The ability of PSCs to differentiate into all the hundreds of distinct cell-types in the human body presents both an opportunity and a challenge for regenerative medicine (a). Exclusive differentiation of PSCs into a pure population of given lineage (amidst multiple alternate lineage options) would be revolutionary for regenerative medicine, but has proven challenging. We focused on differentiation of PSCs into the definitive endoderm germ layer (the embryonic precursor to internal organs, such as the lung, liver, pancreas, stomach, and intestines). In mechanistic studies performed in collaboration with others, we showed how the endoderm-inducing extracellular signal TGFβ functions; we showed that the TGFβ transcriptional effectors Smad2/3 become redistributed on chromatin upon TGFβ signaling (b), culminating in the induction of the key endodermal transcription factor Eomes (c). We subsequently developed an approach to generate a nearly-pure human definitive endoderm population within 2 days of PSC differentiation. We identified signals that enabled the stepwise differentiation of PSCs into primitive streak and subsequently into definitive endoderm while concomitantly inhibiting the formation of unwanted cell-types (namely, mesoderm and ectoderm) (d). This enabled the consistent generation of a ~94% pure endoderm population from nine distinct embryonic and induced pluripotent stem cell lines (d). We further detailed the signals that drive the subsequent progression of endoderm into anterior foregut (anterior-most endoderm), posterior foregut (medial endoderm) and midgut/hindgut (posterior-most endoderm), and the signals that control the bifurcation of posterior foregut into liver vs. pancreatic progenitors (d). Collectively this set another standard for the stem-cell differentiation field and provided a starting point to subsequently generate populations of more differentiated endodermal cell-types.
Loh & Ang et al., 2014. Efficient endoderm induction from human pluripotent stem cells by logically directing signals controlling lineage bifurcations. Cell Stem Cell 14: 237-52
Our most recent work has focused on the generation of liver progenitors from PSCs. Generating new liver cells from PSCs is of urgent importance, as liver failure is the 12th leading cause of adult death in developed nations and there is a dire shortage of livers available for transplantation. However, multiple challenges surround the clinical application of PSC-derived human liver cells (a). First, it is not yet possible to obtain homogeneous and authentic ESC-derived human liver cells that are the molecular and functional equivalent of primary liver cells. Second, the repertoire of extracellular signals that developmentally specify liver fate and the temporal sequence with which they act remains ambiguous. To meet this challenge, we defined various signaling pathways that regulate consecutive stages of liver differentiation, including patterning of human endoderm into foregut; liver bud specification; and segregation of hepatocyte vs. biliary fates (b). This knowledge allowed us to sequentially differentiate human ESCs into a >90% pure AFP+TBX3+ liver bud progenitor population and later, an >80% pure ALB+GS+ more downstream liver precursor population by days 6 and 18 of in vitro differentiation, respectively (b). These ESC-derived ALB+ human liver precursors repopulated the damaged liver of immunodeficient Fah-/- mice, improving the survival of these mice and reducing level of bilirubin (a measure of liver injury) (b). Hence, this study provided insights into the signaling regulation of different phases of liver development and enabled the efficient generation of human liver progenitors that may be potentially useful for therapeutic or industrial applications.
Ang et al., 2018. A roadmap for human liver differentiation from pluripotent stem cells. Cell Reports 22, 2190-2205
We focused on how the mesoderm germ layer (the embryonic precursor to organs such as heart, blood, kidney and bone) develops from PSCs. How these distinct mesodermal organs become diversified from one another was incompletely understood. To address this, we mapped the landscape of mesoderm development by 1) mapping the sequence of pairwise lineage choices through which distinct mesodermal precursors are formed and 2) determining the positively- and negatively-acting extracellular signals that promote or repress mesodermal cell-fate specification at each lineage juncture (a). This knowledge helped guide the efficient differentiation of PSCs into twelve human mesodermal lineages, including bone and heart progenitors, each of which could engraft mouse models (a). We further identified stepwise changes in gene expression and surface marker expression across each successive step of human mesoderm development (a,b), thus detailing the molecular features of the mesodermal developmental landscape.
Loh & Chen et al., 2016. Mapping the pairwise choices leading from pluripotency to human bone, heart, and other mesoderm cell types. Cell 166: 451-67