text
string | predicted_class
string | confidence
float16 |
|---|---|---|
Physicians were identified in publicly available certification agency directories, and hospital or university websites, and invited to participate by regular or electronic mail. They were purposively recruited by specialty (cardiac or vascular surgery, interventional cardiology and orthopaedic surgery), region (Canadian provinces), setting (academic, community) and years in practice (self-reported early, mid or late career). A reminder was sent to non-respondents at 2 and 4 weeks from initial contact. Sampling aimed to interview 10 of each specialty who varied by other sampling characteristics. Of 561 physicians invited to participate, 534 declined or did not respond and 27 consented; we were able to schedule interviews with 22 (Table 1). Table 1Demographic characteristics of interview participantsPhysician specialtySelf-reported career stageSubtotalEarlyMidLateOrthopaedic surgeons10OCE-MB11OTE-MB14OTE-AB15OTE-NS16OTE-NS17OTE-NS06OTM-ON08OTM-MB12OCM-BC03OTL-ON07OTL-ON09OCL-MB12Cardiac or vascular surgeon, or interventional cardiologists02CTE-ON04CTE-ON01CTM-ON05CTM-ON13CTM-MB19CTM-ON20CTM-AB21CTM-ON22CTM-MB18CTL-ON10Subtotal810422C, cardiac; O, orthopaedic; T, teaching; C, community; E, early career; M, mid-career; L, late career; two letter code for province.
|
study
| 100.0 |
Telephone interviews were conducted by the principal investigator, a PhD trained Scientist with many years of experience in qualitative research, between 8 April and 28 September 2015. During the interview, participants were asked about factors that influenced their choice of medical devices and their response to adverse medical device events (findings are reported elsewhere). To address the main purpose of this study and elicit views about PE, participants were asked: ‘What is the role of patients in deciding which type of device to use?’ This was distinguished from telling patients about device risks and benefits during the required informed consent process, and described as having a discussion with patients about device options, characteristics and performance. Interviews were audio-recorded and professionally transcribed. Interview length was an average of 30.6 min (median 30.5, range 22.0–45.0 min).
|
study
| 100.0 |
Analysis was concurrent with data collection, and proceeded until no further unique themes emerged from successive interviews (thematic saturation). Data were organized using Microsoft Office software. The principal investigator identified unique themes using constant comparative technique . First, interview transcripts were read to identify, define and organize themes in participant responses relevant to the main interview questions (first-level coding). Second, a codebook was developed to organize codes reflecting emerging themes, their definition and sample quotes illustrating application of that code. Third, transcripts were reviewed to assess whether and how to expand or merge themes (second-level coding). Saturation was determined through discussion of emerging themes among all members of the research team on three occasions during the iterative data analysis process until it was deemed by consensus that the most recent interviews produced consistent information.
|
study
| 99.94 |
Following qualitative analysis, themes were interpreted in several ways. Level of PE articulated by participants was described according to the Carman et al. Multidimensional Framework for Patient and Family Engagement in Health and Health Care as either consultation (patients receive information about their diagnosis and/or treatment), involvement (patients are asked about their treatment preferences) or partnership (treatment decisions are based on patient preferences, evidence and clinical judgement). For example, if participants said that patients were not involved in decisions because it was the physician responsibility to do so, that was categorized as ‘consulting’ with patients. PE processes were described based on the Prokopetz et al. commentary about medical devices and shared decision-making. The commentary proposed that it was reasonable for physicians to choose the device best suited to patients, but recommended that they engage patients by providing a rationale for the implant chosen, discussing available evidence in support of the device, disclosing relevant financial relationships, eliciting patient concerns and expectations, and confirming patient understanding . For example, if participants said that a barrier of involvement in decision-making was that patients did not have the capacity to understand technical information, this idea was mapped to the Prokopetz et al. recommendation to confirm patient understanding . Feasibility of PE, defined by Grande et al. as multi-level factors influencing PE , were described based on determinants that emerged from the data. For example, if physicians said that they preferred to use familiar devices on which they were trained, this was mapped to physician factors that constrained PE; and if they said that they used specific devices to fulfil purchasing group contracts, this was mapped to health system factors that constrained PE. Then, multi-level factors influencing PE which emerged from qualitative interviews were tabulated with the corresponding mapped Prokopetz et al. recommendations to generate a framework by which physicians can engage patients in decisions about medical devices. All members of the research team met again to review and finalize the interpretation of data.
|
study
| 99.94 |
Twenty-two physicians who implanted cardiovascular (pacemakers, ICDs and stents) and orthopaedic devices (largely hip/knee implants) were interviewed (Table 1). These included 8, 10 and 4 early, mid and late career physicians, respectively, from 5 provinces, and each was from a different hospital. Supplementary Table 1 presents all themes and exemplar quotes. Select quotes are discussed here to illustrate themes, discrepant views and participant characteristics associated with discrepant views.
|
study
| 99.94 |
Most participants favoured ‘consulting’ patients, described as informing patients about the device that they had already decided to use. One participant said: ‘There is a lot of variability in what different physicians would deem as acceptable consent’ (02CTE), suggesting that patients may receive inconsistent information about medical devices.
|
other
| 99.9 |
Some participants said that patients could be ‘involved’ in particular decisions, for example, choosing a category of device (e.g. tissue or mechanical for cardiovascular, metal or plastic bearing surface for orthopaedic) but not a specific device from within that category.
|
other
| 99.9 |
No physicians explicitly stated that they discussed conflicts of interest, evidence for the chosen device, queried patients about concerns or expectations, or confirmed patient understanding. Several participants said that they provided patients with a rationale for choosing a particular device, and discussed the risks and benefits associated with that choice as part of the informed consent process. I give them a three-page handout that talks in generic terms about the indications, the complications, the success rate, the failure rate, the recovery period. I also say it to them verbally. I tell them that if they have questions let me know. (03OTL)You need to talk to the patient about what the device is supposed to do, how it is going to be implanted, what risks are there, what potential benefits are there, and all these are outlined in the informed consent as well as verbal discussion. (18CTL)
|
other
| 99.9 |
Many interacting patient, physician, health system and device/device market factors were said to influence PE. Factors were largely common for disparate types of devices and among physicians with different characteristics. These are summarized in Table 2 along with corresponding recommendations derived from Prokopetz et al. for engaging patients in PE . The summary presented in Table 2 provides physicians with a framework of topics by which to engage patients in discussions related to implantable medical devices. Table 2Factors constraining and enabling PE in decision-making about medical devicesCategoryConstraining factorsEnabling factorsaPatientBest fit for physical and demographic characteristicsPrognosis (life or death scenario)Age (device longevity greater than expected patient lifespan)Individual desire for PE (most prefer to let physicians decide)Capacity to understand complex, technical informationWell informed about manufacturers/devicesExplain why a particular implant is recommendedSolicit patient values and preferences regarding unknown risksProbe for other patient concerns and expectationsConfirm patient understandingPhysicianFamiliarity/comfort with specific device due to training/experienceTime required to educate patientsDisclose relevant financial relationshipsHealth systemFulfilment of purchasing group contractsUse of least costly device for same indicationRefer patient to another physician who uses a device preferred by the patientDevice or device marketComparative advantage of different devices for same indicationNumber of different devices available for the same indicationDiscuss available evidence in support of the deviceUse lay language to distinguish design features and trade-offs between different devicesaConstraining factors were mapped to relevant strategies (enabling factors) for shared decision-making in relation to medical devices that were recommended by Prokopetz et al. .
|
review
| 99.9 |
Some participants said that PE was not pertinent among patients facing death who had no option other than an implantable cardiovascular device, or among older patients for whom orthopaedic devices were likely to last through their lifespan. Several participants said that device selection was largely based on patient physiology and demographic characteristics, and that only the physician could assess these clinical factors to choose the device best suited to each patient.
|
other
| 99.9 |
Many participants questioned patient ability to understand technical information about devices, and said that patients generally wanted physicians to make such decisions. A few acknowledged that this view may be regarded as paternalistic, but emphasized that physicians must be trusted by patients to make decisions in their behalf. At a certain point it becomes absurd. Are we gonna have to discuss what suture material we use? And why we're using that vendor? The average person is not interested. It's just too heavy for them to grasp. Maybe I'm very paternalistic. I don't think I am. An overarching policy of detailed descriptions of different technologies and why we might use one over the other—I'm really not sure that it's relevant. (21CTM)Newspapers assume that people can read a grade 5 or 6 level. I think that the general population, to be quite frank, is not smart enough to engage in that discussion. Physicians quite frankly don't have the time to educate people, even in the basics that they would have to know. That sounds really paternalistic, and I should probably apologize for that, but I just can't see that as being workable. That is one of those areas where providers have to make decisions on behalf of their patients, and the patients have to trust their providers to make those decisions in good faith. (06OTM)
|
other
| 99.9 |
At a certain point it becomes absurd. Are we gonna have to discuss what suture material we use? And why we're using that vendor? The average person is not interested. It's just too heavy for them to grasp. Maybe I'm very paternalistic. I don't think I am. An overarching policy of detailed descriptions of different technologies and why we might use one over the other—I'm really not sure that it's relevant. (21CTM)
|
other
| 99.94 |
Newspapers assume that people can read a grade 5 or 6 level. I think that the general population, to be quite frank, is not smart enough to engage in that discussion. Physicians quite frankly don't have the time to educate people, even in the basics that they would have to know. That sounds really paternalistic, and I should probably apologize for that, but I just can't see that as being workable. That is one of those areas where providers have to make decisions on behalf of their patients, and the patients have to trust their providers to make those decisions in good faith. (06OTM)
|
other
| 99.94 |
These views contrasted with others who said that patients are increasingly well informed, sometimes even more so than physicians. Patients have researched it a lot on the internet and they know what companies have had issues. I think it's a problem now that there's so much information about these implants out there that the patients can tell me more than I know what instrument companies are good and which are not. (07OTL)
|
other
| 99.9 |
Patients have researched it a lot on the internet and they know what companies have had issues. I think it's a problem now that there's so much information about these implants out there that the patients can tell me more than I know what instrument companies are good and which are not. (07OTL)
|
other
| 99.94 |
Several participants said that they primarily use devices familiar to them, largely based on their training, to achieve optimal outcomes. Some differing views were expressed. Some participants who use orthopaedic implants said that proficiency with many devices was needed to best meet patients’ clinical needs. Some people treat everything with one system and some other people treat tailored to the patient. It depends on your philosophy and your training. I'm one of those people that tailor it so I try to look at the patient's issues and find the best solution available. (10OCE)Some people may feel they're only comfortable sticking with one. But being an arthroplastic surgeon is complex. You need a variety, and there are benefits and down sides to every single implant in terms of correcting for deformities, problems, variations on normal anatomy. In order to give the best outcome for patients, the one or two implants that you're comfortable with may not correct those issues and that's why I feel the need to use a wide variety of implants. (14OTE)
|
other
| 97.94 |
Some people treat everything with one system and some other people treat tailored to the patient. It depends on your philosophy and your training. I'm one of those people that tailor it so I try to look at the patient's issues and find the best solution available. (10OCE)
|
other
| 99.94 |
Some people may feel they're only comfortable sticking with one. But being an arthroplastic surgeon is complex. You need a variety, and there are benefits and down sides to every single implant in terms of correcting for deformities, problems, variations on normal anatomy. In order to give the best outcome for patients, the one or two implants that you're comfortable with may not correct those issues and that's why I feel the need to use a wide variety of implants. (14OTE)
|
other
| 99.9 |
Choice of device was often limited to what was approved for use by purchasing groups at the hospital, regional or provincial level. There are contractual obligations that would make me try one device more than another. In cases where I can use multiple devices then I would try and fulfill my contractual obligations. (13CTM)
|
other
| 99.94 |
Contractual limitations were viewed as a cost-saving measure that was not necessarily in the best interest of patients. One of the biggest deciding factors will be cost and not necessarily surgeon comfort, patient anatomy and track record of implant. We've had experience that if you force surgeons to change implants based on a contract that your complication rate goes up for a while. That is problematic when it occurs. So it makes good business sense until you actually go and look at your revision costs over the next months to two years and then, all of a sudden, all of your cost-savings went into pain and suffering of patients and their subsequent care. (08OTM)
|
other
| 99.9 |
One of the biggest deciding factors will be cost and not necessarily surgeon comfort, patient anatomy and track record of implant. We've had experience that if you force surgeons to change implants based on a contract that your complication rate goes up for a while. That is problematic when it occurs. So it makes good business sense until you actually go and look at your revision costs over the next months to two years and then, all of a sudden, all of your cost-savings went into pain and suffering of patients and their subsequent care. (08OTM)
|
other
| 99.94 |
Several participants said that devices were largely interchangeable and a less expensive version of the device was often sufficient, obviating the need for PE. Surgeons and physicians need to be conscientious about the finances in our health care, you can't be implanting the best of the best in every single person. We have to be selective to some degree. (14OTE)
|
other
| 99.9 |
In contrast, others said that some devices were more advantageous or safe than others, which would support the need for PE. There are some devices where I'm not switching because it's doing everything I need it to do, and other situations where an iteration of a device provides very helpful advantages in terms of ease of implantation or safety. (21CTM)
|
other
| 99.9 |
PE was viewed as less feasible if there were few devices to choose from on the market. In the world of implantable ventricular assist devices, there are only two available devices on the market now that are being used predominantly around the world. (04CTE)
|
other
| 99.94 |
In previous research, patients did not achieve desired levels of engagement in discussions about implantable medical devices [18–21]. This study assessed whether and how, from a physician's perspective, it was feasible to engage patients in such discussions. Most participants informed patients about the device they chose, the rationale for that choice and associated risks. Few involved patients in decisions by discussing evidence for the device, eliciting concerns and expectations, confirming understanding or revealing conflicts of interest. None partnered with patients to choose particular devices. Participants described multiple interacting patient, physician, health system and device/market factors that constrained PE, which were common for disparate types of devices and among physicians with different characteristics.
