A genome-wide screen for essential genes that controls the formation of human heart progenitors

Abstract

The heart is a complex organ system composed of multiple types of tissues. These tissues are produced by a diverse set of muscle and non-muscle cells, originated from a few pools of progenitors. During the heart development, these progenitors are able to expand and differentiate in a tightly controlled manner, generating diversified heart cell lineages. The progenies from these progenitors interact with each other and ultimately integrate into distinct heart tissues. The foundation of a healthy and functional heart stems from the state of its progenitor pools. Any errors that occurred during the formation, proliferation, differentiation, and assembly of these progenitors are the potential causes of many congenital heart diseases. To investigate the cellular mechanisms of human heart development and their implications in congenital heart diseases, we face many challenges, two of them are: 1) generation of progenitor cells that can self-assemble into mature cardiac tissue that faithfully resembles native mature adult cardiac tissue; 2) identification of regulators that controls the formation, proliferation and differentiation of these progenitors.

In paper I, we reported the large-scale generation of an enriched pool of human pluripotent stem cells (hPSCs) derived human ventricular progenitors (HVPs). These HVPs can build a function ventricular heart muscle in vivo via a cell autonomous pathway, including controlled proliferation followed by normal growth, maturation, and self-assembly. This tissue generation process of HVPs recapitulates one of the earliest and most essential steps of organogenesis. With these properties, HVPs highly likely resembles the progenitors that contribute to the ventricular cardiac muscle tissue during human cardiogenesis. In the study, we also explore the therapeutic potential of HVPs in heart failure. As a resource for further analyzing the genetic and molecular pathways of HVPs, we also documented the transcriptomic transitions of the progenitor formation and subsequent differentiation via sequential RNA-Seq. With the success of generating HVPs, we next try to identify regulators, specifically, the ones that control the formation of HVPs.

In paper II, we used CRIPSR-Cas9 system to target β-catenin (encoded by CTNNB1), a central component of the canonical WNT signaling pathway. The WNT signaling is a major player in cardiogenesis. By temporal modulating the WNT/β-catenin signaling pathway with small molecules, high differentiation efficiency (>90%) can be achieved. With CTNNB1 mutated hPSCs, we found that Wnt/β-catenin signaling is neither required for hPSC self-renewal, nor for neuroectoderm formation. However, Wnt/β-catenin signaling is absolutely essential for mesendoderm lineage, including cardiac progenitors and cardiomyocytes. This study pinpoints the β-catenin as the master switch of the human cardiogenesis. Another set of important signaling pathways in cardiogenesis are the TGFβ superfamily signaling pathways. Due to the complicate interaction between WNT/β-catenin and the TGFβ superfamily signaling pathways, it is difficult to define the roles of TGFβ superfamily signaling pathways from chemical inhibition studies.

In paper III, we used CRIPSR-Cas9 system to target SMAD4, a central component in the whole superfamily. With SMAD4 mutated hPSCs, we confirmed the dispensable role of SMAD4 for hPSC self-renewal in vitro. Furthermore, we demonstrated the essential requirement of SMAD4 in the formation of human cardiac mesodermal precursor cell. By transcriptome analysis, we identified that SMAD4 mutants failed to differentiate into cardiac mesoderm and, after 6 days, switched to neuroectoderm. Primitive steak (PS) genes were expressed in both the wild type and the mutant cells on day 1. And interestingly, on day 1, the only active ligand in the TGFβ superfamily signaling pathways is NODAL, which specifies the pathway in the family as NODAL/SMAD4 pathway. Together, these data suggest that during human mesoderm induction, the WNT/β-catenin is responsible for triggering the expression of PS genes, while NODAL/SMAD4 is responsible for the feedback enhancement for PS gene expression. This study highlights the essential roles of NODAL/SMAD4 signaling pathway in human cardiac mesodermal induction.

In order to unbiasedly uncover the regulators that control the formation of HVPs, in paper IV, we developed a genome-wide CRISPR screen based on cardiac differentiation from hPSCs. From the screen output, we compiled a list of 15 candidate genes. After validating 7 of these, we identified ZIC2 as an essential gene for cardiac progenitor formation. ZIC2 is known as a master regulator of neurogenesis. hPSCs with ZIC2 mutated still express pluripotency markers. However, their ability to differentiate into cardiomyocytes has greatly reduced. Transcriptome profiling reveals that they have switched to an alternative mesodermal cell fate. Our results provide a new link between ZIC2 and human cardiogenesis and document the potential power of genome-wide unbiased CRISPR screens to identify key steps in heart progenitor fate determination during human cardiogenesis with hPSC model systems. In summary, we have generated HVPs, which can self-assemble into human ventricular muscle tissue and further identified CTNNB1, SMAD4, and ZIC2 as the essential regulators that controlled the formation of HVPs.