Human embryonic stem cells (hESCs) can be induced to differentiate into specialized cell types and offer great promise for the development of novel regenerative therapies. Much work remains to be done, however, before effective, safe, well-tested therapies can be delivered to patients. One of the first tasks at hand is to characterize in great detail the biological mechanisms underlying the ability of hESCs to either differentiate into specialized cell types in culture or to remain in an undifferentiated state. The ability of hESCs to switch developmental fates must be carefully analyzed in order to engineer these cells into helpful therapeutic remedies. The capacity of hESCs to switch to different developmental fates results from powerful gene regulation processes that can silence differentiation-triggering genes in the undifferentiated state and release silencing for some of these genes upon differentiation. The chemical modification (methylation) of DNA is one of the two known silencing mechanisms used by hESCs. Mouse ESCs deficient in their ability to methylate DNA fail to differentiate and die upon induction of differentiation. Similarly, changes in the DNA methylation profiles of adult cells are a hallmark of cancer initiation. Despite the importance of DNA methylation, the precise distribution and stability of this modification in the genomes of hESCs has not been assessed comprehensively. Similarly, the manner by which global DNA methylation profiles are reprogrammed during differentiation is unknown. Here, we propose to develop an innovative method allowing the genome-wide high-resolution mapping of DNA methylation in human cells. This method will be applied both to two distinct lines of hESCs in their undifferentiated state and to hepatocyte-like cells derived from the differentiation of hESCs into liver. Comparing global DNA methylation profiles from two undifferentiated hESC lines will both define a core methylated profile and highlight the extent of natural variation that exists between hESC lines. Analysis of methylated loci in undifferentiated hESCs will contribute new molecular markers for hESCs characterization. This will also provide a useful diagnostic tool by which to control the quality of long term hESCs cultures. For instance, cultures showing drift in the distribution of methylated DNA sequences outside of normal variation should not be used for therapeutic development since these cultures might develop altered properties, including improper differentiation and potential tumorigenicity. Finally, the comparison of methylation profiles of hESCs-derived hepatocyte-like cells to normal hepatocytes from healthy individuals is likely to offer novel clues as to what genes are turned on or off during the differentiation process. This will lead to improved differentiation techniques and represent a gold standard by which to validate a differentiation procedure.
Statement of Benefit to California:
The delivery of novel human embryonic stem cell-based regenerative therapeutics to California patients will require the development of strict quality and safety standards for the culturing and differentiation of these cells. Our work will benefit both aspects in the following way. 1) Our work will lead to the development of new molecular markers to characterize hESCs in their undifferentiated state. It will also provide sensitive diagnostic tools to ensure that these cells have not drifted away from their undifferentiated state during long-term culturing, thus minimizing the risks associated with such drift. 2) Our work will lead to improvements in the techniques used to differentiate hESCs into specialized therapeutic cells by identifying specific genes that are turned on or off in the differentiation process. These genes will provide novel markers for improving and verifying differentiation techniques. Moreover, our work will ultimately allow to compare the genomic characteristics of hESC-derived differentiated cells to those of normal cells of the same intended cell type (our work targets liver cells). This comparison will validate the differentiation techniques and will constitute a guarantee of safety and efficacy.
SYNOPSIS: This proposal aims to study DNA methylation in hESCs. The approach will utilize a DNA methylation tiling array to map all methylation sites in the genome, and changes in methylation patterns will be determined after differentiation is induced. The first specific aim will optimize the experimental procedures in using a commercially available antibody to bind and recover methylated DNA, and investigate methylation sites using commercially-available whole genome arrays. Specific Aim 2 will test the hypothesis that methylation activity affects the reprogramming of gene expression in hESCs. The modification of methylation patterns will be examined in differentiated hepatocytes. SIGNIFICANCE AND INNOVATION: The premise of this proposal is that DNA methylation plays a role in ESC differentiation. This proposal is significant in that it aims to determine the DNA methylation pattern of ESC genome-wide. This work is original in that it has not been done before, but it is not particularly innovative or creative. STRENGTHS: The strength of the proposal is in the generation of a reference map of DNA methylation profiles in the genome. This would be an important contribution to the field. Another strength of the proposal is the PI’s collaborator Dr. Farnham, who has a great track record in genome-wide analyses. WEAKNESSES: The main weakness of this proposal is its low creativity and innovativeness. Genome-wide mapping is an established technology, thus the approach is low risk, but here it is very expensive for what would be gained. If this work is truly a worthwhile endeavor, NIH should fund it. Another weakness is the uncertainty by which true ChIP signal will be discerned from noise, which is dependent upon antibody quality and thus far is not well established. Further, it is not clear how this huge amount of data will lead to a better understanding of ESC regulation. There could be widespread or no changes in DNA methylation upon ESC differentiation. It is not clear how the author will determine how much of this is connected to ESC regulation as opposed to a consequence of cell differentiation. DISCUSSION: Reviewers believe that the applicant could accomplish this work and that solid data might emerge, and they believe that the work needs to be done. The question is since genome-wide methylation studies have not been done in animal cells should hESC be the first? The cost is very high, and because most of the funds will be used to pay for the tiling arrays, there may be no immediate payoff for ESC biology. Questions were raised about whether the data derived from this study would be more important now or further down the road. Another discussion point concerned the heterogeneity of cell populations and the potentially dynamic methylation states that may arise in a heterogeneous background during differentiation. The PI would try to circumvent this problem by differentiating progenitors into a number of cell types prior to looking at methylation patterns, and the PI intends to use a number of signals to monitor the differentiation state. Nonetheless, it is unclear whether any remaining heterogeneity would compromise data interpretation.