|
study
| 99.94 |
While there is little directly comparable research with which to relate these findings, a number of ideas that emerged from this study have also been identified elsewhere including variability among physicians in the informed consent process, patient preference for physicians to choose devices and the use of devices approved in purchasing group contracts. Interviews with 11 American cardiologists revealed substantial variation in the extent to which they discussed ICD risks with patients . In a survey of 364 American orthopaedic surgeons, cost reduction programmes based on volume discounting at their institution was a frequently listed factor that influenced their decision-making . In the same study, among 102 patients undergoing hip or knee arthroplasty, 93.1% said that their orthopaedic surgeon should choose the device; 5.9% said that physicians should consult patients when making the decision. With respect to PE in general, a systematic review found that time constraints, lack of applicability due to patient characteristics and specifics of the clinical situation were the most frequent barriers of shared decision-making .
|
review
| 99.75 |
Our study was unique in that it examined determinants of PE for medical device decision-making. Strengths included the use of rigorous qualitative methods, and analysis of the findings using existing frameworks of PE [1, 2] and PE for medical device decision-making . However, several limitations should be mentioned. The number of participants may appear small, but qualitative research is meant to capture detailed information from few, representative participants. Their views may reflect the Canadian health care setting and may not be transferrable to other settings. However, the devices they use are those used worldwide, and several issues that emerged were also revealed in other studies which support the reliability of these findings [27–29]. To mitigate this, sampling was purposive to capture perspectives from individuals with a variety of characteristics, and achieved thematic saturation which signals that recruitment is sufficient to identify themes, though not necessarily sufficient to explore the characteristics and implications of these themes. This would require larger samples from diverse backgrounds, and would be an interesting future study.
|
study
| 99.94 |
At the patient level, some participants questioned patients’ ability and interest to discuss technical information about devices while others said that patients themselves acquired information about device performance. Research is needed to examine patient capacity to engage in discussions and decisions about various types of medical devices, and interventions that can support PE in this context. The PE literature advocates that partnership is appropriate in all circumstances provided it matches patient preferences about the level of engagement . This study did not interview patients, but other research shows that patient preferences for involvement in device decision-making may vary. For example, in one study, 20 patients who were interviewed about involvement in decision-making for knee implants said that they desired active participation, and 17 said they were not given sufficient information or opportunity . In contrast, in another study, among 102 patients undergoing hip or knee arthroplasty, 93.1% said that their orthopaedic surgeon alone should choose the device . A systematic review of 115 studies of patient preference for involvement in shared decision-making found that the majority of patients undergoing invasive procedures (78.5% across 11 studies) preferred to be involved . This rate was similar to patients with cancer in 43 studies, and higher than patients with other chronic conditions (26 studies) or in the general population (36 studies). Hence, it may behoove physicians to assess patient desire for extent of involvement in decision-making about implantable devices.
|
review
| 99.9 |
However, views among participating physicians about the factors constraining PE raise several implications for practice and for ongoing research. At the physician level, views differed on whether proficiency in one or many devices was ideal. If the latter were true, then surgical mentorship may help physicians to expand their competency in a range of devices such that they could engage patients in a discussion of device options . First, research should establish if patient outcomes differ between physicians who use one or many types of devices.
|
other
| 99.9 |
At the level of devices, participant views contrasted on whether devices were interchangeable. PE is, in part, based on a discussion of the evidence for device safety and effectiveness, however, such information is lacking for many types of devices [13, 14]. The IDEAL (Idea, Development, Exploration, Assessment and Long-term follow-up) framework, originally devised to guide the evaluation of novel surgical techniques, was recently modified to accommodate the evaluation of devices . The IDEAL framework included five stages (first in human, prospective development studies, prospective exploration studies, assessment via RCT or alternatives and long-term study). The IDEAL-D framework includes six stages (first in human, which allows confidential reporting to accommodate intellectual property rights; sequential prospective non-comparative cohort studies to generate insight on operator learning curves and iterative changes to implantation procedures; large uncontrolled prospective cohort studies or audits to build consensus on definitions, quality control and outcome expectations for subsequent trials (for first-of-kind devices, could be omitted for successor devices); assessment via RCT or alternatives (for first-of-kind devices) and long-term study or nested RCTs via device registries). If widely adopted, IDEAL-D processes may lead to greater evidence on the safety and effectiveness of medical devices .
|
review
| 99.9 |
At the health system level, and potentially overriding physician, patient and device factors, contractual obligations may restrict physicians from using devices with which they are proficient, or considered best suited for patients’ needs, which was thought to increase costs due to complications and revisions. If this phenomenon is widespread, it challenges whether decisions about medical devices can be considered preference sensitive, which hinges on patients having legitimate treatment options . First, research should establish whether device restrictions imposed by purchasing groups are associated with poor outcomes, perhaps by comparing outcomes at hospitals with and without such arrangements.
|
other
| 99.9 |
In summary, this study revealed several factors that, apart from potentially variable patient preference, challenge PE. Further research is needed to identify conditions, for example, type of device and patient characteristics, in which different forms of PE (consult, involve and partner) are relevant and feasible. For now, most participants agreed that informing and involving patients were feasible. Yet some noted that discussions about device risks and benefits varied across physicians, and this was also found to be true elsewhere [18–21]. This study generated a preliminary framework (Table 2) by which physicians can more consistently and thoroughly engage patients in discussions about medical devices, even when decisions may not be preference sensitive due to constraints on choice imposed by patient, physician, device or health system characteristics. Further research is needed to evaluate use and impact of this framework on patient and physician satisfaction with consultations about implantable devices.
|
study
| 99.94 |
Understanding the geographical deployment of biodiversity through time is a central theme in historical biogeography . The disjunct distributions of closely related organisms between East Asia and North America have fascinated botanists and biogeographers for over a century and a half [2–5]. In plants, biogeographic studies employing the integration of phylogenetic hypotheses, inference of ancestral ranges, and estimates of divergence times have largely focused on the classic eastern Asian and eastern North American floristic disjunction pattern [5–8]. Few studies have been devoted to investigate the eastern Asian and western North American disjunction [9, 10]. For these two patterns, the Miocene has been regarded as an important period, in which the Bering land bridge likely acted as a major gateway [5, 11–13].
|
review
| 99.7 |
In the Northern Hemisphere, East Asia is a pivotal biogeographic region as it presents high levels of plant species diversity and endemism [14, 15]. Based on Takhtajan’s floristic system, southern East Asia belongs to the Paleotropical Kingdom, whereas northern East Asia is part of the Holarctic Kingdom (Fig. 1). Recent molecular phylogenetic studies also indicate that the Tertiary relict floras within East Asia could be subdivided into two distinct southern and northern regions [17, 18]. The former consists of southern and southeastern China with extending to the Himalayas, while the latter contains Japan, Korea, and northeastern China. Besides, as a continental island adjacent to southeastern mainland China, the Ryukyus Islands, and Philippines, the floristic source of Taiwan is not clear [19–21]. To date, biogeographic relationships among southern East Asia, northern East Asia and Taiwan are far from understood.Fig. 1Geographic range of Coptis species. Doted lines in bold demarcate boundaries of the Holarctic and Paleotropical kingdoms according to Takhtajan
|
study
| 99.56 |
The goldthread genus Coptis Salisb. (Ranunculales, Ranunculaceae, Coptidoideae) is of pharmaceutical and economical importance and is mainly distributed in the warm temperate to the cold coniferous forests of eastern Asia and North America [22, 23]. Among the 15 species recognized by Tamura , C. trifolia (L.) Salisb. has the widest distribution area (including Japan, the Kurile Islands, Kamchatka, and North America), while the other 14 species are restricted to smaller regions: five species are found in southern and southwestern mainland China with extensions to the Himalayas, five in Japan, one in Taiwan, and three in western North America (Fig. 1). Our recent phylogenetic analysis based on three DNA markers indicates that three western North American species of the genus clustered with five mainland Chinese and two Japanese species, and Taiwanese C. morii Hayata and three Japanese species grouped together . The fruits of Coptis are dehiscent follicles and seeds may be autochorously dispersed owing to lacking obvious adaptation to wind-dispersal. Seeds are not thereby expected to disperse over long distance or oceanic barriers. Thus, Coptis provides a remarkable opportunity for studying the eastern Asian and western North American distribution pattern, as well as the biogeographic relationships within East Asia.
|
study
| 99.94 |
In this study, first we reconstruct a dated phylogeny for Coptis based on six DNA markers, using a Bayesian relaxed clock method. Using the resulting dated-phylogenetic framework, we then infer the ancestral range evolution of Coptis by comparing the relative fit of six biogeographic models. Our study contributes to the knowledge on the eastern Asian-western North American distribution pattern and eastern Asian biogeography.
|
study
| 99.94 |
We sampled all 15 species of Coptis recognized by Tamura . Coptis and the monotypic Xanthorhiza Marshall compose the subfamily Coptidoideae, which is sister to a large clade containing the overwhelming majority of genera of Ranunculaceae [24, 25]. Scoring this large clade for geographic areas is a challenge. Here, we only selected Xanthorhiza as the outgroup. The sampled species and their GenBank accession numbers are listed in Additional file 1: Table S1.
|
study
| 99.94 |
Six DNA markers, including five plastid (rbcL, trnL intron, trnL-F spacers, trnD-trnT, and trnH-psbA) and one nuclear (ITS) regions were used in this study. We generated new trnL sequence for C. japonica var. anemonifolia (Siebold & Zucc.) H. Ohba and trnL and ITS for C. morii. These two samples were collected in public land and no specific permits were required. Other sequences were obtained from GenBank. Laboratory procedures and sequence handling followed Wang and Chen . Three difficult-to-align regions in trnL-F (encompassing 20 positions), two difficult-to-align regions in trnH-psbA (48 positions), and one difficult-to-align region in trnD-trnT (24 positions) were excluded from the analyses. The final dataset included 4288 characters: rbcL, 1304 bp; trnL intron, 465 bp; trnL-F, 426 bp; trnD-trnT, 1122 bp; trnH-psbA, 289 bp; and ITS, 682 bp.
|
study
| 100.0 |
We first conducted a likelihood ratio test to determine whether our sequence data were evolving in a clock-like fashion. Because rate constancy along all branches of the phylogeny was rejected (δ = 146.63, d.f. = 14, P < 0.0001), we used a Bayesian relaxed clock methodology as implemented in BEAST v1.8.2 to generate a dated phylogeny for Coptis. Based on our recent broader study of Ranunculaceae , the split time between Coptis and Xanthorhiza was estimated at ca. 16.23 Ma (95% highest posterior density (HPD): 8.51–25.96) and was here used as a secondary calibration point. Following the suggestion of Ho , we assigned a prior normal distribution for the calibration, in which a standard deviation of 2 was set.
|
study
| 100.0 |
Following the result of Baele et al. , we used Bayes factors calculated by marginal likelihoods derived from path sampling (PS) and stepping-stone sampling (SS) to compare the parametric fit of three clock models: exponential, lognormal and random. Since our sampling included all recognized species of Coptis and Xanthorhiza, a birth-death tree prior was used.
