory , cortical granules ( Figure 5A , A′ ) . Moreover , Suf/Spastizin protein also colocalized with MPA on cortical granules ( Figure 5B , B′ ) . Since not all MPA-vesicles were positive for Rab11b or Suf , we hypothesized that colabeling is only observed at the vesicle surface . In contrast , if the optical section is more central in the vesicle , the green Rab11b or Suf-signal was encompassing the luminal MPA-positive cargo . Since Suf was highly enriched in the cortex and hence , colocalization might be caused by protein abundance , we analyzed optical sections in deeper layers of the oocyte with less Suf signal ( Figure 5C–E ) . Indeed , Rab11b and Suf mostly localized outside the granule-lumen ( Figure 5C , D ) . Higher magnifications showed that Rab11 in general stains the entire surface of the vesicle ( Figure 5E , upper row ) , whereas Suf was also found to be restricted to extraluminal microdomains ( Figure 5E , lower row ) . Taken together , the colocalization of Suf with Rab11b and MPA-lectin supports a role of Suf/Spastizin in the formation of cortical granules during zebrafish oogenesis . To analyze the role of Suf/Spastizin in cortical granule formation during oogenesis , we compared wt and mutant oocytes using electron microscopy . More cortical granules were visible in suf/spastizin oocytes ( Figures 6A and S5A ) . Interestingly , their electron dense core was not visible , which is a remarkably strong phenotype compared to other factors involved in dense-core vesicle maturation [75] . We also noted more and smaller lysosomes consistent with the results of the lysotracker staining . To verify an increase in cortical granules , we stained oocytes with MPA-lectin [74] . Wild-type zebrafish oocytes showed a few , large cortical granules in the cytoplasm and numerous , smaller vesicles at the cortex ( Figures 6B and S5B ) . Conversely , mutant oocytes appeared filled with MPA-lectin indicating that Suf/Spastizin is involved in cortical granule formation during oogenesis . The observed additional granules in zebrafish mutants could be generated by two alternative mechanisms . Either Suf/Spastizin could suppress the formation of granules , which in the mutant leads to an increase of mature , fusogenic vesicles . Alternatively , Suf/Spastizin could regulate the sorting of the MPA epitope , which in the mutant oocyte leads to an increased number of MPA-positive vesicles . At egg activation , cortical granules released their contents during exocytosis leading to chorion elevation ( Figure 6C ) [76] . Therefore , the first alternative would lead to a stronger chorion-elevation after exocytosis . However , after triggering exocytosis in wt and suf/spastizin stage V eggs by H2O exposure , we observed no chorion elevation 10 min after egg activation in mutants ( Figure 1B ) . Furthermore , in 30 mpf ( min post fertilization ) embryos from mutant mothers the chorion was weakly elevated ( Figure 6E ) , although the regular exocytosis process is completed within six minutes after sperm entry [77] . We quantified chorion and embryo diameters to exclude growth defects during oogenesis as a cause for the size differences . Embryo size in wt and mutants were similar , while chorion diameters were reduced in embryos from mutant mothers , which was also apparent in eggs activated without sperm , supporting the idea that chorion elevation is impaired ( Figure S5C–G ) . These results indicate that Suf/Spastizin controls the sorting of MPA-positive cargo , whose failure leads to the accumulation of immature vesicles in the mutant egg . This hypothesis predicts that immature cortical granules in the mutant egg do not secrete their cargo . Indeed , after activation , mutant eggs still contained numerous cortical granules , whereas wt lost their vesicles close to the cortex with a few left in the inner cytoplasm ( compare Figures 6B and 6D ) . To directly visualize the fusion of the vesicles , we analyzed the kinetics of exocytosis by labeling cortical actin [74]: 60 s after activation fusing cortical granules generate negatively stained crypts in the wt actin meshwork , whereas suf mutants show no vesicle exocytosis ( Figure 6F ) ; and 180 s after activation the collapsing crypts form scars of accumulating f-actin in wt oocytes , while the cortex of suf mutants did not change . These experiments demonstrate that cortical granules of suf mutants are not fusogenic and thus , support the hypothesis that their maturation is controlled by Suf . In cell culture , immature secretory granules remove specific SNARE proteins from their cytoplasmic surface to acquire the competence to fuse with the plasma membrane [78]–[80] . For instance , depletion of GGA3 in neuroendocrine cells inhibits sorting of VAMP4 away from neuroendocrine secretory granules , which needs to be removed to permit vesicle fusion with the membrane [81] . When we analyzed VAMP4 in zebrafish oocytes , suf/spastizin mutants accumulated this SNARE protein compared to wt ( Figure 6G ) . These results show that in zebrafish oocytes Suf/Spastizin is essential for the maturation of fusion-competent cortical granules . Since chorion elevation provided a sensitive read-out for Suf functionality in zebrafish oocytes , we determined whether wt suf rescues the mutant phenotype . However , the length of the suf gene with almost 8 kb made it problematic to obtain sufficient in vitro transcribed RNA for injection and hence , we injected plasmid DNA into oocytes [82]–[84] . Furthermore , we used stage III oocytes , since they can be incubated for longer periods to allow for protein expression than matured stage V eggs . Moreover , zebrafish stage III oocytes also show spontaneous chorion expansion [82] . After 12 h , the majority of wt oocytes elevated their chorion ( 84 . 3%±6 . 9 ) , whereas suf mutant oocytes rarely showed a perivitelline space ( 7 . 3%±2 . 9 ) ( Figures 6H and S6A ) also after injection of control plasmid ( Figure S6B ) . Wt Suf plasmid partially rescued chorion expansion in mutant oocytes ( 58 . 5%±19 . 0 ) , but the chorion was not elevated to the same extend as in wt ( Figures 6I and S6A ) . Interestingly , when we overexpressed the p96re allele , we also observed partial rescue , but at a lower rate ( 40 . 0%±13 . 4 ) ( Figures 6I and S6A ) . To confirm the rescue with molecular markers , we stained injected oocytes with MPA-lectin and VAMP4 . Both markers were reduced in mutant oocytes after injection of plasmid encoding wt Suf or Sufp96re , but not after injection of control plasmid ( Figure S6B ) . This result confirms our previous hypothesis that the zebrafish p96re allele encodes a hypomorph with reduced activity . Taken together these data demonstrate that Suf controls secretory vesicle maturation probably via sorting during zebrafish oogenesis . In cell culture experiments , VAMP4 is removed from immature secretory granules after it is sorted into Clathrin-coated buds , which finally pinch off [78] , [85] , [86] . The localization of Suf to compartmental microdomains also supports a role for Suf in sorting . Consistent with our results , tissue culture experiments previously proposed a role for Suf in sorting [27] , which is also a prerequisite for vesicle abscission . To analyze in zebrafish oocytes whether suf/spastizin mutants show a defect in vesicle abscission , we returned to the EM analysis and investigated mutants at higher magnification . Interestingly , close to the cortex of mutant oocytes we discovered compartments covered with Clathrin-coated buds , which appeared to not complete budding and abscission ( Figures 7A ( white arrowheads ) and S7A ) . In wt oocytes these prominent compartments with Clathrin-buds were never observed suggesting that Suf/Spastizin controls a step such as sorting to initiate vesicle budding and fission . To confirm that Clathrin-coated buds accumulate , we compared the localization of Clathrin in wt and mutant oocytes . Clathrin accumulated in mutants corroborating the EM data ( Figures 7B , D and S7B , D ) . The compartments accumulating Clathrin in mutant oocytes looked similar to cisternae formed in the temperature-sensitive Dynamin mutant shibire in Drosophila [87] . Dynamin exerts the ultimate step after sorting and budding during the fission process [88]–[90] . Indeed , we observed an accumulation of Dynamin on cortical granules in zebrafish suf/spastizin mutants suggesting that Suf/Spastizin controls a molecular step , which permits Dynamin mediated fission in oocytes ( Figures 7C , D and S7C , D ) . If fission by Dynamin is essential for cortical granule maturation , treatment of zebrafish oocytes with the Dynamin-specific inhibitor Dynasore should mimic the mutant phenotype as long as Suf is required genetically upstream of vesicle fission [91] . Interestingly , wt oocytes accumulated the cortical granule marker MPA similar to mutants after Dynasore treatment ( Figures 7E , F and S8A–F ) . However , we also noted differences such as mature cortical granules in Dynasore-treated wt oocytes , which probably formed during oogenesis before the drug treatment ( Figure S8F ) . Although Dynamin is involved in additional processes as revealed by the Dynasore inhibition , these data confirm that Dynamin mediated fission is required during maturation of secretory granules in the zebrafish egg . Cortical granule formation was previously considered to be an ongoing process during oogenesis , but whether it continued after ovulation remained unclear [70] . To analyze whether cortical granule maturation still continues in ovulated , fully matured stage V eggs , we treated them for 5 , 15 and 30 min with Dynasore and examined chorion elevation . Indeed , the level of chorion elevation corresponded well with the duration of Dynasore treatment ( Figures 7G and S8G ) . Remarkably , the 30 min treatment with Dynasore inhibited chorion elevation completely . Most importantly , Dynasore also reverted the transparency of the egg cytoplasm back to opaqueness similar to suf mutant eggs , which we confirmed by MPA-Lectin and VAMP4 staining ( Figures 7G and S8H ) . The transparency of the egg cytoplasm is caused by changes in the crystal structure of yolk globules , which are considered dormant lysosomes [92]–[94] . Theses intriguing results suggest that cortical granule maturation is critical for the function of lysosomal yolk globules during all stages of oogenesis . In summary , our results show that Suf/Spastizin is a key regulator of secretory vesicle maturation during zebrafish oogenesis probably before the fission of Clathrin-coated buds through Dynamin ( Figure 7H ) .
Here we take advantage of zebrafish genetics and its oocytes with their high vesicle trafficking activity to describe a novel role for the SPASTIZIN homolog Souffle . We show that during zebrafish oogenesis Suf/Spastizin is essential for the maturation of secretory granules . Suf/Spastizin colocalizes with Rab11b to an intermediate compartment during the formation of dense-core vesicles . During this process , Suf/Spastizin is necessary for fission from immature secretory granules in zebrafish oocytes . A striking phenotype in suf oocytes is the accumulation of Rab11b positive vesicles . However , in fibroblasts it is unusual that Transferrin does not colocalize with Rab11b and consistently , none of the other phenotypes besides the defect in cytokinesis supports a role for Suf/Spastizin in endosomal recycling . Interestingly , in certain human cell types , secretory cargoes also pass through Rab11-positive endosomes [reviewed in 61] . Moreover , polarized tissue culture cells spatially and functionally separate Rab11a/Transferrin and Rab11b [68] , [95] . This remarkable separation of Rab11a and −11b is mostly observed in polarized epithelial cells , which form the Rab11-positive subapical compartment ( SAC ) or common endosome as a central sorting hub for recycling [reviewed in 96] . Although the zebrafish oocyte is highly polarized along the animal-vegetal axis , polarity at the level of vesicle transport has not been previously described . Therefore , the Rab11b positive compartment in the zebrafish oocyte requires further analysis to confirm its homology to the subapical compartment of human epithelial cells . Another study performed in rat neuroendocrine PC12 cells implicated Rab11b in the formation of dense-core vesicles [64] . This report showed that Rab11b , but not Rab11a and Rab25 of the Rab11 protein family , are involved in secretion . Moreover , Rab11b was suggested to control the sorting of secretory cargo in neuroendocrine cells . This result in rat tissue culture also provides a molecular mechanism for Suf/Spastizin function most consistent with our results in the zebrafish oocyte . Interestingly , when they repeated their experiments in unpolarized human fibroblasts , e . g . HeLa cells , Rab11b showed different effects [64] . This cell-type dependent role of Rab11b in mammals might also explain why the Spastizin homolog was not implicated in regulated secretion before our study , since it was mostly analyzed in unpolarized human fibroblasts [21] , [26] , [27] . It remains to be determined to which degree our results in zebrafish oocytes are comparable to those in mammalian tissue culture experiments . The data from human fibroblasts showed that Suf/Spastizin interacts with the novel AP5 complex during the formation of multivesicular-bodies ( MVB ) and that knock-down of AP5 generates empty MVBs [26] , [27] . Consistently , the mouse k . o . demonstrates genetically that Spastizin is critical in vivo during this process , which leads to the formation of lysosomes . In zebrafish , we also detected fragmented lysosomes in suf/spastizin mutant oocytes . Moreover , mutations in AP5 and SPASTIZIN cause HSP in humans confirming genetically that both proteins are involved in the same process [20] , [24] . In addition , Suf/Spastizin also binds to Kif13a [21] , which was implicated by tissue culture overexpression experiments to regulate secretion [97] . Since all these studies are carried out in different organisms and different tissues , it is difficult to extract the precise function of Suf/Spastizin . A major question arising from our study is how the various observed phenotypes could be reconciled . Two alternative scenarios integrate all results: Suf/Spastizin primarily acts in MVB/lysosomes and the maturation defect of secretory vesicles is secondary e . g . through retrograde transport to the Trans-Golgi-Network [reviewed in 98] or alternatively , Suf/Spastizin primarily acts in immature secretory granules and the lysosomal defect is secondary e . g . through sorting of the mannose-6-phosphate receptors transporting hydrolytic enzymes into lysosomes [reviewed in 71] , [99] . Currently , we favor the second model , which is supported by proteins such as AP-3 , which play a role in dense core vesicle formation [100] and lysosomal maturation [reviewed in 101] . By contrast , defects in retrograde transport were hitherto not reported to affect the formation of the dense core in secretory granules as in the suf mutant , but rather lead to missorting of lysosomal enzymes [reviewed in 98] . This model would also predict that cargo sorted away in immature secretory granules is necessary for homotypic lysosome fusion ( Figure 7H ) as described in other cells [102] . These open questions make clear that additional studies are necessary to determine the precise role of Suf/Spastizin and whether a defect in lysosome formation or secretion can lead to neuronal degeneration in HSP patients . In humans , Suf/Spastizin encodes one of the more than 50 loci involved in the neurodegenerative disorder HSP controlling diverse cellular processes [18] . However , regulated secretion was so far not considered . Interestingly , defects in regulated secretion as observed in the zebrafish oocyte may explain some of the HSP symptoms . Cortical granules in the zebrafish oocyte appear identical to large dense-core vesicles ( DCV ) in neurons [71]–[73] , [103] , which store neurotrophic factors at the synapse [104] . Neurotrophic factors are responsible for the dynamics and maintenance of synaptic connections during long-term potentiation [105] , [106] . In contrast to neurotransmitter vesicles , DCVs need to be transported from the cell body to the synapse . This transport model would resolve why in HSP preferentially the longest axons degenerate from their synapse . In addition , a defect in long-term potentiation would explain why HSP neurons are not maintained and the symptoms become apparent in juveniles and adults , but not in embryos . Investigating the molecular network regulated by Suf/Spastizin in zebrafish oocytes as well as transferring these results to the nervous system as described for Atlastin ( SPG3A ) in zebrafish [42] and Drosophila [107] could provide novel insights into the biochemical etiology of HSP and bring us closer to a therapeutic treatment for patients .
Fish were maintained as described [108] in accordance with regulations of the Georg-August University Goettingen , Germany . Oocytes were dissected and cultured as previously described [83] and their stage indicated with Roman numerals as published [11] . For in vitro maturation and plasmid injection , stage III oocytes were manually dissected and incubated for 1 hr in 60% L-15 medium . After removing damaged oocytes , meiosis resumption was induced by a 90 min-exposure to 1 µg/ml DHP or EtOH as carrier control . For plasmid injection , stage III oocytes were injected with 1 ng of pCS2+wt-suf or pCS2+mut-suf encoding the p96re allele as described [84] . After injection oocytes were incubated for 12 h at 28°C in 90% L-15 medium ( 0 . 5% BSA; 100 µg/ml Gentamycin ) and then scored for chorion elevation [82] . Suf/Spastizin morpholino injections were performed as previously described [37] . For the trafficking assay , stage III oocytes were incubated with 125 µg/ml Transferrin- Alexa594 ( Molecular Probes ) in OR2 buffer [108] for 10 and 25 min followed by 30 min of chasing in Transferrin-free OR2 buffer ( recycling ) or with 10 µg/ml of LDL Dil ( Molecular Probes ) in OR2 buffer for 10 min ( degradation ) . Then , oocytes were fixed with MEMFA ( 1 M MOPS pH 7 . 4 , 20 mM EGTA , 10 mM MgSO4 , 3 . 7% formaldehyde ) after washing twice with OR2 and twice with PBT and stained with antibodies or fluorescent dyes . For the chorion elevation assay , ovulated stage V eggs were squeezed from gravid females . Eggs were activated by adding E3-medium [108] and imaged at 30 min [108] after activation . Images were used to measure chorion elevation using Fiji software [109] . For Dynasore treatment , live oocytes were collected in OR2 buffer and incubated with 500 µg/ml of Dynasore ( Sigma ) or the carrier DMSO as control for 1 hr at room temperature . Then , oocytes were washed thrice and processed for immunofluorescence staining . qRT-PCR on selected stages of oogenesis and embryogenesis was performed as previously described [83] . The described mapping position of the suf mutation on chromosome 13 [28] was used for fine mapping as previously published for the bucky ball locus [83] . The closest polymorphic markers flanking the mutation are: AL13-10-fw: 5′-GTTCCCACTCAGAGAAACAA-3′ , AL13-10-rev: 5′-GTAATGGTGGGGTTTAATGA-3′ , AL13-13-fw: 5′-TGCTTAAATTGCAGTTACAATAA-3′ , AL13-13-rev: 5′-TGAGATGCGTCTTTAAGTTG-3′ . The cDNA sequence of wt and mut souffle were submitted to gene bank with the accession numbers KC707919 and KC707920 . The souffle mutation is registered in the zebrafish Zfin-database ( zfin . org ) with the ZFIN-ID: ZDB-GENE-070117-691 , the gene as: ZDB-GENE-030131-3286 and in the zebrafish genome database at Ensembl ( www . ensembl . org ) under the gene id: ENSDARG00000040131 . Alignments were performed with CulstalW and the similarity was quantified with vector NTI ( Invitrogen ) . The phylogenic tree was constructed as described [83] with 1000 iterations . SW480 cells were transfected with 5 µg plasmid using Lipofectamine 2000 ( Invitrogen ) . Cell lysates were separated on a 6% PAA-gel and blotted for 24 h at 40 V on PVDF-membranes in blotting buffer ( 5% MeOH , 15% western salts ) . Suf antibody ( 1∶200 ) was detected with anti-rabbit HRP . Oocytes were dissected from gravid female and fixed with MEMFA buffer for 1 h at room temperature after proteinase K ( 50 µg/µl ) treatment for 3 min . Oocytes were washed thrice with PBT and blocked with PBT containing 2% BSA and 2% Horse serum for 2 h at room temperature . Subsequently , oocytes were incubated with primary antibody: Rab5 ( 1∶200 SCBT ) , Rab7 ( 1∶1000 abcam ) [42] , Rab11b ( 1∶500 GeneTex ) , Suf ( 1∶200 ) , Dynamin2 ( 1∶200 GeneTex ) , Clathrin ( 1∶200 abcam ) and VAMP4 ( 1∶200 SySy , Goettingen , Germany ) in blocking solution overnight at 4°C . Secondary antibodies were added at 1∶200 ( Alexa 488 or 594; Molecular probes ) in blocking solution overnight at 4°C . For double labeling , oocytes were incubated with Suf-antibody directly labeled with ATTO590 ( SySy , Goettingen , Germany ) for 2 h at room temperature . The Suf antibody was generated by immunizing two rabbits each with two peptides TEQVKVPAKDRNRE ( aa 187–200 ) and LNKTSTNKGMSKTD ( aa 1007–1020 ) and then purified by affinity-chromatography ( Biogenes , Berlin Germany ) . DNA was stained with 1 µM of Hoechst 33342 . Cortical granules were stained with 50 µg/ml MPA Lectin Texas Red ( EY Labs Inc . ) and lysosomes with 70 nM Lysotracker DND-99 Red ( Molecular Probes ) . Phalloidin staining was performed as described [74] . Images were captured at room temperature using a LSM780 confocal microscope ( Carl Zeiss ) with a Plan Apochromat 63×/1 . 4 NA and 25×/0 . 8 NA oil-immersion and a digital microscope camera ( M27; Carl Zeiss ) . After washing the oocytes with PBT , yolk was cleared with Murray's solution ( Benzylbenzoate ( 66% ) /Benzylalcohol ( 33% ) ) during imaging . A multiple wavelength laser was used to visualize red ( 561 ) fluorescence , green ( 488 , 405 ) and blue ( 405 ) and images were acquired and processed using ZEN 2011 software ( Carl Zeiss ) . For high-pressure freezing EM , living oocytes were placed in aluminum platelets of 150 µm depth containing 1-hexadecen [110] . The platelets were frozen using a Leica Em HPM100 high-pressure freezer ( Leica Mikrosysteme Vertrieb GmbH , Wetzlar , Germany ) . The frozen oocytes were transferred to an automatic Freeze Substitution Unit Leica EM AFS2 . The samples were substituted at −90°C in a solution containing anhydrous acetone , 0 . 1% tannic acid for 24 h and in anhydrous acetone , 2% OsO4 , 0 . 5% anhydrous glutaraldehyde ( EMS Electron Microscopical Science , Ft . Washington , USA ) for additional 8 h . After a further incubation over 20 h at −20°C , samples were warmed up to 4°C and washed with anhydrous acetone subsequently . The samples were embedded at room temperature in Agar 100 ( Epon 812 equivalent ) at 60°C over 24 h . Images were taken in a Philips CM120 electron microscope ( Philips Inc . ) using a TemCam 224A slow scan CCD camera ( TVIPS , Gauting , Germany ) . In all experiments , error bars indicate the standard deviation ( at least three independent experiments ) . The statistical significance ( p-value ) of two groups of values was calculated using a two-tailed , two-sample unequal variance t-test calculated in MS-Excel or www . graphpad . com .
|
Oocytes of egg laying animals frequently represent the biggest cell type of a species . The size of the egg is a consequence of active transport processes , e . g . the import of yolk proteins , which results in the massive storage of vesicles . In addition , secretory vesicles termed cortical granules are stored in the oocyte to be discharged right after fertilization during cortical reaction , which also occurs in mammals . Their secretion leads to chorion expansion , which prevents the lethal entry of additional sperm and protects the developing embryo against physical damage . Mutants with a defect in membrane transport are successful tools to discover genes regulating vesicle formation . We molecularly identify the disrupted gene in the recessive maternal-effect mutation souffle , which encodes a homolog of human SPASTIZIN . SPASTIZIN was previously implicated in endocytosis , but our cellular analysis of mutant oocytes connects this gene also with the regulation of cortical granule exocytosis . More precisely , we show that Suf/Spastizin is crucial for the maturation of cortical granules into secretion competent vesicles describing a novel role for this protein . Since SPASITIZN causes the disease Hereditary Spastic Paraplegia in humans , our results will help to decipher the pathogenesis of this neurodegenerative disorder .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"physiology",
"animal",
"genetics",
"vertebrates",
"animals",
"osteichthyes",
"animal",
"models",
"developmental",
"biology",
"model",
"organisms",
"organism",
"development",
"molecular",
"genetics",
"research",
"and",
"analysis",
"methods",
"fishes",
"organogenesis",
"membrane",
"trafficking",
"zebrafish",
"cell",
"biology",
"gene",
"identification",
"and",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"molecular",
"cell",
"biology",
"organisms"
] |
2014
|
Souffle/Spastizin Controls Secretory Vesicle Maturation during Zebrafish Oogenesis
|
Human infection by malarial parasites of the genus Plasmodium begins with the bite of an infected Anopheles mosquito . Current estimates place malaria mortality at over 650 , 000 individuals each year , mostly in African children . Efforts to reduce disease burden can benefit from the development of mathematical models of disease transmission . To date , however , comprehensive modeling of the parameters defining human infectivity to mosquitoes has remained elusive . Here , we describe a mechanistic within-host model of Plasmodium falciparum infection in humans and pathogen transmission to the mosquito vector . Our model incorporates the entire parasite lifecycle , including the intra-erythrocytic asexual forms responsible for disease , the onset of symptoms , the development and maturation of intra-erythrocytic gametocytes that are transmissible to Anopheles mosquitoes , and human-to-mosquito infectivity . These model components were parameterized from malaria therapy data and other studies to simulate individual infections , and the ensemble of outputs was found to reproduce the full range of patient responses to infection . Using this model , we assessed human infectivity over the course of untreated infections and examined the effects in relation to transmission intensity , expressed by the basic reproduction number R0 ( defined as the number of secondary cases produced by a single typical infection in a completely susceptible population ) . Our studies predict that net human-to-mosquito infectivity from a single non-immune individual is on average equal to 32 fully infectious days . This estimate of mean infectivity is equivalent to calculating the human component of malarial R0 . We also predict that mean daily infectivity exceeds five percent for approximately 138 days . The mechanistic framework described herein , made available as stand-alone software , will enable investigators to conduct detailed studies into theories of malaria control , including the effects of drug treatment and drug resistance on transmission .
Approximately 2 . 5 billion people live in areas whose local epidemiology permits transmission of Plasmodium falciparum , the parasite that causes the most life-threatening form of malaria [1] . Malaria has inflicted a severe toll in morbidity and mortality over the course of human history . Nonetheless , recent studies , however , document significant reductions in malaria mortality over the past decade [2] , [3] . Given these encouraging results , public health experts are planning campaigns to reduce or eliminate transmission from many areas of the world [4] , [5] . To help assess the feasibility of eliminating malaria from an area , efforts are ongoing to model and map the historical and current limits of this transmission . These models and maps also help establish a baseline to judge the success of these efforts [1] , [6] , [7] . The development of these mathematical frameworks , however , is complicated by the diversity of mosquito vectors , varying levels of human immunity , and the extent to which control efforts are applied . The development of mathematical models of malaria is contingent on a detailed understanding of the parasite lifecycle . This begins in humans when motile parasite forms , termed sporozoites , enter the body through the bite of an Anopheles mosquito and travel to the liver where they rapidly proliferate . Upon emerging from the liver , parasites then enter the blood stream as merozoites . These merozoites infect red blood cells ( RBCs ) , develop , replicate , burst from the infected cells , and repeat the cycle of asexual blood stage infection that causes disease . These asexual blood stages are able to avoid clearance in the spleen by expressing surface ligands that enable parasitized red blood cells ( PRBCs ) to adhere to endothelial cells in the microvasculature [8] , [9] . This property of cytoadherence and sequestration results from surface expression of P . falciparum erythrocyte membrane protein ( PfEMP1 ) . Because PfEMP1 presents a prominent antigenic target for the immune system , P . falciparum has evolved a sophisticated system of epigenetically-regulated antigenic variation , whereby individual parasites typically express only a single , antigenically-distinct member of the multigene family var that encodes PfEMP1 [8] , [9] . Expression continuously switches between var genes as a mechanism to continually present new epitopes that escape an already existing antibody response . Separate from the pathogenic asexual blood stages , intra-erythrocytic parasites can also differentiate into sexual stages known as gametocytes [10] . Once parasites have committed to becoming gametocytes , they sequester in the bone marrow or microvasculature and develop through four stages for 7–12 days [11] . They then reenter the circulation to complete their maturation as Stage V gametocytes . Mature Stage V male and female gametocytes are then primed to form gametes and mate in the midgut of the definitive host , the Anopheles mosquito , following blood meal ingestion . Many models of malaria have been developed to describe this cycle of transmission from the mosquito to the human host and back . These models can be broadly classified into two categories: compartmental and mechanistic . A compartmental model is any type of transmission model that simulates populations of individuals transitioning into different compartments at constant rates , with each compartment representing a different state of disease/non-disease . For example , an “SIR” model is a compartmental model in which individuals are grouped into three populations , namely susceptible ( S ) , infectious ( I ) , and recovered ( R ) . Individuals transition between compartments at a constant rate depending on several factors that include the virulence of the disease and the immune responses of hosts . More sophisticated models include additional compartments that each represent a different disease state . For example , the infective compartment may be separated into multiple compartments ( I1 , I2 , I3 , etc . ) each with different levels of infectiousness , or other compartments may be added , for example infected but not infectious hosts ( known as the E compartment ) [12] . The basic units of analysis in compartmental models are populations; the number of individuals within each disease state is tracked over time , but individuals are categorized only to the extent that they occupy one of the various compartments . In contrast , mechanistic malaria models incorporate the within-host mechanisms that determine human infectiousness over time . In such models , individual hosts are the primary units of analysis [13] , [14] . Transitioning among different levels of infectivity occurs because of individual clearance of infections , and parasite densities are modeled at the individual level . Individuals differ in multiple parameters including the intensity and duration of infection and the timing of fever . Each of these two frameworks has a useful role to play . Compartmental models benefit from simplicity , identifiably , and clarity , while mechanistic ones allow for simulations of control measures that are highly non-linear . Regardless of the model type , one of the most important mathematical quantities for theories of disease control aimed at elimination is R0 , the basic reproduction number [15] . R0 is defined as the number of secondary cases that an index case would generate in a population without previous exposure to the disease . R0 serves as a threshold criterion for transmission: if the R0 of an area is below 1 , the disease will eventually become extinct; if above 1 , the disease will spread . For malaria , R0 can be expressed as the product of the vectorial capacity ( the number of infectious mosquito bites that result from mosquitoes taking blood meals on a fully infectious human in a single day ) , the duration of the human infectious period , and the efficiency of transmission from humans to mosquitoes . Vectorial capacity can be estimated using a variety of techniques [7] , [16]–[22] . However , the human component of malaria transmission is difficult to quantify , in part , because the transmissibility of an infection is affected by many competing factors . Although a variety of mathematical models have been built to simulate the progression of malaria infections , [13] , [23]–[27] , no model has yet produced an estimate of net human infectivity over time . Here , we report a stochastic , mechanistic , within-host model that simulates the progression of Plasmodium falciparum infections and human-to-mosquito infectivity . We built upon previously published work by Molineaux and Dietz , who first developed the asexual and gametocyte components of our model from malaria therapy data , in which individuals with tertiary syphilis were infected with P . falciparum to induce a fever and clear the syphilitic bacteria [23]–[25] . This framework has been used to simulate the effects of vaccines on transmission [13] , [23]–[26] . However , this earlier work required that parameters be fitted to an individual patient's case history before simulation . We have extended these earlier studies by choosing stochastic distributions for model parameters , thereby allowing for within-host simulations that generate an ensemble of infection dynamics that are consistent with observed malaria therapy infections . We also included additional components that enable simulations of human-to-mosquito infectivity and onset of symptoms . Using this model , we have examined the levels of human-to-mosquito infectivity over time and isolated the host-related determinants of the basic reproduction number , R0 . This novel analysis of R0 made it possible to analyze overlooked aspects of transmission relevant for elimination campaigns . We calculated that net human infectivity is equivalent to 32 fully infectious days , on average . Further , we calculated the distribution of infectiousness within human populations given the natural variability of individuals' immune responses to infection , as well as the mean infectivity of a population over time . We found that infectiousness from malaria persists for a long duration of time: mean infectiousness is predicted to exceed five percent for 138 days after infection . These results were then compared to outputs from compartmental models [12] , [28] , [29] . We propose that the modeling work described herein provides the most careful estimate yet of the distribution of human responses to malaria infection and the mean human contribution to R0 . Our model also provides a framework to examine how antimalarials may affect malaria transmission , given the complexities of host-parasite dynamics .
The model used to calculate asexual parasitemia is a within-host model that simulates the course of an infection one replication cycle after merozoites have emerged from the liver . Asexual parasitemias are modeled as a system of discrete ( two-day time interval ) difference equations previously elaborated by Molineaux et al . [13] , [23] . The model simulates parasite densities in 50 different subpopulations differentiated by var gene expression type . In each replication cycle , a fixed percentage of parasites in each subpopulation switch into a different population . The switching probabilities are structured such that certain var genes are expressed with higher probability than others; immune pressure against variants also plays a role ( the switching phenomenon is described below ) . Asexual parasitemia densities are regulated by three host immune responses: an innate response that establishes an upper limit for parasite density; a PfEMP1 variant-specific response that regulates short-term periodic oscillations in density; and a variant-transcending response that causes a steady log-linear decrease in density over time , clearing the infection . We do not simulate deaths from malaria , as these are so few in proportion to the very large number of total infections as to not significantly impact overall transmission . Our model was fitted to data from malaria therapy patients , in which individuals with tertiary syphilis and with no acquired immunity to malaria were inoculated with single strains of P . falciparum in order to induce a fever and clear the infection [30] , [31] ( see details below ) . Thus , our asexuals model best reproduces the time course of asexual parasitemias in malaria-naïve adult male patients who exhibited effective immune responses . Our within-host model incorporates three types of immune responses . An innate response represents inflammation , fever , and cytokine responses to parasite replication and is a function of total parasite load , irrespective of PfEMP1 type . The two other immune responses are acquired and are dependent on antibody production . represents the PfEMP1 variant-specific immune response , with the response to each isotype denoted by the subscript i . represents the acquired PfEMP1 variant-transcending immune response . This immune response is provoked by the conserved regions of PfEMP1 ( since some PfEMP1 variants have been shown to induce cross-reactivity ) as well as conserved surface proteins ( such as MSP-1 ) and other parasite antigens . Both of the antibody responses are assumed to decay exponentially over time in the absence of new antigen production in our model . Each of these three responses has a characteristic effect on the progression of parasitemia . The innate response controls the initial densities of asexual parasitemia and is dominant during the initial period of infection . The second response , the variant-specific acquired response , controls the characteristic peaks and dips in parasitemia and interacts with the var switching structure to determine the densities of PfEMP1 variants over time . The third response is the variant-transcending acquired response , which produces the roughly log-linear decline in parasitemias over time and helps to clear the infection . We assumed that the strengths of the innate and variant-transcending immune responses vary among individuals . In our model , PfEMP1 variant densities are explicitly modeled . The parasite population is partitioned into 50 different subpopulations , each representing one antigenic isotype ( i ) . The probability that a given isotype population Pj will switch into the population Pi is given by the probability pi ( t ) , which changes over time ( and is independent of j ) . The probability pi ( t ) is designed to incorporate three aspects of var switching leading to expression of the antigenically distinct PfEMP1 proteins . First , we assume that the PfEMP1 status of parasites is reset during the mosquito stage such that infections start with a single PfEMP1 variant [34] . Second , we assume that the probability of switching into variants is not constant among the variants , but is structured such the likelihood of switching into some variants is greater than the likelihood for others [34]–[36] . This pattern may reflect the distance of the var genes from the telomeric regions or other types of inherent var structure and gene regulation [34]–[36] . Third , we assume that a PfEMP1 variant has a decreased probability of appearing if the immune system has previously mounted a response to that variant . The biological rationale for this assumption is that a prior immune response will decrease the probability of a variant appearing because parasites expressing this variant are more likely to be cleared before reaching densities detectable by smear . It is also possible that more than one variant , even most or all variants , are expressed at the onset of infection [37] , [38] . However , the innate response controls the initial phase of infection ( before antibodies are developed ) , and this response is independent of the PfEMP1 types present . During this early period of infection , there is likely to be selective pressure from the host against some isotypes , such that some isotypes are eliminated [38] . Thus , even if all variants are expressed initially , only some will survive to the period when antibodies are formed . The difference between a model in which all variants are expressed initially and our model is that the former relies entirely on variant cross-reactivity and/or immunodominance to maintain infections [39] , [40] , whereas ours relies on both cross-reactivity and the appearance of less likely variants to maintain infections . In our model , we also assume that parasites expressing different PfEMP1 variants proliferate at different rates . We assign each isotype a growth rate mi chosen from a truncated normal distribution . This assumption is based on experimental evidence that some PfEMP1 variants proliferate faster than others in vivo ( specifically , some variants that are associated with severe disease have been shown to propagate faster than those that are not ) [41] . Further , certain variants may be better adapted to a host's particular biology than others , resulting in differences in net growth rates in vivo [38] . Because we are interested in utilizing this model to simulate drug treatment in low-transmission areas , treatment-seeking behavior is an important consideration . In the absence of diagnostic testing , fever may serve as an indicator of infection for both patient and clinician [42] , [43] . To predict when fever first occurs , we utilize modeling by Dietz et al . [44] who used malaria therapy data to fit probability distributions to the onset of fever . In our model , all patients are assumed to be symptomatic and to experience a fever that begins a variable number of days before reaching maximum parasitemia . To determine the day of first fever following one cycle of replication after emergence of parasites from the liver into the blood stream ( taken as time zero ) , we use a uniform distribution based on an individual's maximum asexual parasitemia [13] , [44] . Specifically , we simulate the time course of an individual infection from inoculation to clearance and record the maximum parasitemia achieved ( denoted ) . We then take a random draw ( denoted d ) from the distribution U ( log10 ( 0 . 0002 ) , 0 ) = U ( −3 . 699 , 0 ) and calculate [13] , [44] . The first day that an individual's parasitemia is greater than or equal to is assumed to be the first day of fever for that individual . Our gametocytemia model equations were first articulated by Diebner et al . and Eichner et al . ( the two models differ slightly; we adhere to the formulation by Eichner et al . ) [24] , [25] . In our model gametocytes are produced by each wave of asexual parasitemia at a stochastic frequency determined by the function γ ( ) . Gametocytes are assumed to sequester for a variable number of days as they develop . Once the mature gametocytes emerge into the blood stream , they are cleared by the immune system , die naturally , or are transmitted to a mosquito . Our gametocyte model simulates levels of circulating ( Stage V ) gametocytes as well as numbers of gametocytes in the earlier stages ( Stages I–IV ) on a daily timescale . Gametocyte lifetimes , in the absence of immune response related to asexual parasitemia , are assumed to follow a Gompertz distribution [24] . We assume that the degree of anti-gametocyte immunity is related to the cumulative levels of previous asexual parasitemia . We do not include any suppressive effect of fever on gametocyte densities , as reported in [45] . As in the asexual model described above , the original gametocyte modeling work [24] , [25] fitted model parameters to each individual patient's malaria therapy data . We modified their model by choosing model parameters from probability distributions such that the resulting outputs matched the observed variability in the malaria therapy data . In the original Ross-Macdonald model , the infectivity of humans to mosquitoes was parameterized by a constant , c [46] . In our model we estimate the probability of human-to-mosquito transmission ( defined as production of an oocyst [47] ) as a function of gametocyte levels . For our baseline simulations , we utilize the nonlinear relationship between gametocytemia and infectivity described by Stepniewska et al . based on mosquito feeding studies on malaria therapy patients [47]–[49] . Net infectivity for an individual is quantified by taking the area under a curve generated from predicted infectivity over time; this quantity is equivalent to the number of fully infectious days . Our model of infectivity is not mechanistic in the same way that our asexual and gametocyte models are . We use the sigmoidal curve reported in [48] to force high gametocyte densities to be substantially less infectious than would be predicted by a proportional model of infectivity . We do not model why this nonlinearity occurs . Two main hypotheses could explain the reduced infectivity of gametocytes at high densities . The first is that gametocytes themselves regulate their infectiousness in a density-dependent manner [50]–[53] such that high densities are proportionally much less infectious than low densities . The second hypothesis is that host factors ( antibodies , cytokines , fever ) affect the infectivity of gametocytes [27] , [54]–[57] , though fever was not found to influence the infectivity of gametocytes in malaria therapy [58] . In our model , we do not include these possible additional factors in the calculation of human-to-mosquito infectivity; however , we did conduct a sensitivity analysis to examine the effects of different density-to-infectivity relationships on our model outputs . A final note regarding infectivity is that of Jeffery and Eyles in their original 1955 study of mosquito feedings on malaria therapy patients [47] . These authors observed that , in the first two to four days after gametocytes were observable in the bloodstream of infected patients , individuals were not infectious to mosquitoes . They attributed this phenomenon to the fact that , when gametocytes are first becoming microscopically detectable , they are immature and are thus unable to infect mosquitoes . We account for the observed non-infectivity of gametocytes appearing very early in the course of infection by adjusting modeled infectivity profiles slightly . Specifically , if the difference between the first observable asexual and sexual parasitemias was 15 days or less for a simulated individual , then this individual becomes infectious two days after gametocytes were first observed . For individuals with larger differences between asexual and gametocyte patency , or that never have an observable gametocytemia , we assume that individuals are not infectious until more than 17 days after asexual blood stage parasites are first detectable . This adjustment roughly corresponds to the feeding study data reported by Jeffery and Eyles [47] .
We have developed a mechanistic model of the progression of malaria within a human host , parameterized such that the model reproduces the median and extremes of the dynamics of infection observed in malaria therapy . For the asexuals model , we fitted five model parameters to the minimum , medians , and maxima of nine different malariometric indices derived from malaria therapy data . For the gametocyte model , we fitted five model parameters to the minima , geometric means , and maxima of three different indices derived from the gametocytemias of malaria therapy patients . Table 1 illustrates those model parameters that were changed from published reports . A mathematical formulation of the model , as well as a description of how it was fitted to data , is described in the Methods . Standalone versions of the model for Macintosh or PC platforms are provided in the Supporting Information ( see MACmodel . zip and PCmodel . zip ) , along with user manuals ( Text S1 ) and an illustration of the graphical user interface ( Figure S1 ) . Figure 1 graphically illustrates the important features of our model by presenting three individual simulations . Figure 1A illustrates the P . falciparum lifecycle for reference . Figure 1B shows simulated asexual parasite densities over time , expressed as log10 PRBC per µL of blood . The black line illustrates the lower limit of detectability by microscopy ( 10 PRBC/µL ) [45] , [61] . The individual depicted in green has patent parasitemias for a period of ∼50 days , lapses into sub-patent parasitemia for ∼60 days , then has a short period of patency before relapsing permanently into sub-patent parasitemias . The infection is completely cleared by ∼day 400 , post emergence of parasites from the liver . The characteristic peaks and dips apparent in the densities are associated with PfEMP1-based antigenic variation . The individual in purple displays three separate periods of patent parasitemia , whereas the individual in blue also has four periods , with the first lasting nearly 100 days . The inset in Figure 1B shows the first 50 days of infection along with the first fever day for each individual ( onsets of fever are indicated by triangles ) . Fever is simulated to occur on day 7 post emergence for the individual depicted in green , day 11 for the blue individual , and day 12 for the purple . Figure 1C shows the daily gametocytemias of the simulated individuals from Figure 1B . Note that the x-axis scale has been reduced from 700 to 600 for clarity . For the green individual , ∼10% of the first wave of asexual parasites converts to gametocytes . However , the later waves of asexual parasitemia have much lower asexual-to-sexual conversion probabilities , resulting in sub-patent gametocytemias after ∼day 60 and essentially no gametocytes after day 340 . The asexual-to-sexual conversion probability is chosen stochastically for each wave of asexual parasitemia for each individual according to the distribution observed from malaria therapy ( the geometric mean probability of conversion is approximately 0 . 7% ) . For the blue individual , the first asexual wave has a lower conversion probability than the second , resulting two gametocyte peaks of roughly equal height; gametocytes disappear from microscopic detection near day 140 and are completely cleared by day 600 . For the purple individual , conversion probabilities are so low that gametocytes are patent only for a very short period between days 20 and 40 post emergence and are cleared completely near day 400 . Figure 1D illustrates the daily probabilities of human-to-mosquito transmission ( i . e . the probabilities that a mosquito bite on these individuals would produce oocysts ) . The x-axis scale is now reduced from 600 to 250 days . To calculate the infectivity curves in Figure 1D , the gametocyte densities in Figure 1C were transformed using a sigmoidal relationship derived from feeding studies on malaria therapy volunteers [48] ( see section below on gametocyte densities and their relationship to human-to-mosquito infectivity ) . Net infectivity is calculated by integrating the daily human-to-mosquito infectivity curves over time ( shaded areas ) . The peaks of patent gametocytemia for the green , blue , and purple individuals in Figure 1C are clearly mirrored in Figure 1D , though the peaks of infectivity are exaggerated due to the transformation from density to infectivity . As illustrated in Figure 1 , an important feature of within-host malaria dynamics is antigenic variation . This variation is governed to a considerable extent by the nature of var gene switching leading to the expression of antigenically distinct PfEMP1 variants . In our model , we assumed that var is reset during infection so that only one variant is expressed after emergence from the liver . We then assumed that a fixed percentage of parasites switch into a new var type per replication cycle , with certain var variants more likely to appear than others . Further , we assumed that immune pressure against a given variant would reduce its likelihood of appearing . Figure 2 illustrates the var ( PfEMP1 ) expression patterns for a representative simulated individual . Figure 2A decomposes the total parasitemia over time into the various var subpopulations , such that each color corresponds to the proportion of parasitemia for each given type . Individual var types are counted as expressed only if their corresponding parasite populations reach 0 . 02 parasites per microliter , the assumed threshold for detection by polymerase chain reaction [62] . Figure 2B shows the total number of var variants expressed at any given time post emergence , and Figure 2C shows the cumulative number of var variants that have been expressed during the course of the infection ( some variants are removed by the immune response ) . This particular simulation has a maximum of 10 variants simultaneously expressed within the first few days of infection , and this level decreases over time because of immune clearance . Because the switch rates for some variants are assumed to be faster than others ( following a geometric series with a common ratio of 1/3 ) , simulations exhibit a substantial var variation early in the infection , with only a few less-favored variants appearing later . Figure 2D illustrates the total parasitemia over time , which is affected not only by the var switch rate but also by the three types of host immune response ( innate , variant-specific , and variant-transcending ) . Another important determinant of human infectivity is the assumed relationship between gametocyte density and parasite infectivity to mosquitoes ( also referred to as human-to-mosquito infectivity ) . A variety of functions relating gametocytes to infectivity have been described and proposed in the literature [48] , [56] , [63] , [64] . All of these relationships share two features: a ) infectivity increases monotonically with density , and b ) high gametocyte densities are proportionally less infectious than low densities . However , the exact shapes of the curves differ . For our best-fit parameterization , we relied upon the functional form fitted by Stepniewska et al . [48] from human feeding studies conducted from malaria therapy patient volunteers . Figure 3 illustrates this sigmoidal relationship ( in red , denoted ‘Median Infectivity , Stage V’ ) , as well as a scatterplot of density versus infectivity data from Carter and Graves [63] , [64] ( blue circles ) and from a meta-analysis by Bousema et al . [56] ( purple squares ) . Figure 4 provides a graphical illustration of two measures of model fit using the best-fit parameters . Figure 4A illustrates a measure of goodness-of-fit for our asexuals model , specifically the cumulative distributions of the durations of infection for both our model and the malaria therapy data [33] . The grey horizontal line illustrates the median durations of infection: our within-host model has a slightly shorter median duration ( 196 days ) than the malaria therapy data ( 215 days ) [33] . The slope of the cumulative distribution function from our model outputs is slightly steeper than that from the malaria therapy , indicating less variation in our modeled durations of infectivity compared to the malaria therapy data . However , the maximum durations of infectivity between model and malaria therapy are very similar . Figure 4A also shows the cumulative durations predicted by the two other models ( in pink and green; see below ) . For our model of gametocyte densities , we visually examined a total of 262 malaria therapy charts provided by Diebner et al . [24] , [65] and recorded the maximum observed gametocytemia from each patient ( data were recorded as log10 values to the nearest tenth ) . We then compared these data to the maximum gametocytemias from 1 , 000 runs of our model using the best-fit parameters . Because the Diebner et al . study only includes individuals who recorded at least four gametocyte-positive observations [24] , we censored out model runs in which gametocyte levels never exceeded 10 per µL , leaving 988 runs remaining . Figure 4B provides the empirical cumulative distributions of the durations of gametocytemia for the two data sets after log-transformation , i . e . , the proportion of data that are less than or equal to a given level of log10 gametocytemia . The malaria therapy values are slightly higher on average initially , with a median of 3 . 10 for malaria therapy versus 2 . 95 from the model ( grey horizontal line ) . Our model had a broader tail than the malaria therapy data , with more elevated gametocytemias than observed in the therapy data . The mean from the malaria therapy data was 3 . 01 , whereas the mean from the model was 2 . 98 . However , in our model , we estimated gametocytemias every day ( i . e . , we captured every maximum possible ) , as opposed to the sparser sampling of the malaria therapy data . Further , some of the individuals included in the patient charts from [24] were treated with chloroquine , chlorguanide , or quinine to terminate the infection after the initial period of continuous patent asexual parasitemia [23] . This treatment may have slightly biased downward the recorded malaria therapy maxima . Once we were able to generate malarial infections in silico that resembled malaria therapy data across a variety of indices , we then attempted to quantify the distribution of human infectivity over time . The basic reproduction number R0 is one of the most important parameters for quantifying the infectivity of a disease [15] . The classical expression for the R0 of malaria was derived by Macdonald and can be formulated with four terms [46] , [66] , [67] . Potential transmission by a mosquito population is described by its vectorial capacity , V0 , which describes the number of infectious bites that would arise from all the mosquitoes that bite one fully infectious individual on a single day . Two parameters , b and c , describe the proportion of blood meals that successfully cause an infection: b is the probability that an infected mosquito will infect an uninfected human upon biting; c is the probability than an infected human will infect an uninfected mosquito during a blood meal . In the Ross-Macdonald model , the infectious period of humans is exponentially distributed , with a daily clearance rate of r and a mean duration of infection of r−1 days . The basic reproduction number of malaria is then described by the classic formula:The Ross-Macdonald model [46] , [66] , [67] assumes that c is a constant over this period , so the ratio c/r describes the net infectiousness of a simple human infection . This net infectiousness fraction can be interpreted as the number of days that a person is fully infectious . In reality , neither V0 , b , c , nor r are constant among individuals over time and R0 is only the first moment of a complicated multivariate distribution . Consider a population of N individuals , none of whom have been previously exposed to malaria . These individuals will differ in their responses to malarial infection , including onset of first fever relative to the initiation of blood stage infection , immune responses to asexual and sexual parasite densities over time , and the time to clearance of infection . We let Di ( t ) denote the probability that individual i will infect a mosquito upon being bitten at time t; this function takes values between 0 and 1 . With our mechanistic model , one can simulate the full variability of Di ( t ) for populations with no acquired immunity . If we first consider the mean of Di ( t ) within a population using the formulathe resulting function D ( t ) is a function of time only . We call this function the mean human infectivity over time . Mean human infectivity is an important function for elimination in many contexts . Calculation of D ( t ) allows for a determination of how likely malaria will be able to persist through droughts or intensive antimalarial campaigns . The function D ( t ) for our mechanistic model is shown in Figure 5 under best-fit model parameters . In Figure 5A , the simulated asexual parasitemias from 1 , 000 runs of the model are illustrated . A large diversity in responses can be observed , with asexual parasitemias differing among individuals by many orders of magnitude post emergence . These differences in asexual parasitemias are also mirrored in large differences among individuals in both gametocyte densities and human-to-mosquito infectivity over time ( not shown ) . Figure 5B illustrates the 25th and 75th percentiles of daily infectivity for these simulated individuals , as well as the mean infectivity over time ( in red ) . The mean infectivity D ( t ) is skewed due to the presence of some individuals exhibiting long-lived infectious periods . One important prediction from our model is that mean infectiousness is greater than five percent for 138 days after infection ( see Discussion ) . If we integrate Di ( t ) over time , rather than over individuals , we obtainWe call Di the distribution of net infectivity within a population . This distribution describes how individuals vary in infectiousness given the natural variability in host-parasite interactions . Our model-predicted Di is shown as a violin plot in Figure 5C . The infectivity of most individuals is clustered around the mean value ( 32 fully infectious days ) ; however , there are an appreciable number of individuals who are predicted to be much more infectious than the mean individual . The maximum observed infectivity is 125 . 2 fully infectious days . If we integrate either the mean human infectivity over time D ( t ) with respect to t , or the distribution of net infectivity Di over a population , we arrive at what we call the mean net human infectivity , D . The quantity D was first described in the supplement to [1]; this malaria map made use of preliminary results from the model described here . D can be calculated in one of two ways:For our mechanistic model D ranges between approximately 31–34 when averaged over a population of 1 , 000 individuals ( the mean of 5 , 000 runs was 32 . 4 ) . The units of D can be considered as fully infectious days , i . e . , the number of days in which an individual has a probability of 1 of infecting a mosquito . This value represents the human contribution to R0 , and we note here that D is invariant across time , space and ecological setting . Once we had computed Di , D ( t ) , and D , we then compared our calculations to values imputed from three other models: those of Lawpoolsri et al . [29] , Okell et al . [12] , or Dietz et al . ( known as the ‘Garki model’ ) [28] . The former two models were designed to simulate the effectiveness of antimalarials at reducing malaria transmission and are the focus of our comparisons . The model of Lawpoolsri et al . was fitted to data from a low-transmission region of Thailand ( PfPR∼0 . 0–1 . 5 ) [29] while the model of Okell et al . was fitted to three regions of medium intensity transmission in Tanzania . Both are compartmental models ( Lawpoolsri et al . has one infectious compartment and Okell et al . has four infectious compartments varying in infectivity and clearance rate ) , and both papers employ their models to predict the constant equilibrium prevalence in untreated and treated cases . We first compared the D ( t ) predicted from our model with those from Lawpoolsri et al . [29] and Okell et al . [12] . Lawpoolsri et al . assume that the mean rate of clearance in infectious individuals is 1/188 day−1 with a constant daily human-to-mosquito infectiousness ( c ) of 0 . 5 . In the model of Okell et al . [12] , each of the four infectious compartments in this model had different clearance rates ( 1/10 . 5 , 1/10 . 5 , 1/31 . 5 , 1/157 . 5 day−1 ) and each compartment had a different proportional infectivity ( 1 . 90 , 3 . 08 , 1 . 53 , 0 . 28 ) of the average daily infectivity c = 0 . 05 . We did not weight these durations of infectivity for age or body surface area , i . e . we calculated the unweighted D ( t ) . Figure 4 illustrates the cumulative distributions of the durations of infection and infectiousness for these two models as well that of the mechanistic model . We see that our mechanistic model matches the malaria therapy curve closely compared to the compartmental models . These latter models have significantly heaver tails , indicating that individuals are infected for longer periods of time in those models . We can derive D ( t ) for the compartmental models [12] , [29] , using the curves from Figure 4A and the c values for each compartment . Figure 6A shows D ( t ) for both of these models as well as our mechanistic model; Figure 6B illustrates the first 200 days of this function for closer inspection . We see that the model of Lawpoolsri et al . predicts that mean infectivity is above 5% for 433 days , the output from Okell et al . is above this threshold for only 45 days , and our mechanistic model output is above this value for 138 days ( or until ∼153 days after emergence of parasites from the liver , discounting the initial period when infectivity is near zero ) . The differences in D ( t ) among the models may have to do with model structure . Lawpoolsri et al . is constrained functionally by the assumption of only one infectious compartment . Okell et al . uses four infectious compartments and thus encompasses a much larger class of distributions ( the hypoexponential distributions ) for the lifetimes of infection . Further , by weighting the infectivity of each of the duration of infectiousness compartments differently , Okell et al . increase the degrees of freedom of D ( t ) , allowing them to more closely fit their target data . Further , these two models differ in the data sets being fitted: the endemicity of the regions being modeled at equilibrium in Lawpoolsri et al . are much lower than those in Okell et al . It is possible that individuals in low-endemicity areas are infectious at higher levels for longer periods than individuals in high-endemicity areas , because acquired immunity may limit the severity and density of repeated P . falciparum infections . This effect may provide a means of identifying the effects of immunity on transmission . However , we would need to fit a variety of endemic equilibria with hypoexponential models such as that of Okell et al . to test such a hypothesis . We cannot generate quantitative conclusions from comparing the models of Lawpoolsri et al and Okell et al directly , given their different model structures . Integrating over time , we find that the D values for these three models are 7 . 2 fully infectious days for the model of Okell et al . , ∼32 days using the current model , and 94 days in the model of Lawpoolsri et al . We can also compare these values to an older field-tested compartmental model , known as the ‘Garki model’ because it was fitted to data from a malaria-endemic site in Garki , Nigeria [28] . This model includes compartments for immunity such that immune individuals clear infections faster than non-immune individuals . To calculate the net human infectiousness D for this model , let V0 be the vectorial capacity of an area . For malaria , . Further , let V be the critical vectorial capacity below which transmission is unstable , i . e . , . Thus , D = 1/V . As derived in the Garki model , , where α1 is the clearance rate of infectivity , δ is the death rate , and g is the probability of becoming infected by the bite of an infected mosquito; thus [28] . Using the values derived from Garki , we find that D = 45 . 5 fully infectious days . For the Garki estimate , the values of α1 and δ were assumed and only g was fitted to data; thus essentially D itself was fitted to data as a single parameter [28] . This fitted value for D accords relatively well with the value generated by our model [28] . Given our calculations of D , we can rescale the plots of D ( t ) by multiplying each curve by a scaling factor so that models of Lawpoolsri et al . and Okell et al . share the same mean net infectivity as the mechanistic model; these results are shown in Figure 6C . Once the models are rescaled , we can see more clearly that the models of Okell et al . and the mechanistic model predict that infectiousness is cleared at very similar rates throughout the population , whereas Lawpoolsri et al . predict a much more gradual loss of infectiousness . The closeness of D ( t ) for the scaled stochastic representation of Okell et al . and the mechanistic model is somewhat surprising , although Okell et al . do parameterize some of their model parameters from malaria therapy data . In the previous section we calculated the mean responses of individuals over time for the models of Lawpoolsri et al . [29] and Okell et al . [12] . However , since these models are both compartmental , they can readily be formulated as stochastic , individual-based models by assuming that individuals are in each infectious compartment for exponentially distributed times . We thus computed the distribution of net infectiousness within a population , Di , for both models . Figure 7 compares the distributions Di for these two compartmental models to the distribution generated by our mechanistic model . As implied by the D ( t ) curves , Figure 7A illustrates that the model of Lawpoolsri et al . predicts that some individuals have very high D values , whereas the distribution Di generated by the model of Okell et al . is much more centered about its mean . If we scale the distributions Di to all have the same mean as the mechanistic model , we see that Di for Lawpoolsri et al . is still much more dispersed than the mechanistic model; however , Di for Okell et al . matches quite well to that of the mechanistic model ( Figure 7B ) . We ran a variety of sensitivity analyses by varying the model parameters and observing the changes in model output . For the asexuals model , we adjusted such that the mean duration of infection varied between 183 and 237 days ( ∼95% confidence interval as reported by Sama et al . [59] ) . We found that the net infectivity for the model varied from 29 . 9 to 37 . 4 net infectious days , versus 32 . 4 for the best-fit parameters [1] . For the gametocyte model , we examined the effects of varying the αG parameter . For our best-fit parameterization , we assumed that αG∼U ( 0 . 06 , 1 ) . If we assumed that αG followed the U ( 0 , 1 ) distribution , then the maximum average circulation time increased to 24 . 0 days ( close to the 22 recorded in malaria therapy; Table 3 ) . The average maximum circulation time was increased because the lower bound of the uniform distribution was changed from 0 . 06 to 0 . 0 , i . e . , in some individuals gametocyte age had no effect on gametocyte longevity . The average infectivity of the population was increased by a small amount to 34 . 6 using the wider bounds for αG , versus 32 . 4 for the model with αG∼U ( 0 . 06 , 1 ) . Further , the maximum number of net infectious days for αG∼U ( 0 , 1 ) was 181 . 5 , versus 125 . 2 for αG∼U ( 0 . 06 , 1 ) . Thus , αG∼U ( 0 , 1 ) produced a very heavy tail in the distribution of infectivity among individuals . Regarding the relationship between gametocyte density and human-to-mosquito infectivity , our default model outputs assumed the relationship from Stepniewska et al . as fitted from malaria therapy [48] . We also simulated the effects of assuming different types of gametocyte density to infectivity relationships . Specifically , we simulated 14 different types of possible functional relationships between gametocyte densities and infectivity ( Figure 3 ) . Our default assumption was called the ‘Median , Stage V’ relationship ( solid red line in Figure 3 ) ; we also assumed both ‘High’ and ‘Low’ Stage V relationships ( illustrated as dashed red lines in Figure 3 ) . These latter relationships were chosen to capture much of the observed variation in the Carter and Graves data [63] , [64] . Further , we ran a logistic regression through the Carter and Graves data to develop another functional relationship ( dark blue line in Figure 3 ) ; this logit fit was similar to the data reported in the meta-analysis of Bousema et al . [56] . Each of these four relationships relates observable ( Stage V ) gametocytes to infectivity , and for each of these four relationships we could apply the Jeffery-Eyles observation that gametocytes are not infectious at the onset of gametocyte appearance [47] to generate a total of eight density-to-infectivity relationships . To develop the six other possible relationships between gametocyte densities and infectivity , we utilized additional information regarding the biology of P . falciparum . Not all gametocytes that are observable are infectious; once gametocytes enter the circulation , they still need a brief number of days to mature further before becoming infectious [47] , [68] , [69] . Circulating Stage V gametocytes can be further discriminated into two categories: Stage VA gametocytes and Stage VB gametocytes [68] . Stage VA gametocytes are circulating but are not infectious; Stage VB gametocytes are both circulating and infectious . Thus we generated three additional functional relationships by assuming that observable gametocytes were infectious only after two additional days of maturation . These three relationships were designed to parallel the ‘Median , ’ ‘High , ’ and ‘Low’ relationships from above but assuming only Stage VB gametocytes are infectious; these are illustrated in light blue in Figure 3 . We then modified each of the three Stage VB assumptions by assuming that there is a short period at the beginning of infections in which gametocytes are not infectious , as above [47] , for a total of 14 possible functional relationships between gametocyte density and infectivity . The mean net infectivity values for seven of the parameterizations are 70 . 0 , 41 . 1 , 23 . 6 , 64 . 1 , 33 . 3 , 16 . 2 , and 36 . 3 net infectious days for the Stage VB , High; Stage VB , Median; Stage VB , Low; Stage V , High; Stage V , Median; Stage V , Low; and Carter & Graves parameterizations , respectively ( without the Jeffery-Eyles corrections and with mintrans = 0 ) . If we include the effects of the Jeffery-Eyles correction , these seven parameterizations yield 68 . 2 , 40 . 2 , 23 . 1 , 61 . 2 , 32 . 0 , 15 . 6 , and 35 . 0 mean net infectious days , respectively ( assuming mintrans = 0 ) . Each mean is from 1 , 000 runs . Varying the assumed relationship between gametocyte density and infectivity will also affect other aspects of transmission by altering the duration between parasite emergence and infectivity and/or the total duration of positive infectivity . Also of note , our model calculated P . falciparum infection dynamics only among adults , as there are no malaria therapy data for children and it is not well-understood how children differ in their overall levels of infectivity from adults . In a companion paper ( Johnston et al . , in prep ) we discuss how our results concerning infectivity among adults may translate to children and the implications of using our model results for malaria control planning .
Here we describe the development of a novel , stochastic , within-host model of the progression of malaria in patients with no acquired malarial immunity . This model utilizes the difference equations originally developed by Molineaux and Dietz to simulate the progression of asexual and sexual parasitemias [23]–[25] . We have parameterized these equations so that the entire range of observed responses in malaria therapy can be reproduced without needing to fit parameters to individual case histories . We also extended the modeling framework from [23]–[25] to include components for simulating the onset of first fever and human-to-mosquito transmission . Once our mechanistic model was formulated , we revisited the analytic Ross-Macdonald model to examine how human infectiousness enters into the formula for the basic reproduction number R0 . We then analyzed human infectiousness in three ways , calculating the mean human infectivity over time D ( t ) , the distribution of net infectivity Di , and the mean net human infectivity , D . We found that D in our mechanistic model is approximately 32 fully infectious days . This quantity is invariant in a population over time and plays a crucial role in determining R0 . We have utilized this value in recent malaria mapping work [1] , although a full mathematical treatment of this quantity was left until the present . Our study included a review of the mathematical literature to determine whether we could impute these quantities from other modeling work to provide a baseline for comparison . We examined the models of Lawpoolsri et al . [29] , Okell et al . [12] , and the Garki model [28] , and found them to vary widely in their calculation of D , D ( t ) and Di . We propose that our new estimate of D is the most appropriate one for R0 , because R0 assumes no acquired immunity and our model is parameterized solely from malaria therapy studies with individuals that were non-immune . The other models cannot easily disentangle the effects of acquired immunity , multiplicity of infection , and control efforts from the effects of immunity acting on a single infection , though we have described how future efforts might begin to separate these quantities . In addition to our calculation of the invariant D and its importance for R0 , we also predict that human infectiousness persists for a long period of time at levels sufficient to promote transmission in areas of high vectorial capacity . While these calculations are for populations with no acquired immunity , they are relevant for malaria elimination efforts because antimalarial immunity wanes rapidly in the absence of infection [17] , [70] . As this immunity wanes , the responses of individuals to infection can be expected to approach those observed in malaria-naïve individuals [28] , [66] . Of note , a recent study in Senegal found that persistent infectiousness prevented interruption of transmission even when incidence had been reduced to very low levels through insecticide-treated bed nets and usage of ACTs [17] . Our model confirms the relevance of persistent low-level infectiousness for elimination efforts . In addition to the usefulness of these results for mapping and control efforts , the modeling platform and analytic framework described herein will help clarify the different assumptions among malaria models . Further , because we calculate asexual and sexual parasite densities daily , and because the model reproduces the entire variability of host-parasite dynamics observed in malaria therapy , our modeling framework provides a powerful new tool for exploring the effects of antimalarial treatments on transmission . As malaria decreases worldwide , our model results will become more relevant to more regions of the world , thus helping to improve targeting of control efforts .
|
We report a new mathematical model of the progression , within a human host , of a malaria infection caused by the parasite Plasmodium falciparum . This model incorporates probability distributions for the key parameters of infection and transmission so that model outputs match the entire range of observed responses in patients , without the requirement for fitting individual data . Further , we simulate the daily densities of both the disease-causing and transmissible forms of the parasite within an individual , as well as the onset of fever and the probability of parasite transmission to mosquitoes . This model allows us to reproduce aspects of infection that are critical for malaria control modeling . As a first application , we calculate the net infectiousness of humans to mosquitoes and predict that net human infectivity from a single infection is on average equal to approximately 32 fully infectious days . This value has been used to help map the worldwide intensity of malaria transmission . We also predict that mean daily infectivity is greater than five percent for approximately 138 days . Our modeling framework , available as downloadable software , will allow researchers to probe the effects of treatment and drug resistance on malaria transmission in unprecedented detail , helping to improve malaria control efforts .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"infectious",
"diseases",
"mathematics",
"tropical",
"diseases",
"(non-neglected)",
"infectious",
"disease",
"epidemiology",
"epidemiology",
"plasmodium",
"malariae",
"applied",
"mathematics",
"population",
"biology",
"biology",
"malaria",
"parasitic",
"diseases"
] |
2013
|
Malaria's Missing Number: Calculating the Human Component of R0 by a Within-Host Mechanistic Model of Plasmodium falciparum Infection and Transmission
|
Many biological problems involve the response to multiple perturbations . Examples include response to combinations of many drugs , and the effects of combinations of many mutations . Such problems have an exponentially large space of combinations , which makes it infeasible to cover the entire space experimentally . To overcome this problem , several formulae that predict the effect of drug combinations or fitness landscape values have been proposed . These formulae use the effects of single perturbations and pairs of perturbations to predict triplets and higher order combinations . Interestingly , different formulae perform best on different datasets . Here we use Pareto optimality theory to quantitatively explain why no formula is optimal for all datasets , due to an inherent bias-variance ( noise-precision ) tradeoff . We calculate the Pareto front of log-linear formulae and find that the optimal formula depends on properties of the dataset: the typical interaction strength and the experimental noise . This study provides an approach to choose a suitable prediction formula for a given dataset , in order to best overcome the combinatorial explosion problem .
Different fields of biology ask how multiple perturbations affect a biological system . For example , to understand the function of DNA sequences such as promoters or coding regions , or to design new ones , it is important to understand how mutations combine to affect function [1–4] . Another widely studied example is how multiple drugs combine to affect cells and organisms . This question is important for developing effective combination therapy [5–9] and to reduce drug resistance [10–14] . A major challenge in these fields is the combinatorial explosion problem: the number of combinations increases exponentially with the number of perturbations . Given n different single perturbations , there are 2n possible combinations . In DNA sequences there are 4n combinations of bases so that sequences of 30bp have 1018 possible combination . Drugs present the additional dimension of doses , so that 8 drugs at 6 doses amount to 68≈106 combinations . Therefore , the number of combinations quickly outgrows experimental ability . To overcome the combinatorial explosion problem , there are two main approaches . In the case of sequences , one can use directed evolution to find sequences with desired function [15–19] . This approach is powerful and is based on exponential expansion of the sequences with highest function . However , experimental evolution still covers only a tiny fraction of sequence space and has the potential to get stuck on local optima . In the case of drugs this approach is not feasible . The other main approach is to use mathematical models to estimate the effects of combinations using only a small number of measurements . Machine learning studies use regression-like models to estimate drug and mutation effects based on a learning set of measurements [20–24] . For example [4] analyzed combinations of mutations on the lac promoter , and [25] analyzed a library of mutation in green fluorescent protein . One limitation of machine learning is that it requires extensive training data , which may exceed experimental ability when samples are rare and perturbations are costly , as in the case of drug combinations . To overcome the lack of large training datasets , another line of research establishes analytical formulae to estimate combination effects based on , for example , measurements of single perturbations and pairs . Analytical formulae can include knowledge about the biology of the system and can therefore be more effective than machine learning when data is scarce . The most common baseline model , that seems to work well as a first approximation in many cases , is Bliss independence [26] in which the effect of a pair of perturbations is the product of the single perturbation effects , sij = sisj . Bliss independence is equivalent to additivity in log-effect space . Another baseline model for drugs is Loewe ( dose additivity ) [27] , but seems to be less accurate than the Bliss approximation for high-order drug combinations [28 , 29] . Baseline models are generally inaccurate because they do not consider the interactions between perturbations . These interactions are called synergy and antagonism , in the case where the combination shows larger or smaller effect than the baseline model , respectively . Several studies have attempted to present formulae that take interactions into account , by including measurements for pairs . Wood et al . [30] introduced an Isserlis-like formula based on singles and pairs . For triplets , the formula is s123 = s1s23+s2s13+s3s12−2s1s2s3 . This formula worked well for combinations of up to 4 antibiotics . Zimmer et al [31] presented a model which used measurements of dose-response for single drugs and drug pairs to compute the dose-dependent effect of higher order combinations , with excellent accuracy on antibiotics and anti-cancer drugs . An additional formula , based only on pairs , performed well on small single-dose drug datasets [32] . Surveying these studies , it seems that there is no best formula that outperforms others on all datasets . Instead , each formula works well on the dataset it was developed on , but typically less well on other datasets . This situation suggests that , because datasets differ in their noise and interaction strengths , there may be a range of formulae to consider . There is therefore a need to compare formulae , to understand when formulae fail , and to develop ways to decide which formula to use when considering a given dataset . Here , we address these questions by studying the tradeoffs inherent in formulae for combinations . We study wide classes of formulae and test them on twelve experimental datasets for drugs and sequences , as well as on synthetically generated datasets . We find that no formula outperforms the others on all datasets . Instead , each dataset has a different optimal formula . On the other hand , many formulae are suboptimal for all datasets . We explain this result using a well-known concept from statistical learning , the bias-variance tradeoff [33–35] . Roughly speaking , good formulae should be complex or expressive enough to capture the true variability of the dataset ( low bias ) . On the other hand , formulae should be simple enough in order to avoid overfitting the noise in the dataset ( Fig 1 ) . Hence , the optimal formula for a dataset should be dependent on the typical effect size ( true variability ) of the dataset as well as the experimental noise . In order to understand this tradeoff , we use Pareto optimality [36–39] . Pareto optimality was previously used to study model selection and hyper-parameter choice in machine learning models [40 , 41] . We use it to define the optimal formula for each dataset , based on its noise and interaction strength . We suggest a method to choose the optimal formula for a new dataset .
For simplicity , we concentrate on the problem of predicting the effect of triplets of perturbations from data on the effects of pairs and single perturbations . We provide formula for the effects of k perturbations in the supporting information ( S1 Text , [29 , 42] ) . To establish notation and terminology , we use the term perturbation as a general term for drug , mutation or other type of change in the system . We define the effect as the measurable outcome of the perturbations on the system function , such as survival of cancer cells for anti-cancer drugs , growth rate of bacteria in case of antibiotics , and the activity of a promoter or a protein in the case of sequence mutations . Three different perturbations will be denoted by 1 , 2 , 3 . The value of the effect in the absence of perturbation ( wild-type ) is S∅ . The effects of single perturbations are S1 , S2 , S3 , of pairs of perturbation are S12 , S13 , S23 . The effect of the triplet perturbation , which we wish to predict given singles and pairs effect data , is S123 . For the effects normalized by the wild-type we use lower case letters sx=SxS∅ Formulae from the literature include the Bliss independence formula: s123=s1s2s3 ( 1 ) Machine learning approaches often use a regression formula: s123=s12s13s23s1s2s3 . ( 2 ) This formula results from regression where one fits the effects of singles and pairs to s = ∑iaixi+∑i , jaijxixj where xi = 0 if mutation i is absent and xi = 1 if it is present . A third formula uses only information from pairs [32]: s123=s12s13s23 . ( 3 ) These formulae belong to the class of log-linear formulae , and hence we focus on this class . The most general formula in this class , taking into account the symmetry in perturbation indices ( re-naming drugs 1 , 2 and 3 should not affect the prediction for S123 ) is: S123=S∅α ( S1S2S3 ) β ( S12S13S23 ) γ To make the calculation linear , we use the logarithm of the un-normalized effects Lx = log ( SX ) , resulting in L123=αL∅+β ( L1+L2+L3 ) +γ ( L12+L13+L23 ) The log-linear formulae thus have three parameters , α , β and γ . They include the previous formula discussed above: Bliss independence is when α = −2 , β = 1 , γ = 0 and regression is α = γ = 1 , β = −1 . We now evaluate the precision of each formula . As an operational definition of precision , we use a Taylor-series approach . We assume that the log effect is a smooth function f of multiple inner variables of the system . Each perturbation is represented by a change in one of these inner variables . Without loss of generality , we can assume that without perturbations , L∅ = f ( 0 , 0 , 0 ) . Then L1 , the log effect of perturbation 1 , is L1 = f ( x , 0 , 0 ) , for some value of x . Similarly , the other two single perturbations are L2 = f ( 0 , y , 0 ) and L3 = f ( 0 , 0 , z ) . The pair log effects are L12 = f ( x , y , 0 ) , L13 = f ( x , 0 , z ) , L23 = f ( 0 , y , z ) . To predict the triplet , we need to estimate L123 = f ( x , y , z ) . Mathematically , this is equivalent to the question of estimating a function on one vertex of a 3D box given its values on the other 7 vertices [42] . Even though in reality perturbations are sometimes not small , we will next assume that they are in order to give an operationalized and analytically solvable way to discuss precision . When the values of x , y and z are such that they represent small perturbations , one can use a Taylor expansion and ask which of the formulae are precise to which order of expansion ( no matter what the exact form of f ) . Here we will derive conditions for a formula to be precise to the 0th , 1st and 2nd orders in Taylor series . But first we explain intuitively what these precisions orders mean . Formulae precise to 0th order have the property that if all effects are equal , L∅ = Li = Lij = C the prediction for the triplet is equal to that effect: L123 = C . Formulae accurate to first order have the property that if all pairs are Bliss independent in the sense that sij = sisj , then the predicted triplet is also Bliss independent s123 = s1s2s3 . We now derive the conditions for precision to different orders . The Taylor expansion of L123 is , to first order: L123=f ( x , y , z ) =f ( 0 , 0 , 0 ) +∂f∂x ( 0 , 0 , 0 ) x+∂f∂y ( 0 , 0 , 0 ) y+∂f∂z ( 0 , 0 , 0 ) z+o ( |x| , |y| , |z| ) We equate this to the Taylor expansion of the log-linear formula: αL∅+β ( L1+L2+L3 ) +γ ( L12+L13+L23 ) ==αf ( 0 , 0 , 0 ) +β[f ( x , 0 , 0 ) +f ( 0 , y , 0 ) +f ( 0 , 0 , z ) ]+γ[f ( x , y , 0 ) +f ( x , 0 , z ) +f ( 0 , y , z ) ]== ( α+3β+3γ ) f ( 0 , 0 , 0 ) + ( β+2γ ) [∂f∂x ( 0 , 0 , 0 ) x+∂f∂y ( 0 , 0 , 0 ) y+∂f∂z ( 0 , 0 , 0 ) z]+o ( |x| , |y| , |z| ) We therefore obtain the condition for a formula to be precise to 0th order: 1=α+3β+3γ From now on , we restrict ourselves to the class of models that are precise to 0th order . We next ask which models are precise to 1st order . The condition for 1st order precision is: β+2γ=1 All the formulae on this line in beta-gamma space give exact approximation to the first order . For example , the Bliss ( β = 1 , γ = 0 ) , the regression ( β = −1 , γ = 1 ) and pairs formula ( β=0 , γ=12 ) fall on this line of first order precision . We can define the deviation from 1st order precision as follows: P1st ( α , β , γ ) = ( 1−β−2γ ) 2 We next ask which formulae are precise to the second order . We find that there is only one log-linear formula which is precise to the second order–the regression formula of Eq 2 ( S1 Text ) ( α = γ = 1 , β = −1 ) . The deviation of other formula from second-order precision can be represented by the sum of the coefficients of the second order error ( S1 Text ) : P2nd ( α , β , γ ) = ( 12−β2−γ ) 2+ ( 1−γ ) 2 The precision findings are summarized in ( Fig 2A and 2B ) . The figures plot contours of accuracy to different orders as a function of β and γ . In the plots , α is evaluated by the zero-order precision demand α = 1−3β−3γ . The plots are therefore restricted to 0th order precise formulae . It is seen that optimal first-order accuracy occurs on a line in model space which includes the Bliss and regression models , and that second-order precision has elliptical contours maximal at the regression model . If precision was the only factor at play , one would expect the regression model to outperform others . However in most real datasets this model does poorly [31] . The reason is that it is sensitive to experimental noise . To estimate the robustness to noise of different models , we model experimental noise in the measured effects , Li = Li+χi and Lij = Lij+χij , where χ are independent Gaussian noise with equal STD σ for all measurements ( similar conclusions apply to the case of non-independent noise , S1 Text ) . This corresponds to log-normal multiplicative noise for the effect measurements . Such log-normal noise is typical for experiments on drug and mutation effects [31 , 32] . Here we derive an expression for the noise in the predicted triplet effect . We must separate between two cases . Case I occurs when there is experimental noise in L∅ ( the wild-type ) , as is the typical case for sequence ( mutation ) data , so that L∅ = L∅+χ∅ . Case II is when L∅ is noiseless , as often happens for drug combinations when the effect is cell survival and L∅ = 0 by definition . To compute the variation in the prediction of a triplet s123 given the noise in the pair and single inputs , we assume independent noise for each variable . The noise ( std ) for case I depends on the three parameters of the model α , β and γ: PN , WTnoise ( α , β , γ ) =σα2+3β2+3γ2 And in case II ( noiseless L∅ ) only on the parameters β and γ: PN , WT=1 ( α , β , γ ) =σ3β2+3γ2 Note that noise is minimal when α = β = γ = 0 , a formula that always predicts 0 . This model is not precise even to 0th order . Considering only models precise to 0th order , we obtain the minima of the noise performance function in case of noisy wild type ( S1 Text ) : argmin ( PN , WTnoise ( α , β , γ ) ) = ( 17 , 17 , 17 ) Which simply averages the inputs L∅ , Li and Lij , and in case of noiseless wild-type simply taking the wild-type value S∅ as the prediction argmin ( PN , WT=1 ( α , β , γ ) ) = ( 1 , 0 , 0 ) Contours of this function in the cases of presence and absence of noise in the wild-type appear in ( Fig 2C and 2D ) . In both cases noise grows with distance from the single minimum . In order to compare models according to the two tasks , precision and noise , we use the Pareto front approach . The Pareto front is defined as the set of formula for which there is no other formula that is better at both tasks . Given the two performance functions of noise and precision , we compute the Pareto front as the set of points of external tangency of the performance contours [43 , 44] . The resulting front is a one-dimensional curve in the space of formulae ( beta-gamma space ) . In the case of first-order precision and noise robustness , the front is a straight line . In the absence of wild-type noise the Pareto front is defined by ( see S1 Text and Fig 3A ) : γ=2β Or in the presence of wild-type noise ( see S1 Text and Fig 3B ) : 5β+2γ=1 If noise and first-order precision are the only tasks faced by formulae , it is expected that all optimal formulae will fall on this line . We next computed the Pareto front where the two tasks are noise robustness and second order precision . In the case of noiseless wild-type , this give the conic defined by the equation ( see S1 Text and Fig 3C ) : −2γ2+7βγ+2β2+γ−6β=0 In the case of noisy wild-type we find ( see S1 Text and Fig 3D ) : 5β2+28βγ−22β+16γ2−20γ+5=0 It is now possible to compute the entire Pareto front which consists of optimizing the three performances together . The boundary of the Pareto front is defined by the Pareto fronts of the pairs of tasks . The entire Pareto front in the cases of noiseless and noisy wild-type is composed of two thin triangle-like shapes that meet at a vertex , as shown in Fig 4A and 4B ( black region ) . We note in passing that the typical solution for a Pareto front with three tasks resembles a single triangle with the optima for the three tasks at the three vertices[43 , 44]; the elongated two-shape pattern found here results from the fact that the optima for one task , first order precision , falls on a line and not a single point . The present approach can be applied to any class of formulae . To illustrate this we compute the Pareto front for a class of generalized mean formulae in S1 Text . In order to test the relevance of the Pareto front to real data , we compiled a set of thirteen published experimental datasets for drugs and mutations ( Table 1 ) . This includes data on the effects of drugs ( antibiotics , cancer drugs ) on cells , and the effect of mutations on proteins and organisms . The datasets include the effects of singles , pairs and triplets of perturbations . For each dataset , we scanned formulae ( scanning β and γ with α = 1−3β−3γ to provide 0th order precision ) and found the formula that gives the smallest root-mean-square error for triplet predictions . This formula , a point in the β , γ plane , is the optimal formula for that dataset . In order to control for outliers and variation in the data , we repeated this for each dataset on 30 bootstrapped datasets , in which we built a new dataset sampled from the original data with replacements . Thus , each dataset yields 30 additional optimal formula points . We find that the optimal formula for all datasets lie close to the Pareto front ( Fig 4A ) . The large datasets fall neatly on the Pareto front ( E . coli antibiotics 1 , A549 and others ) , whereas smaller datasets tend to deviate more due to their larger bootstrapping variance ( Dihydrofolate reductase , H1299 , E . coli antibiotics 2 ) . Note also that the main direction of variability of the bootstrapping distribution is parallel to the Pareto front [45] . In the presence of noise in the measured wild-type effect ( case II above ) , the datasets also fall on the Pareto front ( Fig 4B ) . In this case the datasets are larger , hence they have less variability in the bootstrapping . Here , we used an expansion trick to increase the amount of usable data from small fully-factorial datasets . In the expansion trick , we consider treatment with a single perturbation Li as wild-type L∅ . We then consider treatments with an additional second perturbation Lij as a single perturbation on the wild-type background , Li , treatments with three perturbations Lijk as the pairs Ljk , in order to predict the triplet Ljkm given by the quadruplet Lijkm in the original data ( Fig 5 ) . We also used pairs and higher order combinations as wildtype to the extent allowed by the dataset . This increases the number of triplets in the fully factorial dataset of order k from ( k2 ) to at most ( k2 ) 2k in its most extended form ( Table 1 shows both original and expanded triplet number ) . We further tested 61 datasets from the UniProbe database [46] on protein-DNA binding interactions in fully factorial datasets of 8 mutations . We use the expansion trick using 1000 randomly chosen starting point sequences as a wild-type from each fully factorial dataset . We find that the optimal formulae for these datasets all fall close to the Pareto front ( Fig 4C ) . The results are near the noise-robustness archetype , suggesting that noise is a dominant source of variation in these protein-binding microarray experiments . We see that optimal formulae for different datasets are close to the Pareto front . We next asked how the properties of the dataset affect which formula is optimal for that dataset . To do so , we generated synthetic datasets with different parameters , so that we could control the level of noise and the level of interaction strength ( deviation from the Bliss formula , see S1 Text ) , the two factors that influence the performance of the formula . To generate simulated data we used a third order polynomial f ( x , y , z ) with random coefficients , sampled at different random points , with Gaussian noise added ( which varies between datasets ) . The goal is to predict triplets from pairs and singles , that is to predict f ( x , y , z ) from the projections on axes and planes ( S1 Text ) e . g . f ( x , 0 , 0 ) , f ( x , y , 0 ) etc . The noise amplitude of each dataset is the standard deviation of the Gaussian noise added to log effect . The interaction strength ( that is synergy/antagonism ) of each dataset is given by its mean deviation from the Bliss approximation I=|sij−sisj||sisj| . To control I , we sampled the function at various distance from the origin ( S1 Text ) , where the larger x y and z , the larger the nonlinearity and hence the interaction . For each such synthetic dataset , we computed its optimal formula among the log-linear family and found that for datasets with small interaction strengths , the optimal formula falls close to the curve defining the Pareto front ( Fig 6A ) . Interestingly , when interaction strength become larger , points go a bit beyond the second order precision archetype ( Fig 6A , solid arrow ) , and when interaction strength was increased even further , points start to go back to the ( 0 , 0 ) point deviating from the Pareto front ( Fig 6A , dashed arrow ) . To see the trends described above we plotted the optimal values of γ and β as function of noise and interaction strength ( Fig 6B and 6C ) . We start by considering the region above the solid arrow ( small interaction strength ) , we see that in this region γ increases and β decreases with interaction strength . This is the expected result since larger γ and smaller β means getting closer to the second-order-precision archetype . Second-order precision becomes more important relative to noise as interaction strength gets larger , noise robustness becomes more important than second order precision when the noise in dataset is larger . These results summarize the prediction of Pareto optimality theory . Interestingly , there are simulated datasets for which the optimal formulae are beyond the second order archetype ( Fig 6A , right side ) . It was found that formulae of second order tend to approximate higher order interactions better than expected [47] . The points beyond the Pareto front are example of formulae of second order which are especially better in predicting the value of the third order function . In the region of large interaction strength ( Fig 6B and 6C over the dashed arrow and , Fig 6A dashed arrow ) , we see the opposite trend of decreasing γ and increasing β with interaction strength . The explanation of this surprising result is that formulae in this region no longer satisfy the assumption of precision to 0th order . The interaction strengths in this case are so large , such that the Taylor approximation approach no longer gives the optimal formulae . Fig 6D shows that indeed the formulae found for higher interaction strength no longer gives predictions which are accurate to the 0th order . These results indicate that one can predict the optimal formula for a dataset if one can estimate its noise and interaction strengths .
One general question is to what extent high order interactions exist in biological systems that can’t be explained by pairs . High order interactions in this context are defined as the deviation of the measurement form a null model that includes the effects of single and pair perturbations . Thus , choice of null model can affect the results . For example , a standard definition of pairwise interaction is: ϵ12=s12−s1s2 This formula is based on a Bliss independence null model for the combined single effects: s12 = s1s2 . Different studies of triplets use different null models [48 , 49] . For example , a recently study measured the effects of about 150 , 000 triple gene deletions in yeast , and compared them to single and pair deletions [49] . Third-order interactions were estimated using an Isserlis null model ( s123 = s1s23+s2s13+s3s12−2s1s2s3 ) yielding ϵ123=s123−s1s2s3−ϵ12s3−ϵ13s2−ϵ23s1 Evaluating the triplet interaction using the absolute value of ϵ123 as defined above gives a mean absolute triplet interaction of 0 . 044 . Significant triplet interactions were estimated to be about 100 times more common than significant pair interactions . We used the present approach to predict the optimal model using the noise and effect size in the pair measurements in this study . The best null model is similar to the pairs model ( Eq 3 ) , which is less noise-prone than the Isserlis model . With this null model , the mean absolute triplet effect is 23% lower .
In this study , we find that the problem of predicting the combined effects of perturbations does not have a unique optimal solution . Instead , different solutions and formulae are optimal for different datasets . We analyze the Pareto front of models that trade-off noise robustness and precision . This Pareto front of optimal formulae matches observations on the best formula for a range of real and synthetic datasets . The present study offers a way to predict the best formula based on the noise and effect size of pairs data . By measuring interaction strength based on pairs , and experimental noise using repeats , one can judge where on the Pareto front the optimal formula might lie for a given dataset . One important use of these formula is to estimate high-order effects between genes . For example , a third-order effect ϵ123 is defined by the measured effect of three perturbations minus a null model based on single and pair perturbations . The better the null model , the more accurate the estimation of the high-order effect . We find that the present approach can improve the null model used for estimating the effects of triple yeast gene deletions in a recent large scale study ( Kuzmin , 2018 ) . The improved estimation lowers the number of apparent three-gene interactions that can’t be explained by pairs . This is relevant for the design of systematic gene perturbation experiments , because it indicates that pairs may be enough to capture most of the interactions . Pair scans are much more feasible than triple-perturbation scans , suggesting an optimistic outlook for understanding complex gene interactions . This study used a Taylor expansion to define precision . Taylor series strictly apply only to small perturbations . Despite this limitation the method seems to work well for mutation and drug combination dataset . One reason for this is that higher-order effects in biological systems seem to be smaller than low order ones [29] , which is equivalent to the underlying assumption of the Taylor approximation . It would be fascinating to use the present approach to analyze additional classes of formula , and to understand the effects of multiple perturbations on additional biological systems .
Computations of the maxima of the different performance functions ( Fig 2 ) , and the Pareto fronts of multiple performance functions ( Figs 3 and 4 ) were performed analytically , and are detailed in the result section and S1 Text . All simulations and Figs 1–4 and 6 were produced using MATLAB 2017 . Evaluation metric for formulae performance was RMSE . Therefore , the coefficients of the optimal formula were computed using linear regression on a dataset ( Figs 4 and 6 ) . In Fig 6 , simulated data was generated using random symmetric polynomials of degree 3 according to the formula: f ( x , y , z ) =a0+a1 ( x+y+z ) +a2 ( x2+y2+z2 ) +a3 ( x3+y3+z3 ) +a4xyz+a5 ( xy+xz+yz ) +a6 ( x2y+y2x+x2z+z2x+y2z+z2y ) Where ai were sampled randomly and uniformly between 0 and 1 . To get the simulated dataset such random formulae were evaluated at random points in the box [0 , ϵ]×[0 , ϵ]×[0 , ϵ] . The approximation distance ϵ varied logarithmically between [0 . 06 , 0 . 06∙29] . To the synthetic dataset we added random log-normal noise N ( 0 , σ ) , where σ varied logarithmically between [0 . 01 , 0 . 01∙29] . Each point in Fig 6 is based on average of 10 different simulated dataset generated from 10 different random functions , each simulated dataset consists of 300 points .
|
Sometimes a combination of drugs works much better than each drug alone . Finding such drug cocktails is a pressing challenge in order to combat drug resistance and to improve drug effects . However , it is impossible to test all combinations of multiple drug experimentally . Therefore , researchers are looking for computational rather than experimental approaches to overcome this problem . One approach is to measure the effect of few drugs and plug it into a formula that predicts the effect of many drugs together . Existing prediction formulae typically perform best on the dataset that they were developed on , but less well on other datasets . Here we explain this observation and give a guide for the choice of an optimal prediction formula for a given dataset . The optimal formula depends on two main properties of the dataset: 1 ) The interaction strength between the drugs and 2 ) The experimental noise in the data . This study may help researchers discover effective combinations of multiple drugs and multiple perturbations in general .
|
[
"Abstract",
"Introduction",
"Results",
"Estimating",
"triplet",
"interactions",
"requires",
"an",
"accurate",
"null",
"model",
"Discussion",
"Methods"
] |
[
"antimicrobials",
"deletion",
"mutation",
"medicine",
"and",
"health",
"sciences",
"microbial",
"mutation",
"statistical",
"noise",
"statistics",
"drugs",
"microbiology",
"mathematical",
"models",
"mutation",
"antibiotic",
"resistance",
"mathematics",
"antibiotics",
"artificial",
"intelligence",
"pharmacology",
"mutation",
"databases",
"research",
"and",
"analysis",
"methods",
"gaussian",
"noise",
"computer",
"and",
"information",
"sciences",
"antimicrobial",
"resistance",
"mathematical",
"and",
"statistical",
"techniques",
"biological",
"databases",
"database",
"and",
"informatics",
"methods",
"genetics",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"machine",
"learning"
] |
2019
|
Noise-precision tradeoff in predicting combinations of mutations and drugs
|
How a stimulus or a task alters the spontaneous dynamics of the brain remains a fundamental open question in neuroscience . One of the most robust hallmarks of task/stimulus-driven brain dynamics is the decrease of variability with respect to the spontaneous level , an effect seen across multiple experimental conditions and in brain signals observed at different spatiotemporal scales . Recently , it was observed that the trial-to-trial variability and temporal variance of functional magnetic resonance imaging ( fMRI ) signals decrease in the task-driven activity . Here we examined the dynamics of a large-scale model of the human cortex to provide a mechanistic understanding of these observations . The model allows computing the statistics of synaptic activity in the spontaneous condition and in putative tasks determined by external inputs to a given subset of brain regions . We demonstrated that external inputs decrease the variance , increase the covariances , and decrease the autocovariance of synaptic activity as a consequence of single node and large-scale network dynamics . Altogether , these changes in network statistics imply a reduction of entropy , meaning that the spontaneous synaptic activity outlines a larger multidimensional activity space than does the task-driven activity . We tested this model’s prediction on fMRI signals from healthy humans acquired during rest and task conditions and found a significant decrease of entropy in the stimulus-driven activity . Altogether , our study proposes a mechanism for increasing the information capacity of brain networks by enlarging the volume of possible activity configurations at rest and reliably settling into a confined stimulus-driven state to allow better transmission of stimulus-related information .
How spontaneous brain dynamics are altered under stimulation or task conditions remains an important open question in neuroscience . Empirically , one of the most robust hallmarks of task-driven brain activity is the decrease of variability following an external stimulus input , a phenomenon observed across a variety of species , cortical areas , tasks , stimulus and attentional conditions , and using brain signals observed across multiple spatiotemporal scales including neuronal membrane potentials , neuronal firing rates , field potentials and functional magnetic resonance imaging ( fMRI ) signals [1–5] . A recent fMRI study showed that trial-by-trial variability of BOLD signals decreases following stimulus onset in a visual detection task and that the magnitude of variability reduction was correlated with the magnitude of trial-averaged response [3] . Moreover , the temporal variance of BOLD signals is significantly smaller during the same task as compared with the resting condition [6]—an effect that has also been reported in brain field potentials , neuronal membrane potentials , and neuronal spiking activity [7–9] . This suggests that the multidimensional space outlined by cortical activity is reduced following the stimulus onset [10] . Yet , a detailed mechanistic explanation of these effects is still lacking . In the present work we aimed to model the empirical observations of the fMRI study of [3] , by studying the effect of external inputs on the first- and second- order statistics of a large-scale model of the brain [11] . This model is composed of N local E-I nodes , with one excitatory and one inhibitory neural subpopulations , representing N brain regions that are interconnected through an empirical large-scale connectivity matrix obtained using diffusion imaging data of healthy human subjects [12] . The dynamics of each of the E-I nodes follows the mean field equations derived by [13] and the excitatory firing rate is clamped around 3 Hz by adjusting the connection weight from the I population to the E population , a procedure known as Feedback Inhibition Control ( FIC ) [11] . This large-scale model has been shown to provide an efficient description of resting-state fMRI functional connectivity together with realistic stimulus-evoked activity [11] . Here we assumed that different tasks can be modeled by sets of inputs that co-activate different brain regions . Furthermore , we focused on synaptic activity , since it has been shown that BOLD signals relate to local field potentials ( LFPs ) more closely than to neuronal firing rates [14–17] . Using this model we observed that , as a consequence of single node and network dynamics , the application of an external input impacts the network statistics , so that the entropy of the stimulus-evoked activity is lower than that during spontaneous activity . We confirmed this model prediction using empirical fMRI data and further discussed its functional implications .
