NEW YORK (GenomeWeb News) – A survey of DNA methylation in the genomes of identical twins discordant for autism spectrum disorder uncovered epigenetic patterns associated with the disorder, according to a study published yesterday in Molecular Psychiatry by a team of UK researchers.
Further, the researchers noted that there are differences in DNA methylation patterns seen within twin pairs that both have the disorder but have differing symptoms or severity of the condition.
While ASD appears to be highly heritable, it also is a heterogeneous disease. In 70 percent of cases in which one identical twin has the disorder, the other will as well, but there can be differences in symptoms — difficulties in social interactions, repetitive behaviors, and delays in communication development — and their severity. This, according to the researchers, suggests a possible role for epigenetics in the etiology of the disorder.
In this study, the researchers examined the DNA methylation profiles of 50 pairs of identical twins, including sets of twins who were concordant and discordant for ASD, as well as controls. They identified differences in methylation patterns between participants with autism and controls as well as between twins with varying severity of ASD traits. A number of the genes that are differentially methylated have previously been implicated in ASD.
"We've identified distinctive patterns of DNA methylation associated with both autism diagnosis and related behavior traits, and increasing severity of symptoms," Chloe Wong, from King's College London's Institute of Psychiatry and the paper's first author, said in a statement. "Our findings give us an insight into the biological mechanism mediating the interaction between gene and environment in autism spectrum disorder."
Wong and her colleagues drew upon twin pairs from the UK Twins Early Development Study, including five pairs of twins in which both twins had ASD, 34 pairs discordant for ASD, and 11 control twin pairs. The twins underwent genome-wide DNA methylation analysis that examined some 27,000 sites in the genome.
While the researchers did not see global changes in DNA methylation between discordant twins, they noticed a number of site-specific differences, which they said supported the notion of DNA methylation affecting ASD phenotype. In particular, they noted differences between twins at the NFYC promoter.
Further, in twin pairs exhibiting differences in ASD symptoms and severity, Wong and her colleagues also uncovered DNA methylation pattern differences, with a notable change occurring at the PIK3C3 promoter in twins discordant for social autistic traits. This region was hypomethylated in affected twins as compared to their sibling. This further suggested to the researchers that environmental factors could mediate its effects through the epigenome, and that changes at some CpG sites appeared to be linked to specific symptoms.
In a handful of twins, the researchers also examined if there were methylation differences between sporadic and familial ASD, finding that there were, especially at a spot upstream of C19orf33. They also noted methylation differences near other genes, like MBD4, AUTS2, and MAP2, that had previously been linked to autism and related disorders.
One twin pair, Wong and her colleagues pointed out, was an extreme outlier, both in the sense of having a high score on an autism symptom scale as well as having high levels of DNA methylation at a number of CpG sites. Further analysis of this twin pair showed that they had CNV changes at known ASD-related sites as well. "Given the important role of epigenetic mechanisms in regulating gene expression, it is plausible that, like CNVs, methylomic variation could mediate disease susceptibility via altered gene dosage," the study authors wrote.
There were also sites showing differing levels of methylation between cases and controls. For instance, a region upstream of MGC3207 was "significantly hypomethylated" in ASD cases as compared to controls. A number of other sites also exhibited between-group differences.
While the researchers suggested that their findings indicate a role for epigenetics in the etiology ofASD, they cautioned that some of the subgroups in their study were small, and that their findings should be replicated in a larger population or possibly over time to pinpoint when such epigenetic changes arise, and to develop ways to possibly roll back those changes.
"Epigenetic changes are potentially reversible, so our next step is to embark on larger studies to see whether we can identify key epigenetic changes common to the majority of people with autism to help us develop possible therapeutic interventions," Jonathan Mill, lead author on the paper from King's College London's Institute of Psychiatry and the University of Exeter, said in a statement.