|
study
| 99.94 |
For all BEAST analyses, data partitioning and nucleotide substitution models were determined using PartitionFinder 2.1.1 [34, 35]. The Markov chain Monte Carlo chains were run for 100 million generations, sampling every 10,000 generations. Tracer v1.6 was used to assess appropriate burn-in and the adequate effective sample size values (> 200). A burn-in of 25% was applied, and the maximum clade credibility (MCC) tree with the mean ages and 95% HPD intervals on nodes were conducted in TreeAnnotator v1.8.2 (part of the BEAST package) and edited in FigTree v.1.4.2 (http://beast.bio. ed.ac.uk/FigTree).
|
study
| 100.0 |
Based on the floristic characteristics [16, 18] and distributions of Coptis and Xanthorhiza , we coded five biogeographical areas (Fig. 1): (A) western North America, (B) southern East Asia (including southern and southeastern mainland China and the adjacent Himalayan region), (C) Japan and adjacent islands (including the Kurile Islands and Kamchatka), (D) Taiwan, and (E) eastern North America. The maximum range size was set to three, as no extant species occurs in more than three biogeographical regions. Because the Bering land bridge was periodically available for exchanges of plants between eastern Asia and western North America until 3.5 Ma [37–39], dispersal probabilities between pairs of areas were specified for two separate time slices (Additional file 1: Table S2).
|
study
| 99.94 |
We used the R package BioGeoBEARS for ancestral range estimation (ARE) on the MCC tree from the BEAST run under the optimal clock model and tree speciation prior. Recently, Ree & Sanmartín demonstrated that the likelihood-based models with the +J parameter are invalid because of errors in the estimation of likelihoods. Here we compared the following three models of biogeographical estimation in the maximum likelihood (ML) framework: dispersal-extinction cladogenesis (DEC) model , dispersal–vicariance analysis (DIVA) and BayArea model . The fit for the different models was assessed using the Akaike information criterion scores.
|
study
| 100.0 |
We identified the random clock model as optimal for our data (Table 1). The dated phylogenetic tree generated in the BEAST analysis under the random clock model and birth-death tree prior is indicated in Fig. 2. The relationships among Coptis species are well resolved with strong support (PP > 0.95) except for the node defining the sister relationship of C. quinquefolia Miq. and C. morii. Coptis contains two main clades (I and II). Based on our time estimates (Fig. 2), the stem and crown ages of Coptis are estimated at ca. 15.47 Ma (95% HPD: 11.47–19.37; node 1) and 9.55 Ma (95% HPD: 6.66–12.92; node 2), respectively. Within clade I, three western North American species clustered together and split from their eastern Asian sister group at ca. 7.78 Ma (95% HPD: 5.16–10.52; node 3). Japanese C. japonica Makino and C. lutescens Tamura are nested in the group of mainland Chinese species and the split of these two Japanese species and their sister group occurred at ca. 4.85 Ma (95% HPD: 2.98–6.80; node 4). Within clade II, Taiwanese C. morii and Japanese C. quinquefolia were grouped together with weak support (PP = 0.73). The split time of C. morii and C. quinquefolia was estimated to be at ca. 1.34 Ma (95% HPD: 0.69–2.18; node 5).Table 1Comparison of three clock models in BEAST analyses via Bayes factorsClock modelMarginal likelihoodExponentialLognormalRandomPS implementation Exponential− 8809.70–37.44−32.90 Lognormal− 8828.42−37.44–−70.34 Random− 8793.2532.9070.34SS implementation Exponential− 8810.29–37.62−33.72 Lognormal− 8829.10−37.62–−71.34 Random−8793.4333.7271.34–2ln Bayes factor (BF) was calculated by marginal likelihoods derived from path sampling (PS) and stepping-stone sampling (SS) implementations in BEAST. 2ln BF > 2.0 represents positive evidence, > 6.00 represents strong evidence, and > 10.00 represents very strong evidence Fig. 2Dated phylogeny of Coptis inferred from the combined plastid and nuclear data using BEAST under the random clock model and birth-death tree prior. Gray bars represent 95% highest posterior density intervals. Nodes of interests were marked as 1–5 in bold. All nodes are strongly supported (PP > 0.95) except for one node (in dashed line). Plio., Pliocene; Plt., Pleistocene
|
study
| 100.0 |
2ln Bayes factor (BF) was calculated by marginal likelihoods derived from path sampling (PS) and stepping-stone sampling (SS) implementations in BEAST. 2ln BF > 2.0 represents positive evidence, > 6.00 represents strong evidence, and > 10.00 represents very strong evidence
|
other
| 99.6 |
Dated phylogeny of Coptis inferred from the combined plastid and nuclear data using BEAST under the random clock model and birth-death tree prior. Gray bars represent 95% highest posterior density intervals. Nodes of interests were marked as 1–5 in bold. All nodes are strongly supported (PP > 0.95) except for one node (in dashed line). Plio., Pliocene; Plt., Pleistocene
|
study
| 99.94 |
A DIVALIKE was found to be the best-fitting model (Table 2). The ARE for Coptis using BioGeoBEARS is indicated in Fig. 3 and Additional file 2: Figure S1. Area probabilities of all nodes are high except the root. Our ARE shows that the ancestral range of Coptis and Xanthorhiza is unresolved but likely involved eastern North America, western North America and Japan (node 1). The most recent common ancestor of Coptis was likely distributed in western North America, southern East Asia and Japan (node 2). Within Coptis, two vicariance events and two dispersal events were inferred at the species level (Fig. 3).Table 2Comparison of the fit of three models of biogeographical range evolution and model-specific estimates for the different parametersModelLnLParameter nb d e AICΔAICAICCΔAICCDEC−24.0620.031.00 × 10− 1252.123.5953.043.58DIVALIKE−22.2720.031.00 × 10− 1248.53049.460BAYAREALIKE−28.7720.041.04 × 10−161.5513.0252.4713.01d = dispersal rate; e = extinction rateFig. 3Ancestral range estimation (ARE) for Copits BEAST using BioGeoBEARS under the DIVALIKE model. Labeled nodes (1 to 5, as referred to Fig. 2), with 95% highest posterior densities (grey bars), are discussed in the text. The estimated ancestral ranges with the highest ML probability are shown by boxes on each node. Additional file 2: Figure S1 provides all ARE per node with pies. A pie is placed in this figure at the root with the highest probability less than 50%. The depictions of temperature (in red) and sea level (in black) changes are modified from Zachos et al. and Haq et al. , respectively. Plio., Pliocene; Plt., Pleistocene
|
study
| 100.0 |
Ancestral range estimation (ARE) for Copits BEAST using BioGeoBEARS under the DIVALIKE model. Labeled nodes (1 to 5, as referred to Fig. 2), with 95% highest posterior densities (grey bars), are discussed in the text. The estimated ancestral ranges with the highest ML probability are shown by boxes on each node. Additional file 2: Figure S1 provides all ARE per node with pies. A pie is placed in this figure at the root with the highest probability less than 50%. The depictions of temperature (in red) and sea level (in black) changes are modified from Zachos et al. and Haq et al. , respectively. Plio., Pliocene; Plt., Pleistocene
|
study
| 99.94 |
The phylogenetic relationships in Coptis are highly consistent with the results of Xiang et al. , but are usually resolved with greater support for clades found therein. Our results do not support Taiwanese C. morii as sister to three Japanese species (C. ramose (Makino) Tamura, C. quinquefolia and C. trifoliolata (Makino) Makino), and instead suggest that C. morii is sister to C. quinquefolia, although with moderate support (PP = 0.71). Using the split age of ca. 16.23 Ma (95% HPD: 8.51–25.96) between Coptis and Xanthorhiza , we obtained a similar age estimate for the split (ca. 15.47 Ma; 95% HPD: 11.47–19.37; Fig. 2).
|
study
| 100.0 |
BioGeoBEARS analyses indicate that the crown of Coptis and Xanthorhiza most likely occurred in a widespread area comprising North America and Japan (Fig. 3; node 1), although other somewhat less likely ARE are possible (Additional file 2: Figure S1). The estimated age for the split of these two genera highly coincides with the mid-Miocene Climatic Optimum (MMCO; ~ 15–17 Ma; Fig. 3) . During this period, exchange of temperate plants between East Asia and North America could occur via the Bering land bridge . Paleobotanical data indicate that the mixed mesophytic forest of the early and middle Miocene was continuous from Japan through Alaska and into conterminous North America [47, 48].
|
study
| 99.94 |
The American west encompassing the Colorado Plateau, Basin and Range, the High Plains, and the Rocky and Sierra Mountains began to uplift rapidly by 20–15 Ma . A middle Miocene flora from Carson Pass in the central Sierra Nevada suggests uplift of about 2300 m since that time . The uplift is a key factor in creating an increasingly drier climate in the North American interior around that time [49, 51]. Paleobotanical evidence suggests that by the middle Miocene the arid interior has become an effective barrier to biotic interchange between eastern and western North America [52, 53]. After the MMCO, an increasingly drier climate, as well as global cooling (Fig. 3) , might thus have resulted in a vicariance event responsible for the divergence of Coptis and Xanthorhiza (node 1; Fig. 3).
|
study
| 99.9 |
After Coptis diverged from Xanthorhiza, a subsequent dispersal from Japan to southern East Asia occurred in the early late Miocene (9.55 Ma, 95% HPD: 6.66–12.92; node 2). This time is markedly later than the time of the opening of the Japan Sea (23–15 Ma), which separated the Japanese Islands from the Northeast Asian margins [54–56]. However, during the early late Miocene, a marked drop of sea level occurred (Fig. 3) , which might have resulted in East China Sea seafloor exposure between the Eurasian mainland and the Japanese Archipelago. Hence, Coptis could have migrated westward into continental Asia via this land bridge. Subsequent sea-level rise might have resulted in the interruption of population exchange of the genus between the Asian mainland and the Japanese Islands. Accordingly, Coptis diverged into two clades (I and II).
|
study
| 100.0 |
In clade I, one vicariance episode happened between western North America and southern East Asia in the Late Miocene (ca. 7.78 Ma, 95% HPD: 5.16–10.52; node 3), which overlapped closely with the time of the first opening of the Bering Strait (7.4–5.5 Ma) . Evidence from sedimentology and foraminifera indicates that uplift of the St. Elias Mts. in Alaska began about 8.5 Ma . Palynological analyses suggest that the trends of temperature decline and increasing canopy openness in Alaska and Yukon Territory occurred between 9.7 and 7.0 Ma, owing to global and local tectonic developments . These events may explain the distribution of Coptis between southern East Asia and western North America during the Late Miocene. The split of western North American Polypodium californicum Kaulf. (Polypodiaceae) and its eastern Asian relatives (P. fauriei (Copel.) Makino & Nemoto and P. glycyrrhiza D.C. Eaton) also occurred during the same period (ca. 8.81 Ma, 95% HPD: 5.06-13.08) . Such distribution patterns resulting from orogenic events have been found in some plant lineages and in different biomes, such as Campanulaceae , Orchidaceae , and Rubiaceae .
|
study
| 99.94 |
One dispersal event in clade I occurred in the early Pliocene from southern East Asia to Japan (ca. 4.85 Ma, 95% HPD: 2.98–6.80; node 4). The most recent common ancestor of Japanese Pseudotsuga japonica (Shiras) Beissn. and mainland Chinese P. gaussenii Flous and P. sinensis Dode (Pinaceae) was estimated to occur at ca. 4.64 ± 1.93 Ma . In Eupteleaceae, Chinese Euptelea pleiosperma Hook. f. & Thomson split with Japanese E. polyandra Siebold & Zucc. at ca. 6.04 Ma (95% HPD: 2.89–9.36) . The drop of sea level may have resulted in exchanges of plants between mainland Asia and the Japanese Islands via the East China Sea land bridge, and subsequent rise of sea level and global cooling (Fig. 2) , as well as an increasingly drier climate in Asia , may have caused the interruption of the continuous distribution of ancestral populations of some extant species during the Late Miocene to the Early Pliocene.
|
study
| 100.0 |
Within clade II, one dispersal event from Japan to Taiwan occurred in the Early Pleistocene (ca. 1.34 Ma; 95% HPD: 0.69–2.18; node 5). The eustatic sea-level fluctuation during this period, as well as global cooling (Fig. 2), may have triggered Coptis range expansion from Japan to Taiwan via the Ryukyu Islands, and may have subsequently caused range fragmentation. A similar scenario also explains the current distribution of Taiwanese Chamaecyparis formosensis Matsum. and C. taiwanensis Masam. & Suzuki (Cupressaceae) from hypothetical Japanese ancestors . Our analysis on Dichocarpum W.T. Wang & P.G. Xiao indicates that Taiwanese D. arisanense (Hayata) W.T. Wang & P.G. Xiao could have originated from mainland China in the Early Pleistocene (ca. 1.26 Ma, 95% HPD: 0.48–2.33) . These studies support the hypothesis that temperate elements of the flora of Taiwan recently migrated from mainland China and Japan .
