Genome-Wide Analyses of DNA Methylation in Autism Brains Suggest Epigenetic-Mediated Effects in GABA Signaling

Oral Presentation
Saturday, May 12, 2018: 2:09 PM
Willem Burger Hal (de Doelen ICC Rotterdam)
J. I. Young1, A. J. Griswold2, C. Brandenburg3, J. P. Hussman4, M. A. Pericak-Vance2, G. J. Blatt3 and J. M. Vance5, (1)John P. Hussman Institute for Human Genomics, University of Miami, Miami, FL, (2)John P. Hussman Institute for Human Genomics, University of Miami Miller School of Medicine, Miami, FL, (3)Hussman Institute for Autism, Baltimore, MD, (4)Hussman Institute for Autism, Catonsville, MD, (5)Hussman Institute for Human Genomics, Miami, FL
Background: Ample evidence supports involvement of the striatum and cerebellum in autism pathophysiology. Emerging evidence implicates altered DNA methylation as a likely contributing factor to autism etiology.

Objectives: In order to gain understanding of the role of DNA methylation and transcriptional regulation in autism, we profiled the methylome and transcriptome of striatal and cerebellar tissue from autistic and neurotypical individuals.

Methods: DNA and RNA were extracted from frozen striatum and cerebellar tissue from 5 autism cases and 5 neurotypical individuals obtained from the Maryland Brain and Tissue Bank. Samples were matched for age, sex and post-mortem interval. Methylation profiles were analyzed using MethylationEPIC bead arrays. As CpG methylation is often correlated amongst neighboring sites, we used differentially methylated regions (DMR) for this analysis. Differential gene expression was assayed through RNA-seq analysis. Differentially methylated genes (containing DMRs) as well as genes differentially expressed were further analyzed to determine functional pathways and ontology enrichment using the Enrichr web tool. In addition, transcription factor binding site enrichment was identified using oPOSSUM and GATHER-based analysis of TRANSFAC motifs.

Results: We found significant DMRs between cases and controls, including DMRs in genes previously implicated in autism. Pathway analysis of genes containing DMRs in the striatum indicates significant enrichment in the Kegg pathway “GABAergic synapse”, whereas the Kegg pathway “endocytosis” is enriched in the cerebellum. Both pathways have been previously postulated to contribute to autism pathophysiology.

RNA-seq based analysis of differential gene expression identified differentially expressed genes (DEG) with FDR <0.05 in both brain regions, including several that have been implicated in autism. Interestingly, pathway analysis found that striatal DEG genes were enriched for the Kegg pathway “Neuroactive ligand-receptor interaction” and the gene ontology category “GABA-A receptor activity”. Analysis of TFBS revealed enrichment of PAX6 binding to striatal DEG genes. Our data also shows that PAX6 contains a significantly hypomethylated DMR in striatum in autism individuals as well.

Conclusions: We have identified tissue-specific epigenomic alterations in autism brains, including alterations in known autism-associated genes. The etiology of these changes are not known, but could be inherited or due to environmental exposures during development. Both DMR and differential gene expression point to GABA signaling mechanisms as important in autism pathophysiology. These pilot data suggest that epigenetic evaluation of selected brain regions has high potential for autism studies. In addition, converging molecular alterations may provide insight into mechanisms of disease even with smaller sample sets than those required for genetic studies.

See more of: Genetics
See more of: Genetics