Human induced pluripotent stem cell-based models of congenital diseases : hirschsprung disease and congenital central hypoventilation syndrome

Abstract

Whole-genome sequence (WGS) studies of human patients have identified thousands of disease-associated genetic variants for complex multifactorial congenital disorders, but it is very challenging to shortlist the disease drivers and to understand how these variants attribute to the diseases. In this study, I used human induced pluripotent stem cell (hiPSC)-derived cell models to recapitulate the development of enteric nervous system (ENS) and central nervous system (CNS) to unveil the disease etiology of two closely associated diseases, Hirschsprung disease (HSCR) and congenital central hypoventilation syndrome (CCHS), which affect ENS and CNS respectively.

HSCR is a neurocristopathy caused by the incomplete colonization of ENS in the colon. Our group has previously identified millions of non-coding variants of small effect size from the HSCR patients by WGS. To discover the key disease-driving variants, we first established a hiPSC-derived ENS model, of which the epigenomic landscapes were profiled to identify the key regulatory elements involved in ENS development. By integrating these epigenomic profiles with the WGS data into a newly developed prioritization framework for non-coding variants, we uncovered three novel regulatory elements with HSCR-associated non-coding variants in the RET, RASGEF1A and PIK3C2B loci. Subsequent functional analyses of these non-coding variants by our hiPSC-derived ENS model revealed that these variants significantly alter the gene expression of their own or the neighboring genes in a cell-stage-specific manner, without affecting the overall ENCC differentiation to neuronal lineages. This suggests that these non-coding variants mainly modulate the gene expression dynamics during ENS development, of which the dysregulation may contribute to HSCR susceptibility.

CCHS is a life-threatening disorder characterized by the impaired breathing control to hypercarbia and hypoxemia due to the defective development of the respiratory center in CNS. CCHS is frequently associated with HSCR and the patients carrying different lengths of polyalanine repeat expansion mutations (PARMs) in PHOX2B are susceptible to CCHS alone or in combined with HSCR. To evaluate the roles of PHOX2B-PARMs in CCHS pathogenesis, we established a three-dimensional hiPSC-derived brain organoid (hBO) model to mimic the CNS development in vitro. Using hiPSC lines carrying different PHOX2B-PARMs, we successfully recapitulated the disease phenotypes in the hiPSC-derived ENS and CNS models, including the failure in the generation of enteric dopamine beta-hydroxylase neurons and the loss of the retrotrapezoid nucleus neurons in the respiratory center-like region in the hBO. By single-cell transcriptome analysis of the PHOX2B-PARM mutant hBO, we found that PHOX2B-PARMs interrupt the differentiation path of PHOX2B+ progenitors during CNS development. Subsequent gene ontology analyses suggested PHOX2B-PARMs dysregulate genes implicated in neuronal fate specification, pattern specification, neurogenesis, synapse functions and neurotransmission. In summary, our study demonstrates the feasibility of using hiPSC-derived cell models as disease models for the understanding the underlying disease mechanisms underlying various human congenital disorders.