Functional Genomic Analysis of Chemically Defined Human Embryonic Stem Cells

We have made significant progress during the previous granting period which has resulted in a publication in Genome Research detailing genomic DNA methylation changes in a variety of human embryonic stem cells and their derivatives. We have also been successful in identifying regions of the genome bound by of histone modifications and transcription factors in hESCs and derived endoderm. These targets have led to a greater understanding of how Nodal signaling and chromatin configuration maintain pluripotent state and subsequently trigger differentation. We expect 3 publications to result from this work during the next granting period.

Toward a goal of developing endodermal lineages from hESCs, including liver, pancreas, lung and intestine, we have developed new tools and approaches to identify these subtypes as well as a molecular understanding of how these subtypes emerge. These advances are highlighted in three papers which are currently under review for publication and one in preparation. Two of these papers redefine endodermal subtypes derived from hESCs, including new methods to isolate lineage restricted endodermal populations and a means to distinguish between single endodermal cells. The third paper provides an unprecedented view of the Nodal signaling pathway and its intersection with bivalent domains in both hESCs and derived endoderm. This chromatin signature which consists of Smad transcription factors and both histone repressive and active marks is the most conducive for mediating downstream targets of Nodal, providing inroads into how signaling pathways and chromatin cooperate during fate specification in hESCs. This analysis has led to the elucidation of new proteins that mediate Nodal signaling; ones that play a key role in endoderm specification.

During the past year, we have completed three research projects as a result of this grant. Two of these have recently been published in Genes and Development and Developmental Biology. The third is currently in review at Development. The discoveries reported in these papers are highlighted below: HEB and E2A function as SMAD/FOXH1 cofactors Nodal signaling, mediated through SMAD transcription factors, is necessary for pluripotency maintenance and endoderm commitment. We have identified a new motif, termed SMAD Complex Associated (SCA) that is bound by SMAD2/3/4 and FOXH1 in human embryonic stem cells (hESCs) and derived endoderm. We demonstrate that two bHLH proteins - HEB and E2A - bind the SCA motif at regions overlapping SMAD2/3 and FOXH1. Further, we show that HEB and E2A associate with SMAD2/3 and FOXH1, suggesting they form a complex at critical target regions. This association is biologically important, as E2A is critical for mesendoderm specification, gastrulation, and Nodal signal transduction in Xenopus tropicalis embryos. Taken together, E proteins are novel Nodal signaling cofactors that associate with SMAD2/3 and FOXH1 and are necessary for mesendoderm differentiation. Chromatin and Transcriptional Signatures for Nodal Signaling During Endoderm Formation in hESCs The first stages of embryonic differentiation are driven by signaling pathways hardwired to induce particular fates. Endoderm commitment is controlled by the TGF-β superfamily member, Nodal, which utilizes the transcription factors, SMAD2/3, SMAD4 and FOXH1, to drive target gene expression. While the role of Nodal is well defined within the context of endoderm commitment, mechanistically it is unknown how this signal is manifested at binding regions within the genome and how this signal interacts with chromatin state to trigger downstream responses. To elucidate the Nodal transcriptional network that governs endoderm formation, we used ChIP-seq to identify genomic targets for SMAD2/3, SMAD3, SMAD4, FOXH1 and the active and repressive chromatin marks, H3K4me3 and H3K27me3, in human embryonic stem cells (hESCs) and derived endoderm. We demonstrate that while SMAD2/3, SMAD4 and FOXH1 binding is highly dynamic, there is an optimal signature for driving endoderm commitment. Initially, this signature is marked by both H3K4me3 and H3K27me3 as a very broad bivalent domain in hESCs. Within the first 24 hours, at a few select promoters, SMAD2/3 accumulation coincides with H3K27me3 reduction so that these loci become relatively monovalent marked by H3K4me3. JMJD3, a histone demethylase, is simultaneously recruited to the promoters of the endodermal gene GSC and EOMES, suggesting a conservation of mechanism at multiple promoters genome-wide. The correlation between SMAD2/3 binding, monovalent formation and transcriptional activation suggests a mechanism by which SMAD proteins coordinate with chromatin at critical promoters to drive endoderm specification. Distinguishing Cells with Housekeeping Transcripts Distinguishing between cell types is central to a broad array of biological fields and is at the heart of advancing regenerative medicine and cancer diagnostics. In this report, we use single cell gene expression to identify transcriptional patterns emerging during the differentiation of human embryonic stem cells (hESCs) into the endodermal lineage. Endoderm specific transcripts are highly variable between individual CXCR4+ endodermal cells, suggesting that either the cells generated from in vitro differentiation are distinct or that these embryonic cells tolerate a high degree of transcript variability. Housekeeping transcripts, on the other hand, are far more consistently expressed within the same cellular population. However, when we compare the levels of housekeeping transcripts between hESCs and derived endoderm, patterns emerge that can be used to clearly separate the two embryonic cell types. We further compared 4 additional human cell types, including 293T, iPSC, HepG2 and endoderm derived iPSC. In each case the relative levels of housekeeping transcripts defined a particular cell fate. Interestingly, we find that three transcripts, LDHA, NONO and ACTB, contribute the most to this diversity and together serve to segregate all 6 cell types. Overall, this suggests that levels of housekeeping transcripts, which are expressed within all cells, can be leveraged to distinguish between human cell types and thus may serve as important biomarkers for stem cell biology and other disciplines.