Noncoding de novo mutations contribute to autism spectrum disorder via chromatin interactions

Abstract

Autism spectrum disorder (ASD) is a neurodevelopmental disorder with complex genomic etiologies, with most cases occurring sporadically. De novo mutations (DNMs), accordingly, have been thought to play a critical role in the development of ASD. However, DNMs are known to contribute to ~10%-30% of patients with ASD, which is largely owing to studies focusing on protein-coding regions, occupying up to 2% of the genomes. Although DNMs in noncoding regions also seem to underlie the genomic etiologies of ASD, the pathogenic role of the mutations outside the protein-encoding regions remains poorly understood. Since the genomes are a form of three-dimensional chromatin structures, how the noncoding DNMs affect the chromatin interactions in ASD is to be explored. Here, we generated 931 whole-genome sequences of whole blood DNA acquired from 276 Korean simplex quad families to detect DNMs, and identified target genes that have the chromatin interactions with the noncoding DNMs in regulatory elements, using resources of DNase-seq and Hi-C. Notably, the noncoding DNMs that have the chromatin interactions exhibit transcriptional dysregulation implicated in ASD. Moreover, those target genes are significantly involved in histone modification and prenatal expression of brain development, both of which are implicated in the pathogenesis of ASD. Then, we experimentally validated the effects of the noncoding DNMs defined through the chromatin interactions. Indeed, the noncoding DNMs remotely affect the expressions of target genes in the mutation-carrying neurons derived from probands' induced pluripotent stem cells. Interestingly, the noncoding and coding DNMs collectively contribute to severely low IQ. This strongly suggests that a clinical subtype of ASD can be genetically defined by discovering the functionally active noncoding DNMs. Lastly, we suppose the genomic accountability for ASD is beyond 70% using this method. Analysis results are reproducible in independent 517 probands of MSSNG resource. This work reveals the contribution of the novel genomic risk for ASD via the chromatin interactions, using massive whole-genome sequences.