Transcriptional and Functional Profiling of Human Embryonic Stem Cell-Derived Cardiomyocytes
Feng Cao1., Roger A. Wagner2., Kitchener D. Wilson1,3., Xiaoyan Xie1, Ji-Dong Fu6, Micha Drukker4, Andrew Lee1, Ronald A. Li6, Sanjiv S. Gambhir1,3, Irving L. Weissman4, Robert C. Robbins5, Joseph C. Wu1,2*
Citation: Cao F, Wagner RA, Wilson KD, Xie X, Fu J-D, et al. (2008) Transcriptional and Functional Profiling of Human Embryonic Stem Cell-Derived Cardiomyocytes. PLoS ONE 3(10): e3474. doi:10.1371/journal.pone.0003474
Summary: Many applications of human embryonic stem cells (hESCs) have been explored over the past 48 years; one focus has been the use of human embryonic stem cell-derived cardiomyocytes (hESC-CMs) in promoting repair/recovery after myocardial infarction. There are hopes of not only preserving cardiac function, but actual regeneration of diseased muscle because stem cells provide a potentially limitless source of cells. However, the pluripotency of stem cells also presents the serious risk of teratoma formation.
Feng et al. sought to characterize the molecular networks governing the differentiation of cardiomyocytes with an eye toward the transplantation of hESC-CMs to myocardial ischemia in vivo. To do so, both genomic and noninvasive imaging tools were utilized to understand the biological processes that could form the foundation of future stem cell therapy.
Feng et al. differentiated hESCs into cardiomyocytes utilizing the methods outlined in the Figure 1, below. They utilized RT-PCR analysis to track expression in the hESC-derived EBs as they differentiated into beating clusters. The expected stem cell markers (Oct4, NANOG, Rex1) were all observed early on, while early stage cardiac transcriptional factors (such as Nkx2.5 and MEF2C) appeared later between 14 and 28 days.
Figure 1: Schematic outlining the cardiomyocyte differentiation experimental design. The cells were maintained in an undifferentiated state on a feeder layer of mouse embryonic fibroblasts (MEFs). After the appearance of beating clusters in the embryoid bodies (EBs), the cells were separated by Percoll density gradient purification, which allowed for cardiomyocyte-enriched populations ranging from 40-45% beating EBs expressing the cardiac marker troponin-T as determined by FACS analysis.
Primary ventricular cardiomyocytes were used in microarray analysis. The significance analysis of microarrays (SAM) algorithm was then used to identify those genes which had changed expression significantly during differentiation from pluripotent hESCs to fetal heart cells.
In addition to performing transcriptional profiling of the cells, the authors also sought to compare their cultured cardiomyocyte population with what would likely be an optimal cell population for transplantation. Specifically comparisons of electrophysiological readings and energy metabolism revealed several points. First, these in vitro differentiated cells, while capable of contraction, had not yet faced the biomechanical stresses of in vivo cardiac development. Additionally, of the hESC-CMs, the ventricular-like derivatives were most like the ideal primary fetal heart cardiomyocytes, in terms of resting membrane potential.
Finally, the authors examined the effects of transplantation of hESC-CM into ischemic regions of the left ventricle of mice. These mice showed improvement, as measured by increased angiogenesis as well as left ventricular fractional shortening (LVFS) in comparison with controls. However, based upon a histologic evaluation of the transplanted hESC-CMs, there was very minimal integration of fluorescently tagged cells in the infracted areas as has been noted in previous studies. The authors hypothesize that the improvement seen in week 8 may relate to paracrine factors.
Significance: Through the research conducted, it was shown that hESC-CMs express cardiomyocyte genes at levels similar to those in 20-week fetal heart cells, which makes them promising candidates for use in vivo. Observations also indicated the importance of environment in the development of the hESC-CMs that fit the necessary parameters for potential stem cell therapeutics.
5 comments:
Very interesting article.
First, noting that cells in the body require certain molecular signals for differentiation and survival, do you think that these cardiomyocytes might differ from cardiomyocytes in human hearts since the researchers promoted differentiation with FBS and L-Gln?
2. Cell sorting revealed that only 43% of the cells isolated by Percoll density gradient separation were positive for cardiac troponin-T. What kind of cells are the remaining 57%? The researchers did test to see if undifferentiated hESCs might cause terratomas in the mouse model, but they never described the actual spectrum of cells forming this 57%.
3. Since the hESCs were maintained using mouse embryonic fibroblasts, do you think that might have affected their gene expression and morphology since mouse cells will be releasing molecular signals different from human cells?
4. What do you think caused the death of the hESC-CMs in the murine models? Why are undifferentiated hESCs not affected but instead form terratomas in the same setting?
Did the paper discuss how the hESC-CM were transplanted into the mice? Specifically, how the implanted cells interacted with the natural cells. Did the implanted cells develop the same size, shape, beating, etc. as the natural cells?
Hi Chian, I was wondering if this paper talked about any possible methods of recreating the in-vivo biomechanical environment for the purposes of optimizing the development hESC-CMs for therapeutics. Also, how often were they running RT-PCR? I didn't see mention of any downregulation or "turning off" of genes over the 28 day period, but it seems to me like this would probably happen.
Thanks,
Jason
To Spectator:
1. Yes, it is definitely possible that these cardiomyocytes differentiated from hESCs may differ from those found in human hearts. The researchers induced differentiation using FBS and L-Gln because they somehow affect the molecular signaling cascade for differentiation. However, depending on where within the cascade, these factors influence the signaling, it may or may not have the same/as complete of an effect. In terms of the markers and indicators (i.e. beating), these differentiated cardiomyocytes exhibit the same characteristics as those within the human heart.
2. That is a very good point. Although the researchers did not discuss what the remaining 57% of cells consisted of, they probably included a variety of cells such as undifferentiated hESCs, other muscle cells, etc.
3. The use of mouse embryonic fibroblasts as a feeder layer for the hESCs could very possibly have affected the gene expression and differentiation pathways of the hESCs. This is a concern that has been expressed time and time again, but is difficult to overcome as some sort of maintenance layer is generally necessary.
4. The hESC derived cardiomyocytes may have died because they were unanchored in the mouse models. While these cells may have exhibited some of the markers of cardiomyocytes, there are significant differences in the cellular environment of in vitro cell culture versus the in vivo murine model. Since these cells were grown and differentiated in vitro, they may not have developed the proper ECM and cellular infrastructure to survive and integrate into the murine cardiomyocyte population. Additionally, the hESC-CMs lacked paracrine signaling, which appears to important in the cell population’s survival. However, the undifferentiated hESCs may have been able to survive, and grow into teratomas precisely because of their undifferentiated nature. These hESCs may have been able to respond to cellular signaling within the murine model.
To Dana:
The hESC-CM were introduced into ischemic myocardium via direct injection of 1x106 Fluc+/eGFP+
hESC-derived cardiomyocytes in 40 µl of PBS (n=16), with approximately 90% of the cells introduced dying during the first 3 weeks post-injection. It was observed that the cells remained localized within the heart tissue, but they were not really integrated with the host myocardium. There was little comparison of beating, size, etc. of hESCs to the observations of the natural cells.
To Jason:
The paper briefly mentions that the high hESC-CM death after transplantation might be somewhat alleviated by utilizing tissue engineering to create some synthetic tissue constructs to organize and support the transplanted cells. However, the researchers do not go into any further depth regarding this matter. The RT-PCR analysis was conducted over 42 days, but there was no mention of the frequency of RT-PCR. I would assume roughly daily based upon the level of precision with which the data was reported. I would also agree that there a "turning off" of certain genes and a "turning on" of others should have been noted.
Post a Comment