text
string | predicted_class
string | confidence
float16 |
|---|---|---|
The aim of the present study is to confirm this assumption, describe the hydrogen chemisorption properties on M metals (with M = Pd, Rh, or Ir) and determine the stoichiometric ratios H/MS using a simple methodology (statistical model) by the same philosophy as that developed in our previous work . The proposed statistical model will be confronted with the H/M ratios and particle size values obtained from literature data.
|
study
| 100.0 |
The shape of Pd, Ir, or Rh crystallites (or particles) is assumed to be a perfect fcc cuboctahedron (Figure 1). This particle shape was specially chosen because it appears that the cuboctahedron shape can perfectly mimic the evolution of surface atoms of the equilibrium shape of fcc metal (icosahedron, Marks decahedron, perfect truncated decahedron and truncated octahedron) as a function of the crystallite size . Using the methodologies of Van Hardeveld and Hartog , and our previous work , consisting in a systematic way of atom numbering by using mathematical series (the number of atoms are numerically counted for different cluster sizes, and a program is used to determine the logical series associated), it is possible to determine the statistics of atom distribution (NT, NS, NB, and NCi representing the total number of atoms, surface atoms, bulk atoms, and atoms of i coordination number, respectively), dispersion (D), size (d), metallic specific surface area (SM), and adsorption sites (top, bridge, and hollow sites) for metal cuboctahedron cluster (Figure 1). Based on our previous work, Table 1 summarizes the enumeration and the equations giving statistics of atoms, dispersion, size, metallic specific surface area, and the number of each adsorption site for a given value of m (defined as the number of atoms lying on equivalent edge, corners atoms included, of the chosen crystallite) for Pd, Ir, and Rh metal cuboctahedron clusters, respectively .
|
study
| 100.0 |
For the reason of energetic considerations, hydrogen adsorption sites differ from one metal to another. Ab initio and/or DFT calculations obtained from the literature for Pd, Ir, and Rh are therefore used to firstly determine the most favorable adsorption sites, which are evolving with the cluster size. The latter are finally used to build a unique adsorption repetitive sequence for each metal based on a linear combination of these adsorption sites to finally describe the hydrogen adsorption in the full size range. This is detailed in the following section, and summarized in Table 2. These DFT calculations generally consider pure metals, and therefore, unsupported particles, whereas nanoparticles are experimentally deposited onto a support. This raises the question about the nature of adsorption sites between supported and unsupported particles, and also, about the accessibility of a hydrogen atom over the whole metallic surface when a strong metal support interaction (SMSI) occurs. One may reasonably consider that adsorption sites are not modified by the presence of a support, since it has been demonstrated for Ir that top and bridge sites are the most favorable adsorption sites, whether the metal particle is supported or not . Next, concerning the fraction of metal interacting with the support, the metal support interaction is weakened when H/M ratio increases . This metal support interaction weakening is the direct consequence of hydrogen insertion between the metal and the support. Therefore, the entire metal surface is accessible to hydrogen, even in the case of SMSI.
|
study
| 100.0 |
For the Pd flat surfaces, the most favorable sites for H adsorption are the hollow (4-fold) and the hollow (3-fold) fcc sites for Pd(100) and Pd(111) faces, respectively. These are representative of the large particle size domain. For the large Pd clusters, we can select N4(8,8,8,8) adsorption sites for Pd(100), starting from m = 4, and N3fcc(9,9,9) adsorption site for Pd(111), starting from m = 6. In the case of a smaller Pd cuboctahedron cluster (m = 2, 13 atoms), two stable sites for H adsorption were found by Watari et al. . One is the hollow (4-fold) N4(5,5,5,5) inside the square face, and the other one is the hollow (3-fold) hexagonal close packing (hcp) N3hcp(5,5,5) of the triangular face. It has to be mentioned that these sites exist only for small particle sizes, since for m = 2 most of the surface atoms display a coordination number of 5. For intermediate particle size, several 4-fold adsorption sites are coexisting on the square face, which are a combination of coordination number 5 (corners), 7 (edges), and 8 (faces). This leads to two additional possibilities, which are N4(5,7,7,8) resulting from an edge atom creation, starting from m = 3, and N4(7,7,8,8) resulting from an additional face atom creation, starting from m = 4. In the same way, additional 3-fold hcp adsorption sites on a triangular face have to be taken into consideration as the crystallite size is increasing. These are N3hcp(5,7,7), starting from m = 3 and N3hcp(7,7,9), starting from m = 4. As mentioned above, 3-fold hcp sites are the most favoured for small crystallite sizes, whereas 3-fold fcc are favoured for large sizes. In this way, the additional two 3-fold hcp sites permit the transition between small and large crystallites.
|
study
| 100.0 |
Following these hypotheses, the number of H atoms that can be adsorbed on the Pd cuboctahedron surface (for a given m, denoted NH,Pd) can be calculated as follows (Equation (3)): (3)NH,Pd=N3hcp(5,5,5)+N3hcp(5,7,7)+N3hcp(7,7,9)+N3fcc(9,9,9)+N4(5,5,5,5)+N4(5,7,7,8)+N4(7,7,8,8)+N4(8,8,8,8)
|
study
| 99.94 |
In the case of Ir, the most favorable sites for H adsorption are the bridge and the top sites for Ir(100) and Ir(111) faces, respectively, corresponding to the N2(8,8) and N1(9) adsorption sites, both starting from m = 4. Davis et al. calculated that the most favorable H adsorption sites for 38 atom truncated octahedron Ir cluster are the bridge edge sites , indicating that the equivalent position N2(5,7) and N2edge(7,7) adsorption sites have to be taken into account for small cuboctahedron clusters. Moreover, two types of adsorption sites have been suggested on the basis of DFT calculation for tetrahedron Ir4 cluster. These additional adsorption sites are top (corresponding to the N1(5) adsorption site for cuboctahedron clusters) and bridge position at Ir–Ir bonds (corresponding to N2(5,5) adsorption sites for cuboctahedron clusters) . Starting from m = 3, an additional bridge site N2(7,8) appears and has to be considered as another adsorption site.
|
study
| 100.0 |
According to these energetically favored adsorption sites, the number of H atoms that can be adsorbed on the Ir cuboctahedron surface (for a given m, denoted NH,Ir) can be calculated as follows (Equation (4)): (4)NH,Ir=N1(5)+N1(9)+N2(5,5)+N2(5,7)+N2edge(7,7)+0.5×N2(7,8)+0.5×N2(8,8) where the 0.5 coefficient is used to obtain a coverage of 1 monolayer with N2(7,8) and N2(8,8) .
|
study
| 99.94 |
For Rh, the most favorable sites for H adsorption are the hollow (4-fold) and the hollow (3-fold) fcc sites for Rh(100) and Rh(111) faces, respectively, corresponding to N4(8,8,8,8) (starting from m = 4) and N3fcc(9,9,9) (starting from m = 6) adsorption sites. DFT calculations over small sized Rh clusters (tetrahedron Rh4 and octahedron Rh6) indicated that bridge sites are the most stable , corresponding to N2(5,5), for a small cuboctahedron cluster (m = 2). When the cluster size increases, N2(5,7) (starting from m = 3) and N2edge(7,7) (starting from m = 4) equivalent adsorption sites are created, due to the additional appearance of edge atoms. As shown for Pd clusters, the N4(8,8,8,8) sites for (100) faces can lead to the creation of additional 4-fold sites (N4(5,5,5,5)+N4(5,7,7,8)+N4(7,7,8,8)) as the cluster size decreases. Finally, the number of H atoms that can be adsorbed on the Rh cuboctahedron surface (for a given m, denoted NH,Rh) can be calculated as follows (Equation (5)): (5)NH,Rh=N2(5,5)+N2(5,7)+N2edge(7,7)+N3fcc(9,9,9)+N4(5,5,5,5)+N4(5,7,7,8)+N4(7,7,8,8)+N4(8,8,8,8)
|
study
| 100.0 |
The values obtained from this statistical model have subsequently been confronted with numerous literature data reported in Table 3. Results depicted in Figure 2a–c show that the model accurately predicts the literature values of H/Pd, H/Ir, and H/Rh, respectively. In addition, the model predicts H/M values in the range 0–1.08 for Pd, 0–2.77 for Ir, and 0–2.31 for Rh. The latter result clearly indicates that a single stoichiometry for Pd, Ir, and Rh cannot be used.
|
study
| 100.0 |
In order to have a representative view of the surface adsorption properties over Pd, Ir, and Rh, the H/MS theoretical chemisorption stoichiometric factors versus the theoretical H/M ratio are depicted in Figure 2d. The adsorption of one hydrogen atom per surface M atom (MS) is reasonably constant (near unity) for H/Pd < 0.54, H/Ir < 0.28, and H/Rh < 0.36, which corresponds to the large particle size domain. However, when H/Pd ≥ 0.44, H/Ir ≥ 0.28, and H/Rh ≥ 0.36 (small particle size domain), the H/MS ratio increases with the H/M ratio to reach a maximum value of 1.17, 3.00, and 2.50 for Pd, Ir, and Rh, respectively. This particular behavior directly originates from the different sites considered for hydrogen adsorption (Equations (3)–(5)), as well as their relative proportion (Table 1).
|
study
| 100.0 |
The knowledge of the different parameters determined by the model (NT, NS, NH, D (%), d (nm), and SM (m2gM−1)) for any value of m allows drawing correlations with the value of H/M (M corresponding to the chosen metal), the latter being accessible from a chemisorption experiment (Figure 3a–c). It can be seen that the evolution of dispersion, particle size, as well as metallic surface area, are clearly differing from one metal to another. The physical reason for these differences lies in the different adsorption sites between Pd, Rh, and Ir. For a convenient determination of D (%), d (nm), and SM (m2gM−1), a general fifth order polynomial trend line (with the R2 value equal to 1) is provided. The expression of dispersion, reciprocal particle size, and metallic surface area (see Table 1) are given below (Equations (8)–(10)), and are plotted as a function of H/M on Figure 3: (8)D(%)=aD×(HM)5+bD×(HM)4+cD×(HM)3+dD×(HM)2+eD×(HM) (9)1d (nm−1)=a1/d×(HM)5+b1/d×(HM)4+c1/d×(HM)3+d1/d×(HM)2+e1/d×(HM) (10)SM(m2gM−1)=aSM×(HM)5+bSM×(HM)4+cSM×(HM)3+dSM×(HM)2+eSM×(HM)
|
study
| 100.0 |
Equations (6)–(8) can be generalized by the following single equation (Equation (11)): (11)YM=aY×(HM)5+bY×(HM)4+cY×(HM)3+dY×(HM)2+eY×(HM) where aY,bY,cY,dY, and eY are constants depending on the nature of the metal M considered (where M = Pd, Rh, or Ir). The values of these empirical constants for Equation (11) are listed in Table 4.
|
study
| 100.0 |
The methodology described for determining stoichiometric factors for Pt clusters has been successfully generalized to 3 other fcc metals, Pd, Ir, and Rh. The use of this model clearly explains the fundamental reason for overstoichiometries experimentally observed on small particle sizes, and is related to multiple adsorption sites whose relative proportions are strongly size sensitive. The model can also be easily adapted to other shapes, provided that the surface statistics are known. The systematic use of this model for determining metallic specific surface areas from chemisorption experiments is therefore highly recommended for the accurate and meaningful calculation of turnover frequencies (TOF), which is one of the most important parameters to be determined in catalysis. We are currently investigating this aspect in our lab.
|
study
| 100.0 |
Engineering microorganisms to overproduce interested products is an important practice in metabolic engineering. In the successful examples, overexpressing key genes of metabolic pathways is a widely used strategy for achieving overproduction (Ajikumar et al., 2010, Alper et al., 2005). The purpose is to up-regulate the flux for substrate synthesis or to intensify the shunt at key metabolic nodes toward an improved flux to targeted metabolites. Since the overproduction of natively synthesized metabolites is usually achieved by genetically manipulating metabolic pathways, identifying the key pathways and gene targets is a key step to determine gene overexpression strategies for consequential manipulations. Traditionally, completion of such tasks was largely relying on the experience of metabolic pathways and enzymatic kinetics. However, with the increasing practices of metabolic engineering in overproducing fuels, chemicals and natural products (Stephanopoulos, 2012), empirical predictions have been hardly satisfying the analysis of sophisticated pathways, such as the multiple-repeated pathways in fatty acid synthesis and the rarely explored secondary metabolite biosynthesis. Therefore, it is critical to establish a standard procedure for identifying gene overexpression strategies.
