FASD study creates a blueprint for understanding the role of epigenetics in neurodevelopmental disorders

August 19, 2016

NeuroDevNet researchers have made the first steps towards identifying an epigenetic signature for Fetal Alcohol Spectrum Disorder (FASD), drawing from the largest-ever study of children and youth exposed to alcohol before birth.

The multi-disciplinary group found a diverse pattern of DNA methylation that possibly indicates modified genetic expression in 110 children diagnosed with FASD, or known to have a history of prenatal alcohol exposure, compared to 96 age- and sex-matched controls aged 5-18. 

“With this publication we are providing some deeper insights into potential mechanisms associated with FASD that had not been previously appreciated,” says NeuroDevNet Deputy Scientific Director James Reynolds. “We also provided evidence that may help explain the overlap between FASD and other neurodevelopmental disorders.”

The research entailed a complex analysis of genetic and epigenetic material collected from saliva samples and cheek swabs, together with a battery of neurobehavioral assessments. Six hundred and fifty-eight differentially methylated sites were identified, with 41 of these showing a methylation change greater than 5% - a common threshold for biological relevance.

"DNA methylation signature of human fetal alcohol spectrum disorder" was published online in Epigenetics and Chromatin on June 29.

“This is the first paper that takes the power of this large cohort and integrates detailed clinical observation, genetics and epigenetic data,” says Dr. Elodie Portales-Casamar, who was the Network’s neuroinformatics manager at the time the research was conducted. “All of our signal changes are subtle, though consistent, and statistically significant.”

Now an investigator within BC Children’s Hospital, she wrestled with the complexity of a massive data set looking at more than 2 million genetic markers and 450 thousand epigenetic markers.

Both depth and breadth of national expertise was necessary to produce the study. ASD research group co-lead Dr. Stephen Scherer, based at SickKids in Toronto ran the saliva samples, and Marie-Pierre Dubé at the Université de Montréal ran an analysis to identify potential candidate genes linked to FASD. Neuroinformatics lead Dr. Paul Pavlidis at the University of British Columbia made sure the statistical approach and methodology were rock solid, while Network investigator Dr. Michael Kobor performed and guided the DNA methylation analyses from his lab based at the BC Children’s Hospital.

Among the 16 co-authors were NeuroDevNet Investigator Dr. Albert Chudley, who has assembled an independent cohort to validate the study, and trainee Alexandre Lussier, who played a key role in the data analysis and preparation of the manuscript.

While the findings fall short of identifying a definitive biomarker for FASD to aid in early diagnosis, the researchers involved are encouraged. “All this information puts us closer,” says Reynolds.  “Now we’re down to a manageable number of candidates – a few dozen actual genes to investigate in more depth.”

Dr. Kobor is also pleased. “This is a very good start – it’s the largest study of its kind,” he says. “It sets a blueprint for what we could do in differentiating neurodevelopmental disorders in terms of methodology and approach. In FASD, we can use epigenetic tools to stratify – to characterize different profiles along the spectrum of FASD.

“With bigger cohorts and more coming online,” adds Dr. Kobor. “NeuroDevNet is primed to be a leader in that nationally – if not world-wide, in the three domains (FASD, autism and cerebral palsy) that we have.”

It is not yet known what mechanisms underlie alcohol’s impacts on the developing brain. Next steps in pursuing this understanding include the integration of massive data sets drawn from neuroimaging, together with these genetic findings. The sheer volume of data is daunting, Kobor and colleagues acknowledge, but advances in machine learning – computer programs that can teach themselves to grow and change when exposed to new data – hold promise for the future and the ability to “predict” FASD based on epigenetic patterns.

DNA methylation signature of human fetal alcohol spectrum disorder
Portales‑Casamar et al. Epigenetics & Chromatin (2016) 9:25
DOI 10.1186/s13072‑016‑0074‑4