|
study
| 100.0 |
We present a dated phylogeny for all species of Coptis, a genus of pharmaceutical and economical importance. Our biogeographical inference indicates that a vicariance event between Japan-western North America and eastern North America occurred in the Middle Miocene, resulting in the split of Coptis and Xanthorhiza. The most recent common ancestor of Coptis occurred in western North America, southern East Asia and Japan. In Coptis, two vicariance episodes, involving Japan and western North America-southern East Asian and western North America and southern East Asian, took place at ca. 9.55 Ma and 7.78 Ma, respectively. Two dispersal events happened from mainland Asia to Japan at ca. 4.85 Ma and from Japan to Taiwan at ca. 1.34 Ma, respectively. This study shed light on the past floristic exchanges between East Asia and North America, as well as within East Asia.
|
study
| 99.94 |
Additional file 1:Table S1. GenBank accession numbers and vouchers/references for the sequences used in this study. Table S2. Manual dispersal multipliers. (PDF 36 kb) Additional file 2:Figure S1. Raw PDF outputs from biogeographic estimations in BioGeoBEARS. (PDF 358 kb)
|
other
| 99.9 |
Astrocytes, the principal macroglial cells in the central nervous system (CNS), are fundamental to brain function and health [1, 2]. In pathological conditions, astrocytes react to almost all forms of CNS insults. Astrocytes’ pathological responses typically involve increased expression of glial fibrillary acidic protein (GFAP), cellular hypertrophy and enhanced proliferation. This pattern is indicative of reactive astrogliosis, a hallmark of diverse neurological disorders including trauma, ischemia, neurodegeneration, multiple sclerosis, epilepsy and autism [3–5]. A growing body of evidence indicates that reactive astrocytes are not bystanders during disease development . However, their functions and regulatory mechanisms remain elusive.
|
review
| 99.8 |
Reactive astrocytes responding to pathological insults are distinct from normal astrocytes in morphology, functions and molecular profiles [1, 5–7]. To explore these astrocytes’ regulatory mechanisms, many molecules have been investigated over the past few decades. However, many of them exhibit paradoxical implications [5–7]. Further complicating the picture is the fact that the regulatory mechanisms of reactive gliosis have been most frequently studied in cell culture models. Findings from such models are increasingly being tested in genetically engineered mice, with some studies failing to confirm in vitro findings in vivo. For example, cytoplasmic or nuclear signaling proteins, such as Akt, mechanistic target of rapamycin, extracellular signal-regulated kinase (Erk) and signal transducer and activator of transcription 3 (STAT3), are upregulated in reactive astrocytes, and their positive involvement in regulating astrogliosis has been confirmed by genetic approaches in vivo [8–14]. Receptor tyrosine kinases are upstream activators of these signaling pathways [15–17]. Among these, fibroblast growth factor 2 (FGF2) and its receptor (FGFR) increase in reactive astrocytes, promote GFAP expression in cultured astrocytes, and are thus thought to mediate astrocytes’ reactive responses. Surprisingly, recent loss- and gain-of-function studies on genetically engineered mice show that FGF signaling inhibits astrocyte reactivity under both uninjured and injured conditions . These findings emphasize that the upstream signaling that regulates astrogliosis remains largely undetermined.
|
review
| 99.9 |
ErbB receptors (ErbB1–4), another family of receptor tyrosine kinases, and their ligands, including epidermal growth factor (EGF), neuregulin (NRG) and transforming growth factor α (TGFα), have been reported to increase in tissues with reactive astrogliosis [19–21]. In the nervous system, ErbB receptors are differentially expressed across various neural cell types and regulate many developmental and pathological events [16, 17, 22]. Once ligand bound, ErbB receptors dimerize and activate multiple intracellular signaling pathways, including Akt/mechanistic target of rapamycin, Erk and STAT3, to potently regulate cell proliferation, survival, differentiation, migration and inflammatory responses [16, 17, 22]. Both mutations and post-transcriptional alteration of ErbB receptors have been implicated in neurological disorders, including demyelination, stroke, epilepsy and psychiatric disorders [16, 22]. However, it remains unknown whether aberrant ErbB signaling in astrocytes participates in disease progression.
|
review
| 99.8 |
To examine whether ErbB signaling has a role in astrocytes of injured or diseased brains, we manipulated ErbB receptor activity in mature astrocytes by adopting a pan-ErbB strategy in mice. By conditionally expressing either a dominant-negative ErbB mutant that inhibited any ErbB receptor when coexpressed in the same cell or a constitutively active ErbB mutant that promoted ErbB receptor activation, we circumvented the limited information on ErbB receptor composition in astrocytes and focused on the function of ErbB signaling. Through in vivo studies combining loss- and gain-of-function approaches, we found that ErbB signaling positively regulated astrocyte reactivity, exerting a direct effect on hypertrophic remodeling and a cooperative effect on other features of reactive astrocytic responses.
|
study
| 100.0 |
In situ hybridization studies have revealed that the epidermal growth factor receptor (EGFR/ErbB1) and ErbB2 are expressed in astrocytes . After non-detection of ErbB receptors in astrocytes of the adult brain by immunostaining, we generated Mlc1-tTA;TRE-dnEGFR (Mlc1-dnEGFR) bi-transgenic mice by crossing Mlc1-tTA mice with TRE-dnEGFR mice to investigate the existence of functional ErbB signaling in mature astrocytes. The Mlc1 gene encodes megalencephalic leukoencephalopathy with subcortical cysts-1 (Mlc1), a membrane protein expressed specifically in GFAP-positive cells in the adult brain . Thus, in our transgenic mice, the Mlc1 promoter drove the expression of tetracycline-controlled transactivator (tTA) in astrocytes to activate tetracycline-responsive element (TRE)-controlled transcription through a ‘Tet-off’ system .
|
study
| 100.0 |
To confirm the specificity of Mlc1-tTA, an adeno-associated virus (AAV) harboring a TRE-yellow fluorescent protein (YFP) reporter was stereotaxically injected into the brain of 1-month-old Mlc1-tTA mice (Figure 1a). To avoid confusion with a reactive response, Mlc1-tTA targeting cells in the normal brain were identified by YFP fluorescence 1 day after virus injection. YFP-positive (YFP+) cells in the hippocampal hilus were immunopositive for GFAP (Figure 1a). It is now well known that astrocytes in the brain are heterogeneous. Although GFAP, an intermediate filament protein, is a reliable marker for astrocytes in the white matter, cerebellum and hippocampus of the adult brain, it is normally undetectable in the cerebral cortex [7, 11, 25]. Therefore, to confirm the specificity of YFP+ cells in the cerebral cortex, we immunostained AAV-injected cortical slices with an antibody to Acyl-CoA synthetase bubblegum family member 1 (Acsbg1), a protein marker for a wider spectrum of astrocytes [11, 26]. YFP+ cells in the cortex were positive for Acsbg1 (Figure 1a). In addition to adult astrocytes, Mlc1-tTA also targeted astrocyte-like adult neural stem cells (NSCs), as indicated by the presence of YFP-labeled cells coexpressing the NSC marker, nestin (Supplementary Figure 1a and b). Nevertheless, nestin+ cells in the normal adult brain were strictly localized in NSC niches, such as the subgranular zone, subcorpus callosum zone and subventricular zone.
|
study
| 100.0 |
After confirming the specific targeting of Mlc1-tTA to astrocytes in the cortex, we employed Mlc1-dnEGFR double transgenic mice to study ErbB receptor function in astrocytes. TRE-dnEGFR transgenic mice encode a dominant-negative mutant of EGFR (dnEGFR) that is ectopically expressed under the control of a TRE with a cytomegalovirus minimal promoter (PminCMV) . DnEGFR is a truncated form of EGFR that lacks the intracellular kinase domain but retains the ability to bind to other ErbB family members . In this study, efficient suppression of ErbB receptor activity by dnEGFR was first verified by coexpressing dnEGFR with a comprehensive panel of ErbB receptors in HEK293 cells. When coexpressed with dnEGFR, the activity of each ErbB receptor was reduced 50–60%, indicating haploinsufficient function (Figure 1b and c). When not fed the tetracycline analog doxycycline (Dox), bi-transgenic Mlc1-dnEGFR mice expressed dnEGFR in Mlc1-controlled cells (Figure 1d). After failing to detect dnEGFR with available antibodies, we examined dnEGFR transcripts through real-time reverse transcription polymerase chain reaction (RT-PCR) and observed a 20- to 60-fold increase in the cerebral cortices of Mlc1-dnEGFR mice compared with those of littermate controls (Supplementary Figure S2a and b). Variations in the increase of dnEGFR transcripts in different Mlc1-dnEGFR mice may have been caused by varied insertion copies of the ectopic gene. Notably, dnEGFR expression in astrocytes during development did not change animal weight, body size or gross brain structures. In addition, immunostaining of GFAP or Acsbg1 in brain slices did not reveal significant differences in astrocyte number, distribution or size (Supplementary Figure S2c–g), suggesting that reduced ErbB receptor activity did not alter astrocyte development in Mlc1-dnEGFR mice.
|
study
| 100.0 |
To investigate whether ErbB signaling is required for reactive astrogliosis, we used an acute brain injury model in which the parietal cortices of mice, aged postnatal day 30 (P30), were exposed and injured via a needle puncture. Three to 21 days later, brains were isolated, sectioned and immunostained for GFAP. Increased GFAP expression is a clear hallmark indicating the transition of normal astrocytes to reactive astrocytes [5–7]. Both Mlc1-dnEGFR and littermate control mice exhibited increased GFAP+ cells adjacent to the injury sites, with comparable distributions (Supplementary Figure S3a). Reactive astrocytes possess a molecular profile distinct from that of normal astrocytes [5, 7, 28]. For example, nestin, the protein marker for NSCs, appears to be a molecular hallmark of reactive astrocytes induced by various insults [29, 30], and it appeared in reactive astrocytes in either Mlc1-dnEGFR or control cortices 3 days after injury (Supplementary Figure S3b).
|
study
| 100.0 |
Using this model, we observed similar increases in dnEGFR transcripts in injured and intact Mlc1-dnEGFR cortices (Supplementary Figure S3c and d). To investigate whether ErbB receptor activity was altered in the reactive astrocytes of Mlc1-dnEGFR mice, injured cortical slices were costained with antibodies to GFAP and the active forms of ErbB1–4. Although other antibodies did not detect signals, immunoreactivity of phosphorylated ErbB3 (pErbB3) was clearly observed in the reactive astrocytes of control mice (Figure 1e). Notably, pErbB3 was hardly detected in the reactive astrocytes of Mlc1-dnEGFR mice, indicating inhibition of endogenous ErbB signaling by dnEGFR (Figure 1e). Note non-detection of pErbB3 was not due to loss of the ErbB3 protein because ErbB3 immunoreactivity was present in the reactive astrocytes of Mlc1-dnEGFR mice in a manner similar to that observed in controls (Figure 1f). The involvement of other ErbB receptors in reactive astrocytes could not be excluded, as their absence could have been due to the used assays being insensitive to small amounts of protein. In order to test this idea, we cultured primary astrocytes isolated from TRE-dnEGFR or Mlc1-dnEGFR mouse brains. All four ErbB receptors were detected in the primarily cultured astrocytes by Western blotting (WB) (Supplementary Figure S3e). We treated the primary astrocytes with recombinant human EGF (rhEGF), a ligand specifically binding to EGFR, and revealed activation of all four ErbB receptors in control astrocytes (Supplementary Figure S3e and f). Unsurprisingly, ErbB receptor activities induced by rhEGF were significantly reduced in primary Mlc1-dnEGFR astrocytes (Supplementary Figure S3f). Treatment with recombinant human NRG1, another ligand that specifically binds to ErbB3 or ErbB4, induced ErbB4 activation in primary control astrocytes. Remarkably, it was significantly reduced in primary Mlc1-dnEGFR astrocytes (Supplementary Figure S3e and f). These results consolidated the inhibitory effects of dnEGFR on ligand-induced ErbB receptor activation. It was noticeable that basic activities of EGFR, ErbB2 and ErbB3 were suppressed in primary Mlc1-dnEGFR astrocytes (Supplementary Figure S3f), suggesting an inhibition on efficiency of endogenous ErbB ligands. Therefore, dnEGFR targeting of ErbB receptor kinase activity helped us circumvent the limited knowledge of ErbB receptor composition in vivo and focus on characterizing their function in reactive astrocytes.