|
review
| 99.9 |
The rapid advances of constraint-based models provide the possibilities for quantitative evaluation of cellular metabolism (Bordbar et al., 2015, Kauffman et al., 2003), allowing to develop the standard method for rational pathway design. According to the annotated genome information, the reconstructed constraint-based models could represent the current knowledge of full metabolic reactions and their associated genes for an organism. With those constraint-based models, algorithms such as flux balance analysis (FBA) were developed to perform the in silico analysis of metabolic fluxes (Orth et al., 2010). Relying on the principle of mathematical optimization and mass balance, metabolic fluxes can be simulated within determined constraints. Such efforts have advanced the development of modeling approaches such as OptKnock (Burgard et al., 2003) that facilitates the procedures for identifying gene targets and pathway design.
|
review
| 99.9 |
Unlike gene knockout based simulation, in silico identifying gene overexpression targets has more uncertainties to be experimentally verified because of the difficulties for exactly manipulating fluxes to certain values. To overcome this challenge, methodologies have been developed for simulating gene overexpression, such as OptForce (Ranganathan et al., 2010) and FSEOF (Choi et al., 2010), as well as their derivatives (Chowdhury et al., 2014, Park et al., 2012). By using enforced flux and flux variability analysis, gene targets with desired up-regulation were successfully simulated and experimentally verified. However, those overexpressed gene targets were mostly identified to coordinate with additional manipulations (e.g. knockouts or down-regulation), whereby overexpressing some targets such as targets in glycolysis may not always independently contribute to an overproduction. Therefore, it is important to know the contribution of each candidate targets toward the theoretical maximum yield to fulfill the growing needs on customized pathway design. In addition, most current modeling methods still require specific programming skills that restricts the access for biologists and broad users. It is highly desirable to develop the software platform that can bridge the technical gap between computational modeling and bench works. In this paper, we present a software package, UP Finder that facilitates the identification of gene overexpression strategies for the metabolic engineering of targeted overproduction. It highlighted the quantitative evaluation for each overexpression candidate on yield contribution. The graphical user interface of the UP Finder also provided easier access for broad users. Two typical examples in metabolic engineering that lycopene precursor and fatty acyl-ACP overproduction were used to evaluate feasibilities of the UP Finder for analyzing biosynthesis pathways of natural products and biofuels. The identified gene targets by the UP Finder showed high degree of agreement with the reported key genes in the literatures.
|
study
| 99.9 |
The metabolic reconstructed model of Escherichia coli iJO1366 (Orth et al., 2011) was used for analyzing gene overexpression strategies in lycopene precursor overproduction. And the reconstructed model of Synechocystis sp. PCC 6803 iJN678 (Nogales et al., 2012) was used for the analysis of fatty acyl-ACP overproduction.
|
study
| 100.0 |
FBA was used for all model analysis. For wild-type model, the defaulted biomass formulation was used as the objective function for maximizing cell growth. For theoretical maximum yield model, the targeted product was used as the objective function for maximizing the production of targeted product, such as farnesyl pyrophosphate and fatty acyl-ACP discussed in Results.
|
study
| 100.0 |
All computation was performed on Mac OS × 10.6.8, 1.86 GHz Inter Core 2 Duo Processor, 2 GB 1067 MHz DDR3 Memory. COBRA toolbox v2.0.5 was added to the path of MATLAB_R2012b, including SBML Toolbox_4.1.0 bundled in the package. libSBML_5.7.0 was installed to access the Systems Biology Makeup Language. Gurobi_5.1.0 was used as the LP solver.
|
other
| 99.94 |
The parameter fluxwt represents wild-type flux that is the flux solution of the wild-type model, and fluxopt represents the optimum flux that is the flux solution of the theoretical maximum yield model. The up-regulation ratio (Ratio) is defined as the ratio of fluxopt to fluxwt of a reaction (Ratio = fluxopt / fluxwt). And the Yield is simulated product yield of the targeted product by using fluxopt of a reaction as the constraint, in which maximizing cell growth is the objective function.
|
study
| 99.94 |
UP Finder is an interfacial modeling tool based on the COBRA toolbox in MATLAB, which is developed by the MATLAB Graphical User Interface Development Environment (GUIDE). It is used to identify all the key gene targets for overexpression that directly related to the overproduction of a metabolite in a microorganism. The working procedure of the UP Finder is composed of following steps (Fig. 1):(1)Identification of up-regulated fluxes. The main concept is to compare the flux distributions between the wild-type and overproducing metabolic networks by calculating theoretical maximum yield of a targeted product. Thus, up-regulated fluxes and their associated pathways (termed as up-regulated pathways in this study) can be identified through this comparison.(2)Re-verification of identified pathways. Since not all the identified pathways from Step (1) are directly related to the overproduction, a re-verification is necessary to filter the low-relevant targets. For these identified pathways, their fluxes under overproducing networks were considered as the optimum fluxes (fluxopt) to achieve theoretical maximum yield of the product. The simulated product yields (Yield) constrained by each fluxopt for the wild-type network were used to evaluate the best contribution of each up-regulated pathway toward overproduction. Pathways with Yield > 0 are considered as the key targets that directly lead to the overproduction.(3)Rank of the output. The output of the UP Finder is the abbreviated reaction names of the selected key pathways in Step (2). A termed parameter, Ratio, which is the ratio of each fluxopt over their associated wild-type fluxes (fluxwt) was used for ranking the output from high to low. Because Ratio reflects the up-regulated level for each reaction, the one with the highest Ratio value suggests the highest preference when considering gene overexpression in engineering of the targeted overproduction.Fig. 1The working procedure of the UP Finder. Mutant model, the overproducing metabolic network, the model with flux distribution under the theoretical maximum yield conditions (flux distribution for reaching theoretical maximum yield of a metabolite); fluxwt, wild-type flux, which is the flux distribution of the wild-type conditions; fluxopt, optimum flux, which is the flux distribution of the theoretical maximum yield conditions; Ratio, up-regulation ratio, which is the ratio of the optimum flux to the wild-type flux of a reaction; Yield, simulated yield of the targeted product by using the optimum flux of a reaction as the constraint.Fig. 1
|
study
| 100.0 |
Identification of up-regulated fluxes. The main concept is to compare the flux distributions between the wild-type and overproducing metabolic networks by calculating theoretical maximum yield of a targeted product. Thus, up-regulated fluxes and their associated pathways (termed as up-regulated pathways in this study) can be identified through this comparison.
|
study
| 100.0 |
Re-verification of identified pathways. Since not all the identified pathways from Step (1) are directly related to the overproduction, a re-verification is necessary to filter the low-relevant targets. For these identified pathways, their fluxes under overproducing networks were considered as the optimum fluxes (fluxopt) to achieve theoretical maximum yield of the product. The simulated product yields (Yield) constrained by each fluxopt for the wild-type network were used to evaluate the best contribution of each up-regulated pathway toward overproduction. Pathways with Yield > 0 are considered as the key targets that directly lead to the overproduction.
|
study
| 100.0 |
Rank of the output. The output of the UP Finder is the abbreviated reaction names of the selected key pathways in Step (2). A termed parameter, Ratio, which is the ratio of each fluxopt over their associated wild-type fluxes (fluxwt) was used for ranking the output from high to low. Because Ratio reflects the up-regulated level for each reaction, the one with the highest Ratio value suggests the highest preference when considering gene overexpression in engineering of the targeted overproduction.
|
study
| 99.94 |
The working procedure of the UP Finder. Mutant model, the overproducing metabolic network, the model with flux distribution under the theoretical maximum yield conditions (flux distribution for reaching theoretical maximum yield of a metabolite); fluxwt, wild-type flux, which is the flux distribution of the wild-type conditions; fluxopt, optimum flux, which is the flux distribution of the theoretical maximum yield conditions; Ratio, up-regulation ratio, which is the ratio of the optimum flux to the wild-type flux of a reaction; Yield, simulated yield of the targeted product by using the optimum flux of a reaction as the constraint.
|
study
| 83.25 |
Through the interface of the UP Finder, after loading a Systems Biology Markup Language (SBML) model in the Organism item, all metabolites included in this model will be shown in the Targeted product item. Users can simply specify one metabolite as the target for overproduction. By choosing UPA in the Method option and running the program, a list of reaction names presented with their associated genes, reaction formulas, Yield and Ratio values will be returned as the output. All computation is based on the COBRA toolbox and MATLAB, and all optimization uses FBA for the solutions (Orth et al., 2010, Schellenberger et al., 2011). Initializing the COBRA toolbox is necessary in MATLAB before loading SBML models. The default uptake and growth constraints of the reconstructed model are used for the analysis. Users can also adjust the uptake and growth conditions to simulate metabolisms with special requirements. In addition, the UP Finder integrates FBA optimization in the Method option, which allows the basic function for computing growth rates under different conditions (Fig. 2). The UP Finder is freely available from GitHub (https://github.com/MEpathway/UP-Finder.git).Fig. 2The interface of the UP Finder. The interface contains 5 major functional units, including the Organism, Condition/Growth, Targeted product, Method and Output. Condition is the exchange reactions of SBML models with their constraints, and Growth indicates the specific biomass objective function (e.g. autotrophic or heterotrophic growth for Synechocystis sp. PCC 6803). Method contains two computational methods: UPA (up-regulated pathway analysis) and FBA (flux balance analysis).Fig. 2
|
study
| 99.94 |
The interface of the UP Finder. The interface contains 5 major functional units, including the Organism, Condition/Growth, Targeted product, Method and Output. Condition is the exchange reactions of SBML models with their constraints, and Growth indicates the specific biomass objective function (e.g. autotrophic or heterotrophic growth for Synechocystis sp. PCC 6803). Method contains two computational methods: UPA (up-regulated pathway analysis) and FBA (flux balance analysis).
|
other
| 99.75 |
As an important isoprenoid, lycopene overproduction is a textbook example in metabolic engineering. In E. coli, overproducing farnesyl pyrophosphate (FPP) is critical for increasing the product yield of lycopene because FPP is the native precursor in lycopene biosynthesis (Alper et al., 2005). Herein, we presented the analysis of FPP (frdp[c]) as the Targeted product for overproduction in the UP Finder (Fig. 2). By loading the constraint-based model of E. coli iJO1366 in the Organism item, the output presented key pathway/gene targets for overexpression toward FPP overproduction in E. coli. As shown in Fig. 3, ten metabolic reactions (9 genes involved) were identified as the key pathways for FPP overproduction. The identified reactions were ordered by their associated Ratio values, which can be used as a quantitative evaluation for the overexpression preference. According to this ranked preference for gene overexpression, it was observed that the higher preferred gene targets showed the closer metabolic distance to the targeted product (FPP), which represented the higher relevance to the related overproduction. Compared with the reported key genes for FPP/lycopene overproduction in the literatures (Wang et al., 2009), results from the UP Finder identified all 9 key genes in isoprenoid biosynthesis that directly related to FPP overproduction, showing 100% identity. Also, comparing to the key genes identified by FSEOF (Choi et al., 2010), those gene targets in central carbon metabolic metabolism were not included in the results of UP Finder (Table 1).Fig. 3Analysis results from the UP Finder for farnesyl pyrophosphate (FPP) overproduction in E. coli. Identified gene targets are also shown in the metabolic pathway of FPP biosynthesis presented with gene overexpression preference toward FPP overproduction. GAP, glyceraldehyde-3-phosphate; DXP, 1-deoxy-D-xylulose 5-phosphate; MEP, 2-C-methyl-D-erythritol 4-phosphate; CDP-ME, 4-diphosphocytidyl-2-C-methyl-D-erythritol; CDP-MEP, 4-diphosphocytidyl-2C-methyl-D-erythritol-2-phosphate; MEC, 2C-methyl-D-erythritol-2,4-cyclodiphosphate; HMBPP, (E)−4-hydroxy-3-methylbut-2-enyl-diphosphate; IPP, isopentenyl diphosphate; DMAPP, dimethylallyl diphosphate; GPP, geranyl pyrophosphate; FPP, farnesyl pyrophosphate. dxs, 1-deoxy-D-xylulose-5-phosphate synthase; dxr, 1-deoxy-D-xylulose reductoisomerase; ispD, 2-C-methyl-D-erythritol 4-phosphate cytidylyltransferase; ispE, 4-(cytidine 5′-diphospho)−2-C-methyl-D-erythritol kinase; ispF, 2-C-methyl-D-erythritol 2,4-cyclodiphosphate synthase; ispG, 2C-methyl-D-erythritol 2,4 cyclodiphosphate dehydratase; ispH, 1-hydroxy-2-methyl-2-(E)-butenyl 4-diphosphate reductase; idi, isopentenyl-diphosphate D-isomerase; ispA, geranyltranstransferase (farnesyl diphosphate synthase); crtE, GGPP synthase, crtB, phytoene synthase; crtI, phytoene desaturase. See Supplementary Table S1–S2 for the abbreviations of reaction and metabolite names shown in the UP Finder results.Fig. 3Table 1Comparison of identified gene targets for overproducing lycopene precursor (farnesyl pyrophosphate, FPP) in E. coli.Table 1Reported key genes (Wang et al., 2009)Identified key genes by UP FinderIdentified key genes by FSEOF (Choi et al., 2010)dxsdxsdxsdxrdxrdxrispDispDispDispEispEispEispFispFispFispGispGispGispHispHispHidiidiidiispAispAispApgipfkABfbaAtpiAgltAacnABicdAsucABsucCDsdhABCDfumABmdh
|
study
| 100.0 |
Analysis results from the UP Finder for farnesyl pyrophosphate (FPP) overproduction in E. coli. Identified gene targets are also shown in the metabolic pathway of FPP biosynthesis presented with gene overexpression preference toward FPP overproduction. GAP, glyceraldehyde-3-phosphate; DXP, 1-deoxy-D-xylulose 5-phosphate; MEP, 2-C-methyl-D-erythritol 4-phosphate; CDP-ME, 4-diphosphocytidyl-2-C-methyl-D-erythritol; CDP-MEP, 4-diphosphocytidyl-2C-methyl-D-erythritol-2-phosphate; MEC, 2C-methyl-D-erythritol-2,4-cyclodiphosphate; HMBPP, (E)−4-hydroxy-3-methylbut-2-enyl-diphosphate; IPP, isopentenyl diphosphate; DMAPP, dimethylallyl diphosphate; GPP, geranyl pyrophosphate; FPP, farnesyl pyrophosphate. dxs, 1-deoxy-D-xylulose-5-phosphate synthase; dxr, 1-deoxy-D-xylulose reductoisomerase; ispD, 2-C-methyl-D-erythritol 4-phosphate cytidylyltransferase; ispE, 4-(cytidine 5′-diphospho)−2-C-methyl-D-erythritol kinase; ispF, 2-C-methyl-D-erythritol 2,4-cyclodiphosphate synthase; ispG, 2C-methyl-D-erythritol 2,4 cyclodiphosphate dehydratase; ispH, 1-hydroxy-2-methyl-2-(E)-butenyl 4-diphosphate reductase; idi, isopentenyl-diphosphate D-isomerase; ispA, geranyltranstransferase (farnesyl diphosphate synthase); crtE, GGPP synthase, crtB, phytoene synthase; crtI, phytoene desaturase. See Supplementary Table S1–S2 for the abbreviations of reaction and metabolite names shown in the UP Finder results.