We first evaluated the variability of the synaptic activity of single E-I nodes ( Fig 1A ) . We calculated two types of variability: i ) the variance across stochastic realizations of synaptic activity ( trial-by-trial variance ) , noted σ2 , and ii ) the autocovariance ( temporal variance ) of synaptic activity , defined as the covariance of the synaptic activity with itself at pairs of time points and noted Fu ( t+τ , t ) . Explicitly these statistics are given by: σ2=Var[u]=〈[u ( t ) −〈u ( t ) 〉]2〉 , ( 1 ) Fu ( t+τ , t ) =〈[u ( t+τ ) −〈u ( t+τ ) 〉] . [u ( t ) −〈u ( t ) 〉]〉 , ( 2 ) where u is the synaptic activity and the angle brackets < . > denote the average over stochastic model realizations ( i . e . the average over simulated trials ) . The autocovariance measures the strength of the influence of the past dynamics of the system on its future dynamics [its normalized version , Fu ( t+τ , t ) /Fu ( t , t ) , which is insensitive to the absolute amount of fluctuation , is the autocorrelation function ( ACF ) ] . The equations governing these statistics can be analytically calculated by assuming that the noise is sufficiently weak to allow for a linearized treatment of the fluctuations or linear noise approximation ( see Methods ) . We examined how the application of an external stimulus Iext to the E population ( Fig 1A ) changes the variability of synaptic activity . The stationary trial-by-trial variance of the synaptic activity under external input was compared to its stationary spontaneous level ( Iext = 0 ) , and the relative change was quantified by: Δσ2 ( Iext ) =100×[σ2 ( Iext ) σ2 ( 0 ) −1] . ( 3 ) The relative change of the mean synaptic activity ( Δm ) was also computed: Δm ( Iext ) =100×[m ( Iext ) m ( 0 ) −1] . ( 4 ) Fig 1B shows that an external input monotonically reduces the trial-by-trial variance of the synaptic activity of both E and I populations , and increases the mean synaptic activity for both populations . In S1 Text we explicitly solved the equations for the variance and showed that the decrease of synaptic activity’s variance in response to an external input is determined by nonlinearities and connectivity parameters ( see also S1 Fig ) . Moreover , the external input reduces the spread of the autocorrelation function ( ACF ) of the synaptic activity of both E and I populations ( Fig 1C and 1D ) . In conclusion , application of an external input attenuates the trial-by-trial fluctuations and shortens the temporal memory of the synaptic activity of an E-I local node . We next evaluated the first- and second-order statistics of task-driven activity in a large-scale network composed of local E-I nodes interconnected through empirically derived anatomical connectivity ( see Methods; see also [11] ) . The model has a single free parameter G that determines the strength of connectivity , called global coupling parameter ( see Methods ) , which in the following is fixed to G = 2 . 15 , this value falls in the range of G values ( between 1–4 . 45 ) for which the model fits closely to the resting-state functional connectivity of fMRI data [11] . Given the previous results for an isolated node , we predict that external inputs to local nodes propagate through the dynamical system , reducing the trial-by-trial variance of other nodes in the network via direct or indirect pathways . Fig 2A shows the response of the large-scale network when eight brain regions receive an external input ( equal to Iext = 0 . 02 nA ) . To simulate the results of [3] , in which human subjects performed a visual detection task , the selected brain regions receiving external inputs are related to visual processing . Two observations can be made: First , as expected from the response of isolated nodes , trial-by-trial variance reduces under simulated task condition for nodes directly receiving external inputs ( Fig 2B ) . Second , consistent with the above prediction , many nodes not directly receiving external inputs also exhibit trial-by-trial variance reduction upon external stimulation to ( other nodes in ) the network . Notably , the change of trial-by-trial variance with respect to the spontaneous activity ( Δσ2 ) is negative for all nodes ( Fig 2B ) and Δσ2 is negatively related to the change of synaptic activity Δm ( Fig 2C ) . This negative relation is consistent with the empirically observed negative correlation between the magnitude of variability reduction and the amplitude of evoked response in fMRI signals [3] . This relation is expected for a large variety of connectivity matrices , since it arises from the propagation of the stimulus to nodes separated by direct and indirect links . However , using synthetic connectivities with different levels of clustering , we found that the relation holds for connectivity matrices with low or intermediate clustering , as it is the case of human connectomes , but it breaks for excessively clustered connectivity matrices for which recurrent connections highly dominate ( see S2 Fig ) . The temporal dynamics of the model ( Fig 2D ) show that , during the application of the stimulus , the mean synaptic activity increases , while its variance decreases , and , after a period of relaxation of ~1–2 s , the system settles into a stable stimulus-evoked state . In the stationary spontaneous and stimulus-induced states , the power spectral density ( PSD ) of fluctuations of the system in the presence of stochastic perturbations can be calculated using the linear approximation ( see Methods , Eq 26 ) . The change of variance in the frequency domain is given by the relative change of the power spectral density ( ΔPSD ) in the task-driven synaptic activity with respect to the spontaneous condition , defined as: ΔPSD=100×[PSDtaskPSDspont−1] . ( 5 ) Interestingly , the effect of imposing an external input is different for different frequencies and , as a result of network interactions , the PSDs of the brain regions are differently affected by the external input ( Fig 2E ) . For both excitatory and inhibitory units , most of the brain regions directly receiving the external input showed reduced power in frequencies lower than 100 Hz , with a maximal reduction at ~9Hz ( 9 . 65 Hz for excitatory units; 7 . 42 Hz for inhibitory units ) , but those not directly receiving external input showed increased power in frequencies below 0 . 9 Hz and decreased power in frequencies between 0 . 9 and 100 Hz , with a maximal reduction at ~9Hz ( 9 . 35 Hz for excitatory units; 8 . 42 Hz for inhibitory units ) . These results are consistent with empirical electrophysiological findings of prominent desynchronization in alpha/beta frequency ranges during task performance [18] and human ECoG observations of decreased power in <1 Hz range only in task-relevant brain regions [7] . To show that the above results are not specific to the particular hypothetical “visual” task , we produced a large set of hypothetical tasks , by imposing an external input ( equal to 0 . 02 nA ) to the excitatory population of 8 randomly selected brain regions . The negative relation between Δσ2 and Δm was found for all tested stimuli ( Fig 2F ) . Interestingly , while the external stimulus highly impacts the covariances with respect to the spontaneous case ( Fig 2G ) , with a tendency to increase them , it only slightly changes the correlations between nodes ( Fig 2H ) . This indicates that functional connectivity amongst nodes , classically measured using correlation matrices , is not dramatically changed by imposing an external stimulus . We next allowed the global parameter G to vary and observed that the above results are qualitatively the same for a large parameter space ( within G = 1 and 3 ) ( Fig 3 ) , namely that Δσ2 and Δm are negatively related ( Fig 3A ) , that the task-driven functional connectivity is very similar to the spontaneous functional connectivity ( Fig 3B ) , and that the input prominently reduces the power of frequency fluctuations lower than 40 Hz ( Fig 3C and 3D ) . Within this parameter range the model captures both the observed behavior of the stimulus-driven activity and the resting functional connectivity ( as shown previously in [11] ) . In contrast , for G>3 , the model correctly predicts the resting functional connectivity , but the behavior of the stimulus-driven activity is not consistent with the empirical observations . We next calculated the dynamic change of the temporal variance ( autocovariance ) of the synaptic activity when an external input is applied to the large-scale model . The external input was applied at time t = 0 and lasted for 2 s . We used direct stochastic simulations of the network to estimate the time evolution of the autocovariance ( Fig 4A ) . During the application of the external input , the temporal correlation length is reduced . To quantify this effect we calculated the characteristic time scale of the ACF , noted T95 , given by the time lag at which its value is equal to 0 . 05 ( i . e . 95% percent of correlation decay ) . T95 was calculated using the linear approximation ( Eqs ( 21–24 ) ) in stationary spontaneous and stimulus conditions ( Fig 4B ) . We found that temporal correlations lasted more than twice as long in the spontaneous state than in the task state ( for the excitatory synaptic activity: T95 = 290 ms vs . T95 = 140 ms; for the inhibitory synaptic activity: T95 = 170 ms vs . T95 = 30 ms ) . Hence , the temporal memory of the synaptic activity of the large-scale model is shortened after the stimulus onset . Up to now we have focused on the dynamics of the synaptic activity . Because BOLD fMRI is widely used to study brain dynamics under both resting state and cognitive tasks , an important question pertains to whether the previous results apply to the dynamics of BOLD signals . To test this , we used a hemodynamic model to convert the total synaptic activity ( the sum of excitatory and inhibitory synaptic activity ) into BOLD activity . We used the Balloon-Windkessel model for the Hemodynamic response that describes the transduction of neural activity to BOLD changes , though non-linear dynamic equations of blood flow and deoxyhemoglobin content [19] . The model parameters were chosen as in [3] . Using this nonlinear model we found that an external stimulus input increases the trial-averaged BOLD activity , while reducing the averaged trial-by-trial variance of BOLD signals ( Fig 5A ) , leading to a linear negative relation between the relative change of trial-averaged BOLD activity and the relative change of its trial-by-trial variance during the application of the external input ( Fig 5B ) . However , the relative change of variance is positive for some of the brain regions ( 23 over 66 ) . In the model , this is due to the low-pass filtering of the hemodynamic model , since the Balloon-Windkessel model acts as low-pass filter of the synaptic activity that passes frequencies under 1 Hz [20 , 21] . As shown in Fig 2E , the stimulus-induced decrease of the synaptic variance is not negative for all brain regions for frequencies under 1 Hz . As a consequence , those brain regions for which the synaptic activity presents an elevation of the spectral power under 1Hz have a positive relative change of the variance of the BOLD activity ( Fig 5C ) . The stimulus-induced reduction of the autocovariance ( Fig 5D ) is moderate for the same reason: the memory of the BOLD signal is highly dominated by the slow hemodynamic response . We next investigated the functional implications of the change in network statistics induced by external inputs . To this end , we calculated the differential entropy H of the synaptic activity . The differential entropy is an extension of the Shannon entropy for a continuous random variable and it is related to the volume occupied by the continuous random variable . H can be easily calculated for a multivariate normal distribution , an assumption that is met in our case for the level of noise used in this study ( S3 Fig ) . In such cases , H depends on the covariance matrix which can be calculated using the linear noise approximation ( see Methods ) . We evaluated the differential entropy of the spontaneous activity and of the stimulus-driven activity for different model tasks determined by external inputs to a given subset of brain regions ( Fig 6A ) . We found that external stimulation systematically reduces the entropy of the synaptic activity ( Fig 6B ) . We next asked how much entropy ( or uncertainty ) in the synaptic activity is explained by the intrinsic noise present at each node of the model . In other words , we asked how much uncertainty is produced by the dynamical system due to the intrinsic noise of each node propagating into the network . To answer this question we calculated the Kullback-Leibler divergence ( KLD ) , also called relative entropy , between the distribution of intrinsic noise and the distribution of synaptic activity . Because the intrinsic noises are normally distributed with covariance Qn and the distribution of synaptic activity is normally distributed ( for weak noise ) with covariance Cv , the KLD can be calculated using Eq ( 32 ) ( see Methods ) . We found that the relative entropy of the spontaneous synaptic activity is systematically higher than that of the stimulus-driven synaptic activity , indicating that in the spontaneous state the dynamical system adds more uncertainty to the intrinsic stochastic process than it does in the stimulated condition ( Fig 6C ) . Thus far we have considered that the intrinsic noise of each brain region is independent between nodes ( i . e . , Qn is diagonal ) . However , it is reasonable to think that during a task and even at rest different brain regions share some noise , possibly due to shared sensory/proprioceptive background inputs . We thus calculated the entropy and the relative entropy in the case of non-diagonal noise covariance matrices . As for the diagonal case , we found that the stimulus-driven synaptic activity has lower differential entropy and lower relative entropy than the spontaneous activity ( S4 Fig ) . Thus , external stimulation reduces the entropy of synaptic activity even in the presence of common noise . We tested the model prediction of higher entropy in the spontaneous activity than in the task-driven activity using empirical data from [3] . The data consists of fMRI time-series from 33 ROIs , covering five cortical networks , as well as the hippocampus , thalamus and cerebellum , acquired in 17 healthy subjects ( see Methods ) . Each subject completed 8 fMRI runs , each lasting ~7 min , including 4 runs in resting-state conditions and 4 runs in a visual detection task condition . Each task run contains 20 stimulus presentations that the subject was to detect by pressing a button as quickly as possible . The inter-stimulus interval ranged from 17 . 3–30 . 2 s . First , for each subject and condition , we concatenated the time-series of the different runs and estimated the entropy using two methods: i ) by assuming normality and using Eq ( 30 ) , and ii ) by using the Nilsson-Kleijn non-parametric estimator ( see Methods ) . Using both methods , we found that the differential entropy in the resting activity is significantly higher than that of the task-driven activity ( p<0 . 01 , Wilcoxon signed-rank test ) ( Fig 7A ) . Second , we performed a time-resolved analysis in which the differential entropy was calculated using sliding windows of 5 frames ( 10 . 8 s ) shifted in steps of 1 frame ( 2 . 16 s ) . Using Eq ( 30 ) , we computed the time course of the differential entropy , averaged across subjects , in the task condition , Htask ( t ) , and during rest , Hrest ( t ) ( Fig 7B ) . The entropy values were referenced to the rest entropy H0 averaged across subjects and across time windows , i . e . H0=1T∑t=1THrest ( t ) , where T is the total number of time steps in a run . When the task data was aligned to the stimulus onset , we found that the differential entropy significantly decreases after the stimulus onset ( p<0 . 01 , paired t-test ) and , after ~8 s , it recovers its resting level H0 ( the significant difference at –17 . 3 s is due to the previous stimulus ) . Interestingly , the resting-state functional connectivity and the task functional connectivity were very similar , with differences in correlation coefficients ranging between ±0 . 02 ( Fig 7C ) , a feature that is captured by the model ( see Fig 2H ) . The above results show that the task-driven synaptic activity has lower trial-by-trial variance , lower temporal variance , and lower entropy than the spontaneous synaptic activity . Altogether , this indicates that the space occupied by the synaptic activity is reduced when external inputs are impinging upon the network . To illustrate this effect , we represented the synaptic activity at a given time point or in a given trial as a point in the state space . Fig 8A shows the simulated synaptic activity of three brain regions in a three-dimensional space defined by the activity of these brain regions when no external input is applied ( spontaneous condition ) and when an external input is applied ( task condition ) . The mean activity was removed for each brain region; thus , here , the activity represents deviations from the mean . In the spontaneous condition , the network explores a volume of the state space that is larger than the volume explored in the task condition . In the temporal domain , the space occupied by the synaptic activity is also reduced in the task condition compared to the spontaneous condition ( Fig 8B ) . This is shown by plotting the synaptic activity of a given brain region i in the three-dimensional space , or Poincaré map , defined as the synaptic activity in three different time points t , t+τ , and t+2τ . The volume of the space in the Poincaré map outlined by the spontaneous synaptic activity is larger than that occupied by the task-driven synaptic activity .
Reducing the space occupied by the synaptic activity , as a consequence of reducing the trial-by-trial and the temporal variability and increasing covariances , has relevant implications for information processing . It has been shown that temporal variance and entropy of BOLD signals change with chronological age and that young adults who are also faster and more consistent performers exhibited significantly higher brain variability across tasks [22–24] . In addition , the reduction of trial-by-trial variability is highly predictive of better performance [4] . These observations are likely complementary . Indeed , in the view of Information Theory , the mutual information between the brain activity and a given stimulus can be decomposed as the difference between the entropy of the full set of response patterns for all stimuli ( total entropy ) and the entropy conditioned to one stimulus ( evoked entropy ) . Thus , there are two ways of increasing the information carried by the brain activity: by increasing the total entropy or by decreasing the evoked entropy . There is growing evidence that the entropy at rest is an upper bound of the total entropy , since stimulus-evoked patterns reoccur during spontaneous activity [10 , 25 , 26] . In other words , more variability at rest is associated with a larger repertoire of potential brain states and greater information capacity [27] while the ability to reliably settle in a stimulus-evoked brain state allows better transmission of the information about the stimulus . Information Theory provides quantification of the amount of potential information that is available given the distribution of brain activity . How the brain decodes this available information is a topic of active research . Classification of multivariate fMRI patterns has been used to decode different stimuli or behavioral conditions from the fMRI signals [28–30] . In this context , the reduction of trial-by-trial variability under task would improve the discriminability of the fMRI multivariate patterns , which in turn improves the decoding performance . Moreover , if multivariate patterns have to be estimated using short time windows , as is likely during dynamical task processing , reducing the temporal correlations of the fluctuations would improve the estimation of the patterns ( since the reduction of the autocorrelation leads to an increase of the effective number of independent samples within the time window ) . It is possible that the brain uses similar coding schemes to efficiently represent the incoming sensory information and evolving mental states , although exactly how such decoding schemes are implemented by neural systems remains an open question . In the present study we focused on the dynamics of the synaptic activity to model the empirical BOLD fMRI signals . Concentrating on the synaptic activity is justified since it has been shown that BOLD signals relate more closely to Local Field Potentials ( LPF ) rather than neuronal firing rates [14–17] ) . As in previous studies of large-scale models [11 , 21 , 31] , we converted the synaptic activity into simulated BOLD signals via a non-linear hemodynamic model , known as the Balloon-Windkessel model [19] . We found prominent stimulus-induced decrease of BOLD variance and a negative correlation between the relative change of trial-averaged BOLD responses and the relative change of trial-by-trial BOLD variance , as reported empirically by [3] . This supports a previous conclusion [3] that the observed BOLD variability reduction is unlikely to be an effect of the nonlinearities in the hemodynamic response but rather is likely due to the underlying synaptic activity . Nonetheless , for some brain regions the simulated BOLD activity has slightly more variance in the stimulus-driven activity than in the spontaneous activity ( Fig 5B ) . Using a voxel-wise analysis across the whole brain on empirical fMRI data , it was found that whereas some voxels showed increased variance after stimulus onset , none of them were statistically significant after correction for multiple comparisons ( Figure 4 in [3] ) . Thus , whether the observation of task-induced increase in trial-to-trial variance in selected regions in our model has physiological importance remains to be seen . As mentioned above , the increase of variance in the present model is due to the low-pass filtering of the hemodynamic model that suppresses the fluctuations frequencies above 1 Hz . Indeed , consistent with previous empirical observations using human ECoG recordings [7] , the present model shows a prominent task-induced decrease of synaptic variance for frequencies >1Hz , but , for frequencies <1Hz , this is mostly evident in directly activated brain regions only . This suggests that the present dynamic mean field model might be too simple to reconcile these two features and should be extended to consistently reproduce the change of spectral power in both synaptic and BOLD activities . Another alternative is that the hemodynamic model needs to be refined to completely describe the neurovascular coupling between the BOLD signal and the synaptic activity at different frequencies . Indeed , experimental evidence shows that BOLD fluctuations correlate with broadband LFP signals and that the alpha ( 8–12 Hz ) , beta ( 18–30 Hz ) , and gamma ( 40–100 Hz ) LFP bands were informative about the spontaneous BOLD signals from an individual brain area [32] . The mechanism underlying stimulus-induced decrease of neural variability has been recently studied in theoretical works . Among the proposed mechanisms , spontaneous multi-stability has received much attention [5 , 33 , 34] . Under this scenario , the spontaneous activity of local neural networks with an underlying clustered connectivity is highly variable due to transitions through multiple spontaneous states . These transitions render the spontaneous activity more heterogeneous , but are suppressed when a stimulus stabilizes the network in a single evoked state and , as a result , the variability decreases in the stimulus-driven activity . This scenario naturally predicts an important feature of spontaneous activity , namely that the different spontaneous states are similar to the stimulus-evoked states [35 , 36] , a phenomenon reported in studies of neuronal membrane potentials and spiking activity at the microcircuit level [10 , 25] and in resting-state fMRI studies at large-scale network level [26 , 37–40] . By contrast , in the present study , the reduction of variability is due to single node synaptic dynamics ( Fig 1 ) without the need of multi-stability originating from clustered connections . We showed that variance decrease results from nonlinearities and local E-I connectivity ( see S1 Text ) . When the local nodes interact through long-range connections a pattern of stimulus-induced variance reduction is observed as a result of direct and indirect inputs—a phenomenon that is expected for a large variety of connectivities , as soon as large-scale recurrent connections do not strongly dominate ( S2 Fig ) . The model for local nodes presented herein is a mean-field model that describes the mesoscopic dynamics of synaptic activity . This model can be extended by introducing multi-stability in the local dynamics , a direction that requires further investigation . In the present data and model the functional connectivity is only slightly changed between rest and task . Several studies have reported high similarity between resting and task-related functional connectivity [37–40]; however , other studies have demonstrated reorganization of functional networks during task performance [41–43] . Brain dynamics might be engaged into task activity through diverse mechanisms . Here we modeled task-driven activity by imposing sets of inputs that co-activate different brain regions . There are other possible models for task effects on the brain , such as neuromodulation-mediated changes of network parameters that modify the neural excitability , the synaptic efficacy , or the gating of inputs . How these mechanisms alter the statistics of task-driven activity in a large-scale model have only recently been examined and awaits further investigation [44–46] . Finally , we here focused on the effect of imposing a stationary input to the large-scale brain model . A natural extension of the present work would be to study the effect of time-varying ( sinusoidal ) inputs and compute the frequency-dependent response function of different network statistics . Considering an input of small amplitude would allow to linearize the response and to study the eventual resonances . Moreover , these resonances may be partly determined by transmission delays , given by the experimental distance matrix between the different brain regions , a scenario that is not consider in the present work . In conclusion , we have shown that the stimulus-driven shrinkage of cortical activity space can be understood as a property of mesoscopic dynamics embedded in large-scale brain networks , a property that has important implications for information processing .
This research was conducted in agreement with the Code of Ethics of the World Medical Association ( Declaration of Helsinki ) and informed consent was obtained from all subjects before performing the study , in accordance with institutional guidelines . The study design was approved by the Human Studies Committee of Washington University in St . Louis and the local Ethics Committee of Lausanne University . Blood-oxygen-level dependent ( BOLD ) fMRI data ( 4x4x4 mm3 voxels , TE 25 ms , TR 2 . 16 s ) were acquired in 17 normal right-handed young adults ( 9 females , age 18–27 years ) using a 3T Siemens Allegra MR scanner . All subjects gave informed consent in accordance with guidelines set by the Human Studies Committee of Washington University in St . Louis . Each subject completed 8 fMRI runs , each 194 frames ( ~7 min ) in duration . They consisted of two alternating run types . The first run type was a resting-state study in which a white crosshair was presented in the center of a black screen . Subjects were instructed to look at the crosshair , remain still , and to not fall asleep . The second run type was a task study in which the identical crosshair was presented , but now it occasionally changed from white to dark gray for a period of 250 ms , at times unpredictable to the subjects , with an inter-stimulus interval of 17 . 3–30 . 2 sec . The subjects were instructed to press a button with their right index finger as quickly as possible when they saw the crosshair dim . This data set has been previously used in [3 , 6 , 47] . Thirty-three regions of interest ( ROIs ) covering five cortical networks—the attention , default-mode , motor , saliency and visual networks , as well as the hippocampus , thalamus and cerebellum were defined based on previous task-related functional neuroimaging studies . The preprocessing of the fMRI data and definition of ROIs are described in detail in [3] . We used the model of [11] to describe the global dynamics of the whole cortex . This model binds the dynamics of N local nodes , composed of excitatory—inhibitory subnetworks ( E—I networks ) , through the underlying anatomical structure which is estimated using diffusion-imaging data from healthy human subjects . The stochastic differential equations of the model describe the time evolution of the mean synaptic activity of each local node ( i . e . , brain region ) and there are given by: ui ( E ) =I0 , E+wEESi ( E ) +G∑jCijSj ( E ) −wEI , iSi ( I ) +Iext , i , ( 6 ) ui ( I ) =I0 , I+wIESi ( E ) −wIISi ( I ) , ( 7 ) ri ( E ) =ΦE ( ui ( E ) ) =aEui ( E ) −bE1−exp ( −dE ( aEui ( E ) −bE ) ) , ( 8 ) ri ( I ) =ΦI ( ui ( I ) ) =aIui ( I ) −bI1−exp ( −dI ( aIui ( I ) −bI ) ) , ( 9 ) dSi ( E ) dt=−Si ( E ) τE+ ( 1−Si ( E ) ) γri ( E ) +βηi ( E ) ( t ) , ( 10 ) dSi ( I ) dt=−Si ( I ) τI+ri ( I ) +βηi ( I ) ( t ) , ( 11 ) where SiE , I denotes the average excitatory or inhibitory synaptic gating variable ( i . e . , fraction of open channels ) at the local area i ( i ∈ [1 , … , N] ) . In Eqs 10 and 11 ηi ( E ) ( t ) and ηi ( I ) ( t ) are uncorrelated Gaussian noises and the noise amplitude at each node is β = 0 . 01 . riE , I denotes the population firing rate of the excitatory ( E ) or inhibitory ( I ) population in the brain area i . The population firing rates are sigmoid functions ( ΦI and ΦE ) of the input synaptic currents to the excitatory or inhibitory population i is given by uiE , I . Synaptic currents are the sum of i ) local currents within the local E—I networks , ii ) excitatory currents from the other local nodes , and iii ) external inputs Iext . The local currents in node i are the sum of constants inputs to excitatory and inhibitory populations , noted I0 , E and I0 , I , respectively , local excitatory-to-excitatory currents wEESi ( E ) , local inhibitory-to-excitatory currents wEI , iSi ( I ) , local excitatory-to-inhibitory currents wIESi ( E ) , and local inhibitory-to-inhibitory currents wIISi ( I ) . The weights of these local connections are given by: wEE = 0 . 21; wIE = 0 . 15; wII = 1; and the feedback inhibition weight , wEI , i , is adjusted for each node i so that the firing rate of the local excitatory neural population is clamped around 3Hz , whenever nodes are connected or not—this regulation is known as Feedback Inhibition Control ( FIC ) and the algorithm to achieve it is described in [11] . It has been shown that the FIC constrain leads to a better prediction of the resting functional connectivity and a more realistic network evoked activity [11] . Local E—I networks interact through excitatory connections given by the N-by-N anatomical connectivity matrix , noted C . The connectivity matrix is scaled by a single global parameter , G , that changes the network from weakly to strongly connected and determines the dynamical state of the system . As shown in [11] the model has one single stable fixed point of low firing activity in all cortical areas , for all values of G within the region where the FIC regulation can be achieved . For larger values of G , long-range interactions are too strong to be compensated by FIC and the activity diverges . Finally , Iext represents external stimulation for simulating task evoked activity: it is zero for all neural populations under resting state condition , and Iext>0 for those populations excited in the task condition . The values of all parameters are taken from [11] and are presented in S1 Table . Neuroanatomical structure was obtained using Diffusion Spectrum Imaging ( DSI ) data and tractography from five healthy right-handed male human subjects [12] . The grey matter was subdivided into 998 regions of interest ( ROIs ) which are grouped into 33 cortical regions per hemisphere ( 66 areas in total ) according to anatomical landmarks ( S2 Table ) . White matter tractography was used to estimate the fiber tract density connecting each pair of ROIs , averaged across subjects . Anatomical connectivity among the 66 cortical regions was calculated by summing all incoming fiber strengths to the corresponding ROIs of the target region , and dividing it by its region-dependent number of ROIs , resulting in a non-symmetric connectivity matrix . This normalization by the number of ROIs—which have approximately the same surface on the cortex , i . e . the same number of neurons—is required because neuronal activity is sensitive to the number of incoming fibers per neuron in the target region . As the dynamical model of one region already takes into account the effect of its internal connectivity ( see below ) , the connection of a region to itself was set to 0 in the connectivity matrix for the simulations . In the following we derive approximated equations for the statistics of the gating variables and the synaptic activity . To estimate the network’s statistics , we assume that the noise is sufficiently weak so that the state variables fluctuate around their mean value and , by linearizing the equations , we concentrate on linear fluctuations . In this way , we express the system of stochastic differential Eqs ( 6–11 ) in terms of the first- and second-order statistics of the distribution of synaptic gating variables: μi ( m ) , the expected mean gating variable of a given local neural population of type m ( where m = E or I ) of the cortical area i , and Pij ( mn ) , the covariance between gating variables of neural populations of type m and n of local cortical areas i and j , respectively . The statistics are defined as: μi ( m ) ( t ) =〈Si ( m ) ( t ) 〉 , ( 12 ) Pij ( mn ) ( t ) =〈[Si ( m ) ( t ) −μi ( m ) ( t ) ][Sj ( n ) ( t ) −μj ( n ) ( t ) ]〉 , ( 13 ) where the angular brackets < . > denote the average over realizations or “trials” . Note that , for the model , a “trial” means a realization of the system of differential Eqs ( 6–11 ) . In vector form , the system of equations writes: ddt ( S→ ( E ) S→ ( I ) ) = ( f ( E ) ( S→ ( E ) , S→ ( I ) ) f ( I ) ( S→ ( E ) , S→ ( I ) ) ) +η→ ( t ) , ( 14 ) where S→={S→ ( E ) , S→ ( I ) }={S1 ( E ) , … , SN ( E ) , S1 ( I ) , … , SN ( I ) } , η→ is uncorrelated Gaussian noise , fi ( E ) ( S→ ( E ) , S→ ( I ) ) =−Si ( E ) τE+ ( 1−Si ( E ) ) γΦ ( E ) ( ui ( E ) ) , and fi ( I ) ( S→ ( E ) , S→ ( I ) ) =−Si ( I ) τI+Φ ( I ) ( ui ( I ) ) for i = 1 , . . , N . In the following we use a linear approximation of the fluctuations . As shown in [11] , Taylor expanding S→ around μ→=〈S→〉 , i . e . Si ( m ) =μi ( m ) +δSi ( m ) , up to the first order , we obtain the differential equations for the means of the gating variables and the covariance of the fluctuations around the mean . For the mean values: dμi ( E ) dt=ddt〈S→i ( E ) 〉=−μi ( E ) τE+ ( 1−μi ( E ) ) γΦE ( ui ( E ) ) , ( 15 ) dμi ( I ) dt=ddt〈S→i ( I ) 〉=−μi ( I ) τI+ΦI ( ui ( I ) ) , ( 16 ) where ui ( m ) is the mean input current to the neural population m = E , I of cortical area i , defined as: u→= ( u→ ( E ) u→ ( I ) ) =WS→+I→0+I→ext , ( 17 ) where W is a block matrix defined as: W=[wEEIN+G . C−D ( w→EI ) wIEIN−wIIIN] , where C is the NxN anatomical matrix , G the global coupling parameter , IN is the NxN identity matrix , D ( w→EI ) is a NxN diagonal matrix containing the weights of the feedback inhibition wEI , i as diagonal elements , and I→0 and I→ext are the vectors containing the constant and external inputs . Let P being the covariance matrix between gating variables S→ . P is a block matrix defined as: P=[P ( EE ) P ( EI ) P ( IE ) P ( II ) ] . The differential equation of the covariance matrix is [11]: dPdt=AP+PAT+Qn , ( 18 ) where the superscript T is the transpose , Qn is the covariance matrix of the noise , given by Qn=〈η→ ( t ) η→ ( t ) T〉 , and A is the Jacobian matrix given by first-order partial derivative of the nonlinear function f with respect to each variable S , evaluated at μ→ . A is a block matrix defined as: A=[A ( EE ) A ( EI ) A ( IE ) A ( II ) ] , where Aij ( mn ) ={∂fi ( m ) ( μ→ ) ∂Sj ( n ) } . Note that the Jacobian matrix depends on the point μ→ at which it is evaluated . The synaptic input variables u→ are a linear combination of the gating variables S→ and , thus , covariance matrix between synaptic input variables u→ is given by: Cv=WPWT . ( 19 ) Knowledge of the Jacobian matrix and the stationary covariance gives the stationary autocovariance of the gating variables S→ , defined as the covariance of the process with itself at pairs of time points and given as: FS ( t+τ , t ) =〈[S→ ( t+τ ) −μ→ ( t+τ ) ][S→ ( t ) −μ→ ( t ) ]T〉 . ( 20 ) In the stationary regime FS ( t+τ , t ) depends only on τ and is given by: FS ( τ ) =eτAFS ( 0 ) =eτAP , ( 21 ) where the exponential matrix is defined as: eτA=I+τA+12 ! ( τA ) 2+13 ! ( τA ) 3+… ( 22 ) The stationary autocovariance of the synaptic input variables u→ is , thus , given by: Fu ( τ ) =WFS ( τ ) WT . ( 23 ) The autocorrelation function ( ACF ) of the i-th synaptic input variable is given by: ACFi ( τ ) =Fu , i ( τ ) /Fu , i ( 0 ) . ( 24 ) Finally , the power spectral density ( PSD ) of fluctuations around the fixed points is also determined by the Jacobian matrix . The cross-spectrum of the gating variables S→ is given as [11]: ΠS ( ω ) =〈δS˜ ( ω ) δS˜ ( ω ) †〉= ( A+iω ) −1Qn ( AT−iω ) −1 , ( 25 ) where δS˜ ( ω ) is the Fourier transform of δS→ ( t ) and the superscript † is the conjugate transpose . The cross-spectrum of the synaptic input variables u→ is , thus , given by: Πu ( ω ) =〈δu˜ ( ω ) δu˜ ( ω ) †〉=WΠS ( ω ) W† . ( 26 ) The PSD of synaptic activity as a function of the frequency ω is given by the diagonal of ∏u ( ω ) . Note that the different network’s statistics ( variances , covariances , and PSD ) are determined by the Jacobian matrix A that depends on the state of the nonlinear system ( the elements of the A are derivatives evaluated at μ→ ) . Because the application of an external input changes the state of the system , therefore changing the derivatives , the network’s statistics are also changed . In other words , the nonlinear nature of the system renders the network’s statistics state-dependent . In summary , to get the stationary network’s statistics we simulated the deterministic Eqs ( 15–18 ) and , once the stationary values of the mean synaptic gating variables ( μ→ ) , the covariance matrix ( P ) , and the Jacobian matrix ( A ) were reached all other statistics were computed using Eqs 19–26 . All differential equations used in the present study were solved using the Euler’s method with a time step equal to dt = 0 . 1 ms . The total number of simulation steps was 105 , this simulation length ensures that the system reaches the stationary regime . Once we have obtained the linear prediction of the covariance we can estimate the extent of all possible configurations of the network given by the differential entropy H , which expresses the entropy of a continuous variable with n-dimensional probability density function ( p . d . f . ) f , and writes: H ( f ) =−∫Df ( x→ ) lnf ( x→ ) dx→ , ( 27 ) where D ∈ ℝn is the support set of f , i . e . , D = {x|f ( x ) > 0} . The entropy is related to the spread of the p . d . f . , i . e . , it relates to the volume occupied by a continuous random variable . The volume of the support set D is defined as: Vol ( D ) =∫Ddx1dx2…dxn . ( 28 ) The volume of the smallest set that contains most of the p . d . f is approximately 2nH ( f ) [48] . Thus , low entropy implies that the random variable is confined to a small effective n-dimensional volume and high entropy indicates that the random variable is widely dispersed . For a n-dimensional normal distribution ( μ , ∑ ) with covariance matrix ∑ , the differential entropy in bits is given by the following form [48]: H=12ln[ ( 2πe ) ndet ( Σ ) ]/ln ( 2 ) =n2ln ( 2 ) ( 1+ln ( 2π ) ) +12ln ( 2 ) det ( Σ ) , ( 29 ) where det ( Σ ) is the determinant of the covariance matrix . We also calculated the differential entropy for the fMRI time-series used in [3] . For these empirical data we used two different calculations of the differential entropy . The first measure assumes that the data follows a n-dimensional multivariate normal distribution ( n = 33 ) and is given by , first , estimating the covariance matrix of the fMRI signals for each subject ( averaged across runs of the same condition , rest or task ) , noted Σ^ , second , calculating the determinant of Σ^ as the product of the k non-zero singular values ( λ ) to elude singularity , and , finally , calculating the entropy as follows: H=k2ln ( 2 ) ( 1+ln ( 2π ) ) +12ln ( 2 ) ∑j=1kln ( λj ) . ( 30 ) For 16/17 subjects we found that k = 29 for both rest and task . For only one subject we found that k = n = 33 for both rest and task . As a second measure we used the Nilsson-Kleijn non-parametric estimator that does not assume normality and calculates the differential entropy based on nearest neighbors of a sample set [49] . Both ways of calculating the differential entropy H gave very similar results: the values of H obtained using the two methods were highly correlated ( rc = 0 . 91 for rest data and rc = 0 . 90 for task data ) . Following [50] , we defined the relative entropy as the Kullback-Leibler divergence between the intrinsic noise and the synaptic activity of the network . In its general form the Kullback-Leibler divergence between two distributions f and g is defined as: KLD=∫flnfg . ( 31 ) The intrinsic noise and the synaptic activity are normally distributed ( see S3 Fig ) and , in this case , it can be shown that the relative entropy between the intrinsic noise and the synaptic activity writes [50]: KLD ( u→ , η→ ) =12[trace ( Qn-1Cv ) −lndet ( Cv ) det ( Qn ) −2N]/ln ( 2 ) . ( 32 ) The relative entropy can be seen as the amount of uncertainty that is produced by the dynamical system .
|
Task- or stimulus-related changes of brain dynamics have been the subject of intense investigation during the last years . One of the most robust hallmarks of task/stimulus-driven brain dynamics , as measured using diverse recording techniques , is the decrease of variability with respect to the spontaneous level . This has led several researchers to focus on the second-order statistics of evoked activity and to study their functional consequences for information processing . In particular , it was observed that the trial-to-trial variability ( related to variable responses to an identical stimulus from one presentation to the next ) and the temporal variance of functional magnetic resonance imaging ( fMRI ) signals decrease in the task-driven activity . Here , we built a computational model of the whole brain to understand how local and large-scale brain dynamics contribute to these effects . The model allowed us to derive equations for the network statistics of both spontaneous and evoked activity . We observed that , as a consequence of single node and network dynamics , stimulus input impacts network statistics in such a way that the entropy of the stimulus-driven activity is lower than that during spontaneous activity . We confirmed this model prediction using empirical fMRI data and we further discuss its functional implications .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2015
|
Task-Driven Activity Reduces the Cortical Activity Space of the Brain: Experiment and Whole-Brain Modeling
|
Scavenger receptor class B type 1 ( SR-B1 ) and low-density lipoprotein receptor ( LDLR ) are known to be involved in entry of hepatitis C virus ( HCV ) , but their precise roles and their interplay are not fully understood . In this study , deficiency of both SR-B1 and LDLR in Huh7 cells was shown to impair the entry of HCV more strongly than deficiency of either SR-B1 or LDLR alone . In addition , exogenous expression of not only SR-B1 and LDLR but also very low-density lipoprotein receptor ( VLDLR ) rescued HCV entry in the SR-B1 and LDLR double-knockout cells , suggesting that VLDLR has similar roles in HCV entry . VLDLR is a lipoprotein receptor , but the level of its hepatic expression was lower than those of SR-B1 and LDLR . Moreover , expression of mutant lipoprotein receptors incapable of binding to or uptake of lipid resulted in no or slight enhancement of HCV entry in the double-knockout cells , suggesting that binding and/or uptake activities of lipid by lipoprotein receptors are essential for HCV entry . In addition , rescue of infectivity in the double-knockout cells by the expression of the lipoprotein receptors was not observed following infection with pseudotype particles bearing HCV envelope proteins produced in non-hepatic cells , suggesting that lipoproteins associated with HCV particles participate in the entry through their interaction with lipoprotein receptors . Buoyant density gradient analysis revealed that HCV utilizes these lipoprotein receptors in a manner dependent on the lipoproteins associated with HCV particles . Collectively , these results suggest that lipoprotein receptors redundantly participate in the entry of HCV .
More than 160 million individuals worldwide are infected with hepatitis C virus ( HCV ) , which is especially troubling because HCV-induced cirrhosis and hepatocellular carcinoma are life-threatening diseases [1] . Current standard therapy combining peg-interferon ( IFN ) , ribavirin ( RBV ) and a protease inhibitor has achieved a sustained virological response in over 80% of individuals infected with HCV genotype 1 [2] . In addition , many antiviral agents targeting non-structural proteins and host factors involved in HCV replication have been proven highly effective for chronic hepatitis C patients [3] . HCV belongs to the Flaviviridae family and possesses a single positive-stranded RNA genome with a nucleotide length of 9 . 6 kb . There are many reports on candidate molecules for the transportation of HCV into cells . CD81 , which directly binds to HCV envelope glycoprotein E2 , was first identified as an HCV receptor [4] . Scavenger receptor class B type 1 ( SR-B1 ) was also identified as a co-receptor responsible for E2 binding to human hepatic cells by comparative binding studies [5] . Upon introduction of pseudotype particles bearing HCV envelope proteins ( HCVpp ) [6] , claudin-1 ( CLDN1 ) and occludin ( OCLN ) were identified as entry receptors for HCVpp into human kidney-derived HEK293 cells and mouse embryonic fibroblast-derived NIH3T3 cells , respectively [7 , 8] . CD81 , SR-B1 , CLDN1 and OCLN are regarded as essential factors for HCV entry because mouse NIH3T3 cells and hamster CHO cells expressing these four factors permit entry of HCVpp [8] . In addition , development of a robust in vitro propagation system of HCV based on the genotype 2a JFH1 strain ( HCVcc ) has led to the identification of several entry factors , including epidermal growth factor receptor ( EGFR ) [9] , Niemann-pick C1 Like 1 protein ( NPC1L1 ) [10] and cell death-inducing DFFA-like effector B ( CIDEB ) [11] . Previous reports have shown that HCV particles derived from patient sera interact with lipoproteins and apolipoproteins to form complexes known as lipoviroparticles ( LVPs ) [12 , 13] . The formation of LVPs is considered to have significant roles in HCV assembly and entry . Because several HCV receptor candidates are known to play crucial roles in lipid metabolism , these molecules are suggested to participate in HCV binding through interaction with virion-associated lipoproteins . SR-B1 is highly expressed in liver and acts as a binding receptor for mainly HDL to facilitate lipid uptake into hepatocytes . Low-density lipoprotein receptor ( LDLR ) is also a binding receptor for lipoproteins and widely expressed in various tissues including liver . However , the roles of SR-B1 and LDLR in HCV entry are not yet fully understood . Recently , novel genome-editing techniques involving the use of zinc finger nucleases , transcription activator-like effector nucleases , and clustered regularly interspaced short palindromic repeats ( CRISPR ) and CRISPR-associated protein ( CRISPR/Cas9 ) systems have been developed [14–16] . The CRISPR/Cas9 system is composed of guide RNA containing protospacer adjacent motif ( PAM ) sequences and Cas9 nuclease , which form RNA-protein complexes to cleave the target sequences; this system has already been used for the quick and easy establishment of gene-knockout mice and cancer cell lines [17 , 18] . Because of the narrow host range and tissue tropism of HCV , robust in vitro HCV propagation is limited to the combination of HCVcc and human hepatoma-derived Huh7 cell clones . These novel genome-editing techniques have enabled the establishment of target gene-knockout Huh7 cells , which provide reliable tools to determine the precise roles of host factors in the lifecycle of HCV . In this study , Huh7 cell lines deficient in both the SR-B1 and LDLR genes were established by using the CRISPR/Cas9 system and revealed that SR-B1 and LDLR redundantly participate in the entry of HCV . In addition , very low-density lipoprotein receptor ( VLDLR ) , which is expressed highly in the peripheral tissues but only slightly in the liver and Huh7 cells , plays a role in HCV entry redundant to those played by SR-B1 and LDLR .
Many receptor candidates and entry factors are known to be essential for HCV entry . Although previous reports have shown that CD81 , SR-B1 , CLDN1 and OCLN participate in HCV infection [8] , the interplay among these molecules and precise roles in HCV entry are not fully understood . To clarify the involvement of these receptors in HCV entry in more detail , we used the CRISPR/Cas9 system to establish 2 clones for each of 4 knockout ( KO ) Huh7 cell lines respectively deficient in the CD81 , SR-B1 , CLDN-1 and OCLN genes ( Fig 1A ) . Frame shift mutations in all alleles were confirmed by direct sequencing ( S1A Fig ) . Cell viability was determined by the Cell Titer-Glo Luminescent Cell Viability Assay ( S1B Fig , upper panel ) . Luciferase activities in these KO Huh7 cells are comparable to those of parental Huh7 cells . In addition , localization of lipid droplets which participate in lipid metabolism and in encapsidation of HCV was determined by the immunofluorescence assay ( S1B Fig , middle panel ) . The mean numbers of lipid droplet per cell were determined by using ImageJ software ( S1B Fig , lower panel ) . These KO and parental Huh7 cells exhibited similar localization and numbers of lipid droplets and morphologies . To examine the roles of these receptors in HCV entry , HCVcc was inoculated into these KO cells at a multiplicity of infection ( MOI ) of 1 , and intracellular HCV RNA levels were determined by qRT-PCR at 24 h post-infection ( Fig 1B ) . Huh7 cells deficient in either CD81 , CLDN1 or OCLN exhibited a drastic reduction of the intracellular HCV RNA levels compared to those of parental Huh7 cells , in contrast to a slight reduction in those of SR-B1 ( SR-KO ) cells . In addition , infectious titers in the culture supernatants at 72 h post-infection exhibited little decrease in SR-KO cells , in contrast to the much greater decreases in CD81 , CLDN1 or OCLN KO Huh7 cells ( Fig 1C ) . To further confirm the effect of SR-B1 deficiency in HCV infection , SR-KO Huh7 . 5 . 1 cells were established ( S2 Fig ) . The intracellular HCV RNA levels in SR-KO Huh7 . 5 . 1 cells at 24 h post-infection were slightly reduced compared to parental Huh7 . 5 . 1 cells , as seen in Huh7 cells . These results suggest that SR-B1 is dispensable for HCV entry into both Huh7 and Huh7 . 5 . 1 cells . Because both SR-B1 and LDLR have been reported to be entry factors for lipid-associated HCV particles , we hypothesized that LDLR can compensate for the role of SR-B1 in HCV entry . To examine the potential of a redundant role between SR-B1 and LDLR , the effect of siRNA-mediated knockdown of LDLR on HCV entry was examined in parental and SR-KO Huh7 cells . The efficiencies of siRNA-mediated knockdown were confirmed by immunoblotting ( Fig 2A ) . Although intracellular viral RNA levels in cells infected with HCVcc were drastically reduced in both parental and SR-KO Huh7 cells by the knockdown of CD81 , those in SR-KO cells were lower than those in parental cells by the knockdown of LDLR ( Fig 2B ) . To further examine the role of SR-B1 and LDLR in HCV entry , 2 clones for LDLR KO ( LD-KO ) Huh7 cells and 2 clones for SR-B1 and LDLR double KO ( SR/LD-DKO ) Huh7 cells were established by the CRIPSR/Cas9 system ( Fig 3A upper panel ) . Frame shift mutations in all alleles were confirmed by direct sequencing ( S1A Fig ) . Intracellular viral RNA levels in SR/LD-DKO Huh7 cells infected with HCVcc at an MOI of 1 were about 30 times lower than those in parental Huh7 cells at 24 h post-infection , in contrast to the slight reduction of RNA replication in SR-KO and LD-KO cells ( Fig 3A , lower panel ) . In addition , intracellular HCV RNA levels in SR/LD-DKO Huh7 cells were lower than those in SR-KO and LD-KO Huh7 cells at all time points after infection ( Fig 3B ) . To visualize the dissemination of HCV infection , a fluorescence-based live cell reporter system was used [19] . Translocation of GFP from the cytoplasm to nucleus was observed in Huh7 cells stably expressing GFP-NLS-IPS upon infection with HCV through cleavage of the IPS-1 sequence by NS3-4A protease . Nuclear localization of GFP was observed from 24 h post-infection in parental , SR-KO , and LD-KO Huh7 cells upon infection with HCVcc at an MOI of 1 , while it was detected from 48 h post-infection in SR/LD-DKO Huh7 cells ( Fig 3C and 3D ) . To further confirm the redundant role of SR-B1 and LDLR in HCV entry , the effects of exogenous expression of SR-B1 or LDLR in SR-KO , LD-KO and SR/LD-DKO Huh7 cells were examined ( Fig 3E ) . Although HCV RNA levels in SR-KO , LD-KO and SR/LD-DKO Huh7 cells were lower than those in parental Huh7 cells at 24 h post-infection with HCVcc at an MOI of 1 , exogenous expression of either SR-B1 or LDLR enhanced the RNA in SR-KO , LD-KO and SR/LD-DKO Huh7 cells to levels comparable to those in parental cells . In addition , HCV RNA levels were increased in accord with the expression levels of SR-B1 and LDLR ( Fig 3F ) , suggesting that SR-B1 and LDLR redundantly participate in HCV entry . To rule out the possibility that expression of SR-B1 and LDLR enhances HCV RNA replication , in vitro-transcribed subgenomic HCV RNA of the JFH1 strain was electroporated into SR-KO and LD-KO cells with or without expression of SR-B1 and LDLR and cultured in medium containing G418 for a month . Exogenous expression of SR-B1 and LDLR in each of the KO cells exhibited no significant effect on the colony formation of SGR cells ( S3A Fig ) . Next , in vitro-transcribed Fluc RNA and subgenomic HCV RNA containing NanoLuc were electroporated into parental , SR-KO , LD-KO and SR/LD-DKO Huh7 cells and luciferase activities were determined at 24 , 48 , 72 h post-electroporation ( S3B Fig ) . Luciferase activities were similar in parental and KO Huh7 cells , indicating that replication efficiencies of HCV RNA are comparable among these cells . To further examine the roles of SR-B1 and LDLR in HCV replication , replication of HCV RNA and expression of viral protein in each three clones derived from parental , SR-KO , LD-KO and SR/LD-DKO Huh7 cells were examined by qRT-PCR and immunoblotting , respectively ( S3C Fig ) . HCV RNA ( upper panel ) and NS5A protein ( lower panel ) in these cells were comparable among all clones , suggesting that SR-B1 and LDLR are not involved in HCV replication . In addition , to examine the effect of SR-B1 and LDLR in particle production , in vitro-transcribed full-length HCV RNA was electroporated into SR/LD-DKO Huh7 cells expressing SR-B1 or LDLR , and infectious titer in the culture supernatants at early phase post-electroporation was determined by focus forming assay ( S3D Fig ) . Deficiencies of SR-B1 and LDLR gene exhibited no significant effect on infectious titers in the supernatants , suggesting that neither SR-B1 nor LDLR is involved in particle production of HCV . A search using the web-based search engine NextBio ( NextBio , Santa Clara , CA ) revealed that VLDLR , a member of the LDLR family , is expressed at a high level in peripheral tissues , and at a low level in the liver ( Fig 4A ) . Furthermore , expression levels of SR-B1 and LDLR are high in Huh7 cells and primary human hepatocytes ( PHH ) , while those of VLDLR are quite low ( Fig 4B ) . VLDLR belongs to the family of lipoprotein receptors and is structurally homologous to LDLR . Therefore , we considered that VLDLR may also be involved in HCV entry . To examine the role of VLDLR in HCV entry , VLDLR was expressed in SR-KO , LD-KO and SR/LD-DKO Huh7 cells by a lentiviral vector ( Fig 4C , upper panel ) . HCVcc was inoculated into cells at an MOI of 1 and intracellular HCV RNA levels were determined by qRT-PCR at 24 h post-infection ( Fig 4C , lower panel ) . Exogenous expression of VLDLR rescued HCV entry in SR/LD-DKO cells but not in parental Huh7 cells , suggesting that VLDLR expression can compensate for the roles of SR-B1 or LDLR in HCV entry . Very recently , Ujino et al . showed that VLDLR variant 2 participates in HCV entry independent from CD81-mediated HCV entry [20] . To further examine the role of VLDLR in HCV entry , VLDLR was expressed in CD81 KO , CLDN1 KO and OCLN KO Huh7 cells . In contrast to expression of VLDLR in SR/LD-DKO Huh7 cells , exogenous expression of VLDLR in CD81 , CLDN1 and OCLN KO cells exhibited no effect on HCV entry upon infection with HCVcc ( Fig 4D ) . Furthermore , HCVcc was inoculated into CD81 KO and SR/LD-DKO cells expressing either variant 1 or variant 2 of VLDLR at an MOI of 1 and intracellular HCV RNA levels and expression of NS5A were determined by qRT-PCR and immunofluorescense assay , respectively ( S4 Fig ) . The levels of HCV RNA and NS5A expression in VLDLR expressing CD81 KO cells were comparable to that of control CD81 KO cells . These results suggest that the role of VLDLR cannot compensate the role of CD81 , CLDN1 and OCLN . To examine the role of lipoprotein receptors in the entry of other genotypes of HCV , chimeric HCVcc of genotype 1b and 2a , the Con1-JFH1 and Jc1 viruses were used . Expression of SR-B1 , LDLR or VLDLR enhanced the entry of Con1-JFH1 and Jc1 viruses in SR/LD-DKO Huh7 cells at 24 h post-infection ( Fig 4E ) , as seen in JFH1 infection . To further examine the redundant role of lipoprotein receptors in the entry of HCV derived from in vivo , sera of mice with chimeric human livers infected with HCVcc were inoculated into SR/LD-DKO Huh7 cells expressing either SR-B1 , LDLR or VLDLR , and the HCV RNA levels were determined by qRT-PCR ( Fig 4F ) . Exogenous expression of SR-B1 , LDLR or VLDLR recovered susceptibility of SR/LD-DKO Huh7 cells to the mice-derived HCV . These results suggest that VLDLR has a similar role with SR-B1 and LDLR in the entry of HCV derived from not only cell culture but also from in vivo . To determine the roles of lipoprotein receptors in HCV entry in greater detail , binding assay was performed in SR/LD-DKO Huh7 cells expressing SR-B1 , LDLR or VLDLR . HCVcc were inoculated into cells , incubated at 4°C for 1 h and then washed three times with phosphate-buffered saline ( PBS ) to remove unbound particles . HCV RNA levels were determined by qRT-PCR immediately following binding ( Fig 5A and 5B ) . The intracellular HCV RNA levels were significantly lower in CD81 KO and SR/LD-DKO Huh7 cells , but were comparable in CLDN1 KO and OCLN KO Huh7 cells in compared with parental Huh7 cells . In addition , exogenous expression of SR-B1 , LDLR or VLDLR in SR/LD-DKO Huh7 cells rescued the binding step , suggesting that lipoprotein receptors participate in the binding step of HCV entry . Previous studies revealed that mutations of S112F and T175A in SR-B1 were observed in patients with high-HDL cholesterol levels in sera [21] . These mutations are located in the large extracellular loop region of SR-B1 and abrogate binding to HDL and uptake of lipid [22] . SR/LD-DKO Huh7 cells expressing either the wild-type or mutant SR-B1 by lentiviral vectors were established . To confirm the cell surface expression of mutants of SR-B1 , biotinylated cell surface proteins were purified and examined by immunoblotting ( S5A Fig ) . Although GFP-HA and actin in the cytoplasm were not biotinylated , the wild-type and mutants of lipoprotein receptors were biotinylated and detected in the plasma membrane fractions similar to EGFR used as a plasma membrane marker , suggesting that both wild-type and mutant lipoprotein receptors are similarly expressed on the cell surface . To determine lipoprotein uptake activity , lipid transfer assay was performed in SR/LD-DKO Huh7 cells expressing several mutants by using fluorescent-labeled HDL and LDL ( S5D–S5F Fig ) . Expression of wild-type but not of S112F and T175A mutants facilitates to uptake of HDL and LDL in SR/LD-DKO Huh7 cells . To examine the roles of the lipid uptake machinery in HCV entry , SR/LD-DKO Huh7 cells expressing either the wild-type or mutant SR-B1 were inoculated with HCVcc at an MOI of 1 , and intracellular HCV RNA levels were determined by qRT-PCR at 24 h post-infection ( Fig 6A ) . Expression of wild-type SR-B1 but not of the S112F and T175A mutants completely rescued HCV entry , suggesting that lipoprotein binding and lipid uptake of SR-B1 participate in HCV entry . LDLR and VLDLR have several repeats in the ligand binding domain that is responsible for the uptake of LDL and VLDL . To examine whether these repeats participate in HCV entry , LDLR and VLDLR mutants with these repeats deleted were expressed in SR/LD-DKO Huh7 cells and the intracellular HCV RNA levels were determined upon infection with HCVcc at an MOI of 1 by qRT-PCR at 24 h post-infection ( Fig 6B ) . Lipid transfer assay revealed that these deletions almost abrogated the lipid uptake ability ( S5 Fig ) . Expression of deletion mutants in the repeats domain of LDLR and VLDLR in SR/LD-DKO Huh7 cells failed to recover HCV entry , in contrast to the rescue of entry by the expression of the intact LDLR or VLDLR , suggesting that the repeats in the ligand binding domain of LDLR and VLDLR are important for HCV entry . To rule out the possibility that deletions in the repeat-containing domain caused a loss of original domain structures , point mutants of LDLR were constructed ( S5 Fig ) . A previous study revealed that substitution of asparagine to tyrosine in the repeat 5 and in the repeats from 2 to 7 was crucial for binding to LDL and VLDL [23] . To examine the roles of the binding ability of LDLR to LDL and VLDL in HCV entry , SR/LD-DKO Huh7 cells expressing either wild-type or mutant LDLR by lentiviral vectors were inoculated with HCVcc at an MOI of 1 , and intracellular HCV RNA levels were determined by qRT-PCR at 24 h post-infection ( Fig 6C ) . Expression of the wild-type LDLR but not of the mutants rescued HCV entry , suggesting that that ability of LDLR to bind LDL and VLDL is important for HCV entry . To further examine the involvement of the interaction between viral envelope proteins and the receptors in the lipoprotein receptor-mediated HCV entry , the roles of SR-B1 , LDLR and VLDLR in entry of the HCV pseudotype particles ( HCVpp ) were determined ( Fig 6D ) . HCVpp bearing HCV envelope glycoproteins was generated in 293T cells that were deficient in lipoprotein production . Although CD81 KO Huh7 cells did not show any susceptibility to HCVpp infection , the infectivity of HCVpp to SR/LD-DKO Huh7 cells was comparable to that in SR/LD-DKO Huh7 cells expressing lipoprotein receptor , suggesting that viral particle-associated lipoproteins participate in the lipoprotein receptor-mediated HCV entry . Although buoyant density gradient analyses have shown that viral RNA and infectious particles were broadly distributed at various densities , it is not clear whether the utilization of lipoprotein receptors is associated with viral particle densities . To determine the involvement of LVP density in lipoprotein receptor-mediated HCV entry , HCVcc was fractionated by buoyant density gradient ultracentrifugation and each fraction was inoculated into SR/LD-DKO Huh7 cells expressing SR-B1 , LDLR or VLDLR . The infectious titer at 72 h post-infection and intracellular HCV RNA levels at 24 h post-infection in each fraction were determined by focus forming assay and qRT-PCR , respectively ( Fig 7 ) . HCV in the high-density and low-density fractions exhibited higher affinity for SR/LD-DKO Huh7 cells complemented with SR-B1 and to those complemented with LDLR and VLDLR , respectively . These results suggest that lipoprotein receptors such as SR-B1 , LDLR and VLDLR participate in HCV entry in a manner that is dependent on the density of LVP .
In this study , we demonstrated that HCV-associated lipoproteins are involved in HCV entry via lipoprotein receptors such as SR-B1 , LDLR and VLDLR . In addition , our data indicated that these lipoprotein receptors redundantly participate in HCV entry in a manner that is dependent on the density of virion-associated lipoproteins . Previous studies have shown that inhibition of HCV entry by knockdown of both SR-B1 and LDLR was comparable to that by knockdown of either receptor alone , suggesting that SR-B1 and LDLR independently participate in HCV entry [24] . In this study , we employed gene-knockout techniques with a CRISPR/Cas9 system to obtain data more reliable than that by knockdown experiments and demonstrated that SR-B1 and LDLR have a redundant role in HCV entry ( Fig 3 ) . Several reports have revealed the involvement of lipoproteins in HCV entry via SR-B1 and LDLR . Binding of SR-B1 to HCV particles derived from patient sera was inhibited by the treatment of ApoE and VLDL [25] , and lipid transfer activity of SR-B1 was shown to be involved in HCV entry [26 , 27] . In addition , HCV entry was inhibited by the treatment with antibodies against LDLR , ApoE and ApoB [28] , and the interaction between HCV-associated apolipoproteins and LDLR facilitated efficient HCV entry [12 , 29] . On the other hand , Catanese et al . showed that HCVcc efficiently infects Huh7 . 5 cells in the absence of serum lipoproteins [30] . However , we demonstrated herein that the ligand binding activity of SR-B1 , LDLR and VLDLR is crucial for HCV entry , and overexpression of these lipoprotein receptors in SR/LD-DKO Huh7 cells has no effect on infection with HCV pseudotype particles ( Fig 6D ) . In addition , a recent report showed that HCVpp exhibited NPC1L1-independent cell entry and the cholesterol-abundant HCVcc exhibited enhanced NPC1L1-dependent entry , suggesting that virion-associated cholesterol is involved in viral entry via NPC1L1 [10] . Furthermore , knockdown of CIDEB affected entry of HCVcc but not of HCVpp [11] . Both CIDEB and NPC1L1 are involved in the regulation of lipid metabolism , and therefore HCV may utilize lipid metabolism for its cellular entry . The present results suggest that HCV-associated lipoproteins are involved in lipoprotein receptor-mediated entry of HCV . Although our current data demonstrated that deficiency or expression of SR-B1 , LDLR and VLDLR had no effect on entry of HCVpp ( Fig 6D ) , previous study showed that HCVpp entry is dependent on SR-B1[27] . They examined HCVpp assay in rat hepatocarcinoma cells ( BRL cells ) expressing CD81 , CLDN1 and SR-B1 and showed that lipid transfer activity of SR-B1 were required for HCV entry . There is a possibility that other factors can compensate the lipid transfer activity of SR-B1 in Huh7 cells , in contrast to BRL cells . Actually , deficiency of SR-B1 and LDLR cannot block HCV entry completely , compared to CD81 or CLDN1 KO Huh7 cells ( Figs 1B and 3A ) . In addition to SR-B1 and LDLR , we demonstrated that VLDLR , which is structurally similar to LDLR , plays roles in HCV entry similar to those of SR-B1 and LDLR ( Fig 4 ) . Very recently , Ujino et al . showed that HCV utilizes VLDLR for entry independently form CD81 , CLDN1 and OCLN-mediated pathway[20] , in contrast to our current results suggesting that expression of VLDLR had no effect on entry of HCV in CD81KO Huh7 cells . These discrepancies might be attributable to the difference in experimental procedures . First , Ujino et al . established CD81 KO Huh7 . 5 cells whereas we utilized Huh7 cells . Second , they applied knockdown to evaluate the role of CLDN1 and OCLN in contrast to knockout in our study . Third , they used untagged VLDLR while we used HA-tagged VLDLR . Last , they used VLDLR variant 2 lacking O-linked sugar domain while we used variant 1 . Further studies are needed to clarify the precise roles of VLDLR in HCV entry . Although we demonstrated that expression of VLDLR in SR/LD-DKO Huh7 cells significantly rescued HCV entry , enhancement in parental and SR-KO Huh7 cells were not significant . There are possibilities that the affinity of HCV to VLDLR is weaker than that to SR-B1 and LDLR . VLDLR is widely expressed in peripheral tissues but not in hepatocytes , in contrast to the abundant expression of SR-B1 and LDLR in the liver . Both SR-B1 and CLDN1 are specifically expressed in the liver , and their expression might be involved in determining the tissue tropism of HCV infection [31] . However , previous reports have shown that CLDN6 and CLDN9 expressed in various tissues play roles in HCV entry into non-hepatic human cells that are comparable to those played by CLDN1 in HCV entry into human hepatic cells [32–34] . Therefore , it might be feasible to speculate that expression of VLDLR and CLDN6/9 enables HCV to be internalized into various non-hepatic tissues , leading to development of the extrahepatic manifestations that sometimes occur in chronic hepatitis C patients . A number of host factors have been shown to participate in HCV entry into human hepatocytes . Heparan sulfate , LDLR and SR-B1 are thought to mediate the initial attachment of the lipoprotein-associated HCV particles to the cell surface of hepatocytes . After the initial binding , CD81 , CLDN1 and OCLN initiate HCV internalization and induce clathrin-mediated endocytosis [32 , 35 , 36] . A previous study reported that SR-B1 is involved in the early step of HCV entry [37] , as seen also in our current study . In addition , we have shown that not only SR-B1 but also LDLR and VLDLR are involved in the binding step of HCV ( Fig 5 ) . On the other hand , several reports have shown that SR-B1 is involved in the post-binding step of HCV entry [26 , 38] . Although we demonstrated that LDLR and VLDLR can compensate the role of SR-B1 in HCV entry , the roles of LDLR and VLDLR in post-binding step were not evaluated in this study . Previous reports have shown that lipid transfer activities of SR-B1 are required as a post-binding step in HCV entry [26 , 27] . Therefore it might be feasible to speculate that other factors such as LDLR and VLDLR are involved in the post-binding step by using lipid transfer activity or that other mechanisms participate in HCV entry in the absence of SR-B1 . Further studies are needed to clarify the roles of lipoprotein receptors in the post-binding step . Wünschmann et al . demonstrated that low-density HCV particles but not intermediate-density particles bound to LDLR-expressing cells [39] . In addition , Thi et al . showed that SR-B1 mediates primary attachment of HCV particles of intermediate density to cells [27] . These data are consistent with our present finding that the lipoprotein receptor usage of HCV is dependent on viral particle density ( Fig 7 ) . The apolipoproteins on lipoproteins differ according to the lipoprotein density , and HCV particles derived from patient sera are associated with the exchangeable apolipoproteins ApoA-1 , ApoB48 , ApoB100 , ApoC-1 , ApoC-3 and ApoE [40–44] , suggesting that LVPs engage high- and low-density lipoprotein receptors during uptake into hepatocytes . Previous reports have demonstrated the presence of cell-to-cell infection of HCV , which means that HCV particles are directly transmitted to neighboring cells without viral particles production in the extracellular space [45–47] . Although CD81 is a critical factor in cell-free HCV infection , it was shown that CD81 is dispensable for the cell-to-cell spread of HCVcc [48] . Several reports have shown that SR-B1 is involved in not only cell-free but also cell-to-cell infection by using SR-B1 antibodies [38 , 49] . In addition , recent reports showed that cell-to-cell infection was affected by siRNA-mediated knockdown and expression of ApoE [50 , 51] , suggesting that lipoprotein receptors including SR-B1 , LDLR and VLDLR redundantly participate in not only cell-free but also cell-to-cell infection of HCV . Several reports have shown the involvement of lipoprotein receptors in the entry of other viruses . Bovine viral diarrhea virus ( BVDV ) and GB virus C ( GBV-C ) , members of the Flaviviridae family , were reported to use LDLR in entry [28] , and the cell surface expression of LDLR was increased upon infection with dengue virus ( DENV ) [52] . Furthermore , siRNA-mediated knockdown of SR-B1 abolished enhancement of DENV infection through the interaction between SR-B1 and virion-associated ApoA-1 in various cell lines including Huh7 [53] , and the core protein of DENV was shown to bind to VLDL via ApoE in vitro [54] . Finally , levels of total plasma cholesterol , HDL and LDL were shown to be significantly lower in patients with severe dengue hemorrhagic fever than in mild cases or healthy controls [55 , 56] , suggesting that lipid metabolism participates in Flavivirus infection , as seen in HCV infection . Although direct-acting antivirals ( DAAs ) have been applied in a clinical setting , their use is still limited to severe hepatitis , transplantation , HIV/HCV-co-infection or immune-compromised patients [57] . Viral entry is one of the most important steps in the HCV lifecycle , especially in the reinfection of HCV in the graft after liver transplantation . Treatment with inhibitors of HCV entry might be an attractive strategy to prevent reinfection in the transplanted liver . Previous reports have shown that antibodies against SR-B1 inhibit HCV entry not only in vitro but also in vivo . ITX5061 , a SR-B1 antagonist that inhibits HDL catabolism in the liver by targeting SR-B1 , was shown to successfully inhibit HCV entry and spread in vitro [58] . Although ITX5061 has entered clinical development and was found to be safe and tolerated in a Phase 1b trial , the viral loads were not significantly reduced [59] . In addition , prophylactic administration of monoclonal antibodies against SR-B1 into uPA-SCID mice prior to xenotransplantation with human liver cells can prevent infection and spread of HCV [60–62] . In spite of the potent efficacy of SR-B1 inhibitors in clinical settings , we showed that deficiencies of SR-B1 achieved only marginal reduction in HCV entry into Huh7 cells . This discrepancy might be explained in either of two ways . First , inhibitors of SR-B1 may affect the functions of other lipoprotein receptors . Second , SR-B1 may play a major role in HCV entry in vivo , in contrast to the redundant participation of lipoprotein receptors in vitro . In any case , SR-B1 is a promising target for development of novel anti-HCV therapeutics . In summary , we have shown that the lipoprotein receptors SR-B1 , LDLR and VLDLR possess redundant roles in HCV entry through their interaction with the viral-associated lipoprotein .
All cell lines were cultured at 37°C under the conditions of a humidified atmosphere and 5% CO2 . The human hepatocellular carcinoma-derived Huh7 and human embryonic kidney-derived 293T cells were maintained in DMEM ( Sigma ) supplemented with 100U/ml penicillin , 100 μg/ml streptomycin , and 10% fetal calf serum ( FCS ) . The Huh7-derived cell line Huh7 . 5 . 1 was kindly provided by F . Chisari . The primary human hepatocyte ( PHH ) was purchased from PhoenixBio . The cDNA clones of SR-B1 , LDLR , VLDLR , and AcGFP were inserted between the XhoI and XbaI sites of lentiviral vector pCSII-EF-RfA , which was kindly provided by M . Hijikata , and the resulting plasmids were designated pCSII-EF-SR-B1 , pCSII-EF-LDLR , pCSII-EF-LDLR-HA , pCSII-EF-VLDLR-HA , pCSII-EF-AcGFP , and pCSII-EF-GFP-HA respectively . The point mutants of SR-B1 and LDLR and the deletion mutants of LDLR and VLDLR were amplified by PCR and introduced into pCSII-EF . The plasmid GFP-NLS-IPS encodes the green fluorescent protein with a simian virus 40 ( SV40 ) nuclear localization signal fused to IPS-1 residues 462 to 540 , which have the site of HCV NS3-4A cleavage and mitochondrial localization sequence [19] . The plasmid pHH-JFH1 encodes a full-length cDNA of the JFH1 strain . pHH-JFH1-E2p7NS2mt contains three adaptive mutations in pHH-JFH1 [63] . pJFH1 encodes full-length cDNA of the JFH1 strain , and pSGR-JFH1 encodes subgenomic cDNA of the JFH1 strain . The secreted nano-luc ( Nlucsec ) fragment from the pNL1 . 3 vector ( Promega ) was replaced with the neomycin gene of pSGR-JFH1 and the resulting plasmid was designated pSGR-JFH1-Nlucsec . The plasmid pX330 , which encodes hCas9 and sgRNA , was obtained from Addgene ( Addgene plasmid 42230 ) . The fragments of guided RNA targeting the CD81 , SR-B1 , CLDN1 , OCLN and LDLR gene were inserted into the Bbs1 site of pX330 and designated pX330-CD81 , pX330-SR-B1 , pX330-CLDN1 , pX330-OCLN , and pX330-LDLR , respectively . The plasmids used in this study were confirmed by sequencing with an ABI 3130 genetic analyzer ( Life Technologies ) . Mouse monoclonal antibody to β-actin was purchased from Sigma . Mouse anti-CD81 antibody was purchased from Santa Cruz Biotechnology . Rabbit anti-SR-B1 , OCLN , and CLDN1 antibodies were purchased from NOVUS Biologicals , Proteintech and Life Technologies , respectively . Chicken anti-LDLR antibody was purchased from Abcam . Rat anti-HA antibody was purchased from Roche Diagnostics . Rabbit anti-NS5A antibody was prepared as described previously [64] . Alexa Flour ( AF ) 488-conjugated anti-rabbit IgG antibody and BODIPY558/568 lipid probe were purchased from Life Technologies . 4’ , 6-diamidono-2-phenylindole ( DAPI ) was purchased from Vector Laboratories , Inc . A small interfering RNA ( siRNA ) pool targeting LDLR , CD81 and control nontargeting siRNA were purchased from Dharmacon , and transfected into cells using Lipofectamine RNAi MAX ( Life Technologies ) according to the manufacturer’s protocol . pHH-JFH1-E2p7NS2mt , pHH-JFH1 were introduced into Huh7 . 5 . 1 cells; HCVcc in the supernatant was collected after serial passages; and infectious titers were determined by a focus-forming assay and expressed in focus-forming units ( FFU ) [64] . RNA transcribed from pJFH2/AS/mtT4 was electroporated into Huh7 . 5 . 1 cells; HCVcc in the supernatant was collected after serial passages; and infectious titers were determined by a focus-forming assay and expressed in FFU . All mouse studied were conducted at Hiroshima University ( Hiroshima , Japan ) in accordance with the guidelines of the local committee for animal experiments . Chimeric mice transplanted with human hepatocytes were generated as described previously[65] . The experimental protocol was approved by the Ethics Review Committee for Animal Experimentation of the Graduate School of Biomedical Sciences ( Hiroshima University ) . The chimeric mice were infected with a 4 x 105 titer of HCVcc . Serum samples were collected at 2 to 8 weeks after infection . HCV pseudotype particles ( HCVpp ) containing E1 and E2 glycoproteins of JFH1 were produced as previously described [6 , 66] . The lentiviral vectors and ViraPower Lentiviral Packaging Mix ( Life Technologies ) were co-transfected into 293T cells by Trans IT LT-1 ( Mirus ) , and the supernatants were recovered at 48 h post-transfection . The lentivirus titer was determined by using a Lenti XTM qRT-PCR Titration Kit ( Clontech ) , and expression levels and AcGFP were determined at 48 h post-inoculation . Cells lysed on ice in lysis buffer ( 20 mM Tris-HCl [pH7 . 4] , 135 mM NaCl , 1% Triton-X 100 , 10% glycerol ) supplemented with a protease inhibitor mix ( Nacalai Tesque ) were boiled in loading buffer and subjected to 5–20% gradient SDS-PAGE . The proteins were transferred to polyvinylidene difluoride membranes ( Millipore ) and reacted with the appropriate antibodies . The immune complexes were visualized with SuperSignal West Femto Substrate ( Pierce ) and detected by using an LAS-3000 image analyzer system ( Fujifilm ) . Huh7 cells were transfected with pX330-CD81 , pX330-SR-B1 , pX330-CLDN1 , pX330-OCLN , and pX330-LDLR by Trans IT LT-1 ( Mirus ) , and single cell clones were established by the single cell isolation technique . To screen for gene-knockout Huh7 cell clones , mutations in target loci were determined by using a Surveyor assay ( Transgenomic ) according to the manufacturer’s protocol . Frameshift of the genes and deficiencies of protein expressions were confirmed by direct sequencing and immunoblotting analysis , respectively . For quantification of HCV-RNA , total RNA was extracted from cells by using a PureLink RNA Mini Kit ( Invitrogen ) , and the first-strand cDNA synthesis and qRT-PCR were performed with a TaqMan RNA-to-CT 1-step Kit and ViiA7 system ( Life Technologies ) , respectively , according to the manufacturer’s protocol . The primers for TaqMan PCR targeted to the noncoding region of HCV RNA were synthesized as previously reported [64] . For quantification of gene expression , the synthesis of the first-stranded cDNA was performed by using a PrimeScript TR reagent Kit ( Perfect Real Time ) ( Takara Bio ) and quantitive RT-PCR was performed by using Platinum SYBR Green qRT-PCR SuperMix UDG ( Life Technologies ) according to the manufacturer’s protocol . The primers sequences for amplification of the SR-B1 gene were the following: forward primer 5’-ACCGCACCTTCCAGTTCCAG-3’ and the reverse primer 5’-ATCACCGCCGCACCCAAG-3’ . The primers sequences for amplification of the LDLR gene were the following: forward primer 5’-TGCTCTGATGGAAACTGCATCC-3’ and the reverse primer 5’- AGAGTGTCACATTAACGCAGCC-3’ . The primers sequences for amplification of the VLDLR gene were the following: forward primer 5’-CTAGTCAACAACCTGAATGATG-3’ and the reverse primer 5’-AAGAATGGCCCATGCGGCAGAA-3’ . The primers sequences for amplification of the GAPDH gene were the following: forward primer 5’-ACCACAGTCCATGCCATCAC-3’ and the reverse primer 5’-TCCACCACCCTGTTGCTGTA-3’ . Fluorescent signals were analyzed with the ViiA7 system . The plasmid pSGR-JFH1 , pJFH1 and pSGR-JFH1-Nlucsec were linearized with XbaI , and treated with mung bean exonuclease . The linearized DNA was transcribed in vitro by using a MEGAscript T7 kit ( Life Technologies ) according to the manufacturer's protocol . Capped and polyadenylated firefly luciferase ( Fluc ) RNAs were synthesized by using a mMESSAGE mMACHINE T7 Ultra kit ( Life Technologies ) according to the manufacturer’s protocol . The in vitro transcribed RNA ( 5 μg ) was electroporated into cells at 5×106 cells/0 . 4 ml under conditions of 190 V and 950 μF using a Gene Pulser ( Bio-Rad ) and plated on DMEM containing 10% FCS . The medium was replaced with fresh DMEM containing 10% FCS and 1 mg/ml G418 at 24 h post-transfection of transcribed RNA . The remaining colonies were fixed with 4% paraformaldehyde ( PFA ) and stained with crystal violet at 1 month post-electroporation . Cell viability was determined by the Cell Titer-Glo Luminescent Cell Viability Assay ( Promega ) according to the manufacturer’s protocol and expressed in relative light units ( RLU ) at 24 , 48 and 72 h post-seeding . The NextBio Body Atlas application presents an aggregated analysis of gene expression across various normal tissues , normal cell types , and cancer cell lines . It enables us to investigate the expression of individual genes as well as gene sets . Samples for Body Atlas data are obtained from publicly available studies that are internally curated , annotated , and processed . Body Atlas measurements are generated from all available RNA expression studies that used Affymetrix U133 Plus or U133A Genechip arrays for human studies . The results from 128 human tissue samples were incorporated from 1 , 067 arrays; 157 human cell types from 1 , 474 arrays and 359 human cancer cell lines from 376 arrays . Gene queries return a list of relevant tissues or cell types rank-ordered by absolute gene expression and grouped by body systems or across all body systems . In the current analysis , we determined the expression levels of the SR-B1 , LDLR and VLDLR in tissues . The details of the analysis protocol developed by NextBio were described previously [67] . Culture supernatants of Huh7 . 5 . 1 cells infected with HCVcc at 72 h post-infection were passed through 0 . 45-μm-pore-size filters and concentrated by a Spin-X Concentrator ( 100 , 000-molecular-weight cutoff column; Corning , Lowell , MA ) . One milliliter of concentrated sample was layered onto the top of a linear gradient formed from 10% to 40% of OptiPrep ( Axis-Shield PoC ) in PBS and spun at 35 , 000 rpm for 16 h at 4°C by using an SW41-Ti rotor ( Beckman Coulter ) . Each fraction collected from the top was analyzed by qRT-PCR and focus-forming assay . HCVcc was bound to the cells for 1 h at 4°C , washed three times with PBS , and HCV RNA levels were determined by qRT-PCR immediately following binding . For lipid droplet staining , cells were incubated in medium containing 10μg/ml BODIPY for 20 min at 37°C , washed with prewarmed fresh medium , and incubated for 20 min at 37°C . Cells cultured on glass slides were fixed with 4% PFA in PBS at room temperature for 30 min . Cell nuclei were stained with DAPI . Cells were observed with a FluoView FV1000 laser scanning confocal microscope ( Olympus ) . Quantification of images was performed with ImageJ software ( National Institutes of Health ) . Cell surface proteins were biotinylated and purified by Cell Surface Protein Biotynylation and Purification Kit ( Thermo Fisher Scientific ) according to the manufacturer’s protocol . Total cell lysates and biotynylated proteins were applied in immunoblotting analysis . HDL and LDL labeled with 1 , 1’-Dioctadecyl-3 , 3 , 3 , 3’-tetramethylindocarbocyanine percholorate ( DiI-HDL and DiI-LDL ) were purchased from Alfa Aesar . Cells were washed with DMEM containing 0 . 5% fatty acid-free bovine serum albumin ( Sigma ) ( medium A ) , and medium A containing 5μg DiI-HDL or DiI-LDL was added to each well . After incubation for 2 h at 37°C , cells were washed twice with PBS and observed with a FluoView FV1000 laser scanning confocal microscope . HDL uptake was determined by using HDL Uptake Assay Kit ( Bio Vision ) according to the manufacturer’s protocol . LDL uptake was determined after incubation with medium containing 10μg/ml BODIPY FL LDL ( Thermo Fisher ) without FCS for 24 h at 37°C by using PowerScanHT ( DS Pharma Biomedical ) according to the manufacturer’s protocol . The data for statistical analyses are the averages of three independent experiments . Results were expressed as the means ±standard deviations . The significance of differences in the means was determined by Student’s t-test .
|
Hepatitis C virus ( HCV ) utilizes several receptors to enter hepatocytes , including scavenger receptor class B type 1 ( SR-B1 ) receptor and low-density lipoprotein receptor ( LDLR ) . HCV particles interact with lipoprotein and apolipoproteins to form complexes termed lipoviroparticles . Several reports have shown that SR-B1 and LDLR participate in the entry of lipoviroparticles through interaction with lipoproteins . However , the precise roles of SR-B1 and LDLR in HCV entry have not been fully clarified . In this study , we showed that SR-B1 and LDLR have a redundant role in HCV entry . In addition , we showed that very low-density lipoprotein receptor ( VLDLR ) played a role in HCV entry similar to the roles of SR-B1 and LDLR . Interestingly , VLDLR expression was low in the liver in contrast to the abundant expressions of SR-B1 and LDLR , but high in several extrahepatic tissues . Our data suggest that lipoprotein receptors participate in the entry of HCV particles associated with various lipoproteins .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"binding",
"cell",
"physiology",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"molecular",
"probe",
"techniques",
"hepacivirus",
"pathogens",
"gene",
"regulation",
"vector-borne",
"diseases",
"microbiology",
"immunoblotting",
"viruses",
"lipoprotein",
"receptors",
"rna",
"viruses",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"small",
"interfering",
"rnas",
"infectious",
"diseases",
"lipids",
"proteins",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"lipoproteins",
"hepatitis",
"c",
"virus",
"hepatitis",
"viruses",
"molecular",
"biology",
"biochemistry",
"signal",
"transduction",
"rna",
"cell",
"biology",
"flaviviruses",
"nucleic",
"acids",
"viral",
"pathogens",
"genetics",
"biology",
"and",
"life",
"sciences",
"non-coding",
"rna",
"organisms"
] |
2016
|
Lipoprotein Receptors Redundantly Participate in Entry of Hepatitis C Virus
|
Dengue is one of the most serious and rapidly spreading arboviral diseases in the world . Despite many acute febrile illnesses in Ethiopia , the burden of illness due to dengue in the country is largely unknown . Thus , the present study aimed to provide the first baseline data on seroprevalence and associated risk factors of dengue virus ( DENV ) infection in the country . A cross-sectional study of febrile patients who were visiting Metema and Humera hospitals in Northwest Ethiopia from March 2016 to May 2017 was conducted . Blood samples were collected from each participant and serum samples were separated and tested for IgM and IgG antibodies against DENV infection by enzyme-linked immunosorbent assay ( ELISA ) . Risk factors associated with the prevalence of anti-DENV antibodies were tested using logistic regression analysis . Of the 600 samples tested , the overall seroprevalence against DENV infection was 33 . 3% , while the seroprevalence by the study area was 40% in Metema and 27 . 5% in Humera . The overall prevalence of IgM and IgG antibodies against DENV infection was 19% and 21% respectively . Of these , 6 . 7% were positive for both IgM and IgG antibodies . Residence and occupational status were significantly associated with the prevalence of anti-DENV IgM seropositivity and anti-DENV IgM-/G+serostatus . The seasonal variation was significantly associated with the prevalence of anti-DENV IgM but not with anti-DENV IgM-/G+serostatus . The prevalence of anti-DENV IgM-/G+serostatus was significantly higher in Metema than Humera . High prevalence of anti-DENV IgM seropositivity was found in the summer and spring , with a peak in the month of August . The presence of uncovered water either indoor or outdoor and lack of mosquito net use was identified as risk factors for DENV infection . These findings provide the preliminary data on seroprevalence and associated risk factors of DENV infection in the country . The presence of antibodies against DENV infection indicates dengue as one of the causes of undifferentiated febrile illnesses in the study areas . This suggests that prevention and control measures should be designed considering the risk factors identified by this study . Furthermore , we recommend a large-scale study to include DENV infection in the differential diagnosis of all febrile illnesses in Ethiopia .
Dengue is one of the most important mosquito-borne viral diseases and can be caused by any one of the four dengue virus serotypes ( DENV1-4 ) of the genus Flavivirus [1] . Dengue virus is a non-segmented , positive-sense , single-stranded , enveloped RNA virus; transmitted mainly by the bite of Aedes aegypti , a tropical and subtropical mosquito species that inhabits mostly in urban areas in proximity to houses [2] . It breeds in small bodies of fresh water , most commonly in various containers found around homes [3] . It is estimated that 390 million DENV infections occur every year [4] , and it is endemic in more than 100 countries across the Americas , East Mediterranean , Western Pacific , Africa and South-East Asia and Europe [5] . In Africa , dengue has been reported in 34 countries , mostly in Eastern Africa [6] . In the countries bordering Ethiopia such as Sudan , Eritrea , Kenya and Djibouti dengue has also been reported [7–11] . Four of the DENV serotypes have been detected in Africa , with the DENV-2 serotype reported to cause the majority of epidemics [6] . The DENV-2 serotype and the main dengue vector , Ae . aegyptus were recently reported in Dire Dawa Ethiopia [12]; even prior to the DENV outbreak the presence of Ae . aegypti in different regions of Ethiopia had been reported [13 , 14] . Although dengue has a global distribution , the vast majority of the data on DENV infections are from the WHO South-East Asia and Western Pacific Regions [2] . In Ethiopia , prior to 2013 , there were no reports of dengue . However , recently , DENV infection has emerged in Dire Dawa and Somalia regions [12 , 15] . In the present decade , dengue has expanded into new countries where it had not existed earlier [2] . The expansion of dengue is expected to increase due to several factors such as human population growth , climate change , and increased urbanization with sub-standard housing , irregular water supply , and poor environmental sanitation . Together with increased mobility of both vectors and human populations all over the world , further spread of dengue from the endemic areas to many previously unaffected areas is anticipated [16] . Dengue is an acute febrile illness , which occurs after an incubation period of 4–10 days . Dengue disease severity varies from asymptomatic infection to a variety of illnesses ranging from an influenza-like self-limiting illness to a potentially lethal disease such as dengue hemorrhagic fever ( DHF ) or dengue shock syndrome ( DSS ) [17 , 18] . It is a complex disease with various clinical presentations , which often go unrecognized or misdiagnosed as other common fever-causing tropical diseases . For instance , malaria is endemic throughout the African region including Ethiopia; the majority of the febrile illnesses including dengue , are likely to be misdiagnosed and treated as presumptive malaria [6] . Because of this , dengue can progress from a mild , nonspecific viral disease to severe disease . Without good clinical management , mortality from the complications of DHF/DSS can be as high as 20 percent , whereas if the case is recognized early and properly managed , mortality due to the complications declines to less than 1 percent [19 , 20] . Hence , early and rapid laboratory diagnosis of dengue is important for proper management and prevention of complications like DHF/DSS [20 , 21] . Studies in Ethiopia have indicated that the unknown causes of acute febrile illnesses are high [22] . Among the causes of non-malaria febrile diseases around the globe , DENV infection is currently considered as the leading cause of febrile illness [23] . However , data are not available on the prevalence of dengue in Northwest Ethiopia . Considering the current situation of DENV infection and unavailability of data in the country , this study was therefore conducted to document the first baseline seroprevalence and risk factors associated with DENV infection in the country . The present study will be helpful in providing information on DENV infection to healthcare authorities for better clinical management of patients and to design and implement appropriate control measures .
The study was reviewed and approved by the Ethical Review Committees of the University of Gondar and Armauer Hansen Research Institute . The study participants were informed about the study before collecting any data or samples . Written informed consent was obtained from each participant or assent from each parent/guardian of the children . Blood samples were collected by experienced laboratory technologist as part of the routine sample collection . Participants had full right to continue or withdraw from the study . The confidentiality of all participants was maintained throughout the study . The study was carried out in Northwest Ethiopia; Metema and Humera Kahsay Abera hospitals ( Fig 1 ) . The Metema hospital is located in Northwest Ethiopia on the border with Sudan , 897 km North of Addis Ababa and 197 km from the ancient city of Gondar . This town has a latitude and longitude of 12°58′N 36°12′E with an elevation of 685 m above sea level . The Humera Kahsay Abera hospital is also located in Northwest Ethiopia , 252 km from Gondar city and 974 km from Addis Ababa and located in the western zone of the Tigray Regional state , bordered on the west by Sudan , and on the north by the Tekezé River which separates Ethiopia from Eritrea . This town has a latitude and a longitude of 14° 18’N36° 37’E with an elevation of 602 m above sea level . Both Metema and Humera areas are one of the hottest and malaria-endemic areas in the country and the most fertile agricultural zones with the large scale of farming of cash crops . A cross-sectional hospital-based study was carried out among febrile patients attending the two hospitals from March 2016 to May 2017 . The sample size was estimated to be 600 by using single proportion formula at 95% confidence interval , an expected prevalence of 50% and 4% marginal error . The study participants were all febrile patients who have been presumed for DENV infection . Dengue was presumed in patients who live or traveled to an endemic area and presented with fever and two of the following criteria such as nausea or vomiting , rash , aches and pains , tourniquet test positive , leukopenia , and any warning sign based on WHO-TRD 2009 criteria; further suggested dengue cases based on the levels of severity were classified as dengue without warning signs , dengue with warning signs , and severe dengue . Dengue without warning signs was defined as laboratory-confirmed dengue cases without signs of plasma leakage . Dengue with warning signs included abdominal pain or tenderness , clinical fluid accumulation , persistent vomiting , lethargy , restlessness , mucosal bleeding , liver enlargement >2 cm , an increase in hematocrit ( HCT ) concurrent with a rapid decrease in platelet count . Severe dengue; in addition to aforementioned criteria , includes signs of severing plasma leakage and/or severe bleeding , severe organ impairment [2] . A febrile patient was defined as a patient who came to either the outpatient or inpatient department and to either the pediatric or medicine unit at the participating hospital with fever ≥ 38°C . Prevalence of DENV infection was defined as the proportion of participants with IgM and /or IgG ELISA positive . Study participants who met the case definition were screened by clinicians . When a participant was willing to participate in the study , clinical features , demographic information , and risk factors were collected by a nurse using a structured questionnaire that underwent validation and editing after small pilot study . A 3 to 5ml of blood was collected from each eligible study participant by venepuncture into a new sterile plain test tube to obtain serum . The blood samples were centrifuged for 5 min , aliquoted and stored at –20°C until processed . Serum samples were tested for IgM and IgG antibodies to DENV with ELISA ( EUROIMMUN ) . The anti-DENV antibodies ( IgM and IgG ) ELISA were performed as per the manufacturer's instructions [24] . Data were entered and analyzed using SPSS version 20 software . Simple frequency tables were generated , and categorical variables were compared using chi-square test . A univariate logistic regression analysis was used to identify risk factors associated with the prevalence of anti-DENV IgM and IgG antibodies . Those independent variables found p < 0 . 2 in univariate analysis were then used in multivariate logistic regression analysis . Odds ratios ( ORs ) at 95% confidence intervals ( CIs ) were calculated to measure the degree of association . A p-value < 0 . 05 was considered as statistically significant and data were presented in the form of tables and figures .
All the study participants who have fulfilled the inclusion criteria of the study were enrolled and completed the study . A total of 600 febrile patients who were presumed for DENV infection were enrolled in this study . Of these , 394 ( 65 . 7% ) of the participants were males and the mean age of the participants was 25 years , ranging from 1 to 78 years . The majority of the participants 302 ( 50 . 3% ) were between 15 to 29 years of age groups , while the 68 ( 11 . 3% ) were greater or equal to 45 years of age groups . Three hundred sixteen ( 52 . 7% ) study participants were urban dwellers . Most of the participants 301 ( 50 . 2% ) were farmers , followed by 165 ( 27 . 5% ) students . Among the study participants , 184 ( 30 . 7% ) were illiterate ( Table 1 ) . Of the 600 study participants screened for DENV specific IgM and IgG antibodies , 114 ( 19% ) were positive for IgM and 126 ( 21% ) for IgG and 40 ( 6 . 7% ) samples were positive for both IgM and IgG ( Table 2 ) . This showed a total of 200 ( 33 . 3% ) positive cases for IgM/G antibodies . Of the total study participants , 74 ( 12 . 3% ) study participants had IgM+/G- and 86 ( 14 . 3% ) had IgM-/G+ serostatus . As per WHO-TDR ( Tropical Diseases Research ) 2009 guidelines ( 2 ) ; among the 114 anti-DENV IgM seropositive patients , 84 ( 73 . 7% ) were classified as dengue without warning signs , 30 ( 26 . 3% ) as dengue with warning signs , and none of the patients were classified as severe dengue ( Table 3 ) . Out of 600 study participants , the overall seroprevalence of DENV infection was 33 . 3% ( 200/600; 95%CI , 29 . 7–37 . 3 ) . Of this , Metema accounts 56% ( 112/200 ) and Humera 44% ( 88/200 ) of the cases . The seroprevalence by study area was 40% ( 112/280 ) in Metema hospital and 27 . 5% ( 88/320 ) in Humera Kahsay Abera hospital ( Fig 2 ) . None of the study participants had been vaccinated against yellow fever . The month-wise distribution of anti-DENV antibodies showed a high proportion of only IgM in June to November with the peak in the month of August , which is conceded with the monsoon and post-monsoon periods while from December to May its prevalence was law and irregular . The prevalence’s of both anti-DENV IgM and IgG antibodies were irregular with high prevalence in the months of August and October . Seropositivity of only anti-DENV IgM was detected in almost every month of the year except January while only anti-DENV IgG was detected in each month of the year . Seropositivity of both anti-DENV IgM and IgG antibodies were detected in every month of the year except September and November ( Fig 3 ) . In a univariate analysis , gender , residence , occupational status and season showed statistically significant association with the prevalence of anti-DENV IgM seropositivity ( p-value < 0 . 05 ) while age , study area , the marital and educational status of the study participants were not significantly associated . In multivariate analysis , occupational status and season remained significantly associated with the prevalence of anti-DENV IgM seropositivity . Those study participants who lived in urban areas were 1 . 8 times ( AOR = 1 . 85; 95% CI = 1 . 18–2 . 89 ) more likely to have anti-DENV IgM seropositivity than those who lived in rural areas; farmers were 2 times ( AOR = 2 . 03; 95% CI = 1 . 14–3 . 61 ) more likely to have anti-DENV IgM seropositivity than the students . During summer; June to August , 3 . 2 times ( AOR = 3 . 22; 95%CI = 1 . 58–6 . 56 ) higher anti-DENV IgM seropositivity was observed , followed by spring; September to November ( which is the monsoon and post-monsoon period in Ethiopia ) than winter; December to February ( Table 4 ) . Univariate logistic regression analysis of the investigated variables showed statistically significant associations ( p-value < 0 . 05 ) with the prevalence of anti-DENV IgM-/G+ serostatus was observed with age , gender , residence , occupational status and the study area . In multivariate logistic regression analysis , gender , residence , occupational status , and the study area were significantly associated with the prevalence of anti-DENV IgM-/G+ serostatus . Males had 2 times ( AOR = 2 . 05; 95% CI = 1 . 14–3 . 69 ) higher prevalence of anti-DENV IgM-/G+ serostatus than females . Urban residents had 2 . 3 times ( AOR = 2 . 31; 95% CI = 1 . 38–3 . 87 ) higher prevalence of anti-DENV IgM-/G+ serostatus than rural residents . Farmers had 2 . 8 times ( AOR = 2 . 81; 95% CI = 1 . 23–6 . 41 ) higher prevalence of anti-DENV IgM-/G+ serostatus than students . Study participants who were visited Metema hospital had 1 . 7 times ( AOR = 1 . 75; 95%CI = 1 . 05–2 . 89 ) higher prevalence of anti-DENV IgM-/G+ serostatus than those visited Humera hospital ( Table 5 ) . In a univariate analysis lack of mosquito net use , the presence of uncovered water storage either indoor or outdoor and history of travel abroad were significantly associated with the prevalence of anti-DENV IgM seropositivity . However , use of mosquito repellant , indoor insecticidal spraying , keeping the animal at home , slaughtering an animal , seasonal migrant laborer and a place where to sleep were not significantly associated ( p-value > 0 . 05 ) . In multivariate analysis , history of travel to abroad was not significantly associated with the prevalence of anti-DENV IgM seropositivity while lacking use of a mosquito net and the presence of uncovered water storage either indoor or outdoor remained significantly associated . Lack of mosquito net use was 1 . 7 times ( AOR = 1 . 75; 95% CI = 1 . 00–3 . 06 ) higher risk of DENV infection than those who have used it . Those study participants who stored water without cover either indoor or outdoor had 1 . 6 times ( AOR = 1 . 60; 95% CI = 1 . 05–2 . 43 ) higher risk of DENV infection than those who didn’t store it ( Table 6 ) . In the univariate and multivariate analysis , lack of mosquito net use and the presence of uncovered water storage either indoor or outdoor were significantly associated with the prevalence of anti-DENV IgM-/G+ serostatus . Whereas , the use of mosquito repellent , indoor insecticidal spraying , keeping the animal at home , slaughtering an animal , seasonal migrant laborer , history of travel to abroad and a place where to sleep were not significantly associated with the prevalence of anti-DENV IgM-/G+ serostatus ( Table 7 ) .