|
study
| 100.0 |
In addition to increased expression of GFAP, reactive astrocytes exhibit cellular hypertrophy. We found that most GFAP+ cells in the injured cortices of Mlc1-dnEGFR mice were smaller than those in control mice (Figure 2a). Quantitative analysis showed that despite a similar distribution of GFAP+ cells (Figure 2b), reactive astrocytes with a GFAP-immunoreactive area larger than 300 μm2 were significantly reduced in Mlc1-dnEGFR cortices at several time points after injury (Figure 2c). Consistently, WB showed that the increase in GFAP protein levels in injured cortical tissues from Mlc1-dnEGFR mice was 60.74±23.5% of that from littermate controls (Figure 2d). Reduction of GFAP was caused by an inhibition on GFAP gene transcription, and mRNA level of gfap splicing isoform 1 was significantly reduced (Figure 2e).
|
study
| 100.0 |
Dual mutation of GFAP and vimentin, a fellow intermediate filament protein, causes a small reduction in the extension of cellular processes of reactive astrocytes, whereas normal astrocytes are not affected . These proteins are thought to be effectors mediating hypertrophic remodeling. However, we found that the morphological expansion of astrocytes was independent of GFAP expression during injury-induced reactive responses. Injection of AAV-TRE-YFP into injured Mlc1-tTA cortices caused reactive astrocytes to be labeled with YFP fluorescence (Figure 2f). Specificity of YFP-labeled astrocytes and reactive astrocytes was further verified by co-labeling with antibodies against glutamine synthetase (GS) and nestin, respectively (Supplementary Figure S4a and b). Two and a half days after injury, YFP+ cells were hypertrophic and positive for Acsbg1 (Figure 2f). GFAP did not appear at this stage in every reactive astrocyte, whereas all YFP-labeled astrocytes in the injured cortices exhibited expanded sizes (Figure 2g). Especially, many hypertrophic astrocytes expressed little GFAP (Figure 2g, white arrowheads). Thus, morphological remodeling was likely a reactive response that preceded the onset of GFAP protein expression in astrocytes.
|
study
| 100.0 |
To confirm that YFP distribution represented completely the morphology of astrocytes, we used the method that was described previously [31, 32], and injected Alexa Fluor 568, which has smaller molecular weight into YFP-labeled cells in fixed brain slices. As shown in Supplementary Figure S5a and b, both Alexa Fluor 568 and YFP labeled astrocytic cell bodies and fine processes. However, because of labeling the cell at living status, YFP distribution exhibited more complete morphology than the injected fluorescent dye did (Supplementary Figure S5a and b). Therefore, YFP labeling would reliably help us to evaluate the astrocytic sizes in mouse cortices. Four days after injury, more reactive astrocytes in the injured cortices of both Mlc1-dnEGFR and littermate mice expressed detectable GFAP, as all cells targeted by AAV-TRE-YFP were immunopositive for GFAP (Figure 2h). Suppression of hypertrophy in reactive astrocytes in Mlc1-dnEGFR cortices was more striking when cell sizes were measured according to the distribution of YFP fluorescence, in comparison with that based on GFAP immunoreactivity (Figure 2h). Four and 8 days after injury, the YFP-labeled reactive astrocytes in Mlc1-dnEGFR cortices exhibited sizes 28.06±10.3% and 47.30±14.6%, respectively, of those in littermate controls (Figure 2i). Taken together, these results suggested that the morphological reaction of astrocytes was more likely to depend on actin remodeling than intermediate filament extension. Further, these findings revealed that the regulation of this hypertrophic remodeling was ErbB dependent.
|
study
| 100.0 |
We also characterized the increased proliferation of reactive astrocytes. Some of the reactive astrocytes induced by stab injury in both Mlc1-dnEGFR and control mice were immunopositive for the proliferation marker Ki67 (Figure 3a). In addition, reactive astrocytes were frequently labeled with Olig2 immunostaining (Figure 3b). The basic helix-loop-helix transcription factor Olig2 has recently been identified as a protein hallmark of astrogliosis , despite being a key factor in determining motor neuron and, subsequently, oligodendroglial fate in NSCs during embryonic development . In the normal brain, the majority of Olig2+ nuclei, which indicate oligodendroglial lineage, are distributed in the white matter, with only sparse localization of Olig2+ cells in the cerebral cortex. Notably, increased Olig2+ nuclei in the parenchyma have been discovered in stab-, freeze-, ischemia- or Aβ deposit-injured cerebral cortical tissues [33, 35, 36], with many localized in reactive astrocytes [33, 36]. Lineage tracing studies have demonstrated that Olig2+ astroglial cells are an essential source of increased reactive astrocytes in response to injury . Indeed, some Olig2+ nuclei were also positive for Ki67 in the stab-injured cortex (Supplementary Figure S6). Surprisingly, neither the proportion of Ki67+ nor that of Olig2+ reactive astrocytes were changed in Mlc1-dnEGFR mice (Figure 3c), suggesting that dnEGFR did not affect their proliferation.
|
study
| 100.0 |
Acute injury induces proliferation of reactive astrocytes and extensive intermingling of their elongated processes, forming glial scars surrounding damaged CNS tissues . Although glial scars limit damage in tissues, they may elicit seizures in patients after brain injury . Many factors generated by different cell sources influence scar formation . We employed a more severe injury model by stabbing mouse cortices with a sharp blade, and examined scars formed 7 and 15 days after injury (Figure 3d). Scar thickness was evaluated according to the immunoreactivity of GFAP plus that of AQP4, a membrane protein whose immunoreactivity labeled the full mormorphology of scar-forming astrocytes (Figure 3d and e). We found that astrocytic scars in the injured cortices of Mlc1-dnEGFR mice were formed thinner than those in littermate controls (Figure 3d and f). Thus, with the suppression on morphological expansion, scar formation ability of reactive astrocytes was impaired in Mlc1-dnEGFR mice.
|
study
| 100.0 |
The loss-of-function studies revealed that ErbB signaling was specifically required for hypertrophic remodeling of reactive astrocytes. To examine whether ErbB activation in astrocytes was solely responsible for the hypertrophic response, we generated another mouse line with inducible gene expression in astrocytes by crossing Mlc1-tTA with TRE-ErbB2V664E mice. Among ErbB1–4 receptors, ErbB2 does not bind any known ligand but is the preferred partner of other ErbB members . Moreover, dimerization with ErbB2 potentiates the downstream signaling of ErbB receptors . The ectopically expressed gene in TRE-ErbB2V664E transgenic mice encodes an active form of rat ErbB2 (ErbB2V664E) that contains an amino-acid mutation (Vla664/Glu664) within the transmembrane domain that facilitates its dimerization and activation .
|
study
| 100.0 |
To generate this line, TRE-ErbB2V664E mice were mated with Mlc1-tTA mice. Pregnant mice and their offspring were fed with Dox until weaning occurred at P21, whereas bi-transgenic Mlc1-tTA;TRE-ErbB2V664E (Mlc1-ErbB2V664E) mice and littermate controls were allowed to grow to adulthood. Feeding the bi-transgenic Mlc1-ErbB2V664E mice with Dox blocked ErbB2V664E expression, while withdrawal of Dox initiated it (Figure 4a). It is notable that ErbB2V664E expression induced by Dox withdrawal was accompanied by an increase in GFAP in a time-dependent manner (Figure 4a). Immunostaining showed that GFAP+ cells increased dramatically throughout the brain 20 days after Dox withdrawal (Figure 4b).
|
study
| 100.0 |
To confirm that the detected GFAP+ cells were indeed astrocytes in the cerebral cortex, we costained the cortical slices with antibodies to GFAP and Acsbg1. Acsbg1 immunoreactivity was morphologically reminiscent of the staining observed in the cortical astrocytes of control mice and colocalized well with increased GFAP in Mlc1-ErbB2V664E cortices (Figure 4c). Moreover, Acsbg1+GFAP+ astrocytes possessed enlarged cell bodies and broadened cellular processes, indicating a hypertrophy (Figure 4c). Remarkably, the increased number of GFAP+ astrocytes was due to both GFAP expression in normally GFAP-negative astrocytes and active astrocyte proliferation, as indicated by about half (60.95±4.00% in the cortex, 53.16±4.96% in the corpus callosum and 59.38±12.7% in the midbrain) of GFAP+ cells being immunopositive for Ki67 (Figure 4d).
|
study
| 100.0 |
The increased GFAP expression, hypertrophy and elevated proliferation of mature astrocytes implied the induction of spontaneous astrogliosis in Mlc1-ErbB2V664E mice after Dox withdrawal. Consistent with the ectopic expression of ErbB2V664E, various brain regions in Mlc1-ErbB2V664E mice exhibited immunoreactivity with the ErbB2 antibody (Supplementary Figure S7). This immunoreactivity localized well in GFAP+ cells, confirming cell-targeting specificity in Mlc1-ErbB2V664E mice (Figure 4e). There was neither scarring nor aggregation of reactive astrocytes in the cerebral cortices, consistent with the idea that astrocytes in Mlc1-ErbB2V664E mice reacted to intrinsic factors instead of extrinsic ones released from injury sites or degenerative plaques. To further test this idea, primary astrocytes purified from Mlc1-ErbB2V664E and control mice were examined in vitro. Cultured cortical astrocytes from either control or Mlc1-ErbB2V664E mice exhibited reactive features such as the GFAP expression (Supplementary Figure S8a). However, normal cultured astrocytes appeared to be flat and spreading, adhering well to the bottom of the cell culture dish, with only a few of them expressing the radial glial cell marker RC2, whereas astrocytes from Mlc1-ErbB2V664E mice exhibited small but plump shapes and high levels of RC2, as well as accelerated proliferation rates (Supplementary Figure S8a and b). These results suggest that cell-autonomous ErbB activation promoted astrocyte proliferation. The observation of high RC2 expression in Mlc1-ErbB2V664E astrocytes was in line with the previous report that TGFα stimulates cultured astrocytes to dedifferentiate into NSCs .
|
study
| 100.0 |
As another molecular hallmark for reactive astrocytes, nestin did not appear in the cerebral cortex in control mice, but was observed throughout the brain of Mlc1-ErbB2V664E mice and colocalized well with GFAP+ cells (Figure 4f). Similarly, Olig2+ nuclei increased in various brain regions of Mlc1-ErbB2V664E mice (Supplementary Figure S8c), with many localized in GFAP+ cells (Figure 4g). An increased proportion of Olig2+ cells was correlated with an increase in GFAP+ cells (Figure 4h). As in our experiments using the injury model, we observed that many Olig2+ cells were positive for Ki67 in the brains of Mlc1-ErbB2V664E mice (Supplementary Figure S8d), indicating an actively proliferating status. Further, the percentage of Olig2+ cells positive for Ki67 was close to the percentage positive for GFAP (Figure 4i), suggesting that proliferative Olig2+ cells were mainly astrocytes in Mlc1-ErbB2V664E mice. A similar colocalization pattern was observed for Olig2 and nestin in Mlc1-ErbB2V664E mice (Supplementary Figure S8e). We examined the mRNA levels of a series of genes , and did not reveal a typical transcriptional pattern of subtype A1 or A2 for the reactive astrocytes in Mlc1-ErbB2V664E mice (Supplementary Figure S8f). Both A1 and A2 subtype-specific genes were actively transcribed, suggesting reactive astrocytes induced in the brain of Mlc1-ErbB2V664E mice were phenotypically heterogeneous.