|
study
| 100.0 |
Directly converting CO2 into biofuels is regarded as a promising strategy for producing carbon-neutral renewable energy (Atsumi et al., 2009). Fatty acyl-ACP is the important precursor for the biosynthesis of fatty-acid based biofuel molecules, such as free fatty acids, fatty alcohols and alkanes (Liu et al., 2011). Herein, gene overexpression targets were analyzed for overproducing fatty acyl-ACP in the cyanobacterium Synechocystis sp. PCC 6803. As the dominant component of fatty acyl-ACP, palmitoyl-ACP (C16:0 ACP, palmACP[c]) was chosen as the Target product in the UP Finder. Fig. 4 shows the output results. A total of 30 metabolic reactions, involving 6 genes (accABCD, fabD, fabF, fabG, fabZ, fabI) were identified as the key pathways/genes for palmitoyl-ACP overproduction. 28 out of the 30 reactions are catalyzed by fatty acid synthases in Synechocystis sp. PCC 6803, in which 4 key genes (fabF, fabG, fabZ, fabI) are involved. The reactions were ordered by their associated Ratio values, and the different Ratio values of the 30 identified up-regulated pathways represent different up-regulation levels to achieve the same Yield (theoretical maximum yield). Thus, the higher Ratio values indicated the higher demands of metabolic flux for up-regulation toward palmitoyl-ACP overproduction. On the other hand, the completed fatty acid synthesis pathways of the constraint-based model enabled the detailing of fluxes for each single reaction in multiple-repeated fatty acid biosynthesis. It was found that the reactions with same Ratio values reflected the similar level of metabolic flux for going through, which might be used as a quantitative standard for identifying metabolic modules in complex metabolic pathways.Fig. 4Analysis results from the UP Finder for fatty acyl-ACP (palmitoyl-ACP) overproduction in Synechocystis sp. PCC 6803. Identified gene targets are also shown in the metabolic pathway of fatty acyl-ACP biosynthesis presented with gene overexpression preference toward fatty acyl-ACP overproduction. RB15BP, ribulose 1,5-bisphosphate; 3PG, 3-phosphoglycerate; α-KG, α-ketoglutarate; Mal-CoA, Malonyl-CoA; Mal-ACP, Malonyl-ACP. See Supplementary Table S1–S2 for the abbreviations of reaction and metabolite names shown in the UP Finder results.Fig. 4
|
study
| 100.0 |
Analysis results from the UP Finder for fatty acyl-ACP (palmitoyl-ACP) overproduction in Synechocystis sp. PCC 6803. Identified gene targets are also shown in the metabolic pathway of fatty acyl-ACP biosynthesis presented with gene overexpression preference toward fatty acyl-ACP overproduction. RB15BP, ribulose 1,5-bisphosphate; 3PG, 3-phosphoglycerate; α-KG, α-ketoglutarate; Mal-CoA, Malonyl-CoA; Mal-ACP, Malonyl-ACP. See Supplementary Table S1–S2 for the abbreviations of reaction and metabolite names shown in the UP Finder results.
|
study
| 100.0 |
Similar to the finding of Example 1, it was also found that the higher preferred gene targets presented closer metabolic distance to palmitoyl-ACP. Compared with reported key genes regarding to fatty acyl-ACP overproduction in the literatures (Liu et al., 2011), results identified by the UP Finder show 100% identity.
|
study
| 100.0 |
In this study, a modeling software package named UP Finder was developed based on the COBRA toolbox in MATLAB. It facilitated the rapid identification of gene overexpression strategies for the metabolic engineering of targeted overproduction. Gene overexpression targets can be rationally determined by a quantitative evaluation procedure. Development of this interfacial software package was designed to provide the “one-click” convenience, and to facilitate the access for potential users without specific biochemistry and programming backgrounds. By taking advantage of standardized format of SBML models, UP Finder provided broad access for analyzing various targeted products in different microorganisms. Unlike OptForce and FSEOF, UP Finder specifically identified gene targets that were highly related to overproduction rather than all potential important targets. The UPA method used in the UP Finder investigated product yields for each single potential up-regulated pathway by constraining their fluxes with fluxopt, which evaluated their contribution toward theoretical maximum yield and enabled to pinpoint the results as yield/overproduction related. Although some key upstream pathways, such as glycolysis play an important role in improving target product yields when combining with downstream pathway enhancements, the sole overexpression of these genes might not directly contribute to the overproduction.
|
study
| 100.0 |
The quantitative evaluation that combining Ratio and Yield parameters improved the relevance of identified pathways for directly leading to the overproduction. In the UP Finder, gene overexpression targets were first identified as up-regulated pathways (the reaction with Ratio > 0) through a flux comparison between the theoretical maximum yield model and the wild-type model. By using fluxopt as a constraint, the contribution of each single identified up-regulated pathway on targeted overproduction was quantitatively evaluated based on the yield simulation. The restriction of Yield > 0 was used to exclude the reactions that are indirectly related to the overproduction, such as the pathways in central carbon metabolism. Thus, the UP Finder provided a quantitative procedure for identifying the key pathways toward overproduction. In addition, it was found in the examples that pathways with higher Ratio values showed closer metabolic distance to the targeted product. Since the greater Ratio value (up-regulation level) indicates the higher demands of metabolic flux for up-regulation to approach the theoretical maximum yield, overexpression of the particular gene would contribute more to the overproduction of targeted products. Therefore, results from the UP Finder not only presented all the key pathway/gene targets related to the overproduction, but the ranking of the outputs with their associated Ratio values also reflected the preference for considering overexpression strategies in pathway design.
|
study
| 100.0 |
On the other hand, the Ratio parameter can be used for identifying functional metabolic modules. In the analysis examples, it was found that pathways with similar level of Ratio values presented the adjacent locations with regard to metabolic functions. For example, in Example 1 (Fig. 3), identified key pathways can be divided into 3 metabolic modules by their Ratio levels, including the initial MEP pathway synthesis module (dxs), the isoprenoid unit synthesis module (dxr, ispD, ispE, ispF, ispG, ispH), and the FPP synthesis module (idi, ispA). In Example 2 (Fig. 4), pathways with similar chain length of fatty acid synthesis also showed the similar levels on their Ratio values. In metabolic engineering of secondary metabolites and complex metabolic pathways, the imbalance of metabolic flux is a critical limiting factor for reaching high product yields. To coordinate the metabolic imbalance, engineering of module-based metabolic optimization has been regarded as a promising strategy for optimizing product yields (Ajikumar et al., 2010, Xu et al., 2013, Yadav et al., 2012, Zhao et al., 2013). Therefore, by taking advantage of the Ratio parameter, outputs of the UP Finder could also provide a quantitative basis for identifying functional metabolic modules in developing module-based optimization strategies.
|
study
| 99.94 |
In this version of the UP Finder, not all metabolites listed in the reconstructed model will have valid results. It is usually good for analyzing terminal metabolites, such as secondary metabolites. For metabolites with considerable degradation pathways, it may not have valid results because the up-regulated flux would be further consumed by the degradation without accumulation. Since the evaluation process only works for single pathway, the UP Finder does not provide the best combination of overexpression yet in this version. To achieve accurate prediction and high identity with experimental verification, a high quality of metabolic network reconstruction is necessary. Some current SBML models such as E. coli (iJO1366.xml), and Synechocystis sp. PCC 6803 (iJN678.xml) have been tested with valid outputs in this UP Finder. Analysis examples demonstrated that the UP Finder is feasible to analyze gene overexpression targets for overproducing secondary metabolites and complex metabolic pathways, such as fatty acid biosynthesis. Given the decreasing cost of DNA synthesis, fast strain development for overproducing targeted products is becoming possible based on the large-scale DNA synthesis. Therefore, a user-friendly interfacial modeling tool that provides rapid pathway design would play an important role in the era of synthetic biology (Gibson, 2014, Kosuri and Church, 2014, Wang et al., 2011).
|
review
| 52.75 |
In this study, a modeling tool named UP Finder was developed based on the COBRA toolbox. It facilitated the rapid identification of gene overexpression strategies to assist pathway design in metabolic engineering of targeted overproduction. Gene targets with highly related to overproduction were determined by a quantitative evaluation procedure. The graphical user interface of the UP Finder provided easier access for analyzing various targeted products in different microorganisms. Analysis examples for overproducing lycopene precursor and fatty acyl-ACP by the UP Finder showed high degree of agreement with the reported key genes in the literatures.