This is the first health institution based study that provides evidence on seroprevalence of DENV infection into two towns of Northwest Ethiopia . The results of this study showed a seroprevalence of DENV infection of 40% in Metema and 27 . 5% in Humera . Several factors favor transmission of DENV in the study areas . Such as the existing climatic conditions ( i . e . , high temperature ) , providing the optimal environmental and biological circumstances for vector mosquito breeding and reproduction , and also increased urbanization in the study areas might favor the emergence and survival of DENV infected Aedes mosquitoes [16] . The variations of seroprevalence between the study areas might be due to differences in virus circulation inhabitants of these areas which correspond to the area with low virus circulation might have had low transmission of the virus [25] . Dengue is a preventable disease that causes significant morbidity and mortality in most tropical and sub-tropical countries of the world [26–29] . The evidence generated here are essential not only for patient management but also for undertaking early prevention and control interventions; such as reduction of mosquito breeding in household water containers using larvicides , or elimination of discarded containers , and control of adult mosquitoes by spraying insecticides or prevent them from biting using repellent [2] . The overall seroprevalence of DENV infection in the present study was 33 . 3% , which is lower than the recent study in Dire Dawa Ethiopia , 56 . 8% [12] . The difference of the prevalence rates between the two studies might be due to variations in dengue positive considerations; the previous study considered dengue positives either by ELISA/PCR while the current study considered dengue positives only by ELISA test . A similar result was reported in Eritrea 33 . 3% [9] . However , this result is higher than the findings in Tanzania 7 . 7% [30] , in Kenya , 12 . 5% [10] , in Djibouti , 21 . 8% [11] , in the Northern Province of Sudan , 24% [8] , and lower than the study in Thailand , 51 . 5% [31] , and in Kassala , Eastern Sudan , 71 . 7% [7] . The variations might be due to the differences in the environmental factors such as temperature , rainfall , and humidity which affect dengue transmission [32] . Despite one-third of the study participants had antibody against DENV infection , dengue was underrecognized and underreported in Ethiopia , which is in line with an earlier report in Africa [6] . The possible reasons for underrecognition of the DENV infection in Ethiopia might be due to the similarity of most of the clinical presentations of DENV infection with the other febrile illnesses and the lack of documented previous data on the occurrence of DENV which provides awareness of the disease among health providers . The overall prevalence of anti-DENV IgM seropositivity was 19% , while anti-DENV IgM+/G- serostatus was 12 . 3% , likely indicates recent infection with DENV . Since IgM against the DENV infection can be usually detected after the first 5 days of infection and peaks approximately 14 days after the onset of disease and may persist up to 3 months [33 , 34] . However , the possibility that as the IgM antibodies remain negative for the first few days , and also the IgM reactivity was non-specific; is thus cross-reactive due to infection with another flavivirus cannot be excluded [35] . The prevalence of anti-DENV IgM+/G+ serostatus was 6 . 7% . This might indicate that it is likely that the person became infected with DENV within recent weeks; as IgM response is usually detected after the 5 days of illness , and followed by IgG response [34] . The prevalence of anti-DENV IgM-/G+ serostatus was 14 . 3% , might indicate probably past infection with DENV . This seroprevalence may be overestimated by false positive results; since we could not collect paired samples , the presence of anti-DENV IgM-/G+ might be confounded with the issue of cross-reactivity of other anti-flavivirus antibodies [35] . However , recently in Dire Dawa Ethiopia , of serum samples tested for arboviruses , it was DENV which has been reported [12] , this suggests that cross-reactivity with antibodies from other arboviruses might be less likely in the present study . According to 2009 WHO-TDR classification scheme [2] , suggested dengue cases have been classified into two categories , dengue without warning signs and dengue with warning signs . Similar findings were reported elsewhere [36] , but other studies reported previously three of the categories including severe dengue cases [37 , 38] . In this study severe dengue cases were not noticed , this might be due to probably primary infection with any one of the four DENV serotypes which are usually associated with mild disease . Since it is a subsequent infection of DENV with a second heterologous serotype which increases the severity of dengue [39 , 40] . The immune basis for this hypothesis suggests that antibodies developed during the primary infection instead of mediating viral clearance , assist the new virus in infecting host macrophages , a phenomena known as antibody-dependent enhancement [41] . In this study , anti-DENV IgG seropositivity was observed in every month of the year . This suggests potential endemicity of DENV infection; as studies illustrated IgG against the DENV infection might have persisted for several decades [42] . Previous vaccination against yellow fever may render the anti-DENV IgG ELISA test less specific because of cross-reaction [43] . However , in the present study , none of the study participants had been vaccinated against yellow fever; hence natural antibody response which occurs after anti-yellow fever vaccination is unlikely . In the present study , the analysis of the seasonal variation was significantly associated with the prevalence of anti-DENV IgM seropositivity but not with anti-DENV IgM-/G+ serostatus . The highest prevalence of anti-DENV IgM seropositivity observed during summer ( June to August ) , followed by spring ( September to November ) , and indicated the presence of high viral transmission during monsoon and post-monsoon periods . These findings are consistent with the other studies [44 , 45] . The study suggests effective vector control and preventive measures to be implemented during water stagnation periods after the initial bouts of rainfall and at the end of monsoon which corresponds to the breeding time of the dengue vectors [46] . Findings from this study showed the prevalence of anti-DENV IgM-/G+ serostatus was significantly higher in males than in females , which is in agreement with the study elsewhere [47] while others reported no significant difference between genders [48] . Although the prevalence of anti-DENV IgM seropositivity was not significantly associated with gender , its prevalence in males was higher than in females , which is in line with other studies [9 , 47 , 49] . The higher proportion of anti-DENV IgM seropositivity in males might be associated with DENV vectors , Ae . aegypti and Ae . albopictus breeding sites which are more abundant in outdoors than indoors [50] . In the study areas , it is assumed that males might have greater involvement in outdoor activities than females , and thus possibly increase exposure to day feeding Ae . aegypti mosquito , which is the main vector of DENV [2] . Concerning study area , the prevalence of anti-DENV IgM seropositivity was not significantly associated while anti-DENV IgM-/G+ serostatus was significantly higher in Metema than Humera . This is indicative of an intense and long-lasting exposure of population who lived in Metema area to DENV infection risks . With regards to age groups , studies elsewhere reported that children were the most dengue affected subpopulation [51 , 52] . However , in this study anti-DENV , seropositivity was noticed in all age groups with none significantly higher seroprevalence that was observed in age groups of 30–45 years old . These findings are consistent with other studies [12 , 53] . This may be due to the increased possibility of being exposed to dengue vector bite as age increased and becoming seropositive for DENV infection over the lifetime of the individual [42 , 54] . Educational status of the participants in this study was not significantly associated with anti-DENV antibodies seropositivity and this is consistent with another study [48] . This could be due to the fact that previously , the disease is not known in the areas and even not included under the lists of nationally reportable diseases which daunt to use any preventive measures of DENV infection . In this study , those individuals who reside in urban areas were more affected than those in the rural areas . This is in agreement with the studies elsewhere [55 , 56] , and this is also consistent with the current knowledge on the dengue main vector , Ae . aegypti increases in urban environments in that it breeds mainly in the artificial containers often used in urban water collection [16] . There was also a significant association between occupations of the study participants with anti-DENV antibodies seropositivity . Farmers were more affected than other groups with a different occupation . Since the vector is active at daytime and mostly prefers to rest in the shade of trees or nearby buildings . Because of the high temperature in our study areas , most of the time , farmers take rest in the shade of trees which increase exposure to day feeding outdoor Aedes mosquitoes . In the present study , participants who had lack of mosquito net use during sleeping were found to be more at risk of getting DENV infection . This could be due to sleeping without mosquito net increases the contact between mosquitoes and humans , and thus increase transmission of mosquito-borne diseases including dengue [2] . However , study elsewhere showed that no significant difference between mosquito net users and none users [48] . The variations might be probably differences in the quality of mosquito net and the frequency of impregnating nets with chemicals . The presence of stagnant water either indoor or outdoor was identified as the risk factor of dengue , which is consistent with the other studies in elsewhere [57 , 58] . This finding may be explained by the fact that long-term storage of water in opened containers favors breeding of the mosquito vector . This results in an increase in dengue cases , and thus frequent emptying of long-term stagnant water storage inside and outside the house is recommended to reduce mosquito breeding sites [2] . Although this is the first attempt to study a seroprevalence and risk factors associated with DENV infection in Northwest Ethiopia , the study has several limitations . A comparison of the acute serum with the convalescent serum from the same patients was not done due to the nature of a cross-sectional study design . The study was conducted among febrile patients who were attending only health institutions that may not necessarily reflect the true seroprevalence at the community level where some mild or asymptomatic infection might occur . There was no information on the median duration of fever for enrolled subjects , and also RT-PCR and plaque reduction neutralization tests ( PRNT ) were not done due to lack of facilities . These could be done in the future for more accurate estimation of overall prevalence and for identification of circulating serotypes and genotypes . Moreover , this study has also limitations including the presence of a possibility of false negative dengue cases due to IgM antibodies remains negative for the first few days of fever and thus we cannot state that these subjects with negative IgM did not have acute dengue . There is also a possibility of some false positive cases due to cross-reactivity of other anti-flavivirus antibodies with DENV . Despite these limitations , this preliminary study ultimately provides the first baseline data on seroprevalence of DENV infection and associated risk factors in the country . In conclusion , one-third of patients who presented with fever presumed of dengue had antibodies against DENV infection; this should alert all concerned parties within the health sectors . The prevalence of anti-DENV IgM seropositivity was significantly associated with residence , occupation and season . In addition to this , the presence of uncovered water storage either indoor or outdoor was identified as the risk factor of DENV infections . Therefore , we recommend that the prevention strategies and control measures should be designed in the country considering the risk factors identified by this study . Moreover , nationwide surveillance should be done at large to include the dengue in the differential diagnosis of all febrile cases , in Ethiopia .
|
Despite dengue is currently one of the leading causes of arboviral diseases in the globe , it is unrecognized and underreported in Africa , particularly in Ethiopia . Thus , we conducted a cross-sectional study among febrile patients who were attending health institutions to document seroprevalence and associated risk factors of DENV infection in the country . The study illustrated the presence of antibodies against DENV infection for the first time in both study areas , an awaking message for those who were involved in health sectors . Most of the active DENV transmission was found in monsoon and post-monsoon periods with a peak in the month of August . In multivariate analysis residence , occupational status and seasonal variations were significantly associated with the prevalence of anti-DENV IgM seropositivity . Moreover , individuals who lack a mosquito net use and the presence of uncovered water storages either indoors or outdoors were identified as the risk factors of DENV infection . Therefore , we recommend that preventive measures should be considered . Moreover , nationwide surveillance should be carried out at large .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"dengue",
"virus",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"enzyme-linked",
"immunoassays",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"geographical",
"locations",
"microbiology",
"tropical",
"diseases",
"animals",
"viruses",
"multivariate",
"analysis",
"mathematics",
"rna",
"viruses",
"statistics",
"(mathematics)",
"neglected",
"tropical",
"diseases",
"ethiopia",
"antibodies",
"immunologic",
"techniques",
"insect",
"vectors",
"africa",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"proteins",
"medical",
"microbiology",
"dengue",
"fever",
"mathematical",
"and",
"statistical",
"techniques",
"immunoassays",
"microbial",
"pathogens",
"arboviral",
"infections",
"disease",
"vectors",
"insects",
"arthropoda",
"people",
"and",
"places",
"biochemistry",
"mosquitoes",
"eukaryota",
"flaviviruses",
"physiology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"viral",
"diseases",
"species",
"interactions",
"statistical",
"methods",
"organisms"
] |
2018
|
A serologic study of dengue in northwest Ethiopia: Suggesting preventive and control measures
|
Spinocerebellar ataxia type 2 ( SCA2 ) is an autosomal dominant disorder with progressive degeneration of cerebellar Purkinje cells ( PCs ) and other neurons caused by expansion of a glutamine ( Q ) tract in the ATXN2 protein . We generated BAC transgenic lines in which the full-length human ATXN2 gene was transcribed using its endogenous regulatory machinery . Mice with the ATXN2 BAC transgene with an expanded CAG repeat ( BAC-Q72 ) developed a progressive cellular and motor phenotype , whereas BAC mice expressing wild-type human ATXN2 ( BAC-Q22 ) were indistinguishable from control mice . Expression analysis of laser-capture microdissected ( LCM ) fractions and regional expression confirmed that the BAC transgene was expressed in PCs and in other neuronal groups such as granule cells ( GCs ) and neurons in deep cerebellar nuclei as well as in spinal cord . Transcriptome analysis by deep RNA-sequencing revealed that BAC-Q72 mice had progressive changes in steady-state levels of specific mRNAs including Rgs8 , one of the earliest down-regulated transcripts in the Pcp2-ATXN2[Q127] mouse line . Consistent with LCM analysis , transcriptome changes analyzed by deep RNA-sequencing were not restricted to PCs , but were also seen in transcripts enriched in GCs such as Neurod1 . BAC-Q72 , but not BAC-Q22 mice had reduced Rgs8 mRNA levels and even more severely reduced steady-state protein levels . Using RNA immunoprecipitation we showed that ATXN2 interacted selectively with RGS8 mRNA . This interaction was impaired when ATXN2 harbored an expanded polyglutamine . Mutant ATXN2 also reduced RGS8 expression in an in vitro coupled translation assay when compared with equal expression of wild-type ATXN2-Q22 . Reduced abundance of Rgs8 in Pcp2-ATXN2[Q127] and BAC-Q72 mice supports our observations of a hyper-excitable mGluR1-ITPR1 signaling axis in SCA2 , as RGS proteins are linked to attenuating mGluR1 signaling .
Spinocerebellar ataxia type 2 ( SCA2 ) belongs to the group of neurodegenerative diseases caused by polyglutamine ( polyQ ) expansion . This group includes SCA1 , Machado-Joseph disease ( SCA3 or MJD ) , SCA6 , SCA7 , SCA17 , Huntington's disease , spinal bulbar muscular atrophy ( SBMA ) and dentatorubral-pallidoluysian atrophy ( DRPLA ) . SCA2 is an autosomal dominant disorder leading to motor incoordination which is caused by progressive degeneration of cerebellar Purkinje cells , and selective loss of neurons within the brainstem and spinal cord [1] . As with most autosomal dominant ataxias , symptoms are characterized by a progressive loss of motor coordination , neuropathies , slurred speech , cognitive impairment and loss of other functional abilities arising from Purkinje cells and deep cerebellar nuclei [2 , 3] . In SCA2 , expansion of a CAG repeat in exon 1 of the Ataxin-2 ( ATXN2 ) gene causes expansion of a polyQ domain in the ATXN2 protein . As in the other polyQ diseases , the length of the polyQ repeat is inversely correlated with age of onset ( AO ) in SCA2 [1 , 4] . In contrast to other polyQ diseases , mutant ATXN2 does not enter the nucleus in appreciable amounts in early stages of disease . This is also confirmed by protein interaction studies that have identified ATXN2 interactors with cytoplasmic localization [5–8] . Polyglutamine disorders show their pathology through a toxic gain of function of the protein and larger polyQ expansions have been associated with greater pathology [3 , 9] . ATXN2 is widely expressed in the mammalian nervous system [1 , 10 , 11] . It is involved in regulation of the EGF receptor [12] , and the inositol 1 , 4 , 5-triphosphate receptor ( IP3R ) whereby increased cytosolic Ca2+ occurs with CAG repeat expansion [13] . ATXN2 functions are also associated with the endoplasmic reticulum [14] , and the Golgi complex [15] . Studies in Caenorhabditis elegans support a role for ATXN2 in translational regulation as well as embryonic development [6] . ATXN2 is also important in energy metabolism and weight regulation , as mice lacking Atxn2 , developed obesity and insulin resistance [16 , 17] . Furthermore , ATXN2 interacts with multiple RNA binding proteins , including polyA binding protein 1 ( PABP1 ) , the RNA splicing factor A2BP1/Fox1 , DDX6 , TDP-43 , and has been localized in polyribosomes and stress granules demonstrating its unique role in RNA metabolism [5 , 6 , 8 , 18] . Several SCA2 mouse models have been generated . We have reported two transgenic mouse models in which expression of full-length ATXN2 with 58 or 127 CAG repeats ( ATXN2-[Q58] or ATXN2-[Q127] ) is targeted to Purkinje cells ( PCs ) using the Purkinje cell protein-2 ( Pcp2 ) promoter [19 , 20] . These lines show progressive motor phenotypes accompanied by the formation of insoluble cytoplasmic aggregates , loss of PCs , and shrinkage of the molecular layer associated with the reduction of calbindin staining in PC bodies and dendrites . Onset of the motor phenotype of Pcp2-ATXN2[Q127] mice is associated with reduced PC firing that is progressive with age [20] . Another Atxn2-CAG42 knock-in mouse model demonstrated very late-onset motor incoordination associated , but this was seen only in homozygous knock-in animals . This was associated with Pabpc1 deficiency , and upregulation of Fbxw8 , but without loss of calbindin staining or downregulation of Calb1 mRNA [21] . In order to model human diseases using cis-regulatory elements , recent mouse and rat models have been created by transgenesis using human bacterial artificial chromosomes ( BACs ) [22–27] . In the BAC approach , an entire human gene including introns and regulatory regions is introduced into the mouse genome . BAC models often have lower genomic copy numbers than conventional cDNA transgenic models resulting in more physiological expression levels and a potentially more faithful late onset of disease . We developed new BAC-SCA2 transgenic mouse lines expressing full-length human wild-type or mutant ATXN2 genes including upstream and downstream regulatory sequences . BAC mice with mutant ATXN2 exhibited progressive neurological symptoms and morphological changes in cerebellum . We used this mouse model to confirm changes in key PC-genes identified in a cDNA transgenic model , in particular the effects of mutant ATXN2 on Rgs8 steady state protein levels .
To understand the pathological and behavioral effects in the context of physiologic expression of human wild-type and mutant ATXN2 , we engineered a 169 kb human BAC ( RP11-798L5 ) that contained the entire 150 kb human ATXN2 locus with 16 kb of the 5’ flanking genomic sequence and 3 kb of the 3’ flanking genomic sequence ( Fig 1A ) . The authenticities of these constructs were subsequently verified by Southern blot and restriction site analyses ( S1 Fig ) . The CAG tract was mutation-free when sequenced from both strands . After transgenic microinjection of purified intact BAC DNAs , one line each for control ( BAC-ATXN2-Q22 ) and one for mutant mice ( BAC-ATXN2-Q72 ) was further analyzed . These lines will be designated as BAC-Q22 and BAC-Q72 in the remainder of the text . Quantitative PCR ( qPCR ) analyses of genomic DNA revealed that both BAC-Q22 and BAC-Q72 mice had tandem integrates of 10 and 4 copies of the ATXN2 transgene , respectively . In RT-PCR analyses , both BAC-Q22 and BAC-Q72 mice demonstrated the expression of intact human ATXN2 transcripts throughout the central nervous system ( CNS ) , including cerebral hemispheres , cerebellum and spinal cord ( Fig 1B ) . Non-CNS tissues , including heart and liver also showed ATXN2 transgene expression ( Fig 1B ) . The authenticities of PCR products were confirmed by sequencing . We further determined relative expression of ATXN2 transcripts in the two BAC transgenic lines by quantitative RT-PCR . BAC-Q22 cerebella had higher expression of human ATXN2 than BAC-Q72 cerebella while the expression of endogenous mouse Atxn2 remained unchanged in both compared with wild-type mice ( Fig 1C ) . To assess protein expression , we performed Western blot analysis using cerebellar extracts of 16 week-old animals and a monoclonal antibody ( mAb ) to human ATXN2 . The results showed that BAC mice expressed full-length human wild-type or mutant ATXN2 protein . Of note , protein levels of ATXN2-Q22 were higher than those of ATXN2-Q72 . Furthermore , we confirmed the ATXN2-Q72 protein expression using 1C2 mAb , an antibody against an expanded polyQ epitope in Western blot analyses ( Fig 1D ) . These results demonstrate that human ATXN2 transgenes ( ATXN2-Q22 and ATXN2-Q72 ) were properly expressed in BAC mice . In addition to ATXN2 , three overlapping genes ( U7 . 1–202 snRNA , RP11-686G8 . 1–001 and RP11-686G8 . 2–001 ) are contained in the human BAC . Quantitative RT-PCR analyses of wild-type and BAC transgenic mouse cerebellar RNAs demonstrated that the relative expression of each overlapping gene to that of the ATXN2 transgene did not differ between BAC-Q22 and BAC-Q72 animals indicating these genes did not contribute to the phenotypes associated with CAG expansion in the ATXN2 gene ( S2 Fig ) . The Allen Brain Atlas shows widespread expression of human ATXN2 with very significant expression levels in the cerebellum [28] . Given the nature of ATXN2 expression in brain , we determined the expression of human ATXN2 transgene transcript in sub-regions of mouse brain including spinal cord using qRT-PCR . Expression of endogenous mAtxn2 was evident in many regions including frontal , occipital and olfactory cortex , hippocampus , thalamus , basal ganglia , cerebellum and spinal cord . Human ATXN2 transgene expression was found in all regions tested , but relatively higher expression was observed in the basal ganglia ( S3 Fig ) . As cerebellar degeneration is predominant in SCA2 , we further examined the expression patterns of the ATXN2 transgene in discrete areas of the cerebellum using laser-capture microdissection ( LCM ) . We captured molecular layer ( ML ) , Purkinje cells ( PCs ) , granule cell layer ( GCL ) and dentate nuclear ( DN ) fractions . Relative enrichment was determined by measuring expression of a cell-type specific marker genes using qRT-PCR . Evidence for expression of endogenous mAtxn2 was found in all fractions , but was highest in Purkinje cells . Expression of transgenic ATXN2 was also seen in all fractions , although small differences in expression levels existed between BAC-Q22 and BAC-Q72 ( Fig 2A and 2B ) . LCM was remarkably successful in separating cerebellar neuronal population as shown by expression of marker genes for PCs and molecular layer ( Pcp2 and Calb1 ) , granule cells ( Neurod1 ) and dentate neurons ( Spp1 ) ( Fig 2C and 2F ) . In summary , inclusion of regulatory regions in the human BAC transgene led to expression of the transgene that mirrored expression of mouse Atxn2 including low but detectable expression in GCs and DNs . By visual inspection both BAC transgenic lines ( BAC-Q22 and BAC-Q72 ) had a smaller body size than wild-type littermates beginning at 8 weeks of age . By 24 weeks of age , both BAC transgenic mice weighed about 30% less than their wild-type littermates ( Wild-type = 33 . 9 ±3 . 8; BAC-Q22 = 24 . 6 ±3 . 6 and Wild-type = 32 . 1 ±2 . 8; BAC-Q72 = 22 . 9 ±3 . 7 ) . BAC-Q72 mice did not show an abnormal home cage behavior . To assess the development of motor impairment , both BAC transgenic lines and wild-type littermates were tested using the accelerating rotarod paradigm at several time points ( Fig 3 ) . BAC-Q22 mice performed as well as wild-type littermates at 8 , 16 and 36 weeks of age ( Fig 3 ) suggesting that expression of wild-type human ATXN2 was not detrimental to motor function . BAC-Q72 mice were tested at 5 , 16 and 36 weeks of age and compared with their wild-type littermates . BAC-Q72 mice showed normal performance at 5 weeks ( Fig 3 ) and at 12 weeks ( S4A Fig ) . Of note , testing at 12 weeks was performed on mice housed under slightly different conditions , which may explain the relatively poor performance of wild-type mice . At 16 weeks of age , performance of BAC-Q72 mice became significantly worse than wild-type mice ( Fig 3; p<0 . 05 ) and mice continued to perform poorly as they aged ( 24 and 36 weeks old , S4A Fig and Fig 3 ) . Taken together , these results indicate that BAC-Q72 transgenic mice develop a progressive age-dependent motor impairment . To investigate morphological changes associated with the expression of mutant ATXN2 protein , we compared cerebellar sections from BAC transgenic lines with wild-type mice . Immunostaining with calbindin-28k antibody revealed PC morphological changes in BAC-Q72 mice at 24 weeks of age , but not in BAC-Q22 or wild-type mice ( Fig 4A ) . To more quantitatively assess this change , we performed Western blotting and verified reduction of Calb1 and Pcp2 proteins in BAC-Q72 mouse cerebella ( Fig 4B ) . As observed in the Pcp2-ATXN2[Q127] model , cerebellar morphology was still normal at a time when key mRNA transcripts had already declined . Thus , calbindin-stained cerebellar sections and PC counts of BAC-Q72 mice at 12 weeks showed normal cerebellar morphology and unaltered PC counts [18 . 8 ±1 . 2 in WT , n = 3 animals , and 19 . 4 ±1 . 1 in BAC-Q72 mice , n = 3 animals , p = 0 . 51] ( S4B , S4C Fig ) . We previously reported that steady-state mRNA levels of specific PC transcripts preceded behavioral onset in an SCA2 model targeting transgene expression to PCs [20] . Expression changes in these genes ( Calb1 , Pcp2 , Grid2 and Grm1 ) also preceded the onset of a decrease in PC firing . Expression changes were progressive over time and paralleled deterioration of motor behavior . To investigate whether similar changes occurred in BAC transgenic mice as we previously observed in Pcp2-ATXN2[Q127] , we performed qRT-PCR to measure transcript levels of PC-specific genes at different ages . At 16 and 45 weeks , BAC-Q22 mice were indistinguishable from wild-type mice including expression of endogenous mouse Atxn2 ( Fig 5A ) . In BAC-Q72 mice , however , expression of Pcp2 showed significant reductions ( p<0 . 01 ) as early as 5 weeks . All other genes tested remained unchanged compared to wild-type ( Fig 5B ) . At 9 and 16 weeks of age , significant reductions in Calb1 ( p<0 . 05 ) and Grid2 ( p<0 . 01 ) were seen and were progressive ( Fig 5B ) . Steady-state levels of Grm1 decreased only at 24 weeks ( p<0 . 05 ) . Endogenous mouse Atxn2 expression levels did not change in BAC-Q72 mice at any time point when compared with wild-type . Taken together , these data demonstrated that a subset of PC-enriched genes showed a progressive reduction in steady-state mRNA levels in BAC-Q72 mice , whereas they remained unchanged in BAC-Q22 animals . To further characterize the BAC-Q72 line and compare it with the well-characterized Pcp2-ATXN2[Q127] line , we performed transcriptome analysis by deep RNA-sequencing of cerebellar RNA . We chose time points for both lines just prior to behavioral and morphological changes , i . e . 8 weeks for the BAC-Q72 line and 6 weeks for the Pcp2-ATXN2[Q127] line . For both sets of RNAs , quality of reads and alignments were high ( see methods ) . We observed significant changes of 1417 transcripts in Pcp2-ATXN2[Q127] and 491 transcripts in BAC-Q72 mice with a false discovery rate ( FDR ) of ≥15 and a log2 ratio of change ≥|0 . 30| ( Fig 6A ) . With these filtering parameters , 255 transcripts were only seen in the BAC-Q72 line ( class I ) , 236 transcripts were shared between the two lines ( class II ) and 1181 transcripts were changed only in the Pcp2-ATXN2[Q127] line ( Class III ) . We validated changes in several of the class II transcripts by qRT-PCR using cerebellar RNA samples from BAC-Q72 mice ( 8 weeks old ) and Pcp2-ATXN2[Q127] ( 6 weeks old ) , and compared with their respective WT littermates ( Fig 6B ) . The concordance between RNA-seq and qRT-PCR was high ( Fig 6C ) . The top 50 transcripts changed in the BAC-Q72 line are shown in S1 Table and the top 50 transcripts changed in the Pcp2-ATXN2[Q127] line are presented in S2 Table . This table also shows that most of these transcripts are changed in the BAC-Q72 line as well , although with a smaller degree of change or a lower FDR . S3 Table lists the top class II genes sorted by FDR in the BAC-Q72 line . This represents a subset of the 236 overlapping genes shown in Fig 6A . In order to gain insight into the molecular function of altered transcripts in BAC-Q72 and Pcp2-ATXN2[Q127] mice , we performed Gene Ontology ( GO ) analysis . This is shown in S4 Table and indicates that many of the significant GO terms are shared by the two models . Of note , GO terms relate to known functions of PC such as calcium homeostasis , glutamate-mediated signaling and voltage-gated ion channels . In summary , these data indicate a significant overlap of altered transcripts and shared functions in both SCA2 models at comparable stages just prior to onset of morphological and behavioral changes . We were also interested in the nature and expression pattern of transcripts in class I and class III ( Fig 6 ) . We confirmed changes in several of the class I transcripts by qRT-PCR ( S5 Fig ) . These transcripts showed a progressive reduction in BAC-Q72 mice , but remained unchanged in the Pcp2-ATXN2[Q127] line even at late time points . Of these 50 , 16 genes ( Grm4 , Igfbp5 , Fstl5 , Snrk , D8Ertd82e , Dusp5 , Nab2 , Btg1 , Adrbk2 , Slc25a29 , Sty12 , Crhr1 , Synpr , Lrrtm2 , Rit2 and Cabp2 ) were previously identified as GC-specific using translational profiling [29] . Class III transcripts were those that showed changes only in Pcp2-ATXN2[Q127] mice , but not in BAC-Q72 at an FDR>15 and a log2 ratio of change ≥|0 . 3| . We verified expression changes of six class III transcripts longitudinally in Pcp2-ATXN2[Q127] mice at 4 , 8 , and 24 weeks of age , and BAC-Q72 mice at 5 , 9 , 16 and 24 weeks of age , and their respective WT littermates by qRT-PCR . Five of the six transcripts showed significant and progressive reduction with age not only in Pcp2-ATXN2[Q127] mice but also in BAC-Q72 mice ( S6 Fig ) . This is consistent with the milder behavioral phenotype seen in BAC-Q72 mice and suggests that the overlap of the transcriptomes in the two models may potentially be even greater . Changes in steady-state expression of a subset of genes preceded onset of physiological and behavioral changes in Pcp2-ATXN2[Q127] and BAC-Q72 mice . One of the most significantly down-regulated genes in both models prior to behavioral onset was Rgs8 ( regulator of G-protein signaling 8 ) ( S1 , S2 , S3 Tables ) . RGS proteins are regulatory and structural components of G protein-coupled receptor complexes . RGS proteins ( RGS7 , RGS8 , RGS11 , RGS17 and RGSz1 ) are widely expressed in cerebellum and RGS8 is specifically distributed in dendrites and cell bodies of PCs [30 , 31] . Several reports suggest that the RGS family proteins are also associated with motor neuron functions [32 , 33] . The decreased steady-state level of Rgs8 mRNA was confirmed by qRT-PCR in Pcp2-ATXN2[Q127] mice at 4 , 8 and 24 weeks of age , indicating that these RNAs progressively declined with time ( S7A Fig ) . In parallel , we also measured Rgs8 protein steady state levels in Pcp2-ATXN2[Q127] mouse cerebella at 24 weeks of age . As expected , Rgs8 protein levels were significantly reduced in Pcp2-ATXN2[Q127] mice when compared with wild-type mice ( S7B Fig ) . Next , we investigated the fate of Rgs8 mRNA steady-state levels in our BAC mouse models by qRT-PCR . When tested in BAC-Q72 mouse cerebella , levels of Rgs8 mRNA progressively decreased with time but remained unchanged in BAC-Q22 mice compared with wild-type mice across all ages of mice tested ( Fig 7A ) . To examine whether changes in steady-state mRNA levels led to decreased protein abundance , we performed Western blot analysis to measure Rgs8 protein in wild-type and BAC transgenic mouse cerebella . Western blot analyses indicated reduced steady-state levels of Rgs8 protein in BAC-Q72 mice but not in BAC-Q22 mice when compared with wild-type mice at 24 weeks of age ( Fig 7B ) . To assess whether these findings replicated in human cells we analyzed EBV-transformed lymphoblastoid ( LB ) cells derived from a control individual and two SCA2 patients with expansions of Q46 and Q52 ( Fig 7C ) . We could not use skin fibroblasts as this cell type does not express RGS8 . Two SCA2-LB cells expressing Q46 or Q52 demonstrated decreased expression of RGS8 transcript compared with control cells expressing wild-type ATXN2 with 22 repeats . Unfortunately , LB cells do not efficiently translate RGS8 message , so that Western blots did not allow detection of RGS8 protein in LB cells . To test whether reduction of Rgs8 levels induced by mutant ATXN2 could be recapitulated in vitro , we measured steady-state levels of RGS8 mRNA and protein in hygromycin selected enriched SH-SY5Y cells expressing Flag-tagged ATXN2-Q22 , -Q58 or -Q108 . Western blot analyses of whole cell extracts indicated that expression of ATXN2-Q58 or Q108 resulted in decreased RGS8 levels compared to control or ATXN2-Q22 ( Fig 8A ) . To exclude that decreased RGS8 levels were a consequence of selective cellular toxicity of ATXN2-Q58 or -Q108 expression , we measured expression of endogenous DDX6 and PABPC1 , which have been shown to interact with ATXN2 [6 , 8] and CUG-BP1 , a nuclear protein by Western blot analysis . The levels of DDX6 , PABPC1 and CUG-BP1 were not altered ( Fig 8A ) strongly supporting that the effect of mutant ATXN2 was specific to RGS8 . In parallel , qRT-PCR analyses of SH-SY5Y cell lines expressing Flag-tagged wild-type and mutant ATXN2 demonstrated a moderate reduction of RGS8 mRNA in cell expressing Flag-ATXN2-Q108 ( Fig 8B ) . Decrease of RGS8 levels in mutant BAC mice could be the result of transcriptional control , mRNA stability and processing or translational control . In contrast to other polyQ proteins , ATXN2 does not enter the nucleus [19] and protein interaction studies have not yielded proteins thought to be involved in transcriptional control . To examine translation of RGS8 , we expressed exogenous RGS8 in hygromycin selected SH-SY5Y cells expressing Flag-tagged ATXN2-Q22 , -Q58 or -Q108 . MYC-tagged RGS8 cDNA including 5’ and 3’ UTRs was cloned under the transcriptional control of the CMV promoter . Forty-eight hrs post-transfection , Western blot analyses revealed that the levels of exogenous RGS8 were significantly decreased in cells expressing ATXN2-Q58 or -Q108 compared with cells expressing wild-type ATXN2-Q22 ( Fig 8C ) . To control for equal transfection , we monitored levels of GFP , which was expressed as an independent cassette in the plasmid . Thus , presence of mutant ATXN2 reduced RGS8 protein levels in vivo and in vitro . Reduced protein levels potentially out of proportion to reduced mRNA levels in vivo and in vitro suggested to us that ATXN2 might be directly involved in the translation or stability of specific mRNAs . In addition , ATXN2 is known to interact with RNAs through a “Like Sm ( LSm ) domain” [34–36] . It also interacts with cytoplasmic poly ( A ) -binding protein 1 ( PABPC1 ) and assembles with polysomes [6 , 7] . Therefore , we first tested interaction of ATXN2 with RGS8 mRNA and then performed in vitro translation assays in the presence of wild-type and mutant ATXN2 . We performed Protein-RNA immunoprecipitation ( IP ) experiments in cultured SH-SY5Y cells overexpressing Flag-tagged ATXN2 containing Q22 or Q108 . Whole cell extracts were incubated with Flag-mAb-beads and immunoprecipitates were washed with a buffer containing 200 mM NaCl . Bound protein-RNA complexes were eluted from the beads by Flag peptide competition . The IP products were divided equally into two aliquots and one aliquot was analyzed by Western blot . As shown in Fig 9A , the eluted proteins showed co-IP of DDX6 and PABPC1 , which are known to interact with ATXN2 [6 , 8] . To identify RNAs that immunoprecipitated with ATXN2 , the extracted RNAs from the second aliquot were subjected to RT-PCR and qPCR analyses . Our results showed that RGS8 mRNA precipitated with ATXN2-Q22 and ATXN2-Q108 ( Fig 9A and 9B ) . Binding of RGS8 mRNA with ATXN2-Q108 , however , was significantly reduced compared with ATXN2-Q22 in three independent experiments . We next proceeded to examine in vitro RGS8 translation . For that purpose , we performed assays using Flag-tagged ATXN2 with Q22 or Q108 , respectively , and determined RGS8 protein abundance by Western blot analysis . In three independent experiments , one of which is shown in Fig 9C , levels of RGS8 decreased significantly in the presence of ATXN2-Q108 when compared with the levels in the presence of ATXN2-Q22 . No significant alteration in the levels of RGS8 synthesis was detected between ATXN2-Q22 and control extracts ( Fig 9C and 9D ) . These results suggest a role for ATXN2 in translational regulation and a dysregulation of this process in the presence of mutant ATXN2 .
Mouse models generated with tissue specific or strong promoters facilitate the evaluation of functional and anatomical consequences in many neurological disorders . The Purkinje cell protein 2 ( Pcp2 ) and the Prion protein ( PrP ) promoters have been used to generate mouse models for polyQ ataxias such as SCA1 , SCA2 and SCA3 [19 , 20 , 37–41] . For instance , the use of the Pcp2 promoter for expressing mutant ATXN1 or ATXN2 has been shown to recapitulate the progressive cellular and functional phenotype of human SCA1 or SCA2 [19 , 20 , 37] . Use of a BAC-transgenic approach resulted in a more widespread expression of the transgene mirroring prior observations of endogenous ATXN2 expression in mouse and human [1] . The control regions included in our BAC transgene specified expression in CNS and non-CNS tissues ( Fig 1B ) . In the CNS , expression was seen in the cerebral hemispheres , cerebellum and spinal cord . This is consistent with expression of endogenous mouse Atxn2 [1] and in situ hybridization data as shown in the Allen Brain Atlas [28] . In the cerebellum , expression of the BAC-transgene was seen in PCs , but also in granule cells , and neurons of the dentate nucleus ( Fig 2 ) . As the transgenes were not tagged , we used LCM to establish transgene expression in these sub-regions of the cerebellum . Future physiological experiments using the cerebellar slice preparation will need to examine what role mutant ATXN2 plays in granule cells and dentate nucleus and in overall cerebellar dysfunction in comparison with the PC-targeted expression of mutant ATXN2 [20] . Motor function deficits are common to all SCA2 mouse models , although their ages of onset differ . The accelerating rotarod is used to measure motor coordination and motor learning over a number of days . Our BAC-Q72 mice developed progressive motor deficits beginning at 16 weeks of age ( Fig 3B ) . The motor phenotype of our BAC-Q72 mice was intermediate to that of our Pcp2-ATXN2[Q58] and Pcp2-ATXN2[Q127] mice , although transgene copy numbers and precise developmental expression patterns are difficult to compare . As with our Pcp2-ATXN2[Q22] line [19] , the BAC-Q22 line did not show a motor or cellular phenotype . This study now extends these observations to mRNA measurements of key PC genes out to 45 weeks of age ( Fig 5A ) . Lack of mRNA changes in BAC-Q22 are likely not due to differences in expression levels between lines , as transgenic ATXN2 had higher expression in the BAC-Q22 than in the BAC-Q72 line , both at the level of mRNA and protein ( Fig 1C and 1D ) . Lack of any changes in genes that are typically altered early in Pcp2-ATXN2[Q127] and BAC-Q72 supports the notion that simple overexpression of human wild-type ATXN2 does not cause significant PC pathology . In contrast , motor function deficits in Atxn2-CAG42 knock-in mice were not evident until the age of 18 months [21] . By comparing the motor functions in these four SCA2 transgenic mouse models , it is apparent that motor function deficits are dependent on CAG repeat length . Consistent with this interpretation , knock-in Atxn1-CAG78 SCA1 mice developed neither ataxic behavior nor a neuropathological phenotype [42] , while knock-in Atxn1-CAG154 SCA1 mice did [43] . Our BAC-Q72 transgenic mouse model , although generating lower levels of mutant ATXN2 expression in the cerebellum , develop motor deficits that resemble findings in human SCA2 patients . These observations validate the notion that SCAs can be accurately modeled in mice . Animal models for several polyQ diseases have shown alteration of body weight [21 , 43–45] . In this study , BAC transgenic mice demonstrated reduced body weights . The magnitude was similar to knock-in Atxn2-CAG42 mice and Atxn1-Q154/2Q mouse models [21 , 43] . On the other hand , mice lacking Atxn2 exhibit obesity as a consequence of insulin resistance and altered lipid metabolism pathways [16 , 17 , 46] . Increased weight loss due to reduced body fat has also been reported in other polyglutamine diseases , including Huntington disease [47 , 48] . Of note , reductions in body weight were similar for BAC-Q22 and BAC-Q72 mice suggesting that with regard to the body weight phenotype a simple gain of function may be operative that is mirrored by obesity in loss of function models . RGS proteins comprise a large family of more than 20 members that negatively modulate heterotrimeric G protein signaling . They share a homologous RGS domain that functions to activate the GTPase of Gα proteins . RGS8 is widely expressed in testis , brain , and cerebellar Purkinje cells [56 , 57] . Mice lacking Rgs6 or Rgs9 exhibit motor function deficits and ataxia [32 , 33] . Rgs8 knock-out mice were viable , fertile , and showed normal development , but have not been tested in detail for motor behaviors or PC morphology [57] . Given the importance of a dysregulated mGluR1-ITPR1 axis in SCA2 pathology [13 , 58] , reduction in RGS proteins could further increase abnormally enhanced mGluR1 signaling . We therefore examined RGS8 abundance in BAC-Q72 mice and Epstein-Barr virus immortalized human lymphoblastoid B ( LB ) -cells from SCA2 patients ( Fig 7 ) . The results demonstrated that Rgs8 transcripts and protein abundance were significantly decreased in BAC-Q72 mice ( Fig 7A and 7B ) . Consistent with this , SCA2-LB cells also demonstrated decreased RGS8 transcripts ( Fig 7C ) . Next , we developed an in vitro model using SH-SY5Y cells . Overexpression of mutant ATXN2 resulted in downregulation of RGS8 and this phenomenon was not seen for other known ATXN2 interactors ( Fig 8 ) . As protein levels appeared somewhat depressed out of proportion to the observed reduction in steady-state mRNA levels , we hypothesized that ATXN2 might regulate translation of mRNAs directly . Consistent with this hypothesis , we showed that both wild-type and mutant ATXN2 immunoprecipitated RGS8 mRNA in human cell culture and that this interaction was weaker for mutant ATXN2 ( Fig 9A and 9B ) . This was also reflected in in vitro translation assays as presence of an expanded polyQ tract in ATXN2 reduced translation ( Fig 9C and 9D ) . Our observations are consistent with studies of the Drosophila homolog of ATXN2 ( Atx2 ) . Atx2 regulates PERIOD ( PER ) translation by interacting with TWENTY-FOUR ( TYF ) that is required for circadian locomotor behavior . Depletion of Atx2 or expression of mutant Atx2 protein blocked the recruitment of PABP to the TYF-containing protein complex and decreased abundance of PER , thereby altering behavioral rhythms [59 , 60] . ATXN2 interactions with polyA-binding protein 1 ( PABPC1 ) , the splicing factor A2BP1/FOX1 and poly-ribosomes further support roles for ATXN2 in RNA metabolism [5–7] . Depletion of PABP from a cell free extract prevented initiation of mRNA translation [61] . Our studies now extend these observations to mammalian systems and to a gene abundantly expressed in PCs . It is quite likely that Rgs8 will be just one member of a larger set of mRNAs whose expression is regulated by ATXN2 . Aberrant RNA metabolism including processing , degradation , and translation is now recognized to play an important role in neurodegenerative diseases . Among these diseases are amyotrophic lateral sclerosis ( ALS ) , Spinal Muscular Atrophy ( SMA ) and Fragile X syndrome ( FXS ) [62–70] . Although ATXN2 had been implicated in steps regulating mRNA translation and formation of stress granules [8 , 71 , 72] , to our knowledge we describe for the first time a significant difference in these functions between wild-type and mutant ATXN2 . Our observations may also have implications for ALS as long normal ATXN2 alleles are a risk factor for ALS [18 , 73] and some individuals with full mutant ATXN2 alleles may present as ALS [74] . In summary , BAC-SCA2 transgenic mice represent the first animal model with expression of mutant full-length human ATXN2 under the control of its endogenous human promoter including intronic regulatory sequences . These sequences resulted in widespread expression of ATXN2 mirroring expression of endogenous Atxn2 . Expression of mutant ATXN2-Q72 , but not wild-type ATXN2-Q22 , led to a progressive motor deficit , accompanied by morphological and transcriptome changes . As previously demonstrated in C . elegans and the fly [6 , 59 , 60 , 75] , ATXN2 may exert translational control upon a subset of mRNAs . We showed in two independently generated models that presence of mutant ATXN2 in vivo resulted in reduced steady-state levels of RGS8 mRNA and even further reduction in RGS8 protein . ATXN2 coprecipitated with RGS8 mRNA and mutant ATXN2 reduced translation of RGS8 mRNA . RGS proteins can act via Gαq on G-protein coupled receptors . As mutant ATXN2 enhances Ca2+ release from the endoplasmic reticulum ( ER ) via its abnormal interaction with ITPR1 , reduction of RGS8 might be predicted to further increase intracellular Ca2+ by prolonging mGluR1 stimulated Ca2+ release . Our studies now provide a framework to further examine the aberrant mGluR1-ITPR1 axis in SCA2 pathogenesis .
Human lymphoblastoid B ( LB ) -cells from SCA2 patients and unaffected normal controls were used . All subjects gave written consent and all work was approved by the Institutional Review Board at the University of Utah under IRB# 00035351 and IACUC- University of Utah IACUC committee , protocol number 13–0004 . BAC-SCA2 mice were maintained in FVB background and bred and maintained under standard conditions consistent with National Institutes of Health guidelines and approved by the University of Utah , IACUC protocol . A 169 kb of RP11-798L5 BAC clone ( Empire Genomics . , USA ) containing the 150 kb human ATXN2 locus was engineered to replace the endogenous ATXN2 exon-1 CAG22 with CAG72 repeats . The BAC DNA was prepared according to published protocols [76 , 77] and microinjected into FVB fertilized eggs to produce transgenic mice at the University of California Irvine ( UCI ) Mouse Core Facility . BAC-SCA2 mice were maintained in the FVB background and bred and maintained under standard conditions consistent with National Institutes of Health guidelines and approved by the University of Utah , IACUC protocol . For genotyping of BAC-SCA2 transgenic mice , DNA was isolated from mice tails using Qiagen genomic DNA extraction kit ( Qiagen Inc . , USA ) and genotyping PCR was performed . Three primer sets were used to identify the transgene and the primer sequences are follows: P3 forward: 5’-AATTTATGTGATGTT CACTGTTTCTTCC-3’ , P3 reverse: 5’-TACGGTCCCTCCAAATAGTGTTAC-3’ , P7 forward: 5’-TCTTTTTACAGTACAAGCCCACCACC-3’ , P7 reverse: 5’-TTCAAAATG CACCCTTAGCACACCTG-3’ , SCA2-A forward: 5’-GGGCCCCTCACCATGTCG-3’ , SCA2-B reverse: 5’-CGGGCTTGCGGACATTGG-3’ . For all experiments wild-type and transgenic animals were kept as littermates . From 3 to 5 litters were used per experiment dependent on actual size of litters . Mice were deeply anesthetized with isoflurane . Mouse cerebella were removed and immediately submerged in liquid nitrogen . Tissues were kept at −80°C until the time of processing . Total RNA was extracted from mouse cerebella using the RNeasy Mini Kit ( Qiagen Inc . , USA ) according to the manufacturer’s protocol . DNAse I treated RNAs were used to synthesize cDNAs using the ProtoScript cDNA First Strand cDNA Synthesis Kit ( New England Biolabs Inc . , USA ) . Primers for RT-PCR were designed to prevent amplification from genomic DNA ( annealing sites in different exons or across intron-exon boundaries ) . Human ATXN2 primer sites were in exon 1 and exon 5 , including Exon 1-F ( 5’-CTCCTCGGTGGTCGCGGCGACCTC-3’ ) and Exon 5-R ( 5’-CTCTTTTTGCATAACT GGAGTCC-3’ ) . ATXN2 primers for amplifying CAG repeats were SCA2-A ( 5’-GGGCCCCTCACCATGTCG-3’ ) and SCA2-B ( 5’-CGGGCTTGCGGACATTGG-3’ ) . Gapdh primers were GAPDH-F ( 5’-TGAAGGTCGGA GTCAACGGATTTGG-3’ and GAPDH-R ( 5’-GGAGGCCATGTGGGCCATGAG-3’ ) . Gapdh amplification was conducted in parallel as an internal control for RNA quality and was also employed to evaluate quality the reverse transcriptase reactions . Quantitative RT-PCR was performed in Bio-Rad CFX96 ( Bio-Rad Inc . , USA ) with the Power SYBR Green PCR Master Mix ( Applied Biosystems Inc , USA ) . PCR reaction mixtures contained SYBR Green PCR Master Mix and 0 . 5 pmol primers and PCR amplification was carried out for 45 cycles: denaturation at 95°C for 10 sec , annealing at 60°C for 10 sec and extension at 72°C for 40 sec . The threshold cycle for each sample was chosen from the linear range and converted to a starting quantity by interpolation from a standard curve run on the same plate for each set of primers . All gene expression levels were normalized to the Actin or Gapdh mRNA levels . Primer pairs designed for qRT-PCR are given as forward and reverse , respectively , and listed in supplementary table ( S5 Table ) . Cerebella from 8 weeks old wild-type and BAC-Q72 mice ( 4 animals in each group ) , and 6 weeks old Pcp2-ATXN2[Q127] and wild-type littermates ( 16 animals in each group ) were used for RNA sequence analyses . Total RNA was isolated using miRNeasy Mini Kit ( Qiagen Inc . , USA ) according to the manufacturer’s protocol . RNA quality was determined using the Bioanalyzer 2100 Pico Chip ( Agilent ) . Samples with an RNA integrity number ( RIN ) >8 were used for library preparation using Illumina TrueSeq Stranded Total RNA Sample Prep with Ribo-Zero rRNA Removal Kit for mouse . Single-end 50-bp reads were generated on a Hiseq 2000 sequencing machine at the University of Utah Microarray and Genomic Analysis Shared Resource using Illumina Version 4 flow cells . Reads were then aligned to the mouse reference genome ( mm10 ) by Novoalign ( http://www . novocraft . com ) . Quality of RNA sequencing was extremely high with an average of twenty eight million reads for BAC-Q72 and twenty two million reads for Pcp2-ATXN2[Q127] . Ninety eight percent of the reads for both sets of RNAs were aligned to the reference mouse genome . After read alignment , differentially expressed genes were identified using the DRDS application ( version 1 . 3 . 0 ) in the USeq software package ( http://useq . sourceforge . net/ ) . Gene Ontology ( GO ) annotations were obtained for all differentially expressed genes ( p<0 . 05 ) . GO enrichment results were obtained using the software DAVID [78 , 79] . Overlap of BAC-Q72 and Pcp2-ATXN2[Q127] molecular function GO annotations was accomplished using only level 5 categories ( p<0 . 05 ) . SH-SY5Y cells were cultured and maintained in DMEM media containing 10% fetal bovine serum . Epstein-Barr virus immortalized human lymphoblastoid B ( LB ) -cells from SCA2 patients and unaffected normal controls were cultured in RPMI 1640 medium supplemented with 15% fetal bovine serum , penicillin and streptomycin . All subjects gave written consent and all work was approved by the Institutional Review Board at the University of Utah . Protein extracts were prepared by homogenization of mouse cerebella in extraction buffer ( 25 mM Tris-HCl pH 7 . 6 , 300 mM NaCl , 0 . 5% Nonidet P-40 , 2 mM EDTA , 2 mM MgCl2 , 0 . 5 M urea and protease inhibitors; Sigma; cat# P-8340 ) followed by centrifugation at 4°C for 20 min at 16 , 100 × g . Only supernatants were used for Western blotting . Cellular extracts were prepared by the single-step lyses method [80] . The cells were harvested and suspended in SDS-PAGE sample buffer ( 2x Laemmli Sample Buffer; Bio-Rad; cat# 161–0737 ) and then boiled for 5 min . Equal amounts of the extracts were subjected to Western blot analysis to determine the steady-state levels of proteins using the antibodies listed below . Protein extracts were resolved by SDS-PAGE and transferred to Hybond P membranes ( Amersham Bioscience Inc . , USA ) . After blocking with 5% skim milk in 0 . 1% Tween 20/PBS , the membranes were incubated with primary antibodies in 5% skim milk in 0 . 1% Tween 20/PBS for 2 hrs at room temperature or overnight at 4°C . After several washes with 0 . 1% Tween 20/PBS , the membranes were incubated with the corresponding secondary antibodies conjugated with HRP in 5% skim milk in 0 . 1% Tween 20/PBS for 2 hrs at room temperature . Following three additional washes with 0 . 1% Tween 20/PBS , signals were detected by using the Immobilon Western Chemiluminescent HRP Substrate ( Millipore Inc . , USA; cat# WBKLSO100 ) according to the manufacturer’s protocol . The following antibodies were used throughout the study . ATXN2 mAb [ ( 1:3000 ) , BD Biosciences Inc . ; cat# 611378] , 5TF1-1C2 mAb [ ( 1:3000 ) , Millipore Inc . ; #MAB1574] , RGS8 rabbit polyclonal Ab [ ( 1:3000 ) , Novus Biologicals; #NBP2-20153] , Calbindin-D-28K mAb [ ( 1: 5000 ) , Sigma Inc . ; cat# C9848] , PCP2 mAb [ ( 1: 5000 ) , Santa Cruz Inc . ; cat# sc-137064] , DDX6 rabbit polyclonal Abs [ ( 1:4000 ) , Santa Cruz Inc . ; cat# sc-27127-R] , PABPC1 mAb [ ( 1:4000 ) , Santa Cruz Inc . ; cat# sc-27127-R] , CUG-BP1 mAb [ ( 1:4000 ) , Santa Cruz Inc . ; cat# sc-20003] , Flag M2 mAb [ ( 1:10 , 000 ) , Sigma Inc . ; cat# F3165] , GFP mAb [ ( 1:3000 ) , Santa Cruz Inc . ; cat# sc-9996] and MYC mAb conjugated with HRP [ ( 1:5000 ) , Invitrogen Inc . ; cat# A3858] . To control for protein quality and loading , the membranes were re-probed with β-Actin mAb conjugated with HRP [ ( 1:10 , 000 ) , Sigma Inc . ; cat# A3858] . The secondary antibodies were goat anti-mouse IgG-HRP [ ( 1:5000 ) , Sigma Inc . ; cat# A2304] , and donkey anti-rabbit IgG-HRP [ ( 1:5000 ) , Santa Cruz Inc . ; cat# sc-2057] . Motor behavior of SCA2 mice was determined using the accelerating rotarod . Cohorts were age matched prior to all behavioral experiments . Male and female mice performed equally well; therefore , data were pooled and gender differences were not evaluated further . The motor performance of BAC-Q22 and BAC-Q72 mice and wild-type littermates were evaluated using the accelerating rotarod ( Ugo Basile ) according to our published protocol [20] . For mice clinging to the rod , the time at which a mouse had completed 5 rotations was taken as the final latency . Mice were deeply anesthetized with isoflurane , then transcardially perfused with ice-cold phosphate buffered saline ( PBS ) . Tissue was quickly removed and submerged into cold 4% paraformaldehyde ( Electron Microscopy Sciences ) and kept at 4°C overnight . The following day , PFA was replaced with 10 mM sodium citrate pH 6 . 0 , and then incubated at 4°C overnight , after which the tissue was exposed to microwave radiation 3 times in 10 sec bursts . Following microwave radiation , tissues were cryoprotected by incubating in 20% sucrose in PBS overnight followed by 30% sucrose overnight both at 4°C . Then the samples were mounted in Tissue-Tek O . C . T . Compound ( Sakura Finetek ) and stored at -80°C until the time of sectioning . Tissue sections were cut into 20 μM thick slices and floated into cold PBS . Tissues were washed 3 times with PBS at RT for 15 min each time . Free-floating sections were incubated with blocking/permeabilization solution consisting of 5% skim milk , 0 . 3% Triton X-100 in PBS for 4 hr at RT . Sections were then incubated overnight at 4°C with primary antibodies , calbindin-28kDa mAb at 1:200 dilution . After 3 washes in PBS at 15 min each , sections were incubated with DyLight-550 ( Red ) ( Thermo Fischer Scientific ) fluorescent secondary antibodies at 1:500 dilution for 2 hr at RT . Following incubation , the sections were washed 3 times with PBS , and the sections were transferred to Superfrost Plus microscope slides ( Fischer Scientific ) and mounted with Prolong Gold ( Invitrogen ) . Sections were imaged using confocal microscope ( Nikon Eclipse Ti microscopy ) and analyzed by Nikon EZ-C1 software . PCs were counted in parasagittal slices from 3 mice in each group . Fresh whole cerebella from wild type or BAC-Q22 or BAC-Q72 mice was freeze-mounted in O . C . T . and sectioned onto Arcturus PEN Membrane glass slides . Sections were fixed and H&E stained using the Fast Frozen Stain Kit ( EMS ) . Sections on slides were then dehydrated by passage through a solution series of 95% ethanol , 100% ethanol , and then xylene . Prepared slides were stored in a desiccated chamber until needed . LCM was performed using the Arcturus Veritas LCM system . RNAs were prepared from tissue on LCM caps ( CapSure , Applied Biosystems ) using the Arcturus PicoPure RNA Kit ( Applied Biosystems Inc . , USA ) . RNA yield was typically 5 μg/cap . cDNA was then prepared by using the ProtoScript M-MuLV First Strand cDNA Synthesis Kit ( NEB Inc . , USA ) and used for qRT-PCR as described in Methods above . To identify proteins and RNAs that bind to ATXN2 , we carried out protein-RNA immunoprecipitation ( IP ) experiments from lysates of SH-SY5Y cells expressing Flag-ATXN2-Q22 and Flag-ATXN2-Q108 . Whole cell extracts were prepared by the two-step lyses method [80] . First , cells were lysed with a cytoplasmic extraction buffer ( 25 mM Tris-HCl pH 7 . 6 , 10 mM NaCl , 0 . 5% NP40 , 2 mM EDTA , 2 mM MgCl2 , protease and RNAse inhibitors ) and cytoplasmic extracts were separated by centrifugation at 14 , 000 RPM for 20 min . Second , the resultant pellets were suspended in nuclear lysis buffer or high salt lyses buffer ( 25 mM Tris-HCl , pH 7 . 6 , 500 mM NaCl , 0 . 5% Nonidet P-40 , 2 mM EDTA , 2 mM MgCl2 , protease and RNAse inhibitors ) , and the nuclear extracts were separated by centrifugation at 14 , 000 RPM for 20 min . The nuclear extracts were then combined with the cytoplasmic extracts and denoted as whole cell extracts . Specifically , while combining cytoplasmic and nuclear extracts , the NaCl concentration was adjusted to physiologic buffer conditions ( ~150 mM ) to preserve in vivo interactions . Ninety percent of cell extracts were subjected to Flag monoclonal antibody ( mAb ) IP ( Anti-Flag M2 Affinity Gel , Sigma Inc . ; cat# A2220-1ML ) to immunoprecipitate ATXN2 interacting protein-RNA complexes . The remaining 10% of whole cell extracts were saved as the input control for Western blotting and RT-PCR analyses . The IPs were washed with a buffer containing 200 mM NaCl and the bound protein-RNA complexes were eluted from the beads with Flag peptide competition ( 100 μg/ml ) . Eluted fractions were divided into two equal parts . One part was analyzed by SDS-PAGE followed by Western blotting to determine the efficiency and quality of immunoprecipitation . RNA was isolated from the other fraction and subjected to RT-PCR and qRT-PCR analyses to identify RNAs that bound to wild type or mutant ATXN2 . To determine the role of ATXN2 on RGS8 mRNA translation , in vitro translation assays were performed using the rabbit reticulocyte lysate-based cell free TNT Quick Coupled Transcription/Translation Kit ( Promega Inc . , USA ) according to the manufacturer’s instructions , with minor modifications . Briefly , 1 μg of cDNA plasmids of LacZ ( control ) and Flag-tagged ATXN2 expressing Q22 or Q108 were added to 20 μl of the rabbit reticulocyte lysate kit component , including 20 μM amino acids in a total volume of 25 μl . The translation reaction was carried out for 2 hr at 30°C . Next 1 μg of RGS8 cDNA plasmid was added to each translation reaction with fresh rabbit reticulocyte lysate containing 20 μM amino acids in a total volume of 50 μl , and incubated further at 30°C for 4 hr . Translation assays was analyzed by SDS-PAGE followed by Western blot analyses . For Western blot analyses , the experiments were performed at least three times , and wherever appropriate gel films were scanned and band intensities were quantified by ImageJ analyses . The p values were calculated by pairwise Student’s t-tests . Student’s t-tests were also used to compare mRNA steady state levels between BAC and wild-type mice determined by qRT-PCR . The level of significance was set at p<0 . 05 . In the figures , a single asterisk indicates p<0 . 05 , a double asterisk p<0 . 01 , a triple asterisk p<0 . 001 , and ns represents p≥0 . 05 . For accelerating rotarod analyses , repeated measures ANOVA was used with post-hoc t-tests to compare means .
|
Spinocerebellar ataxia type 2 ( SCA2 ) is an inherited neurodegenerative disorder leading to predominant loss of Purkinje cells in the cerebellum and impairment of motor coordination . The mutation is expansion of a protein domain consisting of a stretch of glutamine amino acids . We generated a mouse model of SCA2 containing the entire human normal or mutant ATXN2 gene using bacterial artificial chromosome ( BAC ) technology . Mice expressing a BAC with 72 glutamines ( BAC-Q72 ) developed a progressive cerebellar degeneration and motor impairment in contrast to mice carrying the normal human gene ( BAC-Q22 ) . We found that even prior to behavioral onset of disease , the abundance of specific messenger RNAs changed using deep RNA-sequencing . One of the mRNAs with early and significant changes was Rgs8 . Levels of Rgs8 protein were even further reduced than mRNA levels in BAC-Q72 cerebella suggesting to us that mutant ATXN2 might have a role in mRNA stability and translation . Using a cellular model , we showed that the ATXN2 protein interacted with RGS8 mRNA and that this interaction differed between normal and mutant ATXN2 . Presence of mutant ATXN2 resulted in reduced RGS8 protein translation in a cellular model . Our studies describe a mouse model of SCA2 expressing the entire human ATXN2 gene and emphasize the role of ATXN2 in mRNA metabolism .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Ataxin-2 Regulates RGS8 Translation in a New BAC-SCA2 Transgenic Mouse Model
|
Extracellular bacteria , such as Pseudomonas aeruginosa and Klebsiella pneumoniae , have been reported to induce autophagy; however , the role and machinery of infection-induced autophagy remain elusive . We show that the pleiotropic Src kinase Lyn mediates phagocytosis and autophagosome maturation in alveolar macrophages ( AM ) , which facilitates eventual bacterial eradication . We report that Lyn is required for bacterial infection-induced recruitment of autophagic components to pathogen-containing phagosomes . When we blocked autophagy with 3-methyladenine ( 3-MA ) or by depleting Lyn , we observed less phagocytosis and subsequent bacterial clearance by AM . Both morphological and biological evidence demonstrated that Lyn delivered bacteria to lysosomes through xenophagy . TLR2 initiated the phagocytic process and activated Lyn following infection . Cytoskeletal trafficking proteins , such as Rab5 and Rab7 , critically facilitated early phagosome formation , autophagosome maturation , and eventual autophagy-mediated bacterial degradation . These findings reveal that Lyn , TLR2 and Rab modulate autophagy related phagocytosis and augment bactericidal activity , which may offer insight into novel therapeutic strategies to control lung infection .
Gram-negative bacteria , such as Pseudomonas aeruginosa ( hereafter Pa ) and Klebsiella pneumoniae ( Kp ) are formidable threats to human health , imposing huge healthcare costs worldwide . Pa is the fourth most commonly isolated nosocomial pathogen , accounting for 10% of all hospital-acquired infections , while Kp is the third most commonly isolated pathogen from the blood of sepsis patients [1 , 2] . Increased incidence of antibiotic resistance and lack of effective treatment approaches further limit current interventional options . Alveolar macrophages ( AM ) are on the front line of host defense with a potential role in eliminating bacterial pathogens by phagocytosis and inflammatory responses [3 , 4] . Despite decades of extensive research efforts , the role of AM in phagocytosis and clearance of extracellular bacteria remains incompletely understood , which hinders the development of effective therapeutic strategies . Autophagy is a highly conserved homeostatic mechanism for degrading bulk cellular components during starvation , or other scenarios , to provide the cell with essential nutrients . It has been linked to a wide variety of normal physiological processes including energy metabolism , organelle turnover , growth regulation , and aging [5] . Impaired autophagy can affect the process of various diseases , such as cardiomyopathy , cancer , and infection [6] . Innate immune effectors , such as toll like receptors ( TLRs ) , are important for host defense against pathogens through initiation of phagocytosis and inflammatory response [7] . Autophagy may be modulated following the recognition of conserved pathogen-associated molecular patterns ( PAMPs ) , which interact with host pattern recognition receptors ( PRRs ) , such as TLRs [8 , 9] . Autophagy can be induced in murine macrophages by several TLR ligands , including poly ( I:C ) ( TLR3 ) , LPS ( TLR4 ) and single strand RNA ( TLR7 ) [7] . Interactions between phagocytes , including AM , and bacteria may critically influence the fate of both pathogens and phagocytes through multiple signaling cascades [10] . However , little is known about whether there is interaction between autophagy and phagocytosis during bacterial invasion . Further characterization of the mechanistic underpinnings required to launch and execute immune defenses to eliminate bacterial infection is expected to significantly improve our knowledge of bacterial pathogenesis , thereby providing insight into the design of novel and effective therapeutics . One of the central themes in effective host defense is to understand how host cells counteract invasive bacteria , especially participating in the transport of bacteria to lysosomal killing environments for proteolytic digestion . A recent study of the intracellular bacterium Mycobacterium showed that the autophagy adaptor SQSTM1 ( p62 ) can enhance delivery of bacterial cytosolic components and increase bacterial killing following phagocytosis [11] . Autophagy adaptors , such as SQSTM1 , NDP52 and optineurin , were shown to mediate LC3 recruitment to the ubiquitinated substrate during ubiquitin-dependent xenophagy . Formation of the isolation membrane takes place in the proximity of the early phagosomes . Subsequently , the autophagosome engulfs the pathogen-containing phagosome . In contrast to the double-membraned autophagosome , which is not formed in LC3-associated phagocytosis ( LAP ) , the phagosomal membrane is impacted directly by LC3 [12 , 13] . Prior studies implicated that the Src kinase Lyn initiates FcγR-mediated phagocytosis and participates in the process of post-phagosome formation by interacting with cytoskeletal proteins [14 , 15] . In the case of the extracellular bacterium Pa , we discovered that Lyn , lipid rafts , and TLR2 may play a role in phagocytosis [16 , 17] . Here , we demonstrate that TLR-2 is required for inducing Lyn activity in host defense against Pa infection by facilitating autophagosome maturation . We hypothesized that Lyn-mediated phagocytosis may link autophagy to phagocytosis in a TLR2-Lyn dependent manner . We report that Lyn is a critical upstream signaling component , which expands the concept of general xenophagy [12 , 18] . In addition , we dissected the molecular and cellular bases regarding how Lyn and autophagy contribute to innate immunity through the eventual degradation of bacterial components .
To analyze the expression pattern of autophagy-related genes , we determined their mRNAs in mouse alveolar macrophage MH-S cells after Pa infection using an autophagy based RT2 Profiler PCR Arrays ( catalogue number: PAMM-084Z , Qiagen , Valencia , CA ) . The array analysis revealed that many autophagy related mRNAs ( i . e . , LC3-II , Atg4C , and Atg16L2 ) were upregulated in macrophages ( S1A and S1B Fig , S1 Table ) , suggesting that autophagy may be involved in bacterial infection . To dissect whether the critical E1 enzyme Atg7 was required for host defense against Pa , we targeted Atg7 by siRNA in MH-S cells or isolated primary AM from wild type ( WT ) and atg7 knockout ( atg7-/- ) mice ( S2A Fig ) . Phagocytosis or clearance assays were performed after shorter ( 1 h ) or longer ( 12 h ) periods of Pa infection [19] . Reduction or depletion of Atg7 showed decreased cell viability and reduced bactericidal activity upon Pa infection ( S2B and S2C Fig ) . Importantly , inhibition of autophagy by 3-MA treatment decreased bacterial killing by macrophages without affecting the cell viability , thus resulting in bacterial evasion from immune response ( S2D and S2E Fig ) , whereas rapamycin ( Rapa , autophagy inducer ) increased bactericidal activity upon infection ( S2D and S2E Fig ) . These results imply that autophagy may facilitate bacterial clearance in macrophages upon Pa infection . A prior study suggested that Lyn may be a critical factor during Pa infection [17] . MH-S cells and mouse primary AM were used to assess phagocytosis and bacterial clearance using CFU assays . After Lyn knockdown , the cell viability was determined using a MTT assay , which showed no difference upon Pa infection ( Fig 1A ) , while bactericidal activity was impaired by Lyn deficiency ( Fig 1B ) . Next , we transfected MH-S cells with Lyn-GFP , scrambled siRNA or Lyn siRNA , and infected the cells with another fluorescent-emitting and lipid-affinitive strain , Pa-Cherry . We observed significant Lyn aggregates ( puncta ) around the invading Pa by confocal laser scanning microscopy ( CLSM ) ( Fig 1C ) . CLSM imaging demonstrated that internalization by macrophages was decreased with Lyn deficiency at the early time , suggesting the involvement of Lyn in phagocytosis ( Fig 1D ) . After removing the floating and surface bacteria by polymyxin B treatment at 1 h , the internalized Pa was killed with time . Lyn deficient AM showed decreased bactericidal ability ( Fig 1D ) . Endogenous LC3 conversion into phosphatidylethanolamine-conjugated LC3-II was drastically inhibited by Lyn siRNA in MH-S cells as determined by immunoblotting ( Fig 1E and 1F ) . Further , Lyn knockdown impeded LC3 puncta formation ( Fig 1G and 1H ) . To reflect the physiological relevance , we also isolated primary AM to determine whether Lyn regulates autophagy in these cells . Similarly , significantly-decreased LC3 conversion was found in primary AM from Lyn-/- mice upon Pa infection ( Fig 1I and 1J ) ; and primary AM from Lyn-/- mice also exhibited a marked reduction in LC3 puncta as detected by CLSM imaging ( Fig 1K and 1L ) . MLE-12 cells , a murine lung epithelial cell line that does not appreciably phagocytose or clear bacteria , also showed LC3 puncta upon Pa infection , which was also impaired by Lyn deficiency or by perturbing autophagy ( 3-MA addition ) ( S2F Fig ) . In vivo experiments were then used to determine autophagy induction upon Pa infection . Lyn-/- mice exhibited increased mortality as compared to WT mice after Pa infection ( Fig 1M ) . The increased bacterial burdens in the lungs also implied severely impaired bacterial killing in Lyn-/- mice ( S2G Fig ) . Lung histological analysis revealed that induction of LC3 puncta was decreased ( Fig 1N ) , and pathophysiological tissue damage was more severe in Lyn deficiency following Pa infection ( S2H Fig ) . Since PMN play an important role in innate immunity to Pa infection [20] , PMN migration into infected lungs was measured and found to be increased in Lyn-/- mice as determined by immunostaining with Ly6G ( S2I Fig ) . These results indicate that Lyn is involved in an autophagy related phagocytosis to benefit Pa clearance in macrophages , which may preserve host cell viability after infection . To distinguish whether Lyn plays a role in Pa infection-induced autophagy , we first blocked internalization of Pa into macrophages using cytochalasin D ( CD ) , a phagocytosis inhibitor . While CD had no effect on cell viability upon Pa infection ( Fig 2A ) , its addition inhibited phagocytosis by macrophages ( Fig 2B ) . Interestingly , CD decreased LC3 conversion and phosphorylation ( Tyr297 ) of Lyn ( pLyn ) upon Pa infection ( Figs 2C and S3A ) . CLSM imaging also indicated that CD decreased LC3 puncta upon Pa infection ( Fig 2D and 2E ) . To define the relationship of CD with autophagy , rapamycin was used as a positive control to induce autophagy . CD did not induce LC3 conversion or phosphorylation of Lyn . Moreover , CD did not inhibit rapamycin-induced LC3 conversion ( Figs 2F and S3B ) . Next , siRNA interference was used to further determine the role of Lyn in autophagy . As shown in Figs 2G and S3C , Lyn deficiency did not affect rapamycin-induced autophagy without infection . CLSM imaging also confirmed that Lyn deficiency did not affect rapamycin-induced LC3 puncta ( Fig 2H and 2I ) . These data suggest that Lyn is required for Pa infection-induced autophagy . We hypothesized that Lyn-mediated phagocytosis is associated with autophagy upon Pa infection . Zymosan ( yeast wall extract ) was used to treat macrophages , which did not alter cell viability ( S3D Fig ) . LC3 conversion was found to be induced by Zymosan treatment . However , Zymosan did not induce Lyn phosphorylation ( Fig 2J and S3E Fig ) . In addition , CLSM imaging showed LC3 puncta formation while no Lyn puncta were formed upon Zymosan treatment of macrophages ( Fig 2K and 2L ) . Furthermore , Gram-positive bacterium S . pyogenes ( Sp ) infection was used to determine the induction of LC3 puncta and Lyn puncta by CLSM imaging ( S3F Fig ) . Collectively , these data indicated the role of Lyn in bacterial infection-induced autophagy . To distinguish conventional autophagy from LAP , the upstream preinitiation complex ( ULK1/2 , which is required for autophagy but not LAP ) , and Rubicon ( which mediates LAP but not autophagy ) were knocked down by siRNA interference strategy , respectively ( S3G Fig ) [21] . Knockdown of Rubicon does not affect the cell viability upon Pa infection ( Fig 2M ) ; however , ULK1 interference resulted in decreased bacterial phagocytosis by macrophages ( Fig 2N ) . In addition , ULK1 interference led to decreased pLyn level and also LC3 lipidation as compared to Ctrl siRNA transfected cells ( Figs 2O and S3H ) . CLSM imaging quantitation also showed decreased LC3 puncta by ULK1 deficiency upon Pa infection ( Fig 2P ) . However , Rubicon interference did not affect Lyn phosphorylation while slightly decreased LC3 lipidation upon Pa infection , with only limited phagocytosis reduction , indicating that LAP is less relevant to this infection model ( Fig 2M–2P , S3H Fig ) . To further determine that LC3-lipidation and dot formation occur upon Pa infection , chloroquine was used to inhibit lysosome activities . Lyn phosphorylation was found with or without chloroquine . Furthermore , we observed that LC3-II conversion had accumulated with chloroquine treatment upon Pa infection determined by immunoblotting ( S3I and S3J Fig ) . Interestingly , CLSM imaging showed that LC3 recruitment and phosphorylation of Lyn localized together with invading Pa ( Fig 3A and 3B ) . To unravel whether the phagocytosis of Pa is involved in autophagosome formation , we detected newly formed autophagosomes using transmission electron microscopy ( TEM ) on Pa-infected MH-S cells . Pa-containing autophagosomes with double membranes were identified by TEM ( Fig 3C and 3D ) , providing additional evidence that Pa-containing autophagosomes had formed , which is also termed as xenophagy . Next , co-immunoprecipitation ( co-IP ) was performed to determine the interaction of Lyn with autophagy-related proteins . As shown in Fig 3E , we observed that Lyn could bind well to Atg12-Atg5 and LC3 complex upon Pa infection . In addition , we found that Lyn associated with Pa ( Fig 3E ) , which suggested that Lyn may play a role in regulating autophagosome formation upon bacterial infection . To further understand the molecular mechanism in Lyn-mediated autophagosome formation , we asked which domain ( s ) of Lyn is required for interaction with the functionally or structurally relevant proteins . Lyn-GST peptides with distinct functional domains ( see schematic , Fig 3F ) were prepared as described previously [16 , 22] . The Lyn-GST fragments were coated on immobilized glutathione agarose beads to pull-down interacting partners from MH-S cell lysates and probed for Atg12-Atg5 , LC3 , or Pa , respectively . The presence of both Src homology 2 ( SH2 ) and SH3 domains of Lyn were found to be required for interaction with signaling proteins within autophagosomes , whereas the kinase domain was dispensable for the interaction upon Pa infection ( Fig 3G , S4A and S4B Fig ) . These data further suggested that Lyn may interact with critical autophagic proteins localized in autophagosomes through SH2 and SH3 binding regions . We next investigated whether this interaction specifically occurred in macrophages with strong phagocytosis . Primary AM and epithelial cells ( MLE-12 ) were infected with Pa and cell lysates were applied for pull-down by Lyn-GST fragments . Although SH2 and SH3 domains of Lyn were required for interactions with autophagic proteins and binding to Pa in macrophages , no apparent Lyn-Pa interaction in MLE-12 cells was observed ( S4C and S4D Fig ) , which may be due to different mechanisms of uptaking Pa by epithelial cells . To confirm that Lyn activation was responsible for the underlying process , MH-S cells were transfected with Lyn WT , Lyn K275D ( dominant negative ) , and Lyn Y508F ( constitutively active ) constructs and treated with Pa . The transfection of different plasmids had no effect on the cell viability upon infection ( Fig 3H ) , whereas Lyn K275D transfected cells showed impaired phagocytic ability upon bacterial infection ( Fig 3I ) . Lyn activation was detected by immunoblotting for pLyn . While Lyn Y508F plasmid-transfected cells showed enhanced activity of Lyn kinase , Lyn K275D-transfected cells showed decreased activity as compared to Lyn WT-transfected cells ( Figs 3J and S4E ) . LC3-II conversion was found to be inhibited in Lyn K275D transfected cells upon Pa infection ( Figs 3J and S4E ) . Furthermore , LC3 puncta revealed that Lyn kinase activities were well correlated with Pa-induced autophagy in macrophages ( Fig 3K and 3L ) . Our data also showed that PP2 ( a commonly used Lyn inhibitor ) reduced infection-induced autophagy ( S4F and S4G Fig ) . Overall , the results demonstrated that the role of Lyn in infection-induced autophagy was dependent on its kinase activities . We next attempted to determine whether a single bacterial component ( i . e . LPS ) , can induce autophagy . Interestingly , heat-killed Pa ( HKPa ) and LPS all caused LC3-II conversion to a similar extent as Pa , as detected by immunoblotting ( Figs 4A and S5A ) . On the other hand , CLSM fluorescence microscopy showed that HKPa or LPS also induced LC3 puncta ( Fig 4B and 4C ) . Importantly , we also observed increased phosphorylation of Lyn by LPS pretreatment ( Figs 4A and S5A ) . These results indicate that Pa or LPS is potentially sufficient to induce autophagy in macrophages . Based on our previous observations [17] , we probed the role of TLR2 and TLR4 as upstream signals to initiate autophagy . After TLR2 or TLR4 knockdown ( Fig 4D ) , MH-S cells were infected with Pa . Both TLR2 and TLR4 deficiency impaired the phagocytosis of Pa ( Fig 4E ) , while cell viability was not altered ( Fig 4F ) . However , only TLR2 knockdown showed impaired LC3 puncta formation upon bacterial infection ( Fig 4G and 4H ) . Immunoblotting also showed that TLR2 deficiency inhibited LC3 conversion and Lyn phosphorylation upon bacterial infection ( Figs 4I and S5B ) . These results suggest that TLR2 may play a role in autophagy induction during Pa infection ( although other TLRs’ role cannot be excluded at this time ) . To delve into the involvement of TLR2 , HEK293 cells with stable TLR2 expression were infected by Pa . Although these cells are not phagocytes , Lyn activation ( shown as pLyn level ) or autophagy induction ( shown as LC3 conversion or LC3 puncta ) was much stronger in cells overexpressing TLR2 than vector control cells ( Fig 4J–4L , S5C Fig ) . Pam3CSK4 , a TLR2 agonist , increased phagocytosis of Pa ( Figs 4M and S5D ) ; however , Pam3CSK4 had no impact on LC3 puncta if left uninfected ( S5E Fig ) . Pam3CSK4 also increased Lyn activation or autophagy in MH-S cells upon Pa infection as determined by immunoblotting ( Figs 4N and S5F ) . These results suggested that Lyn may serve as a bridge between TLR2 and autophagy after Pa infection . To assess whether autophagy has a role on Lyn-dependent elimination of Pa in macrophages , we knocked down Atg5 and Beclin1 using siRNA ( Fig 5A ) . Atg5 or Beclin1 has little effect on cell viability upon Pa infection ( Fig 5B ) ; however , phagocytosis was inhibited by down-regulating these two genes ( Fig 5C ) . Immunoblotting also showed that the phosphorylation of Lyn was decreased by Atg7 , Atg5 or Beclin1 deficiency ( Fig 5D and 5E ) . CLSM imaging also showed the LC3 puncta and pLyn level was dampened by a deficiency in either of these autophagic genes ( Fig 5F and 5G ) . The clearance of phagocytized bacteria was also dampened by inhibiting Atg5 or Beclin1 in AMs upon Pa infection ( Fig 5H and 5I ) . We then performed immunoblotting to verify the digestion of the invading bacteria and found that when autophagic genes were knocked down , the degradation of the bacteria proteins was blocked over time ( Fig 5J and 5K ) . All these results suggest that autophagy plays an important role in Lyn-mediated phagocytosis and clearance upon bacterial infection . To further assess the phagocytosis process , we elucidated whether Rab5 ( early endosome marker ) is involved in Lyn-mediated phagocytosis and subsequent bacterial clearance . We transfected MH-S cells with Rab5-RFP plasmid and infected the cells with Pa for different times . Immunostaining was next performed to probe pLyn to evaluate the internalized bacteria using CLSM microscopy . The number of phagocytized Pa was increased at the early time of infection . Interestingly , Rab5 and Lyn were both found to be surrounded with invading Pa , which implied that Rab5 may play a role in conjunction with Lyn in phagocytosis ( Fig 6A and 6B ) . Next , co-IP was performed to confirm the interaction of Lyn and Rab5 ( Figs 6C and S6A ) . To this end , we isolated phagosomes after Pa infection and the results showed that pLyn was found in the same fraction as Rab5 ( e . g . , phagosomes ) , while PP2 pretreatment impaired this process ( Fig 6D ) . To test our hypothesis that Lyn is involved in Rab5-mediated autophagy related phagocytosis of Pa , we co-transfected Rab5-RPF , LC3-GFP with Ctrl or Lyn siRNA into MH-S cells . After Lyn knockdown , the formation of autophagosomes was impaired upon Pa infection ( Fig 6E and 6F ) . Similar results were found with Rab7 ( late endosome marker ) . We observed an interaction of Lyn with Rab7 ( S6A Fig; Lyn deficiency also decreased autophagosome-lysosome formation upon Pa infection ( S6B and S6C Fig ) . These data suggest that Lyn-mediated phagocytosis may be dependent on the conserved endocytosis pathway involving Rab families . We then used dominant negative plasmids to down-regulate Rab5 and Rab7 and found that Rab5-DN-RFP transfection did not affect cell viability ( Fig 6G ) , while dampened the phagocytosis ( Fig 6H , S6D and S6E Fig ) and colocalization of Rab5 with Pa ( Fig 6I and 6J ) . Importantly , Rab7-DN-RFP transfection also significantly inhibited phagocytosis and colocalization of Rab7 and Pa inside the autolysosome ( S6F and S6G Fig ) . Collectively , these findings suggest that Lyn , together with Rab family members , participated in the transportation of invading bacteria in phagocytosis and clearance upon infection . To determine whether Lyn is involved in actin remodeling in macrophages , we performed co-IP and found that Lyn interacted with actin after Pa infection ( Fig 6K ) . Lyn siRNA-transfected cells showed significantly decreased polymerized F-actin indicated by phalloidin staining ( around Pa containing phagosomes ) vs . Ctrl siRNA-transfected cells at both 2 h and 8 h ( Fig 6L and 6M ) . Also , flotillin-1 expression , which is required for nucleation of actin on phagosomal membrane , correlated with the internalization of Pa , and was inhibited by Lyn siRNA interference ( S6H and S6I Fig ) . These findings prompted us to determine whether Pa-containing phagosomes are involved in polymerized actin , we evaluated the phosphorylation of cofilin , a conserved actin-modulating protein . Uninfected cells allowed the phosphorylation of basal amounts of cofilin , whereas Pa infection led to increased phosphorylation of cofilin . Further , cofilin remained unphosphorylated in Lyn-deficient macrophages following Pa infection ( S6H and S6I Fig ) . Moreover , we found that Lyn interacted with cofilin after Pa infection as evaluated with co-IP ( S6J Fig ) . Thus , interaction between actin and Lyn is required for modulation of the phosphorylation state of cofilin during Pa infection . Thus , Lyn is required for both F-actin network formation and phosphorylation of cofilin during Pa infection . To determine the critical role of Lyn in delivering bacterial components and in enabling intracellular traffic , a process required for autophagosome-lysosome fusion , LAMP1-RFP was transfected into macrophages to track lysosomes . We found that both LAMP1 and pLyn were colocalized with internalized Pa at 8 h post infection ( Fig 7A and 7B ) . To biochemically characterize the role of Lyn and autophagy in Pa trafficking inside the cell , we isolated phagosomes from Pa-infected cells using a sucrose gradient method [16] . Horse radish peroxidase ( HRP ) was used to determine phagosome distribution and Pa Ab was used to show phagosome . We found that both Atg12-Atg5 and LC3 were colocalized with Lyn , and were shifted into early phagosome , late phagosome or phagolysosome fractions following infection ( Fig 7C and 7D ) . Importantly , Pa bacterial proteins were detected using polyclonal Pa antibody in phagolysosome ( 45%; fractions 3 , 4 , 5 ) , early phagosome ( 30%; fraction 6 ) and late phagosome ( 25%; fractions 7 , 8 , 9 ) , while negligible in other fractions as determined by immunoblotting ( Fig 7C ) . The presence of LAMP1 in fractions 9 and 10 indicates the autophagosome fusion with lysosome and a potential role of autophagy in bacterial delivery and clearance ( Fig 7C ) . CLSM imaging showed that the colocalization of LC-3 and Lyn was drastically impaired by Lyn knockdown ( Fig 7E and 7F ) . Importantly , proteolytic assay also showed that the degradation of the phagocytized bacteria ( Figs 7G–7I , S7A ) , while the lysosome inhibitor chloroquine prevented the digestion of the invading bacteria ( Figs 7J and S7B ) . Finally , we used another clinically significant Gram-negative bacterium , K . pneumoniae ( Kp ) , to elucidate whether the autophagy related phagocytosis is a general phenomenon during bacterial infection . As expected , Kp infection also induced LC3 puncta around internalized bacteria , which was inhibited by Lyn deficiency or 3-MA ( S7C–S7G Fig ) . To further dissect that autophagy-dependent innate immunity is generally important for immunity to P . aeruginosa or K . pneumoniae infections , additional strains ( PAK , PA14 , Kp , and Kp-Xen 39 ) were used to determine whether manipulation of autophagy impacts bacterial survival . Reduced bacterial killing was caused by autophagy inhibition or Lyn deficiency , indicating that Lyn-assisted autophagy is a general pathway for innate immunity against these microorganisms . These results collectively demonstrate that Lyn has an essential role in Gram-negative bacteria-induced autophagy , facilitating bacterial clearance . Fig 7K illustrates a model delineating the role of Lyn in Pa-induced autophagy related phagocytosis .
The current study demonstrates that TLR2/Lyn signaling plays a critical role in recruitment of autophagic components to bacteria-containing phagosome . We show that Lyn mediates the formation of the membrane isolation in the proximity of the Pa-containing phagosome in AM . Our data further reveal that Lyn transmits signaling for cytoskeleton protein-mediated intracellular transport following TLR2 activation , resulting in a merge between autophagosome and phagosome . This autophagosome-lysosome fusion helps eliminate the internalized Pa , leading to enhanced bacterial clearance . Further , blocking autophagy and/or depleting Lyn hamper the delivery of bacteria into lysosomes , thereby dampening bacterial eradication . Taken together , we characterize TLR2 and Lyn as new regulators of bacteria-induced autophagy and phagocytosis both in vitro and in vivo , providing new insight into the molecular mechanism of immune responses against acute bacterial infection . Autophagy induced by intracellular bacterial infections has garnered increasing interest and has been shown to play crucial roles in host defense in immunological cells [23 , 24] . This autophagic role is relevant to the bacterial fates after phagocytes engulf some critical pathogenic bacteria , such as M . tuberculosis , L . monocytogenes , S . enterica and S . flexneri [24–27] . Previous reports have defined that macrophages with Atg7 deficiency could express enhanced scavenger receptor to phagocytize and retain M . tuberculosis [28] . However , whether autophagy can influence the uptake of extracellular bacteria , such as Pa or Kp , is largely unknown . A recent report showed that Pa infection enhanced autophagy [29] , and that Atg7-/- mice showed enhanced susceptibility to infection and impaired bacterial clearance [29] . However , there is no in vivo evidence to show reduced bacterial clearance associated with autophagy deficiency in Lyn-/- mice [30 , 31] . Using mouse models , we show that autophagy related proteins contribute to xenophagy of extracellular bacteria Pa or Kp , thus affecting phagocytosis of host macrophages . Our data implicate that the role of autophagy varies with bacteria . Previous studies indicate that PRRs may be activated in a PAMP-specific manner . The TLR5-mediated response was shown to be induced by flagellin , while LPS appears to link to the TLR4-mediated response [32 , 33] . LPS is one of the important structural components in the outer membrane of Gram-negative bacteria . CD14 facilitates the transport of LPS to the TLR4/MD-2 complex and regulates LPS recognition [34] . However , TLR2 and TLR4 recognize different bacterial cell wall components [35 , 36] . We show that Pa induces autophagy and that HKPa initiates host inflammatory responses through TLR2 . We noted that LPS , a TLR4 ligand , also induced autophagy . Previous studies indicated that TLR2-mediated recruitment of LC3 to phagosomes may promote phagolysosome fusion [18] . Consistent with this observation , we showed that Lyn boosts phagocytosis of macrophages and facilitates subsequent intracellular trafficking processes . Our previous studies showed that Lyn plays a role in Pa infection [16 , 17 , 31] . However , whether Lyn can regulate autophagy has remained elusive . Others show that blocking autophagy sensitizes prostate cancer cells towards Src family kinase inhibitors [37] . Interestingly , Lyn is reported to increase the survival of glioblastoma cells under starvation-induced autophagy [38] . Here , we identify a role of Lyn in initiating autophagy during Pa infection . Unlike LAP , typical double-membraned autophagosomes that contain bacteria are formed in our study , which is also specifically relevant to the activity of Lyn . We also noted that blocking autophagy by 3-MA impaired phagocytosis . mTOR is a major negative regulatory axis of autophagy , thus inhibition of mTOR and its associated pathways can induce autophagy . We also found that rapamycin increases autophagy to elevate innate immune defense against Pa infection . These findings indicate that Lyn facilitates phagocytosis through autophagic mechanisms and serves as a link between phagosomes and autophagosomes . Thus , we delineate a novel mechanism by which Lyn mediates autophagy-related phagocytosis to benefit host against Pa . We also demonstrate that Lyn’s activity is required for inhibiting Pa infection , as depletion of Lyn and/or Atg7 in macrophages resulted in increased Pa growth . Hence , the participation of Lyn and Atg7 suppressed Pa replication , whereas the absence of Lyn or Atg7 or other autophagic proteins ( Atg5 and Beclin1 ) allowed for an uncontrolled bacterial growth . Murine macrophages lacking Lyn are defective in migration and in phagocytosis [17] . These observations suggest that the uptake of Pa may be impaired in Lyn-deficient macrophages . Previous studies suggest that intracellular growth of bacteria requires the halt of phagolysosome fusion [39] . This intracellular trafficking defect is observed in permissive macrophages , and most bacteria-containing vacuoles fuse with lysosomes to clear bacteria [40] . How autophagosome-lysosome fusion is modulated upon Pa infection is poorly understood . Our findings demonstrate that Lyn may modulate the delivery of bacteria to lysosomes , facilitating autophagosome maturation and subsequent bacterial clearance . Phagosomes bind microtubules and actin filaments and migrate to interact with other compartments within the cell [41] . Our studies indicate that Lyn is involved in polymerization of F-actin , a cofilin-dependent process . Rab5 , an early endosome marker , enables the recruitment of early endosomes to form nascent phagosomes and is also required for the recruitment of Rab7 ( a late endosome marker ) , thus promoting phagolysosome fusion [42] . By isolating phagosomes to biochemically identify which phagosomal constituents interact with Lyn , we found that both Rab5 and Rab7 interact with Lyn , which may be downstream regulators following Lyn-mediated autophagy related phagocytosis . Thus Rab5 and Rab7 may be required for delivering bacterial components to lysosomes in Lyn-mediated Pa phagocytosis and autophagosome-lysosome fusion . In summary , our studies represent autophagic benefit in host defense against Pa through a TLR2/Lyn axis . The contribution of Lyn to autophagy is found to facilitate the removal of intracellular bacteria within macrophages . Our results also demonstrate that TLR2 activates Lyn . Identification of both Lyn and TLR2 as novel regulatory factors in autophagy may implicate their therapeutic potential for control of Gram-negative bacterial infection .
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocols were approved by the Institutional Animal Care and Use Committee at the University of North Dakota , School of Medicine ( Assurance Number: A3917-01 ) . Dissections and injections were performed under anesthesia that was induced and maintained with ketamine hydrochloride and xylazine , and all efforts were made to minimize suffering . atg7-deficient ( floxed atg7-/- ) mice were provided by Dr . Youwen He at Duke University and these mice were originally constructed by Masaaki Komatsu at Tokyo Metropolitan Institute of Medical Science [43] . To generate mice with a monocyte-specific knockout of Atg7 , we have generated myeloid-specific conditional knockout ( KO ) mice by cross-breeding atg7flox/flox with lyz2tm1 ( Cre ) lfo>/j mice ( Jackson Laboratory , Bar Harbor , ME ) [44] . The knockout was induced by injecting 0 . 1 mg of tamoxifen ( Sigma , St Louis , MO ) daily for 5 days [45] . lyn KO mice were provided by Dr . S . Li at University of Massachusetts . Age and sex matched wild-type ( WT ) mice ( C57BL/6J ) were used as controls [31] . MH-S ( ATCC CRL-2019 ) were cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum ( HyClone Laboratories , Logan , UT ) and 100 U/ml of penicillin/streptomycin ( P/S , Life Technologies , Rockville , MD ) antibiotics in a 37°C incubator with 5% CO2 . Mouse AM were isolated by bronchoalveolar lavage ( BAL ) [46] . In brief , trachea was cannulated with a 20-gauge catheter; 0 . 9 ml BAL buffer was instilled , flushed four times , and retrieved . A total of 3 . 0 ml BALF was retrieved from each mouse and cytospin slides prepared with 0 . 5 ml BALF were stained by HEMA-3 ( Fisher , Rockford , IL ) to enumerate leukocyte subtypes based on their cellular and nuclear morphological properties . After centrifugation at 2 , 000 rpm , AM cells were resuspended and cultured in RPMI 1640 medium as above . MLE-12 ( ATCC CRL-211 ) were cultured in HITES medium as above . Stable TLR2 expression HEK293 cells ( ATCC CRL-157 ) using a pUNO-TLR2 plasmid were obtained from InvivoGen ( San Diego , CA ) and cultured in DMEM medium as above . P . aeruginosa wild-type ( WT ) strain PAO1 was a gift from Dr . Stephen Lory ( Harvard Medical School , Boston , MA ) . The PAO1-GFP strain was obtained from Dr . Gerald Pier ( Department of Microbiology and Molecular Genetics , Brigham and Women’s Hospital , Harvard Medical School ) . A Pa-cherry strain was obtained from Dr . John Singer of University of Maine ( Orono , ME ) . Kp-GFP ( ATCC 43816 serotype II ) was kindly provided by Dr . Steven Clegg ( University of Iowa Carver College of Medicine , Iowa City , IA ) . Bacteria were grown overnight in lysogeny broth ( LB ) at 37°C with vigorous shaking . Next day , the bacteria were pelleted by centrifugation at 8 , 000×g and resuspended in 10 ml of fresh LB broth , in which they were allowed to grow until the mid-logarithmic phase . Thereafter , the optical density ( OD ) at 600 nm was measured , and the density was adjusted to 0 . 25 OD ( 1 OD = 1×109 cells/ml ) . Before infection , cells were washed once with PBS , and replaced with antibiotic-free medium immediately [47] . Cells were infected by Pa at multiplicity of infection ( MOI ) of 10:1 ( bacteria-cells ratio ) for 1 h . After infection , macrophages were washed 3 times with PBS , extracellular bacteria were removed by addition of 100 μg/ml polymyxin B and left in incubation for another 1 h . Phagocytosis was evaluated by counting colony forming unit ( CFU ) [16] . Bacterial clearance by AM cells was also determined using the CFU assay after 1 h infection , following with 1 h polymyxin B incubation and then 10 h culture in antibiotics free medium to gauge eradication of the internalized bacteria , indicating clearance of bacteria [48] . For in vivo experiments , 40 mg/kg ketamine was used for anesthesia . Mice were intranasal instilled with 1×107 CFU of Pa ( suspended in 25 μl PBS ) . Mice were sacrificed 24 h post infection . The lungs were collected for paraffin section . For survival experiments , mice ( 6/group ) were infected as above and monitored till moribund . Cells were transfected with LC3-RFP , LC3-GFP , Lyn-GFP , Rab5-RFP ( dominant negative , DN ) and Rab7-RFP ( DN ) in serum-free medium ( Thermofisher Scientific ) , and achieved high efficiency using LipofectAmine 2000 reagent ( Life Technologies , Grand Island , NY ) following the manufacturer’s instructions . Similarly , we performed siRNA transfection assays [49] . Real-time PCR profiling of mRNAs were conducted on a SYBR Green-based , RT2 Profiler PCR Array System ( Cat# PAMM-084Z , Qiagen , Valencia , CA ) . The array includes 84 key genes that encode components of the molecular machinery and key regulators modulating autophagy in response to both extracellular and intracellular signals . Expression of mRNAs was detected by QuantiTect SYBR Green RT–PCR Kit ( Qiagen , Valencia , CA ) . The separate well 2-ΔΔCt cycle threshold method was used to determine relative quantitative levels of individual mRNA , and these were expressed as the fold difference to GAPDH , respectively . Rabbit polyclonal antibodies against , Atg7 , Atg12-Atg5 , TLR2 , Lyn , pcofilin-1 , cofilin-1 , flotillin-1 , actin and mouse polyclonal antibodies against LC3 , GAPDH were obtained from Santa Cruz Biotechnology ( Santa Cruz , CA ) . Rabbit monoclonal antibody against phosphor-Lyn ( Tyr297 ) was obtained from Cell Signaling Technology ( Danvers , MA ) . Rabbit monoclonal antibody against flagellin type b and Pa were kindly provided by Dr . Gerald Pier of Harvard Medical School . The samples from cells were lysed and quantified . The lysates were boiled for 10 min , and added with protease inhibitor cocktail ( Roche Diagnostics , Indianapolis , IN ) . The supernatants were collected and 20 μg of each sample were loaded onto 12% SDS-polyacrylamide gel electrophoresis ( PAGE ) and electrophoresed for protein resolution . The proteins were then transferred to polyvinylidine difluoride membranes ( Thermofisher , Rockford , IL ) and blocked for 1 h at room temperature using 5% non-fat milk blocking buffer . Membranes were incubated for 2 h at room temperature with appropriate first antibodies diluted at 1:1000 in 5% bovine serum albumin ( BSA ) immunoblotting antibody buffer . After washing ( three times , 10 min once ) with washing solution , the membranes were incubated for 60 min at room-temperature with horseradish peroxidase-conjugated secondary antibody ( Rockland Immunochemicals , Gilbertsville , PA ) diluted 1:3000 . Signals were visualized using an enhanced chemiluminescence detection kit ( SuperSignal West Pico , Pierce ) [46] . Co-immunoprecipitation ( co-IP ) is performed using Protein A/G-argarose beads and then mix with the related antibody . The target proteins were detected using immunoblotting as above . Cells were grown either on coverslips in a 24-well plate or in glass-bottomed dishes ( MatTek , Ashland , MA ) . Rabbit monoclonal antibody against Ly6G was bought from Abcam ( Cambridge , MA ) . For immunostaining , the cells or slide tissues were fixed in 3 . 7% paraformaldehyde , permeabilized with 0 . 2% Triton X-100 in PBS and incubated with blocking buffer containing 2% BSA for 30 min . Samples were incubated with primary antibodies at 1/500 dilution in blocking buffer for 1 h and washed three times . After incubation with appropriate fluorophore-conjugated secondary antibodies , the cover slips were mounted on slides with Vectashield mounting medium . The images were captured using an LSM 510 Meta confocal laser scanning microscope ( CLSM , Carl Zeiss Micro Imaging , Dublin , CA ) , and processed using the software provided by the manufacturer [50] . GST-Lyn constructs with different functional domains were originally obtained from Dr . O . Miura ( Tokyo Medical and Dental University , Tokyo , Japan ) [22] and transformed into BL21-DE3 strain of E . coli . The GST-Lyn fragments were extracted using immobilized glutathione columns following the manufacturer’s instructions [16] . Equal whole cell lysates were incubated with GST-Lyn peptide for protein interactions . The pull-down products were analyzed by immunoblotting with specific Abs . Cell lysates were processed for phagosome isolation , as previously described [51] . Briefly , the cell homogenates were separated by sucrose gradient density centrifugation ranging from 90 to 5% of sucrose cushion . The phagosomes were heavy and separated at 65% density , whereas lysosomes were found at the interface of 35 and 5% densities . Fractions ( 1 . 1 ml ) were collected from top ( total 10 fractions ) corresponding to the density gradient . TEM was employed for identifying autophagosomes using modified Karnovsky’s fixative [52] . Images were taken and processed according to our previous reports [30] . All experiments were performed in triplicate and repeated at least three times . Data are presented as changes compared with the controls from the three independent experiments . Results are shown as means±SD . One-way ANOVA ( Tukey’s post hoc ) was used for statistical analysis . Differences were accepted as significant at p<0 . 05 .
|
It is vital to establish the mechanistic basis for initiation of host defenses and immune responses that are required to eliminate bacterial infection . This line of inquiry will increase knowledge of bacterial pathogenesis and uncover new insights that can enhance design and effectiveness of novel therapeutics . We demonstrate that TLR-2 is required for inducing Lyn activity in host defense against Pa infection through assistance in autophagosome maturation , and may link autophagy to phagocytosis in a TLR-2-Lyn-dependent manner . Thus , these results may further help to alleviate human acute lung injury/adult respiratory distress syndrome ( ALI/ARDS ) caused by Gram-negative bacteria .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2016
|
Lyn Delivers Bacteria to Lysosomes for Eradication through TLR2-Initiated Autophagy Related Phagocytosis
|
Dengue is a leading cause of fever and mimics other acute febrile illnesses ( AFI ) . In 2009 , the World Health Organization ( WHO ) revised criteria for clinical diagnosis of dengue . The new WHO 2009 classification of dengue divides suspected cases into three categories: dengue without warning signs , dengue with warning signs and severe dengue . We evaluated the WHO 2009 classification vs physicians’ subjective clinical diagnosis ( gestalt clinical impression ) in a large cohort of patients presenting to a tertiary care center in southern Sri Lanka hospitalized with acute febrile illness . We confirmed acute dengue in 388 patients ( 305 adults ≥ 18 years and 83 children ) , including 103 primary and 245 secondary cases , of 976 patients prospectively enrolled with AFI . At presentation , both adults and children with acute dengue were more likely than those with other AFI to have leukopenia and thrombocytopenia . Additionally , adults were more likely than those with other AFI to have joint pain , higher temperatures , and absence of crackles on examination whereas children with dengue were more likely than others to have sore throat , fatigue , oliguria , and elevated hematocrit and transaminases . Similarly , presence of joint pain , thrombocytopenia , and absence of cough were independently associated with secondary vs primary dengue in adults whereas no variables were different in children . The 2009 WHO dengue classification was more sensitive than physicians’ clinical diagnosis for identification of acute dengue ( 71 . 5% vs 67 . 1% ) , but was less specific . However , despite the absence of on-site diagnostic confirmation of dengue , clinical diagnosis was more sensitive on discharge ( 75 . 2% ) . The 2009 WHO criteria classified almost 75% as having warning signs , even though only 9 ( 2 . 3% ) patients had evidence of plasma leakage and 16 ( 4 . 1% ) had evidence of bleeding In a large cohort with AFI , we identified features predictive of dengue vs other AFI and secondary vs primary dengue in adults versus children . The 2009 WHO dengue classification criteria had high sensitivity but low specificity compared to physicians’ gestaldt diagnosis . Large cohort studies will be needed to validate the diagnostic yield of clinical impression and specific features for dengue relative to the 2009 WHO classification criteria .
