Studies in embryonic development have guided successful efforts to direct the differentiation of human embryonic and induced pluripotent stem cells (PSCs) into specific organ cell types in vitro 1,2. For example, human PSCs have been differentiated into monolayer cultures of liver hepatocytes and pancreatic endocrine cells3–6 that have therapeutic efficacy in animal models of liver disease 7,8 and diabetes 9 respectively. However the generation of complex three-dimensional organ tissues in vitro remains a major challenge for translational studies. We have established a robust and efficient process to direct the differentiation of human PSCs into intestinal tissue in vitro using a temporal series of growth factor manipulations to mimic embryonic intestinal development 10 (Summarized in supplementary Fig. 1). This involved activin-induced definitive endoderm (DE) formation 11, FGF/Wnt induced posterior endoderm pattering, hindgut specification and morphogenesis 12–14; and a pro-intestinal culture system 15,16 to promote intestinal growth, morphogenesis and cytodifferentiation. The resulting three-dimensional intestinal “organoids” consisted of a polarized, columnar epithelium that was patterned into villus-like structures and crypt-like proliferative zones that expressed intestinal stem cell markers17. The epithelium contained functional enterocytes, as well as goblet, Paneth, and enteroendocrine cells. Using this culture system as a model to study human intestinal development, we identified that the combined activity of Wnt3a and FGF4 is required for hindgut specification whereas FGF4 alone is sufficient to promote hindgut morphogenesis. Our data suggests that human intestinal stem cells form de novo during development. Lastly we determined that NEUROG3, a pro-endocrine transcription factor that is mutated in enteric anendocrinosis 18, is both necessary and sufficient for human enteroendocrine cell development in vitro. In conclusion, PSC-derived human intestinal tissue should allow for unprecedented studies of human intestinal development and disease.

The liver and pancreas are specified from the foregut endoderm through an interaction with the adjacent mesoderm. However, the earlier molecular mechanisms that establish the foregut precursors are largely unknown. In this study, we have identified a molecular pathway linking gastrula-stage endoderm patterning to organ specification. We show that in gastrula and early-somite stage Xenopus embryos, Wnt/␤-catenin activity must be repressed in the anterior endoderm to maintain foregut identity and to allow liver and pancreas development. By contrast, high ␤-catenin activity in the posterior endoderm inhibits foregut fate while promoting intestinal development. Experimentally repressing ␤-catenin activity in the posterior endoderm was sufficient to induce ectopic organ buds that express early liver and pancreas markers. ␤-catenin acts in part by inhibiting expression of the homeobox gene hhex, which is one of the earliest foregut markers and is essential for liver and pancreas development. Promoter analysis indicates that ␤-catenin represses hhex transcription indirectly via the homeodomain repressor Vent2. Later in development, ␤-catenin activity has the opposite effect and enhances liver development. These results illustrate that turning Wnt signaling off and on in the correct temporal sequence is essential for organ formation, a finding that might directly impact efforts to differentiate liver and pancreas tissue from stem cells.

The canonical Wnt pathway is necessary for gut epithelial cell proliferation, and aberrant activation of this pathway causes intestinal neoplasia. We report a novel mechanism by which the Sox family of transcription factors regulate the canonical Wnt signaling pathway. We found that some Sox proteins antagonize while others enhance ␤-catenin/T-cell factor (TCF) activity. Sox17, which is expressed in the normal gut epithelium but exhibits reduced expression in intestinal neoplasia, is antagonistic to Wnt signaling. When overexpressed in SW480 colon carcinoma cells, Sox17 represses ␤-catenin/TCF activity in a dose-dependent manner and inhibits proliferation. Sox17 and Sox4 are expressed in mutually exclusive domains in normal and neoplastic gut tissues, and gain-and loss-of-function studies demonstrate that Sox4 enhances ␤-catenin/TCF activity and the proliferation of SW480 cells. In addition to binding ␤-catenin, both Sox17 and Sox4 physically interact with TCF/lymphoid enhancer factor (LEF) family members via their respective high-mobility-group box domains. Results from gain-and loss-of-function experiments suggest that the interaction of Sox proteins with ␤-catenin and TCF/LEF proteins regulates the stability of ␤-catenin and TCF/LEF. In particular, Sox17 promotes the degradation of both ␤-catenin and TCF proteins via a noncanonical, glycogen synthase kinase 3␤-independent mechanism that can be blocked by proteasome inhibitors. In contrast, Sox4 may function to stabilize ␤-catenin protein. These findings indicate that Sox proteins can act as both antagonists and agonists of ␤-catenin/TCF activity, and this mechanism may regulate Wnt signaling responses in many developmental and disease contexts.

Tbx20 is a member of the T-box transcription factor family expressed in the forming hearts of vertebrate and invertebrate embryos. We report here analysis of Tbx20 expression during murine cardiac development and assessment of DNA-binding and transcriptional properties of Tbx20 isoforms. Tbx20 was expressed in myocardium and endocardium, including high levels in endocardial cushions. cDNAs generated by alternative splicing encode at least four Tbx20 isoforms, and Tbx20a uniquely carried strong transactivation and transrepression domains in its C terminus. Isoforms with an intact T-box bound specifically to DNA sites resembling the consensus brachyury half site, although with less avidity compared with the related factor, Tbx5. Tbx20 physically interacted with cardiac transcription factors Nkx2-5, GATA4, and GATA5, collaborating to synergistically activate cardiac gene expression. Among cardiac GATA factors, there was preferential synergy with GATA5, implicated in endocardial differentiation. In Xenopus embryos, enforced expression of Tbx20a, but not Tbx20b, led to induction of mesodermal and endodermal lineage markers as well as cell migration, indicating that the long Tbx20a isoform uniquely bears functional domains that can alter gene expression and developmental behaviour in an in vivo context. We propose that Tbx20 plays an integrated role in the ancient myogenic program of the heart, and has been additionally coopted during evolution of vertebrates for endocardial cushion development.