Convergent impact of schizophrenia risk genes

Abstract

Schizophrenia is a highly heritable psychiatric disorder with a complex genetic risk architecture that reflects the additive impact of hundreds of risk variants. While many schizophrenia-associated risk variants are thought to regulate the expression of target genes in a cell-type-specific manner, the mechanisms by which the effect of these myriad variants combine to contribute to risk remain unclear. Here we apply a CRISPR-based approach to evaluate in parallel twelve schizophrenia eGenes (that encompass common variation) in human glutamatergic neurons. Querying the shared neuronal impacts across risk genes uncovers a convergent effect concentrated on pathways of brain development and synaptic signaling. Our analyses reveal shared and divergent downstream effects of these twelve genes, independent of their previously annotated biological roles. General convergence of gene expression increases with increasing polygenicity, while the specificity of convergence increases between functionally similar genes. Convergent networks show brain-region and developmental period-specific enrichments, as well as disorder-specific enrichments for both rare and common variant target genes across schizophrenia, bipolar disorder, autism spectrum disorder, and intellectual disability. These gene targets are drug-able and potentially represent novel points of therapeutic intervention. Convergent signatures are also resolved in the post-mortem brain. Overall, convergence suggests a model to explain how non-additive interactions arise between risk genes and may explain cross-disorder pleiotropy of genetic risk for psychiatric disorders.

Non-additive effects of schizophrenia risk genes reflect convergent downstream function

“…The robust overlap between non-additive genes and convergent gene targets suggest a functional redundancy of within-function eGenes, suggesting that the observed subadditive impact reflected a saturation of the downstream effects of individual eGenes converging on common gene targets. Using a pooled CRISPR screen of a partially overlapping group of 10 SCZ eGenes in iGLUTs, we similarly reported that the strength of convergence increased with gene number and that the specificity of convergence increased with the degree of functional similarity and co-expression between eGenes 47 . Convergent genes downstream of SCZ eGenes were showed cross-disorder enrichment targets for psychiatric risk genes, suggesting that they may underlie shared clinical hallmarks and pleiotropy of risk 47 .…”

“…Using a pooled CRISPR screen of a partially overlapping group of 10 SCZ eGenes in iGLUTs, we similarly reported that the strength of convergence increased with gene number and that the specificity of convergence increased with the degree of functional similarity and co-expression between eGenes 47 . Convergent genes downstream of SCZ eGenes were showed cross-disorder enrichment targets for psychiatric risk genes, suggesting that they may underlie shared clinical hallmarks and pleiotropy of risk 47 . Moreover, recent studies of loss-of-function autism spectrum disorder genes, albeit perturbations unlikely to be inherited in combination, also reported convergence, whether evaluated in vitro in human neural progenitor cells 48 and brain organoids 49 , or in vivo in fetal mouse brains 50 and Xenopus tropicalis 51 .…”

“…Our data argues for a re-conceptualization of genetic contribution to disease risk; PRSs that incorporate biological pathways and gene co-expression networks of the risk variants being summed 42,71 , together with their downstream impacts 47 , may improve patient stratification 42,72 or better predict drug response 73 . Moreover, resolving convergent effects shared between eGenes will identify nodes whereby the impact of aggregate genetic risk can be most effectively decreased and prioritize potential therapeutic targets.…”

“…Perturbations of SCZ eGenes resulted in non-additive and convergent transcriptional effects that occur within groups of functionally similar genes (see also 47 ) and increased with number of eGenes perturbed. Across fifteen eGenes, within-function combinatorial perturbations resulted in sub-additive impacts on transcriptional profiles of hundreds of genes, which largely overlapped with the convergent gene targets of the same SCZ eGenes.…”

Schizophrenia Risk Mapping and Functional Engineering of the 3D Genome in Three Neuronal Subtypes

“…Our neuronal subtype-specific 3D genome maps are accessible through the PsychENCODE Knowledge Portal (https://synapse.org) and further expand the number of Hi-C datasets from human brain 8,9,11,17,47,70,71 . Cell type-specific 3D genome analyses across the major neurotransmitter classes implicated in schizophrenia could improve the accuracy of diagnosis 72 and the targets of distal risk genes may represent novel therapeutic targets 73 .…”