Dengue is an important cause of morbidity and mortality worldwide in the tropics and mimics other causes of acute febrile illness ( AFI ) [1–3] . Early recognition of severe dengue could improve outcomes [4] . The 1997 World Health Organization ( WHO ) dengue clinical classification criteria were developed to assist with surveillance , triage , and treatment decisions in the management of dengue , but were difficult to apply clinically and had poor sensitivity in identifying severe dengue [5] . Therefore , World Health Organization ( WHO ) in 2009 revised the clinical criteria to improve the diagnosis of dengue [5] . The new criteria categorized patients with suspected dengue as having dengue without warning signs , dengue with warning signs , and severe dengue , and have improved sensitivity in identifying patients with severe dengue [6] . However , the performance of these revised criteria has not been widely evaluated in Sri Lanka [5] . To compare clinical features of dengue and to assess the performance of the revised 2009 WHO clinical criteria for the diagnosis of acute dengue , we prospectively studied a large cohort with AFI during an epidemic of dengue 1 ( DENV-1 ) in southern Sri Lanka .
Trained study physicians prospectively enrolled patients with AFI admitted to the adult and pediatric wards of the largest ( 1 , 500 bed ) tertiary care hospital in the Southern Province . Consecutive patients ≥1 year of age with documented fever ( >38°C ) at presentation or within 48 hours of hospital admission were eligible . We obtained epidemiologic and clinical data and an acute blood sample at enrollment and convalescent serum at 2–4 week follow-up . We recorded the treating physicians’ subjective clinical diagnosis at presentation . Study staff prospectively extracted all clinical data and the treating physicians’ clinical diagnosis at discharge from the patients’ medical records during hospitalization . Specimens were promptly frozen at -70°C and shipped on dry ice months later for off-site testing . We retrospectively confirmed acute dengue using IgG and IgM ELISA , virus isolation , RT-PCR for DENV , and RT-PCR for flaviviruses , as previously described [7] . Confirmed dengue was defined by 1 ) IgG seroconversion alone with positive IgM or IgM seroconversion ( definitive serologic evidence ) , 2 ) PCR and either isolation or alternative target PCR ( definitive virologic evidence ) , or 3 ) PCR and/or isolation with a positive convalescent IgM ( virologic and serologic evidence ) . Primary and secondary acute dengue were determined by the absence or presence of IgG in the acute sample , respectively . The Chi square test or Fisher exact test were used for categorical variables and t-test or Kruskall-Wallis test for continuous variables . Bivariable logistic regression yielded odds ratios ( OR ) with 95% confidence intervals ( 95% CI ) for features associated with acute dengue vs other AFI and secondary vs primary dengue . Multivariable logistic regression was performed for adults and children separately . All statistically significant ( p<0 . 05 ) non-collinear clinical variables ( symptoms and signs ) were initially included ( full model ) . For laboratory values , leukopenia ( <4 x 103 cells/μL , thrombocytopenia ( <100 x 103/μL ) , and elevated transaminases ( AST or ALT > 120 IU , 3 times upper limit of normal ) were included in the models rather than absolute laboratory values . For hematocrit and hemoglobin , absolute values were used given variability in standard values between males and females and adults and children . Variables were then sequentially removed to yield the most parsimonious model ( p-value for all <0 . 05 ) . Excluded variables were then reintroduced and retained if significant . Both application of the WHO classification criteria and diagnostic testing for dengue were performed retrospectively for the purpose of the study . The treating physicians’ clinical diagnosis both at presentation and discharge reflected his or her subjective clinical impression . Laboratory confirmation of acute dengue was used as the gold standard for evaluating the accuracy of the 2009 WHO classification criteria and physicians’ clinical diagnoses at admission and discharge . To evaluate the performance of the treating physicians’ clinical diagnosis versus the WHO classification criteria at identifying acute dengue , sensitivity , specificity , positive predictive value , and negative predictive value were calculated . Receiver-operating characteristic ( ROC ) curves were generated and area under the curve ( AUC ) was calculated . STATA , version 11 ( STATACorp , College Station , Texas ) was used for all analyses . Ethical approval was obtained from Ethics committee of Faculty of Medicine University of Ruhuna , Institutional review boards of Duke University USA , Duke-National University Singapore and Johns Hopkins University USA . Written informed consent was obtained from all patients ≥ 18 years of age , parental informed consent was obtained from patients 1–17 years of age and additionally written assent was obtained from all those aged 12–17 years to participate in the study . Study doctors ( MBBS ) with pediatric experience obtained informed consent in a pre-designed consent form .
We obtained convalescent sera from 877 ( 89 . 6% ) of 976 patients enrolled; 628 ( 64 . 3% ) were male and 306 ( 31 . 4% ) <18 years . Acute dengue was confirmed in 388/976 ( 40% ) ; 39 with inconclusive results were excluded . Of confirmed dengue , 103 ( 26 . 6% ) cases were primary , 245 ( 63 . 1% ) secondary , and 40 ( 10 . 3% ) could not be classified because of insufficient acute sera . Among 351 virologically confirmed ( dengue PCR or virus isolation positive ) cases , 320 ( 91 . 2% ) were DENV-1 , 25 ( 7 . 1% ) DENV-4 , and 6 ( 1 . 7% ) DENV-2 , with similar proportions in adults vs . children [in children , 60 ( 92 . 3% ) DENV-1 , 2 ( 3 . 1% ) DENV-4 , and 2 ( 3 . 1% ) DENV-2] . In both adults and children , patients with acute dengue were older than those with other AFI and more likely to have joint pain , muscle pain , anorexia , right upper quadrant tenderness , rash , leukopenia ( <4 x 103 cells/μL ) , thrombocytopenia ( <100 x 103/μL ) , and elevated ( >3 times normal ) transaminases ( Table 1 ) . Those with acute dengue were less likely than others to have cough and lung crackles . Additionally , adults with dengue were more likely to have vomiting than those with other AFI and children with dengue were more likely to report headache , rhinitis , sore throat , and abdominal pain . On multivariable analyses , both adults and children with acute dengue were more likely than those with other AFI to have leukopenia and thrombocytopenia ( Table 2 ) . Additionally , adults with acute dengue were more likely than those with other AFI to have joint pain , higher temperatures , and absence of crackles on examination . Children with acute dengue were more likely than others to have sore throat , fatigue , oliguria , an elevated hematocrit ( ≥20% from baseline ) and elevated transaminases . Patients with acute dengue were hospitalized longer than those with non-dengue AFI ( median 5 vs . 4 days , p<0 . 001 ) . However , disease severity overall was low . Nine adults showed signs of plasma leakage and 16 patients ( 14 adults , 2 children ) signs of hemorrhage . Among the 9 adults with plasma leakage 2 were classified as severe dengue and 6 were classified as dengue with warning signs per the WHO 2009 criteria . Among 16 patients with signs of hemorrhage , 1 was classified as severe dengue and 11 were classified as dengue with warning signs . Two adults with acute dengue ( 2 with severe dengue ) and 4 adults with other AFI ( 0 . 6% of total cohort ) required care in an intensive care unit; no patients with acute dengue died . Acute dengue was predominantly associated with secondary dengue , and those with secondary dengue were older . On bivariable analysis , both adults and children with secondary dengue were more likely to have muscle pain and thrombocytopenia and to report a longer duration of fever than those with primary dengue ( Table 3 ) . Adults with secondary dengue were also more likely to have joint pain and flushing . Children with secondary dengue were more likely to have fatigue and leukopenia and less likely to have rhinitis/congestion . On multivariable analysis , presence of joint pain ( 3 . 02 [CI 1 . 62 , 5 . 64] ) , thrombocytopenia ( OR 2 . 05 [1 . 12 , 3 . 75] ) , and absence of cough ( OR 0 . 49 [0 . 28 , 0 . 89] ) were independently associated with secondary vs primary dengue in adults whereas no variables were different in children . Both patients with primary and secondary dengue were hospitalized for a median of 5 days ( p = 0 . 53 ) . All 9 adults with plasma leakage ( 2 were classified as severe dengue and 6 were classified as dengue with warning signs per WHO 2009 WHO criteria ) had secondary dengue . Among 16 patients with hemorrhage ( one classified as severe dengue and 11clasiified as dengue with warning signs ) , 3 had primary dengue ( 2 adults , 1 child ) and 12 had secondary dengue ( 11 adults , 1 child ) . Two adults with dengue who required intensive care ( both classified as severe dengue ) had secondary dengue . No deaths were recorded . Clinical diagnosis by attending physician vs 2009 WHO classification for identifying acute dengue In the febrile cohort , the most common clinical diagnoses at admission were unspecified viral fevers ( 40 . 6% ) , dengue ( 28 . 8% ) , leptospirosis ( 12 . 3% ) , lower respiratory tract infections ( 9 . 1% ) , and upper respiratory tract infections ( 4 . 3% ) and the most common clinical diagnoses at discharge were unspecified viral fevers ( 30 . 5% ) , dengue ( 27 . 6% ) , lower respiratory tract infections ( 9 . 6% ) , leptospirosis ( 9 . 6% ) , and upper respiratory tract infections ( 4 . 7% ) . Among confirmed dengue cases , dengue was the most common clinical diagnosis both at admission ( 49% ) and at discharge ( 58% ) , followed by unspecified viral fever ( 37% at admission and 24% at discharge ) . The sensitivity and specificity of clinical diagnosis of dengue by the attending physician at discharge in adults vs children was similar and was higher for secondary vs primary dengue ( Table 4 ) . In adults , the sensitivity of clinical diagnosis at discharge for secondary dengue was 64% ( 95% CI 57–70 ) vs 49% ( 95% CI 39–59 ) for primary dengue . In children , the sensitivity of clinical diagnosis was 71 . 4% ( 95% CI 56 . 7–83 . 4 ) vs 24 . 4% ( 95% CI 12 . 4–40 . 3 ) for secondary vs primary infections , respectively . The overall accuracy of physicians’ clinical diagnosis on hospital admission was 67 . 1% ( 95% CI 64 . 2–70 . 0 ) and 75 . 2% ( 95% CI 72 . 5–77 . 9 ) at discharge . Dengue was erroneously diagnosed clinically in 15% of cases at admission and 7% at discharge . The 2009 World Health Organization ( WHO ) clinical classification had similar sensitivity for diagnosis of acute dengue in adults vs children ( 78% vs . 70% , p = 0 . 09 ) , but had improved sensitivity in secondary vs primary dengue ( 84% vs . 59% , p < . 001 ) . Overall accuracy of the WHO classification in identifying acute dengue was 71 . 5% ( 95% CI 68 . 3–74 . 6; Fig 1 ) . Of 295 laboratory-confirmed patients with dengue who met the 2009 WHO criteria for dengue , 218 ( 73 . 9% ) were identified as having dengue with warning signs , 74 ( 25 . 1% ) without warning signs , and 3 ( 1 . 0% ) severe dengue . Children with acute dengue identified by WHO criteria were no more likely to be classified as having dengue with warning signs than were adults ( 84 . 5% versus 72 . 6% , respectively , p = 0 . 06 ) . Similar proportions of patients with primary and secondary dengue were classified as having dengue with warning signs ( 75 . 4% versus 74 . 1% , respectively ) , although all 3 patients classified as having severe dengue had secondary dengue .
Early confirmation of acute dengue and identification of those at increased risk for severe dengue is desirable to decrease mortality; however , accurate point-of-care diagnostic tools are not widely available across the tropics: therefore , validated clinical instruments to improve case detection and classification are needed . To better identify patients with dengue , including severe dengue , the WHO published updated criteria for the clinical classification of dengue in 2009 . However , the sensitivity and specificity of the revised criteria is not known . The presence of epidemic dengue , reproducible enrollment criteria , and rigorous laboratory confirmation supported by excellent follow-up allowed us to retrospectively compare the 2009 WHO criteria for diagnosis of acute dengue versus clinicians’ clinical diagnosis on admission and at discharge . To our knowledge , this is the first evaluation of the performance of the 2009 WHO dengue classification criteria at identifying dengue among hospitalized children and adults with undifferentiated fever in the Southern Province , Sri Lanka . In our study , patients with acute dengue were more likely to have joint pain , muscle pain , anorexia , right upper quadrant tenderness , rash , leukopenia , thrombocytopenia , and elevated transaminases compared with those with other AFI . Leukopenia and thrombocytopenia were the independent predictors most strongly associated with dengue . Body aches and joint pain were also found to be more common in patients with dengue vs . other AFI in Puerto Rico [8] . A study in Brazil also identified history of rash but additionally taste disorder , conjunctival redness , and lymph node enlargement [9] . Multiple studies have found leukopenia , thrombocytopenia , and elevated transaminases positively associated with confirmed acute dengue [3 , 9–14] . Although cough and crackles were less common in patients with acute dengue in the present study , more than one-third of patients with dengue in our cohort reported cough . Similar results were reported in a 1986 epidemic in Puerto Rico [15] . Therefore , respiratory symptoms should not be used to rule out dengue . In our 2007 study , dengue was associated with the absence of sore throat and presence of diarrhea , conjunctivitis , jaundice , abdominal pain , leukopenia , and thrombocytopenia [16] . The small difference in symptoms in the current study compared to 2007 may reflect the infecting DENV serotype , the greater proportion of children and patients with secondary dengue , or other factors . Presence of sore throat was independently associated with acute dengue in this study but only in children , which may explain why presence of sore throat was not associated with acute dengue in our previous study focused on adults . In this study , enrollment of substantive numbers of children in addition to adults allowed us to perform multivariable analyses to identity features independently associated with acute dengue vs other AFI separately in adults vs children . On multivariable analyses , both adults and children with acute dengue were more likely than those with other AFI to have leukopenia and thrombocytopenia . Additionally , adults were more likely than those with other AFI to have joint pain , higher temperatures , right upper quadrant abdominal pain , and absence of crackles on examination whereas children with dengue were more likely than others to have sore throat , fatigue , oliguria , an elevated hematocrit ( ≥20% from baseline ) and elevated transaminases . Previous reports have suggested that clinical features of dengue may differ in adults vs children; however , most studies have only included adults with dengue [3 , 17–20] . Vomiting was associated with dengue vs other AFI in adults in Singapore and abdominal pain with dengue vs other AFI in children in Vietnam [21]; however , other studies have not found abdominal pain to differentiate dengue vs other AFI in adults or in children [22 , 23] . Further studies , with sufficient enrollment and objective enrollment ( e . g . , documented fever ) to support controlled analyses , are needed to assess whether the clinical features we found to be independently associated with dengue vs non-dengue AFI in adults and children will be again identified . We found that several features were much more frequent in secondary dengue than in primary dengue , even in the absence of severe disease . Both adults and children with secondary dengue were more likely to have longer durations of fever prior to enrollment , muscle pain , and thrombocytopenia than patients with primary dengue . Additionally , adults presenting with secondary dengue were more likely to have a lower temperature , vomiting , joint pains , and flushing on examination than adults with primary dengue whereas children with secondary dengue were more likely to have rhinitis/congestion than those with primary dengue . On multivariable analysis , presence of joint pain and thrombocytopenia and absence of cough were independently associated with secondary vs primary dengue in adults and no features were identified in children . Studies comparing secondary vs primary dengue in either adults or children are lacking [8 , 23–25] . In a cross-sectional study in Vietnam , Phuong et al found no clinical differences in a cohort of 202 patients with secondary and 32 with primary dengue [23] . However , Gregory et al found that those with secondary dengue were more likely to have body aches , joint pain , nausea , and vomiting compared with patients with primary dengue [8] . In Thailand , Pancharoen et al found that children with primary dengue were more likely to have runny nose , diarrhea , rash , and seizure and less commonly headache , vomiting , and abdominal pain [24] . Of note , we found that most ( 70% ) patients had secondary dengue , as would be expected given progressively larger island-wide , annual dengue epidemics; however , disease severity was generally low [26] . This low severity may be related to the enrollment of a cohort with undifferentiated fever , rather than biased enrollment of patients with classic signs of severe dengue/dengue hemorrhagic fever/ dengue shock syndrome . DENV-1 is generally considered to cause less severe disease than DENV-2 and -3 but disease severity is also influenced by the strain of virus within a serotype . Large population-based studies are needed to determine if features of secondary dengue also independently predict severe disease and to delineate the influence of virus serotype and strain on disease severity . We found that the sensitivity of the treating physician’s clinical diagnosis of dengue at the time of admission was improved relative to our 2007 study ( 58 vs . 14% ) , which likely reflects heightened clinical suspicion owing to our prior study , the 2009 classification criteria , and/or the epidemic transmission in 2012 . However , clinical diagnosis on admission was less sensitive than the 2009 WHO classification . In contrast , the treating physician’s clinical diagnosis of dengue at the time of discharge was more sensitive than the WHO classification , as would be expected given the opportunity for clinical observation and non-etiologic laboratory investigations . The WHO classification also had higher sensitivity for diagnosis of secondary vs . primary dengue ( 84% vs . 59% , p < . 001 , respectively ) . Wanigasuriya et al found that the 2009 classification better identified dengue with warning signs compared with the 1997 WHO classification and Jayaratne et al found that ≥ 5 warning signs predicted severe dengue [27 , 28] . Others have suggested that the 2009 criteria may overestimate disease severity [6 , 29–33] . Although the 2009 WHO classification had relatively high sensitivity in our cohort , the criteria classified almost 75% as having warning signs , even though <10% developed plasma leakage or bleeding . Notably the 25% without warning signs per classification might have not required hospital admission . Since dengue with warning signs is thought to require monitoring in hospital and many patients without warning signs hospitalized , possible overestimation of dengue disease severity has important implications for an already overburdened public healthcare system . Sri Lanka has experienced annual , island-wide large epidemics of dengue in the past decade with increasing numbers of patients being hospitalized [26] . Our study describes the clinical features and performance of the WHO classification criteria among children and adults admitted with undifferentiated fever to the largest tertiary care hospital in the Southern Province . Sri Lanka has a longstanding and effective public health system for the control of communicable diseases , with achievements that include the elimination of malaria , high coverage of the population with childhood vaccinations , and very low prevalence of human immunodeficiency virus [34] . No comprehensive studies detail the etiologies of acute febrile illness , but the annual health bulletin lists notifiable diseases such as dengue , varicella , mumps , dysentery , leptospirosis , viral hepatitides , and enteric fever as being common [34] . Our prior studies in the Southern Province have shown the relatively high prevalence of leptospirosis and rickettsial illnesses as causes of fever among hospitalized patients in the same hospital [35 , 36] . In the current study , we only enrolled patients hospitalized for AFI , so we could not evaluate features associated with hospitalization; however , 92 . 6% of those with acute dengue were hospitalized in our prior study [16] . The relatively small proportion of patients with objective indices of severe disease may be explained by the fact that we enrolled at a public tertiary care hospital where patients may have been referred and admitted for reasons other than disease severity . The frequency of hospitalization with acute dengue is consistent with Sri Lankan guidelines to hospitalize those with dengue and platelet counts ≤ 100 , 000 , and with WHO 2009 guidelines recommending hospitalization for those with dengue with warning signs . Comparisons of clinical features of dengue in children are complicated by underestimation of subjective symptoms in small children ( ascertainment bias ) ; however , the median age of children with acute dengue in this study was 11 years and we identified several features as suggestive of dengue . Our estimates of secondary dengue are conservative , since we required virologic evidence to distinguish secondary acute dengue from past dengue and 10% of patients with acute dengue had insufficient acute phase sera for IgG ELISA . Additionally , we did not perform or record tourniquet tests , since earlier studies have suggested that the utility of the tourniquet test is reduced in patients with darker skin and many studies have excluded this test in evaluations of the performance of the 2009 WHO dengue diagnostic criteria [37] . We may , therefore , have slightly underestimated the sensitivity of the 2009 WHO criteria , which includes a positive tourniquet test as suggestive of probable dengue [5 , 38] . We may also have underestimated mortality if patients died before being able to come to hospital despite free medical care . Finally , our study was conducted in a specific population of children and adults admitted with undifferentiated fever to the largest tertiary care hospital in one province of Sri Lanka . We used reproducible enrollment criteria and rigorous laboratory confirmation , but our findings may not be generalizable to other regions in Sri Lanka or to outpatient populations . Although a tertiary care center , this hospital provides both primary and tertiary care to a large portion of the population in the Southern Province ( only approximately 6% of hospitals admissions during 2012–2013 were due to transfers , per hospital data ) , thus provides a good representation of the general population in the area . In conclusion , we identified clinical features independently associated with dengue vs other AFI in adults vs children in a large cohort of patients enrolled with acute febrile illness . We found high sensitivity but low specificity of the 2009 WHO dengue classification system vs . clinical diagnosis at presentation . Strengths of our study include reproducible , unbiased , prospectively applied enrollment criteria , large sample size ( for children in addition to adults ) , and rigorous laboratory confirmation of dengue supported by 90% convalescent follow-up . This rigorous study design allowed us to evaluate the performance of the most recent ( 2009 ) WHO clinical criteria vs clinical diagnosis in those with reference standard-confirmed acute dengue , including subsets of adults vs children and those with secondary vs primary dengue . Most dengue cases during this epidemic were relatively mild with few cases of severe dengue and no deaths , despite a predominance of secondary dengue and dengue with warning signs according to the 2009 WHO classification . It is possible that different results could be seen in other geographic regions wherein particularly the non- dengue AFI group may differ . Additional prospective studies of dengue , enrolling patients with AFI using standardized protocols with reproducible enrollment criteria , will be needed to fully assess the predictive capacity of specific clinical features in adults vs children and secondary vs primary dengue , in addition to clinical outcomes and costs associated with use of the 2009 WHO classification for management of patients with acute dengue worldwide .
|
Dengue is an important cause of acute fever in the tropics that is difficult to distinguish from other common etiologies of fever . The World Health Organization ( WHO ) revised criteria for the clinical diagnosis and classification of acute dengue in 2009 . The performance of these criteria has not been widely evaluated in countries where dengue is endemic . We confirmed acute dengue in 388 of 976 patients presenting with acute febrile illness ( AFI ) to the largest tertiary care center in the Southern Province of Sri Lanka . We found specific clinical features and laboratory investigations to be predictive of acute dengue versus other AFI . The new WHO 2009 classification was more sensitive than physicians’ clinical diagnosis for identification of acute dengue on admission to hospital , but also over-estimated the severity of illness . Further large cohort studies are warranted to validate the performance of the 2009 WHO criteria for diagnosis and prognosis of dengue in regions where the disease burden is high .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"physicians",
"medical",
"doctors",
"pathology",
"and",
"laboratory",
"medicine",
"medical",
"personnel",
"tropical",
"diseases",
"geographical",
"locations",
"health",
"care",
"throat",
"health",
"care",
"providers",
"signs",
"and",
"symptoms",
"neglected",
"tropical",
"diseases",
"infectious",
"diseases",
"thrombocytopenia",
"sri",
"lanka",
"dengue",
"fever",
"pain",
"hematology",
"people",
"and",
"places",
"professions",
"diagnostic",
"medicine",
"asia",
"anatomy",
"population",
"groupings",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"myalgia",
"neck",
"abdominal",
"pain"
] |
2018
|
Evaluation of the WHO 2009 classification for diagnosis of acute dengue in a large cohort of adults and children in Sri Lanka during a dengue-1 epidemic
|
While in Northern hemisphere countries , the pandemic H1N1 virus ( H1N1pdm ) was introduced outside of the typical influenza season , Southern hemisphere countries experienced a single wave of transmission during their 2009 winter season . This provides a unique opportunity to compare the spread of a single virus in different countries and study the factors influencing its transmission . Here , we estimate and compare transmission characteristics of H1N1pdm for eight Southern hemisphere countries/states: Argentina , Australia , Bolivia , Brazil , Chile , New Zealand , South Africa and Victoria ( Australia ) . Weekly incidence of cases and age-distribution of cumulative cases were extracted from public reports of countries' surveillance systems . Estimates of the reproduction numbers , R0 , empirically derived from the country-epidemics' early exponential phase , were positively associated with the proportion of children in the populations ( p = 0 . 004 ) . To explore the role of demography in explaining differences in transmission intensity , we then fitted a dynamic age-structured model of influenza transmission to available incidence data for each country independently , and for all the countries simultaneously . Posterior median estimates of R0 ranged 1 . 2–1 . 8 for the country-specific fits , and 1 . 29–1 . 47 for the global fits . Corresponding estimates for overall attack-rate were in the range 20–50% . All model fits indicated a significant decrease in susceptibility to infection with age . These results confirm the transmissibility of the 2009 H1N1 pandemic virus was relatively low compared with past pandemics . The pattern of age-dependent susceptibility found confirms that older populations had substantial – though partial - pre-existing immunity , presumably due to exposure to heterologous influenza strains . Our analysis indicates that between-country-differences in transmission were at least partly due to differences in population demography .
In late April 2009 , the first cases of the novel swine-derived H1N1pdm influenza A virus were detected in Mexico and the United States , prompting the World Health Organization ( WHO ) to raise the level of influenza pandemic alert to phase 5 [1] . By the end of 2009 , the H1N1pdm virus had spread to more than 208 countries , resulting in hundreds of thousands of cases and at least 18000 deaths [2] , [3] . Following WHO and Centers for Disease Control and Prevention ( CDC ) recommendations , generalized media coverage and international mobilization , many countries initiated mitigation measures and enhanced surveillance of H1N1pdm virus infection in humans , providing an abundance of epidemiological data for this epidemic [3] , [4] . As a result the H1N1pdm is one of the most documented pandemics with enhanced surveillance established in many regions of the globe , with the exception of Africa [5] , [6] . The H1N1pdm virus was introduced into most northern and southern hemisphere countries during the spring and summer of 2009 . This period is outside the typical influenza season in temperate countries in the Northern hemisphere , but in the typical winter season for influenza transmission for countries from temperate regions of the Southern Hemisphere . In most Southern hemisphere temperate countries , a full epidemic of H1N1pdm influenza was observed and the pandemic strain quickly became the predominant circulating influenza virus , replacing seasonal strains in many countries [7] . Influenza transmission in a given community may depend on several factors: e . g . climatic characteristics as temperature and humidity [4] , [8] , [9] , virus intrinsic transmissibility , acquired immunity in affected populations , contact patterns in the community , collective and individual measures limiting virus spread [10] . The 2009 H1N1 pandemic was a unique opportunity for comparing the spread of a novel influenza virus in a community setting in different countries with different population structures and contact patterns . In this context , countries from temperate regions of the Southern Hemisphere , which present different demographic patterns and experienced the virus during their usual winter season , present an opportunity to evaluate the impact of these characteristics on transmission . Here we use mathematical modelling to assess the transmission characteristics of H1N1pdm virus using epidemiological data from Southern hemisphere countries in temperate regions . We address the question of the origins of the observed differences between countries by investigating the role of seasonality ( with latitude used as a proxy ) , population density and population demography ( with proportion of children used as a proxy ) . We then explore more precisely the contributions of demography in the spread of the disease by fitting different transmission models to the set of countries .
The epidemiological data analysed here were weekly case incidence of laboratory-confirmed H1N1pdm or influenza-like-illness ( ILI ) and the distribution of cumulative incidence by age-group over the study period for seven Southern hemisphere countries and one state ( Argentina , Australia -whole country and Victoria- , Bolivia , Brazil , Chile , New Zealand and South Africa ) . The data were extracted from websites or public reports issued by the countries surveillance systems . Country datasets and corresponding sources are described and listed in Table 1 . Neither daily case incidence nor age-stratified weekly case incidence data were available . Depending on the country , weekly incidence data were either laboratory confirmed H1N1pdm cases ( H1N1CC ) ( Argentina , Australia , Bolivia , Brazil , Chile , New Zealand , South Africa ) or influenza-like-illness ( ILI ) ( Australia , Chile , New Zealand , Victoria ) . All available datasets were used in the analysis , even when multiple datasets were available for a given country . Cumulative distributions of cases by age were extracted from the same data sources ( Table 1 ) . These were generally of H1N1pdm confirmed cases , except for Australia and New Zealand , where we used the age distribution of ILI cases . Due to differences between countries in the age stratification of available H1N1pdm data , country-associated age-groups were broken down into the following: Argentina ( 0–5 , 5–19 , 20–49 , 50–59 , ≥60 years old ) ; Australia ( 0–5 , 5–19 , 20–49 , 50–64 , ≥65 years old ) ; Victoria ( 0–5 , 5–19 , 20–49 , 50–64 , ≥65 years old ) ; Bolivia ( 0–5 , 5–19 , 20–44 , 45–49 , ≥50 years old ) ; Brazil ( 0–5 , 5–14 , 15–49 , 50–59 , ≥60 years old ) ; Chile ( 0–5 , 5–14 , 15–54 , 55–64 , ≥65 years old ) ; New Zealand ( 0–5 , 5–19 , 20–49 , 50–59 , ≥60 years old ) ; South Africa ( 0–5 , 5–19 , 20–49 , 50–64 , ≥65 years old ) . Demographic information was extracted from census data of the national statistics institute of the corresponding countries ( data are presented in details and electronic URL for sources are listed in Table S1 in Text S1 ) . A deterministic model was constructed to describe the spread of the virus in a population structured by age-groups . Model parameters and their values are summarized in Table 2 . Five age-groups were defined in the model ( NA = 5 ) : young children , children , young adults , adults , older adults ( with breakdowns as defined above ) . Population structure was described by the vector Ni , with Ni representing the number of individuals in age-group i . Total population size was noted NP . Individuals in the population were assumed to be either susceptible , infected or recovered ( classical SIR model ) . Each age group of the population was initialized with y0 ( a fitted parameter ) infections at the beginning of the simulation ( ten weeks before the first week of observation ) . The model incorporated heterogeneous mixing by age , with a variety of mixing patterns being explored ( more details are presented below and in section 1 of Text S1 ) . The parameter β defined the transmission coefficient . Susceptibility to infection was hypothesized to vary with age and given by the vector ρi . To avoid confounding with the parameter β , the susceptibility of young children was fixed at 1 ( ρ1 = 1 ) and the susceptibility of other groups was estimated . Therefore , for a given individual of age i , the risk of infection per contact with an infected individual is given by βρi . The generation time was assumed to be Gamma distributed [11] , [12] with mean µ = 2 . 6 days and standard deviation σ = 1 . 3 days [13] . Although some previous studies have suggested that children infected with influenza may be more infectious than adults , there was no evidence of any significant age-specific transmission risk of H1N1pdm [13] , [14] . Consequently , no age-specific infectiousness was considered in the model . We also assumed that only a proportion of infected individuals were effectively reported to the surveillance system , represented in the model by a reporting rate preport ( underreporting included here both unreported symptomatic cases and asymptomatic cases ) . No incubation period or reporting delay was considered , since so long as the generation time distribution is captured accurately , ignoring these factors does not affect transmission parameter estimates . We finally assumed that ILI surveillance data included a constant incidence of non-influenza related cases ( baseline ) , defined as BL . Technical details of the model can be found in section 1 of Text S1 . The basic reproduction number of the virus spread , R0 , was computed as the largest eigenvalue of the next generation matrix K of the model . The next generation matrix defines the next generation of new infected from a previous generation of infected [15] with element Ki , j representing the expected number of new cases from age-group i generated by one infected individual of age-group j . K was defined as:with β being the contact rate , ρ the susceptibilities and M the mixing matrix among age-groups , defined as the proportion of contacts an infected individual in age class j makes with individuals in age class i . The infection attack rate pI was defined as the proportion of individuals in the population having been infected after the epidemic ends . Parameters of the dynamic model were estimated in a likelihood-based Bayesian setting using Markov Chain Monte Carlo ( MCMC ) methods with a Metropolis Hastings sampler to explore the space of parameters . The posterior median and 95% credible interval were reported for each parameter . See Text S1 for more details . Initially , parameters were estimated for each country independently ( country-specific fits ) . In order to better understand the role of demography on H1N1pdm spread , estimation was also run for all the countries together ( global fits ) . We defined three model variants which differed in the assumption made on mixing patterns between age-groups . In the first two models , assortative mixing between age groups was assumed [16] . For a given age group , individuals had a proportion of their contacts θ occurring in their own age-group , with the remaining 1-θ fraction of contacts occurring at random in the whole population . Model variant one ( M1 ) involved a simple assortative mixing in which individuals mixed preferentially in their own age-group ( with fixed probability θ = 0 . 25 ) and randomly with all age-groups with probability ( 1-θ ) . Although higher values for assortative parameter were proposed in previous studies [16] , θ = 0 . 25 was chosen as it was consistent with mixing patterns measured in the UK via diary studies [17] . Model variant two ( M2 ) involved a more elaborate description of mixing . Three different assortativity parameters were defined: θ1 = 0 . 15 for young children; θ2 = 0 . 4 for older children; and θ3 = 0 . 14 for adults . The numerical values were estimated by fitting the mixing matrix to the mixing patterns measured in the UK [17] . For M1 and M2 , the contact rate parameter ( β ) was assumed to be common to all age-groups . Given that contact rates vary among age-groups [17] , this means that the estimates of age-dependent susceptibility obtained for these model variants also implicitly incorporate variation in contact rates as well as actual variation in susceptibility arising from pre-existing immunity . Model variant three ( M3 ) differed from M1 and M2 as it used an empirical contact matrix . The matrix was derived from the POLYMOD study data published for casual contacts in United Kingdom [17] . In order to derive appropriate matrices for each of the studied countries , two assumptions were made . First , we assumed that in a country in which a given age-group is more prevalent than in the UK , any individual will have a higher proportion of his contacts appearing in this age-group than individuals from the same age-group in the UK . Second , we assumed that contact rates varied between age-groups but were constant across countries ( see supplementary material ) . Model parameters and their values ( if these were not fitted ) are listed in table 2 . Firstly , we fitted model variant M1 to weekly case incidence data and to the cumulative age distribution of cases for each country independently , using a negative binomial likelihood with fitted variance parameter ( to allow for over-dispersion in the case data ) . For each country , nine parameters were estimated: reporting rate ( preport ) , four age-related susceptibilities ( ρi ) i = 2 . . 5 , dispersion parameter for the negative binomial likelihood , baseline for ILI incidence in the sample population ( BL ) , initial number of cases at the beginning of the simulation ( y0 ) and reproduction number ( R0 ) . Secondly , to assess the extent to which a single model could explain the patterns seen in different countries' epidemics , we fitted model variants M1 to M3 to all the countries simultaneously , keeping most parameters common to all countries . For these global fits , susceptibilities by age and contact rate were assumed to be common to all the locations ( five global parameters ) whereas reporting rate ( preport ) , ILI incidence baseline ( BL ) , and the initial number of cases ( y0 ) were fitted on a country-specific basis ( four country-specific parameters ) . Further details of the models and fitting procedures are given in the supplementary material . MCMC methods were used to obtain parameter estimates . For the country-specific fits , MCMC samples of 3×106 were generated for each country with the first 100000 iterations discarded to allow the chain to converge . For the global fits equilibration of the MCMC chains was slower , so we generated samples of 6×106 and discarded the first 2×106 of these . In order to assess which factors could influence the spread of the virus in the different countries , the R0 estimates were regressed on countries demographic age-distribution , latitude of the capital city ( except for South Africa where the biggest city was considered ) and densities of populations ( see supplementary material ) . This analysis was conducted for two different set of R0 estimates: the R0 values estimated from the exponential growth of confirmed cases in the early weeks of the epidemic in each country , using the renewal equation [11] , [12] ( supplementary material ) and the median posterior estimates from the country-specific fits . H1N1 confirmed cases were used for those countries where such data was available and ILI data was used for the one area ( Victoria ) where such data were not available .
With the exception of South Africa , the H1N1pdm epidemic started at the end of May ( epidemiological weeks [EW] 20–22 ) and finished by the end of September ( around EW 40 ) . South Africa experienced a first wave of seasonal H3N2 influenza followed by H1N1pdm influenza peaking in early August 2009 [6] ( Table 1 ) ( Figures 1 and 2 ) . Cumulative age-specific incidence is summarized in Table S1 of Text S1 , as well as demographic data and sources . Estimated empirical R0-values derived from the early exponential growth rate of the epidemic were positively correlated with the proportion of children in the population ( p = 0 . 004 ) as illustrated in figure 3a . No significant association was found with latitude and density ( supplementary material ) . Estimates of R0 , attack rate and reporting rate are summarized in Table 3 . For each country and dataset , Figure 1 compares the fits of the model ( grey lines ) with the H1N1pdm incidence data . The match to the age distribution of cases is shown in Figure 2 , and estimates of R0 for the 8 countries are plotted in Figure 3B . Estimated posterior median values of R0 ranged from 1 . 2 and 1 . 8 , with the highest values ( 1 . 5 and 1 . 8 respectively ) being obtained from for Argentina and Chile ( though for Chile , only the ILI data gave a high estimate ) . We found estimated age-related susceptibilities to vary markedly by country . With the exception of Bolivia and Brazil , a consistent pattern of decreasing susceptibility with age and higher susceptibility for children under 20 was found ( Figure 4 ) . We obtained estimated posterior median infection attack rates of between 20% and 50% of the population ( Table 3 ) . These values also varied markedly from one country to another: from 20% for Australia to 40% for Argentina and Brazil . Common and country specific parameter estimates from the fits of the global model are summarized for model variants M1-M3 in Table 4 , while fit quality to the incidence time series is illustrated in Figures 1 and 2 . Overall , the global fits reproduce temporal and age trends in the surveillance data well , albeit not as precisely as the fits of the country-specific model ( see section 6 of Text S1 for evaluation of model fitting ) . Peak incidences were slightly underestimated for Argentina , Chile-ILI and New Zealand-H1N1CC and overestimated for Australia-ILI , Victoria , Chile-H1N1CC and South Africa . Likelihood comparison did not allow one of the 3 model variants examined to be identified as superior ( section 6 of Text S1 ) . The global fits well reproduced the age distribution of cases for Argentina , Australia , Victoria and New Zealand , although the contribution of adult cases were underestimated for Bolivia and Brazil , and overestimated for South Africa and Chile ( Figure 2 ) . Resulting R0 estimates were similar for the three model variants , with still significant ( albeit much reduced compared with the country-specific model ) variation between countries: the highest values were obtained for South Africa and Bolivia and the lowest ones for New Zealand , Australia and Victoria ( Figure 3B ) . Lastly , age-dependent susceptibilities to H1N1pdm were still found to decrease with age ( Figure 4B ) . This effect was higher in model M1 and M2 suggesting that children had both higher susceptibility to the virus and higher numbers of contacts . Estimates from model M3 also suggested that resulting differences in relative susceptibilities among adult age-groups might largely be due to variation in contacts rates between these age-groups . Only two country-specific parameters were fitted for the global fits: the initial number of cases ( y0 ) and the reporting rate ( preport ) . As y0 , and preport mainly influence epidemic timing and the scaling required to match surveillance incidence data , the variation in R0 seen between countries and the qualitatively good fits obtained support the idea that demographic differences between countries may have had a substantial impact on H1N1pdm transmission .
Our results suggest transmission of H1N1pdm in 2009 varied significantly between the eight countries/states included in our analysis . Differences were found in transmissibility ( R0 median estimates ranged between 1 . 2 and 1 . 8 ) and in the size of the epidemic ( estimated median infection attack rates ranging 20–50% ) . Estimates of R0 are relatively low compared with previous estimates from past pandemics , for which values in the range 1 . 7–2 . 2 have been more typical [18]–[24] , though it should be noted that some of the higher values of R0 obtained for previous pandemics assumed a longer mean generation time than we do here . Our estimates are comparable to typical flu seasons ( R0∼1 . 3 ) [25] and consistent with other studies for H1N1pdm in 2009 obtained from other countries [26]–[30] . Our results further reinforce existing evidence that children ( <20 years old ) were substantially more susceptible to infection with H1N1pdm than adults [31]–[33] , with adults having 30–80% the susceptibility of children , depending on the model variant examined . The country-specific fits led to differences in susceptibility estimates among countries , maybe indicating that some over-specification exists in the country-specific model . However , this might also suggest that levels of prior existing immunity differ among the studied populations , which has been documented in some countries [31] , [34] , [35] , playing a role in the variation in patterns of H1N1pdm spread observed . If real , such differences in pre-existing population immunity may have contributed to the unexplained variance of the global fits relative to the country-specific fits . It should be noted that models M1 and M2 assumed simple assortative mixing by age with no age-dependent variation in contact rates , so that estimates of age-dependent susceptibility may be confounded with variation in contact rates with age . Model M3 used data from a diary survey of contact patterns [17] and thus incorporated higher contact rates in children , and the resulting estimated differences in susceptibility between adults and children were therefore lower for that model variant . In addition , in a context of high media coverage and public concern , it is possible that cases in children might have been more likely to lead to health-care seeking behaviour , affecting estimates . Nevertheless , our finding that susceptibility decreased with age is consistent with recent serological study results which demonstrated a significant proportion of immune adults prior to the start of the 2009 H1N1 epidemics [31] , [34] , [35] . Age-dependent susceptibility might arise from the effect of immune system maturation or cross-reactive immunity due to prior infections with other ( non H1N1pdm ) influenza subtypes/strains . In a completely naive population , the reproduction number would therefore be expected to be substantially larger . The lack of serological data during the pandemic prevented explicit incorporation of pre-existing immunity in the model [36] , though age-dependent susceptibility implicitly represents its effects . Sensitivity analyses in which we assumed pre-existing immunity at the beginning of the pandemic suggested including immunity would substantially affect our estimates of R0 ( given the estimates provided here are implicitly in the presence of substantial pre-existing immunity ) , but also of attack rate . Although H1N1pdm was a new virus , our results further reinforce the evidence base that there was substantial pre-existing partial cross-immunity to the virus prior to the 2009 epidemic , particularly in adults . Cross-immunity , an important feature of seasonal influenza epidemiology , was not expected to play such a key role in a pandemic situation . Clearly the experience of H1N1 in 2009 has highlighted the need for more research – both experimental and theoretical - on heterosubtypic immunity ( and perhaps non-HA mediated immunity ) . Pre-existing immunity impeded the estimation of the classic basic reproduction number ( R0 ) from the data examined here . Our R0 estimates are really estimates for R[0] , the reproduction number at the beginning of the epidemic ( at time 0 ) , rather than for the reproduction number in the absence of prior immunity . However , for ease of notation ( and because one might argue that transmission may never occur in a truly immunologically naïve population ) , we have chosen still to refer to the reproduction number of the 2009 virus at the start of each country's epidemic as R0 . Each of the three tested mixing matrices was clearly a simplification of the true mixing patterns that might be observed in the studied countries . M1 and M2 assumed a simple assortativity model ( moderate preference for mixing preferentially within one's own age group ) . The value of 0 . 25 assumed for the assortativity parameter is broadly consistent with the levels of assortativity seen in the mixing matrices provided by the UK POLYMOD survey [17] . However , in order to test whether this choice influenced the estimates , we undertook a sensitivity analysis and looked at values in the range 0–0 . 5 . This indicated that neither reproduction numbers nor susceptibility estimates were strongly affected by varyingθ . The models presented here were intentionally parsimonious . Our aim was to compare in the simplest way possible the initial epidemic of a novel influenza in different countries . The models developed here cannot generate multiple waves of transmission , and do not capture potentially important behavioural changes that may have affected transmission and disease surveillance during the pandemic [37]–[39] , such as early risk avoidance and higher rates of health-care seeking behaviour early in the pandemic . In addition we did not allow for the potential impact of school holidays and seasonal climate variation on transmission [40]–[42] , which may have improved the models fits . Lastly , only local transmission was considered here . Imported cases were not considered in the model as one would expect importations to be a substantial proportion of cases only in the first weeks before the epidemic starts and that the transmission would thereafter be predominantly local . However , by exploring multiple model variants we have demonstrated that estimates of R0 and attack rates are largely robust to uncertainty in the parameterisation of age-specific mixing patterns in the population . The differences in pandemic surveillance [43] in the countries considered may be the most influential factor affecting the reliability of our estimates and the variation found between countries . Surveillance to detect virologically confirmed cases of influenza was likely to have been highly non-systematic in several countries and variable throughout the pandemic , meaning the relationship between measured incidence and true incidence of infection may have been highly non-linear . In particular , many countries which initially undertook highly intensive case finding in 2009 moved to less intensive surveillance once case numbers grew too large for routine virological testing to be undertaken . Syndromic surveillance of ILI , by comparison , is typically more systematic but suffers from ILI being non-specific for influenza . All surveillance systems were subject to the effects of changes in health-care seeking behaviour over time . While we estimate the proportion of infections appearing in surveillance incidence data ( the reporting rate ) , we did not have the statistical power to do anything other than assume that reporting rates were constant over time . Perhaps the most interesting aspect of our results is that demographic differences between countries may have contributed strongly to the differences in observed H1N1pdm spread . In particular , we found countries with higher proportions of children ( under 20 ) had higher estimated R0 values and attack rates . Fits for the global models with shared parameters between countries are clearly poorer than the country-specific fits , but nevertheless capture much of the country to country variation . That said , fit quality for Argentina and for South Africa may indicate other factors playing a role in determining the observed patterns of transmission ( or alternatively may result from imperfections in surveillance ) . Several other factors have been demonstrated to impact the Influenza virus transmission , notably seasonal climatic variations , such as absolute humidity and temperature [8] , [44] . Although the countries examined here have substantial geographical differences between them ( e . g . capital city latitudes between 15°S and 41°S and mean population densities between 3 and 24/km2 ) , no significant association between estimated R0 and latitude or densities of populations were found ( Section 8 and Figure S8 in Supplementary material ) . More generally , our estimates of reproduction numbers did not differ strongly from those obtained from analyses of the spring/summer wave in countries from the Northern Hemisphere ( US , Mexico and UK ) [16] , [27] , [45] , suggesting a limited impact of seasonal variation in H1N1pdm transmissibility . Prior immunity could also explain differences between countries as pointed out by recent serological surveys showing that immunity to H1N1pdm varied by country of tested individuals [31] , [34] , [35] , [46]–[48] . Results presented here suggest there may be country-to-country differences in epidemiology ( driven in part by demographic variation , but not entirely so ) , suggesting some need to allow for appropriate modification of control policies on a country by country basis . In particular , targeting vaccination at children may be more optimal for countries with populations with a high proportion of school-age children . They also support the importance of developing accurate age-structured models for the analysis of influenza epidemics and the potential benefit of extending real time data collection by age-group , on serology and/or reporting rate . To conclude , this study is one of the first attempts to gain insight into the dynamics of disease transmission via inter-country comparison . Our analysis has shown that , although differences in spread of H1N1pdm were observed during the Southern hemisphere winter wave , many features of transmission were shared between countries and could be explained with largely common parameters for all countries . We showed that differences between countries could be partially explained by differences in population demography . Our results confirm that susceptibility to the virus decreased with age but also that higher contact rates in children may have partly shaped the way H1N1pdm influenza spread in 2009 .
|
Although relatively mild , the 2009 H1N1 pandemic reminded us once again of the on-going threat posed by novel respiratory viruses and the need for understanding better how such pathogens emerge and spread . From April to September 2009 , countries in temperate regions of the Southern hemisphere experienced large epidemics of H1N1pdm during their winter season , with the new virus quickly becoming the predominant circulating influenza strain . We use mathematical modelling to analyse H1N1pdm epidemiological data from 8 southern hemisphere countries . We aim at understanding better the factors which may have influenced virus transmission in these countries . We find that transmissibility of the virus was relatively low compared with previous influenza pandemics , largely because of strong pre-existing age-dependent susceptibility to the virus ( older people being less susceptible to infection , perhaps due to pre-existing immunity ) . We suggest that population demography had a strong impact on the virus spread and that higher transmission rates occurred in countries having a younger population . Our results highlight the requirement to use age-structured models for the analysis of influenza epidemics and support the need for country-specific analyses to inform the design of control policies for pandemic mitigation .
|
[
"Abstract",
"Introduction",
"Material",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"infectious",
"diseases",
"public",
"health",
"and",
"epidemiology",
"epidemiology",
"infectious",
"disease",
"epidemiology",
"global",
"health",
"infectious",
"disease",
"control",
"epidemiological",
"methods",
"infectious",
"disease",
"modeling"
] |
2011
|
Transmission Characteristics of the 2009 H1N1 Influenza Pandemic: Comparison of 8 Southern Hemisphere Countries
|
Loss of heterozygosity ( LOH ) at tumor suppressor loci is a major contributor to cancer initiation and progression . Both deletions and mitotic recombination can lead to LOH . Certain chromosomal loci known as common fragile sites are susceptible to DNA lesions under replication stress , and replication stress is prevalent in early stage tumor cells . There is extensive evidence for deletions stimulated by common fragile sites in tumors , but the role of fragile sites in stimulating mitotic recombination that causes LOH is unknown . Here , we have used the yeast model system to study the relationship between fragile site instability and mitotic recombination that results in LOH . A naturally occurring fragile site , FS2 , exists on the right arm of yeast chromosome III , and we have analyzed LOH on this chromosome . We report that the frequency of spontaneous mitotic BIR events resulting in LOH on the right arm of yeast chromosome III is higher than expected , and that replication stress by low levels of polymerase alpha increases mitotic recombination 12-fold . Using single-nucleotide polymorphisms between the two chromosome III homologs , we mapped the locations of recombination events and determined that FS2 is a strong hotspot for both mitotic reciprocal crossovers and break-induced replication events under conditions of replication stress .
Cancer cells contain a variety of genomic changes that result in altered gene expression affecting cell growth . Amplification or over-expression of oncogenes and loss of heterozygosity ( LOH ) at tumor-suppressor genes are both significant contributors to tumorogenesis . Human common fragile sites have been extensively investigated for their contribution to genomic changes that cause tumor initiation and progression . Common fragile sites are large genomic regions of 250 kb–1 Mb that are unstable under conditions that partially inhibit DNA replication ( reviewed in [1] ) . Treatment with aphidicolin , which inhibits DNA polymerases [2] , [3] , or hydroxyurea , which inhibits ribonucleotide reductase and results in unbalanced nucleotide pools [4] , both cause replication stress that induces instability at fragile sites . Several mechanisms have been proposed to explain why breaks form in human common fragile sites , including secondary structure formation within single-stranded DNA ( ssDNA ) at stalled replication forks [5] , [6] , paucity of replication origins [7] , [8] , replication fork pausing between early- and late-replicating regions [9] , [10] , and collision between RNA and DNA polymerases [11] . Multiple mechanisms may contribute to breaks , and each mechanism may be responsible for breaks at a particular site or group of sites . The mutations at common fragile sites appear to often be early drivers of tumorogenesis rather than later “passenger” events [12]–[14] . This may be because replication stress resulting from nucleotide deficiency and oncogene-induced hyper replication occurs early in the progression of cancer [15]–[17] . Research to date has focused on the ability of common fragile sites to cause deletions at tumor-suppressor genes , initiate oncogene amplification by breakage-fusion-bridge cycles , generate non-reciprocal translocations , and promote integrations of human papilloma virus ( reviewed in [18] ) . However , common fragile sites are also hotspots for sister chromatid exchange [19] , and down-regulation of Rad51 in human cells , a key protein in homologous recombination , results in increased gaps and breaks at common fragile sites [20] , which suggests the potential for fragile site lesions to also cause LOH through homologous recombination . Double-strand breaks are the canonical inducer of homologous recombination , but this repair pathway can also be stimulated by single-strand gaps and stalled replication forks , lesions that are likely to occur at fragile sites [21]–[23] . Homologous recombination in mitosis favors use of the sister chromatid as a repair template and use of non-crossover resolution pathways [24]–[27] , but inter-homolog events can occur and result in LOH from crossovers , break-induced replication ( BIR ) , and local gene conversion events [28] , [29] . Mitotic recombination events that cause LOH have been understudied , and it is unknown to what extent the replication stress present early in cancer development causes LOH by mitotic recombination , and whether fragile sites contribute to these events . In Saccharomyces cerevisiae , a fragile site named FS2 was identified on chromosome III [30] . FS2 is composed of two , 6 kb Ty1 elements in inverted orientation separated by ∼280 bp . Like human fragile sites , FS2 is a hotspot for double-strand breaks under conditions of DNA replication stress when DNA polymerases are partially impeded [30] , [31] . In cells with normal levels of polymerase , FS2 is more stable but it is a hotspot for BIR events leading to non-reciprocal translocations between Ty1 elements , indicating that the fragile site is active even in the absence of replication stress [30] , [32] , [33] . In cells with low levels of DNA polymerase , it is likely that long stretches of single-stranded DNA form at the replication fork , which we hypothesize allows the inverted Ty1 elements of FS2 to self-pair into a hairpin structure , and cleavage of this hairpin results in a DSB [30] , [34] . Here , we have used this yeast model to examine the role of fragile site instability in stimulating LOH during mitosis . In diploid cells , we determined the frequency of mitotic recombination events on chromosome III occurring spontaneously and under conditions of replication stress by low levels of polymerase alpha . Frequent single-nucleotide polymorphisms ( SNPs ) between the two chromosome III homologs were used to map the location of crossovers and BIR events . We find that chromosome III has a higher than expected level of spontaneous mitotic BIR , compared to reports for chromosomes IV , XII , and XV [35]–[37] , and that replication stress elevates mitotic recombination by 12-fold . Reciprocal crossovers and BIR events occur at approximately equal frequencies under replication stress , and fragile site FS2 is a strong hotspot for causing LOH by both of these types of events . Our analysis of gene conversions associated with crossovers indicates that lesions at FS2 during replication , and not during G1 , are the primary stimulation for these mitotic recombination events .
The naturally-occurring fragile site FS2 is located on S . cerevisiae chromosome III [30] . To evaluate mitotic recombination stimulated by this fragile site , we constructed diploids based on the detection system developed to study mitotic crossovers on yeast chromosome V [38] , [39] ( also see Text S1 ) . An event that causes loss of heterozygosity at the SUP4-o locus in a mitotic division at the time of plating results in a red/white or red/light pink sectored colony . Therefore by their nature , each sectored colony represents an independent event . The relevant features of the five diploid strains we created are shown in Figure 1 . These strains are homozygous for ade2-1 , which is an ochre stop codon null mutant allele . Cells with mutant ade2 are adenine auxotrophs and appear red due to a build-up of a red precursor in the metabolic pathway for adenine synthesis . We inserted a single copy of the tRNA ochre suppressor SUP4-o on the right arm of one homolog of chromosome III approximately 159 kb distal to the centromere . SUP4-o suppresses the ochre stop mutation; therefore the diploids are adenine prototrophs and light pink in color . The diploids are also homozygous for the GAL-POL1 construct , except for strains AMC324 and AMC331 [30] , [34] . This construct links the GAL1/10 promoter to the POL1 gene , so that the level of Pol1p in the cell is regulated by galactose in the growth medium , which allows us to induce replication stress and instability at FS2 . Under high galactose conditions ( 0 . 05% ) , the level of Pol1p is approximately 300% of wild-type levels , and under low galactose conditions ( 0 . 005% ) , it is limited to approximately 10% of wild-type levels , thereby putting the cell under replication stress [30] . Sectored colonies can result from several types of events that cause loss of heterozygosity at the SUP4-o locus in our diploids: crossover , BIR , local gene conversion , and chromosome loss . A crossover between the chromosome III centromere and SUP4-o is diagrammed in Figure 2A . As shown , one daughter cell is homozygous SUP4-o/SUP4-o , and the other daughter cell lacks this gene . Only half of crossover events are detected , due to chromosome segregation patterns . If the two recombined chromosomes segregate together in cell division , no red/white sectoring will occur . The two possible segregation patterns are equally likely in yeast [40] therefore the frequency of crossovers observed in our experiments is multiplied by two to obtain the total frequency of crossovers . Sectoring can also result from a BIR event initiated between the centromere and SUP4-o that proceeds centromere-distal from invasion , local gene conversion at SUP4-o , or loss of the chromosome containing SUP4-o ( Figures 2B , 2C , and 2D ) , although in these cases the sectoring is red/light pink . BIR initiated by a lesion in the homolog that does not contain SUP4-o results in white/light pink sectoring . This color difference is difficult to consistently detect , and therefore white/light pink sectors were not examined . A BIR event initiated on the right arm that proceeds centromere-proximal would not be detected; however , this type of event is unlikely because BIR is impeded by the centromere [41] . Loss of HygR in the red side of a sectored colony suggests chromosome loss ( Figure 2 ) , although BIR or crossover on the left arm of chromosome III can affect this phenotype . A point mutation in SUP4-o also results in red/light pink sectoring ( not shown ) . Our diploids have ∼0 . 5% sequence divergence between homologous chromosomes , as a result of mating a haploid derived from YJM789 with an S228c-related haploid [42] . This divergence in sequence does not cause a significant change in the rate of mitotic crossovers [39] . In our diploids , the S228c-related haploid is MS71 , and it contains fragile site FS2 on chromosome III [31] . The YJM789-derived chromosome III does not contain FS2; therefore , to provide homology for recombination , we inserted one Crick-orientation Ty1 element in the corresponding location on this chromosome . We used single nucleotide polymorphisms ( SNPs ) between homologs that change a restriction enzyme site to map and analyze recombination events . Because BIR events and chromosome loss are readily detectable in our system as red/light pink sectors only when the initiating lesion occurs on the homolog containing SUP4-o , we created two different experimental diploid strains ( Figure 1 ) . In Experimental Diploid #1 , both SUP4-o and FS2 are on the MS71-derived homolog of chromosome III , which allows us to evaluate the frequency of BIR and chromosome loss from initiating lesions on this homolog . To evaluate BIR and chromosome loss that result from initiating lesions on the YJM789-derived chromosome III , which does not contain FS2 , we created Experimental Diploid #2 ( strain AMC 310 ) by moving SUP4-o to the YJM789-derived homolog . Spontaneous mitotic events on chromosome III were initially evaluated in Experimental Diploid #1 ( Y332 ) grown in medium with high galactose . This diploid is homozygous for the GAL-POL1 construct and the single copy of SUP4-o is located on the same homolog of chromosome III as fragile site FS2 ( Figure 1 ) . We identified 31 sectored colonies among 30 , 543 total colonies . The event responsible for each sectored colony was determined through a combination of phenotype analysis and SNP genotyping , and frequencies for all event classes are reported in Table 1 . In Experimental Diploid #1 on high galactose , the total frequency of spontaneous mitotic events resulting in LOH on the right arm of chromosome III is 115×10−5 . We observed three categories of events: reciprocal crossovers , BIR , and chromosome loss . The spontaneous frequency of crossovers is 26×10−5 . Since the interval between CEN3 and SUP4-o is 159 kb , this is 1 . 65×10−6 crossovers per kb . The frequency of spontaneous BIR events initiated between CEN3 and SUP4-o is 46×10−5 , or 2 . 89×10−6 BIR events per kb . Because BIR is unidirectional in its transfer of genetic information , only BIR initiated by a lesion in the homolog containing SUP4-o results in red/light pink sectoring ( Figure 2 ) . Initiating lesions on the other homolog that are repaired by BIR result in white/light pink sectoring that is not easily detected . Therefore , BIR in Experimental Diploid #1 reported in Table 1 reflects only events initiated by a break on the MS71-derived homolog of chromosome III , which contains both SUP4-o and fragile site FS2 . Similarly , loss of the SUP4-o containing homolog of chromosome III is detectable by red/light pink sectoring . The frequency of spontaneous loss of the MS71-derived chromosome III is 43×10−5 in Experimental Diploid #1 . Cells with GAL-POL1 grown on high galactose contain an excess of Pol1p and have a modest increase in instability at fragile site FS2 relative to strains with POL1 under its native promoter [30] . To evaluate the effect of excess Pol1p on mitotic recombination , we created Control Diploid #1 , which is isogenic to Experimental Diploid #1 but homozygous for POL1 under its native promoter . After growth in medium with high galactose , the total frequency of spontaneous mitotic events resulting in LOH is reduced by half in this control diploid compared to Experimental Diploid #1 on high galactose ( p = 0 . 0187 ) ( Table 1 ) . However , the relative proportions of each type of mitotic event ( crossover , BIR , and chromosome loss ) are not significantly different between these two diploids ( p = 0 . 089 ) . FS2 is a hotspot for Ty1-mediated translocations under normal polymerase conditions [32] , [33] . To evaluate the effect of FS2 instability on mitotic recombination in cells with normal levels of POL1 , we modified Control Diploid #1 by replacing the entire FS2 region on the MS71-derived homolog , including both Ty1 elements and the nucleotides between them , with the NAT gene [43] . The same region on the YJM789-derived homolog was also replaced with the NAT gene . This diploid is referred to as Control Diploid #2 . We found that there is no difference in the total frequency of spontaneous mitotic LOH events on the right arm of chromosome III between Control Diploids #1 and #2 after growth in medium with high galactose ( p = 1 . 0 ) ( Table 1 ) . Partial inhibition of replication by lowering the level of DNA polymerase alpha causes breaks on yeast chromosome III at fragile site FS2 . In haploid cells these breaks are frequently repaired by BIR or result in loss of chromosome III and can be detected by increased illegitimate mating [30] , [34] . In diploid cells , low polymerase alpha increases mitotic reciprocal crossovers within the yeast rDNA array by 7-fold [44] . Here , we have further evaluated the role of replication stress in stimulating events that cause LOH in diploid cells , and the role of fragile site instability in initiating these events . To study stress-induced mitotic events on yeast chromosome III , we grew Experimental Diploid #1 in medium with no galactose for six hours to lower the level of polymerase alpha , followed by plating on high galactose . We identified 140 sectored colonies among 22 , 640 total colonies . Replication stress in this diploid increases the total frequency of mitotic LOH events by 6 . 5-fold relative to high galactose conditions ( p<0 . 001 ) ( Table 1 ) . However , the relative proportions of the categories of crossover , BIR , and chromosome loss in this diploid are the same in both high galactose and no galactose ( p = 0 . 463 ) . As explained above , BIR events and chromosome loss are readily detectable in our system only when the initiating lesion occurs on the homolog containing SUP4-o . In Experimental Diploid #1 , both FS2 and SUP4-o are on the MS71-derived homolog . We used Experimental Diploid #2 , which has SUP4-o on the non-FS2 containing YJM789-derived homolog , to evaluate the frequency of BIR and chromosome loss from initiating lesions on this homolog of chromosome III . We grew this diploid in medium lacking galactose for six hours to induce replication stress then plated cells on high galactose . We identified 47 sectored colonies among 14 , 876 total colonies . Under replication stress , the total frequency of mitotic LOH on chromosome III in Experimental Diploid #2 is half that of Experimental Diploid #1 ( p<0 . 001 ) ( Table 1 ) . The frequency of crossovers is similar in Experimental Diploids #1 and #2 under replication stress ( p = 0 . 543 ) ( Table 1 ) . This result is consistent with our expectation , given that crossovers are detected in our system irrelevant of which chromosome III homolog contains the initiating lesion . The frequency of replication stress-induced BIR is one-third lower in Experimental Diploid #2 than in Experimental Diploid #1 ( p = 0 . 0629 ) . This difference results from the absence of FS2 in the SUP4-o marker homolog . In Experimental Diploid #2 , only BIR events that are initiated by lesions on the YJM789-derived chromosome can be detected as red/light pink sectors . Events caused by a lesion at FS2 on the MS71-derived homolog of chromosome III will result in white/light pink sectoring , which is not easily detected and thus not scored in this diploid . The lower frequency of BIR in Experimental Diploid #2 suggests that FS2 instability drives 1/3 of the stress-induced BIR observed in Experimental Diploid #1 . There is an even stronger reduction in the frequency of chromosome III loss in Experimental Diploid #2 under replication stress , such that the loss frequency is below that observed in Experimental Diploid #1 with high levels of polymerase . Therefore , the primary cause of chromosome III loss in Experimental Diploid #1 is FS2 instability . To further evaluate the role of FS2 instability in driving mitotic LOH under replication stress , we created Control Diploid #3 ( Y382 ) , in which we stabilized fragile site FS2 . Normally , the inverted Ty1 elements at FS2 are separated by ∼280 bp . We inserted the NAT gene [43] between these two Ty1 elements , separating them by ∼1 . 8 kb . The increased distance effectively stabilizes the fragile site [30] . We grew Control Diploid #3 in medium with no galactose for six hours to induce replication stress conditions , and then plated cells on high galactose medium . We identified 7 sectored colonies among 21 , 666 total colonies , and analyzed these sectored colonies as before . Under replication stress , the total frequency of mitotic LOH on chromosome III in Control Diploid #3 is less than half that of Experimental Diploid #1 ( p<0 . 0001 ) ( Table 1 ) . However , the relative proportions of the categories of crossover , BIR , and chromosome loss events in these diploids is similar ( p = 0 . 1106 ) . We used the ∼0 . 5% sequence divergence between the two homologs of chromosome III in our diploids to map the locations of the events causing sectoring . This divergence results in many single nucleotide polymorphisms ( SNPs ) between the homologs , some of which alter restriction enzyme sites . We purified a single cell from each half of the sectored colony and evaluated a set of 27 SNPs on the right arm of chromosome III by PCR and restriction enzyme digest ( Table S3 ) . On average , these SNPs are spaced 6 . 9 kb apart . The closest SNP centromere-distal to FS2 that changes a restriction enzyme site is at chromosome III base 175324; this is 5 . 4 kb from the end of the fragile site . The closest SNP centromere-proximal to FS2 is at base 167720; this is 0 . 8 kb from the end of the fragile site . As shown in Figure 3A , in the case of BIR , all SNPs in the light pink cell remain heterozygous while in the red cell , SNPs proximal to the event remain heterozygous and SNPs distal to the event are homozygous for the homolog lacking SUP4-o . In the case of BIR events initiated by a lesion in unique sequence , we assume that invasion of the broken end into the corresponding region on the homologous chromosome results in homozygosity for SNPs distal to the BIR . However , if the broken end occurs near a Ty1 element , it may invade a Ty1 or Ty2 element on a non-homologous chromosome to initiate replication . In such cases of non-allelic repair , SNPs distal to the BIR would be hemizygous . In Experimental Diploid #1 we detect only BIR initiated by lesions on the MS71-derived homolog of chromosome III , which contains both SUP4-o and FS2 . As anticipated , only YJM789-derived SNPs are present in the red cell distal to each BIR event . Our SNP mapping indicates that fragile site FS2 is a hotspot for initiation of BIR in this diploid ( Figure 4A ) . Of 66 total BIR events under replication stress , 18 were initiated between the SNPs flanking FS2 . BIR can be initiated centromere-proximal to a break location due to exonuclease processing at the break that usually exposes 3–6 kb of ssDNA [45]; therefore , the nine BIR events initiated between the pair of SNPs immediately centromere-proximal to FS2 likely also result from lesions at FS2 , for a total of 27/66 events ( 41% ) stimulated by FS2 . We evaluated the significance of this distribution by dividing the CEN3 – SUP4-o interval into four equal-sized bins of 39 . 7 kb , then counting the number of BIR events initiated within each bin . There are 16 events in bin #1 ( CEN3 to SNP152 ) , 43 in bin #2 ( SNPs 164 to193 ) , 5 in bin #3 ( SNPs 195 to 233 ) , and 2 in bin #4 ( SNPs 246 to SUP4-o ) . This distribution is significantly different from random ( p<0 . 0001 by chi-square goodness-of-fit ) . To determine whether BIR events were allelic or non-allelic , we determined the sizes of chromosome III in a subset of 35 BIR events from Experimental Diploid #1 under replication stress . Allelic events will produce chromosome III repair products of normal size , and non-allelic events will produce a chromosome that may be smaller or larger than the normal chromosome III size . Intact yeast chromosomes from each event were separated by pulsed-field gel electrophoresis , and chromosome size was evaluated by Southern blotting with a CHA1 probe to the right arm of chromosome III . Of events tested that were initiated between the SNPs flanking FS2 or within 6 kb proximal of FS2 , 7/18 ( 39% ) had non-allelic BIR products ( Figure S1 ) . All of the BIR events tested that were initiated more than 6 kb proximal of a Ty1 element were allelic . In high galactose conditions that permit high levels of POL1 transcription , the number of BIR events initiated in Experimental Diploid #1 at or within 6 kb proximal to FS2 is reduced to 4/14 ( 29% ) ( Figure 4B ) . This fact that this reduction is relatively modest is likely attributable to excessive polymerase alpha causing FS2 instability in this diploid , because Control Diploid #1 , which has POL1 under its native promoter , and Control Diploid #3 , which has a stabilized version of FS2 , do not have any BIR events initiated at or within 6 kb proximal of FS2 . We note that the pair of tandem-oriented Ty1 elements centromere-proximal to FS2 on the MS71-derived homolog is not a BIR hotspot in Experimental Diploid #1 ( Figure 4A ) , although this was a frequent site of recombination in the illegitimate mating assays previously used to study fragile sites on yeast chromosome III [30] , [34] . This difference will be further discussed below . In Experimental Diploid #2 , 10/28 BIR events ( 36% ) were initiated at or within 6 kb proximal of the location allelic to FS2 ( Figure 4C ) . In this diploid , only events initiated by a lesion on the YJM789-derived homolog are detected . As before , we evaluated the significance of this distribution by dividing the CEN3 – SUP4-o interval into four equal-sized bins of 39 . 7 kb , then counting the number of BIR events initiated within each bin . There are 7 events in bin #1 , 15 in bin #2 , 4 in bin #3 , and 2 in bin #4 . This distribution is significantly different from random ( p = 0 . 0029 by chi-square goodness-of-fit ) . Therefore , despite the fact that FS2 is not present on the YJM789-derived homolog , the site allelic to this fragile site is a hotspot for initiation of BIR events . The YJM789 homolog of chromosome III contains a pair of inverted delta elements ( the ∼300 bp long terminal repeat portion of Ty1 elements ) at the location allelic to FS2 . The spacing between these inverted deltas is the same as between the Ty1 elements of FS2 . As explained above , we modified the YJM789 homolog to expand the Crick-orientation delta element to a full Ty1 , to provide homology for recombination without creating a fragile site . However , the inverted delta elements also have the potential for intra-strand base pairing to form a hairpin under conditions of replication stress . Since the overall frequency of stress-induced BIR is lower in Experimental Diploid #2 than in Experimental Diploid #1 , the frequency of BIR stimulated by the “full” version of FS2 is higher than that stimulated by the “delta only” version of FS2 ( frequencies of 123×10−5 and 67×10−5 FS2-stimulated BIR , respectively ) . There were three BIR events in Experimental Diploid #2 that had adjacent gene conversion tracts; two with a 4∶0 tract ( SC100 and SC104 ) and one with a 3∶1 tract ( SC121 ) ( Figure S2 ) . Gene conversion associated with BIR has previously been reported , and appeared to result from repair of two double-strand breaks in the same location on both sister chromatids , [35] , [36] . The 3∶1 tract observed here does not fit that mechanism , and may instead represent repair of heteroduplex mis-matches in the region of invasion for BIR initiation . The 4∶0 tracts are unusual and may represent an internal deletion prior to BIR initiation . Figure 3B shows an example of the SNP pattern in a sectored colony from a reciprocal crossover on the right arm of chromosome III . For crossovers un-associated with gene conversion , SNPs proximal to the crossover remain heterozygous , and distal to the crossover , are homozygous for the homolog lacking SUP4-o in the red cell , and homozygous for the homolog containing SUP4-o in the white cell . Gene conversion that is associated with a crossover can be of two types , either a typical 3∶1 segregation in which SNPs are heterozygous in one cell and homozygous in the other ( as shown in Figure 3B ) , or a 4∶0 pattern in which SNPs are homozygous for the same version in both the red and white cells [39] . The 3∶1 conversions appear to result from repair of damage that occurs during S-phase and 4∶0 conversions result from DNA double-strand breaks that occur during G1 that are replicated , followed by repair of both broken sister chromatids in G2 using the unbroken homolog as a template [46] . As shown in Figure 5A , our SNP mapping results indicate that fragile site FS2 is a hotspot for crossover events under replication stress caused by low Pol1p . We identified 41 crossover events in Experimental Diploid #1 under stress ( Figure 5B ) . These crossover events were collected in two ways; 29 crossover events were collected among the 22 , 640 colonies in Table 1 that were fully analyzed for crossover , BIR , and chromosome loss events , and 12 crossover events were collected among another set of 14 , 792 colonies that was not fully analyzed for BIR and chromosome loss events . Of the 41 crossovers in Experimental Diploid #1 , 21 have no associated gene conversion tract , 19 have a gene conversion adjacent to the crossover , and one has a conversion tract that is not contiguous with the crossover . Of the 21 crossovers without gene conversion , 8 occur between the SNPs flanking FS2 . Of 19 crossovers with adjacent gene conversion tracts , 12 tracts cross a SNP flanking FS2 . Therefore , 20/41 crossover events ( 49% ) in Experimental Diploid #1 under replication stress are associated with FS2 . The crossover data from Experimental Diploid #2 under replication stress is similar to Experimental Diploid #1 , which is consistent with the expectation that our system detects all crossovers between CEN3 and SUP4-o irrespective of which homolog contains SUP4-o or which homolog has the initiating lesion . In Experimental Diploid #2 , 8/15 events ( 53% ) are associated with FS2 ( Figure 5B ) . Of 15 total crossover events , 11 are unassociated with gene conversion , and 6 of these occur between the SNPs flanking FS2 . Of the 4 crossovers with adjacent gene conversion , two have tracts that cross a SNP flanking FS2; these two tracts have information transferred from the non-FS2 containing homolog indicating the initiating event was at or near the fragile site ( Figure 5B ) . In Experimental Diploid #1 in high galactose conditions , five crossovers were collected . These crossover events were collected in two ways; 4 crossover events were collected among the 30 , 543 colonies in Table 1 that were fully analyzed for crossover , BIR , and chromosome loss events , and 1 crossover event was collected among another set of 4 , 792 colonies that was not fully analyzed for BIR and chromosome loss events . Of these five crossovers , one is located at FS2 ( Figure 5D ) . This crossover is associated with a gene conversion tract in which the transfer of genetic information indicates that the initiating event is on homolog with the “delta-only” FS2 . In Control Diploid #1 , which has POL1 under its native promoter , two of the five crossovers are located between the nearest SNPs flanking FS2 ( Figure 6 ) . Although the number of events detected under high galactose and in Control #1 is low , it is intriguing that 20–40% of crossovers in these diploids were at FS2 . We address this result in the discussion below . In Control Diploid #2 , which has POL1 under its native promoter and no FS2 , the single crossover detected was not near the deleted fragile site , and in Control Diploid #3 , which is GAL-POL1 and has a stabilized version of FS2 , one of the three crossovers was at FS2 ( Figure 6 ) . Several characteristics of the crossover-associated gene conversion tracts in Experimental Diploids #1 and #2 under replication stress are of interest . First , 12 of the 14 tracts crossing a SNP at FS2 have three copies of the information from the chromosome lacking FS2 . This result is consistent with damage at FS2 responsible for crossover stimulation , since in both mitotic and meiotic recombination events , the damaged chromosome typically receives genetic information from the unbroken homolog [47]–[49] . However , the two tracts that were stimulated by an initiating lesion on the YJM789-derived homolog are consistent with our BIR results above , in which a “delta-only” version of FS2 is capable of stimulating a lower level of recombination than the “full” version of FS2 . Second , of 24 total tracts , only two are 4∶0 type tracts and the others are 3∶1 . The 3∶1 conversions have been reported to result from repair of S-phase damage and the 4∶0 conversions result from DNA double-strand breaks in G1 that are replicated , followed by repair of both broken sister chromatids during G2 [46] . Therefore , our results indicate that the crossover-associated gene conversion tracts under replication stress are consistent with damage occurring primarily during S phase . Third , six of the gene conversion tracts associated with FS2 cross both SNPs flanking FS2 , four cross only SNPs centromere-proximal , and four cross only SNPs centromere-distal . Therefore , repair of a lesion at FS2 that occurs during S-phase can result in gene conversion that extends either bi- or uni-directionally . Fourth , our mitotic gene conversion tracts are relatively long , with a median length of 14 . 7 kb ( 95% confidence interval of 7 . 0 kb to 34 . 5 kb ) for the 23 tracts contiguous with a crossover . Both 4∶0 and 3∶1 tracts were included in our analysis of median tract length .
St Charles and Petes [28] defined the microStern ( µS ) as a unit to measure mitotic crossovers , with 10−6 crossovers/division equal to one microStern , and they estimated the entire yeast genome has a mitotic genetic map length of 620 µS . The portion of chromosome III we evaluated accounts for 1 . 3% of the physical yeast genome , therefore we expect a genetic map length of 8 µS . We detect a map length of 6 µS for the right arm of chromosome III in Control Diploid #2 , which has normal Pol1p levels and does not contain FS2 . In Control Diploid #1 , which has normal Pol1p but contains FS2 , we detect a map length of 310 µS , and two of the five crossover events are at FS2 . These data are in accordance with reports that FS2 can be unstable under normal polymerase conditions [32] , [33] . However , there is no difference in the total frequency of spontaneous mitotic LOH events between these two diploids ( p = 1 . 0 ) ( Table 1 ) . Previous studies of mitotic LOH in yeast have reported that BIR is less frequent than crossovers . On yeast chromosomes IV and XII , spontaneous BIR is three to four-fold less frequent than crossovers [35] , [37] , and on chromosome XV , repair by BIR of a mitotic double-strand break from an I-SceI cut site is extremely rare compared to repair that results in crossover or non-crossover outcomes [36] . The exception to this pattern is in old yeast mother cells , in which nearly 90% of spontaneous mitotic LOH results from BIR [37] . We observed that BIR is nearly 5-fold more frequent than crossovers in Control Diploid #2 , and that BIR is only 20% less frequent than crossovers in Control Diploid #1 . None of the BIR events in Control Diploid #1 were initiated at or near FS2 , which indicates that a mechanism other than fragile site instability drives spontaneous BIR on yeast chromosome III in this strain . The BIR pathway is primarily used to repair one-ended double-strand breaks , such as those that exist at collapsed replication forks [50] . Therefore , our results may suggest a higher frequency of spontaneous replication fork stalling and collapse on the right arm of chromosome III than on other chromosomes similarly examined to date . Here , we report that the total frequency of mitotic LOH is elevated 12-fold in Experimental Diploid #1 with low levels of polymerase , relative to Control Diploid #1 with wild-type levels of polymerase ( Table 1 ) . In our analysis , replication stress induces reciprocal crossovers , BIR , and chromosome loss with approximately equal frequency . In haploids with low levels of polymerase alpha , physical analysis of chromosome III indicates that a double-strand break at FS2 occurs in approximately 7% of cells [34] . If a similar percentage of diploid cells with low polymerase alpha have breaks at FS2 , then our results indicate that LOH is a rare outcome in responding to these breaks . This is not unexpected , because LOH as a result of mitotic recombination requires crossover and BIR events involving the homologous chromosome . However , during mitosis the sister chromatid is favored as a repair template during S-phase [27] , [51] , [52] , and crossover resolution of Holliday junctions is normally suppressed [53] . A related issue is the detection of gene conversion events at FS2 that are un-associated with crossover . Our system does not permit analysis of such events unless those gene conversions are large enough to also encompass SUP4-o . Recent studies have demonstrated that approximately 35% of conversions are crossover-associated [35] , [36] , [54] . Therefore , we would not expect that undetected local gene conversion events at FS2 would change the relative rarity of LOH at this site compared to the frequency of breaks . Here , we report that FS2 is a hotspot for driving mitotic events that result in LOH on the right arm of yeast chromosome III . Unexpectedly , a smaller inverted repeat consisting of two long terminal repeat delta elements separated by the same ∼280 bp distance as between the two full Ty1 elements of FS2 , is similarly a hotspot for mitotic recombination under replication stress . However , the delta-only FS2 stimulates only half as many BIR events as the full FS2 . There are no other inverted delta-delta pairs on the right arm of chromosome III to investigate for fragile site activity . However , inverted pairs of delta elements have been reported to fuse and generating acentric and dicentric chromosomes in yeast when replication is impeded; faulty template switching at stalled replication forks was proposed as a mechanism to generate these rearrangements [55] . Human Alu sequences , which are similar in length to the yeast delta element , stimulate breakage and recombination when inserted in inverted orientation on yeast chromosome II , although this is strongly influenced by the distance between the repeats [56] . It is somewhat surprising that crossovers and BIR events are stimulated approximately equally at FS2 . Under replication stress , it is hypothesized that extended single-stranded DNA at the replication fork allows a hairpin to form between the pair of inverted Ty1 elements of FS2 [30] , [34] , and cleavage at this secondary structure would result in replication fork collapse to a one-ended double-strand break , which should primarily drive BIR ( Figure 7 ) [50] . Crossover formation requires a double Holliday junction intermediate . The stimulation of crossovers at FS2 is primarily replication-dependent , because the gene conversion tracts adjacent to crossovers are nearly all of the 3∶1 type . Two possible ways that a double Holliday junction intermediate could form at FS2 are ( 1 ) template switching at a stalled replication fork or single-strand gap left at FS2 during replication , or ( 2 ) convergence of a collapsed fork with replication from a nearby origin , producing a canonical double-strand break ( Figure 7 ) [22] , [23] . Although physical analysis of chromosome III demonstrates that double-strand breaks do form at FS2 under replication stress [34] , it is unclear whether breaks are the initiating lesion for crossovers at this fragile site , since template switching during replication can generate a double Holliday junction in the absence of a break . Mitotic crossovers are rare in cells with wild-type levels of polymerase , but of the five events we collected in Control Diploid #1 , two were at FS2 and did not have an adjacent 3∶1 gene conversion tract . The two inverted Ty1 , Ty2 pairs on chromosome IV have been reported as hotspots for spontaneous crossovers [28] . Crossover events at these are usually associated with 4∶0 gene conversion , indicating an initiating lesion in G1 . Although the number of spontaneous crossover events we collected is too low for a conclusive comparison with this data on inverted Ty1 , Ty2 pairs from chromosome IV , FS2 likely behaves as a similar hotspot for G1-lesion stimulated crossover events in un-stressed cells . In cells with low levels of polymerase , over 90% of gene conversion tracts associated with a crossover on the right arm of chromosome III are of the 3∶1 pattern , indicating an initiating lesion during S-phase . Analysis of crossovers induced by low alpha DNA polymerase on yeast chromosomes IV and V indicates that these also are associated primarily with 3∶1 rather than 4∶0 conversion tracts ( W . Song and T . D . Petes , personal communication ) . These results are consistent with stalled or collapsed forks under replication stress stimulating crossover formation . In studies of un-stressed cells , crossover-associated gene conversion tracts on yeast chromosomes IV and V are either a mixture of 3∶1 and 4∶0 , or are primarily 4∶0 [35] , [39] . Our median gene conversion tract length in cells under replication stress is 14 . 7 kb , which is much longer than the 1–4 kb tracts observed during meiosis [57]–[59] . Other studies of mitotic crossovers in yeast have highlighted similarly extensive gene conversion with median tract lengths of 4 . 7 to 20 . 3 kb [28] , [29] , [35] , [39] , [46] . The density of SNP markers evaluated affects our ability to evaluate how often crossovers have adjacent gene conversion . In a previous report on chromosome IV where a high density of SNPs was used , 87% of crossovers had an adjacent gene conversion [28] . In our cells under replication stress , which were evaluated using fewer SNPs , only 41% of crossovers had an adjacent gene conversion . As discussed , we find that spontaneous mitotic BIR events on the right arm of chromosome III are more frequent than expected , compared to other yeast chromosomes . Under replication stress by low levels of polymerase alpha , the frequency of mitotic LOH is elevated approximately 12-fold , resulting from crossover , BIR , and chromosome loss . Fragile site FS2 is a hotspot for initiating LOH events under replication stress , and S-phase lesions at this site stimulate crossovers and BIR events approximately equally . More than one-third of the BIR events initiated at or near FS2 are non-allelic , resulting in gross chromosomal alteration of chromosome III . These results have important implications for adding to the mechanisms in which human common fragile sites promote tumorogenesis . Human common fragile sites , like yeast FS2 , are unstable under conditions of replication stress [60] and replication fork stalling at sequences with secondary-structure forming potential has been observed in some fragile sites [6] , [7] . The contribution of deletions , amplifications , and translocations at common fragile sites to tumor development and progression has been extensively documented [18] , [61] , [62] . However , LOH at tumor suppressor genes has long been known as a driver of tumorogenesis [63] , and this mechanism has not been well studied at fragile sites . Based on our results , further research is warranted to determine the role of common fragile sites in stimulating LOH in tumors through BIR and reciprocal mitotic crossovers .
The five diploid strains used for analysis of mitotic recombination were Y332 , Y382 , AMC310 , AMC324 , and AMC331 ( Figure 1 ) . Each of these diploids was created by mating an MS71-derived haploid cell [64] with a YJM789-derived haploid cell [42] , resulting in ∼0 . 5% sequence divergence between homologous chromosomes [42] . Each diploid is homozygous for the ade2-1 mutation and contains one copy of SUP4-o . Strains Y332 , Y382 , AMC310 , and AMC324 contain one copy of fragile site FS2; strain AMC331 does not contain any Ty1 elements at the location of FS2 on either chromosome III homolog . Strains Y332 , Y382 , and AMC310 are homozygous for the GAL-POL1 construct [30]; strains AMC324 and AMC331 are homozygous for POL1 driven under its native promoter . The configuration of genes and markers in each diploid strain is diagrammed in Figure 1 . The steps in construction of these strains are described in the Text S1 , and construction details and genotypes for all strains are in Tables S1 and S2 . All transformations and matings were done using standard protocols . All five diploid strains , whether they contained GAL-POL1 or not , were maintained at 30°C in standard rich media [65] , with the exception that the medium contained 3% raffinose instead of dextrose . Raffinose was used as a carbon source because it does not suppress the GAL1/10 promoter , allowing us to control expression of the GAL-POL1 construct by varying the amount of galactose in the medium . All diploid strains were purified to individual colonies , and were inoculated for growth overnight in standard rich media containing high galactose ( 0 . 05% ) . Three or four cultures were inoculated for each diploid in each condition . Cells were then washed and diluted 1∶5 in fresh rich media liquid culture with no galactose for 6 hours ( to induce replication stress by lowering the level of polymerase alpha ) , or were diluted 1∶5 in rich media liquid culture with high galactose for 6 hours . The galactose treatment for each strain is indicated in Table 1 . The density of each culture was determined , and cells were spread at low density to form colonies ( ∼350 colonies per plate ) on plates containing high galactose and 10 µg/ml adenine ( two-fold less than standard omission medium ) . Twenty to forty plates were spread from each culture , to obtain 7 , 000 to 14 , 000 colonies per culture . Cells were allowed to grow at 30°C for 3 days and then plates were incubated overnight at 4°C to intensify red color development in the colonies . The total number of colonies was counted for each culture , and culture counts from each diploid were totaled . If a crossover , BIR event , local gene conversion at SUP4-o , or chromosome loss event occurs in the first or second division at the time the diploid is plated , a sectored colony is produced . Therefore , each sectored colony is an independent event . All red/white and red/light pink sectored colonies in which the red portion was at least one-fourth of the colony were identified , and a single cell from each half of the sector was purified for subsequent analysis of the mitotic event that resulted in sectoring . The frequency of BIR events and of chromosome loss events reported in Table 1 was calculated as [number of sectored colonies of the event type/total colonies] . The frequency of crossovers reported in Table 1 was calculated for each strain as [2*number of crossover sectored colonies]/[total number of colonies] . 95% confidence intervals for the proportion [66] of each mitotic event were calculated using VassarStats ( http://vassarstats . net/ ) . Chi-square contingency tables were used to compare the frequencies of mitotic events between strains . Phenotype analysis was initially used to classify sectored colonies . Sectors from Y332 , Y382 , AMC324 , or AMC331 with phenotype His+ HygS ( red cell ) and His+ HygR ( light pink cell ) usually result from chromosome loss . Sectors from AMC310 of phenotype His− HygR ( red cell ) and His+ HygR ( light pink cell ) are usually chromosome loss . Sectors from all strains in which cells from both sides of the sectored colony remain His+ HygR are crossover , BIR , local gene conversion at the SUP4-o locus , or mutation at the SUP4-o locus . There is ∼0 . 5% sequence divergence between the two homologs of chromosome III in the experimental diploids ( Wei et al . 2007 ) , and several of the single nucleotide polymorphisms ( SNPs ) alter restriction enzyme sites . For example , a SNP on chromosome III at base 266045 results in an HpyCH4III site on the YJM789-derived chromosome but not the MS71-derived chromosome . This region was amplified by PCR , generating a 374 bp product ( Table S3 ) . If the site is heterozygous in the cell being examined , digestion of the amplified product with HpyCH4III followed by gel electrophoresis reveals three band sizes: the uncut 374 bp product and the cut 259 bp and 115 bp products . Genotype analysis of the SNP at chromosome III base 266045 , which is 7 kb centromere-proximal of SUP4-o , was used for initial evaluation of event type in all sectored colonies . In chromosome loss and BIR events , the red cell of a sector has only the form of the SNP from the copy of chromosome III lacking SUP4-o , and the light pink cell is heterozygous at this site . In crossover events , the red cell of a sector has only the form of the SNP from the copy of chromosome III lacking SUP4-o , and the white cell has only the form of the SNP from the copy of chromosome III containing SUP4-o . Sectors with red cells that remained homozygous at SNP 266045 may result from a point mutation in SUP4-o or a small gene conversion tract surrounding SUP4-o; these were not further analyzed . All sectored colonies with a change of zygosity at SNP 266405 were further analyzed . Genomic DNA was harvested from purified cells from each side of sectored colonies and subjected to polymorphism analysis . For all sectored colonies , an initial set of 8 SNPs were tested for zygosity to reconfirm the event type . BIR and crossover events were then tested with additional SNPs to further refine the location of the event . In total , we used 25 SNPs in the 159 kb interval between CEN3 and SUP4-o on chromosome III , plus 2 additional SNPs centromere-distal from SUP4-o . Polymorphic sites , primers , and diagnostic restriction enzymes are listed in Table S3 . Gene conversion tract lengths were calculated as described in Lee et al . ( 2009 ) with the modification that the size of Ty1 elements present in the MS71-derived homolog that are not present in the sequenced genome are accounted for in our distance calculations when this homolog is used as the template for repair . Genomic DNA from 1×108 cells was harvested in agarose blocks to prevent shearing as described in [67] . Chromosomes were separated by PFGE in a 1 . 2% gel in 0 . 5× TBE at 14°C using a Gene Navigator system ( Pharmacia Biotech ) . Switch times at 6 V/cm were as follows: 50 sec switch for 4 . 5 hr , 90 sec switch for 5 . 5 hr , 105 sec switch for 7 . 5 hr , 124 sec switch for 7 . 5 hr , 170 sec switch for 7 . 5 hr . DNA was transferred to Hybond N+ membrane ( GE Healthcare Life Sciences ) by a neutral transfer according to standard protocol , then probed with CHA1 , a gene located on the left arm of chromosome III . The CHA1 probe was made by PCR , using primers 5′ CTGGAAATATGAAATTGTCAGCGAC and 5′ TGAATGCCTTCAACCAAGTGGCCCTTTC . Probes were radioactively labeled by random-prime labeling using Ready To Go beads ( -dCTP ) ( GE Healthcare Life Sciences ) . Southern blot hybridization and washes were standard . Membranes were exposed to a phosphor screen and images were captured with an FLA-3000 scanner ( Fujifilm ) . There are two normal sizes for chromosome III in our diploids; the YJM789-derived homolog of this chromosome is ∼18 kb smaller than the MS71-derived homolog , because the YJM789 homolog has only one Ty1 element and the MS71 homolog has four Ty1 elements on the right arm of chromosome III . Diploids with two normal-size copies of chromosome III were considered allelic BIR events . Diploids with one normal-size III and one chromosome III of abnormal size were considered non-allelic BIR events .
|
Loss of heterozygosity ( LOH ) at tumor-suppressor genes contributes to cancer , and deletions resulting in LOH are frequently observed in tumor cells at certain chromosomal regions known as common fragile sites . LOH can also result from repair of DNA damage by mitotic recombination , if the homologous chromosome rather than the sister chromatid is used as a repair template . The extent to which fragile site instability causes LOH by mitotic recombination with the homologous chromosome is unknown . We evaluated mitotic recombination on the yeast Saccharomyces cerevisiae chromosome III , which contains a naturally-occurring fragile site known as FS2 . We report that yeast chromosome III has a high frequency of spontaneous mitotic recombination that involves the homologous chromosome . Under conditions that stimulate instability at the fragile site , LOH resulting from mitotic recombination on yeast chromosome III is increased 12-fold , and FS2 is a hotspot for initiating these events . These results suggest that instability at human common fragile sites may drive mitotic recombination repair pathways that cause LOH and promote tumorogenesis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
Fragile Site Instability in Saccharomyces cerevisiae Causes Loss of Heterozygosity by Mitotic Crossovers and Break-Induced Replication
|
The history of Chagas disease control in Peru and many other nations is marked by scattered and poorly documented vector control campaigns . The complexities of human migration and sporadic control campaigns complicate evaluation of the burden of Chagas disease and dynamics of Trypanosoma cruzi transmission . We conducted a cross-sectional serological and entomological study to evaluate temporal and spatial patterns of T . cruzi transmission in a peri-rural region of La Joya , Peru . We use a multivariate catalytic model and Bayesian methods to estimate incidence of infection over time and thereby elucidate the complex history of transmission in the area . Of 1 , 333 study participants , 101 ( 7 . 6%; 95% CI: 6 . 2–9 . 0% ) were confirmed T . cruzi seropositive . Spatial clustering of parasitic infection was found in vector insects , but not in human cases . Expanded catalytic models suggest that transmission was interrupted in the study area in 1996 ( 95% credible interval: 1991–2000 ) , with a resultant decline in the average annual incidence of infection from 0 . 9% ( 95% credible interval: 0 . 6–1 . 3% ) to 0 . 1% ( 95% credible interval: 0 . 005–0 . 3% ) . Through a search of archival newspaper reports , we uncovered documentation of a 1995 vector control campaign , and thereby independently validated the model estimates . High levels of T . cruzi transmission had been ongoing in peri-rural La Joya prior to interruption of parasite transmission through a little-documented vector control campaign in 1995 . Despite the efficacy of the 1995 control campaign , T . cruzi was rapidly reemerging in vector populations in La Joya , emphasizing the need for continuing surveillance and control at the rural-urban interface .
An estimated 8 million people in Latin America are infected by the protozoan parasite Trypanosoma cruzi , the causative agent of Chagas disease [1] . Trypanosoma cruzi is typically transmitted to humans and other mammals through contact with feces of an infected blood-feeding triatomine insect . The primary vector species in southern Peru is Triatoma infestans , which has adapted to live in and around human dwellings . Infection can also occur via congenital transmission , blood transfusion , or organ transplantation [2] . Infection is generally life-long and while most infected individuals remain asymptomatic , 20–30% progress over a period of decades to chronic clinical manifestations , including life-threatening cardiac and/or gastrointestinal disease [3] . As a result , Chagas disease is estimated to be responsible for the loss of 670 , 000 disability-adjusted life years ( DALYs ) and 14 , 000 human lives annually [4] . Trypanosoma cruzi transmission by T . infestans has been interrupted in several South American countries through household application of pyrethroid insecticides , but a comprehensive approach to vector control has only recently been instituted in southern Peru [1] , [5] . Throughout Latin America , however , Chagas disease vector control is complicated by the processes of urbanization and migration [6] , [7] . In recent decades in southern Peru , extensive urbanization has occurred at the periphery of cities as well as within previously rural areas [8] . New localities are typically established by rural migrants and share the trait of being situated – geographically as well as socio-culturally – at a rural-urban interface [9] . To improve understanding of T . cruzi transmission in the peri-rural context , we performed cross-sectional serological and entomological surveys in four contiguous localities located 30 km from the city of Arequipa . We evaluated spatial and temporal patterns of T . cruzi infection , utilizing a multivariate catalytic model [10] and Bayesian methods to estimate incidence of infection over time .
