Expression quantitative trait loci in the developing human brain and their enrichment in neuropsychiatric disorders

Abstract

BackgroundGenetic influences on gene expression in the human fetal brain plausibly impact upon a variety of postnatal brain-related traits, including susceptibility to neuropsychiatric disorders. However, to date, there have been no studies that have mapped genome-wide expression quantitative trait loci (eQTL) specifically in the human prenatal brain.ResultsWe performed deep RNA sequencing and genome-wide genotyping on a unique collection of 120 human brains from the second trimester of gestation to provide the first eQTL dataset derived exclusively from the human fetal brain. We identify high confidence cis-acting eQTL at the individual transcript as well as whole gene level, including many mapping to a common inversion polymorphism on chromosome 17q21. Fetal brain eQTL are enriched among risk variants for postnatal conditions including attention deficit hyperactivity disorder, schizophrenia, and bipolar disorder. We further identify changes in gene expression within the prenatal brain that potentially mediate risk for neuropsychiatric traits, including increased expression of C4A in association with genetic risk for schizophrenia, increased expression of LRRC57 in association with genetic risk for bipolar disorder, and altered expression of multiple genes within the chromosome 17q21 inversion in association with variants influencing the personality trait of neuroticism.ConclusionsWe have mapped eQTL operating in the human fetal brain, providing evidence that these confer risk to certain neuropsychiatric disorders, and identifying gene expression changes that potentially mediate susceptibility to these conditions.Electronic supplementary materialThe online version of this article (10.1186/s13059-018-1567-1) contains supplementary material, which is available to authorized users.

Large eQTL meta-analysis reveals differing patterns between cerebral cortical and cerebellar brain regions

“…Additionally, we compared our eQTL to a publically available fetal brain eQTL resource 40 and found good replication of these eQTL as well (estimated replication rate π 1 = 0.909 for the cortical meta-analysis, and π 1 = 0.861 for cerebellum), though somewhat lower than the replication observed in the GTEx cohorts, which are comprised of adult-derived samples.…”

Cross-disorder GWAS meta-analysis for Attention Deficit/Hyperactivity Disorder, Autism Spectrum Disorder, Obsessive Compulsive Disorder, and Tourette Syndrome

“…The top result of our four-disorder combined transcriptome-wide analysis was from transcript of LRRC37A4P. Even though it is a pseudogene, altered transcription levels of this probe have been observed in multiple neurological conditions, including Alzheimer disease and ASD (86)(87)(88)(89). Furthermore, the implicated region has been previously found as contributing to various neurodevelopmental or neurodegeneration processes (90)(91)(92)(93)(94).…”

Whole-Genome and RNA Sequencing Reveal Variation and Transcriptomic Coordination in the Developing Human Prefrontal Cortex

“…To assess the sensitivity of our cis-eQTL discovery analysis relative to previous work, we evaluated the relationship between sample size and eGene discovery for: the BrainVar prenatal, postnatal, and complete sample analyses as run using the HCP-and SVA-adjusted expression data, GTEx v7 analyses by tissue (gtexportal.org), postnatal human frontal cortex by the CommonMind Consortium (Fromer et al, 2016), and prenatal human whole brain (O'Brien et al, 2018). Using the sample size reported in each analysis, or for each tissue (GTEx), and the number of genes with at least one eQTL reaching significance of FDR ≤ 0.05, we plotted the relationship between eGene discovery and sample size ( Figure S4).…”

“…We evaluated the performance of the eQTLs that we identified in our analyses with published sets of eQTLs identified in the human postnatal frontal cortex (The GTEx Consortium et al, 2017) and in the human prenatal whole brain (O'Brien et al, 2018). For the postnatal frontal cortex data, we downloaded significant variant-gene pairs from the GTEx v7 data release from gtexportal.org and used R to write out the variant locations to a bed file format.…”