|
study
| 100.0 |
Duchenne muscular dystrophy (DMD), caused by mutations in the gene encoding for the cytoskeletal protein dystrophin, is a severe illness characterized by progressive muscle weakness and degeneration. In affected patients this eventually leads to loss of ambulation, respiratory failure, and premature death. In healthy muscle cells, dystrophin interacts with numerous proteins of the so-called dystrophin-associated protein complex (DAPC),1,2 thereby serving as a linker between the cytoskeleton and the extracellular matrix. Disruption of this link in case of dystrophin-deficiency renders muscle tissue vulnerable to mechanical stress.3,4
|
review
| 89.2 |
Besides skeletal muscle degeneration, dystrophin-deficiency in DMD also initiates severe cardiac complications. Among those, cardiac arrhythmias and the development of a dilated cardiomyopathy considerably contribute to the morbidity and mortality associated with the disease.5,6 Although the precise mechanisms causing these cardiac complications are largely unknown, there is evidence that impaired cardiac ion channel expression and function are involved. For example, dystrophin-deficient ventricular cardiomyocytes derived from mouse models for DMD express less cardiac Nav1.5 sodium channel protein than healthy myocytes.7,8 Dystrophic cardiomyocytes also show significantly reduced sodium current densities and altered sodium channel gating properties.7-9 Reduced sodium currents in murine dystrophic cardiomyocytes match with the impairments in cardiac impulse conduction observed in DMD patients (e.g.10). Since ion channels in cardiomyocytes do not function in isolation, but instead in an orchestrated fashion as part of complex protein networks,11,12 it is not surprising that dystrophin-deficiency impairs the properties of Nav1.5 channels. Nav1.5 is considered a member of the DAPC and as such a direct interaction partner of the dystrophin-binding protein syntrophin.7,13,14 The lack of dystrophin may therefore disturb regulatory interactions of Nav1.5 within the DAPC, which normally are a prerequisite for the proper expression and function of this channel. Accordingly, syntrophin mutations were linked with an abnormal sodium current through Nav1.5 channels.15
|
study
| 97.06 |
Apart from the well described regulation of Nav1.5 sodium channels by the DAPC in cardiomyocytes, evidence has been accumulating that this protein complex also plays a role in regulating Kir2.x inward rectifier potassium channels. Thus, like Nav1.5, Kir2.x channels do interact with the DAPC protein syntrophin through their PDZ-binding motifs.12,16,17 Moreover, Nav1.5 and Kir2.1 colocalize and modulate each other's surface expression in cardiomyocytes,14,18 suggesting organization of these 2 channels in a common protein complex. Based on this evidence, the authors of a recently published review article12 hypothesized that the DAPC is important in the regulation of Kir2.x channel expression and function. Functional evidence for this hypothesis, however, has been lacking so far. Kir2.x channels in ventricular cardiomyocytes (most prominently Kir2.119,20) account for the inward rectifier potassium current IK1, which controls the resting membrane potential, and the final phase of action potential repolarization.19,21 Clinically, mutations in the KCNJ2 gene encoding for Kir2.1 induce diseases associated with severe cardiac arrhythmias and increased risk of sudden cardiac death.21,22
|
review
| 99.9 |
In order to test the hypothesis that the DAPC is important in the regulation of Kir2.x channels, here we have studied potential IK1 abnormalities in dystrophin-deficient ventricular cardiomyocytes derived from DMD mouse models. Besides the classical dystrophin-deficient mdx mouse,23 we also used mice additionally carrying a mutation in the utrophin gene (mdx-utr).24,25 The latter DMD mouse model develops a more severe cardiomyopathy with an earlier onset compared to mdx.25,26
|
study
| 100.0 |
Figure 1A shows typical original traces of whole cell potassium currents recorded from a wild type (wt) cardiomyocyte, and from dystrophic (mdx and mdx-utr) cardiomyocytes. The currents were elicited by depolarizing and hyperpolarizing voltage steps outgoing from a holding potential of −100 mV (pulse protocol, displayed in the inset of Fig. 1B). A summary of the current density-voltage relationships, derived from a series of such experiments, is presented in Fig. 1B. It can be seen that the current densities of dystrophic cardiomyocytes were substantially decreased when compared with those in wt cells. A comparison of the current densities in wt and dystrophic cardiomyocytes at −100 mV revealed a highly significant difference (Fig. 1C). Figs. 1B and 1C do also show that both types of dystrophic cardiomyocytes displayed similar potassium inward currents. Thus, the current densities of mdx and mdx-utr cardiomyocytes were almost identical over the whole voltage range studied (Fig. 1B). Together these data suggests that dystrophin-deficiency (mdx and mdx-utr) in cardiomyocytes generated the observed decrease in IK1 density when compared to the wt, with no additional effect of utrophin-deficiency (only mdx-utr). Figure 1.IK1 in wt and dystrophic ventricular cardiomyocytes. (A) Typical original potassium current traces of a wt, mdx, and mdx-utr cardiomyocyte elicited from a holding potential of −100 mV by 500-ms steps to various voltages (the pulse protocol is shown in the inset of Fig. 1B). The dashed line indicates the zero current level. (B) Current density-voltage relationships derived from a series of experiments as shown in Fig. 1A (n = 38 for wt, 36 for mdx, and 13 for mdx-utr, respectively). The current levels at the end of the test pulse were evaluated and plotted against the applied voltages. The lines through the data points represent fits with the function: y = Y0/(1 + exp((x-V05)/k)). (C) Current density values at −100 mV (means ± SEM) are compared between wt and dystrophic (mdx and mdx-utr) cardiomyocytes. *** indicates that ANOVA revealed a highly significant difference between the tested groups (p < 0.001). (D) Current density values at −100 mV of wt and dystrophic cardiomyocytes from all experiments on female (♀) mice (n = 14 for wt, 15 for mdx, and 13 for mdx-utr). ** p < 0.01, ANOVA. E: Current density values at −100 mV of wt and mdx cardiomyocytes from all experiments on male (♂) mice (n = 24 for wt and 21 for mdx). * p < 0.05 (p = 0.011), Student's t-test.
|
study
| 100.0 |
IK1 in wt and dystrophic ventricular cardiomyocytes. (A) Typical original potassium current traces of a wt, mdx, and mdx-utr cardiomyocyte elicited from a holding potential of −100 mV by 500-ms steps to various voltages (the pulse protocol is shown in the inset of Fig. 1B). The dashed line indicates the zero current level. (B) Current density-voltage relationships derived from a series of experiments as shown in Fig. 1A (n = 38 for wt, 36 for mdx, and 13 for mdx-utr, respectively). The current levels at the end of the test pulse were evaluated and plotted against the applied voltages. The lines through the data points represent fits with the function: y = Y0/(1 + exp((x-V05)/k)). (C) Current density values at −100 mV (means ± SEM) are compared between wt and dystrophic (mdx and mdx-utr) cardiomyocytes. *** indicates that ANOVA revealed a highly significant difference between the tested groups (p < 0.001). (D) Current density values at −100 mV of wt and dystrophic cardiomyocytes from all experiments on female (♀) mice (n = 14 for wt, 15 for mdx, and 13 for mdx-utr). ** p < 0.01, ANOVA. E: Current density values at −100 mV of wt and mdx cardiomyocytes from all experiments on male (♂) mice (n = 24 for wt and 21 for mdx). * p < 0.05 (p = 0.011), Student's t-test.
|
study
| 100.0 |
In order to check for potential gender differences, the data presented in Fig. 1C are displayed separately with respect to the sex of the animals used for cardiomyocyte isolation (Figs. 1D and 1E). In Fig. 1D the potassium current densities from wt and dystrophic (mdx and mdx-utr) cardiomyocytes derived from female mice only are compared. In Fig. 1E the respective current density data (wt and mdx) from only male mice are presented. Our data imply that dystrophin-deficiency significantly decreased the IK1 densities, irrespective of the gender of the animals used for cardiomyocyte isolation.
|
study
| 100.0 |
Kir2.1 potassium channels are the major determinants of IK1 in murine ventricular cardiomyocytes.19 In order to test if the decreased IK1 in dystrophic cardiomyocytes can be explained by a reduced expression of Kir2.1 channels, we performed western blot experiments with membrane fractions of cardiac ventricular tissues isolated from wt and dystrophic (mdx and mdx-utr) mice. Fig. 2 (A and B) shows that Kir2.1 protein levels in wt and dystrophic ventricles were similar. This suggests that Kir2.1 channel expression is normal in dystrophin-deficient cardiomyocytes. Finally, Fig. 2C compares typical Kir2.1 immunostainings of an isolated wt and a mdx cardiomyocyte. Cross-striations, typical for T-tubular localization of Kir2.1 channels in mouse ventricular cardiomyocytes,27 can be observed both in the normal and the dystrophic cell. Similar staining patterns showing cross-striations were obtained in all the studied wt and mdx cardiomyocytes, which were isolated from 3 normal and dystrophic mice, respectively. These immunostaining data implies normal T-tubular Kir2.1 channel localization in dystrophic cardiomyocytes. Figure 2.Kir2.1 protein exparession and localization in wt and dystrophic ventricular cardiomyocytes. (A) Representative western blot experiment of membrane lysates from adult wt and dystrophic (mdx and mdx-utr) ventricular tissues stained for Kir2.1 and the β-subunit of a Gs protein (AS7). The latter was used as loading control. (B) Relative band intensities of Kir2.1 normalized to the respective band intensities of the loading control plotted as means ± SEM for wt (n = 5) and dystrophic (mdx, n = 6; mdx-utr, n = 2) animals. A Student's t-test did not reveal a significant difference between wt and mdx (p = 0.11; n.s., not significant). (C) Typical examples of Kir2.1 immunostainings of an isolated wt (left) and mdx (right) cardiomyocyte.
|
study
| 100.0 |
Kir2.1 protein exparession and localization in wt and dystrophic ventricular cardiomyocytes. (A) Representative western blot experiment of membrane lysates from adult wt and dystrophic (mdx and mdx-utr) ventricular tissues stained for Kir2.1 and the β-subunit of a Gs protein (AS7). The latter was used as loading control. (B) Relative band intensities of Kir2.1 normalized to the respective band intensities of the loading control plotted as means ± SEM for wt (n = 5) and dystrophic (mdx, n = 6; mdx-utr, n = 2) animals. A Student's t-test did not reveal a significant difference between wt and mdx (p = 0.11; n.s., not significant). (C) Typical examples of Kir2.1 immunostainings of an isolated wt (left) and mdx (right) cardiomyocyte.
|
study
| 100.0 |
Voltage-dependent ion channels in cardiomyocytes are not solely regulated by the transmembrane voltage, but also through interactions with numerous proteins organized in complex networks.11,12 Such “ion channel multiprotein assemblies” comprising (but not limited to) anchoring proteins, adaptor proteins, regulatory proteins, and enzymes regulate the trafficking, expression, localization, and function of an associated channel.
|
study
| 99.94 |
A recently published review article12 deals with protein assemblies of cardiac Nav1.5 sodium and inward rectifier Kir2.1 potassium channels, that control cardiac excitability and, in case of dysregulation, introduce arrhythmogenesis. Interestingly, these 2 channels not only share a number of common protein interaction partners (e.g., caveolin-3, synapse-associated protein-97, and syntrophin) (12 and references therein), but also modulate each other's cell-surface expression.14,18 These findings suggest organization of Nav1.5 and Kir2.1 in one or several common protein complexes. Here the DAPC stands out as a likely candidate. Thus, at the cardiomyocyte membrane, both channels interact through the PDZ domain-scaffolding protein syntrophin (see Fig. 2A in12), which is a direct binding partner of dystrophin within the DAPC.
|
review
| 99.9 |
The well-known fact that disruption of the DAPC (due to dystrophin-deficiency) in cardiomyocytes significantly impairs Nav1.5 channel expression7,8 and function7-9 suggests that an intact DAPC is required to maintain normal Nav1.5 channel properties. In contrast, the notion that the DAPC is also important in the regulation of Kir2.x channel expression and function12 has remained purely speculative till this day. In the present study, we provide the first functional evidence that the DAPC indeed also impacts Kir2.x channels by showing that IK1 in dystrophin-deficient cardiomyocytes is substantially diminished. This suggests that, like Nav1.5 channels, Kir2.x channels require interaction with the DAPC for their proper functioning. In both cases, a disruption of dystrophin leads to a marked reduction in the respective current (INa and IK1, respectively). A reduced IK1 due to dystrophin-deficiency may provide an explanation for the low resting potentials measured in dystrophic cardiac8,28 and skeletal29-31 myocytes when compared to their healthy counterparts. It also represents a so far unknown potential mechanism to cause cardiac arrhythmias in DMD patients.
|
study
| 100.0 |
An obvious difference between the regulatory action of the DAPC on Nav1.5 and Kir2.x channels in cardiomyocytes, which has emerged from our findings and previous studies, concerns the effect of the protein complex on channel expression. Thus, dystrophin-deficiency induces a substantial reduction in Nav1.5 channel expression,7,8 but has no (the present study) or only minor7 effect on Kir2.1 protein levels. This suggests that the decreased IK1 in dystrophin-deficient cardiomyocytes is not due to reduced Kir2.1 channel expression, but is caused by another mechanism. Because Kir2.1 channel localization in dystrophic cardiomyocytes may also be normal (Fig. 2C), channel inhibition by cytoplasmic regulatory factors is likely. Follow-up studies are needed to identify the responsible regulator(s). A better insight in how the DAPC regulates Kir2.x channels to control cardiac excitability, and how mutations in genes encoding for the involved proteins impair this regulation should increase our understanding of diseases associated with arrhythmias such as DMD, and may lead to improvements in therapy. Finally, DAPC regulation of inward rectifier potassium channels may also be relevant in other organs, e.g. the brain. Thus, in brains from mdx mice, dystrophin-deficiency was associated with a significant reduction in Kir4.1 mRNA expression and protein content.32
|
study
| 99.94 |
Wild type (wt) C57BL/6 mice, dystrophin-deficient mdx,23 and dystrophin/utrophin-deficient double mutant mdx-utr24,25 mice, backcrossed on the C57BL/6 background for more than 12 generations, were used for the study. Details about the mutations are described in our earlier work.33 Through the text, the dystrophin−/− status (utrophin+/+) is named “mdx,“ and double mutant mice (dystrophin−/− and utrophin−/−) are termed “mdx-utr.” The mice were genotyped using standard PCR-assays. For comparisons between mdx and mdx-utr, littermates were used.