The ethical review committees of the Johns Hopkins Bloomberg School of Public Health , the Universidad Peruana Cayetano Heredia , and the University of Pennsylvania approved the research protocol . The ethical review committee of the University of Arizona approved the usage of de-identified study data . All individuals ≥1 year old residing within the study area were invited to participate in the serological study . Signed informed consent was obtained prior to participation by adults and parents of participating children . Children also provided signed informed assent prior to participating . All households in the study area were invited to participate in the entomological study . Signed informed consent was obtained prior to participation by an adult resident of each household . The district of La Joya ( population 24 , 192 ) is located approximately 30 km southwest of the city of Arequipa ( population 864 , 250 ) and encompasses a mosaic of rural and peri-rural communities [8] . In this article , peri-rural refers to communities with high-density human habitation within an otherwise rural landscape , whereas peri-urban describes localities with high-density human habitation located at the periphery of an urban center [9] . The La Joya study area ( Figure 1 ) was comprised of four contiguous peri-rural localities , with 2 , 251 persons living in 678 households within a 41-hectare area . From August through November 2008 , we conducted human and entomological surveys . The human survey included a collection of demographic data and a detailed history of where each individual had lived from birth to the present time . The human survey and vector collections were timed to coincide with a household insecticide spray campaign conducted by the Arequipa Regional Office of Health . No documentation of any previous household insecticide campaign in the area was found in the Chagas Control Program records prior to the study . Detailed socio-demographic data , including a comprehensive migration history , were collected from each consenting study participant . Migration data were coded to indicate whether each individual lived inside or outside of the study area during each calendar year over the span of his or her lifetime . Five ml of venous blood was drawn from each study participant by trained medical personnel ( 3 ml for children younger than 5 years ) . Blood was maintained at 4°C and separated by centrifugation on the day of collection . Serum and cellular fractions were stored at −20°C until testing . Serum specimens were evaluated for antibodies to T . cruzi using a commercially available enzyme-linked immunosorbent assay ( ELISA ) ( Chagatek , Laboratorio Lemos ) . The assay is reported to have 100% ( 95% CI: 82 . 9–100% ) sensitivity; specificity estimates have varied from 87 . 3% ( 80 . 9–92 . 2% ) to 98 . 1% ( 93 . 8–99 . 7% ) [11] . All specimens with positive ELISA results and a 20% random sample of specimens with negative results were tested by the indirect immunofluorescent antibody test ( IFA ) using a titer of 1:32 as the positive cutoff [12] . Specimens with a positive result for both ELISA and IFA were considered confirmed positive for T . cruzi infection . Immediately after application of the residual pyrethroid insecticide deltamethrin 5% Wettable Powder ( K-Othrine , Bayer ) at 25 mg a . i . per m2 , two trained triatomine collectors systematically searched all domestic and peri-domestic areas , which included human and animal living spaces , for a total of one person-hour for each participating household [13] . Captured T . infestans were transported the same day to the Arequipa-based Universidad Peruana Cayetano Heredia laboratory . Insect life stage , sex ( adults ) , and count were documented by site of collection . Gut contents of all insects ( except first instars ) were evaluated microscopically to determine T . cruzi infection status [14] . Household geographic coordinates were collected using a global positioning system ( GPS ) unit ( Juno ST , Trimble Navigation Limited ) . Geographic coordinates; household and individual participant codes; and diagnostic , socio-demographic , and entomological-parasitological data were maintained in a relational database management system and imported into a geographic information system ( GIS ) ( ArcGIS 9 . 3 . 1 , ESRI ) for subsequent analysis and visualization .
Of 2 , 251 study area residents , 1 , 333 ( 59 . 2% ) participated in the serological survey . Compared with study participants , non-participants were younger ( mean age = 22 . 6 versus 26 . 6 years; p<0 . 001 ) and were more likely to be male ( 57 . 1% versus 42 . 2%; p<0 . 001 ) . One hundred forty-four participants had positive results by T . cruzi ELISA ( 10 . 8%; 95% CI: 9 . 1–12 . 5% ) , and 101 ( 7 . 6%; 95% CI: 6 . 2–9 . 0% ) had confirmed positive results by IFA . Forty-three individuals had ELISA-positive/IFA-negative diagnostic results and were considered seronegative in subsequent analyses . Of 238 randomly selected ELISA-negative specimens , 226 were IFA-negative and 12 were IFA-positive . All were considered seronegative in subsequent analyses . ELISA-IFA inter-diagnostic concordance was evaluated using the kappa statistic: kappa = 0 . 687 ( 95% CI: 0 . 610–0 . 764 ) . All T . cruzi seropositive individuals were evaluated by Arequipa Regional Office of Health medical personnel and were offered treatment in accordance with the Ministry of Health of Peru guidelines . Study participants contributed a total of 35 , 501 person-years at risk for T . cruzi infection , 19 , 514 ( 55 . 0% ) of which were experienced inside the study area and 15 , 987 ( 45 . 0% ) of which were experienced outside the study area . The prevalence of T . cruzi infection increased with increasing age , years of residence inside the study area , and years of residence outside the study area ( Table 1 ) . Of 513 study participants aged 18 years or younger , only five ( 1 . 0%; 95% CI: 0 . 1–1 . 8% ) had confirmed positive serological results , and three of these were children of T . cruzi seropositive mothers . There were no significant associations between individuals' serostatus and the presence or density of T . infestans or T . cruzi-infected T . infestans in their households ( Table 2 ) . Of 678 study area households , 405 ( 59 . 7% ) had one or more residents who participated in the serological survey . Eighty-five ( 21 . 0%; 95% CI: 17 . 0–25 . 0% ) households surveyed contained at least one person with T . cruzi infection . Six hundred twenty-three ( 91 . 9% ) households participated in the entomological survey , and T . infestans were collected in 171 ( 27 . 4%; 95% CI: 23 . 9–31 . 0% ) households surveyed . Of the 170 infested households with vectors examined , 46 ( 27 . 1%; 95% CI: 20 . 3–33 . 8% ) contained T . cruzi-positive vectors ( Figure 2 ) . K-function difference analysis demonstrated statistically significant spatial clustering of vector-infested households at spatial scales from 10 to 500 m . Similar analysis found significant spatial clustering of households with T . cruzi-infected vectors at spatial scales from 10 to 130 m and 390 to 410 m . No significant clustering was found at any spatial scale evaluated for households containing T . cruzi-infected humans ( Figure 3 ) . The migration-transmission interruption model provided the best fit to the seroprevalence data ( Table 3 , Figure 4 ) . This model indicated that transmission was interrupted around 1996 , approximately 12 years prior to our study . Based on model estimates , the mean annual incidence of T . cruzi infection in the study area was approximately 1% prior to transmission interruption and fell to 0 . 1% from then onward . The mean incidence per year of residence outside the study area was approximately 0 . 1% . The intercept value was estimated to equal 0 . 5% . This value represents the probability of T . cruzi infection independent of time of exposure , including the probability of congenital infection . Based on model results , we interviewed Ministry of Health personnel who had been working since at least the early 1990s . Several officials recounted that then-President Alberto Fujimori visited La Joya in 1995 as part of his re-election campaign and authorized financing for an insecticide spray campaign in response to local political demands . Ministry field personnel confirmed that a single , district-wide application of the pyrethroid insecticide 10% lambda-cyhalothrin at 25 mg a . i . per m2 was implemented in May to October 1995 . We subsequently searched local newspaper archives and encountered an article documenting the control campaign [23] . Data obtained from personal interviews and newspaper records strongly support model estimation of transmission interruption ( Figure 5 ) .
The transmission of Trypanosoma cruzi by Triatoma infestans can be interrupted through coordinated and sustained vector control efforts , as has been demonstrated in Brazil , Chile , and Uruguay [24] . The history of Chagas disease control in Peru and many other nations , however , is marked by sporadic and poorly documented vector control campaigns [25] . These campaigns , although ephemeral , can affect the patterns of infection in a population . Through a catalytic model we were able to reconstruct the history of T . cruzi transmission in the peri-rural community of La Joya , Peru , and in the process describe the effect of a nearly forgotten vector control campaign . The classic catalytic model has been utilized widely to retrospectively estimate incidence from age-prevalence data for a variety of lifelong infections [26]–[28] , including T . cruzi infection [29]–[31] . The basic model assumes that incidence of infection is constant with time and age; we show here that the model can be easily expanded to accurately describe a complex history of transmission and control , even in a peri-rural community with a highly mobile population . We found that incidence of T . cruzi infection was high prior to a vector control campaign in 1995 . This control campaign successfully disrupted transmission of T . cruzi . Our estimates from the expanded catalytic model show prevalence in the study population increasing from 5% in twenty year olds , up to 10% among thirty year olds and 20% among those over sixty . Little of the prevalence could be attributed to infection outside of the study area . Peri-rural and peri-urban places play a crucial role in contemporary dynamics of migration and urbanization across Latin America [9] . In turn , migration and urbanization are changing the geography and epidemiology of many parasitic infections , including Chagas disease [6] , [7] . The incursion of T . infestans and T . cruzi into nearby urban and peri-urban Arequipa [32] , [33] may have its source in peri-rural communities like La Joya . Inhabitants of peri-urban Arequipa typically migrate to nearby rural and peri-rural localities in search of seasonal agricultural employment [34] . The cycle of seasonal migration provides a plausible means for the rural-to-urban transportation of vector and parasite . The timing , at least , is suggestive . Triatoma infestans emerged in peri-urban communities of Arequipa in the 1980s and 1990s [34] . Our analysis here shows that transmission was very common in La Joya in the 80s and midway through the 90s . Molecular studies [35] of the vector and parasite could provide a definitive answer to the question of provenance of Chagas disease in and around the city of Arequipa . The intervention that nearly eliminated T . cruzi transmission in the study communities consisted of a single application of pyrethroid insecticide . The intervention effectively stopped T . cruzi transmission for many years . However , as is often the case , over time T . infestans and T . cruzi reemerged in the end [25] . The spatial patterns we observed in La Joya are precisely those to be expected following a reasonably effective intervention campaign . A large number of older individuals were infected , and they presumably were infected prior to the intervention campaign by vector populations that were spatially diverse . The vectors we observed during our study were clustered , suggesting that the vector population is so young that it was still in the process of re-dispersing through the community . Households with vectors carrying the parasite were even more spatially clustered , suggesting that the re-dispersal of T . infestans has outpaced that of T . cruzi . We found very low prevalence of infection in children , and some of the few children diagnosed may well have been infected congenitally . The resurgent T . cruzi transmission clearly had not yet caused many infections in humans . However , had the parasite been allowed to continue to spread there is little doubt that it would have resulted in significant disease in the human population . The contrasting presence of T . cruzi in the vector population and absence of new human infections suggest that parasite reemergence was limited to animal reservoirs . There are diverse and dense domestic animal populations in peri-rural communities of Peru , and La Joya is no exception . The guinea pig is particularly common , and is a highly competent reservoir for T . cruzi . We speculate that the presence of high densities of guinea pigs has allowed for reemergence of T . cruzi . At least three important study limitations merit mention . First , T . cruzi diagnostics lack a gold standard , and are especially difficult to interpret in Arequipa [36] . We considered only confirmed ( ELISA-positive/IFA-positive ) cases of infection; spatial analyses elsewhere in Arequipa suggests that ELISA-positive/IFA-negative results likely represent true positives [37] . Accordingly , our estimates of seroprevalence and incidence in the study population may be lower than the true values . Second , non-participants were on average slightly younger than participants , which might lead to a slight underestimate of seroprevalence in the population , though this would not affect the incidence estimates from the catalytic model . Third , we were unable to consider a more detailed model of geographic variability in risk of T . cruzi infection . This limitation resulted from the vast variety of study participants' migration histories , which included many small towns that we could not locate . We attempted to gather information on the previous insecticide control activities in the study site prior to conducting our study . In-depth interviews conducted with community members failed to elicit recall of the 1995 insecticide application campaign [34] . Insecticide application was simply not a memorable event in the lives of community members . We may have lacked some due diligence in examining files in the health post; these were not made readily available to us . We note that had we included exact information on the timing of the intervention campaign in the catalytic model , our point estimates of incidence of infection before and after the campaign would not likely have changed , although the credible intervals around those point estimates may have been narrower . We present our method for making such estimates without perfect knowledge of the history of control activities because we believe they may prove useful for others conducting similar serologic studies on Chagas , or other diseases , when historical information on control activities is incomplete . The geography and epidemiology of Chagas disease – like that of many parasitic diseases – is changing . Decreased funding and insecticide resistance are endangering gains achieved by the insecticide-based interventions such as the Southern Cone Initiative against T . infestans [38] . Waning political interest further complicates implementation of sustained vector control , and economically and politically marginalized populations may suffer disproportionately . Transmission cycles of T . cruzi are emerging in peri-urban communities [32] , [33] , and re-emerging in peri-rural communities like La Joya . Integrated epidemiological , entomological , environmental and historical data are needed to better elucidate past processes driving the changing geography of Chagas disease , and to facilitate control of new cycles of transmission , especially at the complex rural-urban interface .
|
The historically rural problem of Chagas disease is increasing in urban areas in Latin America . Peri-rural development may play a critical role in the urbanization of Chagas disease and other parasitic infections . We conducted a cross-sectional study in an urbanizing rural area in southern Peru , and we encountered a complex history of Chagas disease in this peri-rural environment . Specifically , we discovered: ( 1 ) long-standing parasite transmission leading to substantial burden of infection; ( 2 ) interruption in parasite transmission resulting from an undocumented insecticide application campaign; ( 3 ) relatively rapid re-emergence of parasite-infected vector insects resulting from an unsustained control campaign; ( 4 ) extensive migration among peri-rural inhabitants . Long-standing parasite infection in peri-rural areas with highly mobile populations provides a plausible mechanism for the expansion of parasite transmission to nearby urban centers . Lack of commitment to control campaigns in peri-rural areas may have unforeseen and undesired consequences for nearby urban centers . Novel methods and perspectives are needed to address the complexities of human migration and erratic interventions .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"infectious",
"diseases/neglected",
"tropical",
"diseases"
] |
2011
|
A History of Chagas Disease Transmission, Control, and Re-Emergence in Peri-Rural La Joya, Peru
|
The Gram negative bacterium Vibrio cholerae is the causative agent of the secretory diarrheal disease cholera , which has traditionally been classified as a noninflammatory disease . However , several recent reports suggest that a V . cholerae infection induces an inflammatory response in the gastrointestinal tract indicated by recruitment of innate immune cells and increase of inflammatory cytokines . In this study , we describe a colonization defect of a double extracellular nuclease V . cholerae mutant in immunocompetent mice , which is not evident in neutropenic mice . Intrigued by this observation , we investigated the impact of neutrophils , as a central part of the innate immune system , on the pathogen V . cholerae in more detail . Our results demonstrate that V . cholerae induces formation of neutrophil extracellular traps ( NETs ) upon contact with neutrophils , while V . cholerae in return induces the two extracellular nucleases upon presence of NETs . We show that the V . cholerae wild type rapidly degrades the DNA component of the NETs by the combined activity of the two extracellular nucleases Dns and Xds . In contrast , NETs exhibit prolonged stability in presence of the double nuclease mutant . Finally , we demonstrate that Dns and Xds mediate evasion of V . cholerae from NETs and lower the susceptibility for extracellular killing in the presence of NETs . This report provides a first comprehensive characterization of the interplay between neutrophils and V . cholerae along with new evidence that the innate immune response impacts the colonization of V . cholerae in vivo . A limitation of this study is an inability for technical and physiological reasons to visualize intact NETs in the intestinal lumen of infected mice , but we can hypothesize that extracellular nuclease production by V . cholerae may enhance survival fitness of the pathogen through NET degradation .
The Gram negative facultative human pathogen Vibrio cholerae is the causative agent of cholera , which is defined as an acute , secretory diarrheal disease . Today , the global burden of cholera is estimated to reach several million cases per year , with the majority located in the endemic areas of Africa and Asia [1] . However , explosive outbreaks facilitated by natural disasters , high population density and poor sanitation can occur worldwide as recently demonstrated by the cholera epidemic in Haiti , where cholera cases have not been reported before 2010 [2] . The lifecycle of clinically relevant V . cholerae serogroup O1 and O139 is marked by two distinct phases . V . cholerae is not only a pathogen of the human gastrointestinal tract , but also a natural inhabitant of aquatic ecosystems , which serve as important reservoirs during periods between epidemics . Biofilm formation on chitinous surfaces provided by zoo- and phytoplankton as well as entry into a viable but non-culturable state are thought to be important for persistence within these nutrient limited environments [3] , [4] . Infection usually starts with the oral ingestion of V . cholerae with contaminated food or water . The infectious dose is quite high and ranges from 106 to 108 depending on the acidity in the stomach and overall health of the human being [5] , [6] . After passage through the stomach , V . cholerae reaches the small bowel , its primary site of colonization , and induces virulence factors such as the toxin coregulated pilus and the cholera toxin . Incubation periods from 12 h up to several days have been described , before the first symptoms can be recognized [5] , [6] . Due to the activity of the cholera toxin , the patient develops a massive watery diarrhea with volumes of up to 20 l stool per day , which can rapidly lead to life threatening dehydration , hypotensive shock and organ failure . Without treatment the case-fatality rate for severe cholera can exceed 70% [5] . V . cholerae leaves the host in a transient phenotype called hyperinfectivity , which is characterized by a infectious dose 10 to 100-fold lower compared to in vitro-grown bacteria [7] . In addition , V . cholerae exhibits an exceptional growth rate in the gastrointestinal tract and exits the human host at relatively high numbers of up to 108 CFU per ml patient stool with the onset of the diarrhea [6] , [8] . These observations provide some explanation for the rapid transmission and explosive spread of cholera during outbreaks . In general , cholera is still considered to be rather a noninflammatory secretory disease . However , microscopical studies conducted by Mathan and coworkers in 1995 revealed an activation and increase in inflammatory cells in the gut of cholera patients [9] . Subsequent studies demonstrated a broad upregulation of inflammatory cytokines as well as recruitment of polymorphonuclear neutrophils ( PMNs ) in the gastrointestinal tract of patients during the acute phase of cholera [10]–[12] . Signals stimulating the inflammatory response seem to be diverse . Recent studies demonstrated that Vibrio flagellins activate the inflammatory response via TLR5 , lipopolysaccharide ( LPS ) is recognized through TLR4 and lipoproteins by TLR1/2 [13]–[15] . Interestingly , V . cholerae toxins seem to have some anti-inflammatory activity . In vitro studies demonstrated that pretreatment with cholera toxin suppresses the induction of cytokines in LPS-stimulated macrophages [16] . These results provide some explanation for the observed side effects including a pronounced inflammatory response in human volunteers vaccinated with live-attenuated oral nontoxigenic V . cholerae strains , which were deleted for the cholera toxin [17] , [18] . In addition , it was recently suggested that the cytolytic and proteolytic activity of the accessory toxins could affect innate immune cells , including neutrophils , and may therefore facilitate prolonged colonization in the host [19] , [20] . A first line of defense and central part of the human innate immunity are neutrophils , which are recruited to the site of infection by chemokines secreted from macrophages or local cells upon contact with microbial pathogens . Besides phagocytic activity , activated neutrophils can release NETs , which are composed of decondensed chromatin associated with granular and cytoplasmic proteins [21] , [22] . These NETs can effectively entrap and promote extracellular killing of bacteria , most likely due to their high serine protease content [21] . Extracellular nucleases of the Gram positive pathogens Staphylococcus aureus , Streptococcus pneumoniae and group A Streptococcus have been reported to play a role in the evasion of neutrophil defenses via degradation of the NET-backbone , enabling the liberation of the bacteria from NETs [23]–[25] . Interestingly , V . cholerae encodes two extracellular nucleases Dns and Xds , which exhibit endo- and exonuclease activity , respectively . Recently , we identified xds as a gene induced at a late stage of infection [26] . While a minor fraction of these late genes were required for maintenance of colonization , most of them were found to facilitate the transition fitness of V . cholerae from the host into the aquatic lifestyle [26] . Thus , we initially investigated the physiological function of the two extracellular nucleases Dns and Xds of V . cholerae during biofilm formation and characterized them as modulators for extracellular DNA in the V . cholerae biofilm matrix [27] . Nevertheless , the induction in vivo prompted us to investigate whether the nucleases also play a role during infection and provide a mechanism against NETs . In the present study , we show that the extracellular nucleases increase the colonization fitness in immunocompetent mice , but are not required for colonization in mice depleted for neutrophils ( neutropenic mice ) . Based on this observation , we comprehensively characterized the interplay between neutrophils and V . cholerae and demonstrate that the extracellular nucleases are essential for NET degradation , mediate evasion from NETs and thereby increase survival upon contact with neutrophils .
To determine whether there is a difference in the colonization efficiency of wild type and double nuclease mutant , we conducted single strain infections using an adult mouse model according to Nygren et al . [28] . The streptomycin-treated adult mouse model was chosen over the otherwise frequently used infant mouse model , because infant mice have been reported to exhibit an innate immunity with impaired cytokine response , i . e . interleukin IL-1 and IL-6 [29]–[31] . Furthermore , the streptomycin-treated adult mouse model allows a stable colonization of V . cholerae for at least 72 h , which offers the possibility to investigate the impact of the innate immune response on maintenance of the infection [28] . Stable colonization with V . cholerae in the streptomycin-treated adult mouse is limited to the cecum and colon , while the pathogen is rapidly cleared from the small intestine . That is why , the results of the current study were obtained from cecal or colonic tissues . Longer persistence of V . cholerae in the small intestine requires the additional use of ketamine anesthesia [32] . Unfortunately , ketamine has been reported to affect neutrophil activation and recruitment [33]–[36] . Thus , ketamine treatment might have masked innate immune mechanisms relevant for this study and was not applied . C57BL/6 mice were inoculated intragastrically with V . cholerae wild type or the ΔdnsΔxds mutant and colonization was analyzed after 24 h and 72 h ( Fig . 1A , open bars ) . Although all mice showed a considerable amount of recoverable CFU at both time points , the double nuclease mutant showed a tendency towards a slightly reduced colonization . This trend consolidated over time and resulted in a significantly reduced median colonization of the double nuclease mutant compared to wild type at 72 h ( Fig . 1A , open bars ) . Induction of inflammatory and regulatory cytokine and chemokine expression was assessed by qRT-PCR in tissue of V . cholerae infected mice and compared to streptomycin-treated , but uninfected control mice referred to as mock-inoculated controls from hereon ( Fig . 1B ) . At 24 h the expression levels of IL-6 , the macrophage inflammatory protein 2 alpha ( MIP-2 ) and the keratinocyte derived cytokine ( KC ) , representing a functional homolog of the human IL-8 [37] , [38] , were already 2-fold elevated in mice infected with V . cholerae wild type or ΔdnsΔxds mutant . At 72 h the inflammatory response was generally more apparent and characterized by 2 to 7-fold upregulated expression levels of the inflammatory markers in mice infected with V . cholerae wild type or ΔdnsΔxds mutant compared to the mock-inoculated control group . In concordance with the current literature , these data demonstrate an inflammatory response in the gastrointestinal tract upon infection with V . cholerae [10]–[12] , [19] . Noteworthy , at 72 h distinct differences in the inflammatory response between wild type and ΔdnsΔxds mutant infected mice could be observed ( Fig . 1B ) . While the induction of IL-6 was quite comparable in both infection groups , ΔdnsΔxds mutant infected mice exhibited a significant higher expression of MIP-2 and KC compared to wild type infected mice . MIP-2 ( CXCL2 ) and KC ( CXCL1 ) belong to the CXC chemokines , which are potent chemoattractants involved in neutrophil recruitment to the site of infection [39]–[42] . In further consistency with previous reports [9] , [10] , [43] , [44] , infiltration of neutrophils was detected for mice colonized with wild type and ΔdnsΔxds mutant by histological analysis of cecal tissue sections ( Fig . 1C and S1 ) . At 24 h post infection mice colonized with wild type and ΔdnsΔxds mutant exhibited a significant higher amount of neutrophils infiltrating the tissue compared to the mock-inoculated control group ( Fig . 1C ) . Immunofluorescence staining of histone H1 and DNA along with myeloperoxidase ( MPO ) , a marker for neutrophils [45] , was used to support the neutrophil infiltration in the epithelium ( Fig . 1D ) . At 72 h the effects became indistinct due to a notable increase of neutrophils in the mock-inoculated control group , which is most likely caused by side effects from the continuous treatment with antibiotics ( Fig . 1C ) . Furthermore , the peak of neutrophil infiltration in mice colonized with V . cholerae for 72 h might have already passed . Thus , most of the neutrophils might have reached the intestinal lumen and underwent activation , while histological analysis allows only the quantification of the remaining neutrophils . Based on these results , we hypothesized that the observed colonization defect of the double nuclease mutant at 72 h might result from an impaired evasion from neutrophils . Thus , colonization fitness of the wild type and the ΔdnsΔxds mutant was assayed in neutropenic mice , which were generated by treatment with Anti-Ly6G mAb prior to infection [46] . In the case of the infection with V . cholerae wild type almost comparable colonization levels in immunocompetent and neutropenic mice were observed ( Fig . 1A , right panel ) . In contrast , the median colonization of the ΔdnsΔxds mutant in neutropenic mice was significantly increased by 10-fold compared to immunocompetent mice ( Fig . 1A , right panel ) . These data suggest that the different colonization fitness of the double nuclease mutant in immunocompetent and neutropenic mice depends on the presence of neutrophils . These in vivo results prompted us to investigated the direct interplay of neutrophils and V . cholerae , since the interaction between neutrophils and V . cholerae is largely uncharacterized . First , we analyzed the reactive oxygen species ( ROS ) production of neutrophils upon contact with V . cholerae , since the oxidative burst is a key feature for neutrophils , which is involved in signaling processes as well as microbial killing [45] , [47] . ROS production of human neutrophils incubated with V . cholerae wild type , ΔdnsΔxds , Δdns or Δxds mutant was measured in a luminometric based plate assay over a time period of 6 h . As a positive control neutrophils were stimulated with phorbol myristate acetate ( PMA ) , which is a potent , non-physiological activitor for ROS production . Unstimulated neutrophils served as negative control . As expected PMA-stimulated neutrophils showed pronounced ROS production , while no increase in ROS occurred in unstimulated neutrophils ( Fig . S2A ) . A robust ROS production in a MOI dependent manner was observed for all V . cholerae strains tested ( Fig . 2A and B ) , indicating that V . cholerae is recognized by human neutrophils . Furthermore , the peak of ROS production is reached at approximately 4 h ( MOI 4 ) or at approximately 3 h ( MOI 40 ) independent of the strains tested ( Fig . S2B and C ) . Since the detected ROS levels and dynamics are comparable for wild type and nuclease mutants , the recognition and activation potential of neutrophils by V . cholerae is independent of the presence of the two extracellular nucleases . Interestingly , bacterial extracellular nucleases have been shown to degrade NETs , which are released upon microbial stimulation and represent chromatin decorated with antimicrobial proteins [25] . Moreover , ROS has been reported to be essential for NET formation [45] , [47] . To test whether V . cholerae is able to induce NET formation , a fluorescent-based quantification of DNA released by neutrophils was performed . In detail , DNA was quantified using the cell impermeant fluorescent dye Sytox green , which only detects extracellular DNA or DNA not surrounded by an intact membrane . In the case of NET formation the plasma membrane of neutrophils gets ruptured and DNA decorated with antimicrobial peptides is released . Consequently , released DNA is stained by Sytox Green , which results in an increase in the fluorescent signal and is a well established technique to quantify NET formation [45] . Neutrophils were stimulated with wild type , ΔdnsΔxds , Δdns or Δxds mutant ( Fig . 2C–F ) . Again , PMA-stimulated and unstimulated neutrophils served as positive and negative control , respectively ( Fig . 2G ) . Release of DNA started approximately after 2 h for the PMA-stimulated neutrophils , while the fluorescent signal remained at very low levels in the unstimulated control . To confirm that the assay indeed measures DNA , Dnase I was added to neutrophils 6 h after PMA-stimulation ( Fig . 2G ) . As expected the signal rapidly decreased due to the degradation of extracellular DNA . In the case of neutrophils stimulated with ΔdnsΔxds or Δdns mutant the NET level increased over time after a lag-phase of 4 to 6 h and reached a plateau at approximately 80% of total DNA ( Fig . 2C and D ) . Notably , wild type stimulated neutrophils showed only a slight increase in the NET level with the maximum at 20% followed by a decline . The single nuclease mutant Δxds displayed an intermediate phenotype . An increase of the MOI had neither an effect on the general trends nor on the maximal level of NET formation , but resulted in a slightly shortened lag-phase ( Fig . 2C and D ) . Again , addition of DNAse I to neutrophils stimulated with the double nuclease mutant after 6 h resulted in a rapid decrease of the fluorescence intensity confirming that the fluorescence signal reflects liberated DNA ( Fig . 2H ) . As demonstrated above all V . cholerae strains tested showed similar induction of ROS-levels ( Fig . 2A and B ) . Thus , the lower NET formation in the wild type or Δxds cannot simply be explained by a lower level of neutrophil activation . Furthermore , complementation to wild type levels or even below could be achieved for the double or single nuclease mutants by expression of dns or xds in trans on a plasmid for both MOI doses tested ( Fig . 2E and F , Fig . S3A and B ) . The vector control showed no effect on the observed phenotypes in all strains tested . In summary , reduced NET formation observed in this assay correlates with expression of the extracellular nucleases . Since the readout of this assay is based on the detection of the released DNA from neutrophils , the data suggests that the V . cholerae nucleases degrade NETs very efficiently . In order to eliminate the possibility that the observed NET formation is strain specific , the assay was repeated in another V . cholerae clinical isolate including the respective nuclease mutants with essentially the same result ( Fig . S3C and D ) . Recently , Valeva et al . reported that the V . cholerae cytolysin attacks the membrane of granulocytes and thereby stimulates ROS production , activates degranulation and release of elastase [48] . In order to test whether the cytolysin also affects liberation of DNA by neutrophils we constructed a triple mutant deleted for the cytolysin ( HlyA ) as well as for the two extracellular nucleases and measured the amount of released DNA by the fluorescence assay . Indeed , the amount of liberated DNA is reduced by 50% in the triple mutant compared to the double mutant ( Fig . 2H ) . Thus , the cytolysin seems to be a potent stimulus for degranulation and DNA release . The residual activity in the triple mutant indicates additional stimuli of V . cholerae , which can induce NET formation . To confirm the hypothesis that Dns and Xds degrade DNA of NETs , we performed a degradation assay using the same fluorescent-based assay as for the NET quantification ( Fig . 3A ) . In contrast to the previous assay , NET formation was stimulated with PMA prior to the addition of the wild type , ΔdnsΔxds , Δdns or Δxds mutant . This allowed a quantification of the degradation capacity of all strains on already formed NETs . In samples incubated with ΔdnsΔxds or Δdns mutant no degradation was visible , while incubation with wild type or Δxds resulted in pronounced degradation of the NETs within 8 h . Expression of dns in trans restored the degradation capacity of the double nuclease mutant even beyond wild type levels . In both assays presented in Fig . 2 and 3A , Δdns shows more pronounced effects than Δxds . Thus , the endonuclease Dns seems to be more efficient in degradation of NETs compared to the exonuclease Xds . However , the wild type exhibits the most prominent degradation capacity , which suggests that both nucleases act somewhat synergistically . In addition , live cell microscopy was performed over a period of 8 h to monitor neutrophils stimulated with wild type or ΔdnsΔxds mutant in presence of the cell impermeant fluorescent dye Sytox green ( Movie S1 and S2 ) . This allowed temporal detection of NET formation as well as NET degradation on the cellular level by appearance and disappearance of a fluorescent signal . Images of selected time points along the experiment are provided in Fig . 3B . Within the first hours a similar progression of NET induction was monitored for wild type and ΔdnsΔxds mutant , reaching a maximum level of intensity at 5 . 5 h . From there the fluorescent signal remained quite stable in case of the ΔdnsΔxds mutant over the entire imaging period , which indicates no NET degradation . In contrast , the fluorescent signal steadily vanished in the presence of the wild type leaving only few remaining signals at the endpoint of the experiment . Taken together , these results demonstrate that wild type and ΔdnsΔxds mutant stimulate NET formation to an equal level , but only the wild type has the ability to degrade NETs by the activity of the two extracellular nucleases . To visualize NET formation induced by V . cholerae in more detail , neutrophils were stimulated with PMA , wild type , ΔdnsΔxds mutant or left unstimulated and subsequently analyzed by confocal microscopic immunofluorescence ( Fig . 4A ) . In the unstimulated control sample intact neutrophils with lobulated nuclei and a granular staining for neutrophil elastase within the cytoplasm of cells could be observed . Micrographs of PMA stimulated neutrophils showed web-like structures of DNA with neutrophil elastase dispersed over areas larger than original cells indicating NET formation [47] . In wild type stimulated neutrophils large areas with neutrophil elastase surrounded by bacteria were visible . However , only few spots of DNA with low intensity and nearly no web-like structures could be detected . Especially , the absence of DNA in areas of elastase signal strongly suggests degradation of the DNA in NETs by the extracellular nucleases . In the case of neutrophils stimulated with ΔdnsΔxds mutant intense web-like DNA structures and neutrophil elastase signals could be observed . Consistent with the results of the degradation assay and live cell imaging , the ΔdnsΔxds is not capable to degrade DNA of the NETs . In comparison to the wild type , the ΔdnsΔxds mutant exhibits a stronger colocalization with decondensed neutrophil elastase and DNA . NET formation induced by V . cholerae was also microscopically quantified according to previously described methods [45] , [49] . The technology is based on the determination of the DAPI DNA signal area occupied by intact cells or NETs ( in µm2 ) , which provides an appropriate measure to distinguish NETs from intact or non-NETotic dead neutrophils and allows a quantification of NET formation based on morphological structures ( see Materials and Methods for details ) . Using this approach we determined that 90% of the PMA-stimulated neutrophils , 15% of wild type incubated neutrophils and 30% of the double mutant incubated neutrophils underwent NET formation after 6 h ( Fig . 4B ) . Overall these values are quite comparable to the percentage DNA fluorescence at the 6 h time point measured by the extracellular DNA assay ( Fig . 2 ) and therefore confirm these results . Again , lower levels of NET formation in the wild type can be explained by the degradation of extracellular DNA due to the activity of the two nucleases as observed in microscopic images ( Fig . 4A ) . Thus , the real level of NET formation is most likely underestimated for the wild type by assays based on extracellular DNA levels . Noteworthy , complete degradation of NETs by V . cholerae wild type requires several hours ( Fig . 2 , 3 and 4 ) . This might indicate , that the extracellular nucleases are not constitutively expressed , but rather induced upon contact with NETs . To determine if gene expression of nucleases is induced in presence of NETs , qRT-PCR of samples extracted from wild type V . cholerae incubated with PMA stimulated neutrophils was performed ( Fig . 5A ) . Bacteria incubated in absence of neutrophils served as control condition . V . cholerae incubated with NETs were characterized by upregulated expression of dns ( 3-fold ) and xds ( 8-fold ) compared to the control condition . Interestingly , a similar induction was observed by incubating V . cholerae just with commercially available herring sperm DNA . Thus , V . cholerae induces expression of the two extracellular nucleases in presence of extracellular DNA , which is an abundant component of NETs . The colocalization of V . cholerae and NETs observed in Fig . 4A already provides some evidence that the ΔdnsΔxds mutant gets efficiently trapped within NETs . Indeed , NETs have been shown to trap and kill bacteria mediated through antimicrobial granule proteins decorating the NET structure and chromatin [21] . To investigate whether V . cholerae is ensnared in NETs , a plate based NET entrapment assay was performed . NET formation was stimulated with PMA followed by incubation with V . cholerae wild type or ΔdnsΔxds mutant ( Fig . 5B ) . A significant higher amount of ΔdnsΔxds mutant bacteria was found to be entrapped compared to wild type bacteria . In concordance with the NET visualization and degradation results , the reduced entrapment rate for the wild type is most likely due to degradation of NETs by the activity of the two extracellular nucleases . NETs not only entrap microbes and prevent spreading of the infection , but can also effectively kill bacteria due to antimicrobial effectors present at high concentrations in NETs [22] . To test whether V . cholerae is killed within NETs , a appropriate killing assay was conducted . Bacterial killing after coincubation of PMA stimulated neutrophils with wild type V . cholerae or ΔdnsΔxds mutant was measured ( Fig . 5C ) . A significantly higher total and extracellular killing rate of ΔdnsΔxds mutant bacteria compared to wild type bacteria was observed . Noteworthy , the phagocytic killing was similar for wild type and ΔdnsΔxds mutant , whereas the increased total killing of the ΔdnsΔxds mutant predominantly resulted from increased extracellular killing representing NET-mediated killing .
Until now , only little attention has been drawn to the innate immune response during a V . cholerae infection . Naturally , the overall inflammatory response to V . cholerae colonization is moderate compared to that seen during infections with enteroinvasive bacterial pathogens , e . g . Yersinia enterocolitica or Salmonella enterica , which are capable of penetrating the gut epithelium of the host and cause systemic infection [50] , [51] . However , emerging evidence by the recent literature indicates a substantial inflammatory response during cholera infection , which is marked by an induction of inflammatory cytokines as well as by recruitment of innate immune cells to the site of infection [10]–[12] . In agreement with these reports , we also observed a significant upregulation of inflammatory markers as well as infiltration of neutrophils into the murine gastrointestinal tract upon colonization with V . cholerae . Neutrophils act as first innate immune cells , which migrate to the site of infection for containment and clearance of the pathogens . Furthermore , activated neutrophils release a variety of chemokines for activation as well as recruitment of macrophages and dendritic cells [52] , [53] . Antimicrobial strategies of neutrophils range from phagocytosis of the respective microbes to the formation of NETs , which are able to bind and kill pathogens due to a high concentration of antimicrobial compounds [21] , [22] . NET formation of neutrophils upon contact with Gram negative and Gram positive bacteria as well as fungi has been described [21] , [54] . To withstand these attacks pathogens have evolved several mechanisms to circumvent neutrophil killing , which range from prevention of phagocytosis and intracellular survival to evasion of NETs . The later relies on extracellular nucleases degrading the chromatin fibers , which is the major structural component and scaffold of NETs . So far , this mechanism was solely described for the Gram positive bacteria , such as S . aureus , S . pneumoniae or group A Streptococcus causing skin or lung infections [23]–[25] . The data presented herein demonstrate that the Gram negative enteric pathogen V . cholerae also uses the activity of two extracellular nucleases , Dns and Xds , to efficiently degrade NETs and thereby evade entrapment and killing by NETs . The impact of this evasion mechanism is highlighted by a colonization defect of the double nuclease mutant in immunocompetent mice at 72 h . It seems reasonable that the neutrophils have to reach the colonization site first , before they can affect colonization . Accordingly , neutrophil infiltration and an increase of inflammation are already detectable at 24 h , while the impact of neutrophils on colonization requires their action on the luminal side and consequently manifests with a delay . The fact , that the observed colonization defect of the double nuclease mutant is negated in neutropenic mice corroborates the contribution of the extracellular nucleases on the evasion of NETs as part of the innate immune response against cholera . Despite intensive analyses detection of NET-like structures , defined by colocalization of MPO , histones and DAPI [21] , [47] , [55] , in the lumen of the gastrointestinal tract failed , while the visualization of neutrophil infiltration in the epithelium of infected mice was successful . As demonstrated by this study , the wild type rapidly degrades NETs , which reduces the likelihood to detect NETs in vivo . More stable NETs might exist in double nuclease mutant infected mice , but several technical limitations have to be taken into account . To our knowledge NET visualization has so far not been achieved in the lumen of the gastrointestinal tract , which has apparent degradative activities , exhibits a pronounced auto-fluorescence and mechanical forces act on the luminal content by the peristaltic movements . Thus , NETs could rapidly loose their characteristic shape and will not be detectable above the background . Recently , Queen and Satchell reported for a nontoxigenic ( cholera toxin-deficient ) V . cholerae strain that spread of the infection to spleen or liver within the first 6 h post infection is enhanced in neutropenic mice , while depletion of neutrophils did not affect the colonization levels in the gastrointestinal tract [19] . Based on these results , the authors hypothesized that neutrophils are required for containment of the infection to the gastrointestinal tract . Consistent with their findings , we also observed quite comparable colonization levels of a toxigenic V . cholerae wild type in the gastrointestinal tract in neutropenic and immunocompetent mice . In addition , Queen and Satchell demonstrated that the enhanced clearance of a multi-toxin deficient V . cholerae is negated in neutropenic mice , suggesting that neutrophils can impact the disease progression [19] . The colonization data obtained in the present study using the double nuclease mutant of a toxigenic strain reinforce the role of neutrophils in controlling the intestinal colonization of V . cholerae . In the V . cholerae wild type infection the impact of neutrophils on colonization levels seems to be masked by potent mechanisms of the bacterium , such as extracellular nucleases , which allow evasion from neutrophils . In addition to the secretion of extracellular nucleases , anti-inflammatory effects of the cholera toxin and cytolytic activity of accessory toxins have been described , which could also counteract the innate immune response [16] , [56] , [57] . Since these mechanisms of V . cholerae very likely act synergistically , one might only see the full impact of neutrophils by deletion of all effectors . Thus , it is quite remarkable to already observe a fitness defect of V . cholerae in vivo just by deletion of the two extracellular nucleases . Although the colonization level of the double nuclease mutant was significantly reduced compared to wild type , a successful colonization by the mutant was still established . Thus , the contribution of neutrophils on reduction of the overall colonization level seems to be moderate and might not affect the disease progression in the individual host to a large extend . However , rapid proliferation in the gut and leaving the host in high numbers is a key factor for transmission of cholera [6] , [8] . Reduction of the total colonization level in the host by 60% , as observed for the double nuclease mutant , could consequently decrease the amount of bacteria released in the stool and might minimize reinfections and the spread of the disease . In the case of Group A Streptococcus , it has been demonstrated that G-actin effectively blocks bacterial nuclease activity and enhances neutrophil killing of the pathogen in vitro [23] , but whether therapeutic approaches using nuclease inhibitors are applicable and can be used in the case of cholera outbreaks needs further investigation . Noteworthy , the expression of MIP-2 and KC , representing chemokines for attraction of innate immune cells , in the host tissue was significantly higher in mice infected with double nuclease mutant compared to those infected with wild type . One explanation could be that prolonged entrapment of the double nuclease mutant in NETs allows recruitment of additional innate immune cells to the site of infection and a continuous presentation of the bacterium to the immune system , which might enhance the chemokine response . However , V . cholerae wild type is present at even higher numbers and consequently should also serve as a potent stimulus for the innate immune cells . Interestingly , Uchiyama et al . recently showed that group A Streptococcus degrade their own DNA by the activity of the Sda1 nuclease , which abrogates recognition of bacterial DNA by TLR9 and prevents TLR9-dependent cytokine release , such as IFN-α and TNF-α [58] . Indeed , we confirmed the expression of these cytokines is significantly induced in murine Tlr4−/− macrophages or human neutrophils by incubation with genomic DNA derived from V . cholerae compared to DNAse I-treated DNA controls ( Fig . S4 ) . Thus , degradation of V . cholerae DNA by the activity of the extracellular nucleases could contribute to the silencing of the innate immune system during colonization of V . cholerae . Recently we hypothesized that one signal for induction of both nucleases in mature biofilms might be nutrient limitation , since both nucleases are induced under conditions of phosphate limitation [27] . As demonstrated in this study , the presence of NETs , i . e . extracellular DNA , seems to be an alternative signal for induction of the nucleases . Recently , xds was characterized as a gene induced at late stages of the infection [26] . Indeed , extracellular DNA as a major component of NETs seems to be a potent inducing signal for the nucleases inside the host . Ongoing research in our laboratory currently aims to elucidate the regulatory pathways and sensors involved . As described previously Dns exhibits endonuclease activity , while Xds has been characterized as an exonuclease [27] , [59] , [60] . Notably , NET degradation was more affected by the deletion of dns compared to the deletion of xds . This might argue for Dns being the dominant nuclease . One explanation might be the simple fact that an endonuclease has more target sites at the DNA compared to an exonuclease requiring free ends of DNA . Given the fact that endonuclease activity creates free DNA strand ends , which can be further cleaved by exonucleases , it seems rational that Dns is the predominant nuclease . Dns in turn does not require the exonuclease activity for full functionality . The double mutant showed the most prominent effects with no measurable degradation of host DNA . Consequently , we solely used the double mutant for following assays throughout the manuscript . The absence of any detectable NET degradation activity in the double nuclease mutant strongly suggests that Dns and Xds are the only extracellular nucleases essential for NET degradation . Consistent with the effects on biofilm formation reported previously [27] , it seems that Dns and Xds can at least partially compensate for each other and they may act somewhat synergistically . As a facultative human pathogen , the lifecycle of V . cholerae is marked by transitions between the environmental state in the aquatic ecosystems and the in vivo state inside the gastrointestinal tract of the host . Together with the results from our previous study [27] , we can now assign a dual role for the two extracellular nucleases of V . cholerae . On the one hand , they play a critical role in the development of a mature biofilm morphology by modulating the extracellular DNA content of the biofilm matrix and allow the utilization of DNA as a phosphate source in nutrient limited aquatic environments [27] . On the other hand , the data presented herein suggest that extracellular nucleases may facilitate the colonization fitness in vivo by liberation of the bacteria through degradation of NETs , which reduces NET mediated entrapment and killing . Similar to their role in the aquatic environment , the two nucleases of V . cholerae could also act as a nutrient scavenging mechanism in the host and facilitate the utilization of DNA as a nutrient source . Thus , V . cholerae efficiently uses the activity of the extracellular nucleases along its lifecycle to facilitate its survival in both habitats , the aquatic environment and the human host .
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the “Bundesgesetzblatt fuer die Republik Oesterreich” and the National Institutes of Health . The protocol was approved by the Committee on the Ethics of Animal Experiments of the University of Graz as well as the Austrian Federal Ministry for Science and Research BM . W-F ( Permit Number: 39/53/30 ex2012/13 ) . Mice were housed with food and water ad libitum and monitored under the care of full-time staff and in accordance with the rules of the Institute of Molecular Biosciences at the University of Graz . Human neutrophils were harvested from peripheral blood of healthy volunteers according to the recommendations of the local ethical committee . Fully written informed consent was provided by study participants , and all investigations were conducted according to the principles expressed in the Declaration of Helsinki . Bacterial strains and plasmids used in this study are listed in Table S1; oligonucleotides are listed in Table S2 . V . cholerae strain AC53 , a spontaneous streptomycin ( Sm ) -resistant derivative of the clinical isolate E7946 ( O1 El Tor Ogawa ) was used as wild type ( WT ) [61] . E . coli strain SM10λpir was used for genetic manipulations . Unless stated otherwise , strains were inoculated at an OD600 of 0 . 05 and grown to late log phase ( OD600 of 1 ) in LB medium with aeration at 37°C . Bacteria were washed 1 x in PBS , resuspended in PBS and adjusted to the respective multiplicity of infection ( MOI ) , being the ratio of V . cholerae to eukaryotic cells . If required , antibiotics and other supplements were used in the following final concentrations: Sm 100 µg/ml; ampicillin ( Ap ) 50 µg/ml; isopropyl-β-thiogalactopyranosid ( IPTG ) 0 . 5 mM . The isolation of chromosomal DNA , PCRs , the purification of plasmids or PCR products , the construction of suicide and expression plasmids as well as the subsequent generation were carried out as described previously [27] . Neutrophil isolation was performed as described previously [49] . Peripheral blood containing neutrophils was collected from healthy volunteer donors by venipuncture in combination with Vacuette blood collection tubes ( K3 EDTA , Greiner Bio-One ) . Briefly , neutrophils were separated from blood by two subsequent gradient centrifugation steps ( 800 g , 25 min ) using Histopaque 1119 ( Sigma-Aldrich ) and percoll ( Amersham ) , respectively . If not stated otherwise , cells were resuspended in RPMI 1640 without phenol red supplemented with 10 mM Hepes . Macrophages were isolated from the peritoneal cavitiy from Tlr4−/− mice 72 h after injection of 2 ml Brewer thioglycollate medium ( 3% ( w/v ) ) as previously described [62] . Tlr4−/− mice were kindly provided by the Institute of Experimental and Clinical Pharmacology ( Medical University of Graz ) [63] . Harvested Tlr4−/− macrophages were washed once and resuspended in RPMI-1640 containing 10% heat-inactivated fetal bovine serum . Tlr4−/− macrophages ( 5×105 ) were seeded into a 24 well plate and incubated for 24 h before fresh medium with genomic bacterial DNA was added . ROS production of human neutrophils was measured by a luminometric assay as described previously [49] . Briefly , 5×104 cells per well were seeded in a white 96 well plate . After addition of 50 µM luminol and 1 . 2 U/ml horseradish peroxidase , NET formation was stimulated either with 100 nM PMA , the indicated V . cholerae strain ( MOI 4 or 40 ) or left untreated . The chemiluminescence resulting from ROS production was measured every 3 min for a period of 6 h in a Tecan Infinite 200 plate reader at 37°C and 5% CO2 . The ROS amounts are presented integral of the area under the curve or as relative light units ( RLU ) for ROS dynamics . Presence of extracellular DNA was measured as described previously [45] , [49] . Briefly , 5×104 cells per well were seeded in a black 96 well plate . After addition of Sytox Green ( 2 . 5 µM ) , neutrophils were stimulated either with 100 nM PMA , the indicated V . cholerae strain ( MOI 4 and 40 ) , left untreated or lysed with 1% Triton X-100 as a 100% lysis control . Fluorescence was monitored every 10 min in a plate-based fluorescence spectrometer ( Fluostar , Omega , BMG ) , for a time period of 14 h , under cell culture conditions ( 37°C , 5% CO2 ) . In some wells DNAse I ( 100 U/ml ) was added after 6 h to PMA or ΔdnsΔxds mutant stimulated neutrophils as a control for DNA degradation . Percent NET formation was calculated as percentage of lysis control . Data were presented as percent DNA fluorescence . For live cell microscopy 105 neutrophils were seeded per chamber ( dish diameter: 35 mm; microwell diameter: 14 mm; coverglass: 0 . 16–0 . 19 mm; MatTek Corp . ) and NET formation was stimulated by addition of the respective V . cholerae strain ( MOI 4 ) in presence of the cell impermeant fluorescent DNA dye Sytox Green . Live cell imaging was performed and the movies were recorded by Nikon Eclipse Ti-E , using appropriate filters and a Splan Fluor , ELWD 40× objective ( Nikon ) . The movies and images were analyzed by Nis-Elements AR version 3 . 2 . 0 software . The degradation assay was performed similar to the DNA fluorescence assay described above , with some modifications . Neutrophils were stimulated with PMA ( 100 nM ) prior the addition of V . cholerae to induce a uniformly high level of NET formation . After 6 h , the indicated V . cholerae strain ( MOI 40 ) was added and changes in the fluorescence were monitored every 10 min in a plate-based fluorescence spectrometer ( Fluostar , Omega , BMG ) for another 8 h to allow detection of NET degradation . Immunostaining of neutrophils was performed essentially as described previously [49] . Briefly , 105 neutrophils per well were seeded on glass cover slips coated with 0 . 01% poly-L-lysine in a 24 well plate and stimulated either with PMA ( 100 nM ) , the indicated GFP expressing V . cholerae strain ( MOI 40 ) , or left untreated for 6 h at 37°C and 5% CO2 . Cell were fixed with paraformaldehyde ( 2% ) , blocked ( 3% cold water fish gelatin , 5% fetal calf serum , 1% bovine serum albumin , and 0 . 25% Tween 20 in PBS ) , incubated with the primary antibody anti-neutrophil elastase ( 5 µg/ml , #BM382 , Acris ) in combination with a secondary antibody conjugated to Cy3 ( Jackson Immuno research ) and stained with DAPI ( 4 µg/ml ) to detect DNA . Specimens were mounted in Mowiol 4–88 and microscopy was performed using a Leica HCX PL Apo 40× oil immersion objective ( NA 1 . 25 ) on a confocal microscope ( Leica SP2 , Leica Microsystems ) . For visualization the Leica LAF software was used . To quantify NET formation microscopically a similar method was used as previously described [45] , [49] . At least 9 random images and not less than 300 neutrophils for each condition were analyzed using the DAPI DNA signal . The ImageJ 1 . 46 software was used to calculate pixel areas of the DAPI signal by adjusting the threshold above background . Calculated pixel areas were converted to µm2 . Particles covering an area less than 45 µm2 were excluded from analysis . The average diameter of human neutrophils is 10 µm , therefore the area of neutrophils in an unstimulated stage is approximately 80 µm2 . Signals exceeding 100 µm2 were considered as neutrophils undergoing NET formation , because they were larger than the whole intact cell area , representing either released NETs or cells with decondensed nuclei , which is an essential step prior to NET release . The percentage of NET formation was calculated per image as ( neutrophils undergoing NET formation/total number of neutrophils ) ×100 . A modified version of the assay described by Berends et al . was used to determine the bacterial NET entrapment [24] . Briefly , 5×104 neutrophils per well were seeded in a 96 well plate and stimulated for 6 h with 100 nM PMA ( 37°C , 5% CO2 ) to induce NET formation , followed by addition of the respective V . cholerae strain ( MOI 40 ) . The plate was centrifuged 10 min at 800 g to bring the bacteria in contact with the previously formed NETs . After incubation ( 10 h , 37°C , 5% CO2 ) the supernatants were carefully aspirated and plated for CFU counting . The wells were gently washed with prewarmed medium to remove residual bacteria not entrapped within the NETs . Subsequently prewarmed medium containing DNAse I ( 100 U/ml ) was added and incubated for 15 min at 37°C to dissolve the NETs and release the entrapped bacteria . This fraction was also plated for CFU counting . The percentage of entrapped bacteria was calculated as ( entrapped bacteria/total number of bacteria ) ×100 . The total number of bacteria in the well represents the sum of bacteria in the supernatant and entrapped bacteria . The NET killing assay was essentially performed as previously published using a 96 well plate assay [64] . Briefly , 106 neutrophils per well were seeded in a 96 well plate . Neutrophils were stimulated with PMA ( 100 nM ) and incubated with the indicated V . cholerae strain ( MOI 40 ) . Noteworthy , V . cholerae used in this assay were grown in presence of extracellular DNA to allow induction of nucleases . In some wells DNAse I ( 100 U/ml ) was directly added as a control to allow degradation of NETs and inhibit NET-mediated killing . Furthermore , phagocytic killing was inhibited in some wells by the addition of cytochalasin D ( 100 µg/ml ) 15 min prior to the addition of PMA and bacteria . The plate was centrifuged 10 min at 800 g , to bring bacteria in contact with neutrophils or NETs and incubated 4 h at 37°C and 5% CO2 . Subsequently , all wells having not received DNAse I , were treated with DNAse I for 15 min ( 37°C , 5% CO2 ) to degrade the NETs as well , which results in release of all remaining bacteria for total CFU count in the respective wells as suggested by Menegazzi et al . [65] . Finally , Triton X-100 ( 0 . 01% ) was added to all wells , the well content was passed three times through a 25 gauge needle to disrupt neutrophils as well as clumped NETs and plated for CFU counting . According to Young et al . [64] , neutrophils incapable of killing seemed to enhance bacterial recovery . Hence , “zero killing” was determined by the CFU recovered from control wells consisting of neutrophils treated with DNAse I and cytochalasin D to inhibit NET-mediated and phagocytic killing , respectively . NET-mediated killing was calculated by subtracting the extent of killing in the presence of DNAse I ( i . e . phagocytic killing ) from the extent of killing in absence of DNAse I and cytochalasin D ( total killing ) . For evaluation of the in vivo colonization fitness we used the previously established adult mouse model by Nygren et al . , which results in a stable colonization of the cecum and colon [28] . Briefly , eight to ten weeks old C57BL/6 immunocompetent or neutropenic mice were given Sm ( 5 mg/ml ) in their drinking water , two days prior to infection and kept on Sm-water ( 0 . 2 mg/ml ) after inoculation with the respective V . cholerae strain ( 7×109 to 1010 CFU per mouse ) or left uninfected for mock-inoculated controls . At 24 h or 72 h post infection mice were euthanized and the ceca and colons removed . Relatively large amounts of fluid can accumulate in the ceca of infected mice , which can be released together with bacteria as soon as the cecum is harvested . Thus , cecum samples are not ideal candidates for reproducible quantification of the colonization levels and were consequently only used for histological analysis . Ceca were fixed in 4% buffered formaldehyde , embedded in paraffin , cut in sections ( 5 µm ) and used for histological evaluation . The proximal 1 cm of the ascending colon was used for RNA preparation to determine the inflammatory response . The residual colon was homogenized in LB medium and plated for CFU counting . To induce neutropenia in mice the monoclonal antibody 1A8/Anti-Ly6G mAb ( BioXCell ) was used as described previously [19] , [46] . 16 h prior to oral inoculation with V . cholerae , mice were injected intraperitoneally with 0 . 8 mg of 1A8 or not injected in case of the immunocompetent mice . A decrease of neutrophils by at least 2-fold compared to immunocompetent mice was considered as neutropenia , which was confirmed by analyses of peripheral blood smears by trained medical staff at the time point of inoculation as well as 72 h post infection . For immunostaining , specimens were processed similarly as described previously [66] . Briefly , samples were deparaffinized , rehydrated in decreasing concentrations of EtOH , and subjected to antigen retrieval by cooking in 10 mM citrate buffer , pH 6 . 0 , for 10 min . Specimens were blocked with 2% BSA and 36 µl/ml mouse Ig blocking reagent ( Vector Laboratories ) in PBS/0 . 1% Triton for 1 h at room temperature . For visualization of neutrophils , a primary antibody directed against MPO ( A0398 , Dako ) diluted in blocking solution was applied over night at 4°C . Additionally , chromatin was detected using a primary antibody directed against histone H1 ( clone AE-4 , Acris Antibodies ) . Primary antibodies were detected with Alexa Fluor 568- and 488-conjugated secondary antibodies ( Life Technologies ) diluted in 2% BSA in PBS/0 . 1% Triton , respectively . DNA was visualized with DAPI ( Life Technologies ) and slides were mounted with fluorescence mounting medium ( Dako ) . Images were captured with a C1 confocal microscope ( Nikon Instruments ) at 60× magnification and are presented as maximum intensity projections from Z-stacks . To measure gene expression of the two extracellular nucleases of V . cholerae in presence of NETs , 106 neutrophils per well were seeded in a 24 well plate , stimulated with PMA ( 100 nM , 4 h ) , followed by incubation for 6 h with V . cholerae wild type strain ( MOI 40 ) . In case of DNA as a stimulus , V . cholerae wild type strain was incubated for 6 h with 2 . 5 mg/ml herring sperm DNA ( Sigma ) . V . cholerae incubated in absence of neutrophils or extracellular DNA served as a control . Bacterial RNA was extracted using the RNeasy Mini Kit ( Qiagen ) and chromosomal DNA was removed using RQ1 RNAse-Free DNase ( Promega ) according to the manufacturer's protocol . To measure the induction of inflammatory gene expression upon V . cholerae infection , the proximal 1 cm of the ascending colon was collected and homogenized with a Power Lyzer24 ( Mobio ) in 1 ml Trizol . For DNA-mediated induction of gene expression , Tlr4−/− murine macrophages ( isolation see above ) were stimulated with V . cholerae wild type genomic DNA ( 2 . 5 µg/ml ) for 12 h or human neutrophils ( 106 ) were stimulated with V . cholerae ΔmsbB genomic DNA ( 2 . 5 µg/ml ) for 8 h using a 24 well plate , respectively . Incubation with DNAse I digested wild type or ΔmsbB genomic DNA ( 2 . 5 µg/ml ) served as control conditions . After the stimulation time Tlr4−/− macrophages or human neutrophils were resuspended in Trizol , RNA was extracted using chloroform extraction and precipitated with isopropanol . The cDNA synthesis was performed with an iScript Select cDNA Synthesis Kit ( Bio-Rad ) using 200 ng bacterial RNA or 1 µg mouse/human RNA . Quantitative RT-PCR was performed with SYBR GreenER qPCR SuperMix for ABI PRISM instrument ( Invitrogen ) utilizing a Rotor-Gene 600 and Rotor-Gene 600 Series Software 1 . 7 ( GenXpress ) according to the manufacturer's instructions . Each reaction contained primers ( 400 ng ) and template ( 10 ng for bacterial cDNA as well as 50 ng for mouse and human cDNA ) and was tested in triplicate . The sequences of the primers used for qRT-PCR are listed in Table S2 , labeled as x_fw and x_rv , in which x stands for the respective gene . For each sample , the mean cycle threshold of the test transcript was normalized to the housekeeping gene 16S rRNA ( bacterial samples ) or 36B4 ( mouse and human samples ) , also known as RPLP , and to one randomly selected control sample . Unless stated otherwise the data is presented as the median with interquartile range . Data were analyzed using the Mann-Whitney U test for single comparisons or Kruskal-Wallis test followed by post-hoc Dunn's multiple comparisons . Differences were considered significant for P values of <0 . 05 . For all statistical analyses the GraphPad Prism 4 . 0a software was used .
|
Although several reports describe an inflammatory component of the diarrheal disease cholera , the innate immune response to V . cholerae and its impact on the pathogenesis of the disease is poorly understood . In the present study we can link the presence of host neutrophils with a colonization defect of a V . cholerae mutant deleted for both extracellular nucleases , Dns and Xds . Neutrophils can be seen as a first line of defense of the innate immunity and can effectively entrap and kill pathogens in neutrophil extracellular traps ( NETs ) . Herein we show for the first time that V . cholerae induces NET formation , but effectively uses its two extracellular nucleases to degrade NETs and evade from this innate immunity weapons . Interestingly , we recently characterized the two extracellular nucleases as modulators of extracellular DNA during biofilm formation , which is rather associated with environmental lifestyle of this facultative human pathogen in aquatic ecosystems . Thus , V . cholerae seems to utilize the activity of the extracellular nucleases under both stages of its lifecycle , inside the host as a defense mechanism against NETs and during biofilm formation in the environment .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"gram",
"negative",
"immunity",
"innate",
"immunity",
"microbial",
"pathogens",
"host-pathogen",
"interaction",
"biology",
"microbiology",
"bacterial",
"pathogens"
] |
2013
|
Vibrio cholerae Evades Neutrophil Extracellular Traps by the Activity of Two Extracellular Nucleases
|
Enterovirus 71 ( EV71 ) causes life-threatening epidemics in Asia and can be phylogenetically classified into three major genogroups ( A∼C ) including 11 genotypes ( A , B1∼B5 , and C1∼C5 ) . Recently , EV71 epidemics occurred cyclically in Taiwan with different genotypes . In recent years , human studies using post-infection sera obtained from children have detected antigenic variations among different EV71 strains . Therefore , surveillance of enterovirus 71 should include phylogenetic and antigenic analysis . Due to limitation of sera available from children with EV71 primary infection , suitable animal models should be developed to generate a panel of antisera for monitoring EV71 antigenic variations . Twelve reference strains representing the 11 EV71 genotypes were grown in rhabdomyosarcoma cells . Infectious EV71 particles were purified and collected to immunize rabbits . The rabbit antisera were then employed to measure neutralizing antibody titers against the 12 reference strains and 5 recent strains . Rabbits immunized with genogroup B and C viruses consistently have a lower neutralizing antibody titers against genogroup A ( ≧8-fold difference ) and antigenic variations between genogroup B and C viruses can be detected but did not have a clear pattern , which are consistent with previous human studies . Comparison between human and rabbit neutralizing antibody profiles , the results showed that ≧8-fold difference in rabbit cross-reactive antibody ratios could be used to screen EV71 isolates for identifying potential antigenic variants . In conclusion , a rabbit model was developed to monitor antigenic variations of EV71 , which are critical to select vaccine strains and predict epidemics .
Enterovirus 71 ( EV71 ) is a non-enveloped RNA virus of the family Picornaviridae and contains a positive sense ssRNA with a single open reading frame ( ORF ) . The ORF is expressed as a large polyprotein that can be cleaved into P1 , P2 and P3 regions . The P1 gene encodes four structural proteins VP1–VP4 , while P2 and P3 genes encode the non-structural proteins responsible for virus replication and virulence [1] . The viral icosahedral capsid is composed of 60 identical units that consist of VP1–VP4 structural proteins [2] , [3] . Variation of capsid proteins , except VP4 , is responsible for the antigenic diversity among the enteroviruses , while neutralizing epitopes and phylogenetic classification are mainly based on VP1 and VP2 [4]–[7] . According to analysis of VP1 sequences , EV71 was phylogenetically divided into three distinct genogroup: A , B , and C [8] , [9] . Genogroups B and C can be further divided into genotypes B1–B5 and C1–C5 , respectively [10] . Recently , genogroup D was identified in India and genogroups E and F were identified in Africa [11] , [12] . Genogroup A composes of the prototype EV71 strain ( BrCr-CA-70 ) which was isolated in 1970 in the United States but had not been detected afterwards until 2008 . In contrast , genogroup B and C viruses have been causing large scale of epidemics in Asia since 1997 and are targeted for vaccine development [10] , [13] . Most EV71 infections manifest as mild cases of hand-foot-mouth disease ( HFMD ) or herpangina in young children , who are potentially at risk for severe neurological and cardiopulmonary complications [8] , [9] . The neurovirulence of EV71 first came to people's attention in California in 1969 [14] . Since then , EV71 has caused several outbreaks sporadically in the 1970s , i . e . 1975 in Bulgaria , 1978 in Hungary [15] , [16] . Since 1997 , EV71 has been further identified as the causative agent responsible for the epidemics of central nervous system disease occurring in Asia-Pacific countries [9] , [17] . In Taiwan , phylogenetic analyses revealed that different predominant genotypes occurred in 1998 ( C2 ) , 2000–2001 ( B4 ) , 2004–2005 ( C4 ) , and 2008 ( B5 ) [10] , [18] . This genotype replacement has also been observed in Malaysia and Vietnam [10] , [19] , [20] . Therefore , continuous monitoring genetic and antigenic evolution of EV71 are critical to vaccine development and epidemic control . Although EV71 has one single serotype as measured using hyper-immune animal sera , recent human studies using post-infection sera obtained from children to measure cross-neutralizing antibody titers against different genotypes have detected antigenic variations among different EV71 strains [21]–[24] . Due to the limitation of small amount of sera available from young children with EV71 primary infection , suitable animal models should be developed to generate a panel of antisera for monitoring EV71 antigenic variations . In the present study , 12 reference viruses representing the eleven EV71 genotypes were collected and purified to immunize rabbits for generating EV71-specific rabbit antisera . Then , cross-reactive neutralizing antibody titers among the 12 reference viruses and sera were measured for evaluating antigenic changes . Finally , 5 recent circulating strains in Taiwan were analyzed genetically and antigenically using the 12 reference antisera .
The animal protocol was approved by the NHRI Institutional Animal Care and Use Committee ( approval no . NHRI-IACUC-100008-A ) following the Institutional Animal Care and Use Committee Guidebook published by the US Office of Laboratory Animal Welfare ( http://grants . nih . gov/grants/olaw/guidebook . pdf ) . Twelve reference viruses representing 11 EV71 genotypes were collected and used in the present study . Eight of these viruses were isolated in Taiwan and the other four viruses ( genotype A , B2 , B3 and C3 ) were isolated in other countries ( Table 1 ) [24] . Besides , five recent strains include 4 strains ( one B5 , one C2 and two C4 ) isolated in 2011 and 2012 and one C2-like virus ( C2L-101-08 ) which was isolated in 2008 and identified as an antigenic variant using post-infection sera obtained from children in a previous study ( Table 2 ) ( Figure 1 ) [22] . Viral isolation methods had been described previously [22] . All viruses were amplified in rhabdomyosarcoma ( RD ) cells using Dulbecco's Minimum Essential Medium ( DMEM ) containing 2% v/v fetal bovine serum and penicillin/streptomycine . The 50% tissue culture infective dose ( TCID50 ) of virus was calculated in RD cells using the Reed-Muench method . Total viral RNA was extracted from culture supernatant by QIAamp Viral RNA Kit ( Qiagen ) according to the manufacturer's instruction . The reverse transcription polymerase chain reaction ( RT-PCR ) and sequencing were performed as previously described [24] . The P1 regions ( 2586 bp ) of the 12 reference and five recent circulating strains were aligned and analyzed using a Clustal W in MEGA 4 . Identification and genotyping was carried out using phylogenetic analysis , which was conducted with 1 , 000 replications of bootstrap analyses using a Neighbor-Joining model and the prototype coxsackievirus A16 ( CA16 ) G-10 strain was used as an outgroup virus [25] . It is well known that two forms of EV71 viral particles , full and empty particles , existed during propagation in cells [3] , [26] . Based on historical poliovirus studies , the full particles are infectious and immunogenic but the empty particles are not [27] . Therefore , we purified EV71 infectious ( full ) particle of the reference viruses for rabbit immunizations . The EV71 culture supernatant was concentrated 10-fold with a Amicon 100K centrifugal filter ( Millipore ) . The crude virus concentrate was loaded onto a 15–65% continuous sucrose gradient and centrifuged at 28000 rpm for 4 hr . Fractions ( 2 mL per fraction ) were collected and the viral titer and protein concentration of each fraction were determined by TCID50 and BCA assays ( Thermo Scientific ) , respectively . Fractions with high infectious virus titers in 32–38% sucrose concentration were merged and concentrated by diafiltration using Amicon 100K centrifugal filter and centrifugation at 3500 g . The purified EV71 viruses were further verified by using Western blot and electron microscopy analysis . Purified EV71 particles were deposited on a carbon-coated 200 mesh copper grid for 1 min at room temperature . The excess sample was removed by filter paper and then the copper grid was stained with 2% phosphotungstic acid solution for 1 min , which was then removed by filter paper . The stained grid was dried for 1 day at room temperature and observed under a JEM 1200EX transmission electron microscopy [26] . Two-month-old New Zealand White rabbits were immunized three times subcutaneously with purified infectious EV71 viruses ( 1×107 TCID50 in 1 mL PBS ) . The immunized rabbits were bled one week after the final boost , and the sera were collected and stored at −20°C for further analysis . The animal protocol was approved by the NHRI IACUC . Serum neutralizing antibody titers were detected using TCID50 assay according to the standard protocol [28] . Serum samples from immunized rabbits were inactivated for 30 min at 56°C , and then diluted two-fold serially in DMEM with an initial dilution of 1∶8 . Fifty µl of diluted sera and 100 TCID50 viruses were added to 96-well microplates and incubated at 37°C for 1 hr . Later , 100 µl of RD cell suspension containing 3×104 cells was added , and cytopathic effect ( CPE ) was observed in the inverted microscopic after an incubation at 37°C for 3–4 days . The neutralization titers were defined as the highest dilution that could result in a >50% reduction in the CPE . Each test sample was run simultaneously with positive serum control , cell control and virus back-titration . Antigenic cartography is a way to visualize and increase the resolution of serological data [21] , [24] . In an antigenic map , the distance between a serum point S and antigen point A corresponds to the difference between the log2 of the maximum titer observed for serum S against any antigen and the log2 of the titer for serum S and antigen A . Thus , each titer in a neutralization assay can be thought of as specifying a target distance for the points in an antigenic map . In this study , an antigenic map was generated using a web-based analytic tool [29] . Recently , 3-dimensional ( 3-D ) structures of P1 protein of two EV71 virus strains ( genotype B3 and genotype C4 ) were generated using X-ray crystallography ( PDB 4AED and PDB 3VBS ) [2] , [3] . The 3-D structure of the genotype C4 virus was based on infectious particles so it was employed as template to locate specific amino acid positions using the RasMol software ( http://rasmol . org/ ) . To compare human and rabbit cross-reactive neutralizing antibody profiles , cross-reactive antibody ratios between homotypic and heterotypic antibody titers were calculated because measurement of neutralizing antibody titers tends to be highly variable and cross-reactive antibody ratios won't be affected by absolute antibody titers . In our previous study , 21 sera collected from children infected with EV71 genotype C2-1998 ( 6 sera ) , C4-2005 ( 2 sera ) , C4-2010 ( 3 sera ) , B4-2002 ( 5 sera ) and B5-2008 ( 5 sera ) viruses were available to measure cross-reactive neutralizing antibody titers against all 12 reference viruses [24] . Serum antibody titers of children infected with EV71 in the same year were merged to calculate geometric mean titers . Therefore , 55 cross-reactive antibody ratios between homotypic and heterotypic antibody titers were available from human serology data . For rabbit serology data , there were 132 cross-reactive antibody ratios among 12 reference viruses . After combining human and rabbit serology data , 55 cross-reactive antibody ratios were available to evaluate the correlation between human and rabbit cross-reactive antibody profiles . Differences between homologous and heterologous neutralizing antibody titers were tested for statistical significance by using Student's t-test with log2-transformed data . The P value<0 . 05 is taken to indicate statistically significance .
Twelve EV71 reference viruses representing 11 EV71 genotypes were collected ( Table 1 ) and their P1 genes were sequenced to confirm their genotypes using phylogenetic analysis ( Figure 1 ) . Two genotype C4 viruses were selected due to high diversity existing in this genotype . These 12 reference viruses were amplified in RD cells and purified by sucrose gradient ultracentrifugation . After ultracentrifugation , all fractions of sucrose gradients were collected for quantification of sucrose concentration , measurement of infectious virus titers and detection of viral proteins using Western blot analysis . The fractions with the highest virus titers usually located at fractions with 32–38% sucrose and they were merged for further analysis ( F32–38 ) and used to immunize rabbits for generating antisera . In addition , several fractions with low virus titers but high concentration of viral proteins usually located at fractions with 24–30% sucrose and they were also merged for further analysis ( F24–30 ) . In electron microscopy analysis , the viruses were purified by second 15–65% continuous sucrose gradient ultracentrifugation and empty particles ( F24–30 ) and full particles ( F32–38 ) were observed in negative staining ( Figure 2 ) . Some previous studies had demonstrated that two different types ( full and empty types ) EV71 particles were separated by using sucrose gradient ultracentrifugation and contained different protein conformations [3] , [26] . To generate rabbit antisera for determining cross-reactive neutralizing antibody profiles among EV71 genotypes , rabbits were immunized with the full particles of the reference viruses and these antisera were collected for measuring cross-reactive neutralizing antibody titers against 12 reference strains and 5 recent strains ( Table 2 ) . The 5 recent strains included 4 strains ( one B5 , one C2 and two C4 ) isolated in 2011 and 2012 and one C2-like virus ( C2L-101-08 ) which was isolated in 2008 and identified as an antigenic variant using post-infection sera obtained from children in a previous study ( Table 2 ) ( Figure 1 ) [22] . As shown in Table 2 , the 12 reference EV71 viruses induced high homotypic neutralization titers ( 1∶256 to 1∶4096 ) . Regarding to cross-reactive antibody responses , genotype A virus ( A-70 ) consistently has >8-fold difference between homotypic and heterotypic neutralizing antibody titers but no clear pattern could be identified for genogroup B and C viruses . Interestingly , genotype B2 and B5 viruses seems to be highly immunogenic and could induce high homotypic and heterotypic neutralizing antibody titers against all genogroup B and C viruses except the C2-like virus isolated in 2008 . Among the 5 recent strains , the C2-like virus is an antigenic outlier and the other 4 recent viruses were antigenically similar to their homotypic reference viruses . To better visualize the cross-reactive serological data , antigenic cartography was further employed to analyze the serological data and showed that genotype A virus and the C2-like virus could be antigenically differentiated from other EV71 viruses ( Figure 3 ) , which are consistent to previous human studies . In our previous study , 21 sera collected from children infected with EV71 genotype C2-1998 ( 6 sera ) , C4-2005 ( 2 sera ) , C4-2010 ( 3 sera ) , B4-2002 ( 5 sera ) and B5-2008 ( 5 sera ) viruses were available to measure cross-reactive neutralizing antibody titers against all 12 reference viruses [24] . Serum antibody titers of children infected with the same EV71 genotype were merged to calculate geometric mean titers ( GMTs ) . Therefore , 55 cross-reactive antibody ratios between homotypic and heterotypic GMTs were available from human serology data . For rabbit serology data , there were 132 pair-wise cross-reactive antibody ratios among the 12 reference viruses . After combining human and rabbit serology data , 55 cross-reactive antibody ratios were available to evaluate the correlation between human and rabbit cross-reactive antibody profiles . Scatter plot between human and rabbit cross-reactive antibody ratios are shown in Figure 4 , which shows that cross-reactive antibody ratios ranged from 1 to 7 in human data and from 1 to 256 in rabbit data . Overall , the cross-reactive antibody ratios calculated using human and rabbit serology data correlate to each other ( correlation coefficient R = 0 . 63 , P<0 . 01 ) ( Figure 4 ) . Based on influenza studies , ≧50% differences in GMT of cross-reactive antibody titers in humans ( i . e . ≧2 in cross-reactive antibody ratios ) may cause decreased vaccine efficacy [30] . Using the same criteria for human EV71 serology data , 18 cross-reactive antibody ratios were ≧2 and 16 of them ( 89% ) could be identified by using ≧8-fold differences in rabbit cross-reactive antibody ratios . Therefore , ≧8-fold difference in rabbit cross-reactive antibody ratios could be used to screen EV71 isolates for identifying potential antigenic variants . Then , the potential EV71 antigenic variants identified using rabbit antisera would be further verified using post-infection sera obtained from children . Since two antigenic variants ( A-70 and C2L-101-08 ) were identified , we further tried to identify amino acid positions related to the observed antigenic variations . As shown in Table 3 , amino acid diversity rates were 0% , 8% ( 21/254 ) , 3% ( 8/242 ) , and 8% ( 25/297 ) for VP4 , VP2 , VP3 and VP1 regions , respectively . Five amino acid residues ( VP2-143N , VP1-18K , VP1-116H , VP1-167D , and VP1-275S ) are specific signatures for A-70 virus but no amino acid residue is specific for C2L-101-08 virus ( Table 3 ) . We further located these 5 residues in the 3-D structure of P1 polyprotein . As shown in Figure 5 , 4 of the 5 residues are located on the surface and 3 ( VP1-116 , VP1-275 and VP2-143 ) of them are located in or near to neutralizing epitopes previously identified using mouse monoclonal antibodies ( VP1 211-225 and VP2 136-150 ) [7] . Although no clear pattern could be identified by analyzing cross-reactive antibody profiles between genogroup B and C viruses , six amino acid residues ( VP2-45D , VP2-126I , VP3-234H , VP3-240S , VP1-43E and VP1-58T ) and seven amino acid residues ( VP2-45S , VP2-198I , VP2-200A , VP3-100L , VP3-232A , VP3-239G and VP1-240T ) were identified to be specific signatures for genogroup B and C viruses , respectively ( Table 3 ) .
Although EV71 has one single serotype as measured using hyper-immune animal antisera , antigenic variations of EV71 have been identified in several human studies using post-infection sera obtained from children [22]–[24] , [31] . Since it is hard to collect large amount of children sera for measuring cross-reactive neutralizing antibody titers , it would be desirable to establish an animal model for monitoring EV71 antigenic variations . In this study , a virus purification platform was established to purify EV71 infectious virus particles which were then used to generate rabbit antisera for measuring cross-reactive neutralizing antibody titers against reference and recent EV71 strains . The cross-reactive neutralizing antibody profiles defined using rabbit antisera were similar to those observed using post-infection sera obtained from children [23] , [24] , [32] . Moreover , ≧8-fold differences of cross-reactive antibody titers measured using rabbit antisera could be used as a screening criterion to select EV71 isolates for further evaluation using post-infection sera obtained from children , similar to ferret antisera used for global influenza surveillance [33] , [34] . Two types ( empty and full ) of EV71 particles were produced in cell cultures and they could be separated using sucrose gradient ultracentrifugation in our study , which are similar to those observed in previous EV71 and poliovirus studies [3] , [26] , [35] . Based on historical poliovirus studies , full particles but not empty particles are highly immunogenic to induce neutralizing antibody responses so vaccine potency assay is based on the quantification of full particles [36] . Therefore , we used the EV71 full particles to generate rabbit antisera for monitoring antigenic variations . Based on the cross-reactive neutralizing antibody profiles measured using the rabbit antisera , we identified two antigenic outliers ( A-70 and C2L-101-08 ) . The genotype A virus was first isolated in California in 1970 and disappeared for 38 years until 2008 . Although genotype C4 viruses have been the predominant circulating viruses for the last 7 years in China [10] , genotype A viruses have been sporadically detected in central ( Anhui and Hubei ) , northern ( Beijing ) and western ( Yunnan ) China since 2008 , which indicates widespread of genotype A viruses in China [13] , [37] . Currently , genotype C4 and B4 viruses were used to develop EV71 vaccines in China and Taiwan , respectively and it would be critical to collect post-vaccination sera obtained from children to measure cross-reactive neutralizing antibody titers against the circulating genotype A viruses . Moreover , mechanism of the reemergence of genotype A viruses in China is not clear and could be clarified through genomics studies . The novel C2-Like virus ( C2L-101-08 ) was isolated in 2008 , Taiwan . Children infected with genotype B4 , B5 , C4 , and C5 viruses showed a maximum of 128-fold decrease in cross-reactive neutralizing antibody titers against C2L-101-08 compared with those of homogenous viruses . In the present study , all 12 rabbit antisera displayed low cross-reactive neutralizing antibodies ( ≦8-fold ) against C2L-101-08 strain ( Table 2 ) and the results were similar with post-infection children serological data [38] . It is indicated that C2L-101-08 strain exhibited evident antigenic diversity from other genotypes . Interestingly , the C2-like viruses were a minor group ( 2 strains/989 strains ) in 2008 in Taiwan and have never been detected globally since 2009 to 2013 [22] , [38] , which indicate that the C2-like viruses do not have evolutionary advantages or become highly successful viral strain causing no/mild symptoms in humans . Surprisingly , no amino acid signature in the C2-like virus could be identified to be related to the observed antigenic variations . Several previous studies had demonstrated that some EV71 viruses were aggregated strains and they were poorly neutralized by anit-EV71 serum in a common neutralizing assay condition [39] , [40] . In the present study , C2-like virus was treated with trypsin or chloroform for deaggregation but it still cloud not be neutralized by all 12 rabbit antisera . In addition , we tried to use the C2-like virus to generate rabbit antisera but did not succeed . Comprehensive genomic analysis and reverse genetics would be required to identify molecular determinants of the observed antigenic variations between the C2-like viruses and other EV71 viruses [41] , [42] . In addition to the C2-like virus , four recent circulating strains including one genotype B5 ( B5-3172-11 ) , one genotype C2 ( C2-9552-12 ) , and two genotype C4 ( C4-3591-11 and C4-0184-12 ) viruses isolated in Taiwan were also evaluated using the rabbit antisera in this study . The B5-3172-11 strain caused nation-wide epidemic in 2012 and was likely imported from Xiamen , China in 2011 . In addition , the two recent genotype C4 viruses were also closely related to the C4 viruses circulating in China ( Figure 1 ) . Based on enterovirus surveillance in northern Taiwan , no genotype C2 virus was isolated in 2009–2011 and the C2-9552-12 virus was the only C2 strain isolated in 2012 and is phylogenetically related to the C2 strains isolated in Canada , France , Netherland , and Singapore in recent years ( Figure 1 ) . Based on rabbit cross-reactive neutralizing antibody profiles , the recent C2 , C4 and B5 viruses did not antigenically significantly differ from their homotypic viruses . Interestingly , the C4 viruses were sporadically detected in 2010∼2011 in Taiwan but the B5-3172-11-like viruses were first detected in late 2011 and caused nation-wide epidemics in early 2012 . It is likely the B5-2011 viruses had evolutionary advantages in replication and transmission efficiency in humans but did not cause antigenic drifts . Suitable EV71 animal transmission models would be desirable to elucidate the mechanism . In the antigenic analysis using the 12 reference viruses and antisera , B2-86 , B3-97 , B4-02 and B5-08 viruses are in the same genogroup and have similar P1 sequences ( Table 3 ) . However , the B3-97 antisera could not efficiently neutralize the B2-86 , B4-02 and B5-08 viruses ( Table 2 ) . Interestingly , the B2-86 , B4-02 and B5-08 antisera could well neutralize the B3-97 virus . Comparing amino acid sequences of genogroup B in Table 3 , one amino acid residue ( VP1-145G ) is the specific signature for B3-97 virus and it may be related to the observed antigenic variation , which is consistent to findings of other studies [38] , [41] . In a recent study , a genotype C4 virus was used as an immunogen to generate 186 monoclonal antibodies ( MAbs ) and the MAbs with high neutralizing antibody titers were purified for antigenic analysis of eighteen EV71 clinical isolates [31] . The results showed that even EV71 strains in the same genotype do not generally produce similar antigenic profiles . These results indicate that the current genotyping of EV71 did not reflect their antigenicity , which is consistent with our study using rabbit antisera . However , the mouse monoclonal antibody profiles were not compared with human data . In contrast to the mouse monoclonal antibody study , our study showed that rabbit cross-reactive neutralizing antibody profiles are similar to the profiles measured using post-infection sera obtained from children . Since human antibody cross-reactive antibody profiles were polyclonal responses , it would be more suitable to use rabbit antisera other than mouse monoclonal antibody to monitor EV71 antigenic variations . Due to significant impacts in public health , five organizations in Asia are developing EV71 vaccines using vaccine strains representing B2 , B4 , and C4 genotypes [10] . As shown in the present study and several previous studies , significant antigenic variations could be detected among different EV71 strains , especially the genotype A virus [10] , [24] , [43] . Currently , antigenic analysis would not be regularly conducted in the enterovirus 71 surveillance system . The rabbit model developed in the present study could be readily integrated into the national enterovirus surveillance system to monitor EV71 antigenic variations .
|
Enterovirus 71 ( EV71 ) has caused several life-threatening epidemics in children in the Asia-Pacific region since 1997 . EV71 has one single serotype as measured using hyper-immune animal antisera but can be phylogenetically classified into three major genogroups ( A , B and C ) and eleven genotypes ( A , B1–B5 , and C1–C5 ) . Recently , epidemiological studies in the Asia-Pacific region have found that large-scale EV71 epidemics occurred cyclically with different genotypes . This observation of genotype replacement , in conjunction with the observed antigenic variations among different EV71 genogroups in human studies , has kindled the interest to establish animal models to monitor the antigenic variations of EV71 . In this study , we developed a rabbit model to monitor antigenic variations of EV71 , which could be further integrated into national enterovirus surveillance systems .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"infectious",
"diseases",
"medicine",
"and",
"health",
"sciences",
"emerging",
"infectious",
"diseases",
"epidemiology",
"biology",
"and",
"life",
"sciences",
"microbiology",
"viral",
"diseases"
] |
2014
|
Monitoring Antigenic Variations of Enterovirus 71: Implications for Virus Surveillance and Vaccine Development
|
DNA cytosine methylation is involved in the regulation of gene expression during development and its deregulation is often associated with disease . Mammalian genomes are predominantly methylated at CpG dinucleotides . Unmethylated CpGs are often associated with active regulatory sequences while methylated CpGs are often linked to transcriptional silencing . Previous studies on CpG methylation led to the notion that transcription initiation is more sensitive to CpG methylation than transcriptional elongation . The immunoglobulin heavy chain ( IgH ) constant locus comprises multiple inducible constant genes and is expressed exclusively in B lymphocytes . The developmental B cell stage at which methylation patterns of the IgH constant genes are established , and the role of CpG methylation in their expression , are unknown . Here , we find that methylation patterns at most cis-acting elements of the IgH constant genes are established and maintained independently of B cell activation or promoter activity . Moreover , one of the promoters , but not the enhancers , is hypomethylated in sperm and early embryonic cells , and is targeted by different demethylation pathways , including AID , UNG , and ATM pathways . Combined , the data suggest that , rather than being prominently involved in the regulation of the IgH constant locus expression , DNA methylation may primarily contribute to its epigenetic pre-marking .
DNA methylation is a common epigenetic regulation mechanism in vertebrates and is involved in gene expression regulation during development and differentiation as well as in defense of the genome against transposable elements . DNA methylation provides a robust epigenetic mechanism for cell fate decisions , cell identity and tissue homeostasis . The importance of this epigenetic regulation is highlighted by the finding that its absence is lethal and aberrant DNA cytosine methylation is often associated with disease such as cancer [1] . Mammalian genomes are predominantly methylated at cytosines in the context of CpG dinucleotide . Mammalian genomes are mostly CpG-poor and these CpG motifs are globally methylated . However , a minority of CpGs occur in CpG-dense regions called CpG islands ( CGIs ) and are generally refractory to DNA methylation . While unmethylated CpG sites and CGIs are generally associated with active promoters , methylated CpGs ( mCpGs ) and mCGIs are closely associated with transcriptionally silent promoters . This pattern is less obvious when it comes to transcription elongation as mCpGs and mCGIs in gene body did not block elongation , leading to the notion that it is transcription initiation that is more sensitive to cytosine methylation [2–4] . B lymphocytes are derived from pluripotent hematopoietic stem cells and develop in fetal liver during embryonic development , then shift to the bone marrow around birth [5] . B cell development requires assembly of its antigen receptor loci through V ( D ) J recombination which occurs in developing B cells in fetal liver and bone marrow [6 , 7] . Further development leads to migration to peripheral lymphoid organs such as the spleen where , upon antigen encounter , mature B cells can undergo another recombination process called class switch recombination ( CSR ) . CSR enables IgM-expressing B cells to switch to the expression of other antibody classes , specified by different constant genes . Each constant gene is part of a transcription unit where transcription , termed germline ( GL ) transcription , initiates at an inducible promoter ( called I promoter ) and terminates downstream of the constant exons [8] . GL transcription is associated with various induced epigenetic changes ( e . g . [9 , 10] ) and is controlled by different cis-regulatory elements including enhancers and insulators ( e . g . [11–14] ) . In particular , the 3′ regulatory region ( 3′RR ) , which contains four enhancers located downstream of the IgH locus , effects a long-range enhancing activity on the multiple I promoters [15] . While V ( D ) J recombination targets all antigen receptor loci in B and T lymphocytes [7] , CSR is strictly B-cell specific and targets exclusively the immunoglobulin heavy chain ( IgH ) locus [8] . This highly restricted targeting raises important developmental questions . For instance , it is still unknown whether all the epigenetic features of the IgH constant locus are acquired de novo in the B cell lineage and at the right B cell developmental stage , i . e . when GL transcription occurs , or whether the locus is at least in part epigenetically pre-marked . Here , we focused on DNA methylation and used bisulphite sequencing to analyze the methylation profiles of multiple cis-acting elements at the IgH constant locus . We show that the methylation patterns of most cis-acting elements are established and faithfully maintained independently of B cell activation or GL transcription . Moreover , one I promoter , but not enhancers , was hypomethylated early during ontogeny and recruited different demethylation pathways .
Splenic B cells can be activated by various extracellular signals ( mitogen , cytokines… ) . Each stimulation condition induces a specific ( set of ) I promoter ( s ) and directs CSR to the corresponding constant gene ( s ) [8] . We checked induction of GL transcription and as expected , RT-qPCR and FACS revealed high levels of GL transcripts and robust CSR upon appropriate stimulation ( S1 Fig ) . To analyze methylation profiles of I promoters , we used bisulphite sequencing . Because this technique does not discriminate 5-methylcytosine from 5-hydroxymethylcytosine , a fraction of methylated cytosines may include 5-hydroxymethylcytosines . Conversely , a fraction of unmethylated cytosines may include 5-carboxylcytosines and 5-formylcytosines . Throughout this study , we did not quantify the levels of the oxidized methylcytosines . In order to determine if and how CpG methylation patterns are affected upon induction of GL transcription , we first compared the methylation state of all CpGs at I promoters and flanking sequences ( Fig 1A ) , in resting and activated splenic B cells . We focused on the promoters’ CpGs to establish the link between DNA methylation and transcription initiation , but we also analyzed I exons and different constant exons as sites of transcriptional elongation . Analysis of some CpGs upstream of the promoters , located outside the transcription units and the known regulatory regions , served as “negative controls” as we anticipated them to be hypermethylated ( S2 Fig and S3 Fig ) . Inspection of the data revealed various unexpected aspects of CpG methylation in the IgH constant locus . In particular: Most CpGs upstream of the promoters were heavily methylated in resting B cells and remained so after activation ( Fig 1B–1E and S3 Fig ) . Strikingly , some promoters’ CpGs , notably the unique CpG at Iγ3 ( see discussion ) , three CpGs at Iγ2b , and one CpG at Iα promoters , were fully unmethylated in resting B cells ( Fig 1B and 1E ) . At the promoters , there was no obvious correlation between promoter activation and CpG demethylation ( Fig 1B–1E ) , except for the Iγ1 promoter’s unique CpG , which lost all methylation upon IL4 activation ( Fig 1D ) . The nature of the stimulus did not alter the CpG methylation state of Iγ2b promoter as a similar pattern was observed following either LPS or TGFβ stimulation , which both activate this promoter ( Fig 1B and 1E ) . A positive correlation between induction of GL transcription and CpG demethylation could be seen for specific , mostly proximal , CpGs at Iγ3 , Iγ1 , Iγ2b and Iγ2a exons ( hereafter Iγ exons ) . In contrast , the CpGs of Iε and ( more markedly ) Iα exons remained hypermethylated ( Fig 1B–1E ) . CpG methylation status of all constant exons studied ( Cγ3 , Cγ1 , Cγ2b , and Cα ) was unchanged upon appropriate activation ( Fig 1B , 1D and 1E ) . The targeting of CpGs for ( de ) methylation is highly focused , i . e . , there is no evidence for spreading of this epigenetic mark as best illustrated by the hypomethylated CpG of Iα promoter ( Fig 1E and S3 Fig ) ( see below ) . In order to determine whether CpG demethylation occurs as a consequence of B cell activation or whether it is a direct consequence of GL transcription per se , we investigated CpG methylation in genetic contexts where Iγ3 and Iγ2b promoters were silenced in activated B cells , or constitutively active in resting B cells . In ZILCR mouse line , the chicken β-globin core insulator was inserted upstream of the 3’RR , resulting in a complete silencing of Iγ3 and Iγ2b promoters upon LPS stimulation ( Braikia and Khamlichi , in preparation ) . In the second mouse model , the 5’hs1RI CTCF insulator within the Cα constant gene was deleted , leading to constitutive activity of Iγ3 and Iγ2b promoters in resting B cells [13] ( Fig 2A and 2B ) . The unmethylated state of Iγ3 and Iγ2b promoters remained essentially unchanged in LPS-activated ZILCR B cells ( Fig 2A ) , and in unstimulated 5’hs1RI splenic B cells ( Fig 2B ) . In LPS-activated ZILCR B cells , the methylation pattern of Iγ3 and Iγ2b exons was comparable to that seen in WT resting B cells ( Fig 1B and Fig 2A ) . When Iγ3 and Iγ2b promoters were active in the absence of B cell activation , a lack of methylation was seen at exons Iγ3 and Iγ2b that was globally similar to that in LPS-activated WT B cells ( Fig 1B and Fig 2B ) . Taken together , the data from WT and mutant splenic B cells demonstrate that the unmethylated state of Iγ3 and Iγ2b promoters is locally established prior to B cell activation and transcription induction , and is maintained independently of B cell activation and promoter activity . Additionally , insulation of the 3’RR does not affect the methylation pattern of Iγ3 and Iγ2b promoters . In contrast , the relative demethylation of Iγ3 and Iγ2b exons results from GL transcription and not from B cell activation . Iγ3 and Iγ2b promoters were unmethylated prior to , and following B cell activation , reminiscent of Eμ enhancer and the 3’RR [16–19] . This led us to explore the methylation pattern of other cis-acting elements , with known or suspected regulatory function . We focused on three CpG-rich clusters at Cδ-Iγ3 intergenic region ( 3’δ1 to 3’δ3 ) ( Fig 3A ) . Two clusters ( 3’δ1 and 3’δ2 ) flank a region that is highly enriched in transcription factors binding sites and may play a role in early B cell development [20]; the other , located further downstream , is used as a negative control . We also examined two DNase I hypersensitive sites within Cγ1-Iγ2b intergenic region ( hereafter 3’γ1E and 5’γ2bE ) that bind various transcriptional/architectural factors [21 , 22] and are involved in long-range interactions with multiple regulatory elements of the IgH locus in early B cells [21] . Additionally , 3’γ1E displays enhancer activity in pro-B cells [22] . Finally , we analyzed the intragenic 5’hs1RI insulator region whose CTCF binding site does not contain any CpG but is flanked by two clusters of 3 and 11 CpGs [13] . The data showed distinct CpG methylation patterns: 3’γ1E was largely unmethylated , both in resting and LPS-activated splenic B cells ( Fig 3A ) and its pattern was unchanged upon insulation of the 3’RR or deletion of 5’hs1RI ( Fig 3B ) . The 3’δ1–3 , 5’γ2bE and 5’hs1RI elements were hypermethylated in resting B cells as well as after LPS activation ( Fig 3A ) . 5’γ2bE CpGs were also methylated in TGFβ-activated splenic B cells ( Fig 3C ) . The finding that Iγ3 and Iγ2b promoters and 3’γ1E enhancer were essentially unmethylated in resting splenic B cells led us to investigate when their non-methylated state was established , and whether this state was B cell-specific . To this end , we analyzed CpG methylation in various tissues and cell types . As controls , we assayed the Eμ enhancer , known to undergo lymphoid-specific demethylation and to remain unmethylated throughout B cell development [17 , 19] , and 5 CpGs upstream of Iγ2b which were heavily methylated in splenic B cells ( Figs 1 and 2 ) . Indeed , The 5 CpGs upstream of Iγ2b were hypermethylated regardless of the cell type analyzed ( Fig 4A and 4B ) . In contrast , Eμ was only minimally methylated in CD4+ T cells ( 10% of mCpGs ) , and was fully unmethylated in WT fetal liver B cells and in pro-B cells derived from the bone marrow of Rag2-deficient mice ( Fig 4B ) . However , Eμ was relatively more methylated in mature sperm ( 64% ) , in serum-grown embryonic stem cells ( ESCs ) ( 51% ) and in the tail tissue of Rag2-deficient mice ( 76% ) ( Fig 4A ) . The 5’γ2bE was heavily methylated in all tissues and cell types analyzed except in ESCs where it was relatively less methylated ( 58% ) ( Fig 4A and 4B ) . Interestingly , 3’γ1E underwent a strict B cell-specific demethylation , contrasting with Eμ enhancer whose demethylation was more pronounced in T cells ( Fig 4B ) . Importantly , Iγ3 promoter was markedly hypomethylated in sperm ( 31% ) ( Fig 4A ) , whereas Iγ2b promoter ( 69% ) ( Fig 4A ) and Iγ1 ( 100% ) and Iα ( 81% ) promoters ( S4 Fig ) were heavily methylated . Importantly , Iγ3 promoter and , to lesser extent , Iγ2b promoter underwent further demethylation in ESCs ( 8% and 51% respectively ) ( Fig 4A ) . In non-B cells , compared to ESCs , Iγ3 and Iγ2b were more methylated in Rag2-/- tail ( 40% and 65% respectively ) ( Fig 4A ) , whereas in CD4+ T cells , Iγ3 underwent moderate methylation ( 31% ) while Iγ2b was further demethylated ( 25% ) ( Fig 4B ) . Remarkably , in the B cell lineage , Iγ2b promoter was more demethylated than Iγ3 promoter in fetal liver ( 7% and 35% of mCpGs ) . Iγ3 promoter became fully unmethylated in pro-B cells of Rag2-deficient mice ( Fig 4B and S5 Fig ) . Altogether , the data revealed that , among the cis-acting elements analyzed , Iγ3 promoter was already hypomethylated in sperm and ESCs , and fully unmethylated in pro-B cells of adult mice . Iγ2b promoter , Eμ and 3’γ1E enhancers were hypermethylated in sperm but underwent massive demethylation in fetal liver B cells . The above data showed that Iγ3 and Iγ2b promoters displayed different dynamic methylation patterns during embryonic development and cell differentiation , and that in sperm and ESCs , Iγ3 promoter was hypomethylated compared to Iγ2b . One possibility is that the two promoters are targeted by different demethylation machineries . In an attempt to identify the demethylation pathways involved , we assayed for CpG methylation at Iγ3 and Iγ2b promoters in mature sperm and resting splenic B cells of mice with Activation-induced cytidine deaminase ( AID ) , Uracil DNA glycosylase ( UNG ) , Ataxia telangiectasia mutated kinase ( ATM ) , or the Tumor suppressor protein p53 deficiency ( see discussion ) . Strikingly , Iγ3 promoter displayed a hypermethylated pattern in AID- , UNG- , and ATM-deficient sperm compared to WT control ( Fig 5A ) . In contrast , the methylation pattern of Iγ3 promoter did not significantly change in p53-deficient sperm ( Fig 5A ) . The methylation pattern of Iγ2b promoter was not significantly affected regardless of the genetic deficiency ( Fig 5A ) . For all deficiencies analyzed , the methylation pattern of Iγ3 and Iγ2b promoters was similar to WT in B cells ( Fig 5B ) . The data established that in mature sperm , Iγ3 and Iγ2b promoters displayed different methylation patterns , and that Iγ3 promoter was specifically hypermethylated in AID- , UNG- , and ATM-deficient sperm . In resting B cells however , the unmethylated profile of both promoters was essentially insensitive to AID , ATM , UNG , or p53 deficiency .