|
study
| 100.0 |
Heterogeneity of reactive astrocytes might be caused by a complication of reactive microgliosis in Mlc1-ErbB2V664E brain. In addition to effects on astrocytes, reactive astrogliosis in diseased or injured brains is always accompanied by recruitment of reactive microglia . Indeed, we observed an increased prevalence of reactive microglia (Iba1+) in the brains of Mlc1-ErbB2V664E mice comparable to that of GFAP+ cells (Supplementary Figure S9a and b). These results showed that active ErbB signaling in mature astrocytes induced reactive responses with a molecular and cellular profile similar to that observed in reactive astrogliosis in injured or diseased brains. Considering the lack of ectopic ErbB activation in microglia or leukocytes that were infiltrated through the blood–brain barrier (Supplementary Figure S9c), our findings suggest that reactive responses of microglia in Mlc1-ErbB2V664E mice were induced by immunogenic factors released from reactive astrocytes. To confirm this idea, we examined Mlc1-ErbB2V664E mouse brains 3 days after Dox withdrawal. At this early stage, GFAP-expressing astrocytes started to appear in the cortices of Mlc1-ErbB2V664E mice, with some of them possessing Ki67+ nuclei. In contrast, Iba1+ cells in the same cortices did not increase, and did not have Ki67+ nuclei (Supplementary Figure S10a–d). These results were consistent with previous reports that reactive astrocytes release many factors including cytokines , and astrocytic gene targeting-induced astrogliosis results in reactive microglia responses . Indeed, we detected increase of cytokines including TGF-β2, interleukin-6 (IL-6), IL-1β, C-C motif chemokine ligand 2 (CCL2), and Ciliary neurotrophic factor (CNTF) in primary Mlc1-ErbB2V664E astrocytes (Supplementary Figure S10e). Noticeably, astrogliosis and microgliosis did not result in cell apoptosis. There were no terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labeling (TUNEL)+ cells revealed in Mlc1-ErbB2V664E brain or littermate controls on either 3 or 20 days after Dox withdrawal (Supplementary Figure S10h).
|
study
| 100.0 |
Compared with their littermate controls, Mlc1-ErbB2V664E mice consumed much less food and water after Dox withdrawal. Twenty days after Dox withdrawal, they exhibited malnourishment with significantly smaller body sizes and lighter weights (Figure 5a and b). To explore whether there was a peripheral problem causing anorexia, we dissected their digestive systems and discovered atrophy of the gastrointestinal tract and accessory organs (Figure 5c). Further investigation of gastrointestinal tract sections with hematoxylin and eosin (H & E) staining showed no pathological changes, although the stomachs were constricted because of lack of content (Figure 5d). There were no cells targeted by Mlc1-tTA in the gastrointestinal tract of Mlc1-ErbB2V664E mice (Supplementary Figure S11a and b). Therefore, anorexia was caused by an inhibition of neural circuits for feeding behavior. Remarkably, hypothalamic astrocytes have crucial roles in controlling feeding behavior , and hypothalamic inflammation is linked to anorexia . Indeed, astrocytic ErbB activation induced prevalent astrogliosis and associated inflammation in the brain including the hypothalamus, as indicated by intensively stained GFAP+ and Iba1+ cells there (Figure 5e). To screen for other possible pathological changes, mouse brains were sectioned for H & E staining. Astrocyte dysfunction and associated inflammation would disrupt the integrity of the brain–blood barrier and influence cerebrospinal fluid production . As a result, the ventricles of Mlc1-ErbB2V664E mice were larger than those of littermate controls. Nevertheless, other brain regions exhibited comparable sizes in the two groups (Figure 5f). Therefore, in animals with reactive astrogliosis and associated inflammation throughout the brain, anorexia was the predominant result, suggesting hypothalamic susceptibility.
|
study
| 100.0 |
Next, we investigated which downstream signaling pathways were activated by cell-autonomous activation of ErbB signaling in astrocytes. Because cultured astrocytes differ from in vivo astrocytes [25, 47], we first examined various candidate signaling proteins in brain tissues. Based on WB assays, Akt and Erk, the classic ErbB downstream signaling proteins that are important for astrogliosis, were indeed active in Mlc1-ErbB2V664E cortices (Figure 6a and b). In addition, the classic inflammatory signaling protein, STAT3, which is critical for astrogliosis [10, 13, 48], markedly increased in both level and activity in cortical tissues from Mlc1-ErbB2V664E mice (Figure 6a and b). Moreover, our immunostaining results revealed specific activation and increased levels of STAT3 protein in reactive astrocytes in Mlc1-ErbB2V664E mice (Figure 6c).
|
study
| 100.0 |
In addition, WB revealed that Src and FAK activity were significantly increased in the cortices of Mlc1-ErbB2V664E mice (Figure 6a and b). Moreover, these two non-receptor tyrosine kinases were specifically activated in reactive astrocytes, as indicated by the immunoreactivities of their multiple active forms being localized in GFAP+ cells (Figure 6d). Both Src and FAK are downstream of ErbB receptors [49, 50], and participate in regulating multiple brain functions [51, 52]. Note that the subcellular distribution pattern of FAK with phosphorylation at Y397 in reactive astrocytes was similar to that of the active form of Src (pY418) (Figure 6d), consistent with a previous report that FAK interacts with Src through phosphorylated Y397 . Src and FAK both mediate signaling pathways that regulate actin polymerization [49, 54]. Consistent with this idea, both protein and phosphorylation levels of profilin, an actin-binding protein that regulates actin polymerization and astrocytic morphology , was increased in the cortices of Mlc1-ErbB2V664E mice (Figure 7a and b).
|
study
| 100.0 |
Although cell-autonomous activation of ErbB signaling in mature astrocytes induced spontaneous astrogliosis with canonical features (Figure 4), inhibiting ErbB signaling in reactive astrocytes specifically suppressed their ability to develop hypertrophy (Figure 2). To investigate which signaling pathways mediated the effects of ErbB activation on the hypertrophy of reactive astrocytes, ErbB-activated downstream signaling proteins were examined in the injured cortices of Mlc1-dnEGFR and littermate control mice. Interestingly, WB revealed no reductions in phosphorylated Akt, Erk or STAT3, despite the fact that ErbB3 activity was reduced in injured tissues of Mlc1-dnEGFR mice (Figure 7a and b). Instead, phosphorylation of FAK and Src was significantly suppressed in injured tissues from Mlc1-dnEGFR mice (Figure 7a and b). Further, immunostaining revealed that stab wound injury in control brains induced upregulation of the active forms of FAK (pY861, pY397 and pY925) and Src (pY418) specifically in reactive astrocytes (Figure 7c and Supplementary Figure S12). However, both the number of cells with active FAK or Src and the average activity per cell were significantly reduced in the injured cortices of Mlc1-dnEGFR mice (Figure 7c–e). In concordance with the dnEGFR-induced suppression of reactive astrocytic expansion, phosphorylation of profilin, as well as its upstream regulating kinase ROCK , were suppressed in the injured cortices of Mlc1-dnEGFR mice (Figure 7a and b). Consistent with the finding that hypertrophic remodeling did not rely on GFAP expression (Figure 2), the discovery of a Src/FAK/profilin signaling pathway emphasized the involvement of active actin remodeling in the reactive responses of astrocytes.
|
study
| 100.0 |
Given the clear phosphorylation of STAT3 in the astrocytes of Mlc1-ErbB2V664E mice after Dox withdrawal (Figure 6a–c), we were surprised to see comparable STAT3 phosphorylation (pSTAT3) levels between injured cortical tissues from Mlc1-dnEGFR and littermate control mice (Figure 7a and b). However, as it is a typical inflammatory signal, STAT3 can be activated by various cytokines in addition to ErbB receptors . To ensure that other activators of the STAT3 pathway were unaffected in injured Mlc1-dnEGFR brains, we examined the inflammatory status of the injured cortices by Iba1 immunostaining to label reactive microglia and by real-time RT-PCR to assess the levels of the following cytokines: CXCL10, CCL2, IL-1β, IL-6, CNTF and TGF-β2. We found no difference in Iba1+ cell densities between Mlc1-dnEGFR and littermate mice in cortical regions adjacent to the sites of injury (Figure 8a and b). Moreover, with the exception of CXCL10 and CNTF, which showed no change, mRNA levels of CCL2, IL-1β, IL-6 and TGF-β2 were significantly increased in injured tissues from Mlc1-dnEGFR mice (Figure 8c). These results indicated that ErbB inhibition in reactive astrocytes did not suppress local inflammation induced by brain injury. Moreover, many studies have revealed that reduced astrocyte reactivity aggravates local inflammation [10, 13, 58–60]. Therefore, the observed increase in cytokine transcription in injured Mlc1-dnEGFR cortices may indirectly reflect attenuated astrocyte reactivity induced by cell-autonomous inhibition of ErbB signaling.
|
study
| 100.0 |
In summary, our findings revealed differentiated regulation of aspects of astrocyte reactivity by ErbB activation. Reactive astrocytes were situated in a niche comprising autonomously released pathological stimuli from various cell types. The non-altered signaling pathways in injured Mlc1-dnEGFR cortices, such as STAT3, were likely affected by multiple upstream regulators. In contrast, Src and FAK were directly regulated by ErbB signaling in reactive astrocytes for hypertrophic regulation (Figure 8h). In supporting our hypothesis, activities of downstream signaling stimulated by rhEGF or rhNRG1, including Akt, STAT3, FAK or Src, were compromised in primary Mlc1-dnEGFR astrocytes in comparison with that in control astrocytes (Figure 8d and e). In contrast, STAT3 activity induced by cytokine CNTF was not reduced in primary Mlc1-dnEGFR astrocytes. Instead, it was more increased by CNTF in primary Mlc1-dnEGFR astrocytes than that in control cells, in line with the in vivo observation that inflammatory status was aggravated in the injured cortices of Mlc1-dnEGFR mice (Figure 8f and g). The dissociated regulation of aspects of astrocyte reactivity indicated a multifaceted implication of reactive astrocytes in diseased or injured brains.
|
study
| 100.0 |
Reactive astrocytes are being considered as therapeutic targets for various neurological disorders [4, 39, 58]. Here, we showed the role of ErbB signaling as a positive upstream regulator in reactive astrogliosis. Different from overexpressing a secreted ligand , we targeted receptors, which specifically regulated the signaling in astrocytes in vivo by using mice genetically engineered for inducible gene expression. We showed that ErbB activation was sufficient to mediate the reactive responses of mature astrocytes, prompting molecular and morphological changes characteristic of injury and disease-induced astrogliosis (Figure 4). We also revealed that many intracellular signals critical for astrogliosis, including STAT3, were stimulated by ErbB activation in astrocytes (Figures 6 and 8). This is in marked contrast to FGF signaling that inhibits astrocyte reactivity , in spite of both ErbB receptors and FGFRs being able to activate similar sets of downstream signaling proteins [15, 16].
|
study
| 100.0 |
Interestingly, the inhibition of brain injury-induced ErbB signaling in reactive astrocytes suppressed hypertrophic remodeling without affecting proliferation (Figures 1, 2, 3). Further, Src/FAK activities were specifically blocked by inhibiting ErbB signaling in reactive astrocytes (Figure 7). Src and FAK are frequently connected to the regulation of cellular processes [53, 62, 63], and most of their effects occur through regulating actin remodeling [49, 50, 54, 64]. Given that hypertrophy could be established independently of GFAP in cortical reactive astrocytes (Figure 2g), we postulated that it was most likely caused by actin remodeling. Consistent with this hypothesis, we determined that phosphorylated profilin, which regulates actin polymerization, was reduced following ErbB inhibition in reactive astrocytes (Figure 7a and b). GFAP expression was also suppressed in injured cortices of Mlc1-dnEGFR mice (Figure 2d and e), which helped limit cellular process extension . Together, these results suggested that ErbB signaling positively regulated astrocyte reactivity, with prominent and direct effects on hypertrophic remodeling. Being a well-known sign of reactive responses, hypertrophy might actively participate in cell–cell contact, endocytosis, phagocytosis and migration by the mediation of actin remodeling. In-depth work is worth pursuing to define the pathological contribution of hypertrophy or other actin-involved cellular activities of glial cells in reactive astrogliosis.
|
study
| 100.0 |
There have been several reports suggesting that morphological changes and proliferation are dissociated in reactive astrocytes. First, injury-induced astrogliosis in GFAP/vimentin-deficient mice exhibits astrocyte proliferation comparable to that in wild-type mice . Moreover, spontaneous astrogliosis in mice deficient in FGFRs does not result in proliferation of reactive astrocytes . Immunogenic capacity is also likely dissociated from the proliferation and hypertrophy in reactive astrocytes, as spontaneous astrogliosis in mice deficient in Bax does not recruit reactive microglia . Nevertheless, the dissociated regulation of aspects of astrocyte reactivity is not clearly understood. The discrepancies between results obtained from Mlc1-ErbB2V664E and Mlc1-dnEGFR mice indicated that the proliferation of reactive astrocytes was regulated by signaling downstream of the initial reactive responses induced by ErbB activation and that it could be independent of ErbB receptor activation in astrocytes.