|
study
| 100.0 |
15–25 week-old female mice (2 wt, 2 mdx, and 2 mdx-utr animals) and male mice (3 wt and 3 mdx animals) were killed by cervical dislocation, and cardiomyocytes were isolated from the ventricles of their hearts by using a Langendorff setup as described in our previous work.9
|
study
| 100.0 |
Applying the whole cell patch clamp technique and established pulse protocols, IK1 was recorded at room temperature (22 ± 1.5°C) from ventricular cardiomyocytes up to 8 hours after preparation using an Axoclamp 200B patch clamp amplifier (Axon Instruments, Union City, CA). Pipettes were formed from aluminosilicate glass (AF150-100-10; Science Products, Hofheim, Germany) with a P-97 horizontal puller (Sutter Instruments, Novato, CA). Their resistances were between 0.8 and 2 MΩ when filled with pipette solution (see below). Data acquisition was performed with pClamp 6.0 software (Axon Instruments) through a 12-bit A-D/D-A interface (Digidata 1200; Axon Instruments). Data were low-pass filtered with 1–10 kHz (−3 dB) and digitized at 10–100 kHz. Data analysis was performed using Clampfit 10.2 (Axon Instruments) and GraphPad Prism 5.01 (San Diego, USA) software. The cells were bathed in 140 mM NaCl, 4 mM KCl, 2 mM CaCl2, 2 mM MgCl2, 5 mM HEPES, 5 mM Glucose, pH = 7.4 (adjusted with NaOH). The pipette solution contained 10 mM NaCl, 140 mM KCl, 2 mM EGTA, 1 mM MgCl2, 0.1 mM Na-GTP, 5 mM Mg-ATP, 10 mM HEPES, pH = 7.2 (adjusted with KOH). Potassium currents were elicited by 500-ms hyper- and depolarizing voltage steps between −140 and −30 mV from a holding potential of −100 mV (see inset in Fig. 1B). Around the resting membrane potential and at more hyperpolarized voltages, the ventricular IK1 conductance is much larger than that of any other potassium current.21 In murine ventricular cardiomyocytes, currents through Kir2.1 channels represent the major component of IK1 consistent with the finding that myocytes isolated from Kir2.1−/− mice completely lacked detectable whole cell IK1 in 4 mM external potassium.19 A potential “contamination” of IK1 by other inward rectifying potassium currents (IK,ATP,34 IK,ACh,35 and IK,Ca35) in our experiments was excluded by the composition of our experimental solutions. Thus, IK,ATP was inhibited by the presence of mM concentrations of ATP in the pipette solution,36,37 and activation of IK,ACh was prevented by the lack of acetylcholine. IK,Ca activation could be excluded because calcium channels do not activate at potentials more negative than −60 mV at which significant IK1 was detectable. Further, calcium was absent from the pipette solution which additionally contained 2 mM of the calcium buffer EGTA (see above). For the determination of IK1 density-voltage relations, the current amplitudes at the end of the test pulses to various potentials were measured. These were then divided by the cell capacitance to obtain current densities. IK1 recordings from both wt and dystrophic cardiomyocytes were always started 5 min after whole cell access was attained to avoid potential artifacts due to time-dependent shifts in current properties.
|
study
| 100.0 |
The ventricles from 5 wt (3 female, 2 male), 6 mdx (3 female, 3 male), and 2 mdx-utr (both female) mouse hearts were used for the western blot experiments. Membrane proteins were isolated using the following protein isolation protocol carried out at 4°C. Frozen cardiac ventricles were cut into 15 μm slices at −20°C using a cryoslicer. Samples were weighed, blood was removed by washing the samples with ice cold PBS, and solution A (10 mM HEPES, pH = 7.4, 10% sucrose, 5 mM EDTA, 1 mM Dithiothreitol (DTT), 2 μg/ml aprotinin, 10 μg/μl leupeptin, 1 mM pefablock) was added (4 μl of solution A / mg of tissue). The tissues were homogenized on ice. After gentle centrifugation to remove bigger tissue pieces, samples were centrifuged at 1000 x g to separate the nuclei from the cytosol and the membrane protein fraction. In the next step the supernatant was centrifuged for 45 min at 100.000 x g to separate the cytosolic protein fraction from the membrane protein fraction. Membrane proteins were re-suspended in 200 μl of solution B (10 mM HEPES, pH = 7.4, 10 % sucrose, 2 mM EDTA, 1 mM DTT, 2 μg/ml aprotinin, 10 μg/μl leupeptin, 1 mM pefablock), and protein concentrations of the lysates were determined by spectrophotometry. The membrane protein lysates of equal protein concentrations were blotted on a 9% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) for 80 to 90 min at a voltage of 150 V. Separated proteins were transferred to a nitrocellulose membrane (GE Healthcare, United Kingdom) using the wet transfer method for 50 min on ice. The membrane was blocked for 2 hours at room temperature using 5% bovine serum albumin (BSA) and incubated over night with the primary antibody (anti-Kir2.1 polyclonal antibody produced in rabbit; APC-026, Alomone Labs, 1:1000 in 2% BSA) at 4°C. On the next day, after 3 PBS washing steps, the membrane was incubated with the secondary antibody (anti-rabbit IgG, horse radish peroxidase-linked antibody; 7074, Cell Signaling Technology, 1:10000) at room temperature for 60 min. The protein bands were visualized using SuperSignal Western Pico Chemiluminescent Substrate (Thermo Scientific, USA) and ECL hyperfilms (GE Healthcare, United Kingdom). The protein band intensities were measured using ImageJ software (http://rsbweb.nih.gov/ij) for western blot analysis. To subtract background signals, the specific intensity peaks were defined, and the area under the curve was normalized to the area under the curve of the loading control protein intensities. An antibody (AS7) recognizing the β1/β2 subunit of membrane bound GS proteins was used as loading control (AS7 was generated by M. Hohenegger;38 AS7 complies with the original non-specific antiserum K521, and was raised against a peptide enclosing residues 8–23 in the sequence of the beta1/beta2 subunit). Western blot experiments were performed several times per heart lysate. The obtained values for every lysate were averaged. These averaged values (resulting in n = 5 for wt, n = 6 for mdx, and n = 2 for mdx-utr) were used for the statistical analyses, whereby only wt and mdx were compared.
|
study
| 100.0 |
After the Langendorff cell isolation procedure the ventricular cardiomyocytes were plated on cover slips, and, 90 min later, fixed in 3.5% paraformaldehyde for 10 min. Cell culture medium was removed, the cells were washed 3 times with PBS, permeabilized in 0.1% Triton X-100 for 5 min at room temperature, and washed again 3 times with PBS. This was followed by a 2 hour block with 10% horse serum and 0.01% azide in PBS. Thereafter, the cells were incubated with Anti-Kir2.1 antibody (APC-026, Alomone Labs; 1:500 in PBS) at 4°C overnight. On the next day, the cells were washed 3 times with PBS, and incubated for 60 min with the corresponding secondary antibody (Alexa Fluor 594, #A21207, Invitrogen; 1:500 in PBS) at room temperature. Subsequently, after 3 more PBS washing steps, the cells were mounted, dried and stored at 4°C. The slides were finally analyzed using a LSM 510 confocal microscope (Zeiss, Jena, Germany). For the immunostaining experiments 3 wt and 3 mdx mice (15–25 week-old) were used for cell isolation.
|
study
| 100.0 |
Data represent means ± SEM. Statistical comparisons between wt, mdx, and mdx-utr were made using oneway ANOVA (for independent samples, GraphPad Prism Software, La Jolla, USA). In case only 2 groups had to be compared, an unpaired 2-tailed Student's t-test was performed. A p value < 0.05 was considered significant.
|
study
| 99.94 |
Trehalose, known as a non-reducing disaccharide of glucose, has been found in wide range of diverse organisms, for example, bacteria, fungi, invertebrates, and plants (Lyu et al., 2013). In higher plants, trehalose accumulates at quite low concentrations, with the exception of resurrection plants (Paul et al., 2008). It was found that in trehalose-rich organisms, trehalose functions both as a carbon source and as an osmoprotectant such in certain microorganisms and insects (Elbein et al., 2003; Zang et al., 2011). There is crucial evidence that trehalose helps prevent heat, osmotic, nutrient, dehydration stress, and toxic chemicals from damaging organisms (Chary et al., 2008). Trehalose synthesis in plants only occurs via the TPS/TPP pathway (Avonce et al., 2006; Jiang et al., 2010). In the first step of this pathway, trehalose-6-phosphate synthase (TPS) joins uridine diphosphoglucose with glucose 6-phosphate, thus forming trehalose 6-phosphate (T6P; Jiang et al., 2010). Then, T6P is dephosphorylated as catalyzed by trehalose-6-phosphate phosphatase (TPP) to produce the disaccharide trehalose. TPS expression levels in cotton increased under drought stress (Mu et al., 2016). When maize was under salt and temperature stress, TPS genes were upregulated (Jiang et al., 2010). Genes which controlled the production of enzymes of trehalose synthesis, including the AtTPS1 gene from Arabidopsis thaliana (Blázquez et al., 1998; Jiang et al., 2010). These findings made it practical to attain transgenic plants altered by a TPS gene of plant origin. Most transgenic plants with overexpressed TPS and/or TPP genes exhibited great tolerance to abiotic stress (Garg et al., 2002; Jang et al., 2003; Karim et al., 2007; Ge et al., 2008).
|
review
| 99.7 |
Although numerous studies found that over-expressing TPS and TPP genes could improve abiotic stress tolerance (Holmström et al., 1996; Jang et al., 2003; Schluepmann et al., 2004; Jiang et al., 2010), observed stress tolerance of transgenic lines did not correlate with trehalose amount (van Dijken et al., 2004). Researchers have investigated function of plant trehalose biosynthetic enzymes by identifying mutations in Arabidopsis TPS1 gene and seeing which were still effective in recovering homologous gene function for yeast mutants deficient in TPS functions (Blázquez et al., 1998; Müller et al., 2001; Vogel et al., 2001). Beyond these stress response, recent intriguing evidence has implicated that TPS genes are important modulators in plant development and inflorescence architecture (Satoh-Nagasawa et al., 2006).
|
review
| 99.8 |
The plant TPS gene family belongs to a small gene family with multiple copies. Members of this gene family showed widespread functional diversification (Lunn, 2007; Chary et al., 2008; Vandesteene et al., 2010; Yang et al., 2012). The Arabidopsis TPS gene family contains 11 members (AtTPS1-11; Shima et al., 2007); rice also contains 11 members (OsTPS1-11; Tuskan, 2006; Vandesteene et al., 2010), and poplar contains 12 members (PtTPS1-12; Vandesteene et al., 2010). The TPS family is divided into two distinct clades, class I and class II, and this dichotomy occurred early in the evolution of green plants (Jang et al., 2003; Shima et al., 2007; Suzuki et al., 2008; Ramon et al., 2009). Apart from polyploid and allopolyploid species, Arabidopsis is quite uncommon among all the angiosperms which have large number class I TPS genes (Delorge et al., 2015). Particularly, distinct characteristics of class I and class II TPS genes are displayed in gene expression patterns, copy number, physiological functions, and enzyme activity.
|
review
| 97.56 |
Sacred lotus (Nelumbo nucifera, Gaertn.) is a basal eudicot which is of significant cultural, agricultural, medicinal, as well as religious importance. TPS plays important role in plant response to environmental stimuli, and thus the study of TPS would be very important for lotus breeding and the research of stress-resistance mechanism in lotus. In this study, we investigated the lotus TPS genes from whole genome-wide studies and their evolutionary relationship with other TPS genes. We explored TPS gene family members' expression patterns in various tissues and in response to various stresses. Having conducted a thorough analysis of gene sequences, molecular evolution, gene structures, and gene expression patterns, we offer an effective framework for further functional characterization of lotus TPS gene families.
|
study
| 100.0 |
The lotus genome DNA data (Ming et al., 2013) was downloaded from COGE (v2, id16885, https://www.genomevolution.org/coge/) and gene prediction was performed with hidden Markov model (HMM) program (Eddy, 1998). Hidden Markov models of TPP and TPS domain were obtained from PFAM (http://pfam.xfam.org/). Motif scanning in TPS was done by PROSITE scan (Hulo et al., 2006). Gene structures of TPS genes were analyzed on the Gene Structure Display Server 2.0 (GSDS; http://gsds.cbi.pku.edu.cn/).