Four main conclusions emerge from this study . DNA methylation does not play a significant role in IgH constant genes expression . Acquisition of DNA methylation by the constant exons is not mediated by transcriptional elongation . The hypomethylated pattern of the late B cell-specific Iγ3 promoter was manifest in mature sperm and ESCs already , in contrast to Eμ and 3’γ1E enhancers and other I promoters . Iγ3 and Iγ2b promoters recruited different demethylation pathways . Except for Iγ1 , B cell activation and induction of GL transcription did not perturb the methylation patterns of I promoters . Explanations such as the nature of the stimulus or the number of promoter CpGs cannot explain these patterns . For instance , Iγ1 and Iε promoters are both induced by IL4 stimulation , but while Iγ1 underwent full demethylation , Iε did not . On the other hand , both Iγ3 and Iγ1 promoters contain a single CpG , but while Iγ3 was already unmethylated in resting B cells , Iγ1 became fully unmethylated only after induction . In this regard , various studies showed that methylation of a single CpG can have important functional or pathological consequences [23–26] . GL transcription at the IgH constant locus is largely controlled by the 3’RR [15] , which was shown to engage in long-range interactions with I promoters through chromatin looping , in a stimulus-dependent manner [27–30] . Additionally , the 3’RR controls various active histone modifications at I promoter/exon regions [31] . Our findings strongly suggest that the formation of IgH loops and the set-up of active histone marks associated with I promoters activation do neither require nor induce demethylation of I promoters . Significantly , Iγ3 and Iγ2b promoters are the most sensitive to 3’RR mutations ( e . g . [11 , 12] ) . Nonetheless , their unmethylated pattern did not change upon insulation of the 3’RR , which fully repressed these promoters . The 3’RR thus controls Iγ3 and Iγ2b promoters through mechanisms that do not involve DNA methylation , contrasting in this regard with other Ig enhancers ( e . g . [17 , 19 , 32] ) . It remains to be established whether the 3’RR displays a demethylating activity at earlier B cell developmental stages . Paradoxically , Iγ1 , known to be relatively 3’RR-independent ( e . g . [11 , 12] ) , was the only I promoter whose demethylation was induced . This may relate to the presence of specific regulatory elements with demethylating activity such as the putative Iγ1 promoter-associated enhancer [33] , and/or the 3’γ1E enhancer . Testing these hypotheses still awaits appropriate knock-out models . Induction of GL transcription led to a moderate hypomethylation of essentially the most proximal CpGs of Iγ exons . This may be due to pausing of RNA pol II that takes place 30–60 nucleotides downstream of the transcription start site ( s ) . Accordingly , high-levels of RNA pol II p-Ser5 were detected at Iγ3 exon upon LPS stimulation [9] , which may protect some CpGs against methylation . Methylation of Iε and Iα exons , however , was not impacted by stimulation , suggesting that the mechanisms that underlie pausing at I exons may differ . Seminal studies using transformed cell lines and methylation-sensitive restriction enzymes found a positive correlation between DNA hypomethylation and constant genes transcription [34–37] . However , this correlation was not observed in primary B cells [38] . Accordingly , we found that Cγ3 , Cγ1 , Cγ2b and Cα exons were already hypermethylated in unstimulated splenic B cells and remained so after induction of GL transcription , regardless of the nature of the stimulus . This indicates that transcriptional elongation across the chromatin of constant exons does not bring about any obvious change of their hypermethylated pattern . Interestingly , this hypermethylated pattern coincides with transcription-associated deposition of H3K36me3 at Cγ exons [9] . In genomic imprinting for instance , acquisition of DNA methylation through transcription-associated H3K36me3 has been demonstrated for some imprinting control regions . In this process , H3K4 methylation , which prevents the action of DNMT3A-DNMT3L de novo methyl-transferase complex , is first removed from chromatin , this enables transcription-associated H3K36me3 to recruit DNMT3A-DNMT3L complex that will methylate DNA [39] . This is clearly not the case for IgH constant exons which are likely methylated through a different mechanism . Previous work indicated that H3K36me3 and intragenic DNA methylation contribute to the silencing of alternative , intragenic promoters [40 , 41] . Low levels of antisense switch transcripts have been detected ( e . g . [42 , 43] ) , but antisense promoters have not been precisely defined . Intragenic methylation may contribute to down-regulation of the antisense promoters and/or other cryptic promoters . An attractive possibility could be that DNA hypermethylation and H3K36me3 across the constant exons protect these regions from AID attack by favoring a compacted chromatin structure after nucleosome displacement induced by RNA pol II passage . This chromatin-based protection mechanism is physiologically relevant as the constant exons are coding sequences whose reading frame must be preserved if the Ig heavy chain is to be produced . In this regard , the highly cytosine-rich , non-coding switch sequences , which are preferentially targeted by AID during CSR are strikingly poor in CpG [44] compared to constant exons . Two major waves of DNA methylation reprogramming occur during development , shortly after fertilization and in primordial germ cells ( PGCs ) . After the massive methylation erasure in PGCs , de novo DNA methylation is acquired in prenatal prospermatogonia before birth . The methylation patterns are fully established at birth and are maintained before the cells enter meiosis [45 , 46] , and it was shown that sperm cells display the highest global DNA methylation level [47] . In stark contrast to I promoters , and to Eμ and 3’γ1E enhancers , Iγ3 was already hypomethylated in mature sperm . Moreover , Iγ3 promoter underwent further demethylation in serum-grown ESCs despite the fact that ESCs grown in this condition display high DNA methylation levels [48] , comparable to those of mature sperm [47] . These findings suggest that Iγ3 promoter is hypomethylated in pre-implanted embryo . Upon differentiation however , Iγ3 promoter moderately acquires DNA methylation and is fully unmethylated only in B cells of adult mice . Altogether , the above findings strongly suggest that the cis-acting elements analyzed are targeted by ( de ) methylating activities in a highly specific manner . The differential targeting is also evident from the patterns of Iγ3 and Iγ2b in sperm with AID , UNG , ATM or p53 deficiency . The role of AID in DNA demethylation is still controversial ( e . g . [45 , 46 , 49–51] ) . AID was implicated both in vitro and in vivo at various stages of mouse embryonic development [47 , 52 , 53] . Our data show that Iγ3 , but not Iγ2b , is hypermethylated in AID-deficient sperm . This indicates that AID demethylation pathway is involved , and preferentially targets Iγ3 promoter . Whether it is AID itself , or a cofactor , that is directly implicated is presently unclear . Also , we do not infer that AID-mediated demethylation occurs in mature sperm . Demethylation may have occurred in PGCs , and the hypomethylated pattern subsequently maintained during the establishment of the male germ line . In this regard , low levels of AID expression were detected in PGCs but not in the germ line [52 , 54] . Preferential targeting of Iγ3 promoter was also evident in UNG-deficient sperm . The base excision repair pathway [54] , and in particular UNG which excises uracil from DNA , has been implicated in DNA demethylation in zygotes and PGCs [53 , 55 , 56] . In antibody diversification mechanisms in B cells , AID deaminates a non-methylated cytosine to generate a U:G mismatch that can be processed by UNG [57] . A somewhat analogous scenario has been proposed for cytosine demethylation in mouse zygotes [55] . Whether , similarly , UNG acts downstream of AID in PGCs is presently unclear . However , it is possible that UNG is involved through an AID-independent pathway . ATM is a major component of the DNA damage response , and it has recently been implicated in the establishment of DNA methylation patterns during spermatogenesis , as global DNA methylation was reduced in ATM-deficient testis [58] . Based on this , we expected a hypomethylated pattern in ATM-deficient sperm . However , Iγ3 was hypermethylated while Iγ2b was unaffected . This suggests that ATM-mediated demethylation of Iγ3 implicates different , yet unknown mechanisms . In contrast , methylation patterns of Iγ3 and Iγ2b promoters did not significantly change in p53-deficient sperm . p53 has been shown to down-regulate the de novo DNMT3A and DNMT3B methyl-transferases and up-regulate TET1 and TET2 in naïve ESCs , whereas in differentiated cells , p53 became a repressor of Tet1 and Tet2 genes [59] . None of these modes of regulation seems to target Iγ3 and Iγ2b promoters although an effect of p53 at a discrete developmental stage can not be excluded . Overall , the methylation pattern of Iγ3 , but not of Iγ2b promoter , was perturbed in AID , UNG or ATM-deficient sperm . This suggests that different pathways somehow contribute to the setting of the methylation patterns of these promoters . Whether these pathways act at the same developmental stage and whether they interact with each other and/or with other pathways is presently unknown . In contrast , none of the pathways studied was significantly required for the maintenance of the unmethylated state of Iγ3 and Iγ2b promoters in resting splenic B cells . Thus , once the demethylation mark has been set up , the involved pathways seem dispensable for the maintenance of the mark at subsequent B cell developmental stages . Though still debatable , accumulated evidence supports the notion that at least some of the epigenetic features that underlie tissue-specific expression are somehow stamped at earlier developmental stages , prior to the specification of the relevant lineage [39 , 60 , 61] . For instance , asynchronous replication , set up early during development , was suggested to epigenetically mark antigen receptor loci for mono-allelic recombination at the right developmental stage [62 , 63] . Some , but not all , B cell-specific enhancers are primed in hematopoietic stem cells ( e . g . [64–67] ) . Other tissue-specific genes are epigenetically marked in ESCs [68 , 69] . Regarding DNA methylation specifically , different tissue-specific enhancers , but not promoters , displayed a subset of hypomethylated CpGs in ESCs [25 , 70] . What could be the functional significance of the overall hypomethylated pattern of Iγ3 promoter ? Splenic marginal zone B cells represent a special population of the adaptive immune system . These “innate-like” lymphocytes [71] play an important role in rapid protective responses against blood-borne antigens . They are in a state of active readiness and switch to IgG3 preferentially in response to T-cell-independent antigens [71] . We speculate that the early set-up of the hypomethylated pattern of Iγ3 may be part of an epigenetic programme that predisposes this promoter for fast activation in marginal zone B cells . In conclusion , methylation patterns of IgH constant locus elements are essentially transcription-independent . The mature B cell-specific Iγ3 promoter is hypomethylated early during ontogeny . Iγ3 and Iγ2b promoters recruit different demethylation pathways that are dispensable for the maintenance of the demethylation mark once established in the B cell lineage . Further investigations are required to unravel the multiple facets of DNA methylation regulation at the IgH locus during development and to elucidate the mechanisms that control the process .
The experiments on mice were carried out according to the CNRS Ethical guidelines and were approved by the Regional Ethical Committee ( Accreditation N° E31555005 ) . ESCs ( CK35 line , of 129Sv background ) were provided by Chantal Cress ( Institut Pasteur , Paris , France ) . The WT and homozygous Rag2-/- , ZILCR , 5’hs1RIΔ/Δ were of 129Sv genetic background . AID-/- , ATM-/- , UNG-/- , and p53-/- mutant mice were enriched in 129Sv genetic background through at least 8 back-crosses , and both their chromosomes 12 ( harbouring the IgH locus ) were derived from 129Sv . All the mice used were 6–8 week-old . ATM-deficient mice were purchased from Jackson labs and p53-deficient mice were from the European Mutant Mouse Archives , Orléans , France . AID-deficient mice were provided by T . Honjo , through C-A . Reynaud and J-C . Weill . UNG-deficient mice were provided by T . Lindahl , through C . Rada and the late M . S . Neuberger . Single cell suspensions from the bone marrows or spleens were obtained by standard techniques . Rag2-deficient pro-B cells ( from the bone marrow of Rag2-/- mice ) and WT fetal liver B cells ( at day 14 post-coitum ) were positively sorted by using B220- and CD19-magnetic microbeads and MS columns ( Miltenyi ) . Splenic B cells were negatively sorted by using CD43-magnetic microbeads and LS columns ( Miltenyi ) . Splenic CD4+ cells were sorted as B220-IgM-CD4+ population . ESCs cells were serum-grown in the presence of LIF ( 106 units/ml ) throughout: first on mitomycin-treated feeder cells for 2 days , trypsinized and amplified for additional 2 days without feeders . After trypsinization , the cells were plated on gelatinized dishes for 2 hours , and the ESC-enriched supernatant carefully pipetted off and plated again for additional 2 hours in order to get rid of contaminating feeders . Sperm was collected from the cauda epididymis of adult males by the “swim-up” method [72] . To induce GL transcription , negatively sorted CD43- splenic B cells were cultured for 2 days , at a density of 5 x 105 cells per ml in the presence of LPS ( 25 μg/ml ) + anti-IgD-dextran ( 3 ng/ml ) ( hereafter LPS stimulation ) , LPS ( 25 μg/ml ) + anti-IgD-dextran ( 3 ng/ml ) + IL4 ( 25 ng/ml ) ( IL4 stimulation ) , LPS ( 25 μg/ml ) + anti-IgD-dextran ( 3 ng/ml ) + IFNγ ( 20 ng/ml ) ( IFNγ stimulation ) or LPS ( 25 μg/ml ) + anti-IgD-dextran ( 3 ng/ml ) + IL4 ( 10 ng/ml ) + IL5 ( 5 ng/ml ) + BLyS ( 5 ng/ml ) + TGFβ ( 2 ng/ml ) ( TGFβ stimulation ) . Genomic DNAs were purified from the following sources: sorted resting splenic B cells from WT , 5’hs1RIΔ/Δ , AID-/- , UNG-/- , ATM-/- , and p53-/- mutant mice; from WT , ZILCR , AID-/- splenic B cells at day 2 post-stimulation; from WT ESCs , resting splenic CD4+ T cells , and fetal liver B220+ cells; from pro-B cells or from the tail of Rag2-/- mice; from mature sperm of WT , UNG-/- , AID-/- , ATM-/- , and p53-/- mutant mice . Purified genomic DNAs were assayed by sodium bisulphite sequencing by using a bisulphite conversion kit ( Diagenode ) . Modified templates were amplified by PCR using converted primers listed in S1 Table . Converted primers were designed by using the public MethPrimer software . PCR products were separated by agarose gel electrophoresis , purified using QIAquick gel extraction kit ( Qiagen ) , and cloned into pCR2 . 1-TOPO vector ( Invitrogen ) . Transformed bacteria were plated immediately after transformation without pre-culture , and randomly picked clones were sequenced ( Eurofins Genomics ) . Sequence analysis showed 99%-100% bisulphite modification efficiency . Allophycocyanin ( APC ) -conjugated anti-B220 , fluorescein isothiocyanate ( FITC ) -conjugated anti-IgG1 , Phycoerythrin ( PE ) -conjugated anti-IgG2b , PE-conjugated anti-IgG2a , and PE-conjugated anti-CD4 antibodies were purchased from BioLegend . FITC-conjugated anti-IgG3 and FITC-conjugated anti-IgA were from BD-Pharmingen . LPS was purchased from Sigma , anti-IgD-dextran from Fina Biosolutions , TGFβ , B-LyS , IFNγ and IL5 from R&D , and IL4 from eBiosciences . At day 4 post-stimulation , B cells were washed and stained with anti-B220-APC and either anti-IgG3-FITC , anti-IgG2b-PE , anti-IgG1-FITC , anti-IgG2a or anti-IgA-FITC . Activated B cells from AID-deficient mice ( unable to initiate CSR ) were included throughout as negative controls . Data were obtained on 5 x 105 viable cells by using a BD FACSCalibur flow cytometer . Total RNAs were prepared from B cells at day 2 post-stimulation , reverse transcribed ( Invitrogen ) and subjected to qPCR using Sso Fast Eva Green ( BioRad ) . Actin transcripts were used for normalization . The primers used have been described [13] . Results are expressed as mean ± SD ( GraphPad Prism ) and overall differences between values from day 0 and day 2 post-stimulation were evaluated by paired t-test , and from WT and AID- , UNG- , ATM- and p53-deficient sperm by unpaired t-test . The difference between means is significant if p value < 0 . 05 ( * ) , very significant if p value < 0 . 01 ( ** ) , and extremely significant if p value < 0 . 001 ( *** ) .
|
DNA methylation mainly occurs at CpG dinucleotides and strongly influences gene expression during development . Deregulation of DNA methylation is often associated with disease . In mammalian genomes , unmethylated CpG dinucleotides are generally associated with active regulatory sequences , while methylated CpGs are often associated with silent promoters . The immunoglobulin heavy chain constant locus comprises multiple inducible constant genes and is expressed exclusively in B lymphocytes . We show that methylation patterns of most of the locus cis-elements , including promoters , enhancers and insulators , are established and faithfully maintained independently of B cell activation or transcription initiation . Acquisition of DNA methylation by the constant genes exons occurs independently of transcriptional elongation . One late B cell specific promoter is hypomethylated early in ontogeny . Constant genes promoters recruit different demethylation pathways that become dispensable for the maintenance of the mark in the B cell lineage . The data suggest that , rather than playing a prominent role in transcriptional regulation , DNA methylation may contribute to the epigenetic pre-marking of the IgH constant locus .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"chemical",
"compounds",
"immunology",
"nucleotides",
"organic",
"compounds",
"dna",
"transcription",
"germ",
"cells",
"pyrimidines",
"epigenetics",
"dna",
"dna",
"methylation",
"chromatin",
"sperm",
"white",
"blood",
"cells",
"chromosome",
"biology",
"animal",
"cells",
"gene",
"expression",
"chemistry",
"chromatin",
"modification",
"dna",
"modification",
"genetic",
"loci",
"biochemistry",
"antibody-producing",
"cells",
"cytosine",
"cell",
"biology",
"nucleic",
"acids",
"b",
"cells",
"organic",
"chemistry",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences"
] |
2019
|
Developmental regulation of DNA cytosine methylation at the immunoglobulin heavy chain constant locus
|
Nigeria has a significant burden of lymphatic filariasis ( LF ) caused by the parasite Wuchereria bancrofti . A major concern to the expansion of the LF elimination programme is the risk of serious adverse events ( SAEs ) associated with the use of ivermectin in areas co-endemic with Loa filariasis . To better understand this , as well as other factors that may impact on LF elimination , we used Micro-stratification Overlap Mapping ( MOM ) to highlight the distribution and potential impact of multiple disease interventions that geographically coincide in LF endemic areas and which will impact on LF and vice versa . LF data from the literature and Federal Ministry of Health ( FMoH ) were collated into a database . LF prevalence distributions; predicted prevalence of loiasis; ongoing onchocerciasis community-directed treatment with ivermectin ( CDTi ) ; and long-lasting insecticidal mosquito net ( LLIN ) distributions for malaria were incorporated into overlay maps using geographical information system ( GIS ) software . LF was prevalent across most regions of the country . The mean prevalence determined by circulating filarial antigen ( CFA ) was 14 . 0% ( n = 134 locations ) , and by microfilaria ( Mf ) was 8 . 2% ( n = 162 locations ) . Overall , LF endemic areas geographically coincided with CDTi priority areas , however , LLIN coverage was generally low ( <50% ) in areas where LF prevalence was high or co-endemic with L . loa . The extensive database and series of maps produced in this study provide an important overview for the LF Programme and will assist to maximize existing interventions , ensuring cost effective use of resources as the programme scales up . Such information is a prerequisite for the LF programme , and will allow for other factors to be included into planning , as well as monitoring and evaluation activities given the broad spectrum impact of the drugs used .
Lymphatic filariasis ( LF ) is one of the most debilitating neglected tropical diseases ( NTD ) in the world [1] . It is caused by the parasitic worms Wuchereria bancrofti , Brugia malayi and B . timori and is transmitted by Anopheles , Culex , Aedes , Ochlerotatus and Mansoni mosquitoes [1] . Wuchereria bancrofti is transmitted throughout the tropics in Africa , Asia , the Pacific and the Americas while B . malayi and B . timori are found in east and south Asia . The disease is endemic in 73 countries with an estimated 120 million people infected and 40 million people with clinical manifestations including lymphoedema ( elephantiasis ) of the limbs and urogenital disorders , especially hydrocele in men [2] [3] . In Africa , 34 countries are endemic , and Nigeria is believed to bear the highest burden of LF , with an estimated 80 to 120 million people at risk [3]–[5] . The Global Programme to Eliminate LF ( GPELF ) was launched in 2000 with the goal of eliminating LF as a public health problem by 2020 [1] . The principal elimination strategy is to interrupt transmission using Mass Drug Administration ( MDA ) with the combinations of albendazole plus ivermectin or albendazole plus diethylcarbamazine ( DEC ) administered once a year for at least five consecutive years . [1]–[3] . Overall , significant progress has been made , however , the scale up of programmatic activities has been slow in Africa , especially in countries with logistical challenges , conflict , instability and fragile infrastructures [6] . The wide and overlapping distribution of the filarial parasite Loa in Africa [7] is also a major impediment due to the risk of severe adverse events ( SAEs ) in co-infected individuals when treated with ivermectin [8] [9] . These constraints pose significant problems for the national LF programmes and GPELF with the potential to severely hinder the 2020 goal of LF elimination globally . To begin to address these complexities , a number of specific objectives and strategies have been developed . First , the GPELF strategic plan aims to achieve full geographical coverage with MDA by 2016 , targeting the countries with the highest burden , including Nigeria [1] . Second , the use of integrated vector management ( IVM ) [10] is advocated in malaria co-endemic areas where both diseases are transmitted by Anopheles mosquitoes [11] . Finally , a provisional strategy for interrupting LF transmission in loiasis endemic countries recently developed recommends albendazole ( 400 mg ) twice yearly in combination with vector control in all co-endemic areas [12] . Finally , mapping LF and L . loa at the lowest possible administrative unit is also considered important to identify small areas that can be treated for LF using the most appropriate regimes to reduce the risk of SAEs , which is considered to be highest when L . loa microfilaremia ( mf ) prevalence is ≥20% . The coordinated effort of global disease control programmes is becoming increasingly important as many operate in the same countries and distribute interventions that have multiple benefits [13]–[16] . GPELF is likely to benefit from the activities of the Global Malaria Programme , including the recent scale up of insecticide treated/long-lasting insecticidal mosquito nets ( ITNs/LLINs ) and indoor residual spraying ( IRS ) [11] . These interventions have also been shown to impact LF transmission in a range of ecological settings [17] , thus more synergy between the programmes in Africa could optimize resources and increase the impact on both diseases [15]–[18] . In countries such as Nigeria where malaria and LF are co-endemic and both transmitted by Anopheles mosquitoes [19] [20] , the use of ITNs has shown to be effective at reducing LF transmission in L . loa co-endemic areas [21] . ITNs have also been successfully integrated with MDA activities in Central Nigeria with report of an increase in ITN ownership and retention [22] [23] . However , to take advantage of these programmatic links , more data on LF vectors is critical as there are many gaps in our knowledge as highlighted in the Anopheles database recently compiled for Nigeria [19] . Integrating activities and combining resources across the various NTD programmes will also have many advantages [13] [24] . For example , the African Programme for Onchocerciasis Control ( APOC ) has developed a sustainable community-directed treatment with ivermectin ( CDTi ) for the parasitic disease caused by the filarial worm Onchocerca volvulus , [25]–[27] . The CDTi approach has been successful in reaching millions of people across high transmission areas of onchocerciasis in Africa , and has also been used to distribute other health interventions including LF treatment and bed nets for malaria control [26] . Moreover , the maps of CDTi priority areas highlight the potential geographical overlap of onchocerciasis with LF , and it is likely that the wide and frequent use of ivermectin has reduced transmission in co-endemic areas [28]–[31] . However , the extent of this impact is yet to be determined at a large scale and needs to be quantified so that benefits from this and future NTD control programmes can be better understood and fully exploited [5] [32]–[34] . These issues are particularly relevant for Nigeria , given the large population at risk of W . bancrofti infection [3]–[5] . The National Lymphatic Filariasis Elimination Programme ( NLFEP ) is yet to complete LF mapping [3] [35] and will need significant financial and technical support to scale up MDA activities across this large , populous country . The aim of this paper , therefore , is to use the Micro-stratification Overlap Mapping ( MOM ) approach [24] to review and synthesize the current knowledge of the distribution of W . bancrofti in Nigeria , and factors that will impact on the control and elimination of LF such as loiasis co-endemicity , onchocerciasis control programmes , and malaria bed net distributions . This information is a prerequisite for effective planning and will help to optimize the future LF MDA implementation strategy to ensure safety , maximum cost effectiveness as well as impact .
Nigeria is a Federal Republic comprising 36 States and its Federal Capital Territory , Abuja [35] [36] . The states are grouped into six geopolitical zones , the North Central ( NC ) , North East ( NE ) , North West ( NW ) , South West ( SW ) , South East ( SE ) and South ( SS ) . Nigeria covers an area of approximately 923 , 768 sq . km , and has a large low plateau intersected by two major rivers , the Niger and Benue , in the central region of the country ( Figure 1 ) . It shares borders with Benin in the west , Chad and Cameroon in the east , and Niger in the north . Its coast in the south lies on the Gulf of Guinea on the Atlantic Ocean and Lagos , the former capital , is an important port city . Nigeria is Africa's most populous country with the total population estimated to be 160 million in 2012 , with approximately 50% living in urban areas . To review and synthesize the current knowledge of the human distribution of LF in Nigeria , a systematic search for data in peer-reviewed published literature and national reports was carried out . The search was conducted using PubMed , JSTOR , Google , SCOPUS and other online scientific and historical databases . References were also obtained from the references listed within articles , and then from the references within those articles . Studies and reports with data on the prevalence of i ) LF infection as circulating filarial antigen ( CFA ) from using immunochromatographic tests ( ICTs ) , antibodies by ELISAs , and microfilaria ( Mf ) from blood slides , and ii ) disease cases ( hydrocele , lymphodema ) [2] [37] were identified and collated into a database . Information on the location/collection site ( village , local government area ( LGA ) and State ) , and time period ( month , year ) , was also collected for mapping and descriptive analyses . Specific information on whether MDA for LF had been administered prior to the LF prevalence measure was recorded and considered in the analysis . The range of methods used to detect LF in the different studies was recorded , as well as information on the mosquito species , which was cross-checked with the Nigerian Anopheles database [19] . The locations of the community or collection site were geo-referenced using the latitude and longitude coordinates obtained from references directly or by cross-checking the names with data from the GEOnet Names Server , Directory of Cities and Towns in the World databases [38] [39] . The coordinates of the midpoint of the LGA was used as a proxy for the locations that could not be allocated exact latitude and longitude coordinates . This is considered to be a limitation of the review and restricts any accurate detailed mapping . It is also acknowledged that LF prevalence distribution has a degree of bias as the data are based on the locations selected by the investigators in the original study , and does not take sampling methodologies between studies into account , which may affect the outcome . In addition , selected data from the Federal Ministry of Health ( FMoH ) collected during LF mapping activities were collated and included in the database . The LF data available for this study were based on Mf prevalence rates collected in selected LGA sentinel sites during baseline surveys in 31 LGAs across 18 States of Nigeria . The WHO standard protocol was used to collect blood samples at night and examined for the presence of Mf . The coordinates of the midpoint of each LGA was used to map the LF prevalence . A national LF endemicity map by LGA was also available from the FMoH , which provided an overall CFA prevalence based on ICT survey in each State carried out between 2000 and 2010 . Specific LGA data is not publicly available and not included in this database , however , the State-level information on the number of LGAs surveyed , prevalence range and year of survey is available in the recently published Master Plan for NTDs [35] . All the relevant information was entered into an Excel worksheet and data analysis was performed using Stata software ( version 12 , StataCorp , Texas , USA ) . All data were mapped using the geographical information systems ( GIS ) software ArcGIS 10 . 0 ( ESRI , Redlands , CA ) to produce maps of LF prevalence distributions , and to examine the geographical overlaps with loiasis-endemic areas , and the different intervention distributions . To examine the potential extent of LF and L . loa co-endemicity , the recent map of the predicted loiasis prevalence produced from a Rapid Assessment Procedure of Loiasis ( RAPLOA ) based on eye worm history carried out between 2004 and 2010 across Africa , including Nigeria [7] , was imported into ArcGIS . Three levels of predicted loiasis prevalence were digitised ( i . e . outlined , shaded ) based on the defined distribution boundaries , which included low <20% , medium 20–40% and high >40% prevalence areas; the latter is equivalent to mf prevalence of >20% . The different levels of loiasis prevalence and the overlap with LF prevalence distributions were highlighted to help identify potential low risk ( i . e . loiasis <20% ) and medium to high risk ( i . e loiasis >20% ) SAE areas .
In total , 41 studies [41]–[81] from 68 published and unpublished filariasis studies identified in the literature were found to have examined the prevalence , clinical manifestations and entomological aspects of LF in Nigeria ( Table S1 ) . The studies excluded from the review had reported data on L . loa and/onchocerciasis only . The majority of LF studies ( n = 30 ) were conducted post 2000 [41]–[69] , nine studies were conducted between 1980 and 2000 [70]–[78] , and three studies were conducted pre 1980 [79]–[81] . The studies indicate that LF is present in 19 States across all six geopolitical zones of the country ( Figure 2a ) . The majority of studies were from the NC geopolitical zone , with the most comprehensive studies carried out in Plateau and Nassarawa States [49] [53] [54] [63] [66] . The FMoH sentinel site Mf prevalence data were carried out more widely in 31 LGAs across 18 States . All information was added to the database ( Table S1 ) and mapped with the other specific Mf data described above ( Figure 2a ) . The FMoH national endemicity map indicated that out of the 774 LGAs in Nigeria , 541 were classified as endemic , 164 were classified as non-endemic and 69 remained to be mapped ( Figure 2b ) . The related state-level data are found in Nigeria Master Plan for NTDs [35] . The range of methods used to detect the presence of LF in Nigeria included serological methods ( using ICTs or ELISA ) , parasitological methods ( blood films for Mf ) and physical examination for clinical manifestations ( lymphodema , hydrocele ) , and were used either alone or in combination . Studies carried out before the 1980s only used parasitological examination of blood films , whereas post 2000 , a combination of serology and parasitological methods were most widely used ( Table S1 ) . In total there were 258 individual data points from 152 unique locations where the prevalence of W . bancrofti was measured by CFA or Mf . The average CFA and Mf rates by State are summarized in Table 1 and 2 . The overall mean CFA prevalence rate across the country was 14 . 0% ( n = 134; range 0% to 66 . 0% ) , and the overall mean Mf prevalence rate was 8 . 2% ( n = 162; 0 to 47 . 4% ) . The CFA and Mf prevalence rates by geopolitical zones indicate that the highest rates occur in the NC , NE , NW and SS zones . The highest CFA prevalence rate recorded was 66% recorded at Ogi-Utonkon , Ado LGA , Benue State , NC [64] , while the highest Mf rate was 47 . 4% recorded at Zing LGA , Taraba State , NE [55] ( Table S1 ) . Overall , there were marked differences in W . bancrofti prevalence at sites that had not received MDA i . e . pre-MDA , compared to those sites that had received MDA i . e . post-MDA . The average CFA prevalence in pre-MDA sites , was 20 . 3% ( n = 68; range 0% to 66 . 0% ) , which was approximately 3 times higher than the average CFA prevalence in post-MDA sites , 7 . 6% ( n = 66 , range 0 . 2% to 31 . 5% ) ( Figure 3a , b ) . The average Mf prevalence in pre-MDA sites was 10 . 1% ( n = 124; range 0% to 47 . 4% ) , which was approximately 5 times higher than the average Mf prevalence in post-MDA sites , 2 . 0% ( n = 66 , range 0% to 12 . 1% ) ( Figure 3c , d ) . The distribution of post-MDA sites occur in the two States of Plateau and Nassarawa , and are the result of an extensive MDA programme delivering the combination of ivermectin and albendazole as detailed in the study publications [49] [53] [54] [63] . Baseline LF mapping was conducted in 1999 and 2000 in 30 LGAs of the two States , and MDA launched in 2000 , and monitored from 2000 to 2009 . Details are contained in the specific reference [63] ( Table S1 ) . The most extensive study on clinical manifestations was conducted by Nwoke et al . [82] , who used hydrocele as a clinical marker to estimate LF prevalence . A rapid epidemiological mapping survey ( REM-LF ) was conducted across 25 States and 536 villages in Nigeria . Details of the specific study sites are not available , however , the survey found that hydrocele was absent in 339 ( 63 . 3% ) villages , and present in 197 ( 36 . 8% ) villages , which were found to have different levels of hydrocele severity . Hydrocele was absent in Jigawa and Kano ( NW ) , and Ogun ( SW ) States . Very few hydrocele cases ( 1–3% ) were found in northern Borno ( NE ) , Kaduna and Zamfara ( NW ) , Edo ( SS ) , Imo ( SE ) , and in Ekiti , Ondo , Osun , Oyo ( SW ) States . The highest hydrocele rates were found in the NE States of Adamawa , Bauchi , Gombe , Taraba and southern Borno , in the NC states of Kogi , Plateau , Nassarawa , and in the northern part of Akwa Ibom State in the SS ( Figure 2c ) [82] . The clinical signs that were reported included limb lymphodema , hydrocele . chyluria and elephantiasis , and were from a few specific areas of the country [45]–[49] [59]–[62][65] [68]–[71] [74]–[75] [77] . For hydrocele , there were 22 sites with prevalence data ranging from 0 . 1% to 50% , while for lymphodema , 12 sites recorded prevalence rates ranging from 1% to 49% . The prevalence of limb elephantiasis was also recorded in 5 sites , which ranged from 1 . 7% to 11 . 8% . The distribution of these study sites is shown in Figure 2c , together with the high hydrocele prevalence states described by Nwoke et al . [82] , and highlight the geographical concordance with CFA and Mf distributions which occur in the central south and eastern regions of the country ( Figure 2a ) . LF prevalence was examined in relation to the L . loa distribution in Nigeria defined by the RAPLOA surveys reporting eye worm history , which were carried out in 381 villages between 2002 and 2010 [7] . Loiasis was found predominately in the southern region of the country , with the highest risk in east along the border with Cameroon , which had a localized area >40% in the States of Taraba and Benue ( Figure 4 ) . Overall , there was minimal geographical overlap with the number of LF prevalence sites determined by CFA . The majority of sites with medium to high LF prevalence rates >25% , were found in low loiasis prevalence areas ( <20% ) where the risk of SAEs are considered to be low ( Figure 4a ) . Similarly , there was minimal geographical overlap with the number of LF prevalence sites determined by Mf , however , more sites with medium to high LF prevalence rates >25% were found in medium loiasis prevalence areas ( 20–40% ) where the risk of SAEs is potentially high ( Figure 4b ) . The overlap with the LF endemicity map available from the FMoH shows a combination of endemic and non-endemic LGAs in the low ( <20% ) to medium ( 20–40% ) loiasis prevalence areas ( Figure 4c ) . Only LF non-endemic LGAs were found in the high risk loiasis area ( >40% ) , which is highlighted in the close up of the map in Figure 4d . The onchocerciasis CDTi priority areas [31] are shown in Figure 5a , and illustrate that large areas across the central region of the country are being targeted with ivermectin treatment . Since 1992 , ivermectin has been distributed annually to 80% of the total population at risk , estimated at ∼38 , 331 , 140 people in 430 endemic LGAs [35] . The LF endemic areas to potentially benefit from CDTi priority areas are extensive and include large areas of NC , NW and NE zones of the country . The LF programme could readily add albendazole to the ivermectin being distributed in these areas ( Figure 5b ) . The potential risks associated with ivermectin treatment for O . volvulus are related to potential SAEs in areas where W . bancrofti and L . loa are co-endemic in the southern region of the county , especially in the States of Benue , Cross River , Ebonyi , Enugu , Osun , Ekiti and regions of Edo , Ondo and Ogun States ( Figure 5c ) . The distribution of LLIN coverage across the six different geopolitical zones is shown in Figure 5d . The highest LLIN coverage occurred in the NE ( 61 . 8% ) and NW ( 58 . 2% ) zones , followed by the SS ( 43 . 5% ) , SE ( 32 . 1% ) , NC ( 32 . 1% ) and SW ( 20 . 3% ) zones respectively . This shows that the highest LLIN coverage occurred in the northern region of the country where LF does not appear to be highly endemic ( Figure 5e ) . The lowest LLIN coverage occurred across the southern region of the country , which coincides with areas where both LF and loiasis are considered endemic and co-endemic ( Figure 5f ) .
This MOM work builds on the recent study carried out in Democratic Republic of Congo [24] , which first used the new overlap mapping approach to collate and map all available country data on W . bancrofti , examine the extent of L . loa co-endemicity and determine the risk and benefits of different intervention strategies . Collectively the two studies address two important countries in Africa with respect to the elimination of LF , as they have the highest burdens of disease , collectively accounting for more than 170 million people at risk . Furthermore , their LF Programmes are yet to scale up to reach full national MDA coverage taking into account the co-endemic areas of L . loa , which may require alternative treatment strategies in selected areas , and coordination with other NTD elimination and malaria control programmes . We advocate that the MOM approach should be used more widely over time and space , and at different geographical scales to better monitor and understand the impact of single and multiple interventions , and to assess progress towards elimination of LF and other diseases . Such an approach is also necessary for national planning purposes as well as increasing the cost effectiveness and coordination of programmes where different strategies are deployed , and where there have been previous interventions which will impact on the goals of the LF programme .
|
Nigeria is estimated to have the highest burden of lymphatic filariasis ( LF ) , a disease also known as elephantiasis , which is transmitted by mosquitoes and caused by the parasite Wuchereria bancrofti . The National LF Elimination Programme is planning to scale up the elimination programme through mass drug administration of ivermectin and albendazole . However , a major constraint to this expansion is the risk of serious adverse events ( SAEs ) associated with the use of ivermectin in areas co-endemic with Loa , the causative agent of tropical eye worm ( loiasis ) . To better understand this and other factors that may impact on LF elimination , we collated and mapped all available LF data , and highlighted the overlaps with predicted loiasis prevalence distributions , onchocerciasis ivermectin treatment areas , and bed net distributions for malaria . This study provides a baseline overview for the LF Programme and will help to maximize existing disease interventions , ensuring cost effective use of resources as the programme scales up .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"infectious",
"diseases",
"mosquitoes",
"soil-transmitted",
"helminths",
"filariasis",
"elephantiasis",
"tropical",
"diseases",
"(non-neglected)",
"global",
"health",
"neglected",
"tropical",
"diseases",
"loiasis",
"onchocerciasis",
"parasitic",
"diseases",
"malaria",
"lymphatic",
"filariasis",
"vectors",
"and",
"hosts",
"infectious",
"disease",
"control"
] |
2013
|
Lymphatic Filariasis in Nigeria; Micro-stratification Overlap Mapping (MOM) as a Prerequisite for Cost-Effective Resource Utilization in Control and Surveillance
|
Immunity to Plasmodium falciparum ( Pf ) malaria is only acquired after years of repeated infections and wanes rapidly without ongoing parasite exposure . Antibodies are central to malaria immunity , yet little is known about the B-cell biology that underlies the inefficient acquisition of Pf-specific humoral immunity . This year-long prospective study in Mali of 185 individuals aged 2 to 25 years shows that Pf-specific memory B-cells and antibodies are acquired gradually in a stepwise fashion over years of repeated Pf exposure . Both Pf-specific memory B cells and antibody titers increased after acute malaria and then , after six months of decreased Pf exposure , contracted to a point slightly higher than pre-infection levels . This inefficient , stepwise expansion of both the Pf-specific memory B-cell and long-lived antibody compartments depends on Pf exposure rather than age , based on the comparator response to tetanus vaccination that was efficient and stable . These observations lend new insights into the cellular basis of the delayed acquisition of malaria immunity .
To date , most successful vaccines have targeted pathogens that induce long-lived protective antibodies after a single infection , such as the viruses that cause smallpox , measles and yellow fever [1] . It has proved more difficult to develop highly effective vaccines against pathogens that do not induce sterile immunity such as the human immunodeficiency virus type-1 ( HIV-1 ) , Mycobacterium tuberculosis ( Mtb ) , and Plasmodium falciparum malaria [2] . However , unlike HIV-1 and Mtb , clinical immunity to malaria can be acquired , but only after years of repeated Pf infections [3] . Passive transfer studies indicate that antibodies ultimately play a key role in protection from malaria [4] , yet several studies show that antibodies to Pf antigens are inefficiently generated and rapidly lost in the absence of ongoing exposure to the parasite ( reviewed in [5] ) . Elucidating the cellular basis of the inefficient acquisition of malaria immunity may ultimately prove critical to the design of an effective malaria vaccine . Despite the key role that antibodies play in protection from a variety of infectious diseases , remarkably little is known about the cellular basis of acquiring humoral immunity in response to natural infections in humans . This gap in our knowledge is due in large part to the difficulty in studying natural infections in humans when we cannot predict who within a population will be infected with a given pathogen at a given time . Thus , our current understanding of the acquisition of immunity is largely derived from animal models and studies of humans after vaccination . These studies have established that long-lived , antibody-based immunity requires the generation and maintenance of memory B cells ( MBCs ) and long-lived plasma cells ( LLPCs ) ( reviewed in [6] , [7] ) . This process begins when naïve B cells bind antigen near the interface of B and T cell areas of secondary lymphoid organs . Several studies suggest that high-affinity binding drives naïve B cells to differentiate into short-lived , isotyped switched plasma cells ( PCs ) within the extra-follicular region which contributes to the initial control of infection . In contrast , lower affinity binding selects for entry of naïve B cells into follicles where germinal centers are formed . After a period of 7–10 days , through the CD4+ T-cell dependent process of somatic hypermutation , the germinal center reaction yields MBCs and LLPCs of higher affinity than the initial wave of short-lived plasma cells ( SLPCs ) . MBCs recirculate and mediate recall responses after re-exposure to their cognate antigen by rapidly expanding and differentiating into PCs , whereas LLPCs residing in the bone marrow constitutively secrete antibody and provide a critical first line of defense against re-infection . The mechanisms by which antibody responses are maintained over the human life-span remains an open question . In one model , LLPCs survive indefinitely in the bone marrow and independently maintain steady-state antibody levels [8] . Alternative models predict that PCs are replenished by MBCs that proliferate and differentiate in response to persistent [9] or intermittent exposure to antigen , and/or through non-specific by-stander activation ( e . g . cytokines or TLR ligands ) [10] . Unlike PCs , which are terminally-differentiated , MBCs may be maintained through homeostatic proliferation [11] , possibly through exposure to polyclonal stimuli [10] . To address fundamental questions related to the generation and maintenance of MBCs and Abs specific for Pf malaria in children in malaria endemic areas , we conducted a year-long prospective study in a rural village of Mali that experiences an intense , sharply-demarcated six-month malaria season annually . We determined whether Pf infection generates MBCs specific for Pf blood stage antigens , and if so , whether they accumulate with age and cumulative Pf exposure , and also whether their frequency correlates with protection from malaria . In addition , we determined whether acute , symptomatic Pf infection resulted in an increase in the number of Pf-specific MBCs and the levels of specific antibodies , and if so , whether this increase remained stable over a six-month period of markedly reduced Pf transmission . By taking advantage of the tetanus immunization schedule in Mali in which infants and women of child-bearing age are vaccinated , we compared the relative efficiencies of the acquisition of tetanus toxoid ( TT ) - and Pf-specific MBCs and Ab , and also tested three hypotheses: 1 ) that growth of the MBC compartment depends on immunological experience rather than age , 2 ) that Pf infection induces non-specific activation of bystander B cells [12] , [13] , and 3 ) that polyclonal activation during heterologous immune responses is a general mechanism for maintaining MBCs and LLPCs [10] .
In May 2006 we initiated an observational cohort study in Mali to investigate the mechanisms underlying naturally-acquired malaria immunity . A detailed description of the study site and cohort has been reported elsewhere [14] . The study population was an age-stratified , random sample representing 15% of all individuals living in a small , rural , well-circumscribed , non-migratory community where antimalarial drugs were provided exclusively by the study investigators . During a two-week period one month prior to the abrupt onset of the six-month malaria season , we enrolled 225 individuals in four age groups: 2–4 years ( n = 73 ) , 5–7 years ( n = 52 ) , 8–10 years ( n = 51 ) , and 18–25 years ( n = 49 ) . Attendance at scheduled follow-up visits was >99% for children ( 2–10 years ) and 82% for adults ( 18–25 years ) during the one-year study period indicating a high degree of study awareness and participation . For the MBC analysis reported here , a subset of 185 individuals was randomly selected within each of the four age categories . All subsequent data and analysis refer to these 185 individuals . The baseline demographic and clinical characteristics of this subset are shown in Table 1 , according to age group . As previously reported [14] , only three of the characteristics shown in Table 1 were associated with decreased malaria risk in multivariate analysis—increasing age , sickle cell trait ( HbAS ) , and asymptomatic Pf parasitemia at study enrollment . During the one-year study period there were 380 unscheduled clinic visits , during which 219 cases of malaria were diagnosed , five of which met the WHO criteria for severe malaria [15] . Malaria episodes were defined as an axillary temperature ≥37 . 5°C , Pf asexual parasitemia ≥5000 parasites/µL , and a non-focal physical exam by the study physician . As expected in this region of Mali , all malaria cases were confined to a six-month period that began in July , peaked in October , and ended by January ( Fig . 1A ) . The incidence of malaria and the proportion of individuals experiencing at least one malaria episode decreased with age , whereas the time to the first malaria episode increased with age ( Table 2 and Fig . 1B ) . Thus , despite intense annual Pf transmission at this study site , malaria immunity is acquired slowly . We first established baseline levels of IgG+ AMA1- , MSP1- and TT-specific MBCs and Abs in Pf-uninfected , healthy children and adults in May just before the malaria season , a point at which there had been little to no Pf transmission for five months . For this analysis we excluded individuals with asymptomatic Pf parasitemia ( 8 . 7% of total cohort; Table 1 ) , because they showed a decreased risk of malaria and tended toward higher frequencies of AMA1- and MSP1-specific MBCs and levels of Ab ( data not shown ) . We focused our analyses on MBCs and Abs specific for Pf blood-stage antigens because humoral responses are known to be critical to blood-stage immunity [4] . We examined the response to two blood stage proteins , Apical Membrane Antigen 1 ( AMA1 ) and Merozoite Surface Protein 1 ( MSP1 ) , because we had previously studied the MBC and Ab responses to these antigens in vaccine trials of Pf-naïve individuals [16] . This afforded the opportunity to compare the acquisition of B cell memory to the same antigens after vaccination versus natural Pf infection . We express MBC data as ‘MBCs per 106 PBMCs’ , where ‘MBCs’ refers to the number of antibody secreting cells derived from MBCs during the six-day culture , and ‘106 PBMCs’ refers to the number of PBMCs after culture . In the present study , the mean frequency of AMA1-specific MBCs per 106 PBMCs increased with age ( Fig . 2A; 2–4 yr: 1 . 2 [95% CI: 0 . 45–1 . 9]; 5–7 yr: 5 . 0 [95% CI: −0 . 2–10 . 1]; 8–10 yr: 8 . 9 [95% CI: 4 . 9–12 . 9]; 18–25 yr: 37 . 8 [95% CI: 10 . 4–65 . 3]; P<0 . 001 ) , as did the proportion of individuals with detectable AMA1-specific MBCs ( 2–4 yr: 8 . 1%; 5–7 yr: 30 . 8%; 8–10 yr: 50 . 0%; 18–25 yr: 54 . 8%; P<0 . 001 ) . Similarly , AMA1-specific Ab levels and the proportion of individuals seropositive for AMA1-specific Abs increased with age ( Fig . 2A; P<0 . 001 for both comparisons ) . There was a positive correlation between the frequency of AMA1-specific MBCs and Ab levels ( Spearman's correlation coefficient = 0 . 35; P = 0 . 005; Fig . S1 ) . We observed a similar age-associated increase in the frequency of MSP1-specific MBCs , although the overall frequency was lower than that for AMA1-specific MBCs ( Fig . 2B; 2–4 yr: 1 . 2 [95% CI: 0 . 55–1 . 9]; 5–7 yr: 3 . 2 [95% CI: 1 . 2–5 . 2]; 8–10 yr: 5 . 9 [95% CI: 2 . 9–9 . 0]; 18–25 yr: 10 . 3 [95% CI: 6 . 3–14 . 3]; P<0 . 001 ) . Likewise , the proportion of individuals who had detectable MSP1-specific MBCs ( 2–4 yr: 9 . 1%; 5–7 yr: 27 . 8%; 8–10 yr: 34 . 3%; 18–25 yr: 47 . 6%; P = 0 . 001 ) was similar to that for AMA1 . MSP1-specific Ab levels and the proportion of individuals seropositive for MSP1-specific Abs also increased gradually with age ( Fig . 2B; P<0 . 001 for both comparisons ) . There was a positive correlation between the frequency of MSP1-specific MBCs and Ab levels ( Spearman's correlation coefficient = 0 . 34; P = 0 . 004; Fig . S1 ) . Remarkably , despite exposure to 50–60 infective mosquito bites per month at the peak of each malaria season in this area [17] , only approximately half of adults had detectable MBCs specific for AMA1 and MSP1 , even though most had detectable AMA1- and MSP1-specific antibodies . Of the 72 individuals without detectable AMA1-specific MBCs before the malaria season , 64 ( 88 . 9% [95% CI 79 . 3–95 . 1] ) did not have detectable MSP1-specific MBCs , suggesting that failure to generate MBCs to one Pf antigen is associated with failure to generate MBCs to other Pf antigens . To understand if the expansion of Pf-specific MBCs with age was driven by repeated exposure to Pf antigens or simply a function of age , we determined the frequency of MBCs specific for an unrelated antigen , tetanus toxoid ( TT ) , with age . In Mali , a single TT vaccine is administered to infants less than six months of age and a second TT vaccine is administered to females around 15 years of age to prevent neonatal tetanus . Thus , we measured TT-specific antibody and MBC responses at least 18 months after TT vaccination , a point at which the TT-specific response is likely to be at steady state . In contrast to what was observed for AMA1- and MSP1-specific MBCs , the frequency of TT-specific MBCs among males did not change significantly from age 2 to 25 years ( Fig . 2C ) ( 2–4 yrs: 10 . 8 [95% CI −7 . 4–29 . 0] , 5–7 yrs: 7 . 3 [95% CI 0 . 7–13 . 9] , 8–10 yrs: 8 . 0 [95% CI 3 . 1–12 . 8] , 18–25 yrs: 4 . 7 [95% CI 1 . 4–8 . 1]; P = 0 . 80 ) . Similarly , the proportion of male adults who were positive for TT-specific MBCs did not differ significantly from male children ( 2–4 yrs: 25 . 0% , 5–7 yrs: 33 . 3% , 8–10 yrs: 40 . 9% , 18–25 yrs 28 . 6%; P = 0 . 80 ) . The slightly higher frequency of TT-specific MBCs in male versus female children was not statistically significant . However , the frequency of TT-specific MBCs was significantly higher in female adults compared to female children ( Fig . 2C; mean frequency of TT-specific MBCs per million PBMC by age group ( 2–4 yrs: 2 . 9 [95% CI 1 . 1–4 . 7] , 5–7 yrs: 3 . 2 [95% CI 0 . 2–6 . 1] , 8–10 yrs: 3 . 4 [95% CI 1 . 1–5 . 7] , 18–25 yrs: 58 . 7 [95% CI 34 . 2–83 . 3]; P<0 . 001 ) presumably the result of booster vaccination . Likewise , the proportion of female adults who were positive for TT-specific MBCs was significantly higher as compared to female children ( 2–4 yrs: 28 . 1% , 5–7 yrs: 25 . 0% , 8–10 yrs: 27 . 3% , 18–25 yrs 88 . 0%; P<0 . 001 ) . For both females and males , TT-specific Ab levels mirrored MBC frequencies ( Fig . 2C ) —clearly increasing from female children to female adults ( P<0 . 001 ) , while not changing significantly by age in males ( P = 0 . 44 ) . Overall , TT-specific Ab levels and MBC frequencies correlated ( Spearman's correlation coefficient = 0 . 48; P<0 . 001; Fig . S1 ) . The observation that Pf-specific MBCs increased with age while TT-specific MBCs in individuals who received no booster vaccine did not increase and tended to decrease slightly with age indicates that the increase in Pf-specific MBCs is driven by repeated antigen exposure and is not simply a function of age . Of note , the size of the total IgG+ MBC compartment , as reflected in the peripheral blood , increased with age ( Fig . 3; P<0 . 001 ) , consistent with the maturation of the total MBC compartment with immunological experience . To assess the Pf-specific MBC and Ab responses to acute malaria , and to determine the stability of this response during a period of little to no Pf transmission , we measured the frequencies of MBCs and Ab levels specific for AMA1 and MSP1 14 days after the first episode of malaria ( convalescence ) , and in a cross-sectional survey at the end of the following dry season ( month 12 ) , and compared these frequencies to the pre-malaria season baseline ( month 0; as detailed above ) . Malaria episodes were defined as an axillary temperature ≥37 . 5°C , Pf asexual parasitemia ≥5000 parasites/µL , and a non-focal physical exam by the study physician . Because few adults experienced malaria ( Table 2 ) , this analysis only included children aged 2–10 years ( see Fig . 4 for sample sizes at each time point ) . The mean frequency of AMA1-specific MBCs in children aged 2–10 years increased from month 0 to convalescence ( Fig . 4A; month 0: 4 . 7 [95% CI: 2 . 8–6 . 6]; convalescence: 13 . 4 [95% CI: 2 . 7–24 . 1; P = 0 . 006] and then decreased from convalescence to month 12 ( Fig . 4A; month 12: 5 . 9 [95% CI: 2 . 4–9 . 4]; P = 0 . 93 versus convalescence ) to a point just above the frequency at month 0 ( Fig . 4A; P = 0 . 021 , month 0 vs . month 12 ) . Likewise , the level of AMA1-specific Ab increased from month 0 to convalescence ( Fig . 4A; month 0: 422 . 8 [95% CI: 228 . 7–617 . 0]; convalescence: 797 . 2 [95% CI: 460 . 0–1134 . 7; P<0 . 001] , and then decreased from convalescence to month 12 ( Fig . 4A; month 12: 535 . 5 [95% CI: 283 . 8–787 . 2]; P<0 . 001 versus convalescence] , to a point just above month 0 levels ( Fig . 4A; P = 0 . 040 , month 0 vs . month 12 ) . The MSP1-specific MBC and Ab responses followed a similar pattern . The mean frequency of MSP1-specific MBCs in children aged 2–10 years increased from month 0 to convalescence ( Fig . 4B; month 0: 3 . 3 [95% CI: 2 . 0–4 . 6]; convalescence: 4 . 8 [95% CI: 2 . 9–6 . 8; P = 0 . 002] and then decreased from convalescence to month 12 ( Fig . 4B; month 12: 4 . 5 [95% CI: 2 . 4–6 . 6]; P = 0 . 71 versus convalescence ) to a point just above the frequency at month 0 ( Fig . 4B; P = 0 . 156 , month 0 vs . month 12 ) . Likewise , the level of MSP1-specific Ab increased from month 0 to convalescence ( Fig . 4B; month 0: 14 . 6 [95% CI: 10 . 5–18 . 6]; convalescence: 302 . 6 [95% CI: 111 . 7–493 . 4; P<0 . 001] , and then decreased from convalescence to month 12 ( Fig . 4B; month 12: 31 . 1 [95% CI: 5 . 5–56 . 6]; P<0 . 001 versus convalescence] , to a point just above month 0 levels ( Fig . 4B; P = 0 . 052 , month 0 vs . month 12 ) . To determine if malaria induces non-specific activation of ‘bystander’ MBCs , we compared the frequencies of TT-specific MBCs and Ab levels before the malaria season ( month 0 ) to that 14 days after acute malaria ( convalescence ) . We observed a small , but statistically significant increase in the frequency of TT-specific MBCs from month 0 to convalescence ( Fig . 4C; month zero: 7 . 1 [95% CI: 3 . 1–11 . 2]; convalescence: 8 . 4 [95% CI: 5 . 0–11 . 8; P = 0 . 012 ) that did not change significantly at month 12 ( month 12: 9 . 1 [95% CI: 3 . 2–15 . 4]; P = 0 . 974 versus convalescence] . In contrast , TT-specific Ab levels decreased slightly from month 0 to convalescence , and again from convalescence to month 12 , although neither decline was statistically significant ( Fig . 4C; month 0: 0 . 58 [95% CI: 0 . 5–0 . 7]; convalescence: 0 . 57 [95% CI: 0 . 5–0 . 7; P = 0 . 063]; month 12: 0 . 54 [95% CI: 0 . 4–0 . 6]; P = 0 . 525 versus convalescence ) . Collectively these results indicate that malaria infection results in an increase in the frequencies of both Pf-specific , and bystander MBCs . However , malaria selectively induces Pf-specific Ab production but does not appear to drive the differentiation of bystander naïve and memory B cells into PCs . By FACS we determined the proportion of B cell subsets in individuals ( 2–4 yrs [n = 38] , 5–7 yrs [n = 21] , 8–10 yrs [n = 23] , 18–25 yrs [n = 27] ) before the malaria season ( Fig . 5A ) . With increasing age , and as a percentage of total CD19+ B cells we observed a decrease in immature B cells ( CD19+ CD10+; P<0 . 001 ) and naïve B cells ( CD19+ CD27− CD21+ CD10−; P = 0 . 047 ) and an increase in resting IgG+ MBCs ( CD19+ CD27+ CD21+; P<0 . 001 ) and activated IgG+ MBCs ( CD19+ CD27+ CD21− CD20+ CD10−; P<0 . 001 ) . The increase with age of classical MBCs is consistent with the increase in total IgG+ MBCs we observed using the MBC ELISPOT assay ( Fig 3 ) . In a subset of 87 individuals from this same study cohort , we previously reported that Pf exposure is associated with an expanded subset of ‘atypical’ MBCs that express FCRL4 and are hyporesponsive to in vitro stimuli [18] , similar to the ‘exhausted’ MBCs described in viremic , HIV-infected individuals [19] . Atypical MBCs are defined as CD19+ CD27− CD21− CD20+ CD10− and typically represents <4% of circulating CD19+ B cells in healthy U . S . adults [19] . Here , analyzing a larger number of individuals in the cohort , we confirmed that this subset of MBCs is expanded in Malian children and adults compared to malaria-naïve U . S . adults ( U . S . adults: 1 . 4% [95% CI: 0 . 9–1 . 8]; Malian children aged 2–10 years: 10 . 2% [95% CI: 8 . 7–11 . 8] , P<0 . 001 versus U . S . adults; Malian adults aged 18–25 years: 14 . 8% [95% CI: 11 . 0–19 . 1] , P<0 . 001 versus U . S . adults ) . Thus , in addition to the increase in classical MBCs , an ‘atypical’ MBC subset is expanding with age in this study population . We investigated the impact of acute malaria on the relative proportion of B cells in each subset in children aged 2–10 years . Compared to the pre-malaria season baseline ( month 0 ) , there were no significant changes in the percent of lymphocytes that were CD19+ 14 days after acute malaria . Within the CD19+ B cell population there were no significant changes in the percent of immature B cells , naïve B cells , or resting MBCs , after acute malaria . Moreover , there was no change in the proportions of resting and atypical MBCs that were IgG+ . However , we observed a decrease in the percentage of total atypical MBCs ( Fig . 5B; month 0: 10 . 9% [95% CI: 9 . 4–12 . 4] , convalescence: 8 . 7% [95% CI: 7 . 3–10 . 2]; P = 0 . 027 ) , and an increase in activated MBCs following acute malaria ( month 0: 1 . 6 [95% CI: 1 . 2–2 . 0] , convalescence: 1 . 9 [95% CI: 1 . 4–2 . 4]; P = 0 . 09 ) . Within the activated MBC subset there was a significant increase in the proportion that were IgG+ ( month 0: 59 . 0% [95% CI: 56 . 0–62 . 1] , month 12: 62 . 8% [95% CI: 59 . 2–66 . 3]; P<0 . 001 ) . The decrease in the proportion of atypical MBCs in the peripheral blood suggests that this subpopulation may be trafficking out of the circulation into tissues in response to acute malaria . We determined prospectively whether AMA1- or MSP1-specific Ab levels or MBC frequencies measured just prior to the six month malaria season were associated with the subsequent risk of malaria . For this analysis a malaria episode was defined as an axillary temperature ≥37 . 5°C , Pf asexual parasitemia ≥5000 parasites/µL , and a non-focal exam by the study physician . Because the incidence of malaria was very low in adults during the study period ( Table 2 ) , they were excluded from this analysis . Three measures of malaria risk were analyzed: 1 ) whether or not malaria was experienced , 2 ) the incidence of malaria , and 3 ) the time to the first malaria episode . In the corresponding multivariate regression models ( logistic , Poisson , and Cox regression ) which controlled for age , sickle cell trait , and concurrent asymptomatic Pf parasitemia , we found no correlation between malaria risk and AMA1- or MSP1- specific Ab levels or MBC frequencies . As discussed below , this finding was not unexpected based on the observation that the malaria vaccine candidates AMA1 and MSP1 did not confer protection against malaria in clinical trials [20] , [21] .
In this year-long prospective study of children and adults in an area of intense , annual , sharply demarcated Pf transmission , we show that MBCs specific for Pf can be acquired , but only gradually in a stepwise fashion over years of repeated Pf exposure . MBCs specific for two Pf antigens , AMA1 and MSP1 , increased in frequency in response to acute Pf infection , and then contracted during a six-month period of decreased Pf exposure to a point slightly above pre-infection levels . Cross-sectional analysis of individuals aged 2–25 years just before the malaria season indicated that this step-wise , incremental increase in Pf-specific MBCs with each malaria season contributes to the gradual expansion of the Pf-specific MBC compartment with cumulative Pf exposure . By comparison , the stable frequency of TT-specific MBCs with age after immunization in infancy indicates that growth of antigen-specific MBC compartments does not simply occur with age , but requires repeated antigen exposure . We do not formally know if the gradual gain in Pf-specific MBCs is in fact due to an increase in long-lived MBCs , or whether those MBCs require Pf-stimulation and would be lost if Pf transmission did not resume after the six-month dry season . In another setting , namely in an area of Thailand with low Pf transmission , Wipasa et al . [22] recently reported that nearly half of adults studied had acquired long-lived Pf-specific MBCs as a result of infrequent malaria infections . It will be of genuine interest to understand the cellular and molecular mechanisms at play in the generation of MBCs under these very different conditions of exposure of children versus adults as these could have significance with regard to vaccine development . Moreover , recent studies in mouse models are revealing multiple , phenotypically and functionally distinct populations of MBCs [23] , [24] and it will be of interest to further characterize Pf-specific MBCs in different malaria endemic settings . The study described here provides a rare view of the acquisition and maintenance of human B cell memory . Most prospective studies of human B and T cell immunological memory have evaluated responses to vaccination rather than natural infection , in part because of the difficulty of predicting who within a population will be infected with a given pathogen at a given time . In response to a single vaccination , several studies have described an expansion and contraction of vaccine-specific MBCs [25] , [26] and CD8+ memory T cells [27] . In one of the few longitudinal studies of the MBC response to natural infection , Harris et al . examined antigen-specific MBC responses of patients presenting with acute Vibrio cholerae infection , a pathogen that elicits long-term protective immunity against subsequent disease [28] . In contrast to our results , they observed that the majority of patients acquired IgA and IgG MBCs specific for two Vibrio cholerae antigens and that these persisted up to one year after infection . Whereas MBCs mediate recall responses to reinfection by rapidly expanding and differentiating into PCs , LLPCs residing in the bone marrow constitutively secrete antibody in the absence of antigen and thus provide a critical first line of defense against reinfection [6] . Logistical constraints precluded the direct measurement of circulating PCs in this study . However , we took advantage of the discrete six-month dry season , a period of little to no Pf transmission , to infer the relative contributions of SLPCs and LLPCs to the Pf-specific IgG response based on a serum IgG half-life of ∼21 days [29] . Two weeks after acute malaria , AMA1- and MSP1-specific Ab levels increased significantly and then decreased over a six-month period to a point just above pre-infection levels , indicating that the majority of PCs generated in response to acute Pf infection were short-lived . This observation is consistent with previous studies that described rapid declines in Pf-specific Ab within weeks of an acute malaria episode [30] , [31] . We infer that the small net increase in Pf-specific Ab at the end of the six-month dry season represents the acquisition of Pf-specific LLPCs . Because Pf transmission resumes after the six-month dry season , we cannot estimate the long-term decay rate of Pf-specific Ab in the absence of reinfection . It remains to be seen whether long-term decay rates of Pf-specific Ab are comparable to rates of Ab decay after exposure to common viral and vaccine antigens such as mumps and measles , for example , which elicit Ab with half-lives exceeding 200 years [32] . The small incremental gains in AMA1- and MSP1-specific Abs in response to acute malaria mirrors the gradual exposure-related increase in Pf-specific MBCs , consistent with the long-lived Abs being the products of LLPCs derived from MBCs . Unlike the response to some other pathogens , such as measles , which induce long-lived protective Abs after a single exposure , it may be that repeated exposure to the Pf parasite is necessary to ‘fill’ the Pf-specific LLPC compartment to the point where basal levels of circulating Abs to any given Pf antigen reach a protective threshold . In a separate study of this cohort , we observed a similar pattern of transient increases during the malaria season of Abs specific for a large number of Pf antigens using protein microarrays [33] suggesting that malaria induces a relatively high SLPC-to-LLPC ratio that is not exclusively a function of the inherent qualities of any given antigen per se . In contrast to the highly efficient immune response to a single smallpox vaccination , which generates long-lived ( >50 years ) MBCs in nearly all vaccinees [34] , a remarkably high proportion of adults in the present study did not have detectable AMA1- or MSP1-specific MBCs despite annual exposure to 50–60 infective mosquito bites per person per month at the height of the malaria season [17] , similar to what Dorfman et al . observed in a cross-sectional study in Kenya [35] . Importantly , most female adults in the present study had detectable TT-specific MBCs three to ten years after a single TT booster vaccine in adolescence . We previously reported that AMA1- and MSP1-specific MBCs were reliably generated in Pf-naïve U . S . adults following just two vaccinations [16] . Taken together , these observations indicate that the relatively inefficient generation and/or maintenance of Pf-specific MBCs in response to natural Pf infection cannot be ascribed entirely to inherent deficiencies in the antigens themselves . Collectively , these observations raise a central question: if AMA1 and MSP1-specific MBCs and Abs can be efficiently generated by vaccination of Pf-naïve adults , and TT-specific MBCs and Abs can be efficiently generated by vaccination of Pf-exposed individuals in this cohort , what underlies the inefficient acquisition and/or maintenance of AMA1 and MSP1-specific MBCs and Abs in response to natural Pf infection ? One simple answer , in addition to parasite antigenic variation [36] , [37] , might be that the enormous number of antigens encoded by the over 5 , 400 Pf genes overwhelms the immune system's capacity to select for and commit a sufficient number of MBCs and LLPCs specific for any given Pf antigen to a long-lived pool [38] . If immunity to clinical malaria requires high levels of antibodies to a large number of Pf proteins , then the inability to commit large numbers of MBCs and LLPCs specific for any given Pf antigen during any given infection , as shown here , may explain , in part , why malaria immunity is acquired slowly . In this scenario the Pf-infected individual is capable of the normal generation and maintenance of MBCs and LLPCs , but acquiring a sufficient number of MBCs and LLPCs to a large number of antigens may simply take years . It is also possible that Pf infection disrupts the immune system's ability to generate or maintain MBCs or LLPCs . The differentiation of B cells into long-lived MBCs depends to a great extent on the affinity of their BCRs for antigen . Recently , evidence was presented that affinity maturation of B cells may fail to occur in the absence of adequate Toll-like receptor ( TLR ) stimulation [39] . We recently reported that Malian adults were relatively refractory to CpG , a TLR9 agonist incorporated into two subunit malaria vaccine candidates [40] , raising the possibility that the slow acquisition of MBCs observed here may be due to a failure of B cells to undergo affinity maturation during Pf infection . Although our data do not directly address the role of apoptosis in the gradual acquisition of Pf-specific MBCs , it is worth noting that we found no evidence of Pf-induced ablation of Plasmodium-specific MBCs , as was observed in mice four days after Plasmodium yoelii infection [41] . The relatively inefficient response to natural Pf infection also does not appear to be due to a persistent , Pf-induced general immunosuppression as the frequency of TT-specific MBCs increased significantly in most adult females in response to a single TT booster vaccination , an increase that appeared to be maintained for years . In an experimental model of lymphocytic choriomeningitis virus ( LCMV ) infection , a high antigen-to-B cell ratio disrupted germinal center formation and the establishment of B cell memory [42] . It is plausible that a similar mechanism is at play during the blood stage of Pf infection when the immune system encounters high concentrations of parasite proteins . Indeed , germinal center disruption is observed in mice infected with P . berghei ANKA [43] and P . chabaudi [44] . It is also possible that specific parasite products selectively interfere with the regulation of B cell differentiation [45] or with the signals required for sustaining LLPCs in the bone marrow [46] . It is also conceivable that the disproportionately high level of class-switched SLPCs we observed in response to Pf infection arises from pre-diversified IgM+IgD+CD27+ ( marginal zone ) B cells—analogous to the rapid protective response against highly virulent encapsulated bacteria that do not elicit classical T-dependent responses [47] . These and other hypotheses could be tested by applying systems biology methods [48] and targeted ex vivo and in vitro assays to rigorously conducted prospective studies of Pf-exposed populations . We previously reported that Pf exposure is associated with a functionally and phenotypically distinct population of FCRL4+ hypo-responsive atypical MBCs [18] , similar to the ‘exhausted’ MBCs described in HIV-infected individuals [19] . In this study , with a larger sample size , we confirmed that Pf exposure is associated with an expansion of FCRL4+ MBCs . The accumulation of atypical MBCs could be linked to the slow acquisition of Pf-specific MBCs , as naïve B cells in response to Pf infection could have a propensity to differentiate into atypical rather than classical MBCs . We also observed that the FCRL4+ MBC population decreased in the peripheral circulation two weeks after acute malaria suggesting that these MBCs are directly involved in the response to Pf infection , possibly trafficking to secondary lymphoid tissues . Although the function of FCRL4+ MBCs is not established , Moir et al . [19] suggested that FCRL4+ ‘exhausted MBCs’ contribute to the B cell deficiencies observed in HIV-infected individuals . In contrast , Ehrhardt et al . [49] , who first described FCRL4+ ‘tissue-like MBCs’ in lymphoid tissues associated with epithelium , suggested that these cells may play a protective role during infections . At present , the factors that underlie the expansion of atypical MBCs in this study population are not known . Genetic or environmental factors that are associated with Pf transmission but not accounted for in this study could explain this observation . It will be of interest to understand the origin , antigen-specificity , and function of FCRL4+ MBCs in the context of Pf infection and the potential impact of these MBCs on the ability of children to respond to malaria vaccines . In multivariate analysis we found no correlation between the frequency of MBCs and levels of Abs specific for AMA1 or MSP1 and malaria risk . This is not necessarily unexpected in light of recent clinical trials that showed that vaccination with either AMA1 or MSP1 did not confer protection [20] , [21] . Furthermore , we suspect that the frequency of MBCs per se may not reliably predict clinical immunity to malaria regardless of antigen specificity . Malaria symptoms only occur during the blood stages of Pf infection and can begin as early as three days after the blood stage infection begins [50] . Because the differentiation of MBCs into PCs peaks ∼6–8 days after re-exposure to antigen [10] , there may not be sufficient time for MBCs specific for Pf blood stage antigens to differentiate into the antibody-secreting cells that would prevent the onset of malaria symptoms . In contrast , the longer incubation period of other pathogens allows MBCs to differentiate into protective antibody-secreting cells before symptoms develop . For example , follow-up studies of hepatitis B vaccinees have shown that protection can persist despite the decline of hepatitis B-specific antibodies to undetectable levels [51] , presumably due to the recall response of persistent MBCs . Thus , protection against the blood stages of malaria may depend on achieving and maintaining a critical level of circulating antibody that can rapidly neutralize the parasite . MBCs may contribute to the gradual acquisition of protective immunity by differentiating into LLPCs with each Pf infection . Here we also provide evidence concerning the mechanism by which MBCs and LLPCs are maintained . We observed a modest but statistically significant increase in TT-specific MBCs two weeks after acute malaria , in support of the hypothesis that MBCs are renewed by polyclonal or ‘bystander’ activation [10] . The stable frequency of TT-specific MBCs with age suggests that the small increases associated with Pf-induced polyclonal activation are matched by the rate of loss of senescent TT-specific MBCs . It has also been proposed that non-specific polyclonal stimulation maintains long-lived Ab responses by driving MBCs to differentiate into SLPCs or LLPCs [10] . Similarly , it has been hypothesized that Plasmodium infection generates large amounts of non-specific Ig [52] through polyclonal B cell activation [12] , [13] . However , despite the presence of TT-specific MBCs and their expansion following Pf infection , we did not observe a concomitant increase in TT-specific IgG . This finding is consistent with recent human studies that demonstrate a lack of bystander IgG production after heterologous vaccination or viral infection [32] , [53]; as well as studies in mice that demonstrate PC persistence after MBC depletion [54] , and the failure of MBCs to differentiate into PCs in vivo upon TLR4 and 9 activation [55] . This finding does not represent an overt inability of TT-specific MBCs to differentiate into PCs , since adult females in this study had a sharp increase in tetanus IgG after a single tetanus booster . It is possible that bystander MBCs specific for antigens other than TT differentiate into PCs after Pf infection , but based on the results of this study we hypothesize that the preponderance of IgG produced in response to malaria is specific for the ∼2400 Pf proteins expressed during the blood-stage of infection [56] , and that increases in ‘non-specific’ IgG reflect boosting of cross-reactive B cells [57] , [58] . From a basic immunology perspective , these data support a model in which non-specific stimuli contribute to MBC self-renewal , but not to the maintenance of LLPCs . Studies of other Ab specificities and isotypes before and after malaria and other infections would test this hypothesis further . Although a recent mouse study showed that MBCs do not proliferate in vivo after immunization with an irrelevant antigen [59] , this may reflect the difference in requirements for MBC maintenance in mammals with relatively short life spans . It is of general interest to determine which parasite products are responsible for the polyclonal activation of MBCs observed here . Studies in vitro suggest that Pf drives polyclonal MBC activation by the cysteine-rich interdomain regions 1α ( CIDR1α ) of the Pf erythrocyte membrane protein 1 ( PfEMP1 ) [13] , [60] , but it is conceivable that Pf-derived TLR agonists [61] , [62] or bystander T cell help [63] , [64] , [65] also contribute to MBC proliferation in the absence of BCR triggering [66] . Animal models have provided important insights into the immunobiology of Plasmodium infection [67] , but ultimately , despite obvious experimental limitations , it is critical to investigate the human immune response to Pf in longitudinal studies since findings from animal models do not always mirror human biology or pertain to the clinical context [68] , [69] . Key challenges for future studies will be to determine the molecular basis of the inefficient generation of MBCs and LLPCs in response to Pf infection and to determine the longevity of these cells in the absence of Pf transmission over longer periods of time . Greater insight into the molecular and cellular basis of naturally-acquired malaria immunity could open the door to strategies that ultimately prove useful to the development of a highly effective malaria vaccine .
The ethics committee of the Faculty of Medicine , Pharmacy , and Odonto-Stomatology , and the institutional review board at the National Institute of Allergy and Infectious Diseases , National Institutes of Health approved this study ( NIAID protocol number 06-I-N147 ) . Written , informed consent was obtained from adult participants and from the parents or guardians of participating children . This study was carried out in Kambila , a small ( <1 km2 ) rural village with a population of 1500 , located 20 km north of Bamako , the capital of Mali . Pf transmission is seasonal and intense at this site from July through December . The entomological inoculation rate measured in a nearby village was approximately 50–60 infective bites per person per month in October 2000 and fell to near zero during the dry season [17] . A detailed description of this site and the design of the cohort study has been published elsewhere [14] . In May 2006 , during a two-week period just prior to the malaria season , 225 individuals aged 2–10 years and 18–25 years were enrolled after random selection from an age-stratified census of the entire village population . Enrollment exclusion criteria were hemoglobin <7 g/dL , fever ≥37 . 5°C , acute systemic illness , use of anti-malarial or immunosuppressive medications in the past 30 days , or pregnancy . All analysis in the present study pertains to an age-stratified subset of individuals ( n = 185 ) randomly selected from those who had complete sets of PBMC samples over the entire study period . From May 2006 through May 2007 , participants were instructed to report symptoms of malaria at the village health center , staffed 24 hours per day by a study physician . For individuals with signs or symptoms of malaria , blood smears were examined for the presence of Pf . Patients with positive smear results ( i . e . any level of parasitemia ) were treated with a standard 3-day course of artesunate plus amodiaquine , following the guidelines of the Mali National Malaria Control Program . Anti-malarial drugs were provided exclusively by the study investigators . Children with severe malaria were referred to Kati District Hospital after an initial parenteral dose of quinine . For research purposes , a malaria episode was defined as an axillary temperature ≥37 . 5°C , Pf asexual parasitemia ≥5000 parasites/µL , and a nonfocal physical examination by the study physician . Severe malaria , as defined by the WHO [15] , was included in this definition . Three clinical endpoints were used to evaluate the relationship between Pf-specific immune responses and malaria risk: 1 ) whether or not malaria was experienced , 2 ) the incidence of malaria , and 3 ) the time to the first malaria episode . Blood smears were prepared and venous blood samples collected during the two-week enrollment period ( month 0 ) , 14 days after the first episode of malaria ( convalescence ) , and during a two-week period at the end of the six-month dry season ( month 12 ) . Hemoglobin was typed from venous blood samples . Stool and urine were examined at enrollment for the presence of helminth infections . Venous blood samples from ten healthy U . S . adult blood bank donors were analyzed as controls . Travel histories for these U . S . adults were not available , but prior exposure to Pf is unlikely . Blood samples ( 8 ml for children and 16 ml for adults ) were drawn by venipuncture into sodium citrate-containing cell preparation tubes ( BD , Vacutainer CPT Tubes ) and transported 20 km to the laboratory where they were processed within three hours of collection . Plasma and PBMCs were isolated according to the manufacturer's instructions . Plasma was stored at −80°C . PBMCs were frozen in fetal bovine serum ( FBS ) ( Gibco , Grand Island , NY ) containing 7 . 5% dimethyl sulfoxide ( DMSO; Sigma-Aldrich , St . Louis , MO ) , kept at −80°C for 24 hours , and then stored at −196°C in liquid nitrogen . For each individual , PBMC and plasma samples from all time points were thawed and assayed simultaneously . Thick blood smears were stained with Giemsa and counted against 300 leukocytes . Pf densities were recorded as the number of asexual parasites/µl of whole blood , based on an average leukocyte count of 7500/µl . Each smear was evaluated separately by two expert microscopists blinded to the clinical status of study participants . Any discrepancies were resolved by a third expert microscopist . Hemoglobin was typed by high performance liquid chromatography ( HPLC; D-10 instrument; Bio-Rad , Hercules , CA ) as previously described [14] . At enrollment , duplicate stool samples were examined for Schistosoma mansoni eggs and other intestinal helminths using the semi-quantitative Kato-Katz method . To detect Schistosoma haematobium eggs , 10 ml of urine were poured over Whatman filter paper . One or two drops of ninhydrine were placed on the filter and left to air dry . After drying , the filter was dampened with tap water and helminths were eggs detected by microscopy . Latitude and longitude coordinates of study subjects' households were measured by a handheld global positioning system receiver ( GeoXM; Trimble ) and reported earlier [14] . ELISAs were performed by a standardized method as described previously [70] . For both AMA1 and MSP1 , a 1∶1 mixture of FVO and 3D7 AMA1 and MSP1 isotypes was used to coat the ELISA plates . The limit of detection for the AMA1 and MSP1 ELISA is based on the range of values that gives reproducible results at the Malaria Vaccine and Development Branch at NIAID where the assay is routinely performed . More specifically , the limit of detection is the ELISA unit value at the lowest point on the standard curve , multiplied by the dilution factor at which samples are tested . The minimal detection levels for the MSP1 and AMA1 ELISA assays were 11 and 33 ELISA units , respectively . For analysis , all data below the minimum detection level were assigned a value of one half the limit of detection ( i . e . 6 units for MSP1 , 17 units for AMA1 ) . The limit of detection for the TT ELISA was not determined because we did not have access to TT-naïve serum . Antigen-specific MBCs were quantified by a modified version of the method developed by Crotty et al [71] . We found that adding IL-10 to the cocktail of polyclonal activators resulted in a six-fold increase in the efficiency of the assay ( Weiss et al . , unpublished observation ) . Briefly , PBMCs were thawed and cultured in 24 well plates at 37°C in a 5% CO2 atmosphere for six days in media alone ( RPMI 1640 with L-Glutamine , Penicillin/Streptomycin 100 IU/ml , 10% heat-inactivated FBS , 50 µM β-Mercaptoethanol ) or media plus a cocktail of polyclonal activators: 2 . 5 µg/ml of CpG oligonucleotide ODN-2006 ( Eurofins MWG/Operon , Huntsville , AL ) , Protein A from Staphylococcus aureus Cowan ( SAC ) at a 1/10 , 000 dilution ( Sigma-Aldrich , St . Louis , MO ) , pokeweed mitogen at a 1/100 , 000 dilution ( Sigma-Aldrich ) , and IL-10 at 25 ng/ml ( BD Biosciences ) . Cells were washed and distributed on 96-well ELISPOT plates ( Millipore Multiscreen HTS IP Sterile plate 0 . 45 um , hydrophobic , high-protein binding ) to detect antibody-secreting cells ( ASCs ) . ELISPOT plates were prepared by coating with either: a 10 µg/ml solution of polyclonal goat antibodies specific for human IgG ( Caltag ) to detect all IgG-secreting cells; a 1% solution of bovine serum albumin ( BSA ) as a non-specific protein control; or 5 µg/ml solutions of either tetanus toxoid ( TT ) , AMA1 , or MSP1 to detect antigen-specific ASCs . For AMA1 and MSP1 , a 1∶1 mixture of FVO and 3D7 isotypes was used to coat the ELISPOT plates . Plates were blocked by incubation with a solution of 1% BSA . For the detection of antigen-specific ASCs , cells were plated in duplicate in eight serial dilutions beginning with 5×105 cells/well . For detection of total IgG ASCs cells were plated at six serial dilutions beginning at 4×104 cells/well . After a five hour incubation of the cells in the ELISPOT plates , plates were washed four times each in PBS and PBS-Tween 20 0 . 05% ( PBST ) , and incubated overnight with a 1∶1000 dilution of alkaline phosphatase-conjugated goat antibodies specific for human IgG ( Zymed ) in PBST/1% FCS . Plates were washed four times each in PBST , PBS , and ddH2O; developed using 100 µl/well BCIP/NBT for 10 minutes; washed thoroughly with ddH2O and dried in the dark . ELISPOTS were quantified using Cellular Technologies LTD plate-reader and results analyzed using Cellspot software . Results are reported as frequencies of MBCs per 106 PBMCs after the six-day culture . The limit of detection of the MBC ELISPOT assay for this analysis was five ASCs per 106 PBMC based on the average number of ASCs on the BSA control . Assay failure was defined as fewer than 1000 IgG+ ASCs per 106 PBMCs after the six-day culture which resulted in the exclusion of 15% of individuals at month 0 , 13 . 2% 14 days after the first malaria episode , and 7 . 3% at month 12 . For individuals with a limited number of PBMCs , priority was given to performing the ELISPOT assay for MSP1 , then TT , and then AMA1 . All phenotypic analyses were performed using mouse mAbs specific for human B cell markers conjugated to fluorophores as previously reported [18] . Fluorophore-conjugated mAbs specific for the following markers were used: PECy7-CD19 , PE-CD20 , APC-CD10 , APC-CD27 , PE-IgG ( BD Biosciences , San Jose , CA ) and FITC-CD21 ( Beckman Coulter , Fullerton , CA ) . A four-color , two-stain strategy was used to identify B cell subsets as follows: plasma cells/blasts ( CD19+ CD21− CD20− ) , naive B cells ( CD19+ CD27−CD10− ) , immature B cells ( CD19+ CD10+ ) , classical MBCs ( CD19+ CD27+ CD21+ ) , atypical MBCs ( CD19+ CD21− CD27− CD10− ) and activated MBCs ( CD19+ CD21− CD27+CD20+ ) . FACS analyses were performed on a FACSCalibur flow cytometer ( BD Biosciences ) using FlowJo software ( Tree Star , Ashland , OR ) . Data were analyzed using STATA ( StataCorp LP , Release 10 . 0 ) and GraphPad Prism for Windows ( GraphPad Software , version 5 . 01 ) . The Kruskal-Wallis test was used to compare continuous variables between groups , and the Fisher's exact test was used to compare categorical variables . The Wilcoxon matched-pairs signed-rank test was used to compare measurements of the same parameter at two time points for the same individual . The correlation between different continuous measures was determined by using the Spearman correlation coefficient . The malaria-free probability over the twelve-month study period was estimated by the Kaplan-Meier curve , and the time to the first malaria episode was compared by the log rank test . Cox's proportional hazards model was used to assess the effect of the following factors on the hazard of malaria: age , gender , weight , ethnicity , distance lived from study clinic , self-reported bednet use , hemoglobin type , antigen-specific MBC frequencies and Ab levels . The same list of variables was included in logistic and Poisson regression models to determine their impact on the odds and incidence of malaria episodes , respectively . For all tests , two-tailed p values were considered significant if ≤0 . 05 .
|
Plasmodium falciparum ( Pf ) is a mosquito-borne parasite that causes over 500 million cases of malaria annually , one million of which result in death , primarily among African children . The development of an effective malaria vaccine would be a critical step toward the control and eventual elimination of this disease . To date , most licensed vaccines are for pathogens that induce long-lived protective antibodies after a single infection . In contrast , immunity to malaria is only acquired after repeated infections . Antibodies play a key role in protection from malaria , yet several studies indicate that antibodies against some Pf proteins are generated inefficiently and lost rapidly . The cells that are responsible for the maintenance of antibodies over the human lifespan are memory B-cells and long-lived plasma cells . To determine how these cells are generated and maintained in response to Pf infection , we conducted a year-long study in an area of Mali that experiences a six-month malaria season . We found memory B-cells and long-lived antibodies specific for the parasite were generated in a gradual , step-wise fashion over years despite intense Pf exposure . This contrasts sharply with the efficient response to tetanus vaccination in the same population . This study lends new insights into the delayed acquisition of malaria immunity . Future studies of the cellular and molecular basis of these observations could open the door to strategies for the development of a highly effective malaria vaccine .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"infectious",
"diseases",
"immunology/immunomodulation",
"immunology/immune",
"response",
"public",
"health",
"and",
"epidemiology/global",
"health",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"immunology",
"infectious",
"diseases/protozoal",
"infections",
"infectious",
"diseases/tropical",
"and",
"travel-associated",
"diseases",
"immunology/immunity",
"to",
"infections"
] |
2010
|
The Plasmodium falciparum-Specific Human Memory B Cell Compartment Expands Gradually with Repeated Malaria Infections
|
Self-organized criticality is an attractive model for human brain dynamics , but there has been little direct evidence for its existence in large-scale systems measured by neuroimaging . In general , critical systems are associated with fractal or power law scaling , long-range correlations in space and time , and rapid reconfiguration in response to external inputs . Here , we consider two measures of phase synchronization: the phase-lock interval , or duration of coupling between a pair of ( neurophysiological ) processes , and the lability of global synchronization of a ( brain functional ) network . Using computational simulations of two mechanistically distinct systems displaying complex dynamics , the Ising model and the Kuramoto model , we show that both synchronization metrics have power law probability distributions specifically when these systems are in a critical state . We then demonstrate power law scaling of both pairwise and global synchronization metrics in functional MRI and magnetoencephalographic data recorded from normal volunteers under resting conditions . These results strongly suggest that human brain functional systems exist in an endogenous state of dynamical criticality , characterized by a greater than random probability of both prolonged periods of phase-locking and occurrence of large rapid changes in the state of global synchronization , analogous to the neuronal “avalanches” previously described in cellular systems . Moreover , evidence for critical dynamics was identified consistently in neurophysiological systems operating at frequency intervals ranging from 0 . 05–0 . 11 to 62 . 5–125 Hz , confirming that criticality is a property of human brain functional network organization at all frequency intervals in the brain's physiological bandwidth .
Critical dynamics are recognized as typical of many different physical systems including piles of rice or sand , earthquakes and mountain avalanches . Dynamic systems in a critical state will generally demonstrate scale-invariant organization in space and/or time , meaning that there will be similar fluctuations occurring at all time and length scales in the system . In other words , there is no characteristic scale to critical dynamics , which will be optimally described by scale-invariant or fractal metrics . Thus , power law or fractal scaling has been widely accepted as a typical empirical signature of non-equilibrium systems in a self-organized critical state [1] , although the existence of power law scaling does not by itself prove that the system is self-organized critical ( SOC ) . For example , turbulence is a conceptually distinct class of dynamics , which is also characterized by self-similar or scale-invariant energy cascades , that can be empirically disambiguated from criticality [2] , [3] . The existence of power laws for the spatial and temporal statistics of critical systems is compatible with the related observations that the dynamics of individual units or components of such systems will show long-range correlations in space and time , and change in state of a single unit can rapidly trigger macroscopic reconfiguration of the system . Many of these phenomena can be studied using computational models of dynamic systems such as the Ising model of magnetization ( see Figure 1 ) and the Kuramoto model of phase coupled oscillators ( see Figure 2 ) . In both these models , the dynamics can be controlled by continuous manipulation of a single parameter . For the Ising model , this control parameter is the temperature; whereas for the Kuramoto model it is the strength of coupling between oscillators . In both cases , as the control parameter is gradually increased ( or decreased ) , the dynamics of the systems will pass through a phase transition , from an ordered to a random state ( or vice versa ) , at which point the emergence of power law scaling and other fractal phenomena will be observed at the so-called critical value of the control parameter . Self-organized critical systems differ from these computational models in the sense that they are not driven to the cusp of a phase transition by external manipulation of an control parameter but instead spontaneously evolve to exist dynamically at that point . Self-organized criticality is an intuitively attractive model for functionally relevant brain dynamics [4]–[7] . Many cognitive and behavioral states , including perception , memory and action , have been described as the emergent properties of coherent or phase-locked oscillation in transient neuronal ensembles [8]–[11] . Critical dynamics of such neurophysiological systems would be expected to optimize their capacity for information transfer and storage , and would be compatible with their rapid reconfiguration in response to changing environmental contingencies , conferring an adaptive ability to switch quickly between behavioral states [12] . In support of the criticality model for brain dynamics , there is already considerable evidence for fractal or power law scaling of anatomically localized neurophysiological processes - including spike frequency , synaptic transmitter release , endogenous EEG and fMRI oscillations [13]–[15] - measured on a wide range of spatial and time scales . However , there have been fewer direct demonstrations of critical dynamics of anatomically distributed neurophysiological systems . Beggs , Plenz and colleagues [16]–[18] have provided empirical evidence of criticality for neuronal network dynamics , represented by a power law probability distribution for the number of electrodes simultaneously recording spike activity in multielectrode array recordings of cortical slices , consistent with the fairly frequent occurrence of neuronal “avalanches” . At the much larger spatial scale of human magnetoencephalography ( MEG ) , the topology of small-world human brain functional networks was found to be self-similar over a range of frequency scales , and the network's topology at each scale was consistent with dynamics close to the critical point of transition from macroscopically chaotic to ordered states [12] . Here we provide more direct evidence for critical dynamics of human brain functional networks measured using both functional magnetic resonance imaging ( fMRI ) and MEG . We focused on two measures of the phase synchronization between component processes of a dynamic system ( which are defined more formally later ) : the phase lock interval ( PLI ) and the lability of global synchronization . The phase lock interval is simply the length of time that a pair of bandpass filtered neurophysiological signals , simultaneously recorded from two different MEG sensors or two different brain regions in fMRI , are in phase synchronization with each other . Thus it is a measure of functional coupling between an arbitrary pair of signals in the system . The lability of global synchronization is a measure of how extensively the total number of phase locked pairs of signals in the whole system can change over time . A globally labile system will experience occasional massive coordinated changes in coupling between many of its component elements . In this sense , global lability is informally analogous to the measure of neuronal “avalanches” introduced by Beggs & Plenz ( 2003 ) to describe simultaneous spiking of large numbers of cells in a multielectrode array measurement of spontaneous neuronal activity . In order to calibrate the behavior of these two synchronization metrics in relation to unquestionably critical dynamics , we first applied them to analysis of the Ising and Kuramoto models as their control parameters were manipulated systematically . These preliminary analyses of two mechanistically distinct computational models demonstrated that the probability distributions of both synchronization metrics followed a power law specifically when the models were in a critical state . This suggested that power law scaling of network synchronization was indicative of critical dynamics regardless of differences in the mechanistic interactions between components in the two models . On this basis , we proceeded to investigate the behavior of these synchronization metrics in neurophysiological data recorded from healthy human volunteers using functional MRI and MEG .
All experimental studies on human subjects were conducted according to the principles of the Declaration of Helsinki and the standards of Good Clinical Practice . All participants provided informed consent in writing . The study protocols were ethically approved by the Addenbrooke's NHS Trust Research Ethics Committee and the Cambridgeshire 2 Research Ethics Committee , Cambridge UK .
