A critical stage of human development is the specification of the three primary germ layers that will comprise the embryo: a process known as gastrulation. During gastrulation pluripotent, unspecialized, cells of the epiblast differentiate into either endoderm, mesoderm or ectoderm lineages, a process coordinated in part by extrinsic signaling cues. The first event of gastrulation is the directed differentiation of epiblast cells towards mesendoderm, a bi-potent progenitor that eventually becomes restricted to either endoderm and mesoderm cell types. Concomitantly with differentiation, nascent mesendoderm cells undergo an epithelial- to-mesenchymal transition (EMT) at the primitive streak, where the cells migrate from the epiblast layer to form a new layer of cells. Because of obvious ethical and legal reasons, it is not possible to study this stage of human development in vivo. However, studies suggest that human embryonic stem cells (hESCs) are in vitro equivalents of pluripotent epiblast cells. They therefore represent a feasible system for modeling human gastrulation in vitro. Since WNTs are known regulators of gastrulation in all metazoan species, it seemed logical to test the role of this pathway in epiblast differentiation. Specifically, I used hESCs to dissect the role of the transcriptional repressor TCF7L1, a downstream transcription factor of the canonical WNT pathway, in the regulation of pluripotency. I found that TCF7L1 acts as a key regulator of mesendoderm differentiation in hESCs. I also identify a novel regulatory interaction between the BMP4 pathway (another key regulator of mesendoderm/primitive streak formation in vivo) and TCF7L1, whereby activation of the BMP4 signaling pathway induces TCF7L1 down-regulation. Based on these findings I hypothesize that the action of BMP4 on TCF7L1 leads to the loss of repression of mesendoderm genes and activates mesendoderm differentiation during gastrulation. Thus, this study marks TCF7L1 as a key suppressor of mesendoderm differentiation in hESCs and provides valuable insights into the regulation of gastrulation in vivo. My studies also make several predictions about the control of signaling events in pluripotent stem cells in vitro and in vivo and provide new avenues in this exciting area of research.

Human embryonic stem cells (hESC) are stem cells capable of differentiating into cells representative of the three primary embryonic germ layers. There has been considerable interest in understanding the mechanisms regulating stem cell pluripotency, which will ultimately lead to development of more efficient methods to derive and culture hESC. In particular, Oct4, Sox2 and Nanog are transcription factors known to be important in maintenance of hESC. However, many of the downstream targets of these transcription factors are not well characterized. Furthermore, it remains unknown whether additional novel stem cell factors are involved in the establishment and maintenance of the stem cell state.

Here we show that a novel gene, L1TD1 (also known as FLJ10884 or ECAT11), is abundantly expressed in undifferentiated hESC. Differentiation of hESC via embryoid body (EB) formation or BMP4 treatment results in the rapid down-regulation of L1TD1 expression. Furthermore, populations of undifferentiated and differentiated hESC were sorted using the stem cell markers SSEA4 and TRA160. Our results show that L1TD1 is enriched in the SSEA4-positive or TRA160-positive population of hESC. Using chromatin immunoprecipitation we found enriched association of Nanog to the predicted promoter region of L1TD1. Furthermore, siRNA-mediated knockdown of Nanog in hESC also resulted in downregulation of L1TD1 expression. Finally, using luciferase reporter assay we demonstrated that Nanog can activate the L1TD1 upstream promoter region. Altogether, these results provide evidence that L1TD1 is a downstream target of Nanog.

Taken together, our results suggest that L1TD1 is a downstream target of Nanog and represents a useful marker for identifying undifferentiated hESC.