|
study
| 100.0 |
Multiple sequence alignments in full-length protein sequences were executed by using MUSCLE 3.52, and this was followed by manual refinement and comparisons (Edgar, 2004; Cao et al., 2014). Phylogenetic analysis was performed with MEGA 6.0 using a maximum likelihood method with the JTT (Jones, Taylor, and Thornton) amino acid substitution model. Bootstrap support values were estimated using 1000 pseudo-replicates.
|
study
| 99.94 |
Protein properties of TPS genes, e.g., the molecular weight (MW), isoelectric point (pI), and grand average of hydropathicity (GRAVY) were calculated using ProParam (http://web.expasy.org/protparam). Analysis for conserved motifs in TPS proteins was carried out using MEME (http://meme.sdsc.edu/meme/cgi-bin/meme.cgo) with the default parameters.
|
study
| 99.94 |
To analyze the functional divergence between groups and predict the amino acid residues resulting in functional difference, program DIVERGE (Gu and Velden, 2002) was applied to calculate the coefficient of type I functional divergence θI as well as that of type II functional divergence θII. The likelihood test is approximated to a χ2 distribution with 1 degree of freedom. After duplication, a significant θI means site-specific changed selective constraints (Xun, 2003; Gu, 2006; Yang et al., 2009). After duplication, a significant θII, indicates amino acid physiochemical property's obvious shifts (Xun, 2003; Gu, 2006; Yang et al., 2009).
|
study
| 100.0 |
The value of dN/dS ratio (or ω, the ratio of non-synonymous to synonymous distances) for each group was analyzed with the program codeml from PAML version 4 (Yang, 2007). A likelihood ratio test (LRT) was used to detect variation in ω among sites between M0 vs. M3 and M7 vs. M8, respectively. The χ2 statistics with degrees of freedom equal to the differences between parameters is twice the log likelihood difference between the two models in the LRT (3 for M0/M3 tests and 2 for M7/M8 tests; Mondragon-Palomino and Gaut, 2005). If the LRT was statistically significant and the evolution of genes below groups, then sites are assumed to be under positive selection pressure through Bayes methods (Yang et al., 2005).
|
study
| 100.0 |
To identify expression pattern of TPS genes, expression level in different tissues and abiotic stress treatments, transcriptome datasets of lotus was analyzed. The data sets obtained from GenBank include: rhizome apical meristem (SRS413495), rhizome elongation zone (SRS413496), rhizome internode (SRS414455, SRS414456, SRS414457, SRS414458, SRS414459), leaf (SRS413488, SRS413489, SRS413490, SRS413491, SRS413492), petiole (SRS413498, SRS413499, SRS413500, SRS413501, SRS413502), root (SRS412272, SRS412273, SRS412278, SRS412279, SRS412280). The library preparation methods and statistics of sequencing data were detailed in Kim et al. (2013). To take advantage of our in house transcriptome data, we analyzed the expression pattern of TPS genes in response to submerged and 5 mm Copper-treated conditions. Three-month-old lotus seedlings were used for submergence (24 h) and 5 mm Copper-treatment (24 h). To achieve submergence, seedlings were transferred to a plastic tank and water was added into the tank to a water depth of 20 cm from top of lotus. The sample without copper or submergence treatment was the control (Con). After the treatments, total RNA was isolated using the CTAB-LiCl method. For mRNA-seq, three sets of total RNA were used for cDNA library construction and sequencing (without replicates) at the Beijing Genomics Institute (BGI, Shenzhen, China), following the manufacture's protocols.
|
study
| 100.0 |
All SRA format data were converted into FASTQ using the SRA toolkit. All raw reads were first filtered with NGS QC Toolkit (v2.3.3) to remove adapter sequences, reads containing poly-N, and low-quality sequences (Q < 20). An index of the lotus genome was built using Bowtie 2 and paired-end or single-end clean reads were aligned to the reference genome using TopHat. The expression levels (fragments per kb per million mapped read, FPKM) from the representative transcript were determined using cufflinks program (cufflinks v2.2.142) with default settings. We used a threshold of FPKM ≥ 1 to define a gene as “expressed.”
|
study
| 99.94 |
Two month-old lotus seedlings were used for submergence treatment and copper treatment. Seedlings were completely submerged or treated with 5 mm copper for 24 h (as described above) and then rhizomes were sampled. The sample without copper or submergence treatment was the control (Con). Meanwhile, different tissues of control seedlings were sampled. Tissues were homogenized with mortar and pestle in liquid nitrogen. Total RNA was isolated using CTAB-LiCl method. About 4 μg of total RNA was reverse-transcribed using an oligo(dT) primer and SuperScript Reverse Transcriptase (Invitrogen, USA). Real-time quantitative PCR was conducted using SYBR green (TaKaRa Biotechnology) on Mastercycler ep realplex real-time PCR system (Eppendorf, Hamburg, Germany) with a final volume of 20 μl per reaction. The specific primers were listed in Table S1. The relative transcript abundance was normalized using lotus Actin gene. Gene expression in various tissues were represented by NnTPS/Actin. Fold changes of genes under copper stress and submergence treatment were values relative to control samples after normalization to Actin transcript levels. RT-qPCR data presented were the means ± SE of at least four independent experiments. Differences among treatments were analyzed by on-way ANOVA, taking P < 0.05 as significant according to multiple comparisons or t-test.
|
study
| 100.0 |
Accumulating evidence showed that TPS genes play a vital role in modulating plants response to abiotic stresses, i.e., OsTPS1 could improve rice-seed tolerance to low temperature, salt, and drought (Jiang et al., 2010; Vandesteene et al., 2010; Li et al., 2011; Zang et al., 2011). However, identity and function of lotus TPS gene family was previously undescribed. The lotus genome was sequenced recently, which makes it possible to carry out genome-wide study on TPS genes in lotus. Plant TPS proteins contain a TPS as well as a trehalose-6-phopshate phosphatase (TPP) domain (Yang et al., 2012). Thus, HMM profile of two conserved TPS domains (TPS and TPP domain) was used to search local lotus protein data with HMMER program (Sonnhammer et al., 1998; Ross et al., 2007). Nine lotus TPSs (NnTPSs) gene family members were identified in lotus genome (Table 1). Taking advantage of these identified genes, we further conducted a homology search from genome and amino acid sequences of lotus. However, the search did not yield new potentially TPS encoding genes for lotus beyond the nine initially identified. The identified NnTPSs contained a similar number of amino acid from 845 (NNU_024672) to 937 (NNU_016432). The isoelectric point (PI) ranged from 5.60 (NNU_022788) to 6.33 (NNU_004429). The protein weight ranged from 95.37 (NNU_024672) to 106.21 (NNU_016432) kDa. Nine NnTPS scattered in 7 scaffolds. Scaffold 4 and 5 had 2 NnTPS genes, respectively. Previous studies have identified 11 TPS in Arabidopsis, 11 in rice, and 12 in poplar (Yang et al., 2012), suggesting that the number of genes remains stable in this family and that usually duplicated copies of TPS genes are not retained after a whole-genome duplication event (Hao et al., 2014).
|
study
| 100.0 |
To assess the evolutionary relationship of nine NnTPS genes, a small evolutionary tree was constructed using the nine NnTPS genes to exam NnTPS gene structures which are typical imprints of evolution within some gene families. The length of gene introns was larger and genetic structures were more complicated in NNU_024672, NNU_020115, NNU_000253, and NNU_016432 than the other NnTPS genes (Figure 1). All NnTPS genes contain at least one intron in their CDSs: most of them have three introns, one (NNU_004429) has four introns and one (NNU_022788) has five introns. According to the previous reports (Nuruzzaman et al., 2010; Hu et al., 2016), the rate of intron gain is slower than that of intron loss after segmental duplication in rice. Thus, it can be concluded that NNU_022788 might be the most ancestral with the other NnTPS genes derived from it. Compared with similar CDS length (2535–2814 bp) among 9 NnTPS genes, their gene length is even more variable (3583–8291 bp).
|
study
| 100.0 |
Structure and phylogenetic analysis of NnTPS proteins. The unrooted phylogenetic tree resulting from the full-length amino acid alignment of all the NnTPS proteins is shown on the left side of the figure. Exon-intron structures of the identified NnTPS genes are shown on the right side. The graphic representation of the optimized gene model displayed using GSDS.
|
study
| 99.9 |
A homology analysis of nine full-length protein sequences indicated that nucleotide identity among them ranged from 63.66% (between NNU_016707 and NNU_022788) to 92.78% (between NNU_024672 and NNU_020115; Figure 2; Figure S1). Reflecting what was found in the protein sequences, a high similarity (ranging from 65.14 to 90.26%) was shared among nine NnTPSs full-length sequences. Several residues in the catalytic center were highly conserved in all NnTPS genes (Figure S1) suggested that corresponding genes encoding active enzymes. Nevertheless, relatively high divergence was observed in some regions of amino acid sequences outside of the domain. It is likely that these un-conserved regions might contribute to functional distinction.
|
study
| 100.0 |
Pairwise sequence identity of full-length TPS proteins. Pairwise sequence identities between TPS domain, TPP domain, full length protein sequence, and sequence outside domain were calculated. The boxplot shows the median (black line), interquartile range (box), and maximum and minimum scores (whiskers) for each data set.
|
study
| 99.94 |
Subcellular localization of NnTPS isoforms was predicted using web subcellular localization predictor server (http://cello.life.nctu.edu.tw/; Table 1). Translated proteins of NNU_016707, NNU_004429, and NNU_016432 were predicted as being located on plasma membrane while most of NnTPSs were located in cytoplasm.
|
study
| 100.0 |
The identification of motifs for all NnTPS proteins were performed by software MEME (Bailey and Elkan, 1994) with default settings. We obtained sequences of 20 motifs and the distribution of these motifs in TPS proteins (Figure 3; Table S2). These motifs were conserved in all NnTPS genes except NNU_024672 which lacked motif12. According to Figure 3, motifs 1, 2, 3, 4, 6, 9, 10, 13, 14, 15, 17, and 20 were located in the TPS domain and motifs 8 and 19 were located in TPP domain. These features in conserved motifs of NnTPS were also observed in other plants (Mu et al., 2016).
|
study
| 100.0 |
Schematic diagram of amino acid motifs of TPS protein. Motif analysis was performed using MEME 4.11.2 as described in the methods. The different-colored boxes represent different motifs and their position in each TPS sequence. The sequences of key motifs that located in TPS and TPP domain were indicated.
|
other
| 97.4 |
To explore whether selection pressure had affected NnTPS genes, we calculated ratio (ω) of the synonymous substitution rate (dS) vs. the non-synonymous substitution rate (dN) using the program codeml from PAML (version 4). Generally, dN/dS ratio >1 indicates positive selection, a ratio < 1 indicates negative or purifying selection and a ratio = 1 indicates neutral evolution (Wang et al., 2005; Song et al., 2015). Results in Figure 5 showed that dN/dS ratios were always less than 1 for different domains and regions outside domain, suggesting purifying selection on lotus TPS gene family. The dN/dS ratios outside NnTPS domain were found to be much higher than the ratios inside the NnTPS domains (Figure 4). These results showed that NnTPS domains evolved faster than regions outside TPS domain. The relaxed purifying or positive selection in the regions outside TPS domain may lead to the above results.
|
study
| 100.0 |
A phylogenetic tree was used to reveal homologous relationships and evolutionary roots of TPS from Arabidopsis, rice, poplar (Yang et al., 2012), and soybean (Yang et al., 2012). The amino acid sequence alignment of TPS in these plants was conducted for phylogenetic tree construction using MEGA 5.1 (Figure 5). According to the phylogenetic tree and previous division of Arabidopsis and rice TPS genes, these TPS genes are divided into two subfamilies (I and II; Vandesteene et al., 2010; Zang et al., 2011). Arabidopsis, rice, soybean, and poplar contain four (AtTPS1, 2, 3, and 4), one (OsTPS1), four (GmTPS1, 2, 3, and 4), and one (PtTPS1) subfamily I TPS, respectively. However, no lotus TPS subfamily I members was found. It should be noted that the absence of subfamily I TPS in lotus might be due to incomplete genome sequencing coverage (86.5%). However, we cannot exclude the possibility that these genes have been fatally lost in evolution, as the number of TPS genes in subfamily I is far less than TPS genes in subfamily II in previous reported species (Zang et al., 2011). In order to describe paralogous and orthologous relations among this family, the subfamily II TPS genes were further divided into five group (II-1, 2, 3, 4, and 5) with high bootstrap support, suggesting that genes in each subgroup might share a similar origin. In most cases, subgroups are duplicates of at least one lineage. For example, subgroup II-2 and II-3 were duplicated after the separation between dicot and monocots. We identified nodes that lead to dicots (lotus, Arabidopsis, poplar, and soybeans)-specific and monocots (rice)-specific subfamilies. Group II-2 only contained lotus, Arabidopsis, and poplar Class II TPS genes, indicating that genes might have been lost from the rice genome. Subfamily I-1 only contained three Arabidopsis Class I TPS. Moreover, most proteins in lotus TPS family were contained in paralogous pairs, including 4 pairs of NnTPS accounting for 89% NnTPSs. This result indicated that most lotus TPS genes expanded in a species-specific manner.
