NEW YORK (GenomeWeb) – By studying blood samples taken from infants, researchers have found that higher numbers of alleles associated with risk for autism spectrum disorder are also associated with differential methylation at certain spots in the genome.
Autism spectrum disorder is highly heritable, though environmental factors still influence its risk, possibly through epigenetic variation.
Using blood spot samples collected shortly after birth, a University of Exeter Medical School-led team examined methylomic variation in 1,263 infants, about half of whom were later diagnosed with ASD. While they found no differences in overall methylation between the later affected and unaffected infants, the researchers did find a link between increased polygenic burden for autism and differences in methylation at certain loci. As they reported this week in Genome Medicine, these loci are close to a signal previously uncovered through a genome-wide association study of autism.
"Our data provide evidence for differences in DNA methylation at birth associated with an elevated polygenic burden for ASD," Exeter's Jonathan Mill and his colleagues wrote in their paper. "Our study represents the first analysis of epigenetic variation at birth associated with autism and highlights the utility of polygenic risk scores for identifying molecular pathways associated with etiological variation."
The researchers analyzed a subset of the Danish iPSYCH study population, called MINERvA, that included 629 ASD cases and 634 matched controls, about half of whom were female. For each participant, neonatal blood was collected on Guthrie cards shortly after birth, and the researchers used these samples for both methylation profiling on the Illumina 450K array and SNP genotyping on its Infinium PsychChip.
However, when Mill and his colleagues compared the methylation among the neonates, they found no difference in global methylation between those who went on to develop ASD and those who did not. Additionally, they didn't find any differentially methylated positions (DMPs) between the cases and controls that met experiment-wide significance.
The researchers added in data from the US-based Study to Explore Early Development and the Simons Simplex Cohort to supplement their cohort, but still found little difference in methylation between cases and controls. They said this lack of association could be due to the technical limitations of the assay they used, their reliance on peripheral tissue, or looking too early in disease development.
Instead, the researchers generated polygenic risk scores (PRSs) for each individual based on alleles linked to autism risk in a recent genome-wide association meta-analysis. With that, they performed an epigenome-wide association study of ASD PRSs. They uncovered two DMPs in CpG sites that were significantly associated with increased polygenic burden.
Both highly significant DMPs are located on chromosome 8, about 5 kilobases away from each other, but annotated to different genes: FAM167A and RP1L1. One is in a predicted enhancer region, while the other is in a region predicted to be quiescent.
The two DMPs flank a region uncovered in the recent ASD GWAS, Mill and his colleagues noted. The index SNP from that study was predicted to be in a region that is repressed in blood and lowly expressed in the brain. The researchers tied the PRS-associated variation in DNA methylation at those two sites to the combined effect of a number of genetic variants in the region that are linked to ASD.
By combining methylation and genetic data from the MINERvA cohort, the researchers searched for DNA methylation quantitative trait loci near the ASD-linked GWAS variants to find 91 such sites. These included four sites on chromosome 20, and the researchers said the genes at those sites — KIZ, XRN2, and NKX2–4 — should be investigated further for roles in ASD.
"The mechanisms linking DNA sequence variation to alterations in DNA methylation and other epigenetic modifications are not yet well understood; further exploration of these processes is warranted to provide insight into the functional consequences of disease-associated genetic variation," the authors wrote.