Power law scaling of human neurophysiological processes has been previously described in both functional MRI and MEG or electrophysiological datasets [14] , [15] , [43] . However , we believe this is the first demonstration of power law scaling of synchronization metrics in human brain networks . It was notable that although power law scaling was demonstrated for all frequency intervals in both fMRI and MEG data , and for all anatomical pairs of regions in the fMRI data , the value of the scaling exponent was variable in relation to both the modularity and the frequency interval of the networks . Thus the scaling exponent of the PLI distribution was smaller for ( intra-modular ) pairs of fMRI signals belonging to the same functional module than for ( inter-modular ) pairs belonging to different functional modules , indicating that prolonged periods of phase locking are generally more likely to occur between functionally related processes . These results are consistent with previous findings that intra-modular pairs of fMRI signals are more strongly correlated than inter-modular pairs [39] , probably a direct consequence of their stronger structural connectivity [44] , as demonstrated in simulations of hierarchical dynamics on the cortical network of the cat [45] . The dependency of on frequency interval of the networks was most evident by analysis of the MEG data . Here we found that tended to be smaller , indicating a higher probability of long periods of phase locking , in lower frequency networks . This observation is also consistent with prior work demonstrating that wavelet correlations between MEG sensors increase with decreasing frequency in a theoretically predictable way [46] . These considerations suggest that scaling of synchronization metrics , although novel in the context of human neuroimaging , has a profile of variability that makes sense in the context of prior observations on functional brain networks . However , it is important to bear in mind some methodological caveats when evaluating the empirical demonstrations of power law scaling we have reported . Given that an ideal scale-free system has no well defined limits in space or time , the “lack of infinity” in the data we are studying inevitably has some effect on our results . In particular , the finite length of our time series prevents us from estimating the phase lock interval ( PLI ) distribution for time intervals longer than about 10 minutes , which will be counted in the last bin of the PLI histogram , corresponding to the duration of the time-series . This will also impact on the estimation of the power law exponent and its impact will be greater for estimation of scaling parameters in lower frequency networks , where we have fewer data points per time series . Therefore we should be cautious about strong interpretations of the absolute value of the estimated power law exponents . Finite size effects are also clearly visible in the estimated probability distributions of the lability of global synchronization . These are seen not only in the experimental fMRI and MEG data , but also in both computational models , and are a direct consequence of the finite spatial extent of the system in terms of a limited number of pairs . However , it is important to note that the probability distributions of lability are distinct for surrogate data compared to experimental data in all cases . Also , the probability distribution of the number of synchronized pairs ( not shown ) for the surrogate data is a narrow Gaussian centered around a small , whereas the experimental and simulated data display a much wider non-Gaussian distribution with a comparatively large number of synchronized pairs on average , limited only by the system size . The importance of finite size effects in interpreting the shape of the empirically estimated probability distributions motivated us to test formally for the goodness-of-fit of a power law distribution ( compared to exponential and log normal distributions ) for the probability of PLI and lability of global synchronization in the Kuramoto model at critical coupling strength , and both the fMRI and MEG data , at all scales; see Table 1 . These results indicate that the power law form is quite consistently the best fitting model for the probability distribution across all datasets and scales . The added value of this analysis is arguably twofold . First , it indicates that phase synchronization is likely to be an important mechanism of functional network formation at all frequencies and in the endogenous or resting state . Phase synchronization of spatially distributed neurophysiological processes is already accepted as a key mechanism in the transient formation of neuronal ensembles coding the representation of perceived objects or memories [9] , [10] . However , most attention has focused on phase synchronization in high frequency intervals , e . g . , the gamma frequency band ( 30–80 Hz ) , and in response to experimentally controlled stimulation [47] . Our results show that intermittent periods of phase-locking , sometimes for long time intervals , are characteristic of endogenous human brain network dynamics . By analogy to the experimental data demonstrating changes in gamma synchronization in response to conscious perception of external objects , one might speculate that spontaneously occurring periods of phase synchronization might represent changes in subjective mental state , or conscious perception of internal objects . In any case , it is clear from these data that intermittent phase synchronization of neurophysiological systems is a general intrinsic property of the brain and not restricted to certain frequency bands or stimulus conditions . We would also draw attention to an analogy between the neuronal avalanches previously described in multielectrode array recordings [16]–[18] , which represent rapid simultaneous changes in spiking coordinated across a large number of individual neurons , and the scaling behavior of our measure of the lability of global synchronization , which indicates the potential for whole brain systems to demonstrate rapid and extensive changes in global phase locking . This analogy seems consistent with the fundamental principle of scale invariance in understanding critical systems: qualitatively similar network dynamics can be expected at very different ( cellular versus whole brain ) spatial scales . The second and main theoretical implication of these results is that they provide direct empirical support for the hypothesis that human brain networks exist dynamically in a critical state . Criticality has been studied most intensively to date in simulated neural networks . These studies indicate that networks at a critical point between order and chaos are optimized for information transmission , and generate a maximum number of metastable global states , conferring a high capacity for information storage [7] , [18] . Critical systems rapidly adapt to minimal exogenous perturbation [26] , which could have obvious selection advantages for a nervous system . It has also been shown that critical dynamics can emerge by the operation of biologically plausible rewiring rules on initially random networks . For example , a Hebbian rewiring rule , whereby connections are formed between nodes with highly correlated activity ( and deleted between nodes with poorly correlated activity ) , led to the self-organization of critical dynamics in an initially random network [48] . Likewise , when connectivity between neurons was modified by a spike timing dependent plasticity rule , critical dynamics emerged in a functional network with small-world topology [49] . A small-world network is characterized by short average path length between nodes , but large clustering coefficient [50] . This architecture can deliver high efficiency of information processing for low connection costs and is common to many systems such as the internet , the global air transport network and the proteome , as well as the brain . The link between critical dynamics and small-world topology is also implicit in our results , given the strong prior evidence for small-world properties of human brain functional networks derived from fMRI and MEG data [12] , [42] . A key , unresolved question concerns the cognitive or mental significance of brain systems criticality . There is very little empirical data directly supporting the important theoretical connection between critical brain dynamics and the adaptivity or versatility of the behavioral repertoire the brain can support . However , it has been reported that changes in the power law scaling exponents of human MEG sensors were highly predictive of success in discriminating low intensity visual stimuli [51] , suggesting that critical dynamics can indeed be related to optimal perceptual function . An intriguing study in a substantively different biological system , namely gene expression changes in the macrophage following pathogen challenge , found evidence of critical dynamics in normal intra-cellular signaling and non-critical dynamics in cells that had been behaviorally impaired in their response to pathogens by specific gene knockouts , implying that criticality in this signaling system conferred an adaptivity advantage [52] . A key challenge for future studies will be to define more precisely how the parameters of critical network dynamics , empirically estimated in neuroimaging data , can be related to adaptivity and optimality of human cognitive and behavioral performance .
|
Systems in a critical state are poised on the cusp of a transition between ordered and random behavior . At this point , they demonstrate complex patterning of fluctuations at all scales of space and time . Criticality is an attractive model for brain dynamics because it optimizes information transfer , storage capacity , and sensitivity to external stimuli in computational models . However , to date there has been little direct experimental evidence for critical dynamics of human brain networks . Here , we considered two measures of functional coupling or phase synchronization between components of a dynamic system: the phase lock interval or duration of synchronization between a specific pair of time series or processes in the system and the lability of global synchronization among all pairs of processes . We confirmed that both synchronization metrics demonstrated scale invariant behaviors in two computational models of critical dynamics as well as in human brain functional systems oscillating at low frequencies ( <0 . 5 Hz , measured using functional MRI ) and at higher frequencies ( 1–125 Hz , measured using magnetoencephalography ) . We conclude that human brain functional networks demonstrate critical dynamics in all frequency intervals , a phenomenon we have described as broadband criticality .
|
[
"Abstract",
"Introduction",
"Methods",
"and",
"Materials",
"Discussion"
] |
[
"physics/interdisciplinary",
"physics",
"neuroscience/theoretical",
"neuroscience",
"computational",
"biology/computational",
"neuroscience"
] |
2009
|
Broadband Criticality of Human Brain Network Synchronization
|
Most genetic variants associated with disease occur within regulatory regions of the genome , underscoring the importance of defining the mechanisms underlying differences in regulation of gene expression between individuals . We discovered a pair of co-regulated , divergently oriented transcripts , AQY2 and ncFRE6 , that are expressed in one strain of Saccharomyces cerevisiae , ∑1278b , but not in another , S288c . By combining classical genetics techniques with high-throughput sequencing , we identified a trans-acting single nucleotide polymorphism within the transcription factor RIM101 that causes the background-dependent expression of both transcripts . Subsequent RNA-seq experiments revealed that RIM101 regulates many more targets in S288c than in ∑1278b and that deletion of RIM101 in both backgrounds abrogates the majority of differential expression between the strains . Strikingly , only three transcripts undergo a significant change in expression after swapping RIM101 alleles between backgrounds , implying that the differences in the RIM101 allele lead to a remarkably focused transcriptional response . However , hundreds of RIM101-dependent targets undergo a subtle but consistent shift in expression in the S288c RIM101-swapped strain , but not its ∑1278b counterpart . We conclude that ∑1278b may harbor a variant ( s ) that buffers against widespread transcriptional dysregulation upon introduction of a non-native RIM101 allele , emphasizing the importance of accounting for genetic background when assessing the impact of a regulatory variant .
Incorporating genomic information into clinical practice is a major focus of personalized medicine . Despite the discovery of a large number of disease-associated genetic variants [1 , 2] , few clinical treatments have been developed that take into account such information [3] . Furthermore , most disease-associated variants occur within regulatory regions of DNA [4 , 5] , making it particularly difficult to predict the biological processes they affect . Determining the mechanisms by which variants influence regulation , and hence phenotypic diversity among individuals , is paramount to a thorough understanding of functional genomics . Uncovering the biological mechanisms underlying regulatory variants , as well as how variants interact with the myriad of genetic backgrounds present within a population , is a major focus of current research . It has become increasingly clear that genetic background contributes to phenotypes [6–11] resulting in the seemingly infinite diversity observed in nature , even within an individual species . Recent studies suggest that transcript levels can be both powerful readouts for and determinants of disease states [12–15] . However , similar to other cellular phenotypes , expression differences among individuals are the product of an exceedingly complex genetic landscape . One reason for this complexity is that even subtle mutations in transcription factors can impart distinct regulatory roles when present within alternative genetic contexts [16 , 17] . Strategies linking genetic variants to regulation of gene expression often consist of one of two approaches . Expression quantitative trait loci ( eQTL ) studies combine genome-wide expression data with genome sequence information to uncover expression-influencing variants , including those linked to disease [18–24] . In addition to variants themselves , eQTL studies have uncovered many important genetic principles . For example , expression-influencing variants that occur near the gene being regulated , or in cis , tend to influence a single gene , whereas variants distal to the gene being regulated , or in trans , typically influence expression of many loci [25–29] . In contrast to the genome-wide eQTL approach , which correlates genetic variants with expression differences , much of our knowledge of the mechanisms underlying gene expression regulation comes from detailed , single-locus studies [30–34] . However , such studies usually do not consider the effects of naturally occurring genetic variation on gene expression . We sought to combine attributes of genome-wide and single-locus studies to understand the mechanistic basis by which genetic variants result in altered regulation of gene expression between two strains of Saccharomyces cerevisiae . Here we describe principles underlying the complexity of gene expression regulation and report evidence that genetic background strongly influences the extent to which a variant affects transcript levels throughout the genome .
Two strains of S . cerevisiae , S288c and ∑1278b , are a model system for understanding how intraspecies genome sequence variation impacts phenotype [35] . We initially sought to identify a locus where we could directly test how naturally occurring genetic variation impacts transcription factor ( TF ) binding and associated transcript levels . Ideally , the locus would harbor at least one SNP within a predicted TF binding site , allowing us to directly test the effect of the SNP on TF binding . If TF binding is modulated by the SNP , we could subsequently determine its influence on nearby transcript levels , effectively defining a mechanism by which genome sequence influences expression in cis . Specifically , we aimed to identify a genomic region that displays strain-specific binding of a TF that correlates with a nearby strain-specific expression difference . In order to correlate TF binding with gene expression , we first performed strand-specific RNA-seq on S288c and Σ1278b . Not surprisingly given the high degree of sequence similarity ( 99 . 7% ) between the strains , the majority of transcripts are expressed at similar levels between the strains ( S1A Fig ) . However , about 20% of genes are differentially expressed ( DESeq , n = 1207 , Padj≤0 . 0005 , minimum average expression ≥ 100 reads ) ( S1B Fig and S4 Table ) . Of the differentially expressed genes , gene ontology ( GO ) terms are enriched for categories such as transcription factor activity , mRNA binding , and oxidoreductase activity ( Pval≤0 . 002 ) ( S2 Table ) . In addition to protein-coding genes , we identified 82 differentially expressed antisense transcripts ( DESeq , n = 82 , Padj≤0 . 0005 , minimum average expression ≥ 50 reads ) ( S1C Fig and S5 Table ) . To study how expression patterns are gained or lost throughout intraspecies evolution , we initially focused on loci displaying an extreme differential expression phenotype between the strains ( i . e . on in one strain and off in the other ) ( n = 62 , S1 Table ) . To identify a potentially regulatory SNP involved in the birth or death of a transcript , we examined the promoter regions of all 62 “extreme expressors . ” Roughly 25% of extreme expressors harbor a SNP within 50 basepairs of their transcription start site . In one case , a SNP is located very near the transcription start site of a non-coding RNA , ncFRE6 , which is transcribed in an antisense orientation to the FRE6 ORF in ∑1278b , but not in S288c ( S2B Fig ) . Closer inspection of the DNA sequence surrounding the ncFRE6-associated SNP revealed that a consensus Reb1 binding motif is interrupted in S288c relative to ∑1278b ( S2B Fig , Red line ) . We hypothesized that Reb1 binding activates ncFRE6 expression in ∑1278b relative to S288c . In parallel with RNA-seq , we monitored occupancy of Reb1 in S288c and ∑1278b by ChIP-seq ( S1 Text ) . Because REB1 shares 100% sequence identity between S288c and ∑1278b it is not surprising that most Reb1 binding events are conserved between the backgrounds ( 83% of total binding events ) ( S2A Fig ) . However , there are a small number of strain-unique Reb1 binding events , including one at the position of the ncFRE6-associated SNP in ∑1278b and not in S288c . Because the SNP disrupts a preferred Reb1 binding site in S288c relative to ∑1278b we interconverted the SNP between backgrounds in an attempt to rescue binding in S288c and/or abolish binding in ∑1278b . We used the “delitto perfetto” method [36] to interconvert the SNP and observed that it is indeed necessary and sufficient to cause the Reb1 binding discrepancy between the strains ( S2C Fig ) . However , we were surprised to observe that abolishing Reb1 binding in ∑1278b did not reduce expression of ncFRE6 , and rescuing binding in S288c did not increase expression of ncFRE6 in S288c ( S2D Fig ) . This result highlights the importance of single locus studies for uncovering the causal variant ( s ) driving differences in transcript levels rather than assuming that correlations between TF binding and transcript levels are meaningful . Since differential Reb1 binding does not cause the differential expression of ncFRE6 between S288c and ∑1278b , we asked whether other cis elements influenced expression . AQY2 is a divergently oriented gene that originates ~1kb upstream of the ncFRE6 transcription start site and is also expressed specifically in ∑1278b . AQY2 encodes a water channel that is disrupted by a premature stop codon in the vast majority of sequenced strains of S . cerevisiae , including S288c , but is functional in ∑1278b [37] . The AQY2/ncFRE6 promoter region has undergone significant genetic drift between S288c and ∑1278b . Harboring 21 SNPs , the AQY2/ncFRE6 intergenic region is one of the most sequence-variable promoters between S288c and ∑1278b ( S3 Fig ) . Because a large number of SNPs within the region , both intergenic and within the body of each transcript , disrupt potential TF binding sites ( Fig 1A , Grey box , and S3 Table ) , we hypothesized that one or more of the SNPs drive the differential expression of AQY2 and/or ncFRE6 . Indeed , replacing all 30 SNPs in ∑1278b with those from S288c results in ~75% reduction of both AQY2 and ncFRE6 and replacing only the 15 AQY2-proximal SNPs results in ~50% reduction in the transcripts , indicating that DNA elements in both halves of the intergenic region contribute to expression levels of both AQY2 and ncFRE6 in Σ1278b ( Fig 1B ) . Surprisingly , the expression levels of both transcripts were reduced by nearly identical magnitudes in ∑1278b promoter-altered strains , implying that the two divergently oriented transcripts are co-regulated in cis . However , while introducing the S288c cis context into Σ1278b dramatically reduced expression of the transcripts , incorporation of 30 ∑1278b SNPs into S288c was completely ineffective at increasing AQY2 and/or ncFRE6 transcript levels ( Fig 1C ) . Taken together , these results indicate that AQY2 and ncFRE6 are likely co-regulated and that a trans-acting factor ( s ) ultimately determines whether AQY2 and/or ncFRE6 are expressed . To learn about the genetic nature of the trans factor ( s ) that causes differential regulation of AQY2 and ncFRE6 between S288c and ∑1278b , we crossed the two strains and monitored expression of the transcripts in a heterozygous diploid . Neither AQY2 nor ncFRE6 expression is observed in the diploid strain ( Fig 2A ) , implying that the S288c expression phenotype is dominant . To determine whether one or more trans factors control expression of AQY2 and/or ncFRE6 , we performed tetrad analysis assaying for expression of both AQY2 and ncFRE6 . For each tetrad analyzed [38] , two haploid segregants express both AQY2 and ncFRE6 concurrently while the other two express neither transcript ( Fig 2B ) . This 2:2 pattern of inheritance suggests that a single trans factor controls the on/off state of both transcripts , and that co-expression of the transcripts is conserved even in the unique genetic admixtures of the segregants . We Sanger sequenced the AQY2/ncFRE6 cis context within each haploid segregant to determine whether the S288c cis context exhibits a similar level of promoter activity as the ∑1278b cis context in terms of expression of AQY2/ncFRE6 ( Fig 2B , Red boxes harbor S288c AQY2/ncFRE6 cis context ) . Indeed , both the S288c and Σ1278b cis contexts permit expression of AQY2/ncFRE6 , but only in the absence of the trans factor . Furthermore , expression levels varied considerably between AQY2/ncFRE6-expressing segregants . Contrary to the results of the promoter swapped ∑1278b strain ( Fig 1B ) , the segregants that harbor the S288c cis context tend to express higher levels of AQY2/ncFRE6 than those harboring the ∑1278b cis context ( Fig 2B and 2C ) . This result implies that there are additional factors that alter the expression levels of AQY2/ncFRE6 , but only within genetic backgrounds that lack the epistatic trans-factor . Furthermore , given that aqy2 codes for a non-functional protein in S288c , it is somewhat surprising that the S288c cis context possesses robust promoter activity . In order to map the genetic location of the trans factor that causes differential expression of AQY2/ncFRE6 between S288c and Σ1278b , we combined bulked segregant analysis [39] with high throughput sequencing . A similar approach was developed previously using microarrays to map complex phenotypes influenced by a large number of loci [40] . We reasoned that the single dominant repressor would be present in all of the non-expressing segregants of an S288c x ∑1278b cross . Therefore , the variant driving differential expression of AQY2/ncFRE6 should always segregate according to the repression phenotype ( S4 Fig ) . We isolated genomic DNA from 28 segregants: 14 that express AQY2 and ncFRE6 and 14 that do not . We pooled equal amounts of DNA from each strain in the two sets and performed high throughput sequencing of the pools . We then sought to identify regions of the genome inherited exclusively from S288c in the non-expressing strains and from Σ1278b in the expressing strains . Only one region fit this criteria: an approximately 35kb region near the left arm of chromosome VIII ( S1 Text and Figs 3A and S5 ) . Because the heterozygous diploid does not express AQY2 or ncFRE6 , we reasoned that S288c likely harbors a repressor within this region . To identify the locus within this region responsible for repression of AQY2/ncFRE6 in S288c , we screened the S288c deletion library [41] for expression of ncFRE6 in each of 12 gene deletions within the 35kb region . Of the deletions tested , only one , rim101Δ , de-repressed the transcripts ( Fig 3B ) , strongly suggesting that the RIM101 allele harbors the trans factor that represses AQY2/ncFRE6 in S288c . In support of this hypothesis we note that RIM101 is a well-characterized zinc finger transcriptional repressor and is one of the most sequence-variable transcription factors between S288c and ∑1278b , harboring 18 SNPs , 13 of which are non-synonymous ( S6 Fig and S1 Text ) . To confirm that the polymorphic RIM101 allele controls expression of AQY2/ncFRE6 , we interconverted the entire RIM101 open reading frame ( S288c: ChrVIII 51111–52988 , Σ1278b: ChrVIII 49766–51655 ) between the strains and measured expression of AQY2/ncFRE6 . Interconverting the RIM101 allele is sufficient to repress expression in Σ1278b and to rescue expression in S288c , confirming that the RIM101 alleles confer distinct trans-acting regulatory capacity with regards to AQY2/ncFRE6 expression ( Fig 3C ) . We concluded that one or more of the sequence variations between the strains is responsible for the difference in RIM101 activity . RIM101 is known to contribute to several phenotypes , including being required for haploid invasive growth in ∑1278b [42] . S288c cannot invade agar due to a loss of function mutation in the TF FLO8 and therefore is insensitive to null mutations in RIM101 . We reasoned that the differences within the S288c and ∑1278b RIM101 allele could affect the invasive growth phenotype [43] in ∑1278b . However , the ∑1278b strain harboring the S288c RIM101 allele , ∑1278b ( S2RIM101 ) , did not lose the ability to invade agar , implying that differences between the S288c and ∑1278b RIM101 alleles do not affect the invasive growth phenotype ( S7 Fig ) . To assess the impact of RIM101 on genome-wide expression , we performed RNA-seq on RIM101 deletion strains in S288c and ∑1278b . Consistent with RIM101’s role as a transcriptional repressor , the majority of the genes whose expression level changes upon deletion of RIM101 in S288c became de-repressed ( 771 upregulated , 301 downregulated ) ( S4 Table ) . Surprisingly , the effect of deleting RIM101 in S288c was much larger than in Σ1278b ( Fig 4A ) . While 1072 genes change expression levels in S288c rim101∆ relative to S288c wildtype , only 145 change in ∑1278b rim101∆ relative to ∑1278b wildtype . Furthermore , the ratio of de-repressed to repressed genes is opposite in the Σ1278b RIM101 deletion ( 45 upregulated , 100 downregulated ) . This result suggests that RIM101 is a stronger repressor in S288c than in ∑1278b . Consistent with a loss of repressive capacity in ∑1278b relative to S288c , we note that AQY2/ncFRE6 levels do not change in the ∑1278b rim101∆ strain . Nevertheless , 145 genes do change expression in ∑1278b rim101∆ , implying that the disparate response to deletion of RIM101 is not due to a complete loss of function of the ∑1278b RIM101 allele . We next sought to determine the extent to which genome-wide differential expression between S288c and ∑1278b can be attributed to RIM101 . We reasoned that genes that are differentially expressed between the wildtype strains but not the RIM101 deletion strains are RIM101-dependent because removal of RIM101 from the system eliminates the observed interstrain differential expression . Hence , these differences in expression level between S288c and ∑1278b can be attributed to differences in RIM101-mediated regulation . Surprisingly , of 1207 differentially expressed genes between S288c and Σ1278b , over two-thirds ( 822 ) are in some way dependent on the presence of RIM101 ( Fig 4B , Red ) . We next asked how expression of the 822 RIM101-dependent transcripts ( as defined in Fig 4B ) changes upon loss of the RIM101 allele in each background . Deleting RIM101 in S288c results in a shift in RIM101-dependent gene expression toward ∑1278b wildtype levels ( S8A Fig ) . However , deletion of RIM101 in ∑1278b did not result in a shift toward S288c wildtype levels ( S8B Fig ) . This asymmetric response to RIM101 deletion is consistent with RIM101 possessing augmented repressive capacity in S288c relative to ∑1278b . We sought to distinguish whether the RIM101 allele itself , or the RIM101 pathway , imparts additional repressive capacity in S288c relative to ∑1278b by swapping RIM101 alleles between backgrounds and assaying genome-wide expression by RNA-seq . Surprisingly , upon introduction of the non-native allele only three transcripts undergo statistically significant changes in expression in both backgrounds ( Padj≤0 . 05 ) ( Figs 5A and S9 ) . AQY2 , ncFRE6 , and TIP1—a cell surface mannoprotein—significantly change expression levels in response to incorporation of the non-native allele in both backgrounds . Such a focused , allele-dependent transcriptional response stands in stark contrast to other trans-regulators discovered in eQTL studies , which tend to affect expression of many genes . It remains unclear how only AQY2 , ncFRE6 , and TIP1 are so dramatically influenced by interconversion of RIM101 alleles between backgrounds . Moreover , the allele-dependent expression level of TIP1 is surprising given that the TIP1 allele and promoter region are invariant between S288c and ∑1278b , and especially because no change in AQY2/ncFRE6 expression was observed in the ∑1278b RIM101 deletion strain , nor was TIP1 expression changed in the S288c RIM101 deletion strain . ( S4 Table ) . This high degree of allele-specificity suggests that unique combinations of factors can collaborate to regulate specific sets of genes . Although only three transcripts become significantly differentially expressed in both S288c and ∑1278b RIM101-interconverted strains relative to their wildtype expression levels , many RIM101-dependent genes appear to be more highly expressed in S288c ( ∑RIM101 ) than in S288c wildtype , consistent with the S288c RIM101 allele encoding a stronger repressor than the ∑1278b allele ( Fig 5A ) . In fact , in S288c ( ∑RIM101 ) , expression levels of the 822 RIM101-dependent genes shift toward a pattern more similar to ∑1278b ( Fig 5B and 5C ) , partially phenocopying the expression shift observed in S288c rim101∆ ( Fig 5B–5E ) . However , incorporation of the strong S288c allele into ∑1278b does not result in a shift towards stronger repression of the same subset of genes ( Fig 5B , 5E , 5F and 5G ) . This asymmetry suggests that other background factors , and not solely the RIM101 allele , are responsible for the gain of widespread RIM101-mediated repression in S288c , and that repression of AQY2 , ncFRE6 and TIP1 is independent of such a background effect . These results imply that the same transcription factor can display drastically altered activity depending on the background that it is present within , and that certain backgrounds , such as ∑1278b , buffer against widespread transcriptional dysregulation upon introduction of a RIM101 variant . Given that so few transcripts significantly change expression levels upon interconversion of RIM101 alleles , we sought to better understand the molecular basis of such specificity . In order to infer the genomic feature or features within RIM101 that contribute to repression of AQY2 and ncFRE6 in S288c , we screened two additional strains of S . cerevisiae , each harboring distinct combinations of sequence variation within RIM101 , for expression of AQY2 and ncFRE6 ( Fig 6A ) . The RIM101 DNA sequence includes 18 SNPs between S288c and Σ1278b , 13 of which alter the amino acid sequence of the Rim101 protein ( S10 Fig ) . In addition to the 13 non-synonymous SNPs , a poly-glutamine repeat stretch is expanded from four amino acids in S288c to eight in Σ1278b . We selected RM11-1a [19] and JAY291 [44] for screening because they have distinct combinations of the sequence variations observed between S288c and Σ1278b . RM11-1a exhibits the AQY2/ncFRE6-repressed phenotype , suggesting that this strain harbors a RIM101 allele capable of repressing the transcripts in a similar manner to S288c . However , the other strain , JAY291 , expresses both transcripts as Σ1278b does . These results indicate that the two transcripts are expressed or repressed concurrently , implying the mechanism by which the transcripts are co-regulated is conserved across diverse S . cerevisiae strains . We reasoned that the RIM101 sequence necessary for repression must exist in both S288c and RM11-1a , but not in Σ1278b or JAY291 . Alignment of the amino acid sequences of Rim101 across the strains revealed four non-synonymous SNPs and a truncated poly-glutamine stretch that exist solely in the repressive strains ( Fig 6B ) . We sought to identify one or more of these variants between S288c and Σ1278b that control expression of AQY2/ncFRE6 . Because variable length poly-glutamine tracks have been associated with altered protein structure and function , including altered protein-protein interactions [45] , we first tested whether poly-glutamine repeat length affected RIM101-mediated repression . After expanding the poly-glutamine tract in S288c and truncating it in Σ1278b , we tested for expression of AQY2/ncFRE6 and detected no deviation from either wildtype strain , suggesting that the length of the poly-glutamine tract does not , by itself , affect RIM101 activity at this locus ( Fig 6C ) . Next we tested whether the four conserved amino acids are sufficient to affect Rim101-mediated repression of AQY2/ncFRE6 . Indeed the collection of all four mutations is sufficient to rescue expression in S288c ( Fig 6D ) . Replacing each amino acid individually revealed one critical amino acid residue with regards to regulation of AQY2/ncFRE6 . In S288c , W249L is sufficient to de-repress AQY2 and ncFRE6 ( Fig 6E ) . Furthermore , L249W is sufficient to repress AQY2/ncFRE6 in Σ1278b . Hence , a single nucleotide polymorphism within the RIM101 transcription factor determines whether AQY2 and ncFRE6 are expressed . Finally , we sought to determine whether the amino acid present at position 249 is predictive of expression of AQY2/ncFRE6 in other strains . We could predict expression of AQY2/ncFRE6 in all five additional strains that we tested ( S11 Fig ) . Strains with L249 express AQY2/ncFRE6 , and those with W249 do not . Clearly , position 249 within the Rim101 protein is intimately linked to expression of AQY2/ncFRE6 across a diverse array of strains . However , the ability to predict AQY2/ncFRE6 expression is not conserved in a closely related species , S . paradoxus ( S12 Fig ) . Hence , the effect of the W249L RIM101 mutation appears to be clade specific , indicating that it may be a recently evolved regulatory mechanism .
Our study highlights the complexity of transcriptional regulation , even at a single locus . For example , though Reb1 binding is clearly regulated by a cis variant between S288c and ∑1278b , and its binding pattern correlates with expression of ncFRE6 , Reb1 binding does not affect expression of ncFRE6 . This result underscores the importance of single locus studies for identifying the true sources of differential expression , rather than relying on correlations between TF binding and expression . Furthermore , our results provide a unique example of how cis- and trans-linked DNA elements function in concert to affect gene expression . While both S288c and ∑1278b cis contexts are capable of directing expression of AQY2/ncFRE6 , promoter activity is only apparent in the absence of an epistatic trans-factor that we determined to be the transcription factor Rim101 . Although our study initially focused on transcriptional regulation of a single locus in cis , much of the complexity governing the genome-wide regulatory capacity of RIM101 arises from unknown background-dependent interactions that result in widespread differences in gene expression in trans . RIM101 target genes undergo a widespread shift in expression pattern specifically in S288c ( ∑RIM101 ) but not in ∑1278b ( S2RIM101 ) . While the physical mechanism underlying this asymmetric response is unknown , previous RIM101-based research could offer clues . In particular , Rim101 is extensively post-translationally modified , including by phosphorylation [46] and proteolytic processing [47] . It is not known whether a background-specific , allele-dependent RIM101 interaction influences either of these modifications . Also , W249L resides in close proximity to the C2H2 zinc finger DNA-binding domain of Rim101 , raising the possibility that variation at this position could impact DNA binding in S288c , but not ∑1278b , perhaps endowing Rim101 with altered regulatory capacity in certain genetic contexts . How do alternate RIM101 alleles achieve such remarkable target specificity ? Interconversion of the RIM101 alleles between strain backgrounds strongly impacts only three transcripts , AQY2 , ncFRE6 , and TIP1 , while other genes remain largely unaffected . How W249L , a mutation that has not been previously described , permits such specificity , remains unclear , though it is likely that such a phenomenon arises from allele-specific interactions with other genetic elements . However , the limited impact of the RIM101 allele on expression of other genes implies that if this is the case , the interaction is specific to AQY2/ncFRE6 and TIP1 . Because the change in expression of AQY2/ncFRE6 occurs in the opposite direction as TIP1 ( AQY2/ncFRE6 higher in ∑RIM101 strains , TIP1 lower in ∑RIM101 strains ) , it is possible that the mechanisms by which W249L elicits such a focused response are different between the two loci . Furthermore , expression of AQY2/ncFRE6 or TIP1 did not change in the ∑1278b RIM101 deletion or the S288c RIM101 deletion strains , respectively , further supporting a role for a W249L-specific interaction with other factors to influence AQY2/ncFRE6 andTIP1 expression specifically . Our results suggest that subtle mutations within TFs interact with genetic backgrounds to elicit unique combinations of gene expression patterns , likely expanding the phenotypic diversity observed within a population . Such a focused , allele-dependent transcriptional response to a TF-linked variant stands in contrast to most known trans-regulators that affect expression of many genes [48] . In order to understand the mechanisms by which such subtle mutations affect target selection , it may be necessary to undertake a systematic allele-swapping strategy . Such studies are likely to reveal concepts important not only for understanding the biochemical nature of the variant itself , but also how the effect of the variant is propagated within alternate genetic backgrounds . Moreover , such an approach would afford researchers the ability to learn specifically about how variants within TFs , rather than other categories of genes typically discovered in eQTL studies [48] , affect gene expression . Our finding that a SNP within a TF that regulates hundreds of genes causes large-scale expression differences in so few transcripts supports a model in which specific TF alleles interact in a combinatorial manner to regulate specific sets of genes [48] . One outstanding question is whether our findings regarding background- or allele-dependent activities of a transcription factor will be generalizable to other complex biological systems , including those involved in disease . For instance , transcription factors , including zinc finger TFs , are frequently mutated in cancers [49] and other human diseases , yet little is known about how the mutations relate to disease progression or outcome . With an enormous amount of sequence and functional data now available through consortiums such as The Cancer Genome Atlas ( TCGA ) and the 1 , 000 Genomes Project [50] , tools now exist to test whether different alleles of the same TF can lead to variable expressivity of disease-associated phenotypes by impacting transcriptional profiles . RIM101-mediated regulation is affected not only by the RIM101-allele , but also the background that it is present within , suggesting that even in the relatively simple case of RIM101-mediated regulation of AQY2/ncFRE6 , the regulatory pathways have diverged between S288c and ∑1278b . Furthermore , S . paradoxus , a species closely related to S . cerevisiae , does not conform to the same regulatory paradigm that controls AQY2/ncFRE6 expression in the S . cerevisiae strains we tested . AQY2/ncFRE6 expression is absent in S . paradoxus although it harbors a RIM101 allele that includes the S . cerevisiae expression-permissive Rim101 L249 variant . This suggests that the RIM101-dependent transcriptional regulatory circuit has been rewired at this locus . Clearly , the regulatory pathways underlying even simple , binary expression patterns display extraordinary complexity that could contribute to the plasticity of gene expression regulation observed throughout evolution . The RIM101 allele-dependent interactions that we observed may contribute to the phenotypic diversity observed between S288c and ∑1278b . Because AQY2 is non-functional in S288c , but functional in ∑1278b , the evolutionary pressures affecting expression of AQY2 are likely different between the strains . Perhaps the subtle RIM101 W249L variant , which strongly alters expression of only three transcripts , represents an example of genetic drift between the strains . TIP1 is a cell-surface mannoprotein and AQY2 is a cell surface water channel , raising the possibility that the focused AQY2 and TIP1 expression differences caused by W249L may result in an altered cell surface environment between the strains . Although we showed that the RIM101 allele did not affect haploid invasive growth , such a re-structuring of the cell surface could result in other RIM101 , AQY2 , or TIP1-linked cell surface phenotypes . Cryptic genetic variation ( CGV ) is genetic variation that influences a phenotype in certain environmental or genetic contexts , but not in others [51] . Although it is almost certain that CGV is common in nature , very few examples have been described in detail [52 , 53] . Our study highlights a previously undescribed mechanism by which CGV can manifest . We propose that polymorphic transcription factors likely represent a source of CGV whereby certain genetic backgrounds buffer against widespread transcriptional dysregulation upon introduction of a non-native allele ( as in ∑1278b ( S2RIM101 ) ) , while others are subject to a dramatic shift in gene expression ( as in S288c ( ∑RIM101 ) . The regulatory capacity of RIM101 is highly background-dependent and the interaction of RIM101 with genetic background determines whether a cell will undergo widespread or localized changes in its transcriptional program upon introduction of an alternative RIM101 allele .
S . cerevisiae strains used in this study were derived from BY4742 ( S288c , his3∆1 , lys2∆0 , leu2∆0 , ura3∆0 ) or L6441 ( Σ1278b , ura3-52 , leu2::hisG , his3::hisG ) . Other strains used in Figs 6A and S11 including JAY291 , RM11-1a , CLIB324 , CLIB382 , YPS163 , T7 , and UC5 are homothallic diploids generously donated by Justin Fay ( Washington University ) . The delitto perfetto [36] method was used to edit genome sequences . For gene expression experiments , cells were grown in standard YPD media to mid-log phase in YPD before RNA isolation . Primers and plasmids used in this study are listed in supplementary materials and methods . Invasive growth phenotype assay ( S7 Fig ) was performed as described in [43] by patching cells onto a YPD plate for two days and washing the plate under gently running water before imaging . RNA was extracted using a standard acid phenol chloroform extraction ( Collart , 2001 ) and DNased with RQ1 DNase ( Promega #M6106 ) according to manufacturer’s instructions . 1ug of RNA was reverse transcribed using Multiscribe reverse transcriptase ( Thermo Fisher #4311235 ) with random hexamers , except for ncFRE6 , for which we used a gene specific RT primer due to the need to measure RNA levels strand-specifically . cDNA was measured using targeted qPCR primers ( S1 Text ) and SYBR select ( Life Technologies #4472908 ) on the Biorad CFX qPCR system . Strand specific RNA-seq libraries were made using the NEBNext Ultra Strand-specific RNA-seq library prep kit ( NEB #E7420S/L ) with manufacturers instructions . Briefly , RNA was isolated by standard acid phenol chloroform extraction and poly-adenylated RNA was purified with oligo ( dT ) dynabeads ( Thermo Fisher #61002 ) . RNA was fragmented and first strand synthesis performed with ProtoScript II reverse transcriptase and random hexamers . Second strand synthesis then incorporated Uridine residues into cDNA . cDNA was purified with AMPure beads ( Agencourt ) . cDNA was then dA-tailed and NEBNext adaptors for Illumina were ligated before another AMPure purification . USER excision removed the second strand and libraries were amplified with NEBNext High Fidelity PCR master mix ( NEB ) . NEBNext Multiplex oligos 1–12 ( NEB #E7335 ) were incorporated during PCR . Libraries were quantified with the Qubit ( Life technologies ) before pooling at equimolar concentrations and sequencing on an Illumina Hiseq . Reads were mapped using bowtie2 and differential expression was assessed using DEseq ( S1 Text ) . S288c ( BY4741 ) and Σ1278b ( L6441 ) haploid strains were crossed to generate a heterozygous diploid . The diploid was sporulated on traditional sporulation media and tetrads were dissected . 28 haploid segregants of an S288c x Σ1278b cross were grown to mid-log phase in YPD and tested for expression of AQY2/ncFRE6 by qRT-PCR . Segregants were binned based on whether they expressed AQY2/ncFRE6 or not . Genomic DNA was isolated , treated with RNAse A ( Thermo Fisher #AM2270 ) , and purified by phenol chloroform extraction and ethanol precipitation . DNA concentrations were measured with the Qubit ( Thermo Fisher #Q33216 ) and pooled at 10nM separately for strains either expressing or not expressing AQY2/ncFRE6 . DNA was sheared using the Covaris M220 ultrasonicator to an average size of 500 bp . DNA was blunted and dA-tailed before ligation of Illumina sequencing adapters . Libraries were amplified by Phusion polymerase with Illumina multiplex barcodes 1+2 for ten cycles before analysis on the Bioanalyzer ( Agilent ) . Samples were sequenced on an Illumina HiSeq 2000 . Reads were mapped using bowtie2 and variants identified using GATK ( S1 Text ) .
|
Large-scale genome profiling studies have revealed that disease-associated variants occur most frequently within regulatory regions of DNA . In order to connect variants ( genotypes ) to expression profiles ( phenotypes ) , it is necessary to have a thorough understanding of the mechanisms underlying transcriptional regulation across genetic backgrounds . In this study we used a high-throughput method for identification of the genetic variant responsible for differential expression of a pair of divergently oriented transcripts between two strains of S . cerevisiae . We discovered a single nucleotide polymorphism within the transcription factor , RIM101 , that is required for co-regulation of both transcripts . Furthermore , we showed that the expression pattern of the divergently oriented transcripts was determined by the SNP across a diverse array of genetic backgrounds . The extent to which the SNP affects genome-wide transcriptional profiles was strongly influenced by the genetic background , a phenomenon that likely contributes to the astounding phenotypic diversity observed within a population .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2016
|
A trans-acting Variant within the Transcription Factor RIM101 Interacts with Genetic Background to Determine its Regulatory Capacity
|
Coordinated replication of eukaryotic genomes is intrinsically asymmetric , with continuous leading strand synthesis preceding discontinuous lagging strand synthesis . Here we provide two types of evidence indicating that , in fission yeast , these two biosynthetic tasks are performed by two different replicases . First , in Schizosaccharomyces pombe strains encoding a polδ-L591M mutator allele , base substitutions in reporter genes placed in opposite orientations relative to a well-characterized replication origin are strand-specific and distributed in patterns implying that Polδ is primarily involved in lagging strand replication . Second , in strains encoding a polε-M630F allele and lacking the ability to repair rNMPs in DNA due to a defect in RNase H2 , rNMPs are selectively observed in nascent leading strand DNA . The latter observation demonstrates that abundant rNMP incorporation during replication can be tolerated and that they are normally removed in an RNase H2-dependent manner . This provides strong physical evidence that Polε is the primary leading strand replicase . Collectively , these data and earlier results in budding yeast indicate that the major roles of Polδ and Polε at the eukaryotic replication fork are evolutionarily conserved .
Three DNA polymerases , Polα , Polδ , and Polε , are required for efficient genome replication in eukaryotes [1] , [2] . The Polα holoenzyme complex has both primase activity and DNA polymerase activity and is required to initiate each DNA synthesis reaction . The primase subunit first synthesizes a short RNA primer of ∼10 nucleotides and the DNA polymerase subunit then extends this primer using dNTPs for a further 20–30 nucleotides , thus initiating DNA replication . Polδ or Polε then substitutes for Polα and perform the bulk of DNA replication by elongating these primers . Genomic DNA is replicated faithfully during every cell cycle with an error rate of approximately 1 in 10−10 errors per base pair , ensuring that the genetic blueprint is transmitted largely unaltered through the generations . In eukaryotic cells , DNA replication is initiated bi-directionally from many replication origins . Because of the antiparallel structure of DNA , one strand ( leading strand ) is replicated continuously in the same direction of the replication fork , while the second strand ( lagging strand ) is synthesized discontinuously in the opposite direction to that of replication fork progression . The relatively small ( 200–1000 base ) stretches of DNA synthesized during lagging strand replication are known as Okazaki fragments and are rapidly processed and ligated to complete lagging strand replication . The fidelity of replication is ensured by the nucleotide selectivity of replicases to achieve error rates of 10−4–10−5 , by exonucleolytic proofreading during replication to increase fidelity about 100-fold , and by post-replication DNA mismatch repair to further increase fidelity and lower the mutation rate to 10−8–10−10 [3] . Polα , Polδ , and Polε all belong to the B family of DNA polymerases . The structure of the active site of B family DNA polymerases is highly conserved throughout evolution . As for most polymerases , the precise geometry of the polymerase active site ensures that mismatches are largely precluded from incorporation [4] . The importance of polymerase active site geometry to replication fidelity is illustrated by the fact that substitutions of conserved active site residues often reduce DNA synthesis fidelity . Relevant to the present study are substitutions in Saccharomyces cerevisiae Polε and Polδ ( M644G and L612M , respectively ) that increase error rates during DNA synthesis in vitro and also result in elevated spontaneous mutation rates in vivo [5]–[8] . These polymerases have particular value for studies of replication fidelity in vivo because their error rates are preferentially elevated for only one of two possible mismatches that could result in a particular base substitution in a cell . For example Polδ L612M preferentially generates T-dGTP rather than A-dCTP errors , and this preference yields strand specific A–T to G–C mutations during duplex DNA replication in vivo . These biased error rates result in asymmetric mutation profiles in a URA3 reporter gene that is replicated in only one direction due to its close proximity to an active origin . When present in each of the two possible URA3 orientations relative to the origin , the mutational patterns observed in strains harboring the pol2-M644G ( polε ) and pol3-L612M ( polδ ) mutator alleles imply that S . cerevisiae Polε and Polδ are the primary leading strand and lagging strand replicase , respectively [9] , [10] . The goal of the present study is to identify the major leading and lagging replicases in the fission yeast Schizosaccharomyces pombe . To investigate Polδ , we took advantage of the fact that both S . cerevisiae Polδ L612M [6] and its human equivalent , Polδ L606M [11] , [12] have been shown to have biased DNA synthesis fidelity . Here we report that Schizo . pombe Polδ L591M generates asymmetric mutation profiles in vivo that are consistent with Polδ being the primary lagging strand replicase in Schizo . pombe . To investigate Polε , we attempted to generate a Schizo . pombe Polε mutation ( polε-M630G ) equivalent to that previously studied in S . cerevisiae ( encoding Polε M644G ) . Schizo . pombe strains with this substitution were not viable . We therefore generated a different allele , polε-M630F , because substitution of phenylalanine at the equivalent active site residues in S . cerevisiae Polα [13] and Polζ [14] are viable and have elevated spontaneous mutation rates . We show here that the Schizo . pombe polε-M630F allele is also viable and a spontaneous mutator . Although it did not display a suitable asymmetric mutation profile for strand assignment , we were able to exploit a second infidelity parameter for strand assignment , the propensity to incorporate rNMP into DNA . Previous studies have demonstrated that during DNA synthesis in vitro and in vivo , S . cerevisiae Polε M644G incorporates greater amounts of rNTPs into DNA than does wild type Polε [15] , [16] . Here we exploit this same promiscuity with the Schizo . pombe polε-M630F mutant , to provide a physical demonstration that the majority of leading strand synthesis in Schizo . pombe is performed by Polε .
Which DNA polymerase replicates which strand has only been determined in the budding yeast S . cerevisiae [9] , [10] . We thus wished to determine if this division of labour between the main replicative polymerases is conserved in the distantly related eukaryote , the fission yeast Schizo . pombe . Our strategy was to establish the direction of replication for a specific locus , to create mutants in the genes encoding two replicative polymerases , Polδ and Polε , that exhibit specific and characteristic profiles of misincorporation , and to use these to assign each polymerase to one or the other strand ( or both ) based on the profile of misincorporation at the directionally replicated loci . The catalytic subunits of Polδ or Polε are encoded by the cdc6 ( pol3 ) and the cdc20 ( pol2 ) genes , respectively . For clarity , here we simply refer to them as polδ and polε . We employed recombination-mediated cassette exchange ( RMCE ) to create strains that harbor each specific mutant polymerase [17] . Mutant genes introduced into the genome by this method are flanked by lox ( P and M3 ) sequences . Thus , we also created control strains ( pol+ ) that have the gene encoding the wild-type polymerase flanked by the same lox sites . The Schizo . pombe ura4+ gene allows for both positive and negative selection . Selecting for loss of ura4 function is achieved by growth on medium containing 5-fluoroorotic acid ( 5-FOA ) , which identifies loss-of-function mutants . However , mutations in either the ura4 or the ura5 genes of Schizo . pombe confer 5-FOA resistance , and it has been reported that greater than 50% of spontaneously arising 5-FOA resistant clones harbor mutations in ura5 [18] . In wild type cells , ura4+ is located on chromosome III while ura5+ is located on chromosome II . Therefore , to efficiently identify mutations at a single chromosomal location that confer 5-FOA resistance , we created two artificial loci where ura5+ was placed adjacent to ura4+ on chromosome III . These differ only in the orientation of the ura4+:ura5+ fragment ( Figure 1A ) . We confirmed that this novel ura5+:ura4+ fragment does not function as a replication origin by demonstrating it would not support maintenance of plasmid sequence in cells . We also deleted the genomic ura5+ gene on chromosome II , so that resultant ura4+:ura5+ Δura5 strains have only one copy of the ura4+ and ura5+ genes . The ura4+:ura5+ locus is on Chromosome III , near two autonomous replicating sequences; ars3003/3004 . Both the ars3003 and ars3004 sequences have been well characterized and are known to be highly efficient at initiating replication [19] , [20] . However , more than 50% of Schizo . pombe intergenic regions have the potential to function as origins of replication [21] . Thus , to experimentally determine the direction of DNA replication at the ura4+:ura5+ locus , we employed the method of directional 2-D gel electrophoresis [22] . DNA from an asynchronous population of cells is first digested with HindIII and BlpI and fragments separated in the first dimension without ethidium bromide . The lane is then excised and digested with SpeI , which cleaves within the HindIII-BlpI fragment containing the ura4+ and ura5+ genes . This DNA is then subjected to the second dimension of electrophoresis in the presence of ethidium bromide and DNA in the gel is transferred to a membrane for Southern blot analysis with the ura4-containing HindIII-SpeI fragment . The results revealed the direction of DNA replication , as illustrated in Figure 1B . Most of detectable replication intermediates show the pattern consistent with DNA replication moving from right to left ( Figure 1C , see red arrow bottom panel: its equivalent is similarly indicated in the top panel of Figure 1B ) . Thus , we conclude that a leftward replication fork replicates the ura4+:ura5+ locus in the majority of cells . We then created the polδ-L591M mutant using RMCE . Schizo . pombe Polδ L591 is equivalent to S . cerevisiae Polδ L612 . polδ-L591M cells grow as well as wild type cells ( Figure 1D ) , demonstrating that this mutant of Polδ is proficient for DNA replication in vivo . In wild type and ura4+:ura5+ Δura5 backgrounds , polδ-L591M showed a strong mutator phenotype ( Figure 1E ) . Spontaneous mutation rates are elevated ∼100 fold in polδ-L591M ( 4–5×10−6/cell division ) compared with that in polδ+ ( 4–7×10−8 ) . The elevated mutation rates indicate that most of the mutations seen in polδ -L591M cells reflect the error specificity of this mutant polymerase , rather than background mutations . As shown in Table 1 , more than half of mutations were point mutations , consistent with elevated base misincorporation observed in vivo for the equivalent S . cerevisiae strain and in vitro for the corresponding mutant version ( L612M ) of S . cerevisiae Polδ [5] , [6] , [23] and human ( L606M ) Polδ [11] , [12] . In addition to point mutations , we observed a variety of duplication and deletion mutations . All of these deletions and duplications were observed at repetitive DNA sequences . More than half of the deletions were >100 bp , while the majority of duplications were <100 bp ( Table S1 ) . Possible mechanisms by which such mutations may arise are addressed in the Discussion . Among the point mutations , transition mutations showed significant strand dependence for misincorporation . Figure 2 and Table 2 show the predicted mispairs formed during synthesis of the transcribed strand , which corresponds to lagging or leading strand synthesis in the Forward or Reverse strains , respectively ( illustrated in Figure 2A ) . A:T to G:C mutations can result from either A:dCTP mismatches or T:dGTP mismatches . Depending on which template strand is copied by the mutated polymerase , this will give a bias of mutation resulting in a spectra dependent on the orientation of the DNA sequence ( see Figure 2B ) . We observed that , for A:T to G:C changes , T:dG mispairing is 12 . 5 fold more frequent than A:dC mispairing in the Forward strain , while A:dC is more frequent in the Reverse strain . Since the misincorporation rate of the corresponding mutant S . cerevisiae and human polymerases are much higher for T:dG than for A:dC [6] , [12] , the results in Figure 2 imply that Polδ preferentially replicates the lagging strand template . A similar bias was also observed for G:C to A:G mutations . G:dT is ∼3 fold higher than C:dA in the Forward strain while C:dA is ∼3 fold higher in the Reverse strain . Comparing these data with the published in vitro results is also consistent with Polδ being responsible for replicating the lagging strand template . Strand dependence was not observed in polδ+ ( Table 3 ) , indicating that the bias seen in polδ -L591M cells reflects base misincorporation by the mutant polymerase rather than sequence context or the transcriptional direction of marker genes . We did not observe strong hotspots for particular mutations , but the total number of occurrences is higher for some mutations , e . g . , T to C at ura4 base pair 236 and 76 in the Forward background and C to T at ura4 190 and ( −91 ) in the Reverse background ( Figure 3 ) . S . cerevisiae Polε M644G shows strong bias between A:dA and T:dT mispairs in vitro and the spontaneous mutation rates of the corresponding mutant cells are significantly higher than that of wild type and exhibit strand bias [8] . However , we found that the equivalent Schizo . pombe polε-M630G mutation is lethal , as was a polε-M630K mutation . Analysis of strains expressing Polε M630G or Polε-M630K from an ectopic integrated copy in a polε+ background ( Figure S1 ) suggest this is largely due to catalytic inactivity , as mutation frequencies were not dramatically increased . Thus , we created strains harboring polε-M630F as an alternative . The decision to substitute to phenylalanine was based on earlier studies showing that Polα L868F is error prone in vitro and mutagenic in vivo [13] , Polε M644F is error-prone in vitro with a weak bias in error rates [7] , and Polζ L979F is error prone in vitro [24] and mutagenic in vivo [14] . The Schizo . pombe polε-M630F that we created using the RMCE methodology grows slightly more slowly than polε+ , although the size of mutant colonies becomes comparable to that of wild type after prolonged incubation ( Figure 4A ) . Strains harboring polε-M630F did not exhibit a substantial increase in spontaneous mutation rate in a mismatch repair proficient background . In strains wherein mismatch repair is inactivated by deleting the msh2 gene , polε-M630F increased the mutation rate by 4–5 fold ( Figure 4C ) . However , upon sequencing ura5 and ura4 from 5-FOA resistant clones , a strand bias sufficient to infer which strand is copied by the mutant Polε was not observed ( Table S2 ) . Mutations at Polε M644 in S . cerevisiae affect the rate of rNMP incorporation into DNA [15] . We thus tested this possibility in Schizo . pombe . rNMPs incorporated into DNA are rapidly excised by the activity of RNase H2 , whose catalytic subunit is encoded by the rnh201 gene of Schizo . pombe . Since increased rNMP incorporation increases alkali-dependent DNA fragmentation , we assayed for increased gel mobility of DNA from the endogenous ura4+ locus using Southern blot analysis . As anticipated , genomic DNA prepared from polε-M630F was not particularly sensitive to alkali treatment when compared to genomic DNA from the polε+ strain ( Figure 5A , lanes 1 and 2 ) . However , it becomes significantly sensitive compared to polε+ when rnh201 is deleted ( lanes 3 and 4 ) . This indicates that Polε M630F incorporates rNMP into DNA at higher rate than wild type Polε and that these are largely removed by RNase H2 activity . Based on this observation , we chose to test the strand specificity of rNMP incorporation using alkali treatment and subsequent probing for either the leading or lagging strand using the appropriate single-stranded probes . We prepared two pairs of probe across ars3003/3004 ( Figure 5B ) . The top strand is detected by probe A and C , while the bottom strand is detected by probe B and D . As shown in Figure 5C , only one of each of the two strands from rnh201Δ polε-M630F was sensitive to alkali at each probed site . The alkali sensitive strand was the bottom strand on the left side of the origin , while the top was sensitive on the right side ( Figure 5B and 5C ) . Since those probed sites are inferred to be copied by replication forks emerged at ars3003/3004 , the alkali-sensitive strands correspond to the nascent leading strand products of replication . Similar results were obtained at another origin ( Figure S2 ) . These results strongly suggest that Polε replicates the leading strand template .
To investigate the role of Schizo . pombe Polδ during DNA replication , we created strains that replicate using a Polδ L591M mutant protein . We showed that Polδ L591M is highly mutagenic and induced various types of mutations in Schizo . pombe . Strand dependence in transition mutations allowed us to conclude that the main role of Polδ is during lagging strand synthesis ( Figure 2 and Table 2 ) . However , the mutational bias seen in this mutant is weaker than would be predicted from the in vivo and in vitro studies of the equivalent S . cerevisiae mutant . Because we used mismatch repair proficient cells for this study ( the double mutant was lethal ) , the mutation spectra we observed here reflect mispairs that have escaped mismatch detection and repair . This may influence our interpretations . For example , bacterial MutS protein has variable affinity for different mismatches , with G:T being one of the best substrates [28] , [29] . Thus , the specificity of mismatch repair might have partially masked the bias of misincorporation induced by Polδ L591M . It is also possible that the mutation spectra were affected by spontaneous base damage that results in mismatches that escape mismatch repair . These caveats mean that , while our results are consistent with a function of Polδ as a lagging strand polymerase , we cannot exclude the possibility that Polδ partly participates in leading strand synthesis or that Polε ( or indeed other polymerases ) may partially replicate the lagging strand [10] , [30] . In addition to point mutations that were expected from in vitro studies of S . cerevisiae and human polymerases , we also observed significantly enhanced formation of deletions and duplications in polδ -L591M cells ( Table 1 ) . All deletions and duplications occurred at repetitive DNA sequences . The majority of duplications involved <100 bases ( Table S1 ) , reminiscent of the mutation spectra for S . cerevisiae rad27 mutants . Jin et al . showed that duplication rates were enhanced by mutations in the Polδ exonuclease domain [25] and S . cerevisiae polδ -L612M cells require functional Rad27 for viability [31] . These studies are consistent with our observations and add support to the premise that Polδ is involved directly in lagging strand synthesis in Schizo . pombe . The size of deletions we observed was relatively larger than that of duplications . More than half of the deletions were loss of >100 bp of sequence . Cai et al . have observed that exonuclease deficient E . coli DNA polymerase II generates similar deletions flanked by direct repeat sequences [32] . They proposed a model in which a mismatch made by a mutator polymerase during replication of the first direct repeat promotes primer relocation to the second direct repeat . Furthermore , we observed a low frequency of inversions flanked by inverted repeat sequences and most of these inversions were associated with deletion , duplication , and/or gene conversion . These events can be explained by template switching . Taken together , these observations suggest that a mismatch formed during DNA replication can cause various kinds of genome rearrangements . Interestingly , chromosome abnormalities such as chromatid breaks are substantially elevated in Pold1+/L604G and Pold1+/L604K mouse cells [33] . To investigate a role of Polε during normal DNA replication , we utilized the observation that S . cerevisiae Polε M644G increased rNMP incorporation [15] , [34] . We first demonstrated that Schizo . pombe polε-M630F cells incorporate rNMP into DNA at higher frequency than polε+ cells ( Figure 5A ) . This property of the mutant polymerase made it possible to determine the strand that is copied by the mutant Polε . Incorporation of rNMP in the leading strand was strikingly higher in polε-M630F mutant cells compared to polε+ cells ( Figure 5B and Figure S1 ) . This result strongly suggests that Polε synthesizes the leading strand . On the other hand , we failed to observe a significant difference in rNMP incorporation in the lagging strand . This suggests that Polε has , at most , a limited role in lagging strand synthesis . Schizo . pombe cells that harbor polε-M630G were not viable , while the corresponding mutation does not cause lethality in S . cerevisiae . Interestingly , the N-terminal catalytic domain of Polε can be entirely deleted in both yeasts [35] , [36] , while a catalytically dead Polε , that retains the full-length protein , is inviable . Our mutation frequency analysis of cells expressing Polε-M630G in a polε+ background ( Figure S1 ) suggest the inviability of polε-M630G is because the corresponding protein is catalytically dead , rather than because it increases the mutation burden beyond that which is sustainable . In addition to supporting a role for Polε in leading strand replication , the results in Figure 5 extend to Schizo . pombe two important conclusions derived from earlier studies in S . cerevisiae , namely that large numbers of rNTPs can be incorporated into the nascent leading strand during replication without strongly affecting growth ( Figure 4A ) and the rNMPs that are stably incorporated into the Schizo . pombe genome by a eukaryotic replicase are efficiently repaired in a RNase H2-dependent manner . In S . cerevisiae , unrepaired rNMPs in DNA promote formation of short deletions between short , tandemly repeated DNA sequences , by a mechanism that is unaffected by mismatch repair status [34] and is initiated by topoisomerase 1-dependent cleavage of rNMPs [37] . Many of the deletions occur in a manner that depends on the orientation of the reporter gene in relation to the closest origin of replication [15] , indicating that they result from rNMPs incorporated into the nascent leading strand by Polε . The characteristics of the Schizo . pombe polε-M630F strains described here offer the opportunity to determine if these consequences are conserved in fission yeast , and to also test whether mating type switching , which depends on rNMPs in DNA [38] , is affected by increased rNMP incorporation by replicases and/or by RNase H2 or topoisomerase status . In this study , we examined roles of Polδ and Polε during normal DNA replication in Schizo . pombe using two different methods . The first method was a genetic analysis of mutation spectra asymmetry in polδ mutant cells . The second was a physical rNMP incorporation assay using polε mutant cells . The combination of these analyses indicates that genomic DNA is replicated in Schizo . pombe in similar manner as has been suggested for S . cerevisiae . Because Schizo . pombe and S . cerevisiae are highly diverged in evolutionary terms [39] , [40] our results strengthen the interpretation that replication in all eukaryotes follows similar rules . We also add a physical assay to the previous genetic data , increasing the likelihood that the interpretation of the genetics is indeed correct . We mainly examined DNA replication at the genomic ura4 locus , because replication initiation at this locus is known to be highly efficient ( Figure 1B ) . However , a similar result was obtained for a second independent locus using the physical method for assigning Polε activity ( Figure S1 ) . Thus , it is reasonable to suggest that DNA replication occurs in similar manner throughout the genome . However , it remains possible that cells utilize these two polymerases in a different manner in some specific situations or at some specific loci .
Schizo . pombe cells were grown in yeast extract ( YE ) medium . Standard genetic and molecular procedures were employed as described previously [41] . To examine cell growth on plates , serial dilutions of cells were spotted on YEA ( YE agar ) plates , and incubated at 30°C . The cdc6+ and cdc20+ genes were amplified by PCR and cloned into pUC19 . cdc6-L591F and cdc20-M630F mutant genes were constructed by PCR-meditated site-directed mutagenesis and sequenced to ensure that only the desired mutation was introduced . Both wild-type and mutant genes were introduced into Schizo . pombe at their native loci by recombination-mediated cassette exchange ( RMCE ) [17] . Spontaneous mutation rates were determined by fluctuation assay as described previously [42] . Briefly , 11 independent single colonies were suspended in 5 ml YEP ( YE+polypeptone ) medium and grown to saturation at 30°C . Cells were diluted appropriately and plated on YEA or YEA containing 0 . 1% 5-fluoroorotic acid ( 5-FOA ) . Colonies were counted after 4 days incubation at 30°C . Mutation rates were calculated by the method of median [43] . Genomic DNA from a single 5-FOA resistant colony was isolated and the ura5-ura4 construct was amplified by PCR to be sequenced . Directional 2-D gel analysis was performed as described previously [22] with modifications . Genomic DNA was extracted and digested with HindIII and BlpI as described in [44] . After the first dimension electrophoresis , DNA was digested with SpeI in a gel slice and subjected to the second dimension electrophoresis . Replication intermediates were detected by Southern blot . Genomic DNA was extracted from exponentially growing cells and purified by Qiagen genomic-tip 100/G . 5 µg of undigested or EcoRI digested DNA was incubated in 0 . 3 M NaOH at 55°C for 2 hours and subjected to 1% alkaline agarose gel electrophoresis [15] . Gels were neutralized and stained with ethidium bromide , followed by Southern blot . Southern blotting was performed according to [45] . DNA fragments of interest were amplified by PCR from Schizo . pombe genomic DNA and used as templates to obtain labeled probes . Radioactive nucleotides were incorporated into DNA using Ready-To-Go DNA Labeling beads ( GE Healthcare ) or strand specific primers and TaKaRa Ex Taq ( TAKARA BIO ) .
|
It is important to understand the architecture of the DNA replication machinery and whether this is common to all organisms . Recent work in Saccharomyces cerevisiae has genetically assigned specific DNA polymerases to leading and lagging strand DNA synthesis , Polε and Polε respectively . In this manuscript , we use a similar genetic assay to demonstrate that , in the highly evolutionarily diverged yeast Schizosaccharomyces pombe , Polδ is similarly responsible for lagging strand synthesis . Importantly , we establish a novel physical assay , the incorporation of rNMPs into newly replicated DNA , which demonstrates that Polε is responsible for leading strand synthesis and does not contribute significantly to lagging strand replication . These data strongly support and consolidate the interpretation of previous genetic data and suggest that the division of labour between polymerases is conserved through evolution .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"model",
"organisms",
"genetics",
"biology",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics"
] |
2011
|
The Major Roles of DNA Polymerases Epsilon and Delta at the Eukaryotic Replication Fork Are Evolutionarily Conserved
|
Information processing in the human brain arises from both interactions between adjacent areas and from distant projections that form distributed brain systems . Here we map interactions across different spatial scales by estimating the degree of intrinsic functional connectivity for the local ( ≤14 mm ) neighborhood directly surrounding brain regions as contrasted with distant ( >14 mm ) interactions . The balance between local and distant functional interactions measured at rest forms a map that separates sensorimotor cortices from heteromodal association areas and further identifies regions that possess both high local and distant cortical-cortical interactions . Map estimates of network measures demonstrate that high local connectivity is most often associated with a high clustering coefficient , long path length , and low physical cost . Task performance changed the balance between local and distant functional coupling in a subset of regions , particularly , increasing local functional coupling in regions engaged by the task . The observed properties suggest that the brain has evolved a balance that optimizes information-processing efficiency across different classes of specialized areas as well as mechanisms to modulate coupling in support of dynamically changing processing demands . We discuss the implications of these observations and applications of the present method for exploring normal and atypical brain function .
The human brain is a complex biological structure with specializations for local , modular processing that are distinct from anatomical properties that facilitate integrative processing . Specifically , anatomic projection patterns suggest a division between areas that form domain-specific hierarchical connections [1]–[4] and distinct heteromodal association areas that receive widespread projections from distributed brain systems [5]–[9] . The dichotomy is not absolute . Sensory systems contain divergent projections and display multimodal convergence at advanced processing stages . Nonetheless , dominance for one connectivity profile over the other is present for many areas and suggests a fundamental organizing principle of cortical-cortical connectivity . Early sensory cortical areas are examples of areas with predominantly local hierarchical connections ( e . g . , see [2] ) while prefrontal , lateral temporal , limbic and paralimbic areas form hubs linking widely distributed connections – neural epicenters of large-scale distributed networks [8] . Studies of comparative anatomy suggest that the ratio of local to distributed areal projections may be critical to the evolution of higher-order cognitive functions including language , reasoning , and foresight . The hominin brain has tripled in absolute size over the past 2–3 million years including a disproportionate enlargement of cortical surface area [10] , [11] . However , expansion comes with a cost to information processing efficiency [11] . Proliferation of long-distance connections and increasing brain volume could lead to untenable wiring lengths if they evolved unchecked [12] . Thus , there is a compensatory pressure to modularize information flow within parallel processing pathways and to maximize efficient communication among areas of similar function . Van Essen [13] proposed that there is a specific selection pressure to optimize wiring length between adjacent functionally-similar areas within the same hemisphere . Consistent with this possibility , cortical folding patterns in the macaque brain minimize between-area wiring lengths for sensory ( e . g . , Broadmann's area [BA] 17 to BA 18 ) and motor ( e . g . , BA 4 to BA 6 ) pathways . The relative proportion of association cortex differs further in the human [14] , [15] . The human brain is three times larger than that of modern great apes yet primary motor ( BA 4 ) and visual ( BA 17 ) cortices are about the same absolute size [16] , [17] . Preuss [14] , [18] , in a detailed analysis of cortical growth , concluded that widely distributed associated areas exhibited an increase in absolute surface area during hominin evolution including higher-order parietal and temporal areas as well as prefrontal cortex . Thus , the long-held belief that the prefrontal cortex is preferentially expanded in humans is only partially correct; heteromodal association areas are likely expanded throughout cortex including those areas falling within prefrontal cortex . Bolstering these observations , surface-based analysis of cortical differences between macaque and human based on 23 estimated homologous areas reveals a high degree of expansion in parietal , lateral temporal , and dorsolateral prefrontal regions and a relative compression of sensorimotor and visual areas [19] . The modern human brain also possesses a high proportion of cerebral white matter relative to contemporary primates including the great apes [20] , [21] ( see also [22] for a broad analysis of primates ) . Comparative study of the arcuate fasciculus , the major fiber bundle connecting anterior and posterior heteromodal language zones , shows that it is enlarged in humans as compared to chimpanzees or macaques [15] . Thus long-distance association projections have expanded as well and may have done so in relation to specific functional adaptations . One can presume that there has been considerable pressure to maintain efficient wiring and network properties as the complexity of cortical connectivity and association cortex has increased , especially considering long-distance projections are well represented in the human brain ( see [23] ) . All of these findings converge to suggest that the balance between long-range projections and local areal interactions is important for efficient cortical processing . While this balance has been recognized for some time ( e . g . , see [5] , [7] , [8] ) , recent computational explorations of connectional patterns have brought the issue into sharp focus [24] . Graph theory , in particular , provides informative metrics to analyze properties of complex networks [25]–[31] . When applied to the study of connectional anatomy , analyses consistently reveal that cortical networks exhibit “small world” properties [32] , [33] . Connections are not randomly dispersed among cortical areas but rather show strong clustering patterns and hubs that allow for relatively short path lengths to propagate information through the networks [34] . Moreover , the extent to which an individual area is central to maximizing communication between multiple areas can be quantified and cortical regions possessing hub-like properties can be mapped . Applying this analysis strategy to structural [35] and functional [36] , [37] human connectivity data reveals a core set of regions along the cortex including paralimbic areas and parietal association areas that behave as hubs . The resulting map of these regions in humans includes the many known heteromodal association areas spread throughout prefrontal , parietal , and lateral temporal cortex and bares a strong resemblance to the estimated regions of cortical expansion in human as compared to macaque ( e . g . , contrast [37] with [19] ) . Although previous studies have focused their attention in network topological modularity [38]–[41] and in some aspects of the relationship between physical distance and connectivity [29] , [36] , [42] , connectivity profiles that differentiate local and distant projection patterns have not been fully characterized . Physical distance and network path length , as discussed above , are among of the most central properties to efficient information propagation . There are two likely reasons for this omission . First , human studies using diffusion techniques to measure anatomic connectivity ( diffusion tensor imaging; DTI ) provide poor information about connectivity between areas that are supported by local association fibers ( u-fibers ) and neighborhood association fibers that connect immediately adjacent and nearby areas [43] . Commonly used diffusion imaging techniques capture long association fibers that travel in discrete fascicles within the hemisphere and commissural fibers that pass between the hemispheres ( but see [44] for a recent exception ) , and usually discard fibers or fail to adequately measure information from close or adjacent regions . Second , functional connectivity approaches that measure cortical-cortical interactions indirectly via correlated blood oxygenation level-dependent contrast ( BOLD ) [45]–[47] have not focused on local anatomic correlations because of the relatively poor spatial resolution of the approach . While the blood flow response is locally regulated ( under certain conditions at the level of the cortical column; e . g . , [48] ) , the current practical resolution for exploring large cortical regions is about 3–4 mm [49] . This makes exploring within-area lateral connections challenging . However , the achievable resolution of functional MRI ( fMRI ) is well within the expected resolution needed to provide information about adjacent and nearby areas that are distinct from interactions carried by long association fibers and other long-range connections . Measurements at this intermediate resolution should be rich in information about the connectional architecture of the human brain including information about whether cortical areas possess local modularity . Motivated by this possibility , we developed and applied a novel approach to map the regional balance between local and distant functional connectivity in the human brain . We first extended a computationally efficient approach based on network graph theory [37] to map the degree of intrinsic functional connectivity between regions throughout the brain , taking into account the local neighborhood connections as well as the remote or distant connections ( within and outside 14 mm of a neighborhood area ) ( Figure 1 ) . Control analyses showed that the method successfully and reliably identified distinct local degree values across the brain . Estimates of these values were then used to explore the properties of regions across the brain and to compare these estimates to those derived from well-known network measures including path length , physical cost , and clustering coefficient . Finally , we examined functional connectivity during an active task ( as contrast to rest ) to examine how functional coupling dynamically changes in response to task demands .
Local and distant functional connectivity are plotted separately ( Figure 2 ) as well as combined into maps of preferential connectivity ( local – distant; Figure 3 ) and overlap ( local ∩ distant; Figure 4 ) . Based on these maps regions could be characterized into three broad categories: 1 ) Regions displaying preferential local connectivity with less distant connectivity , involving mainly primary and secondary/modality-selective cortices ( motor , somatosensory , auditory , visual , and a region at or near the supplementary motor area [SMA] proper ) , 2 ) Regions displaying preferential distant connectivity with relatively low local connectivity including heteromodal areas in the lateral parieto-temporal and frontal cortices , and 3 ) Regions that contained both a high degree of local and distant connectivity including prominent midline regions that comprise components of the default network ( posterior cingulate , certain regions within the medial prefrontal cortex ) . The third connectivity profile is most clearly visualized by examining the overlap of the maps ( Figure 4 ) . Figure S1 and S2 display maps at several levels of threshold and left/right projections to illustrate that the topographies of the preferential and overlap maps are qualitatively consistent across thresholds . Volume displays of the maps are also provided for transverse sections in the atlas space of the Montreal Neurological Institute ( MNI ) ( Figure S3 ) . The full volume data are available from the authors upon request . A striking feature of the maps is that regions near primary sensory and motor cortex show strong preferential local connectivity . Examining the topography of the regions in more detail revealed that they track estimated boundaries of primary sensory and motor areas ( Figures 5 and 6 ) . For example , the regions of the visual system that show strong preferentially local connectivity overlap well with the early retinotopic areas that extend from V1 to V3a and V4 ( Figure 5 ) . In this regard , the analytic procedure of mapping local versus distant functional connectivity at rest is sufficient to reveal the well-established distinction between primary/secondary and association cortices . Regions with high distant degree connectivity and high local degree connectivity converged on multiple regions that fall within the default network [50] , [51] . Although functional connectivity patterns measured at rest provide valuable information about the intrinsic architecture of the brain , they are not synonymous with anatomic connectivity and are influenced by the task state ( see [47] for recent review ) . For this reason , we next explored the influence of task performance on local and distant connectivity . Two results emerged ( Figure 7 ) . First , engaging the task influenced both local and distant functional connectivity in regions typically active during performance of the abstract/concrete classification task . The changes were particularly prominent in the local connectivity estimates and included prefrontal cortex along the inferior frontal gyrus , lateral temporal cortex , dorsal anterior cingulate and a posterior parietal region linked to the frontal-parietal control system ( e . g . , [52] ) . Thus , one unexpected observation is that local functional connectivity can be used to measure engagement of task regions in a manner that is distinct from previous approaches to fMRI data analysis . A subtle change was also noted in increased ( relative ) distant connectivity in visual regions perhaps reflecting coupling of sensory regions to association areas during task engagement . Second , the regions of preferential local functional connectivity , as revealed by the direct contrast of the local to distant connectivity maps obtained from the task data , included the primary sensory and motor cortices ( Figure 7; right column ) . Inspection of the data in reference to cortical flattened representations once again showed that the strongest preferential local connectivity estimates were within or near early retinotopically-defined visual areas . That is , despite some relative changes in local and distant functional coupling during the task state , sensory areas still persisted in having preferentially local connectivity profiles . To situate our findings in the context of other well-known network measures , we computed the average path length , physical cost and clustering coefficient in our data set ( Figure 8 ) . Average path length is a measure of how far a node is , on average , from all other nodes in the network . Low path lengths ( blue in our scale in Figure 8; left column ) are those regions that have the shortest path lengths to other regions of the brain . Physical cost reflects , in some sense , the opposite property and plots , in our scale , regions with physically distant connected regions in yellow and orange . Clustering coefficient is a measure of segregation and , in our scale , displays regions with the greatest level of local modular organization in yellow and orange . As shown in Figure 8 , regions with preferential local connectivity fall within regions that are characterized by long path length , low physical cost and high clustering coefficient . Low levels of network topological path lengths and high levels of physical cost are prominent in regions of distant preferential connectivity . This relationship is perhaps most apparent when comparing the local degree map in Figure 2 and the clustering coefficient map in Figure 8 . It is also possible to detect differences between the local/distant preferential map ( Figure 3 ) and the network measures ( Figure 8 ) . The primary results of our analyses are the map estimates of local and distant functional connectivity . Several parameters were set to complete the analyses ( e . g . , the distance threshold ) and therefore processing decisions may have affected the results . A series of control analyses were conducted to boost confidence in the approach and to establish that the reported results are robust . First , test-retest reliability was assessed for the local and distant connectivity maps by comparing maps derived from two independent datasets each comprising 50 participants ( Figure S4 ) . High correlation coefficients between the two samples were obtained ( r = 0 . 95 for local and 0 . 91 for distant degree connectivity ) . Next , the influence of changing the distance threshold on the resulting local connectivity maps was examined by varying the neighborhood from a radius of 6 mm to 18 mm ( Figure S5 ) . The radius of 6 mm yielded a map that did not notably distinguish areal topography consistent with the limited spatial resolution of the technique . Results showed largely stable estimates of local connectivity for neighborhood radius values greater than 10 mm . We conservatively used a distance threshold of 14 mm for all analyses . The influence of Gaussian smooth was examined by comparing maps without spatial smoothing to the chosen 4 mm full-width half-maximum ( FWHM ) smoothing kernel ( Figure S6 ) . Removing the spatial smooth did not qualitatively affect the results; however , the preferential effects in the degree maps were generally less robust consistent with a reduction in signal-to-noise ratio . We further examined whether correlations between the hemispheres across the midline contributed to the observed results . Bilateral contributions might cause a bias in overestimating local connectivity values along midline structures . Maps that included degree connectivity for only one hemisphere were highly similar to those that included both hemispheres ( Figure S7 ) . Masking the cortex to include only the cortical mantle ( excluding subcortical regions including the basal ganglia , thalamus , and midbrain as well as the cerebellum ) also did not qualitatively change the results but did lead to several subtle differences presumably arising from exclusion of distant thalamic , striatal , and cerebellar connections ( Figure 8 ) . As a final exploration we examined the influence of the specific normalization approach and also the effect of grey matter volume correction ( Figure 8 ) . Again , the results were largely robust to analysis variations .
A strong expectation that human primary sensory cortex will possess properties consistent with a high local distance connectivity organization is provided by prior analyses of macaque anatomic connectivity ( e . g . , [2] ) . The absence of methods able to measure local anatomical connectivity , including detecting small u-fibers and local association connectivity , has limited the ability to visualize this basic organizational property . Our results demonstrate that differences in connectivity profiles were prominent across the cortex . Figure 3 summarizes these differences and Figures 5 and 6 provide detailed examination of sensory and motor areas . Regions that encompass estimated boundaries of early sensory and motor areas display high levels of local connectivity , most likely as a result of local anatomical connections between adjacent or nearby areas . Within the visual cortex , the region with the most pronounced local connectivity organization is at or near V1 with a gradual transition to less modular connectivity profiles as one moves along the hierarchical progression from V2 through V3a , supporting a functional gradient . Regions anterior to the estimated boundary of the MT+ complex display preferentially distant connectivity . Regions at or near primary somatosensory , auditory cortex , and motor cortex also display high levels of local connectivity . Some of the details of the mapping are not fully aligned with expectations , particularly the dense local connectivity observed in area 43 but not area 41 in auditory cortex . It is presently unclear whether this is a differential property of the area , an inaccuracy of the present method , or a consequence of inaccurate estimates of the areal boundaries . Of the regions studied , the best alignment between expectations from macaque anatomy and estimated human areal boundaries was for the retinotopically mapped visual areas [53] . Equivalent mapping in human auditory cortex was not available . Thus , an important future direction will be to map sensory areas within individual subjects and explore further the correspondence between areal boundaries and connectivity profiles . We also found that certain regions along the midline , including portions of the anterior cingulate , have preferential local connectivity and less distant connectivity similar to somatosensory/motor , auditory or visual cortices ( see inset detail in Figure 3 ) . In humans the portion of cingulate just anterior to the genu of the corpus callosum includes areas 24 and 32ac , possible homologues to macaque areas 24a/b/c [54] , [55] . In their seminal studies of medial prefrontal cortex in the macaque , Carmichael and Price [56] noted that areas 24a/b , 10m and 32 are tightly interconnected components of the “medial network” . Unlike other areas of high local connectivity , areas 24a/b do not receive significant projections from sensory systems ( barring some olfactory inputs to the region ) . Inputs are predominantly limbic . What is clear from the connectivity profiles is that cingulate cortex just anterior to the corpus callosum shows a markedly different connectivity profile than posterior cingulate and more anterior medial prefrontal regions that include human area 10 . These regions , as will be discussed in the next section , show among the highest levels of both local and distant connectivity . Certain regions were estimated to simultaneously possess both high levels of local and distant connectivity ( Figure 4 ) . The posterior cingulate , medial prefrontal cortex , and inferior parietal lobule were the three most extreme examples of the hybrid connectivity profile . The importance of these regions as connectional hubs , in particular the posterior cingulate , has been noted previously based on anatomical [57] , [58] , diffusion [35] , [59] , and functional [37] , [50] , [60] connectivity data . Prior functional connectivity analyses using graph theory have previously revealed that these regions are hubs of long-distance cortical-cortical interactions with both high degree and betweeness centrality [36] , [37] . What is new here is that a subset of these regions also possess among the highest levels of local functional connectivity . The regions displaying the hybrid connectivity profile overlap regions that belong to the default network [50] , [51] . This network has been implicated in cognitive functions associated with internal thought as contrast to stimulus-based perception . For example , the network is active during autobiographical memory retrieval [61] , imagining the future [62] , and mind wandering [63] ( see [50] , [64] for reviews ) . These observations raise two questions: What are the functional consequences of a hybrid functional connectivity profile and why is it so prominently represented in the default network ? While there are too few constraints to offer more than broad speculations , it is worthwhile to generate hypotheses to encourage further exploration . Given that the regions possessing hybrid connectivity profiles are active during internal modes of thought , including during passive fixation , it is intriguing to hypothesize that information processing that persists independent of strong sensory constraints might require a set of modular , tightly coupled areas to maintain efficient local processing . That is , the default network may possess features so that information processing is able to maintain stable in situ information ( high local network connectivity ) and simultaneously to associate distributed information from key limbic , parietal and prefrontal regions of the brain ( high distant network connectivity ) . This idea extends what has been previously articulated . For example , in Mesulam's foundational work on transmodal areas [6]–[8] , he emphasizes their role as pointers to distributed modular systems . The present hypothesis expands this notion to also explore processing contributions that arise directly from extensive local connectivity between and within certain contiguous association areas . An interesting unexpected result was the detection of clear local functional coupling changes when a task was engaged ( Figure 7 ) . Most of our analyses were of resting-state fixation data similar to the approach common to the literature . When a continuous task involving semantic classification of words was engaged , the preferential local connectivity profiles in sensory and motor regions were largely retained with the addition of strong local coupling within many of the distributed regions typically activated by the task . In fact , the map of increased local functional connectivity in the task state relative to rest fixation ( Figure 7; left column ) could easily be mistaken for a typical blocked-task functional MRI or positron emission tomography ( PET ) paradigm ( e . g . , compare our Figure 7 with Supplementary Figure 3 in [65] ) . There are two separate implications of this observation . First , the result is a reminder that low-frequency functional correlations as measured by BOLD contrast fMRI are not an exact proxy for anatomical connectivity . Coupling among regions changes as a function of task state . Second , measures of local functional coupling may provide a means to investigate regional activity levels during tasks . It is possible , although not explicitly tested here , that local functional coupling may provide a powerful approach for identifying task-activated regions including for brief epochs of task performance . It will be interesting in future studies to explore this possibility and to generally examine what can be learned from task-induced changes in local functional coupling . A radical possibility is that task contrasts will not be required and local functional coupling by itself , or referenced to other aspects of the data or normative data , will be sufficient to estimate properties of brain activity . The present work maps relative connectivity profiles that distinguish local and distant functional connectivity . As such it moves into an arena that is at the resolution boundary of present functional imaging approaches . Further , the connectivity method applied was based on intrinsic activity correlations measured using BOLD contrast fMRI . As we have discussed previously [47] , there are strengths and limitations of this indirect approach for estimating connectional anatomy . Relevant here is the observation that intrinsic activity correlations can reflect more than direct monosynapatic connections including contributions of polysynapic projections and common driving inputs , such as from the thalamus . Our results also reveal clear , task-dependent modulation of functional coupling . Thus , it will be important to employ convergent methods to validate the anatomic properties that we infer . Diffusion-based methods that employ high angular resolution diffusion ( e . g . , DSI ) may soon be able to map u-fibers and local association fibers [44] . To make a reasonable assessment of whether the properties we are observing reflect an artifact of smoothing or other processing steps , or the boundaries of functional resolution , we conducted a large number of control analyses ( see Supplementary Figures S1 , S2 , S3 , S4 , S5 , S6 , S7 ) . The core results are reliable and robust across alternative smoothing and processing decisions . Thus , while we anticipate further refinement of the methods , we expect that the main results of the paper are valid . One potential limitation of our study is that we are approximating real cortical distance with Euclidean distance . In the case of nodes in adjacent gyri , the Euclidean distance is underestimating the true cortical distance , since the white matter tracts often bend around the intervening sulcus . Future work will be needed in order to include additional information regarding surface distance and ideally white matter tract path lengths ( to approximate this approach see an example in [66] ) . A further limitation of the present paper is the focus on group data . An important avenue for future work will be to explore these properties , perhaps even at higher spatial resolution , in the context of task-based estimates of functional areas . For example , it will be useful to explore whether individual differences in estimated boundaries of retinotopic visual areas track estimates of local distance connectivity . We predict they will . Similarly , the regions of preferentially distinct connectivity overlap with regions that are active during tasks of remembering , foresight , and other forms of high-level cognition [50] , [64] . It will be important to study more explicitly the overlap between connectivity profiles and activity during these forms of cognitive task at the individual subject level . Exploration of individual differences has potentially important implications for study of genetics and neuropsychiatric illness ( e . g . Alzheimer's disease , epilepsy , schizophrenia , bipolar disorder , and autism ) . A particularly interesting area of future exploration concerns the development of local and distinct connectivity and the relation of our metrics to atypical development . Certain neuropsychiatric disorders are suspected to result from molecular disruptions that give rise to aberrant connectivity patterns . For example , autism is associated with overgrowth of the brain early in development and adult white-matter abnormalities [67] . Atypical development may affect the fragile balance between local and distant connectivity . Relevant to this possibility , Fransson and colleagues [68] demonstrated functional connectivity in infants at typical birth age ( the infants were preterm ) . They found strong connectivity among sensorimotor networks but did not identify connectivity within the association networks typical of adults , including the default network . Fair et al . [69] reported that the distributed connectivity pattern within regions defining the default network – the prototypical regions in our maps of distant connectivity – is not fully present in young children . An expanded analysis of the phenomenon demonstrated that childhood development is characterized by a general trend toward increases in functional connectivity across widely distributed regions conceptualized as the development of a ‘local to distributed’ organization [70] . These prior analyses suggest that detailed analysis of the development of local- and distant-connectivity profiles may provide important insights for both typical and atypical neurodevelopmental trajectories . Functional connectivity MRI was used to analyze network properties across the human brain introducing spatial distance information . We discovered that mapping regions based on whether they exhibit preferentially local versus preferentially distant functional connectivity at rest easily separates early sensorimotor , heteromodal association cortices and core regions of the default mode network . This observation reveals a parsimonious property of cortical network architecture that divides processing between many parallel systems characterized by extensive local processing and transmodal regions that serve as hubs connecting these local systems . As a practical application of our approach , metrics of connectivity profiles that reflect local and distributed connectivity can be made rapidly in individual participants . These metrics may thus have value for exploring individual differences both in relation to genetics and also in developmental neuropsychiatric disorders where atypical connectivity profiles may be present . More broadly , the observation that connectivity hubs fall within regions of estimated cortical expansion between monkey and humans [19] , and also regions of late child development [71] , reinforces the hypothesis that association areas make an important contribution to higher-order cognitive functions that are especially well developed in humans .
112 healthy young adults participated in MRI for payment . Table 1 shows the participant demographics . All participants had normal or corrected-to-normal vision and were right-handed , native English speakers . Participants were screened to exclude individuals with a history of neurologic or psychiatric conditions as well as those using psychoactive medications . Resting-state data from these participants have been reported previously [37] , [65] and are openly available to the community upon request . Written informed consent was obtained in accordance with guidelines set forth by the institutional review board of Partners Healthcare Inc , and this research has been conducted according to the Declaration of Helsinki . Scanning was performed on a 3 Tesla TimTrio system ( Siemens , Erlangen , Germany ) using the 12-channel phased-array head coil supplied by the vendor . High-resolution 3D T1-weighted magnetization prepared rapid acquisition gradient echo ( MP-RAGE ) images were acquired for anatomic reference ( TR = 2530 ms , TE = 3 . 44 ms , FA = 7° , 1 . 0 mm isotropic voxels ) . Functional data were acquired using a gradient-echo echo-planar pulse sequence sensitive to BOLD contrast ( TR = 2500 ms , TE = 30 ms , FA = 90° , 36–43 axial slices parallel to plane of the anterior commissure-posterior commissure , 3 . 0 mm isotropic voxels , 0 . 5 mm gap between slices ) . Head motion was restricted using a pillow and foam , and earplugs were used to attenuate scanner noise . During the functional runs , the participants fixated on a visual cross-hair ( plus sign , black on white ) centered on a screen for each of two runs ( each run 7 min 24 sec; 148 time points ) . Participants were asked to stay awake and remain as still as possible . For the task condition , we used a data set previously reported in Buckner et al . [37] . Briefly , two runs of continuous task performance and two runs of fixation were acquired in twelve subjects ( each run 5 min 12 sec; 104 time points ) . Participants decided whether centrally presented visual words represented abstract or concrete entities . Order of task was counterbalanced across participants . The visual stimuli were generated on an Apple PowerBook G4 computer ( Apple , Inc . , Cupertino , CA ) using Matlab ( The Mathworks , Inc . , Natick , MA ) and the Psychophysics Toolbox extensions [72] . Stimuli were projected onto a screen positioned at the head of the magnet bore . MRI analysis procedures were optimized for functional connectivity MRI ( fcMRI ) analysis [47] extending from the approach developed by Biswal et al . [45] . The first four volumes were removed to allow for T1-equilibration effects , followed by compensation of systematic , slice-dependent time shifts , motion correction and normalization to the atlas space of the MNI ( SPM2 , Wellcome Department of Cognitive Neurology , London , UK ) to yield a volumetric time series resampled at 2 mm cubic voxels . Temporal filtering removed constant offsets and linear trends over each run while retaining frequencies below 0 . 08 Hz . Data were spatially smoothed using a 4 mm FWHM Gaussian blur ( note that the effect of smoothing was explicitly considered in control analyses below ) . Several sources of spurious or regionally nonspecific variance then were removed by regression of nuisance variables including: six parameter rigid body head motion ( obtained from motion correction ) , the signal averaged over the whole-brain , the signal averaged over the lateral ventricles , and the signal averaged over a region centered in the deep cerebral white matter . Temporally-shifted versions of these waveforms also were removed by inclusion of the first temporal derivatives ( computed by backward differences ) in the linear model . This regression procedure removes variance unlikely to represent regionally specific correlations of neuronal origin . Of note , the global ( whole-brain ) signal correlates with respiration-induced fMRI signal fluctuations [47] , [73] , [74] . By removing global signal , variance contributed by physiological artifacts is minimized . Removal of signals correlated with ventricles and white matter further reduces non-neuronal contributions to BOLD correlations . Removal of global signal also causes a shift in the distribution of correlation coefficients such that there are approximately equal numbers of positive and negative correlations making interpretation of the sign of the correlation ambiguous [47] , [75] , [76] . This effect is not directly relevant to the current analyses as degree connectivity is computed based on the correlations that exceed a positive threshold . Finally , for computational efficiency , we down sampled the data to 4 mm isotropic voxels . The present study used fcMRI to map the local and distant degree of functional connectivity in the human brain . fcMRI measures intrinsic activity correlations between brain regions [45]–[47] . The method assumes that fcMRI is sufficiently constrained by anatomy to reveal informative estimates of connectivity properties . We have previously outlined the reasons for this assumption as well as the caveats and limitations of fcMRI [47] . For our present purposes it is important to make clear that fcMRI can reflect mono- and polysynaptic connectivity , correlations arising from common sources , and task-dependent dynamic functional coupling . Thus , it should not be considered a direct measure of anatomic connectivity . Nonetheless , fcMRI reflects , to a large degree , the statistical properties of anatomical connections and therefore provides a great deal of indirect information about human connectional anatomy . Degree centrality ( or degree ) is a network measure that quantifies the number of links or edges connected to a node [27] . Here brain voxels ( that sample small regions of cortex ) are the nodes and positive correlations between voxels above certain strength are the links or edges in the graph . A computationally-efficient approach was used to map the degree of functional connectivity across the brain at the voxel level in a large number of individuals [37] taking into account topographical neighborhood information for the local and distant distinction . Thus , we computed a variation of the classic degree centrality measure in graph theory ( e . g . , [27] , [77] ) by introducing physical distance restrictions in the whole-brain voxel-by-voxel functional connectivity network . The immediate neighborhood was taken into account to generate a local degree map and functional connectivity outside of this neighborhood was taken into account to generate a distant degree map . Different parameters of neighborhood threshold , in terms of radius sphere , were tested from which we choose 14 mm radius ( approximately 3 voxels around target voxels ) ( see Figure S5 ) . This distance threshold provides information about connectivity that is likely to reflect communication between local ( nearby ) areas and minimizes the correlations that reflect smoothing between adjacent voxels . For these analyses , the time course of each voxel from the participant's brain defined within a whole-brain mask was correlated to every other voxel time course . As a result an n×n matrix of Pearson correlation coefficients was obtained , where n is the dimension of the whole-brain mask . The Pearson R , or product-moment correlation coefficient , computed in the ith row and jth column of this matrix is given by: ( 1 ) where t is the frame count , and are the voxel intensities at the ith and jth voxel location respectively defined by the whole-brain mask at frame count t . The mean voxel intensity across all of the time points at the ith and jth voxel locations is given by and respectively . From the Pearson correlation coefficients , a map of degree connectivity was computed . We computed the local degree map by counting for each voxel the number of voxels above a correlation threshold of r>0 . 25 inside its neighborhood ( defined as a 14 mm radius sphere ) , and for the distant degree map by counting for each voxel the number of voxels above the same threshold but outside the neighborhood . The r threshold was chosen to eliminate counting voxels that had low temporal correlation due to signal noise ( see [37] for analysis of the effects of r threshold ) . No gap for counting voxels was included between the local and distant degree measures . Finally , undirected and unweighted local and distant degree values were estimated for each voxel of the brain . The measure of degree connectivity for a voxel is given by: ( 2 ) where j is the inside or outside neighborhood voxels of the i depending on the measure of local or distant degree map . The degree connectivity map was then standardized by Z-score transformation so that maps across participants could be averaged and compared [37] . The conversion to Z-score does not influence the topography of the individual-participant maps but does cause the values in each participant's map to be comparable across subjects . In the present case , the Z-score transformation was computed separately for the local and distant degree connectivity maps . Moreover , these final Z-score degree maps were used to create the preferential and overlap maps . We refer here to preferential not as an absolute measure of number of links but rather as a relative measure of the overall topography differences ( LocalZ-score minus DistantZ-score ) . Finally , the overlap map ( Figure 4 ) was created by combining the local and distant Z-score maps after thresholding each of them at 1 SD in order to isolate only regions showing the strongest effects ( see Figure S2 for other threshold criteria for the overlap map ) . To situate our findings in the context of other well-known network measures , we computed in the same sample of subjects the following measures per node: path length ( Figure 8; left column images ) , physical cost ( Figure 8; center column images ) and clustering coefficient ( Figure 8; right column images ) , using matlabBGL ( http://www . stanford . edu/~dgleich/programs/matlab_bgl/ ) and Boost Graph Library ( http://www . boost . org/doc/libs/1_42_0/libs/graph/doc/index . html ) . For each subject , the functional connectivity MRI time series was first down-sampled from 4 to 8 mm voxel size for computational efficiency . We then formed a graph following the same principles as explained above but this time at 8 mm resolution . The path length of a node was calculated as the average path lengths from the node to all other nodes in the graph . Since each pair of connected nodes lie in physical space , we can assign a cost to each edge in the graph based on the physical Euclidean distance between the nodes in the brain . The physical cost of a node is therefore the average of the physical costs of all the edges of the node defined for each voxel as:where the sum is over all the neighbors of voxel i , Dj is the degree of voxel i and fij is the Euclidean distance between voxel i and voxel j . The clustering coefficient of a node is the proportion of all pairs of its neighbors that are directly connected to each other in the graph or , in other words , the number of links between the neighborhood vertices divided by the number of links that could possibly exist between them [31] . The measure of clustering coefficient for each voxel is given by:where Ei is the actual number of links between the neighborhood vertices of voxel i and Di is the degree of voxel i . Observe that Di ( Di−1 ) /2 is the number of possible edges that could exist between the neighbors of voxel i . Several control analyses were performed to explore the influences of processing decisions on the degree connectivity maps , to measure test-retest reliability , and to estimate the effects of local anatomy ( Figures 9 and S1 , S2 , S3 , S4 , S5 , S6 , S7 ) . We first evaluated the possible effects of distance threshold as well as spatial smoothing . Another concern is that the local degree connectivity measure along the midline is contaminated by strong correlations between homologous voxels across the left and right hemispheres . To explore this issue , we computed degree connectivity maps that masked the whole-brain ( both hemispheres ) as well as control analyses that restricted connectivity to within the hemisphere . We examined the influence of brain mask by comparing masks that involved the whole brain ( including subcortical structures ) versus a mask the included only cortical gray matter ( excluding the basal ganglia , thalamus , and midbrain as well as the cerebellum ) . We also examined the influence of the local volume of grey matter in the neighborhood of the voxel ( we refer to this as gray matter correction ) . For this analysis , the preferential map of local versus distant connectivity was estimated taking into account the number of grey matter voxels included in the search sphere rather than the absolute count , in order to appropriately weight regions that may contain less grey matter volume than others ( such as voxels in the cortical/non-cortical interfaces ) . The normalized grey matter mask was obtained from SPM MarsBar toolbox ( http://www . fil . ion . ucl . ac . uk/spm ) . Finally , we tested a second method to normalize degree maps in order to verify that our normalization procedure is not biasing our results using percentage normalization ( [Distant Degree×100]/[Distant Degree+Local Degree] ) . While these control analyses do not represent an exhaustive set of possible processing variations , they bolster confidence that the major aspects of obtained results likely reflect properties of intrinsic functional connectivity that are not artifacts of anatomy or a specific processing decision . Data were visualized on the cortical surface using the population-average , landmark- and surface-based ( PALS ) surface and plotted using Caret software [53] , [78] . The PALS atlas is based on the PALS-B12 dataset from [79] and projects estimated areal boundaries from Broadmann's original architectonic scheme [80] to the surface . These area estimates are thus to be considered approximate . Reference boundaries for visuotopic-mapped areas ( e . g . , V1 , V2v/d , V3 ) are based primarily on fMRI studies of human retinotopic mapping ( e . g . , [81]; see [53] for discussion ) .
|
Information processing in the human brain arises from both interactions between adjacent brain areas and from distant projections that form distributed systems . Here we estimated functional connectivity profiles in the human brain using a novel approach to map the regional balance between local and distant functional connectivity . We discovered that the human brain exhibits distinct connectivity profiles across regions with primary sensory and motor areas displaying preferential local connectivity and heteromodal association areas displaying preferential distant connectivity . These findings expand our knowledge of how the human brain has specialized its architecture to optimize processing efficiency and provides an approach to measure , in individuals , the degree to which the typical balance of local and distant connectivity is present .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"neuroscience/cognitive",
"neuroscience",
"neurological",
"disorders/cognitive",
"neurology",
"and",
"dementia",
"neurological",
"disorders/neuroimaging",
"computational",
"biology/computational",
"neuroscience",
"neurological",
"disorders/neuropsychiatric",
"disorders",
"neuroscience/psychology",
"neuroscience",
"neuroscience/theoretical",
"neuroscience"
] |
2010
|
The Organization of Local and Distant Functional Connectivity in the Human Brain
|
Imprinted genes display biased expression of paternal and maternal alleles and are only found in mammals and flowering plants . Compared to several hundred imprinted genes that are functionally characterized in mammals , very few imprinted genes were confirmed in plants and even fewer of them have been functionally investigated . Here , we report a new imprinted gene , NUWA , in plants . NUWA is an essential gene , because loss of its function resulted in reduced transmission through the female gametophyte and defective cell/nuclear proliferation in early Arabidopsis embryo and endosperm . NUWA is a maternally expressed imprinted gene , as only the maternal allele of NUWA is transcribed and translated from gametogenesis to the 16-cell globular embryo stage after fertilization , and the de novo transcription of the maternal allele of NUWA starts from the zygote stage . Different from other identified plant imprinted genes whose encoded proteins are mostly localized to the nucleus , the NUWA protein was localized to the mitochondria and was essential for mitochondria function . Our work uncovers a novel imprinted gene of a previously unidentified type , namely , a maternal-specific expressed nuclear gene with its encoded protein localizing to and controlling the function of the maternally inherited mitochondria . This reveals a unique mechanism of maternal control of the mitochondria and adds an extra layer of complexity to the regulation of nucleus-organelle coordination during early plant development .
