Placental methylome reveals a 22q13.33 brain regulatory gene locus associated with autism

Abstract

Background Autism spectrum disorder (ASD) involves complex genetics interacting with the perinatal environment, complicating the discovery of common genetic risk. The epigenetic layer of DNA methylation shows dynamic developmental changes and molecular memory of in utero experiences, particularly in placenta, a fetal tissue discarded at birth. However, current array-based methods to identify novel ASD risk genes lack coverage of the most structurally and epigenetically variable regions of the human genome. Results We use whole genome bisulfite sequencing in placenta samples from prospective ASD studies to discover a previously uncharacterized ASD risk gene, LOC105373085, renamed NHIP. Out of 134 differentially methylated regions associated with ASD in placental samples, a cluster at 22q13.33 corresponds to a 118-kb hypomethylated block that replicates in two additional cohorts. Within this locus, NHIP is functionally characterized as a nuclear peptide-encoding transcript with high expression in brain, and increased expression following neuronal differentiation or hypoxia, but decreased expression in ASD placenta and brain. NHIP overexpression increases cellular proliferation and alters expression of genes regulating synapses and neurogenesis, overlapping significantly with known ASD risk genes and NHIP-associated genes in ASD brain. A common structural variant disrupting the proximity of NHIP to a fetal brain enhancer is associated with NHIP expression and methylation levels and ASD risk, demonstrating a common genetic influence. Conclusions Together, these results identify and initially characterize a novel environmentally responsive ASD risk gene relevant to brain development in a hitherto under-characterized region of the human genome.

Systemic interindividual epigenetic variation in humans is associated with transposable elements and under strong genetic control

“…However, though grossly underrepresented on the HM450 and EPIC arrays, CoRSIVs are often among top ‘hits’ in existing EWAS (70). Indeed, these stable (36, 60, 71), systemic epigenetic variants are already showing great promise for disease prediction (72–78). We suggest that improving the coverage of CoRSIVs would enhance the utility of the Illumina EPIC array for the study of population epigenetics.…”

Prenatal vitamin intake in first month of pregnancy and DNA methylation in cord blood and placenta in two prospective cohorts

“…In EARLI, whole genome bisulfite sequencing (WGBS) data were available on 63 cord blood samples sequenced on the HiSeq X [51 overlapping with array methylation data]. Sample processing and WGBS quality control and alignment for cord blood samples [ 71 ] and placenta samples [ 72 , 73 ] have been previously discussed. In MARBLES, WGBS data were available for 91 placenta samples sequenced on HiSeq X Ten (89 overlapping with the array methylation data), 45 cord blood samples sequenced on HiSeq 4000 [30 overlapping with array methylation data], and 42 cord blood samples sequenced on the HiSeq X [35 overlapping with array methylation data].…”

Prenatal vitamin intake in first month of pregnancy and DNA methylation in cord blood and placenta in two prospective cohorts

“…In EARLI, whole genome bisulfite sequencing (WGBS) data were available on 63 cord blood samples sequenced on the HiSeq X (51 overlapping with array methylation data). Sample processing and WGBS quality control and alignment for cord blood samples (68) and placenta samples (69,70) have been previously discussed. In MARBLES, WGBS data were available for 91 placenta samples sequenced on HiSeq X Ten (89 overlapping with the array methylation data), 45 cord blood samples sequenced on HiSeq 4000 (30 overlapping with array methylation data), and 42 cord blood samples sequenced on the HiSeq X (35 overlapping with array methylation data).…”