|
study
| 100.0 |
Many molecular signals released from various cell types are involved in regulating astrocyte proliferation after CNS injury. For example, FGF, Sonic hedgehog, endothelin-1 and STAT3 signaling pathways promote the proliferation of reactive astrocytes [18, 67–69]. Intriguingly, inhibiting inflammation abolishes both Sonic hedgehog activation and proliferation of reactive astrocytes in injured brains . STAT3 can be activated by FGFRs, ErbB receptors and cytokine receptors [57, 70, 71]. Note that reactive astrogliosis is usually complicated by interactions among reactive astrocytes, microglia, endothelial cells and other cell types . We found that although reactive astrocytes caused by cell-autonomous ErbB activation induced local inflammation (Supplementary Figure S9), immunogenic stimuli released from non-astrocytes could not be blocked by ErbB inhibition in astrocytes. In fact, the number of reactive microglia (Iba1+ cells) and cytokine mRNA levels were not reduced in injured cortices of Mlc1-dnEGFR mice (Figure 8a–c). Moreover, STAT3 activity (pSTAT3) was not reduced in either injured cortical tissues or CNTF-treated primary astrocytes from Mlc1-dnEGFR mice (Figures 7 and 8). These results suggest that some aspects of astrogliosis, including proliferation, are regulated by many coordinated signaling pathways that are stimulated by reactive astrocytes themselves or by other sources (Figure 8h, working model). Among them, the mechanism by which ErbB receptors and FGFRs antagonize each other in reactive astrocytes and cross-talk with STAT3 pathways through inflammatory signals deserves further investigation.
|
study
| 100.0 |
No matter how heterogeneous astrocytes in different brain regions are , cell-autonomous ErbB activation in astrocytes induced similar reactive responses throughout the brain (Figure 4 and Supplementary Figures S7–S9). Intriguingly, prevalent astrogliosis and associated brain inflammation predominantly caused anorexia in animals (Figure 5). The susceptibility of different brain regions to reactive astrogliosis warrants further investigation. Nevertheless, the functional roles of astrogliosis in pathological development are still debated. For example, glial scar formation over injury sites has been shown to restrain damage but perturb axon regrowth in the CNS [72, 73]. However, a recent study reported astrocytic scars to be beneficial for axon regeneration . This paradox was solved by a more recent work that disrupting the molecular signaling specifically transforming reactive astrocytes into scar-forming astrocytes leads to enhanced axon regrowth, indicating a functional dissociation of different types of astrocytes that are sequentially induced during the development of reactive astrogliosis . Moreover, the type of CNS insult influences the outcome of astrogliosis in a context-dependent manner [5, 7]. Therefore, signaling pathways regulating astrocyte reactivity must be assessed in specific contexts to comprehensively evaluate the targets that contribute to or inhibit astrogliosis. It would be intriguing to determine whether ErbB inhibition alters functions other than scar formation in reactive astrocytes and whether these functions are required for specific pathological processes. Our work shed light on the molecular mechanisms that regulate astrocyte reactivity and described inducible animal models that could be useful for future investigations into the function of astrogliosis and potential therapeutic targets.
|
study
| 99.9 |
Mlc1-tTA transgenic mice were from the RIKEN Bioresource Center (stock no. RBRC05450). Transgenic mice TRE-ErbB2V664E (stock no. 010577) and TRE-dnEGFR (stock no. 010575) were from the Jackson Laboratory. Mlc1-tTA;TRE-ErbB2V664E (Mlc1-ErbB2V664E) and Mlc1-tTA;TRE-dnEGFR (Mlc1-dnEGFR) were obtained by breeding Mlc1-tTA mice with TRE-ErbB2V664E or TRE-dnEFGR mice, respectively. Primers Mlc1U-657 (5′- AAATTCAGGAAGCTGTGTGCCTGC-3′) and mtTA24L (5′- CGGAGTTGATCACCTTGGACTTGT-3′) with a 680- bp PCR product were used for genotyping of Mlc1-tTA, whereas primers 9707 (5′- AGCAGAGCTCGTTTAGTG-3′) and 9708 (5′- GGAGGCGGCGACATTGTC-3′) with a 625- bp PCR product for that of TRE-ErbB2V664E, and primers 9013 (5′- TGCCTTGGCAGACTTTCTTT-3′) and 7554 (5′- ATCCACGCTGTTTTGACCTC-3′) with a 318-bp PCR products for that of TRE-dnEGFR. Unless indicated, mice were housed in a room with a 12-h light/dark cycle with access to food and water ad libitum. Mlc1-dnEGFR, Mlc1-tTA and Mlc1-ErbB2V664E mice with either sex and their littermate control mice with matched sex were used for experiments. Animal experiments were approved by the Institutional Animal Care and Use Committee of the Hangzhou Normal University.
|
study
| 99.94 |
The pregnant mice and their offspring were fed with tetracycline analog, Dox (0.5 mg/ml, 10 ml/day), in drinking water to inhibit the expression of ErbB2V664E in Mlc1-ErbB2V664E mice from embryonic to indicated postnatal days. Water bottles were wrapped with foil to protect Dox from light. Mlc1-dnEGFR mice and their littermate controls were not treated with Dox for they did not exhibit developmental difference. Withdrawal from Dox induces the expression of ErbB2V664E and dnEGFR in astrocytes of Mlc1-ErbB2V664E and Mlc1-dnEGFR mice. All used littermate control mice were treated the same.
|
study
| 100.0 |
Information on commercial rabbit antibodies is as follows: EGFR (1902-1, 1:5000 for WB) and pEGFR (Tyr1068, 1727-1, 1:2500 for WB) were from Epitomics (Burlingame, CA, USA); ErbB2 (sc-284, 1:1000 for WB, 1:200 for immunofluorescence (IF)), ErbB3 (sc-285, 1:5000 for WB, 1:50 for IF), ErbB4 (sc-283, 1:5000 for WB) and pFAK (Tyr397, sc-11765-R, 1:200 for WB, 1:50 for IF) were from Santa Cruz (Dallas, TX, USA); ROCK2 (9029, 1:1000 for WB), Erk1/2 (9102, 1:5000 for WB), pErk1/2 (Thr202/Tyr204, 4370, 1:5000 for WB), Akt (9272, 1:5000 for WB), pAkt (Ser473, 9271, 1:3000 for WB) and pSTAT3 (Tyr705, 9145, 1:2000 for WB, 1:100 for IF) were from Cell Signaling (Danvers, MA, USA); Ki67 (RB-9043-P1, 1:500 for IF) was from Thermo (Fremont, CA, USA); Glyceraldehyde-3-phosphate dehydrogenase (ABS16, 1:2500 for WB), GFAP (AB5804, 1:2000 for IF) and Olig2 (AB9610, 1:500 for IF) were from Millipore (Temecula, CA, USA); pROCK2 (ab182648, 1:500 for WB), pSrc (Tyr418, ab4816, 1:1000 for WB, 1:100 for IF), Acsbg1 (ab65154, 1:500 for IF), pErbB3 (Tyr1328, ab133459, 1:2500 for WB, 1:100 for IF) and pErbB4 (Tyr1284, ab109273, 1:2500 for WB) were from Abcam (Cambridge, MA, USA); pProfilin (Tyr129, PP4751, 1:1000 for WB) was from ECM Biosciences (Versailles, KY, USA); AQP4 (HPA014784, 1:2500 for IF) and GS (G2781, 1:200 for IF) were from Sigma (St Louis, MO, USA). Information on commercial mouse antibodies is as follows: pErbB2 (Tyr1140, AP3781q, 1:2500 for WB) was from Abgent (San Diego, CA, USA); STAT3 (9139,1:1000 for WB, 1:1600 for IF)) was from Cell Signaling; MBP (MAB382, 1:1000 for WB), GFAP (MAB360, 1:3000 for WB, 1:2000 for IF) and Iba1 (MABN92, 1:1000 for IF) were from Millipore; Nestin (ab11306, 1:100 for IF) and Aldh1L1 (ab56777, 1:200 for IF) was from Abcam; RC2 (1:100 for IF) was from Developmental Studies Hybridoma Bank (Iowa city, IA, USA). Information on commercial goat antibodies is as follows: pFAK (Tyr861, sc-16663, 1:200 for WB, 1:50 for IF) and pFAK (Tyr925, sc-11766, 1:200 for WB, 1:50 for IF) were from Santa Cruz. Rat antibodies to CD45 (103101, 1:200 for IF) was from BioLegend (San Diego, CA, USA). Alexa Fluor 488- and Alexa Fluor 594-conjugated secondary antibodies (1:1000 for IF) were purchased from Invitrogen (Rockford, IL, USA), and horseradish peroxidase-conjugated secondary antibodies (1:5000 for WB) were from CWBIO (Beijing, China). Recombinant rat CNTF (cat. no. 557-NT-010/CF), rhEGF (cat. no. 236-EG-200) and recombinant human NRG1 (cat. no. 377-HB-050/CF) were from R&D (Minneapolis, MN, USA). Dox and TRIzol were from Sangon (Shanghai, China). Fluorescent mounting medium was from CWBIO. Protease inhibitor cocktail and bioinchoninic acid assay kit were from Thermo. EZ-ECL was from Biological Industries (Cromwell, CT, USA). PrimeScript Reverse Transcriptase was from Takara (Shiga, Japan). SYBR Green PCR mixture was from Bio-Rad (Hercules, CA, USA). All other chemicals were from Sigma.
|
other
| 99.9 |
Cortical stab injuries were operated under a stereotaxic apparatus (RWD68025) on mice at age of P30. Mice were anesthetized by 1% pentobarbital (50 mg/kg, i.p.). The parietal skull was exposed and made a small hole by a micro-drill. Traumatic brain injury was made by stabbing with a needle (0.5 mm in diameter) in the cortex at 1.5 mm lateral to the midline, 2 mm posterior to bregma and 1 mm deep from the surface of meninges. Different days after the surgery, brains were isolated and fixed with 4% paraformaldehyde (PFA) in 0.1 M phosphate buffer (PB) and sectioned by vibrating microtome. Astrogliosis was verified by immunostaining for GFAP. Image J (NIH, Bethesda, MD, USA) was used to analyze and quantify the size and number of GFAP+ cells, as well as scar thickness, in consecutive sections over the injury sites. To study the scar formation, mouse cortices were stabbed by a scalpel blade (#11, angled with 20-mm cutting edge and 0.4-mm thickness, Shanghai Chengyuan Medical Supplies Factory, Shanghai, China) parallel to the longitudinal fissure at 1.5 mm lateral to the midline, 2 mm posterior to bregma and 1 mm deep from the surface of meninges. For WB and real-time RT-PCR, injured sites were identified through the drilled holes in the skull and injured cortical tissues were isolated from uninjured part. We usually combined the injured cortical tissues from several mice with the same genotype to serve one sample for each batch of experiment.
|
study
| 100.0 |
AAV-TRE-YFP plasmids were constructed by standard methods, packaged as AAV9 viruses, and produced with titers of 1×1012 particles per ml by OBio (Shanghai, China). Mice were anesthetized by 1% pentobarbital (50 mg/kg, i.p.) and mounted at stereotaxic apparatus (RWD68025). AAV-TRE-YFP (1 μl) was injected into the cortex (from bregma in mm, cortex, M-L: ±0.9, A-P: −1.0, D-V: 1.0) under the control of micropump (KDS310) at speed of 0.07 μl/min. Injecting needles (Hamilton NDL ga33/30 mm/pst4, Switzerland) were withdrawn 5 min after injection. To observe the cell specificity of Mlc1-tTA, injected brains were isolated and fixed in 4% PFA in PB 1 day after injection to avoid reactive gliosis induced by virus injection. To analyze the sizes of reactive astrocytes, viruses were injected near the injury sites 1.5, 3 or 7 days post injury, and injected brains were isolated and fixed 1 day later. Fixed brains were sectioned and immunostained for GFAP or Acsbg1, and images were taken by a Zeiss LSM710 confocal microscope (Berlin, Germany). Morphology of cells was observed by maximum projection of Z-stacked images taken under a 40X oil-immersion objectives. When analyzing the Z-stacked images, it was shown that the astrocytic areas around cell bodies within 6-μm Z-axial range were the biggest (Supplementary Figure S5b). For the quantitative purpose as shown in Figure 2, big fields were captured under a 20X objective, and the pinhole was set at 200 in order to acquire YFP+ cells with most of cell bodies and their processes in optical slices with 6.1-μm thickness. Captured cells with complete cell bodies were subjected to size measurement. YFP fluorescence was imaged with exactly same scanning conditions for paired experiments, and sizes of YFP-labeled cells were measured by Image J.