|
study
| 100.0 |
Functional divergence of the duplicate genes has been recognized as an important source of evolutionary novelty (Yang et al., 2012). To evaluate potential functional divergence, type I and type II functional divergence between groups of the TPS family were estimated by posterior analysis (Gu and Velden, 2002). For this analysis, the collected plant TPS proteins were used and estimation was based on multiple alignments of proteins for any two groups (Table 2). Previous work proposed several possible evolutionary fates of duplicate genes (Hughes, 1994; Force et al., 1999). Our results showed that most type I coefficients (θI) of functional divergence were significantly greater than zero (P < 0.01), while few of the type II coefficients (θII) was statistically greater than zero. These results suggested that type I functional divergence was the dominant pattern for evolution of TPS family in these plants, and significantly site-specific altered selective constraints should contribute to most of the TPS members which would, after the diversification, cause a group-specific functional evolution.
|
study
| 100.0 |
Further, the calculation of site-specific profiles on the basis of posterior analysis for all group pairs with functional divergence would predict the critical amino acid residues, which are accountable for the functional divergence. Among aligned sites, most of them had low posterior probabilities. Qk ≥ 0.80 and Qk ≥ 0.95 were utilized empirically as cutoffs to determine type I functional divergence-related residues existing between groups, so as to greatly reduce false positive (Wang et al., 2011). For most of pairs of groups, at least one site had the posterior probability higher than 0.8, and four pairs of groups had one site with posterior probability higher than 0.95 (Figure 6). A cut-off value Q(k) ≥ 0.95 was used for estimated key amino acid residues. The number of amino acid residues that presumably contributed to alter functional constraints was small between TPS groups. Three amino acid residues (669, 704, and 753) in all comparisons were identified as being most important for the functional divergence. These amino acids are all localized in the TPP domain.
|
study
| 100.0 |
To examine if selective pressure between the seven groups of TPSs genes would change, the likelihood ratio tests of positive selection were used via the ML methods as well as codon substitution models (Yang, 1998; Yang and Bielawski, 2000; Yang et al., 2012). Two tests were performed during the analysis process. First, models M0 and M3 were compared in order to assess whether dN/dS ratio variations existed among codon positions below each group. On the whole, the maximum likelihood estimates of dN/dS values as for groups under model M0 were close to zero, with three exceptions of Arabidopsis TPS genes in group I-2. This implied that the principal force which acting on the evolution of TPS family was purifying selection (Table 3). It is obvious that although the majority of protein residues are heavily dependent on constant purifying selection, some sites are also being impacted by positive selection. Positive selection is an important adaption mechanism; thus, sites under positive selection pressure may have sped up functional divergence of TPS; in this case it would allow the plants to adjust to its environment. The results we get are in accordance with this conclusion. As the statistical differences of log likelihood between model M0 and M3 were significant for all groups, it suggests that the overall selective constraint levels varied across the TPS family group lineages. Secondly, the LRT which is applied to compare data fit to models M7 vs. M8 was used to determine whether positive selection promoted divergence of TPS family below groups or not. Two groups (group I-1 and II-2) of the 7 groups analyzed, were believed to have undergone positive selection, because they satisfied that (1) an estimate of ω > 1 under M8, (2) sites identified to be under positive selection, and (3) a significant LRT. This result suggests that positive selection contributed to the evolution of genes in these groups. When the LRT suggested positive selective action had occurred, positively selected sites are identified under model M8 using Bayesian method (Nielsen and Yang, 1998; Yang and Bielawski, 2000). We found 31 and 4 positively selected sites in subgroup I-1 and II-2, respectively.
|
study
| 100.0 |
TPS genes are known to be involved in development and stress responses. In this study, expression pattern of TPS genes in different lotus tissues has been studied using transcriptome data set of lotus (available at NCBI; Kim et al., 2013) as well as our in house transcriptome dataset of lotus under submergence and copper stress. NnTPS genes showed differential expression level in various lotus tissues (Figure 7A). No tissue-specific expressed NnTPS genes were found. Among them, NNU_016432 showed the highest expression level while NNU_004429 showed the lowest transcript level in most examined tissues, in comparisons with other NnTPS genes. Genes from the same group frequently showed similar expression pattern within specific tissues. For example, both NNU_000253 and NNU_016432 had relatively low expression in leaf and rhizome internode, and high expression in root, rhizome apical tip, and rhizome elongation zone. Based on FPKM calculations, the total transcript abundance of NnTPS genes were obviously low in leaf and petiole and high in root and rhizome. These results indicated that NnTPS genes might primarily have functions in sink tissues.
|
study
| 100.0 |
Expression profiles of the nine NnTPS genes in different tissues (A) and upon submergence and copper stresses (B). The color scale represents RPKM normalized log2 transformed data. Red indicates high expression level and blue indicate low expression level. OL, leaf; RTZ, Combined Rhiz Tip Zone; Rhiz, Rhizome internode; R, Root; P, Petiole; T, Rhizome apical tip; Z, Rhizome elongation zone.
|
study
| 99.94 |
In response to submergence and copper stress, the total transcript abundance of NnTPS did not show apparent difference (Figure 7B). Two NnTPS genes (NNU_0044229 and NNU_016432) were highly induced by submergence treatment. In contrast, most NnTPS genes (i.e., NNU_024672, NNU_014679, and NNU_022788) were down-regulated. The abundance of NNU_024672 decreased over 80%. Submergence leads to lower concentration of oxygen in a plant as a result of a lower diffusion rate of molecular oxygen in water as opposed to air, which would ultimately influence energy metabolism. The reaction of NNU_024672 to low oxygen demonstrates possible involvement in energy metabolism in the seedlings and that highly expressed NnTPS genes could play vital roles during this complicated biological process. Some plant metabolic processes contain copper as an essential micronutrient, functioning as a constituent of enzymes and other proteins, and in photosynthetic reactions (Maksymiec, 1998; Hall, 2002; Yruela, 2009; Martins et al., 2014). Excessive copper can lead to oxidative stress and be phytotoxic (Mithöfer et al., 2004). It can inhibit growth, influence plant metabolism and affect the function of several enzymes (Martins and Mourato, 2006; Mourato et al., 2009). Under copper stress, the transcript abundance of most NnTPS genes was not obviously changed, except two NnTPS genes (NNU_014679 and NNU_022788). It is interesting that abundance of two genes were apparently decreased in response to submergence while increased upon copper stress. We speculate that under copper stress more energy was needed to strengthen antioxidant system, while more energy was needed to be preserved under submergence as photosynthesis is inhibited. The contrasting expression pattern under submergence and copper stress indicated that NNU_014679 and NNU_022788 might play important roles in lotus energy metabolism and participate in stress response. Similar to previous work, TPS has been reported to play an important role in the regulation of carbohydrate metabolism and especially the perception of carbohydrate availability (Müller et al., 2001).
|
study
| 99.94 |
In silico analysis showed that NnTPS genes exhibited differential abundance in various tissues and some of them are obviously regulated by environmental stimuli. Some of these obviously changed genes were then validated by RT-qPCR. As shown in Figure 8, RT-qPCR results of genes under various tissues and stresses were similar in magnitude to those obtained by deep sequencing. NNU_020889, NNU_014679, and NNU_022788 were the most abundant genes in rhizome. It is interesting, that while under submergence, NNU_020889 was significantly induced while NNU_000788 was significantly down-regulated. Moreover, NNU_024672, which has the minimum transcript level were further down-regulated upon submergence.
|
study
| 100.0 |
Validation of expression of selected NnTPS genes in various tissues and in rhizome upon exogenous stimuli. Tissues including leaf, petiole, and rhizome were sampled from 2 month old lotus seedlings. Meanwhile, lotus seedlings were treated with submergence stress or copper stress for 24 h and then rhizome was sampled. The sample without 5 mm copper treatment or submergence treatment was the control (Con). Gene expression was analyzed by real-time RT-PCR. The relative transcript abundance was normalized using lotus actin gene. Values are means ± SE of at least four independent experiments. Bars with different letters are significantly different at P < 0.05 according to Turkey's multiple range test. Fold changes of genes under copper stress and submergence treatment were values relative to control samples. Asterisks indicate significant differences compared with control sample at P < 0.05 levels according to t-test.
|
study
| 100.0 |
Collectively, our results characterize the structure, evolutionary, and some functional features of the lotus TPS gene family. Based on systematic analysis of lotus genomes, we identified 9 TPS genes. Gene structure analysis showed that NNU_022788 might be the ancestral member, from which other genes were derived. Regions outside the TPS domain in nine NnTPS genes evolved faster than regions within the TPS domains. Based on phylogenetic tree constructed using TPS genes from lotus, Arabidopsis, polar, soybean, and rice, TPS genes could be classified into two main subfamilies (I-II). Lotus TPS genes showed species-specific expanded manner. Three amino acid residues (669, 704, and 753) in TPP domain were identified as the most important residues for the functional divergence of these TPS genes. Moreover, 2 groups (group I-1 and II-2) were believed to have undergone positive selection based on likelihood ratio test. Expression pattern analysis showed that NnTPS genes might mainly have function in sink tissues and two NnTPS genes (including NNU_014679 and NNU_022788) might play important roles in lotus energy metabolism and stress responses.
|
study
| 100.0 |
This work was supported by Funding of agricultural science and technology innovation of Jiangsu Province, China (CX(15)1030), National Natural Science Foundation of China (31501795, 31400600), the China Postdoctoral Science Foundation funded project (2014M560432, 2015T80563), the Natural Science Foundation of Jiangsu Province in China (BK20151229, BK20140695), and the fundamental research funds for the central universities (KJQN201659).
|
other
| 99.94 |
Macrophages play a significant role in the initiation and perpetuation of innate and adaptive immune responses. In this role, they perform multiple functions, including the uptake, processing and presentation of antigen, microbial killing, phagocytosis of apoptotic cells, and the secretion of cytokines, chemokines and chemical mediators . The expression and secretion of immune modulators imparts the macrophage the ability to orchestrate immune responses, skewing them toward the pro-inflammatory, anti-inflammatory or regulatory arms of immunity. While pro-inflammatory responses are crucial to the elimination of pathogens, they are also central to the pathogenesis of autoimmune (for example type 1 diabetes, multiple sclerosis and rheumatoid arthritis) and pro-inflammatory (for example sepsis) diseases. Accordingly, much research has focused on understanding the responses of macrophages to pro-inflammatory conditions. These investigations often use murine bone marrow derived macrophages (BMDM), as a model mammalian macrophage system [2–4]. This is because BMDMs exhibit phenotypic and functional homogeneity and closely resemble ex vivo primary cells, thereby making them the preferable model, as opposed to other sources of primary macrophages or cell lines. Immortalised cell lines differ significantly in phenotype/function to primary murine macrophages. Primary resident macrophages of the peritoneum or lung are not naïve, and, therefore, differ phenotypically/functionally according to their previous immune experiences. Moreover, these cells are obtained in lower numbers, as compared to yields derived from bone marrow.
|
review
| 99.9 |
The pro-inflammatory response of macrophages is commonly investigated using lipopolysaccharide (LPS; a major component of bacterial membranes), as a biologically relevant inducer of inflammation [5–7]. The progression of this pro-inflammatory response in macrophages has been well characterised, with an initial induction phase of inflammation at 2 h following LPS exposure, a peak inflammatory phase at 6 h, and, finally, a resolution phase at 24 h post-stimulation [8–10]. The analysis of gene expression levels by RT-qPCR is often employed to study this pro-inflammatory response [11–13]. As a powerful tool, RT-qPCR is a rapid and sensitive technique with potential for high throughput, which allows the detection and quantitation of low abundance mRNA . Despite these advantages, the accuracy and reliability of RT-qPCR data interpretation is dependent upon factors intrinsic to the preparation steps prior to RT-qPCR analysis, including RNA quality and quantity and reverse transcriptase (RT) efficiency. Indeed, it is crucial that the quantity of RNA input to the RT reaction be normalised. However, while necessary, this is not sufficient for direct comparisons of RT-qPCR data. A reference/normalisation gene(s), whose expression level is not regulated by the specific experimental conditions, is routinely included in RT-qPCR for all samples, thereby enabling normalisation of expression levels of the gene of interest (GOI) data [15, 16]. It is recommended that for each set of experimental conditions, the optimal reference gene(s) be determined . However, in reality, few studies validate this important optimisation parameter. Rather, “traditional” reference genes, such as glyceraldehyde 3-phosphate dehydrogenase (Gapdh) or β-2 microglobulin (B2m), that are presumed to be stable in their expression levels, are generally selected as reference genes, despite the fact that they may not be expressed at consistent levels under the specific experimental conditions being studied. Normalisation of target gene expression levels to those of a reference gene which is, itself, regulated by the experimental conditions, likely affects measurements of comparative gene expression levels, thereby compromising data interpretation and biological conclusions made [17–19]. Thus, it is crucial that for any given experimental condition, one or more consistently expressed reference genes are identified and used .
