Common schizophrenia variants functioning in developmental human cortical interneurons

摘要

Abstract Schizophrenia (SCZ) is a highly heritable and complex neurodevelopmental disorder. Remarkable advances have been made recently in SCZ genetic studies with an increasing number of risk loci reaching genome-wide significance; however, gleaning biological insight from these loci has been challenging. The majority of SCZ risk loci are located in non-coding regions. As such, it is hypothesized that they function by regulating distal gene expression via 3D chromatin interactions. However, it has yet to be determined which loci are operational in which cells, at what time points, and with what impact. Recent genomic analyses showed enriched SCZ heritability in human fetal brains rather than adult brains, suggesting the role of SCZ risk loci in modulating fetal development for increased SCZ risks. Thus, unraveling SCZ risk loci function during development will be critical for understanding genetic influences on SCZ risks. Genetic influences on gene expression (e.g. expression quantitative trait loci (eQTLs)) are cell-type-specific, and sometimes confer opposing effects depending on the cell type, underscoring the importance of cell-type-specific studies using homogeneous cell populations for a clear mechanistic understanding. Parvalbumin (PV)- or somatostatin (SST)-expressing medial ganglionic eminence (MGE)-derived cortical interneurons (cINs) are consistently affected in SCZ brains. More importantly, SCZ heritability is shown to be enriched in MGE cells in human fetal brains, necessitating the study of these cells to understand the mechanisms of SCZ risk loci. Although there are no postmortem fetal SCZ tissues for mechanistic study, in vitro differentiation of iPSC—which well recapitulates early embryonic development— provides developmental SCZ brain cells with the same genetic makeup as patient brains. We established methods for the efficient generation of homogeneous populations of MGE-derived cINs from healthy control (HC) and SCZ iPSCs. We also extensively validated their functionality and authenticity both in vitro and in vivo, including robust migration and synaptic integration into host brains that results in efficient inhibitory regulation of host circuitry in transplanted mice. Using an unprecedentedly large number of iPSCs to provide homogeneous populations of HC vs SCZ fetal cINs for mechanistic studies, we will address our hypothesis that SCZ risk loci active in developmental MGE-type cINs regulate distal gene expression via 3D chromatin interactions. Employing transcriptome analysis, PrediXcan analysis, and Micro-C analysis, we will map SCZ risk loci with unknown functions to the risk genes they regulate in these vulnerable cell populations during development. Developmental cIN-specific genetic influences on gene expression, identified based on multiple lines of corroborating evidence, will be functionally validated using CRISPRi/CRISPRa approaches. This unbiased genome-wide comprehensive data set from developmental MGE-type cINs with functional validation will provide a road map for unravelling the genetic basis of developmental SCZ risks and help us identify mechanism-based novel therapeutic targets.