Genomic imprinting is a phenomenon in which somatic cells express some genes only from the maternal or paternal chromosome [1] . Genes with parent-of-origin-specific allele-biased expression patterns are called imprinted genes [2 , 3] . Imprinted genes have only been discovered and confirmed in placental mammals and flowering plants and are thought to have evolved independently through convergent evolution in these two groups [4 , 5] . Imprinting is regulated epigenetically and plays important roles in mediating complex traits in both mammals and plants [6–8] . In mammals , imprinted genes are essential not only in embryonic development and placental development of the fetus but also in sensory function and behavior in adults [9–11] . Loss-of-function of these genes always results in severe disease [12 , 13] . In plants , imprinted genes are also involved in early development , and the loss-of-function of these genes can lead to seed-lethal phenotypes [2 , 3] . In comparison with the hundreds of functional imprinted genes identified in mammals , confirmed imprinted genes in plants are rare , and many of them do not have obvious functions , because no phenotypes were observed in the loss-of-function mutants [1 , 13–17] . In addition , products of the imprinted genes in mammals have a variety of subcellular localizations and numerous molecular functions [18] , whereas in plants , products of most of the confirmed imprinted genes are localized in the nucleus , and many are presumed to function in chromatin remodeling [19–28] . Although high-throughput analyses revealed many putative imprinted genes in Arabidopsis [29 , 30] , maize [31] and rice [32 , 33] , most of them have not been confirmed . It is not known whether the imprinted genes in plants are important in many aspects of subcellular and biological processes , as in mammals , and the driving force of the convergent evolution of genomic imprinting therefore remains a mystery . In this study , we have identified a novel maternally expressed imprinted gene NUWA in Arabidopsis , which is named after the well-known goddess in Chinese ancient mythology who created humans by molding figures from earth . NUWA is an essential gene . Loss of NUWA function resulted in defects in cell/nuclear proliferation in early embryogenesis and endosperm development . Moreover , NUWA is a maternally expressed imprinted gene , because de novo transcription of maternal allele-specific NUWA was detected in the embryo sac after fertilization . Different from those previously-identified plant imprinted genes whose products were localized to the nucleus , the NUWA protein was targeted to the mitochondria and was essential for development and proper function of the mitochondria . These results indicate that NUWA is a new type of imprinted gene that maternally controls early embryo and endosperm development through regulating the function of the maternally inherited mitochondria . The discovery of NUWA also reveals a new aspect of subcellular and metabolic processes in which plant imprinted genes are involved .
The mutant nuwa/+ was obtained from genetic screening of a mutant collection with a Basta/Phosphinothricin ( PPT ) -resistant gene in the T-DNA [34] . In nuwa/+ siliques , 37 . 5% of seeds aborted very early; 1 . 6% of seeds aborted somewhat later ( n = 6685 ) ( Fig 1A ) . In selfed nuwa/+ progeny , the PPT-resistant ( PPTr ) to PPT-sensitive ( PPTs ) ratio was approximately 1 . 33:1 ( n = 4387 ) . No nuwa/- homozygous mutant plants were acquired . When we pollinated nuwa/+ with pollen of a fertilization indicator line FAC1-GUS , whose GUS signal appears 3 hours after fertilization in embryo sacs [35] , the percentage of ovules with GUS signals was indistinguishable between wild type and nuwa/+ ( |u| = 0 . 248 < u0 . 05 = 1 . 96 , P > 0 . 05 ) ( S1A to S1C Fig ) , suggesting that all of the defective female gametophytes of nuwa/+ could be fertilized . Therefore , we hypothesized that the reduction in the seed setting rate in nuwa/+ probably resulted from a defect in the embryo and/or endosperm after fertilization . Using differential interference contrast ( DIC ) microscopy , we found that mutant embryos of nuwa/+ were arrested before the 16-cell stage . Most mutant embryos were arrested at the first three rounds of cell division , and some displayed abnormal cell shapes and irregular cell division patterns ( Fig 1B ) . This result suggested that the mutant embryos were arrested at very early developmental stage , with cell division defects . Then , we analyzed endosperm development using confocal laser scanning microscopy . Only 42 . 7% of nuwa/+ ovules at 26 hours after pollination ( HAP ) were at the 8-endosperm-nuclei stage , which was significantly lower than the percentage in wild-type ovules ( 78 . 6% ) ( |u| = 3 . 91 > u0 . 01 = 2 . 58 , P < 0 . 01 ) ( Fig 1C ) . The percentage of ovules at 4-endosperm-nuclei stage and 2-endosperm-nuclei stage in nuwa were significantly higher than those in wild type ( Fig 1C ) , indicating that the endosperm nuclei proliferations were slower in the mutant . After we crossed nuwa/+ with FIS2-GUS [36] , we found that fewer spherical GUS signals indicating endosperm nuclei were observed in mutant ovules compared with normal ovules ( Fig 1D , S1D to S1I Fig ) , further supporting the notion that endosperm nucleus proliferation is delayed in mutant ovules . The facts that both embryo and endosperm of nuwa/+ developed slowly after fertilization and were arrested at early stages demonstrate that NUWA is an essential gene . Using TAIL-PCR , we found that the T-DNA in nuwa/+ mutant was inserted into the single exon of the gene At3g49240 , which contains a transposable element , AT3TE74080 ( Fig 2A ) . When the full-length genomic DNA of At3g49240 was introduced into nuwa/+ , homozygous mutant plants with normal fertility were obtained ( Fig 2B ) , indicating that the phenotypes of nuwa/+ were caused by loss of At3g49240 function . At3g49240 was designated as NUWA , the name of the Chinese goddess for human creation . Sequence similarity and synteny analysis showed that NUWA is a single copy gene with high sequence similarity in different plant species ( S2 Fig ) . We then investigated the nucleotide diversity of a 6-Kb region centered around NUWA in five Brassicaceae species , and found that the nucleotide diversity of the coding region was significantly lower than that of the promoter region and its 3’ downstream sequence; and also lower than genome-wide average ( 0 . 007 ) [37] ( Fig 2C; S2 Fig ) . This relative low level of nucleotide diversity supports that NUWA is conserved , with its evolution driven by purifying selection . We investigated NUWA expression patterns at both transcriptional and protein levels . First , real-time quantitative RT-PCR ( qRT-PCR ) analysis showed that NUWA was transcribed in all organs tested and was highly expressed in dry seeds and imbibed seeds ( Fig 2D ) . We then generated ProNUWA:NUWA-GFP-GUS transgenic plants to test the protein distribution pattern of NUWA . In the inflorescence , GUS signals appeared at the tips of pistils and the petals of young flowers ( Fig 2E ) . In ovules , GUS activity was detected on integuments before ( Fig 2F ) or early after fertilization ( Fig 2G ) , in embryos at 2–4 cell stage ( Fig 2G ) , heart stage ( Fig 2H ) , torpedo stage ( Fig 2I ) and walking stick stage ( Fig 2J ) and later . In seedlings , GUS activity was detected in cotyledons , young leaves , root tips and lateral root primordia ( Fig 2K and 2L ) . Little GUS activity was observed in mature leaves ( Fig 2M ) . These results suggest that NUWA is highly expressed in the reproductive organs of Arabidopsis . When reciprocal crosses were conducted between nuwa/+ and wild-type plants , male transmission was normal , while female transmission was severely reduced to 15 . 6% ( Fig 3A ) . The decreased female gametophyte transmission rate , together with the observed embryo and endosperm defects , implies that nuwa/+ is a maternal effect mutant . To verify this hypothesis , we pollinated nuwa/+ with either wild-type or nuwa/+ pollen . Seed-lethal phenotypes were observed in siliques ( Fig 3B ) , and the percentage of aborted ovules were identical between the two crosses ( |u| = 0 . 3255 < u0 . 05 = 1 . 96 , P > 0 . 05 ) ( S3A Fig ) , suggesting that the main phenotypes of the mutant seeds were controlled by the maternal genotype rather than that of both parents . These results confirmed that nuwa/+ is a maternal effect mutant . To elucidate the cause of the maternal effect in nuwa/+ , we first investigated the expression of NUWA before fertilization . We transformed the ProNUWA:NUWA-GFP-GUS construct into nuwa/+ to genetically complement the mutant and to investigate NUWA protein expression patterns in mature gametophytes in the wild-type background . GUS activity was observed in the egg apparatus ( Fig 3C ) and central cells ( Fig 3D ) , but not in mature pollen located on the stigma ( Fig 3E ) or dispersed ( Fig 3F ) . Real-time qPCR analysis showed that NUWA transcription levels were much lower in pollen than in mature ovules ( Fig 3G ) . These results suggest that NUWA is predominantly expressed in female gametophytes before fertilization , with sparse expression occurring in male gametophytes . We next analyzed the parent-of-origin-dependent expression patterns of the NUWA protein in ovules after fertilization . We reciprocally crossed homozygous ProNUWA:NUWA-GFP-GUS transgenic plants with wild-type ones . Maternal-specific GUS signal was detected in embryos and endosperm from the zygote to the 16-cell-embryo stage ( Fig 3H to 3M ) . In contrast , a paternal-specific , embryo-specific GUS signal was not observed in early seeds ( Fig 3N and 3Q ) until the 16-cell-embryo stage ( Fig 3O to 3P and 3R to 3S ) . These results indicate that only maternal-allele-specific expression of the NUWA protein is detected in early seeds . To further analyze the parent-of-origin-dependent expression pattern of NUWA transcripts in ovules after fertilization , we adopted allele-specific RT-PCR . A single-nucleotide polymorphism ( SNP ) between Arabidopsis ecotypes Col-0 and Ler in the exon of NUWA was used to distinguish parental origins of endogenous NUWA transcripts in seeds resulting from reciprocal crosses . We found that , for Col-0×Ler cross , all the NUWA transcripts detected in 24 and 72 HAP were from Col-0 allele , and a small proportion of the transcripts detected in 120 HAP were from Ler allele . Similarly , for Ler×Col-0 cross , all the NUWA transcripts detected in 24 HAP and 72 were from Ler allele , and a small proportion of the transcripts detected in 120 HAP were from Col-0 allele . These results indicate that , while transcripts derived from the maternal allele were detected in seeds from 24 HAP to 120 HAP , transcripts derived from the paternal allele were not detected in seeds until 120 HAP ( Fig 3T ) . This result suggests that in the early seeds , only maternal-allele-specific NUWA transcripts are present in ovules , with transcripts of paternal allele first emerging between 72 HAP and 120 HAP . These data indicate that the maternal allele of NUWA is specifically detected at both transcriptional and translational levels before and shortly after fertilization . To precisely investigate the maternal expression pattern of NUWA , we analyzed our RNA-seq data of ovules at different developmental stages . We found that the amount of NUWA transcripts reduced at 6 HAP comparing to 0 HAP , then increased at 12 HAP , and decreased again at 24 HAP ( S3B Fig ) . Because fertilization is completed within 6 HAP in Arabidopsis [38] , this result reveals that NUWA is transcribed in ovules after fertilization . As NUWA expresses on integuments at early development stages , we then isolated wild type embryo sacs at 0 HAP , 6 HAP , 12 HAP and 24 HAP ( Fig 4A to 4D ) and examined transcripts of NUWA . The qRT-PCR results showed the transcript level of NUWA in embryo sacs decreased at 6 HAP before increasing at 12 HAP ( Fig 4E ) , which confirmed that NUWA is transcribed de novo in embryo sacs after fertilization . Because the paternal allele of NUWA transcript is not detected before 120 HAP ( Fig 3T ) , this result suggests that the de novo expressed NUWA transcripts are predominantly from the maternal allele . Since imprinted genes are known as genes expressed predominantly from one of their parental alleles during a period of diploid stage [2] , our results also suggests that NUWA is an imprinted gene . To confirm this hypothesis , we isolated the embryo sacs at 12 HAP from progeny plants of reciprocal crosses between wild type Col-0 and Ler , extracted RNA and performed allele-specific RT-PCR . The results showed that in 12 HAP embryo sacs , only maternal-allele-specific NUWA transcripts were detected ( Fig 4F ) , which supports that the de novo expression of NUWA transcripts detected in early developmental stage are predominantly from the maternal allele . These results indicate that NUWA is a maternally expressed imprinted gene . When we investigated the subcellular localization of NUWA , the GFP fluorescence signal was found to co-localize with the mitochondria stained with the specific dye MitoTracker Orange in seedling cells ( Fig 5A ) . The similar GFP signal was also observed in the heart-stage embryo , and the signal did not co-localize with the plastid auto-fluorescence ( S4 Fig ) . Additionally , a putative mitochondrial-targeting peptide ( mTP ) was predicted in the N-terminus of NUWA [39] . These results suggest that NUWA is localized in mitochondria . When we introduced a ProNUWA:NUWAΔmTP-GFP into nuwa/+ , the mitochondrial-co-localized GFP signal disappeared although the GFP transcript was detected ( S5 Fig ) , implying that mitochondrial localization of NUWA is mediated by the mTP . In addition , no homozygous nuwa was identified among 180 T2 transformants , demonstrating that the NUWAΔmTP-GFP fragment could not rescue nuwa . These results suggest that the mitochondrial localization of NUWA is essential for its proper function . To investigate whether NUWA affects mitochondrial development , we adopted transmission electron microscopy ( TEM ) to analyze ultrastructure in early embryos and endosperm . Compared with mitochondria in wild-type embryos ( Fig 5B ) and endosperm ( Fig 5C ) , the mitochondrial matrix of nuwa mutant embryos and endosperm was less dense , with the cristae composed of structurally looser inner membranes ( Fig 5D and 5E , S6 Fig ) . Statistical analysis showed that the percentages of abnormal mitochondria in nuwa/+ mutant embryos and endosperm were 64 . 3% and 51 . 4% , respectively , both significantly higher than those in wild-type plants ( 8 . 60% and 2 . 19% , respectively ) ( Fig 5F and 5G ) ( |u| = 25 . 15 > u0 . 01 = 2 . 58 , P <0 . 01; |u| = 26 . 88 > u0 . 01 = 2 . 58 , P < 0 . 01 ) . This result indicates that abnormal mitochondria were significantly more abundant in nuwa mutant ovules . We then investigated mitochondrial functional status in isolated early embryos by using the mitochondrial trans-membrane potential indicator JC-1 . In healthy cells with mitochondria at high membrane potential , JC-1 exhibits red fluorescence; in abnormal cells with mitochondria at low membrane potential , JC-1 exhibits green fluorescence [40 , 41] . Results show that red signals of JC-1 were visibly stronger in many wild type embryos ( Fig 5H and S7 Fig ) , whereas JC-1 exhibited much stronger green signals in many nuwa mutant embryos at the same development stages ( Fig 5I and S7 Fig ) . Quantitative analysis showed that the red-to-green fluorescence ratios of nuwa embryos were significantly lower than those of wild type embryos ( |t| = 9 . 277 > t0 . 005 = 3 . 591 , P < 0 . 005 ) ( Fig 5J and 5K ) , indicating that the mitochondrial membrane potential levels of the mutant embryos were decreased compared to the wild type embryos . This result suggests that the mutant embryos of nuwa/+ have defective mitochondria , and the abnormal phenotypes of mutant embryos and endosperm might be attributed to mitochondrial dysfunction . Because PPR proteins are widely involved in regulating the post-transcriptional processing of organelle encoded genes through affecting splicing , editing and translation of organelle transcripts [42 , 43] , we intend to investigate whether NUWA is involved in regulating the expression level and splicing of mitochondrial genes . We extracted RNA from isolated endosperm of nuwa mutant and wild type plant at the same developmental stage , and analyzed , by using real-time qPCR , the expression level of those intron-containing mitochondrial genes and of those ovule-highly expressed genes chosen based on our preliminary RNA-seq data and the expression pattern data in Arabidopsis eFP Browser Database [44] . We also analyzed the trans splicing of the three NAD1 genes , two NAD2 genes and three NAD5 genes . The result showed that the expression levels of these genes were higher in the nuwa mutant , and no defect in splicing was detected ( S8 Fig ) , suggesting that the expression level of mitochondria encoded genome is abnormal in the mutant , and that NUWA is probably not involved in splicing and stabilization processes of mRNA encoded by mitochondria . NUWA encodes a P-subfamily pentatricopeptide repeat ( PPR ) protein of 629 amino acids with 11 PPR motifs ( Fig 6A ) . PPR proteins , which form one of the largest protein families in higher plants , are named after the eukaryote-specific PPR repeat that typically comprises 35 amino acid residues , and have RNA-binding activities [45 , 46] . Not all PPR motifs contribute directly to the RNA-binding activity of PPR proteins , and some PPR proteins have protein-binding activity [46–48] . We therefore investigated whether all the PPR motifs in NUWA are equally important to its function . We also examined the importance of the predicted coiled-coil domain , which usually functions in protein-protein binding . We created four GUS-tagged deletion variants of NUWA , each expressed with the NUWA promoter , for genetic complementation ( Fig 6A ) . We observed that NUWAΔ1stPPR , NUWAΔ1st-6thPPR and NUWAΔcoiled-coil failed to genetically complement nuwa/+ ( S1 Table ) . No homozygous nuwa mutant was identified out of 200 or 180 T2 resistance-pre-selected transformants of each variant; in contrast , a NUWA-GUS fusion protein generated in the same manner completely rescued nuwa . These results indicate that both the first PPR motif and the coiled-coil domain are crucial to NUWA protein functions during early embryogenesis and endosperm development . When we used NUWAΔ11thPPR to complement nuwa/+ , however , 10 homozygous nuwa were detected out of 200 T2 resistance-pre-selected transformants ( S1 Table ) . These homozygous nuwa that were complemented by NUWAΔ11thPPR-GUS had longer stems , undeveloped siliques ( Fig 6B ) and flowers featuring shortened stamens and shriveled anthers with undeveloped pollen ( Fig 6C and 6D ) . These results indicate that the 11th PPR motif is not critical to NUWA function during embryogenesis and endosperm development . However , NUWA may have functions beyond embryogenesis and endosperm development , as the GUS signals were detected in the junction region of filament anthers ( Fig 6E ) and specifically in stomatal guard cells on anther walls ( Fig 6E to 6G ) of proNUWA::NUWA-GFP-GUS transgenic plants . We next examined seed-setting rates and arrested seed developmental stages in the T2 transgenic plants in the nuwa/+ background . Seeds aborted as early as those of nuwa/+ were observed in siliques of these transgenic plants , whereas later-aborted seeds were found in siliques of transgenic plants expressing NUWAΔ11thPPR or NUWAΔcoiled-coil fragments ( Fig 6H ) . Deeper analysis revealed that rates of seed abortion in heterozygous transgenic plants expressing NUWAΔ11thPPR and NUWAΔcoiled-coil were 13 . 0% and 23 . 9% , respectively , which is significantly lower than the abortion rate of nuwa/+ in this experiment ( Fig 6I ) . This result confirms that the 11th PPR motif of NUWA is not as important as the other motifs and domains that were analyzed for their biological functions in the protein . These data also indicate that the coiled-coil domain is functionally less crucial than the first PPR motif during embryogenesis and endosperm development .
In this study , we have identified the expression pattern and essential role of a novel imprinted gene , NUWA , in Arabidopsis early seed development . First , NUWA is an essential gene that controls early embryogenesis and endosperm development in Arabidopsis , because loss of function of NUWA resulted in embryo and endosperm lethality . Second , NUWA is a maternal effect gene , because control of seed development is only through the maternal allele of NUWA , and wild-type pollen could not rescue the defective phenotypes . Third , NUWA is imprinted , as only the maternal allele-specific NUWA transcripts and proteins were identified before and early after fertilization , and the maternal allele of NUWA is transcribed de novo after fertilization . Fourth , NUWA is a new type of imprinted gene , because NUWA is localized in the mitochondria and is essential for the proper function of the mitochondria . Only the maternal allele-specific NUWA transcripts and encoded proteins were detectable during a period of several days after fertilization , and de novo transcription of NUWA started shortly after fertilization . As imprinted genes feature parent-of-origin specific allele biased expression patterns [2 , 7] , or , in particular , parental allele-specific de novo transcription after fertilization [16] , we concluded that NUWA is an maternally expressed imprinted gene . In addition , a TE is inserted in the coding sequence of NUWA , and most of the identified imprinted genes in Arabidopsis have TEs flanked by or harbored in coding sequences [29 , 49] . The mechanism of how NUWA imprinting is regulated , however , is not yet clear , and needs to be further investigated . NUWA is a new type of imprinted gene . It is not because that NUWA is not only expressed in the endosperm but also expressed in many other tissues ( Fig 3H to 3M ) . In fact , although some of the plant imprinted genes are only expressed in endosperm ( such as FWA ) , a lot of plant imprinted genes are expressed in multiple tissues . For instance , MEADA , one of the most famous imprinted genes in plant , was expressed in embryo and in many other vegetative tissues from both paternal and maternal allele [20 , 50] . In addition , genes expressed and imprinted in embryo were also discovered in maize [51] , rice [32] and Arabidopsis [16] . NUWA is special because its expression starts before fertilization , and the maternal control of it in early seed development is mediated by both maternally-deposited products and genomic imprinting . Maternal-specific NUWA products ( including both RNA and protein ) are detected in mature gametes before pollination , suggesting that maternal-deposited products of NUWA exist in embryo sac shortly after fertilization . Because NUWA encodes a PPR protein and mRNAs of PPR protein-encoding genes were reported to be rather unstable with a much shorter half-life ( i . e . , less than 6 hours ) than the average mRNA half-lives of other genes [52] , it is possible that the maternal allele of the zygotic NUWA is specifically activated soon after fertilization to make up the shortage of maternally-deposited NUWA products degraded during the maternal-to-zygotic transition [53–55] . Therefore , the products of NUWA detected in embryo sac early after fertilization are composed of the maternally-deposited NUWA products expressed in female gametophyte before fertilization and the maternal-allele specific NUWA products de novo synthesized after fertilization . That is , the parental effect of NUWA is controlled by parental-allele-specific expression not only before but also after fertilization , which is different from other reported imprinted genes in plants . It is not yet known whether the imprinting of NUWA occurs in embryo , or in endosperm or in both . The imprinting of NUWA occurs very early , i . e . , at about zygote stage , and NUWA is expressed in integument and seed coat . Unfortunately isolation and collection of enough amounts of Arabidopsis zygote and endosperm at zygote stage are extremely challenging , making it difficult to investigate the region in NUWA where imprinting occurs and the imprinting mechanism . Even so , we still think that NUWA is very likely to be imprinted in embryo . The expression of maternal allele of NUWA ( i . e . , the GUS signal ) could always be detected in embryos staged from zygote stage to later stages , whereas the expression of paternal allele of NUWA could not be detected in embryo until the 16-cell stage , suggesting that the NUWA products in early embryo are predominantly derived from its maternal allele . From the TEM result , we could observe severe defective phenotypes in mitochondria in nuwa embryos , suggesting an important function of NUWA in mitochondria and in early embryogenesis . Meanwhile , NUWA belongs to PPR family , proteins from which all have very short half-lives [52] . The de novo expression of the maternal allele of NUWA is likely to be present in the embryo to make up the shortage of maternally-deposited NUWA products degraded during the maternal-to-zygotic transition . Therefore it is very possible that NUWA is imprinted in embryo . On the other hand , NUWA is also very possible to be imprinted in endosperm . The products of the paternal allele of NUWA were never detected in endosperm . It is likely that the products of NUWA in early endosperm are also predominantly derived from its maternal allele . Similarly , from the TEM results we could also see that NUWA has important function in early endosperm development . As NUWA protein could be detected in endosperm from zygote stage to 16-cell stage , the de novo expression of the maternal allele of NUWA is also likely to be present in the endosperm . Thus NUWA is also likely to be imprinted in endosperm . NUWA is a new type of imprinted gene , also because it functions in mitochondria . In plants , although many imprinted genes have been identified , functional imprinted genes are still rarely discovered [17] . NUWA is a functional imprinted gene identified in Arabidopsis which controls early embryo and endosperm development . In addition , NUWA is the first discovered imprinted gene in plants whose protein product has an essential function in mitochondria . In mammals , imprinted genes that function in mitochondria were also identified , loss of function of which lead to severe diseases [56–58] . The discovery of NUWA indicates that , like the imprinted genes in mammals , imprinted genes in plants are also involved in many aspects of subcellular biological processes during early embryo development . The specific mitochondria RNAs that are regulated by NUWA remain to be characterized . As we did not find reduced expression level and defected splicing in mitochondrial mRNA of the mutant ( S8 Fig ) , it is possible that NUWA might be involved in regulating translation of mitochondrial mRNA , possibly through binding to mitochondrial mRNA and facilitating the release of a hairpin structure without changing RNA sequence and RNA copy number [59 , 60] . Another possibility is that NUWA may participate in the post-transcriptional processing of tRNAs or rRNAs . However , it is extremely challenging to detect the structural change of mitochondrial mRNAs , and the sequence and expression level changes of mitochondrial tRNAs and rRNAs , especially from the materials of early embryo , endosperm or ovules of Arabidopsis . New technology advances would be required to investigate the specific molecular function of NUWA in mitochondria . The mitochondrion is an important organelle for energy and metabolism in eukaryotic cells . Coordination between mitochondrial and nuclear genomes is important for the efficiency of mitochondrial function , because mitochondrial proteins encoded by both genomic and mitochondrial DNA varies significantly in different cells [61–63] . In most species of higher eukaryotes , mitochondria are maternally inherited after fertilization . In Arabidopsis , mitochondrial DNA in sperm cells is almost completely degraded before fertilization [63] . Moreover , very few mitochondria from Arabidopsis sperm cells are able to enter the embryo sac during double fertilization , and all are degraded shortly after gamete fusion , as in animals [64 , 65] . Therefore , all the mitochondria and mitochondrial DNA in the fertilized Arabidopsis embryo sac are maternally inherited . After fertilization , a series of remarkable changes take place in newly formed zygotes and endosperm , including recombination of the two differentiated nuclear genomes , remodeling of parental chromatin , and fusion and coordination of parental nuclear and cytoplasmic components [54 , 55 , 66] . Consequently , regulating maternally inherited essential mitochondria by nuclear-encoded maternal allele-specific NUWA products could possibly improve the implementation efficacy of the NUWA function and increase the efficiency of mitochondria-nuclear interaction in the complex post-fertilization environment , which might be the advantage for having an essential gene NUWA to be imprinted and selected during evolution . Therefore , the imprinting of NUWA would be an adaptation and enhancement to the maternal control of mitochondrial development . The discovery of NUWA reveals a unique mechanism of maternal control of the mitochondria and adds an extra layer of complexity to the regulation of nucleus-organelle coordination during early plant development .
Arabidopsis thaliana ecotype Columbia-0 ( Col-0 ) was used as wild type plant . All the transgenic lines used in this study were in the Columbia ecotype . Wild type Arabidopsis thaliana Landsberg ereta ( Ler-0 ) was only used in the allele specific RT-PCR . The nuwa mutant was obtained from the T-DNA insertion mutant library ( in Col background ) of our lab [34] . Plants were grown under long-day conditions ( 16 hr light/8 hr dark ) at 22°C . For crosses , maternal partners were emasculated , and pollinated 2 or 3 days after emasculation . The flanking sequence of the T-DNA insertion site in nuwa was determined by TAIL-PCR with the specific and arbitrary degenerate primers described previously [34] . For genotypic analysis of nuwa , primers nuwa_CS_F and nuwa_CS_R were used to amplify the wild-type allele of NUWA . Primers nuwa_CS_R and DL3 were used to amplify the insertion allele of nuwa and described in the S2 Table . Total RNA of most tissues was extracted from liquid nitrogen frozen tissues using TRIzol ( Invitrogen ) according to manufacturer’s instructions . Total RNA extraction of seeds and stem was performed with phenol extraction buffer ( 50% Tris saturated phenol , 0 . 5% SDS , 50 mM LiCl , 50 mM Tris-HCl at pH8 . 0 , 50 mM EDTA at pH8 . 0 ) at room temperature . Residual DNA was removed using the RNase-free DNase ( TaKaRa ) , and cDNA was synthesized using the M-MLV kit ( Invitrogen/Fermentas ) according to the manufacturer’s instructions . In real-time qPCR , diluted cDNA was used as a template , and three biological repeats was performed using SYBR Green real-time PCR Master Mix ( TOYOBO ) as described previously [67] on an ABI 7500 Real-Time PCR System . The relative expression level of each gene was calculated with the cycle threshold ( CT ) 2-ΔΔCT method [68] . Gene expression values were standardized to TUB2 ( AT5G62690 ) or PP2AA3 ( AT1G13320 ) . DIC observation of cleared ovules and bright field observation of GUS stained ovules were performed using Olympus BX51 microscope . Images of other GUS stained samples were taken using a Leica M205C FA stereoscope . Subcellular localization observation was performed using a confocal laser scanning microscope ( Leica TCS SPE confocal microscope ) . To determine the phenotype of endosperm nuclei , confocal observation of ovules was modified from previously described [69] . Artificial pollinated selfed 26 HAP siliques were harvested and fixed in 4% glutaraldehyde ( in 12 . 5 mM cacodylate , pH 6 . 9 ) , and a vacuum was applied for the initial 20 min , after which they were in fixative overnight at room temperature . After fixation , samples were dehydrated through a conventional ethanol series with 20 min per step , then cleared in 2:1 ( v/v ) benzyl benzoate:benzyl alcohol for a minimum of 1 hr . Siliques were dissected , mounted with immersion oil , and observed with a Leica TCS SPE confocal microscope . For transmission electron microscopy , protocol for sample preparation was modified from previously described [70] . Developing seeds at specific stages were collected , vacuumed and fixed in 5% glutaraldehyde and 3% paraformaldehyde in 0 . 1 M sodium cacodylatetrihydrate buffer ( pH7 . 2 ) at room temperature overnight and then at 4°C for 24 hrs . Ovules were then washed four times with 0 . 1 M sodium cacodylatetrihydrate buffer ( more than 30 min for each time ) and fixed in 2% OSO4 at room temperature for 4 hrs and then at 4°C over night , followed by dehydration in a graded ethanol series . Seeds were embedded in a complete resin mixture ( Spi-chem Spurr ) and incubated in 65°C for 16 hrs . Samples were sectioned using a Leica EM UC6 ultramicrotome and stained with 2% uranyl acetate and lead stain solutions . Sections were examined with a FEI Tecnai G2 20 Twin electron microscope at 200 kV . For genetic complementation , we used GW_NUWA_COMP_16_F and GW_NUWA_COMP_R primers to amplify the NUWA genomic DNA and generated the pNUWA::NUWA genomic DNA constructs through BP clonase and subsequent LR clonase ( Invitrogen ) . The amplified fragments were cloned into a reconstructed vector pK7FWG0 , which was made from the GATEWAY-compatible destination vector pB7RWG2 ( Department of Plant Systems Biology , VIB-Ghent University , Ghent , Belgium ) by digesting the 35S promoter via Spe I and Sac I sites . To generate the pNUWA::NUWA-EGFP-GUS constructs , the NUWA genomic DNA was amplified using TOPO_NUWA_GUS_F and TOPO_NUWA_pg+GUS_R primers . This genomic sequence was cloned into pBGWFS7 vector ( Department of Plant Systems Biology , VIB-Ghent University , Ghent , Belgium ) through pENTR/D-TOPO kit ( Invitrogen ) and LR reactions . In the genetic complementation assay with truncated NUWA , five groups of sequences were amplified to generate the pNUWA::NUWAΔmTP , pNUWA::NUWA Δ1st-PPR , pNUWA::NUWA Δ1stto6th-PPR , pNUWA::NUWA Δ11th-PPR and pNUWA::NUWA Δcoil constructs . As NUWA has no intron , genomic DNA was used as templates in the amplifications . P4P1R_NUWAΔX_F and P4P1R_NUWAΔX_R ( X represents one of mTP , 1st-PPR , 1stto6th-PPR , 11th-PPR and coil ) primers were used to amplify the sequence of the promoter . This sequence was then cloned into pDONRP4-P1r by BP reaction to generate pEN-L4-NUWAΔX-R1 . Meanwhile , TOPO_NUWAΔX_F and TOPO_NUWAΔX_R primers were used to amplify the sequence from the base next to the coding region of the mTP to the base right before the terminator . This sequence was then cloned into pENTR/D-TOPO by TOPO reaction to generate pENTR-NUWAΔX . GUS and EGFP coding sequence were amplified by GUS-F_attB2r/ GUS-R_attB3 and EGFP-F_attB2r/ EGFP-R_attB3 primers , and cloned in to pDONRP2R-P3 respectively to generate pEN-R2-GUS-L3 and pEN-R2-EGFP-L3 . The pNUWA::NUWAΔX constructs were generated by LR reaction of four plasmids , including pK7m34GW , pEN-L4-NUWAΔX-R1 , pENTR-NUWAΔX , and pEN-R2-EGFP-L3 . Constructs were transformed into Agrobacterium tumefaciens GV3101 , using a freeze-thaw procedure . Arabidopsis transformation and transgenic plant screening were conducted as reported [71] . The histochemical GUS staining was modified from described previously [72] . After per-fixation by pure acetone at -20°C for 1 hr , tissues immerged in a staining solution containing 0 . 5 mg ml-1 X-gluc ( Sigma ) in 100 mM Na2HPO4/NaH2PO4 ( pH7 ) , 2 mM K3Fe ( CN ) 6 , 2 mM K4Fe ( CN ) 6 , 10 mM EDTA and 0 . 1% Triton X-100 , modified from [35] . Samples were vacuumed for 10 min and then put in a 37°C incubator overnight . The staining buffer was removed , and the samples were fixed in FAA for 1 or 2 hrs before cleared and embedded . For fuchsin basic staining , resin sections on the glass slides was covered by drops of 1% fuchsin basic solution and incubated at 60°C for 6 min . Then wash away the dye by flowing sterile water . Slices were sealed with 50% glycerine before microscopy . MitoTracker Orange ( Invitrogen ) staining of the seedlings was performed according to manufacturer’s instructions , as described before [73] . The GUS stained samples were cleared using 70% ethanol before microscopy . For whole-mount preparations , ovules at different development stages were mounted in diluted Hoyer’s solution , which was introduced on http://www . seedgenes . org/Tutorial . html [74] , and cleared for 5 to 30 min before microscopy analysis . Isolated early embryos were incubated in 12% ( w/v ) mannitol with 10 μg/mL JC-1 ( Molecular Probes , Invitrogen ) for 30 min at room temperature , and then washed with 12% ( w/v ) mannitol . Images were collected using the confocal microscope mentioned above . JC-1 aggregates were detected with red fluorescence ( excitation , 532 nm; emission , 575–590 nm ) , and JC-1 monomers were detected with green fluorescence ( excitation , 488 nm; emission , 515–545 ) . The ratio of red to green fluorescence of JC-1 images was calculated by Imaris 7 . 5 . 0 software ( Bitplane ) . R software was used for the box-plot . cDNA from ovules ( or isolated embryo sacs ) at different development stages from self and crossed Col-0 and Ler was used for allele-specific RT-PCR ( primers and PCR parameters in S2 Table ) . PCR products were sequenced by the Sanger Chain Termination Method . The isolation of embryo sacs at different stages and the extraction of RNA from the isolated embryo sacs were performed according to modified versions of protocols described previously [75] . Embryo sacs were isolated by brief manual dissection combined with enzymatic maceration . Enzymatic solution is composed of 1% cellulose ( Yakult Honsha Co . Ltd , Tokyo , Japan ) and 0 . 8% Macrozyme R-10 ( Yakult Honsha ) . Before manual dissection , ovules at different development stages were incubated in the enzyme solution for 0 . 5 h . The Dynabeads mRNA DIRECT Micro Kit ( Invitrogen ) was used for mRNA isolation . cDNA was synthesized using the SuperScript III kit ( Invitrogen ) according to the manufacturer’s instructions . For each experiment ( or each time point ) , three independent biological replicates were performed . For each independent biological replicate , 11–16 embryo sacs were isolated for RNA extraction .
|
Imprinted genes are only found in mammals and flowering plants , and they express with parent-of-origin-specific patterns . Unlike in mammals , only a few imprinted genes was identified and functionally characterized in plants , and almost all of them encode nuclear proteins . Here , we identified NUWA as a new type of imprinted gene in Arabidopsis . NUWA has maternal-specific expression and controls the early seed development in Arabidopsis , with its encoded protein localizing to and functioning in the maternally inherited mitochondria . The discovery of NUWA reveals a unique mechanism of maternal control of mitochondrial function and adds an extra layer of complexity to nucleus-organelle coordination . It further indicates that , similar to mammals , imprinted genes in plants are also involved in various subcellular biological processes .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"plant",
"anatomy",
"plant",
"growth",
"and",
"development",
"plant",
"embryo",
"anatomy",
"ovules",
"brassica",
"endosperm",
"developmental",
"biology",
"plant",
"science",
"model",
"organisms",
"experimental",
"organism",
"systems",
"mitochondria",
"bioenergetics",
"embryos",
"plants",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"arabidopsis",
"thaliana",
"embryology",
"embryo",
"sac",
"plant",
"embryogenesis",
"fertilization",
"plant",
"development",
"gene",
"expression",
"biochemistry",
"plant",
"and",
"algal",
"models",
"cell",
"biology",
"embryogenesis",
"genetics",
"biology",
"and",
"life",
"sciences",
"energy-producing",
"organelles",
"organisms",
"fruit",
"and",
"seed",
"anatomy"
] |
2017
|
A Novel Imprinted Gene NUWA Controls Mitochondrial Function in Early Seed Development in Arabidopsis
|
Somatic migration of Toxocara canis- and T . cati-larvae in humans may cause neurotoxocarosis ( NT ) when larvae accumulate and persist in the central nervous system ( CNS ) . Host- or parasite-induced immunoregulatory processes contribute to the pathogenesis; however , detailed data on involvement of bioactive lipid mediators , e . g . oxylipins or eico-/docosanoids , which are involved in the complex molecular signalling network during infection and inflammation , are lacking . To elucidate if T . canis- and T . cati-induced NT affects the homeostasis of oxylipins during the course of infection , a comprehensive lipidomic profiling in brains ( cerebra and cerebella ) of experimentally infected C57BL/6J mice was conducted at six different time points post infection ( pi ) by liquid-chromatography coupled to electrospray tandem mass spectrometry ( LC-ESI-MS/MS ) . Only minor changes were detected regarding pro-inflammatory prostaglandins ( cyclooxygenase pathway ) . In contrast , a significant increase of metabolites resulting from lipoxygenase pathways was observed for both infection groups and brain regions , implicating a predominantly anti-inflammatory driven immune response . This observation was supported by a significantly increased 13-hydroxyoctadecadienoic acid ( HODE ) /9-HODE ratio during the subacute phase of infection , indicating an anti-inflammatory response to neuroinfection . Except for the specialised pro-resolving mediator ( SPM ) neuroprotectin D1 ( NPD1 ) , which was detected in mice infected with both pathogens during the subacute phase of infection , no other SPMs were detected . The obtained results demonstrate the influence of Toxocara spp . on oxylipins as part of the immune response of the paratenic hosts . Furthermore , this study shows differences in the alteration of the oxylipin composition between T . canis- and T . cati-brain infection . Results contribute to a further understanding of the largely unknown pathogenesis and mechanisms of host-parasite interactions during NT .
Toxocara canis ( Werner , 1782 ) and T . cati ( Schrank , 1788 ) are globally distributed , intestinal helminth parasites with canids and felids as definitive hosts [1] . Humans and a wide range of other species can act as paratenic hosts after accidental ingestion of infective stages ( L3 ) of Toxocara spp . , resulting in persistence of the larvae in the body tissues [2 , 3] . The infection of paratenic hosts comprises different stages , starting with larvae entering the cardiovascular system and reaching the liver and lungs during the first week post infection . This acute stage of toxocarosis is called the hepato-pulmonary phase . The myotropic-neurotropic phase indicates the beginning of the chronic stage and is characterised by migration and accumulation of larvae throughout somatic tissues [3] . Humans may get infected due to inadequate hygiene , geophagia or via foodborne transmission [4] . Even though toxocarosis is one of the most frequent helminthoses in humans , the global importance of this zoonosis is probably underestimated [2 , 5] . Human toxocarosis may result in a variety of clinical symptoms . Depending on the somatic distribution of the larvae and the occurring symptoms , toxocarosis is currently classified into four syndromes: covert toxocarosis , visceral larva migrans ( VLM ) , ocular larva migrans ( OLM ) and neurotoxocarosis ( NT ) , whereby NT results from accumulation and persistence of Toxocara spp . -larvae in the CNS [2] . Neurotoxocarosis may lead to encephalitis , myelitis , cerebral vasculitis or optic neuritis [6 , 7] and affected patients may suffer from headache , fever , oversensitivity to light , weakness , confusion , tiredness and visual impairment [6–9] . While the localization of Toxocara-larvae in the human brain has not been systematically investigated , it has been demonstrated that larvae exhibit a species-specific tropism in the murine brain . T . canis-larvae are mainly located in the cerebrum , while T . cati-larvae prefer the cerebellum but mainly accumulate in muscle tissue [10] . Consequently , T . canis- and T . cati-induced NT differs in the severity of structural brain damage as well as the severity of neurological symptoms and behavioural alterations [10–12] . However , host- or parasite-induced immunoregulatory processes contributing to pathogenesis as well as molecular pathogenic mechanisms are only marginally identified yet . Bioactive regulatory lipids ( also called oxylipins ) , such as octadecanoids , eicosanoids and docosanoids , constitute an important class of molecules involved in the complex molecular signalling network during infection and inflammation . Regulatory lipids comprise a plethora of structurally and stereochemically different bioactive mediators derived from arachidonic acid ( ARA ) and related ω-6-polyunsaturated fatty acids ( PUFAs ) like dihomo-γ-linolenic acid ( DGLA ) and linoleic acid ( LA ) as well as ω-3-PUFAs such as α-linolenic acid ( ALA ) , eicosapentaenoic acid ( EPA ) and docosahexaenoic acid ( DHA ) . Regulatory lipids are generated from the oxidation of different PUFAs by three major enzymatic pathways ( an overview of these and selected regulatory lipids is given in Fig 1 ) : The cyclooxygenase pathway ( COX-1 and COX-2 ) results in different prostanoids like prostaglandins , prostacyclins and thromboxanes . Leukotrienes as well as several hydroxy fatty acids are derived from the lipoxygenase pathway ( 5-LOX , 12-LOX and 15-LOX ) . The murine 15-LOX ( alox15 ) additionally acts as a 12-lipoxygenating enzyme , converting PUFAs to metabolites similar to those derived from 12-LOX [13] . Thus , hereinafter these metabolites are referred as 12/15-LOX-metabolites . The superfamily of cytochrome P450 ( CYP 450 ) monooxygenase enzymes catalyses the epoxidation of ARA to epoxyeicosatrienoic acids ( EpETrEs ) , which are hydrolysed to corresponding dihydroxy-derivatives ( DiHETrEs ) by soluble epoxide hydrolases ( sEH ) . In addition , CYP 450 enzymes catalyse the ω-hydroxylation of PUFAs , forming terminal ( ω and ω-n ) hydroxylated fatty acids [14 , 15] . Different regulatory lipids have been detected in cerebral tissues , playing important roles in a variety of physiological processes , such as the maintenance of homeostasis and neural functions including spatial learning and synaptic plasticity [16–19] . Under disease conditions , several oxylipins like COX-derived prostanoids and 5-LOX-derived leukotrienes are involved in inflammatory processes including fever , sensitivity to pain , oxidative stress , and neurodegeneration [20] . In contrast , several metabolites formed via the 12/15-LOX pathway , have been described to exhibit anti-inflammatory activities , e . g . by co-activating peroxisomal proliferator activating-receptors ( PPAR ) , regulating cytokine generation and modulating expression of inflammation related genes [21 , 22] . Furthermore , the 12/15-LOX-derived 13-HODE is an agonist for PPAR-γ and exhibit anti-inflammatory properties [23 , 24] . In contrast , 9-HODE is mainly generated by non-enzymatic reactions and activates the G protein-coupled receptor G2A , which mediates intracellular calcium mobilization and JNK activation , promoting inflammatory processes [25 , 26] . Both metabolites derive from linoleic acid and have been suggested as markers for lipid peroxidation in various chronic diseases [27] . Therefore , Tam et al . [28] proposed the ratio of 13-HODE to 9-HODE as useful biomarker to indicate the immunological status of an active infection . The involvement of regulatory lipids in inflammatory processes has been examined in numerous studies . Most of these studies focused primarily on selected oxylipins , and only a few studies have comprehensively examined quantitative changes during the course of bacterial [29 , 30] and viral [28] infections . While these studies were conducted with tibiotarsal tissues [29] , blood and exudates [30] , and nasopharyngeal lavages [28] as sample material , nothing is known about the dynamic lipidomic profile in the brain during cerebral infections . Furthermore , information about a comprehensive description of these processes during parasitic infections is lacking . Therefore , this study aimed to characterise for the first time alterations in the brain pattern of bioactive regulatory lipids during acute , subacute and chronic NT in T . canis- and T . cati-infected mice as a model for paratenic hosts [3] .
Animal experiments were performed in accordance with the German Animal Welfare act in addition to national and international guidelines for animal welfare . Experiments were permitted by the ethics commission of the Institutional Animal Care and Use Committee ( IACUC ) of the German Lower Saxony State Office for Consumer Protection and Food Safety ( Niedersaechsisches Landesamt für Verbraucherschutz und Lebensmittelsicherheit ) under reference numbers 33 . 9-42502-05-01A038 ( experimental infection of dogs and cats ) , 33 . 12-42502-04-14/1520 , 33 . 12-42502-04-15/1869 and 33 . 14-42502-04-12/0790 ( experimental infection of mice ) . Eggs of field isolates of Toxocara canis ( field isolate HannoverTcanis2008 ) and T . cati ( field isolate HannoverTcati2010 ) , maintained at the Institute for Parasitology , University of Veterinary Medicine Hannover , were obtained from faeces of experimentally infected dogs and cats , respectively , by a combined sedimentation/flotation technique . Eggs were cultured in tap water for about 4 weeks in a controlled temperature chamber at 25±1 °C with oxygenation two times per week to allow development of third-stage larvae . Infective eggs were subsequently stored in tap water at 4 °C until use . For the oxylipin analysis , 4-week-old female C57BL/6JRccHsd mice were purchased from Harlan Laboratories ( The Netherlands ) and were allowed to acclimatize for 14 days before the start of the experiment , while for microarray analysis , 5-week-old female C57BL/6JRccHsd were purchased and an acclimatisation time of one week was provided . Mice were housed in Makrolon cages in a 12/12 hours dark/light cycle receiving standard rodent diet ( Altromin 1324 , Germany ) and water ad libitum . With regard to unsaturated fatty acids , the standard rodent diet contained 2 , 210 mg/kg α-linolenic acid and 16 , 152 mg/kg linoleic acid . At the age of 6 weeks , 45 animals each were infected orally with 2000 embryonated T . canis or T . cati eggs , respectively , administered at once in a volume of 0 . 5 ml tap water , whereas 45 control mice received the same volume of the vehicle ( tap water ) only . At each time point of investigation in the acute phase ( day 7 post infectionem [pi] ) , the subacute phase ( days 14 and 28 pi ) and the chronic phase ( days 42 , 70 and 98 pi ) , five mice were sacrificed by cervical dislocation for the oxylipin analysis . Additionally , as of day 14 pi , three mice of each study group were euthanized for microarray analyses at each time point . For the oxylipin analyses , brains were removed , subdivided into cerebrum and cerebellum , and immediately snap frozen in liquid nitrogen . During further processing , specimens were homogenised in liquid nitrogen using a mortar and pestle , and 50±5 mg of homogenised tissue were weighed and stored individually at -150 °C until oxylipin extraction . For the microarray analysis , brains were removed and subdivided into left and right hemispheres as well as cerebrum and cerebellum . Right cerebrum and cerebellum hemispheres were stored individually in RNAlater RNA stabilization reagent ( Qiagen , Hilden , Germany ) at 4 °C overnight and afterwards at -80 °C until RNA isolation [31] . Extraction and analysis of oxylipins in brain tissue was carried out as described with modifications [32 , 33] . Samples were thawed on ice and 300 μl acidified methanol ( 0 . 2% formic acid [Fisher Scientific , Germany] in LC-MS grade MeOH [Fisher Scientific , Germany] ) , 10 μl antioxidant solution ( 0 . 2 mg/mL EDTA [Sigma Aldrich; Germany] , 0 . 2 mg/mL butylated hydroxytoluene [Sigma Aldrich; Germany] , 100 μM indomethacin [Sigma Aldrich; Germany] , 100 μM TUPS [34] ) in MeOH/H2O ( 50/50 , v/v ) [35] ) as well as 10 μl of internal standards ( each 100 nM 2H4-6-keto-PGF1α , 2H4-PGE2 , 2H4-PGD2 , 2H4-TxB2 , 2H4- LTB4 , 2H4-9-HODE , 2H8-5-HETE , 2H8-12-HETE , 2H6-20-HETE , 2H11-14 , 15-DiHETrE , 2H11-14 ( 15 ) -EpETrE , 2H4-9 ( 10 ) -EpOME , 2H4-9 ( 10 ) -DiHOME , 2H4-15-F2t-IsoP , 2H11-5 ( R , S ) -5-F2t-IsoP , Cayman Chemicals ( local distributor: Biomol , Germany ) [32] were added . Subsequently , the samples were homogenised again using two 3 mm metal beads in a ball mill ( Retsch , Germany ) for 8 min at 15 Hz , followed by centrifugation at 20 , 000 x g for 10 min at 4 °C . The supernatant was diluted with 2700 μl of 1 M sodium acetate ( pH 6 . 0; Carl Roth , Germany ) . Solid phase extraction was carried out on cartridges with an unpolar ( C8 ) /anion exchange mixed mode material ( Bond Elut Certify II , 200 mg; Agilent , Germany ) , preconditioned with one column volume of MeOH and one column volume of 0 . 1 M sodium acetate with 5% MeOH ( pH 6 . 0 ) . After sample loading , the cartridge was washed with one column volume of water and one column volume of MeOH/H2O ( 50/50 , v/v ) . Cartridges were dried for 20 min by low vacuum ( ∼200 mbar ) . Oxylipins were eluted with n-hexane/ethyl acetate ( 25/75 , v/v ) ( n-hexane: HPLC grade [Carl Roth Germany]; ethyl acetate [Sigma Aldrich; Germany] ) with 1% acetic acid ( Sigma Aldrich; Germany ) in glass tubes containing 6 μl of 30% glycerol ( Sigma Aldrich; Germany ) in MeOH . The eluate was evaporated in a vacuum centrifuge ( 1 mbar , 30 °C , 40–60 min; Christ , Germany ) until only the glycerol plug was left . Dried residues were immediately frozen at -80 °C for at least 30 min and reconstituted in 50 μl of MeOH containing a second internal standard allowing the evaluation of extraction efficacy afterwards . Samples were centrifuged ( 20 , 000 x g , 10 min , 4 °C ) . Oxylipins were quantified by liquid chromatography-mass spectrometry ( LC–MS/MS ) following negative electrospray ionization in scheduled SRM mode on a QTRAP6500 mass spectrometer ( Sciex , Germany ) injecting 5 μl as described by Rund et al . [32] . Authentic standard substances of oxylipins were purchased from Cayman Chemicals ( local distributor: Biomol , Hamburg , Germany ) . As the standard for NPD1 is not commercially available it was synthesized as follows: The NPD1-methyl ester was synthesized as described for its C10-epimer [36] replacing the ( S ) -1 , 2 , 4-butanetriol by its ( R ) -enantiomer as starting material for the introduction of the E , E-iododiene . Methyl ester-NPD1 was than hydrolyzed with 1 M LiOH in MeOH/H2O ( 50/50 , v/v ) followed by acidification with McIlvains buffer ( pH 5 ) producing NPD1 as a colorless oil in 97% yield . Peak integration and determination of oxylipin concentration was conducted using Multiquant ( Sciex , Germany ) . RNA isolation and whole genome microarray analysis was conducted as described by Janecek et al . [31] . Briefly , the total RNA content from three cerebra and cerebella from each group at time points 14 , 28 , 42 , 70 and 90 pi was isolated using the RNeasy Lipid Tissue Mini kit ( Qiagen , Germany ) . After further processing , quality control and Cy3-labelling of isolated RNA , labelled cRNA was hybridised to Agilent’s 4x44k Mouse V2 ( Design ID:026655 ) for 17 h at 65 °C and scanned as described by Pommerenke et al . [37] . Obtained data for ptgs1 ( COX-1; Probe A_51_P279100 ) , ptgs2 ( COX-2; Probe A_51_P254855 ) , aloxe3 ( A_55_P2023523 ) , alox5 ( A_51_P247249 ) , alox5ap ( A_51_P235687 ) , alox8 ( A_55_P2029957 ) , alox12 ( A_51_P520306 ) , alox12b ( A_55_P2121682 ) , alox12e ( A_51_P471659 ) and alox15 ( A_51_P252565 ) were statistically analysed as described below . Normality of distribution of all sample sets was analysed by the Kolmogorov-Smirnov test . For normally distributed variables the One-way ANOVA or for skewed distributions the Kruskal-Wallis test was used to reveal statistically significant differences between the infection and control group at each time point . To account for multiple comparisons , false discovery rate adjustment of P-values was carried out in R ( version 3 . 1 . 2; [38] ) and a Q-value of 0 . 1 was considered as statistically significant . For Q-values below 0 . 1 , the following post-hoc tests were carried out: unpaired t-test for normally distributed datasets or Mann-Whitney U test ( MWU ) for skewed distributions , whereby P≤0 . 05 was considered as statistically significant . If a metabolite could not be detected in one of the study groups , the lower limit of quantification was used for statistics . Statistical analyses were conducted with GraphPad Prism™ software ( version 6 . 03; GraphPad Software , California , USA ) . Due to normally distributed datasets for the ratio of 13-HODE to 9-HODE as well as the transcription rates of oxylipin-related genes , an unpaired t-test was used to reveal statistical differences between infected and uninfected groups ( GraphPad Prism™ software [version 6 . 03; GraphPad Software , California , USA] ) . Ratios of lipid mediators and transcriptional levels of mentioned genes between the uninfected control and infection groups over the course of infection were calculated by dividing each individual value of the T . canis- , T . cati- and uninfected control group by the mean value of the uninfected control group at the respective point in time . If a metabolite was not detected in the corresponding uninfected control group , the lower limit of quantification was used to calculate the fold change in infected mice . The fold changes were log2 transformed and the means of the log2 transformed fold-changes were presented as heatmaps to identify relative changes . Heat maps were visualised with MeV [39] ( Version 4 . 9 . 0 , TM4 Software suit [http://mev . tm4 . org] ) .
Clinical assessment of mice as well as data on body weight , whole brain weight , cerebrum and cerebellum weight as well as brain to body mass ratio have been published previously by Waindok and Strube [40] , using the same mice to investigate changes in brain cytokine and chemokine patterns during neurotoxocarosis . In short , infected mice showed varying degrees of neurobehavioural alterations as described by Janecek et al . [12] , with T . canis-infected mice developing symptoms like ataxia to paresis and paraplegia or incoordination earlier and more severe than T . cati-infected mice . The brain/body mass ratio in comparison to the uninfected control group was significantly lower at day 14 pi in T . canis- and T . cati-infected mice ( P = 0 . 0038 and P = 0 . 0024 , respectively ) and at day 28 pi in T . canis-infected mice ( P = 0 . 0353 ) . Similarly , the cerebrum/body mass-ratio was significantly lower at day 14 pi in T . canis- and T . cati-infected mice ( P = 0 . 0005 and P = 0 . 0003 , respectively ) and at day 28 pi in T . canis-infected mice ( P = 0 . 0396 ) , but increased significantly at day 70 pi in the latter group ( P = 0 . 0079 ) . Regarding the cerebellum/body mass-ratio , statistically significant differences were not detectable between the infected and uninfected groups . A total of 73 different oxylipins were successfully detected and quantified in brains of Toxocara spp . -infected mice and uninfected controls . To assess the composition of bioactive lipid metabolites over the course of infection , analysed metabolites of different PUFAs were summarised by their major formation routes , i . e . COX , LOXs , and CYP 450 as well as non-enzymatic autoxidation . The proportion of the respective pathways contributing to the analysed oxylipins in the brains of Toxocara spp . -infected and uninfected mice is illustrated in Fig 2 . Concentrations and P-values regarding the comparison to uninfected control mice are provided in S1 Table . Absolute levels of COX-derived metabolites in the cerebra and cerebella did not differ significantly between infected and uninfected control mice with the exception of T . cati-infected cerebella at day 42 pi . Infection with T . canis and T . cati led to similar alterations of CYP 450-derived metabolites in the cerebra , namely a significant increase at days 14 , 42 and 70 pi . In the cerebella , CYP 450-derived metabolites were significantly increased at days 14 and 42 pi in both infection groups . The total amount of 5-LOX-derived oxylipins was significantly higher in the cerebra of T . canis-infected mice at days 28 and 42 pi and of T . cati-infected mice at day 42 pi . In the cerebella , 5-LOX-metabolites were significantly increased at day 7 pi for T . canis- and at day 14 pi for T . cati-infected mice , as well as in both infection groups at day 42 pi . Metabolites derived by 8-LOX were significantly elevated in the cerebra of both infection groups at days 14 , 28 and 42 pi . Regarding the cerebella , a significant increase was observed at days 14 , 28 and 98 pi in T . canis-infected mice and at days 28 and 42 pi in T . cati-infected mice . In addition , levels of 12/15-LOX-derived oxylipins were significantly elevated at days 14 , 28 and 42 pi in the cerebra of both infection groups . Significantly increased levels of 12/15-LOX metabolites were also observed in the cerebella of both infection groups at days 14 and 42 pi . The levels of non-enzymatically derived oxylipins did not differ significantly between the uninfected control group and the two infection groups in both brain parts . While Fig 2 illustrates relative patterns of detected lipid mediators based on their major metabolic formation pathways during infection , Fig 3 displays alterations of individual bioactive lipid mediators and their fold change in infected compared to uninfected control mice . Oxylipin concentrations and P-values regarding the comparison to the control mice are given in S2 Table ( cerebrum ) and S3 Table ( cerebellum ) . Metabolites are classified by their major biosynthetic pathways , however metabolite formation through other enzymes or non-enzymatic peroxidation cannot be excluded . The ratio of 13-HODE to 9-HODE was shifted towards 13-HODE in the cerebra and cerebella of T . canis- as well as T . cati-infected mice during the course of infection ( Fig 5 ) . A significant increase of the 13-HODE/9-HODE ratio was already detected in the cerebra of T . canis-infected mice at days 7 and 14 pi ( P = 0 . 0171 and P = 0 . 0001 ) , reaching a maximum from days 28 and 42 pi ( P≤0 . 0001 and P = 0 . 0011 ) , while in T . cati-infected cerebra , the ratio peaked already at day 14 pi and remained significantly increased at days 28 and 42 pi ( P = 0 . 0098 , P = 0 . 0049 and P = 0 . 0461 , respectively ) . In both infection groups , the 13-HODE/9-HODE ratio declined to homeostatic conditions in the later phase of infection at days 70 and 98 pi . Although the 13-HODE/9-HODE ratio in T . canis- and T . cati-infected cerebella showed a similar development as in the cerebrum , statistically significant alterations were only detected at days 7 and 28 pi ( P = 0 . 0006 and P≤0 . 0001 ) in T . canis- and at days 14 and 28 pi ( P = 0 . 0139 and P = 0 . 0067 ) in T . cati-infected mice . Transcriptional alterations of different ptgs ( encoding for COX enzymes ) and alox ( encoding for LOX enzymes ) -genes are shown as fold changes in Fig 6 . Transcription of ptgs1 was significantly increased in cerebra and cerebella of T . canis-infected mice during the whole study period ( P≤0 . 0418 ) . In T . cati-infected mice , ptgs1-transcription was significantly upregulated at day 28 pi ( P = 0 . 0406 ) in the cerebra , and at days 28 , 70 and 98 pi ( P = 0 . 0056 , P = 0 . 0453 and P = 0 . 0418 ) in the cerebella . By contrast , the transcription rate of ptgs2 was significantly downregulated in cerebra of T . canis-infected animals at day 28 pi ( P = 0 . 026 ) , cerebra of T . cati-infected mice were not affected . In the cerebella , a statistically significant increase was detected at day 98 pi in both infection groups ( T . canis: P = 0 . 0039 , T . cati: P = 0 . 0402 ) . Furthermore , the transcription rate of alox5 and alox5ap ( encoding for the 5-LOX activating protein FLAP ) was significantly elevated as of day 14 pi ( with the exception of alox5 in the cerebra at day 14 pi ) in both brain parts of T . canis-infected mice ( P≤0 . 0348 ) . In brains of T . cati-infected mice , alox5 was significantly elevated at day 28 pi ( P = 0 . 0266 ) , while no significant alterations of alox5ap-transcription were detected . The transcription rate of different alox12-isoforms was mostly unaffected by NT , with a downregulation in both brain areas of T . canis-infected mice at days 14 ( cerebra P = 0 . 0086; cerebella P = 0 . 0229 ) and 42 pi ( cerebra P = 0 . 0061; cerebella P = 0 . 0398 ) . In T . cati-infected mice , a significant downregulation occurred at days 14 and 28 pi ( P = 0 . 0336 and P = 0 . 0229 ) in the cerebra and day 42 pi ( P = 0 . 0268 ) in the cerebella . In contrast , the transcription of alox12e was elevated in the cerebella of T . canis-infected mice at days 28 , 42 and 98 pi ( P = 0 . 0300 , P = 0 . 0021 and P = 0 . 0300 ) , and at day 42 pi ( P = 0 . 0017 ) of T . cati-infected mice . The transcription of alox15 was significantly upregulated at days 28 pi and 42 pi in cerebra of T . canis ( P = 0 . 0037; P = 0 . 0037 ) as well as T . cati-infected mice ( P = 0 . 0239; P = 0 . 0464 ) . In the cerebella , T . canis-infection resulted in a significant upregulation of alox15 at days 14 , 28 and 98 pi ( P = 0 . 0191 , P = 0 . 0252 and P = 0 . 0006 ) , while T . cati-infected mice only showed an upregulation on days 28 and 98 pi ( P = 0 . 0068 , P = 0 . 0251 ) .
The central nervous system exhibits inflammatory reactions in response to injury , infection or disease , comprising the activation of brain microglia , the rapid release of inflammatory mediators and invasion of immune cells , among others [41 , 42] . Nevertheless , neuroinvasive larvae of Toxocara spp . are able to accumulate and persist in cerebral tissues [43] and even though the infection is characterised by neuroinflammatory hallmarks like hemorrhagic lesions , myelinophages , spheroids and activated microglia [10 , 44 , 45] , larvae are not trapped by inflammatory reactions in cerebral tissues [44 , 46] . Knowledge regarding the cerebral immune response during NT is scarce . In the present study , an overall shift to an anti-inflammatory oxylipin pattern during NT was found . In general , this trend was observed for T . canis- and T . cati-infected mice , even though minor differences , especially for pro-inflammatory regulatory lipids , between T . canis- and T . cati-induced NT were noted . COX-derived prostaglandins are potent immunomodulatory mediators , triggering the expression of inflammatory enzymes [47–50] , chemokines and cytokines [51] . It is commonly believed that ptgs1 ( encoding for the enzyme COX-1 ) is expressed constitutively under homeostatic conditions , whereas the expression of ptgs2 ( encoding for the enzyme COX-2 ) is induced in response to inflammatory stimuli [20] . However , recent data suggest that ptgs1 is a major player in neuroinflammatory processes , while ptgs2 activity mediates neurotoxicity or neuroprotection [52] . This hypothesis is supported by the transcriptomic analysis in the present study , which revealed significantly increased levels of ptgs1-transcription in both brain areas of T . canis-infected mice and at three time points in the cerebellum of T . cati-infected mice , whereas ptgs2-transcription remained largely unaffected . Nevertheless , elevated transcription rates of ptgs1 did not result in elevated concentrations of the corresponding oxylipins , which were only moderately altered . Similar observations have been made regarding cytokine secretion during NT . In a recent study , concentrations of pro-inflammatory cytokines were also not elevated during T . canis- and T . cati-induced NT [40] , although a transcriptional upregulation of pro-inflammatory IFN-γ , IL-6 and TNF-α has been shown in brains of T . canis-infected mice [46 , 53] . Metabolites derived by the 5-LOX enzyme ( encoded by alox5 ) are considered to have pro-inflammatory properties . The transcription of alox5ap leading to FLAP , which is necessary to initiate the activation of 5-LOX , was increased throughout the study period in cerebra and cerebella of T . canis- as well as in cerebella of T . cati-infected animals . Alox5 transcription was also elevated at these time points in T . canis- , but not in T . cati-infected animals . An important 5-LOX-derived regulatory lipid is LTB4 , which activates and recruits neutrophils during inflammatory processes [54 , 55] and in interaction with pro-inflammatory cytokines , LTB4 induces the activation of NF-κB , a key regulator of neuroinflammation [56–58] . During the course of T . canis- as well as T . cati-induced NT , LTB4 levels in the cerebellum were mainly unaffected , while significantly increased levels were detected in the cerebrum during the subacute phase and the beginning of the chronic phase of infection . Recruited neutrophils were , besides eosinophils , shown to be present in perivascular lymphocytic cuffs of T . canis- and T . cati-infected mice brains . However , neutrophil infiltration may play only a subordinate role in the pathogenesis of NT as the pathological picture is dominated by eosinophilic meningitis , microglia activation and neurodegeneration , the latter especially in T . canis-infected mice [45] . The murine alox15 gene encodes for an enzyme that besides 15-lipoxygenation further acts as a 12-lipoxygenating enzyme , converting PUFAs to products also formed by 12-LOX [13] thus it is also referred to as 12/15-LOX . In the present study , alox12-transcription only showed minor alterations in both infection groups , whereas alox15-transcription was significantly upregulated at several time points pi . The highly increased levels of 12/15-LOX-metabolites at days 14 , 28 and 42 pi are thus most likely due to elevated alox15-transcription rates . Most 12/15-LOX-metabolites exhibit anti-inflammatory properties . HETEs and HODEs are activators of the peroxisomal proliferator-activating receptor-γ ( PPARγ ) , which plays an important role in the regulation of cell development and homeostasis [59] . Under neuroinflammatory conditions , 12/15-LOX-metabolites mediate effects of anti-inflammatory IL-4 on NF-κB trans-activation in glial cells and protect oligodendrocyte progenitors [60] . In case of NT , these anti-inflammatory and neuroprotective mechanisms may facilitate the persistence of Toxocara spp . -larvae in the brain and the survival of the paratenic host . Multiple enzymatic conversions of PUFAs result in formation of specialised pro-resolving mediators ( SPMs ) , including specific lipoxins , resolvins and protectins [51 , 61 , 62] . Besides their anti-inflammatory properties , SPMs promote the resolution of inflammation by blocking neutrophil recruitment and mediating phagocytosis and lymphatic clearance of apoptotic neutrophils [51] . Arachidonic acid-derived lipoxins , with LxA4 and LxB4 as the most prominent metabolites , are involved in the regulation of leukocyte trafficking [61 , 63 , 64] . Interestingly , even though LxA4 is involved in pro-resolving processes , LxA4 was not detected in any of the study cohorts . In contrast , LxA4 seems to play an important role during cerebral toxoplasmosis , with high levels of lipoxins suppressing the activation of dendritic cells and the secretion of IL-12 [65] . Furthermore , anti-inflammatory and pro-resolving functions are also partially mediated by resolvins , which can be separated in the EPA-derived E-series ( RvEs ) and the DHA-derived D-series ( RvDs ) [66 , 67] . In the present study , neither RvEs nor RvD1 were present in detectable quantities in any of the study groups . However , the DHA-derived NPD1 , a major bioactive effector in both anti-inflammation and neuroprotection [68 , 69] was successfully detected in the brains of the Toxocara-infected groups , but not in the uninfected control group . In the T . canis- and T . cati-infected mice , NPD1 occurred primarily in the subacute phase of infection between day 14 and 42 pi . In the brain , NPD1 has far-ranging properties regarding neural cell survival , protection of cerebral tissues and inhibition of leukocyte infiltration as well as interleukin 1-β-induced NF-κB activation [68 , 70] . According to Ariel et al . [71] , NPD1 is produced by T-helper type 2-skewed peripheral blood mononuclear cells ( Th2 PBMCs ) in dependence of human ALOX-15 activity , but not in Th1 PBMCs . The “type 2” polarisation of CD4+ T-helper cells , as part of the adaptive immune response , is commonly induced by parasitic helminths [72 , 73] . Del Prete et al . [74] demonstrated a stable Th2-like cytokine secretion of human CD4+ T cells derived from the peripheral blood ( PB ) , stimulated with T . canis excretory/secretory ( TES ) antigen . Similarly to NPD1 , 5 , 15-DiHETE and 8 , 15-DiHETE were detected in Toxocara-infected , but not in control mice . Both metabolites are secreted in large quantities from eosinophils , and eosinophilic meningitis is one of the most prominent pathologic features of NT in mice as well as in humans [6 , 45] . The metabolites 9-HODE and 13-HODE as well as the ratio of both metabolites are proposed biomarkers for various chronic diseases and the immunological status of an active infection [27 , 28] . The low 13-/9-HODE ratio in the acute phase of influenza-infected C57BL/6 mice implied a pro-inflammatory state of the immune response , which was shifted to an anti-inflammatory , pro-resolution state at later time points of the infection , indicated by a higher 13-/9-HODE ratio [28 , 75] . This was also illustrated in the Lyme arthritis model of B . burgdorferi-infected C3H/HeJ and DBA/2J mice , with an elevated 13-/9-HODE ratio in the resistant mouse strain ( DBA ) , compared to the susceptible strain ( C3H ) [29 , 75] . In the present study , the 13-/9-HODE ratio in Toxocara-infected mice was clearly shifted towards an anti-inflammatory immune response in comparison to uninfected mice , particularly in the subacute phase of infection between days 14 and 42 pi . Thus , the ratio of 13-/9-HODE reflects the aforementioned trend of minor changes in the biosynthetic pathway of mostly pro-inflammatory prostaglandins ( COX-pathway ) in concert with a significant increase of potentially anti-inflammatory metabolites of the 12/15-LOX -pathway , observed for both infection groups and brain regions . In general , T . canis is regarded as the main causative agent of toxocarosis while the cat roundworm T . cati is probably underestimated as a zoonotic pathogen [76] . In mice , T . cati-induced NT is characterised by a less severe pathogenesis in terms of clinical symptoms , histopathological alterations and behavioural changes compared to T . canis-infections [10 , 11 , 76] . Furthermore , T . canis-larvae exhibit higher neuroaffinity than T . cati-larvae and mainly accumulate in the cerebrum , whereas T . cati prefers the cerebellum but mainly accumulates in muscle tissue [10] . This species-specific tropism is partly reflected by alterations of the oxylipin pattern , as significant elevations of pro-inflammatory COX- and 5-LOX-metabolites were detected in the cerebellum rather than the cerebrum of T . cati-infected mice . Nevertheless , in essence , alterations in the cerebral oxylipin patterns during T . canis- and T . cati-induced NT were comparable . It remains unknown if the shift towards an anti-inflammatory oxylipin pattern is an immunogenic reaction of the host in response to neuroinvasive larvae to prevent excessive tissue damages or a parasite-induced immunomodulatory effect to evade the host’s immune regulation , facilitating persistence in cerebral tissues . Many parasitic helminths are known to manipulate the hosts’ immune response by secreting immunomodulatory components evoking for example the induction of a modified Th2 response [77] and the reduction of pro-inflammatory cytokines [78] . In conclusion , the present study revealed only minor changes in pro-inflammatory metabolites of the COX-pathway in T . canis- and T . cati-infected cerebra and cerebella . Generally , it should be kept in mind that NT is characterised by focally distributed lesions whereas entire cerebra and cerebella were processed for the current analysis . Thus , healthy brain tissue was overrepresented compared to damaged tissue , which may have masked local effects . Nevertheless , a statistically significant increase of metabolites of different LOX-pathways was demonstrated for both infection groups . Neuroprotective NPD1 was detected in T . canis- and T . cati-infected mice , especially in the subacute and at the beginning of the chronic phase of infection , but not in uninfected mice . Other anti-inflammatory and pro-resolving SPMs were not detected ( LxA4 , RvD1 , RvEs ) . The ratio of 13-HODE to 9-HODE revealed a similar anti-inflammatory immune response in the cerebrum and cerebellum of each infection group . Even though T . canis infection resulted in more pronounced alterations in the pattern of lipid mediators , T . canis- and T . cati-induced NT were comparable in terms of alterations of cerebral oxylipins over the course of infection .
|
Neurotoxocarosis ( NT ) is induced by larvae of the zoonotic roundworms Toxocara canis and T . cati migrating and persisting in the central nervous system of paratenic hosts , and may be accompanied by severe neurological symptoms . Toxocara spp . are known to modulate the hosts’ immune response , but data concerning involvement of signalling molecules are lacking . An important class of mediators participating in the complex molecular signalling network during infection and inflammation are bioactive regulatory lipids , derived from arachidonic acid and other polyunsaturated fatty acids . For a better understanding of inflammatory processes in the brain during an infection with Toxocara spp . , a comprehensive analysis of regulatory lipids was conducted . The infection was predominantly characterised by only minor changes in the pattern of pro-inflammatory oxylipins , while anti-inflammatory metabolites , derived from lipoxygenase pathways , were significantly elevated in the subacute phase as well as in the beginning of the chronic phase of infection . This trend was also reflected in the 13-HODE/9-HODE ratio , a biomarker for the immunological status of an active infection . Obtained data provide a valuable insight in the host’s immune reaction as response against neuroinvasive Toxocara spp . -larvae , contributing to the characterisation of the mostly unknown pathogenesis of NT .
|
[
"Abstract",
"Introduction",
"Material",
"and",
"methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"immunology",
"brain",
"developmental",
"biology",
"metabolites",
"signs",
"and",
"symptoms",
"cerebellum",
"lipids",
"inflammation",
"lipid",
"mediators",
"cerebrum",
"metabolic",
"pathways",
"life",
"cycles",
"immune",
"response",
"biochemistry",
"larvae",
"diagnostic",
"medicine",
"anatomy",
"biology",
"and",
"life",
"sciences",
"metabolism",
"cerebral",
"cortex"
] |
2019
|
Multiplex profiling of inflammation-related bioactive lipid mediators in Toxocara canis- and Toxocara cati-induced neurotoxocarosis
|
Severe dengue infection often has unpredictable clinical progressions and outcomes . Obesity may play a role in the deterioration of dengue infection due to stronger body immune responses . Several studies found that obese dengue patients have a more severe presentation with a poorer prognosis . However , the association was inconclusive due to the variation in the results of earlier studies . Therefore , we conducted a systematic review and meta-analysis to explore the relationship between obesity and dengue severity . We performed a systematic search of relevant studies on Ovid ( MEDLINE ) , EMBASE , the Cochrane Library , Web of Science , Scopus and grey literature databases . At least two authors independently conducted the literature search , selecting eligible studies , and extracting data . Meta-analysis using random-effects model was conducted to compute the pooled odds ratio with 95% confidence intervals ( CI ) . We obtained a total of 13 , 333 articles from the searches . For the final analysis , we included a total of fifteen studies among pediatric patients . Three cohort studies , two case-control studies , and one cross-sectional study found an association between obesity and dengue severity . In contrast , six cohort studies and three case-control studies found no significant relationship between obesity and dengue severity . Our meta-analysis revealed that there was 38 percent higher odds ( Odds Ratio = 1 . 38; 95% CI:1 . 10 , 1 . 73 ) of developing severe dengue infection among obese children compared to non-obese children . We found no heterogeneity found between studies . The differences in obesity classification , study quality , and study design do not modify the association between obesity and dengue severity . This review found that obesity is a risk factor for dengue severity among children . The result highlights and improves our understanding that obesity might influence the severity of dengue infection .
Dengue is an acute systemic viral infection caused by the dengue virus . Currently , there are four known dengue virus subtypes ( DEN-1 , DEN-2 , DEN-3 , DEN-4 ) with Aedes aegypti and Aedes albopictus as the principal vectors . The burdens of dengue infection were concentrated in the tropical and sub-tropical regions mainly in Asia , Caribbean , Africa , the Pacific , and the Americas . Globally , an annual average of 9 , 221 fatal dengue cases was recorded between 1990 and 2013 [1] . In 2013 alone , dengue infection was responsible for 576 , 900 years-life-lost ( YLL ) and 1 . 14 million disability-adjusted-life-years ( DALY ) [1] . Only 25% of those infected with dengue were symptomatic [2] . The signs and symptoms of dengue infection include fever , nausea , vomiting , joint pain , muscular pain , retro-orbital pain , and spontaneous bleeding . The most important predictor of adverse dengue infection was the rapid decrease of platelets with increased hematocrit levels during the critical phase . Currently , the most commonly used dengue classification is the WHO 2009 dengue classification . The WHO 2009 classification classified dengue into Dengue without Warning Signs ( DWWS ) , Dengue with Warning Signs ( DWS ) , and Severe Dengue ( SD ) [3] . Previously , dengue infection was classified using the WHO 1997 classification . The WHO 1997 classified dengue differently from WHO 2009 into Dengue Fever ( DF ) , Dengue Hemorrhagic Fever ( DHF ) , and Dengue Shock Syndrome ( DSS ) . Studies have found that the WHO 2009 dengue classification into severity levels was 53 percent more sensitive in capturing severe dengue infection than the WHO 1997 dengue classification [4 , 5] . Other dengue classifications include WHO 1986 , 1999 , and 2011 . Epidemiological studies have recognized several risk factors for severe dengue infection . Among the risk factors were the two extremes of age , under-nutrition , pregnancy , and secondary infections [6–10] . Furthermore , there were specific clinical signs and symptoms of acute dengue infection that links to severe dengue such as abdominal pain , vomiting , spontaneous bleeding , and diarrhea [11–13] . Nonetheless , other potential risk factors may have a role as novel predictors for dengue severity such as obesity , diabetes , chronic allergies , and hypertension [10 , 14–18] . Overweight and obesity defined as abnormal or excessive fat accumulation that presents a risk to health . Depending on age , different methods were available to measure a body’s healthy weight according to age groups . We defined overweight and obesity using the WHO standards . In children and adolescents using age and sex cut-offs: ( 1 ) overweight defined as either body mass index ( BMI ) ≥ 85th percentile and < 95th percentile or weight-for-height percentage of > 110%; ( 2 ) obesity defined as either BMI ≥ 95th percentile or weight-for-height percentage of more than 120% . In adults , overweight defined as BMI ≥ 25 kg/m2 and < 30 kg/m2 , and obesity was defined as ≥ 30 kg/m2 [19] . Obesity is a major Public Health problem . Globally , the prevalence of overweight and obesity rose by 27 . 5 percent for adults between 1980 and 2013 . The number increased from 857 million in 1980 to 2 . 1 billion in 2013 [20] . In 2016 , the WHO reported that the prevalence of overweight and obesity were at: ( i ) 21 . 9% and 4 . 7% in South-East Asia region; ( ii ) 49% and 20 . 8% in Eastern Mediterranean region; ( iii ) 31 . 7% and 6 . 4% in Western Pacific region; ( iv ) 31 . 1% and 10 . 6% in African region; and ( v ) 62 . 5% and 28 . 6% in Americas region , respectively [21 , 22] . An estimate by Brady et al . on the prevalence of dengue showed that 3 . 9 billion people in 128 countries are at risk for dengue virus infection [23] . WHO also reported that dengue has spread to new areas such as France , Europe , China , and Japan [24] . Hypothetically , obesity may affect the severity of dengue infection through the inflammation pathways . The increased deposition of white adipose tissue ( WAT ) in the overweight and obese individual leads to increase production of inflammatory mediators that were known to increase the capillary permeability and causes plasma leakage [25–27] . In 2013 , Huy NT et al . published a systematic review and meta-analysis of 198 studies up to September 2010 on factors associated with dengue shock syndrome . In the sub-analysis of eight primary studies , the author found that obesity was no association between DSS and overweight or obesity [28] . Recently , in 2016 , Trang et al . published a systematic review and meta-analysis of thirteen studies up to August 2013 on the association between nutritional status and dengue infection [29] . The meta-analysis of eight primary studies on overweight and obesity found no significant association between overweight/obese and DSS . However , the author failed to show substantial consistency regarding the relationship between studies and concluded that the effects of nutritional status on dengue outcomes were controversial . Similarly , both systematic reviews by Huy NT et al . and Trang et al . focused only on the association between nutritional status and dengue infection with studies up to August 2013 . There were scarce and inconclusive shreds of evidence linking overweight and obesity with the severity of dengue infection [30 , 31] . The number obese individuals susceptible to dengue will significantly increase with the increasing prevalence of obesity and increasing populations susceptible to dengue infection . With the hypothesized link between obesity and increasing plasma leakage , obese individuals may be at higher risk of developing severe dengue infection . Our review aims to summarize the current evidence on the association between obesity and severe dengue infection and to identify patients with high risk of severe infection . To our knowledge , this is the first systematic review and meta-analysis on the association between obesity and dengue severity .
We included all interventional and observational studies that evaluated the association between obesity and dengue outcomes . We also added any studies that have information on dengue patients together with information on body compositions such as weight , height , body mass index ( BMI ) and waist circumference ( WC ) . We excluded articles with the following characteristics: ( i ) letter , case report , review , or conference paper; ( ii ) animal study; and ( iii ) in vitro study without patients . MSZ and NAJ independently examined and screened the titles and abstracts for eligibility . There is no restriction with regards to age and publication language . We imported the relevant articles retrieved from the searches to EndNote X7 . 7 . 1 software ( Thomson Reuters , Philadelphia ) . Duplicates were removed primarily using the EndNote software and manually by identification of similar title , author , and the year before the screening process . MSZ also translated one Indonesian article into English during the screening process . Next , the remaining studies were carefully appraised from title and abstract to identify those that may be related to the review question . We based the identification process on the PRISMA flowchart [32]; i . e . , identification , screening , eligibility , and study inclusion . We resolved any disagreements in selection by consensus between authors . We acquired the full-text version of all eligible studies from its databases digitally . MSZ and NAJ reviewed all available full text before data extraction independently . We developed a Google form for data extraction that includes information on the first author , publication year , recruitment , study design , data collection process , allocation of patients ( sequential or random ) , and study site . The form also included characteristics of the patient population , sample size , dengue criteria ) , index of obesity ( BMI , BMI Z-score , BMI-for-age , weight , weight-for-age , height , waist circumference ) , measurement tools for body composition , and age distributions of the patients . Data were then extracted independently by MSZ and NAJ , and verified by SR and MD . The outcome measured in this review was severe dengue infection compared to non-severe dengue infection . We defined severe dengue infection as a collective term that includes either DHF grade III , DHF grade IV , dengue with warning signs , DSS , or severe dengue . Non-severe dengue group includes either DF , DHF grade I , DHF grade II , or dengue without warning signs . The exposure was defined as obesity that includes overweight . The reference group of exposure defined as non-obese includes patients with normal weight and underweight . The quality of the included studies was appraised using the Newcastle-Ottawa Scale ( NOS ) [34] . The NOS evaluated study quality based on three quality parameters ( selection , comparability , and outcome ) divided across eight specific items . Each item on the scale was scored from one point to score up to two points . Currently , there was no specific cut-off score for stratification of study into good , poor , and moderate . Based on our literature search , we found that a cut-off score of seven or more was used to consider a study of good quality . A score between five and six was considered as moderate quality study and anything lesser than five was considered as a poor quality study [35] . Thus , we adopted the similar cut-off score and pair the score with traffic light color coding as described in S1 Table for assessment of study quality in this review . We conducted the meta-analysis using Stata 14 ( StataCorp , Texas ) and considered p<0 . 05 as statistically significant . We used random-effects model for meta-analysis of the association between obesity and dengue severity . The random-effect compared to fixed-effect model is a more robust model by accounting for the variation between the actual effect estimates of included studies . Before pooling the odds ratio estimate , we calculated the individual odds ratio of developing severe dengue infection among obese patients compared to non-obese patients using the available information extracted from the included studies . We computed the pooled odds ratio estimate by weighting each estimate using the inverse variance method . Forest plot and I2 statistic were used to illustrate the combined estimate and evaluate heterogeneity , respectively . In our sensitivity analyses , we used meta-regression to explore the effect of heterogeneity by obesity category ( overweight/obesity and obesity ) , study quality ( good , moderate , poor ) , and study design ( cohort , case-control , cross-sectional ) . Publication bias was assessed using a funnel plot and Egger’s Test . We assumed missingness of any outcomes to be missing completely at random .
We identified a total of 13 , 333 potential reports from five databases and references screening of three earlier relevant reviews [28 , 29] , with 3 , 411 duplicate reports . We included sixty-four reports for the full-text review . Furthermore , we excluded forty-nine reports that did not fulfill the inclusion and exclusion criteria . Finally , we included fifteen reports for systematic review and meta-analysis . The process of study selection was presented in the PRISMA 2009 Flow Diagram ( Fig 1 ) [32] . There was a broad range of objectives amongst the included studies . Three studies analyzed the association between nutritional status with dengue severity [10 , 42 , 44] . Five studies assessed the risk factors for dengue severity and dengue shock syndrome among children infected with dengue [31 , 38 , 40 , 41 , 47] . Remaining studies evaluated childhood obesity as a prognostic factor for dengue shock syndrome [45]; described the clinical disease patterns among children hospitalized for dengue [39]; determined the transfusion requirement among dengue infected patients [36]; analyzed the factor involved in thrombocytopenia and platelet aggregation among dengue hemorrhagic fever patients [37]; examined the association between blood zinc levels and its connection to the severity of dengue hemorrhagic fever clinically [43]; assessed the association between serum transaminase levels and the presence of dengue shock syndrome among children [46]; and determined the levels and correlation between interleukin-8 concentration on admission and dengue shock syndrome outcomes in children [48] . The fifteen studies included a total of 7 , 133 dengue patients with age ranged between 0 and 18 years old . All studies had a clear description of the inclusion criteria . Eleven studies included laboratory-confirmed dengue cases , and the other four [36 , 43 , 46 , 48] studies did not report on the method used for confirmation of dengue cases . The diagnosis was confirmed using either serological test ( hemagglutination inhibition or ELISA ) , viral isolation , or polymerase chain reaction ( PCR ) test . From all fifteen studies , ten studies used obesity [10 , 31 , 37 , 39–42 , 45 , 47 , 48] , four studies used overweight [36 , 38 , 44 , 46] , and one study used over-nutrition [43] as the exposure . The measurement of obesity , overweight , and over-nutrition differ amongst studies . Six studies utilized weight-for-age [10 , 31 , 36 , 38 , 41 , 42] , three studies used BMI-for-age [39 , 43–45] , one study used weight-for-height [40] , while five other studies did not mention the method used to measure the exposure [37 , 46–48] . One in fifteen studies evaluated the possibility of obesity as a prognostic factor for dengue shock syndrome [45] . Two studies used nutritional status as their primary exposure for dengue severity [10 , 44] . Seven studies analyzed the risk factors for dengue severity which include age , sex , nutritional status , as well as other clinical and laboratory risk factors [31 , 38–42 , 47] . In remaining five studies , the author presented obesity , overweight , or over-nutrition as ancillary results [36 , 37 , 43 , 46 , 48] . All five case-control studies defined dengue shock syndrome as the primary outcome [10 , 31 , 40 , 41 , 46] . For cohort study , five studies considered dengue shock syndrome as their primary outcome [36 , 38 , 45 , 47 , 48] . The remaining four studies used dengue hemorrhagic fever [42 , 44] and severe dengue infection ( dengue hemorrhagic grade III and IV ) [37 , 39] as their primary outcomes . While in the cross-sectional study [43] , the author described the proportion of over-nutrition among all grade of dengue fever ( grade I—IV ) . All fifteen studies adopted various World Health Organization classifications of dengue . One study used the WHO 1986 criteria [36] . Nine studies used the WHO 1997 criteria [10 , 37 , 38 , 40–42 , 44–46] . Two studies used the WHO 1999 criteria [39 , 43] . One study used the WHO 1999a criteria [31] . One study used the WHO 2009 criteria [47] , and one study used the WHO 2011 criteria [48] . According to the assessment of study quality using the Newcastle-Ottawa Quality Assessment Scale , we classified nine studies as good quality studies [10 , 31 , 37–39 , 42–44 , 48] and six studies as moderate quality studies [36 , 40 , 41 , 45–47] . Table 3 summarized the total score acquired for each quality domain . Based on the selection domain , six studies [36 , 40 , 41 , 45–47] scored two out of four points , six studies [38 , 44–48] scored three out of four points , and three studies [37 , 39 , 42] scored four over four quality score points . In the comparability domain , eleven studies [10 , 31 , 36 , 38 , 40–43 , 45–47] scored one over two points , and four studies [37 , 39 , 44 , 48] scored a full point of two over two . In the exposure domain , all fifteen studies scored three out of four quality score points . By study design , eight [37–39 , 42 , 44 , 45 , 47 , 48] out of nine cohort studies were classified as good quality studies and one [36] as a moderate quality study . Furthermore , one [46] case-control study was classified as good quality study while four case-control [10 , 31 , 40 , 41] studies were classified as moderate quality . The only cross-sectional study [43] in this review was classified as a moderate quality study . We first pooled the odds ratio for all fifteen studies [10 , 31 , 36–48] evaluating the association between obesity and dengue severity . We found that dengue patients who were obese have 38 percent higher odds of developing a severe presentation of dengue infection compared to non-obese dengue patients ( OR = 1 . 38; 95% CI:1 . 10 , 1 . 73; p = 0 . 01; I2 = 36 . 7% ) ( Fig 2 ) . We found no significant heterogeneity between studies ( p = 0 . 08 ) . There was no statistical evidence of publication bias based on the funnel plot ( Fig 3 ) and Egger’s Test ( p = 0 . 06 ) . In sensitivity analyses using meta-regression , there was no statistical evidence of heterogeneity in obesity classification ( p = 0 . 72 ) , study quality ( p = 0 . 84 ) , nor study design ( p = 0 . 67 ) . As a secondary analysis , we compared obese group in thirteen studies [10 , 31 , 36–42 , 44 , 46–48] with different comparison groups that include: ( i ) normal weight and underweight; ( ii ) normal weight; and ( iii ) underweight . From the pooled analyses , we found that the odds of developing severe dengue infection among the obese group were lower and not statistically significant when compared with the underweight group ( OR = 1 . 26; 95% CI:0 . 93 , 1 . 72; p = 0 . 14 ) . A significant higher odds ratio was found when we compared obese to normal/underweight group ( OR = 1 . 31; 95% CI:1 . 04 , 1 . 65; p = 0 . 02 ) and normal weight group ( OR = 1 . 33; 95% CI:1 . 04 , 1 . 70; p = 0 . 03 ) ( Fig 4 ) .
Our systematic review found a significant association between obesity and dengue severity among children . We found that there were 38 percent higher odds of developing severe dengue infection among obese dengue patients compared to non-obese patients . The results of this review support our hypothesis that obese patients have more severe dengue infection compared to non-obese patients . These findings may be useful in the identification of high-risk dengue patients . Based on our literature review , obese patients were found to be at high risk of developing complication and death due to their stronger immune response compared to malnourished patients [10] . Hypothetically , obesity may affect the severity of dengue through inflammation pathways . The increased deposition of white adipose tissue ( WAT ) in obese individuals leads to an increased production of Interleukin-six ( IL-6 ) , Interleukin-eight ( IL-8 ) , and Tumor Necrosis Factor-Alpha ( TNF-α ) [25–27] . IL-6 , IL-8 , and TNF-a were essential mediators of the inflammation pathway that increases capillary permeability . An increased capillary permeability in obese dengue patients may progressively underlie the process of severe plasma leakage . Thus , with increasing prevalence of obesity in regions with high-risk of dengue infection and the hypothesized link between obesity and dengue severity , an early detection of obese dengue patients was required for closer monitoring and early treatment . Our systematic review found that the included studies defined and measured obesity differently using a different index of measurements such as BMI-for-age , weight-for-age , and weight-for-height . We also found that there were differences in study quality and study design . However , based on our statistical analysis , no heterogeneity found between obesity classification , study quality , and study design . Thus , we concluded that the differences in obesity classification , study quality , and study design do not modify the association between obesity and severity of dengue infection . There were also differences in types of dengue classification used between studies . Since 2009 , multiple revisions on the dengue classification have been done by the World Health Organization ( WHO ) to improve dengue patient management . Currently , the latest classification of dengue categorizes dengue patient into dengue fever , dengue with warning signs , dengue without warning signs and severe dengue . Most studies in this review were using the WHO 1997 dengue classification . Among all , only two studies [47 , 48] used the latest WHO classification ( 2009 , 2011 ) to classify their dengue cases . We did not reclassify cases using the WHO 2009 classification due to the unavailability of the raw data and also to the fact that the WHO 2009 classification classified dengue severity differently based on the presence of warning signs and severe dengue . Thus , with currently available information , it was difficult for us to reclassify all patients as it may lead to selection bias towards the more severe group . Apart from that , all studies included in this review focused only on subjects age between 0 and 18 years old . The unavailability of adult patients limits the applicability of our findings to the general population . With regards to study settings , most of the included studies focus only on hospitalized patients . The hospital settings will encounter more severe dengue cases than primary health care settings . By restricting the study subject to hospitalized patients might introduce bias . It also limits the analysis to a more limited severe dengue group and left out the non-hospitalized group which might be less severe . Therefore , studies related to risk factors or prognostic factors should also include primary healthcare facilities . Based on our literature review , there were higher risks of developing DHF and DSS among individuals with co-morbidities such as diabetes and hypertension[49] . We believed the association between these co-morbidities might be an effect modifier that increases the risk of severe dengue infection but limited only to adult patients . The pathophysiology underlying the effects was not clearly understood as there were not many earlier studies were available . We also found that significantly higher rates of general and central obesity were reported among men and women with higher socioeconomic status in South Asian nations [50 , 51] . In most of the developing countries of South Asia region , the magnitude of household out-of-pocket expenditures on health was at times as high as 80 percent of the total amount spent on health care per annum [52] . We believed that there was a better health-seeking behavior among obese individuals compared to a non-obese individual due to their accessibility for better health care options . Thus , more obese patients were recorded from hospital compared to non-obese patients . We could not ascertain the association between the presence of co-morbidities among obese patients as well as effects of the socioeconomic status of obese patients and severity of dengue infection due to scarce information available . This review has its limitations . We acknowledged the misclassification of obesity due to our inclusion of overweight and over-nutrition . However , we expect the bias to be towards the null as both overweight and over-nutrition refer to similar measures of body composition that is above normal weight . Furthermore , we accepted that the exclusion of fifteen patients from the study by Chuansumrit et al . was due to unavailability of body measurement data [36] . Another seventy-four [44] and seven-hundred-thirty-four controls [10] were also not included because our study does not include the non-dengue patients . Moreover , we also acknowledged that no study with adult participants was included in this review . The unavailability of studies with adult participants might be because previously the distributions of dengue infection were more prevalent among children and less severe among adults . Also , all fifteen included studies were heterogeneous regarding classification , exposure , and outcomes , thus making it more challenging to compare and infer the results . However , these problems were adequately addressed by including all studies that associated with the research question , and not excluding it for lack of quality .
In conclusion , our systematic review and meta-analysis identified a significant association between obesity and dengue severity . The results presented highlight the importance of obesity in dengue infection . It also improves our understanding of the effect of obesity in influencing the severity of dengue infection among children . Further large-scale prospective cohort studies on the effects of obesity on dengue severity among adults and children especially in regions with high prevalence of dengue could provide additional information as well as better understanding on the relevance of obesity in dengue infection .
|
Dengue fever , an Aedes mosquito-borne viral infection , is a self-limiting or asymptomatic viral infection . In its severe form , severe dengue , the symptoms include organ failure , life-threatening bleeding , and shock . Patients with obesity were found to be at higher risks of developing complications and severe dengue infection compared to non-obese patients . Our systematic review and meta-analysis critically compared the available shreds of evidence related to obesity and severe forms of dengue infection . A total of fifteen studies that include nine cohort studies , five case-control studies , and one cross-sectional study were reviewed to explore the association between obesity and dengue severity . The results indicated that obese patients have higher odds of developing severe dengue infection compared to non-obese patients . This review highlighted the importance of obesity in the progression of dengue infection . With increasing prevalence of obese individual as well as expanding areas that were at high risk for dengue infection , obesity may play a role in increasing the burden and mortality related to dengue infection . Further large-scale prospective studies in regions with high prevalence of dengue infections would provide additional information and some better discernment on the relevance of obesity in dengue infection .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion",
"Conclusion"
] |
[
"body",
"weight",
"medicine",
"and",
"health",
"sciences",
"tropical",
"diseases",
"database",
"searching",
"pediatrics",
"physiological",
"parameters",
"nutrition",
"mathematics",
"statistics",
"(mathematics)",
"obesity",
"neglected",
"tropical",
"diseases",
"pediatric",
"infections",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"dengue",
"fever",
"mathematical",
"and",
"statistical",
"techniques",
"childhood",
"obesity",
"physiology",
"database",
"and",
"informatics",
"methods",
"body",
"mass",
"index",
"meta-analysis",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"physical",
"sciences",
"statistical",
"methods"
] |
2018
|
The association between obesity and dengue severity among pediatric patients: A systematic review and meta-analysis
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.