|
study
| 100.0 |
AAV-TRE-YFP-labeled cells in fixed brain slices were filled with dye by methods described previously [31, 32]. Briefly, mice with injured cortices 1 day after stereotaxic injection of AAV-TRE-YFP were anesthetized by 1% pentobarbital and transcardially perfused with oxygenated Ringer’s solution (1.35 mM NaCl, 0.05 mM KCl, 0.01 mM MgCl2·6H2O, 0.013 mM Na2HPO4, 0.15 mM NaHCO3, 0.02 mM CaCl2·2H2O, 0.11 mM dextrose, 0.0085 mM xylocaine) and then with 4% PFA in phosphate-buffered saline (PBS; pH 7.4). Isolated brains were postfixed in 4% PFA in PBS for 1 h, and then sectioned into 75-μm slices by vibrating microtome. YFP-labeled astrocytes in the cerebral cortex were identified under a fluorescence microscope, and were patched and impaled with glass micropipettes (o.d., 1.00 mm; i.d., 0.58 mm; resistance, 100–400 MΩ) that had been backfilled with 10 mM Alexa Fluor 568 (Invitrogen) in 200 mM KCl. Patched astrocytes were iontophoretically injected with the dye by using 1-s pulses of negative current (0.5 Hz) for 2 min. After several cells were filled, the slices were placed in ice-cold 4% PFA overnight and then mounted under coverslips. Dye filled astrocytes were observed under a Zeiss710 confocal microscope equipped with a 40X oil-immersion objective, and Z-stack images were captured and projected to evaluate the subcellular distribution of injected dye and YFP.
|
study
| 100.0 |
Mouse brains were isolated and fixed in 4% PFA in 0.1 M PB (0.019 M NaH2PO4, 0.089 M Na2HPO4, pH 7.4) overnight, and then washed with 0.1 M PB twice. The fixed brains were kept in 0.1 M PB with 1% ProClin 200 in 4 °C until sectioned by vibrating microtome. Soft agar-embedded mouse brains were cut into 50 μm sections and subjected to immunostaining as previously described . Briefly, brain slices were incubated with blocking buffer (10% fetal bovine serum and 0.1% Triton-X-100 in 0.1 M PB) for 1 h at room temperature, and then incubated at 4 °C overnight with primary antibodies diluted in blocking buffer. After washing three times with PB, samples were incubated at room temperature for 1 h with Alexa-488 or −594 secondary antibody, and then washed and mounted on adhesion microscope slides (CITOTEST) with fluorescent mounting medium. For co-immunostaining, samples were incubated with the second primary antibody the next morning for 1 h at room temperature before staining with the secondary fluorescence antibodies. Images were taken by a Zeiss LSM710 confocal microscope with exactly same scanning conditions for paired experiments, and analyzed by Image J.
|
study
| 99.94 |
Different brain regions were isolated and homogenized. For injured brains, only the tissues surrounding the injury sites were collected and homogenized. Homogenates in lysis buffer (10 mM Tris-Cl, pH 7.4, 1% NP-40, 0.5% Triton-X 100, 0.2% sodium deoxycholate, 150 mM NaCl, 20% glycerol, protease inhibitor cocktail) at ratio of 1 ml per 100 mg tissue were lysed overnight in 4 °C. Lysates were centrifuged at 12 000 g and 4 °C for 30 min to get rid of the unsolved debris. Concentration of the supernatant was measured by BCA assay. Proteins in samples were separated by 6–12% sodium dodecyl sulfate–polyacrylamide gel electrophoresis, transferred to a polyvinylidene fluoride membrane (Millipore), and then incubated with indicated primary antibodies at 4 °C overnight after blocking by 5% non-fat milk solution in Tris-buffered saline with Tween-20 (50 mM Tris, pH 7.6, 150 mM NaCl, 0.1% Tween 20) for 1 h at room temperature. Next day, the membranes were washed by Tris-buffered saline with Tween-20 for three times and incubated with the secondary antibodies for 1 h at room temperature. Membranes were washed again and incubated with substrate EZ-ECL for visualization of chemiluminescence by exposure to X-ray films or Bio-Rad GelDOCXR+ Imaging System. Intensities of protein bands were measured by Image J (NIH), and statistical analysis was performed after subtraction of the background intensity.
|
study
| 100.0 |
Fixed tissues were kept in 0.1 M PB with 1% ProClin 200 in 4 °C until sectioned by vibrating microtome. Soft agar-embedded mouse brains were cut into 30 μm sections and subjected to H & E staining. In brief, brain slices were stained with hematoxylin solution for 15 min and then rinsed in running tap water for 10 min before staining with eosin solution for 5 min. After rinsing with running tap water for 10 min, stained slices were mounted on adhesion microscope slides with resin-based mounting medium. Phase contrasted images were taken by a Zeiss Observer A1 inverted microscope under bright field with transmitted-light.
|
other
| 99.8 |
Total RNA was extracted from isolated mouse brains using TRIzol following the manufacturer’s protocol. Complementary DNA was synthesized by using the PrimeScript Reverse Transcriptase. Real-time PCR was performed in four repeats for each sample by using SYBR Green PCR mixture with the Bio-Rad CFX96 real-time PCR system as previously described . Relative mRNA levels were analyzed by Bio-Rad CFX Manager. Transcripts of targeted genes were normalized to these of mouse 18s ribosomal RNA gene in the same samples. Primers for 18s ribosomal RNA were 5′- CGGACACGGACAGGATTGACA-3′ and 5′- CCAGACAAATCGCTCCACCAACTA-3′ with a 94-bp PCR product. Primers for mouse gene EGFR and transgene dnEGFR were 5′- TCCTGCCAGAATGTGAGCAG-3′ and 5′- ACGAGCTCTCTCTCTTGAAG-3′ with a 500-bp PCR product. Primers for mouse gene IL-6 were 5′- GGGACTGATGCTGGTGACAACC-3′ and 5′- CATGTGTAATTAAGCCTCCGACTTGTG-3′ with a 128- bp PCR product. Primers for mouse gene IL-1β were 5′- GGCAGGCAGTATCACTCATTGTG-3′ and 5′- TGTCCTCATCCTGGAAGGTCC-3′ with an 84- bp PCR product. Primers for mouse gene CCL2 were 5′- TCACCTGCTGCTACTCATTC-3′ and 5′- GTAGGTTCTGATCTCATTTGGTTCC-3′ with a 205-bp PCR product. Primers for mouse gene CXCL10 were 5′- GTCTGAGTGGGACTCAAGGGATCCC-3′ and 5′- CATCGTGGCAATGATCTCAACACGT-3′ with a 155- bp PCR product. Primers for mouse gene TGF-β2 were 5′- TGCTTCGAATCTGGTGAAGGCA-3′ and 5′- GGAGAGCCATTCACCCTCCGCT-3′ with a 181- bp PCR product. Primers for mouse gene CNTF were 5′- TTTCGCAGAGCAATCACC-3′ and 5′- AATTGTGACAGGCATCC-3′ with a 433-bp PCR product. Primers for mouse gene H2-D1 were 5′-TCCGAGATTGTAAAGCGTGAAGA-3′ and 5′- ACAGGGCAGTGCAGGGATAG-3′ with a 204-bp PCR product. Primers for mouse gene Serping 1 were 5′- ACAGCCCCCTCTGAATTCTT-3′ and 5′- GGATGCTCTCCAAGTTGCTC-3′ with a 299-bp PCR product. Primers for mouse gene H2-T23 were 5′- GGACCGCGAATGACATAGC-3′ and 5′- GCACCTCAGGGTGACTTCAT-3′ with a 212-bp PCR product. Primers for mouse gene Ggta1 were 5′- GTGAACAGCATGAGGGGTTT-3′ and 5′- GTTTTGTTGCCTCTGGGTGT-3′ with a 115-bp PCR product. Primers for mouse gene Iigp1 were 5′- GGGGCAATAGCTCATTGGTA-3′ and 5′- ACCTCGAAGACATCCCCTTT-3′ with a 104-bp PCR product. Primers for mouse gene Gbp2 were 5′- GGGGTCACTGTCTGACCACT-3′ and 5′- GGGAAACCTGGGATGAGATT-3′ with a 285- bp PCR product. Primers for mouse gene Fkbp5 were 5′- TATGCTTATGGCTCGGCTGG-3′ and 5′- CAGCCTTCCAGGTGGACTTT-3′ with a 194-bp PCR product. Primers for mouse gene Psmb8 were 5′- CAGTCCTGAAGAGGCCTACG-3′ and 5′- CACTTTCACCCAACCGTCTT-3′ with a 121- bp PCR product. Primers for mouse gene Srgn were 5′- GCAAGGTTATCCTGCTCGGA-3′ and 5′- TGGGAGGGCCGATGTTATTG-3′ with a 134- bp PCR product. Primers for mouse gene Amigo2 were 5′- GAGGCGACCATAATGTCGTT-3′ and 5′- GCATCCAACAGTCCGATTCT-3′ with a 263-bp PCR product. Primers for mouse gene Clcf1 were 5′- CTTCAATCCTCCTCGACTGG-3′ and 5′- TACGTCGGAGTTCAGCTGTG-3′ with a 176-bp PCR product. Primers for mouse gene Ptx3 were 5′- AACAAGCTCTGTTGCCCATT-3′ and 5′- TCCCAAATGGAACATTGGAT-3′ with a 147-bp PCR product. Primers for mouse gene S100a10 were 5′- CCTCTGGCTGTGGACAAAAT-3′ and 5′- CTGCTCACAAGAAGCAGTGG-3′ with a 238-bp PCR product. Primers for mouse gene Sphk1 were 5′- GATGCATGAGGTGGTGAATG-3′ and 5′- TGCTCGTACCCAGCATAGTG-3′ with a 135-bp PCR product. Primers for mouse gene Cd109 were 5′- CACAGTCGGGAGCCCTAAAG-3′ and 5′- GCAGCGATTTCGATGTCCAC-3′ with a 147- bp PCR product. Primers for mouse gene Ptgs2 were 5′- GCTGTACAAGCAGTGGCAAA-3′ and 5′- CCCCAAAGATAGCATCTGGA-3′ with a 232-bp PCR product. Primers for mouse gene Emp1 were 5′- GAGACACTGGCCAGAAAAGC-3′ and 5′- TAAAAGGCAAGGGAATGCAC-3′ with a 183- bp PCR product. Primers for mouse gene Slc10a6 were 5′- GCTTCGGTGGTATGATGCTT-3′ and 5′- CCACAGGCTTTTCTGGTGAT-3′ with a 217- bp PCR product. Primers for mouse gene Tm4sf1 were 5′- GCCCAAGCATATTGTGGAGT-3′ and 5′- AGGGTAGGATGTGGCACAAG-3′ with a 258-bp PCR product. Primers for mouse gene B3gnt5 were 5′- CGTGGGGCAATGAGAACTAT-3′ and 5′- CCCAGCTGAACTGAAGAAGG-3′ with a 207- bp PCR product. Primers for mouse gene Cd14 were 5′- GGACTGATCTCAGCCCTCTG-3′ and 5′- GCTTCAGCCCAGTGAAAGAC-3′ with a 232- bp PCR product. Primers for mouse GFAP variant 1 gene were 5′- GACTATCGCCGCCAACTGCA-3′ and 5′- CTAAGGGAGAGCTGGCAGGG-3′ with a 447-bp PCR product. Primers for mouse GFAP variant 2 gene were 5′- GACTATCGCCGCCAACTGCA-3′ and 5′- TCACATCACCACGTCCTTGT-3′ with a 453-bp PCR product.
|
study
| 100.0 |
pFlag-ErbB2, pcDNA3-ErbB3, pFlag-ErbB4 and pcDNA3-EGFR were used in previous reports [77–79]. DnEGFR complementary DNA was amplified from genomic DNA of TRE-dnEGFR mice and cloned into pcDNA3.1/myc-His (−)A vector by using cloning sites BamHI and XhoI. Constructed plasmid was purified and sequenced. Fused myc-tagged dnEGFR was verified by detecting a band around 105 kDa in lysates of transfected HEK293 cells using anti-myc antibody (9E10) for WB.
|
study
| 100.0 |
HEK293 cells were grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 50 U/ml penicillin and 50 μg/ml streptomycin, and were transfected using polyethylenimine as previously described . Briefly, cells were cultured in six-well plates to 80% confluence and incubated for 6 h with precipitates formed by 2 μg of plasmid DNA and 2 μl of 0.5% (wt/vol, pH 7.0) polyethylenimine (Sigma-Aldrich, catalog no. 40 872-7). After replacing with fresh medium, cells were cultured in DMEM containing 10% FBS for 24 h before harvesting.
|
study
| 100.0 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.