|
review
| 99.0 |
To date, no optimal reference gene (or combination of reference genes) has been identified and validated for the normalisation of gene expression levels between control and LPS stimulated BMDMs over the time course of the pro-inflammatory response. Thus, the current study aimed to identify the most stably expressed reference genes during the peak (6 h) and resolution (24 h) phases of inflammation. A list of eight genes that have been used commonly for the normalisation of gene expression levels in BMDMs was initially investigated. From microarray analyses of untreated and LPS treated BMDMs at 6 h and 24 h post-LPS stimulation, three additional candidate reference genes, whose expression remained unchanged under these experimental conditions, were added to the list of commonly used reference genes. Thus, the expression levels of a total of 11 candidate reference genes were compared at each time point following LPS stimulation, and the optimal reference gene, or combination of reference genes, for each phase of the pro-inflammatory response was identified using NormFinder , GeNorm and BestKeeper softwares.
|
study
| 100.0 |
This study is the first to identify and validate Hnrnpab and Stx5a, as optimal reference genes for the normalisation of gene expression data during the peak and resolution phases, respectively, of the BMDM response to LPS. Moreover, this study demonstrates the consistency in expression levels of both Hnrnpab and Stx5a for peritoneal and RAW 264.7 macrophages stimulated with LPS. Importantly, the expression levels of these genes are more stable than those of Gapdh and Actinb, which are both commonly regarded as reference genes in the assessment of macrophage inflammatory responses. These observations demonstrate the importance of assessing the stability of reference genes for every experimental condition prior to normalisation of gene expression.
|
study
| 100.0 |
The aim of the current study was to identify the optimal reference gene, or combination of reference genes, for the normalisation of gene expression data in the commonly studied LPS stimulation model using BMDMs. We first identified a list of eight genes routinely used for the normalisation of gene expression data from experiments using BMDMs in general [23–25] or LPS stimulated BMDMs [26–35]. These candidate genes were Actinb, B2m, Gapdh, Gusb, Hmbs, Hprt, Ppia (cyclophilin A), and Rpl13a. We then identified candidate reference genes from a microarray data set comparing gene expression levels between control and LPS stimulated BMDMs at 6 h and 24 h post-exposure to LPS (Additional file 1: Table S1 and Additional file 1: Table S2, respectively). We selected three candidate reference genes whose expression remained unchanged between control and LPS-stimulated cells at both 6 h and 24 h, including chromatid cohesion factor homolog (Mau2; a gene involved in cell cycle), heterogenous nuclear ribonucleoprotein A/B (Hnrnpab; involved in mRNA processing) and Syntaxin 5a (Stx5a; a gene involved in autophagy; ). Thus, a total of eleven candidate reference genes were tested for the stability of their expression levels. The range of Ct values for untreated and LPS treated BMDMs at each time point, for each gene tested, are described in Fig. 1. The level of expression of genes differed, with some being highly expressed, for example, Actinb (mean (standard deviation [SD]): 6 h control; 18.34(0.18)), while others were expressed at much lower levels, for example, Hmbs (mean (SD): 6 h control; 27.73(0.81)).Fig. 1Ranges of Ct values of the 11 pre-selected reference genes in control and LPS stimulated BMDMs at 6 h and 24 h. Ct values were recorded for control and LPS stimulated (10 ng/ml) BMDMs at (a) 6 h and (b). 24 h. Plotted as boxes are the range of Ct values, with the included horizontal line identifying the mean, of triplicate biological replicates. The unfilled boxes represent control BMDMs, and the grey filled boxes represent LPS stimulated BMDMs
|
study
| 100.0 |
Ranges of Ct values of the 11 pre-selected reference genes in control and LPS stimulated BMDMs at 6 h and 24 h. Ct values were recorded for control and LPS stimulated (10 ng/ml) BMDMs at (a) 6 h and (b). 24 h. Plotted as boxes are the range of Ct values, with the included horizontal line identifying the mean, of triplicate biological replicates. The unfilled boxes represent control BMDMs, and the grey filled boxes represent LPS stimulated BMDMs
|
study
| 100.0 |
The variance in expression levels of candidate reference genes, as indicated by the range of Ct values and SD, differed within a single treatment group. For example, within the control group at 6 h, the lowest variance in Ct values, as indicated by the lowest SD, was Actinb (SD = 0.18; Fig. 1a). The highest variance within the untreated group was for Mau2 (SD = 1.06; Fig. 1a). Interestingly, in some cases, genes differed in their expression variability, dependent upon the time point following LPS stimulation. For example, in the control group, at 6 h, the variance in Ct values between replicates recorded for the gene Rpl13a was larger (range(SD) = 26.15–27.96(0.91)) than the variance observed for Rpl13a at 24 h post-LPS stimulation (range(SD) = 25.86–26.17(0.17)).
|
study
| 100.0 |
In order to identify the most consistently expressed reference gene candidate at each time point, the Ct values were inputted into GeNorm, NormFinder and BestKeeper programs. The analysis provided by NormFinder assigns a stability ranking to candidate reference genes using an algorithm which takes into account intra- and inter-variability (i.e. the variability in expression levels within and between the treatment groups, respectively) . GeNorm determines the expression stability of a gene using a stepwise exclusion of the least stably expressed gene, generating an M value for each reference gene, and M values are then ranked. The Bestkeeper software assigns a ranking to each candidate reference gene based on the SD between samples. The rankings provided by each software for samples at 6 h and 24 h after LPS or vehicle exposure are described in Figs. 2a and b, respectively. There was broad agreement between the three programs regarding the two most and least consistently expressed reference genes between control and LPS stimulated groups at 6 h. At this time point, according to NormFinder and GeNorm, the most stable reference genes were Hnrnpab and Stx5a. Similarly, BestKeeper ranked Hnrnpab and Stx5a in the top four most stably expressed genes, at second and fourth, respectively. Application of all three softwares indicated that the two least stable genes were Gusb and Hmbs.Fig. 2Stability ranking of reference genes in control and LPS stimulated BMDMs at 6 h and 24 h by Normfinder, GeNorm and BestKeeper softwares. Ct values were recorded for control and LPS stimulated (10 ng/ml) BMDMs at 6 h and 24 h. Ct values were transformed as instructed and applied to each reference gene analysis software. The tables above show the ranking of most to least stably expressed reference genes between control and LPS stimulated cells, from top to bottom, as identified by Normfinder, GeNorm and BestKeeper softwares, at (a) 6 h and (b). 24 h. The most and least stable reference genes were identified by NormFinder from an initial data set at 6 h and 24 h. The stability in expression levels of these selected genes were then analysed in an independent experiment by RT-qPCR and Ct values are shown at 6 h (c) and 24 h (d). The error bars represent means ± SD. The significance values were calculated by comparison of control and LPS treated samples as a group, at each time point. *p < 0.05, **p < 0.01
|
study
| 100.0 |
Stability ranking of reference genes in control and LPS stimulated BMDMs at 6 h and 24 h by Normfinder, GeNorm and BestKeeper softwares. Ct values were recorded for control and LPS stimulated (10 ng/ml) BMDMs at 6 h and 24 h. Ct values were transformed as instructed and applied to each reference gene analysis software. The tables above show the ranking of most to least stably expressed reference genes between control and LPS stimulated cells, from top to bottom, as identified by Normfinder, GeNorm and BestKeeper softwares, at (a) 6 h and (b). 24 h. The most and least stable reference genes were identified by NormFinder from an initial data set at 6 h and 24 h. The stability in expression levels of these selected genes were then analysed in an independent experiment by RT-qPCR and Ct values are shown at 6 h (c) and 24 h (d). The error bars represent means ± SD. The significance values were calculated by comparison of control and LPS treated samples as a group, at each time point. *p < 0.05, **p < 0.01
|
study
| 100.0 |
After 24 h LPS stimulation, there was less consensus among the softwares regarding the most reliable reference gene(s). NormFinder and BestKeeper were in agreement, ranking Stx5a and Hnrnpab as the two most stably expressed genes. GeNorm, however, identified Hmbs and Rpl13a as the most appropriate combination of genes for normalisation. In comparison, NormFinder and BestKeeper ranked Hmbs and Rpl13a seventh and eighth in stability, respectively. This discrepancy is likely due to the difference in the mathematical algorithms used to rank genes by NormFinder and BestKeeper, as compared to GeNorm. GeNorm assumes that two reference genes are not co-regulated. If they are co-regulated, they would score artificially high on the GeNorm ranking scale . Indeed, the Ct values recorded for Hmbs and Rpl13a are both regulated in the same direction, in the LPS treated group, as compared to controls, that is, expression of each gene is lower in the LPS treated group, as compared to the controls, evident as a significantly higher Ct value (mean(SD); Hmbs: control: 28.7(0.2), LPS: 29.3(0.1), p = 0.0084; Rpl13a: control: 26.0(0.2), LPS: 26.6(0.1), p = 0.0078; Fig. 1b). The least stable reference genes identified consistently between the three softwares were Gusb and B2m.
|
study
| 100.0 |
To validate these findings, an independent experiment was performed, and the expression of the most and least stable reference genes, as identified by NormFinder software, was analysed by RT-qPCR. In agreement with the findings described above, there was no difference in Ct values observed for Hnrnpab, between the control and LPS treated samples, at 6 h (Fig. 2c). In the above-mentioned analysis, Gusb was identified as the most unstable reference gene under these particular experimental conditions, and was clearly regulated by LPS exposure at 6 h. In agreement with these findings, the Ct values observed for Gusb were significantly higher following LPS exposure at 6 h (p = 0.0304), indicating that LPS down-regulated Gusb gene expression levels. After 24 h exposure to LPS, Stx5a was identified as the most stably expressed gene between control and treated cells. This was validated in the independent experiment, with no difference seen in Ct values between the two treatments (Fig. 2d). In agreement with the above-mentioned analysis, Ct values observed for B2m were significantly lower in the LPS treated BMDMs, as compared to controls (p = 0.0034), indicating that LPS up-regulated B2m expression at this time point.
|
study
| 100.0 |
To investigate the applicability of both Hnrnpab and Stx5a as reference genes in differently sourced macrophages, we compared the expression levels of six of the 11 candidate genes (three of the most stable and three of the least stable) in murine peritoneal macrophages and in the RAW 264.7 murine macrophage cell line. For all genes the Ct values were recorded for control and LPS stimulated (10 ng/ml) cells at 6 h (Figs. 3a-d) and 24 h (Figs. 3e-h). Consistent with the stability of genes in LPS stimulated BMDMs, there was no difference in Ct values observed for Hnrnpab, between the control and LPS treated cohorts, at 6 h. In addition, Stx5a was again identified as the most stably expressed gene between control and treated cells 24 h after LPS treatment, with no difference seen in Ct values between the two treatments. This additional study therefore validates Hnrnpab and Stx5a as suitable reference genes in these multiple sources of macrophages and at multiple time points.Fig. 3Stability ranking of select reference genes in LPS treated RAW 264.7 cells and peritoneal macrophages at 6 h and 24 h. Ct values were recorded for control and LPS (10 ng/ml) stimulated RAW 264.7 cells and peritoneal macrophages at 6 h (a-d) and 24 h (e-h). Ct values were transformed as instructed and applied to the 3 reference gene analysis softwares. The tables above show the ranking of most to least stably expressed reference genes between control and LPS stimulated cells in (a, e). RAW264.7 cells and (b, g) peritoneal macrophages, from top to bottom, as identified by Normfinder, GeNorm and BestKeeper softwares. The Ct values are plotted for LPS stimulated (c, f). RAW 264.7 cells and (d, h) peritoneal macrophages. The error bars represent means ± SDs. The significance values were calculated by comparison of control and treated samples as a group, at each time point. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001
|
study
| 100.0 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.