DNA Methylation and Cross-Tissue Integration Shows Utility Of Blood-Based Epigenetic Research for ASD

Oral Presentation
Saturday, May 12, 2018: 1:57 PM
Willem Burger Hal (de Doelen ICC Rotterdam)
M. D. Fallin1, S. Andrews2, B. K. Sheppard3, K. M. Bakulski4, D. Schendel5, L. Schieve6, G. C. Windham7, L. A. Croen8, I. Hertz-Picciotto9, S. Warren10, P. Chopra10, R. Alisch11, S. Ellis12, J. Mill13, E. Hannon14, C. J. Newschaffer15, D. Arking16 and C. Ladd-Acosta1, (1)Wendy Klag Center for Autism and Developmental Disabilities, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, (2)Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, (3)Johns Hopkins School of Public Health, Baltimore, MD, (4)University of Michigan, Ann Arbor, MI, (5)Aarhus University, Aarhus, Denmark, (6)Centers for Disease Control and Prevention, Atlanta, GA, (7)Environmental Health Investigations Branch, California Department of Public Health, Richmond, CA, (8)Division of Research, Kaiser Permanente, Oakland, CA, (9)University of California at Davis, Davis, CA, (10)Human Genetics, Emory University, Atlanta, GA, (11)Psychiatry, University of Wisconsin, Madison, WI, (12)Johns Hopkins University, Baltimore, MD, (13)University of Exeter Medical School, University of Exeter, Exeter, United Kingdom, (14)Exeter Medical School, University of Exeter, Exeter, United Kingdom, (15)AJ Drexel Autism Institute, Philadelphia, PA, (16)Johns Hopkins University School of Medicine, Baltimore, MD
Background: Epigenetic mechanisms have been implicated in ASD, including evidence from rare-variant genetics findings. Further, common variant GWAS studies for ASD have recently increased in size and precision. Few have integrated genetic and epigenetic findings in the same study of ASD, partly due to skepticism about the utility of blood-based DNA methylation studies for this brain-based disorder.

Objectives: This work aimed to take genome-wide approaches at the DNA methylation (epigenome-wide, EWAS) and SNP variant (GWAS) levels and to integrate information from both to elucidate biological insights not possible via either type of information in isolation.

Methods: DNA methylation was measured on 968 child blood samples from the Study to Explore Early Development (SEED 1) and 343 child-sibling pairs from the Simons Simplex Collection (SSC) using the Illumina 450K Beadchip array. We performed EWAS meta-analysis, after extensive data cleaning and adjustment, and compared top EWAS results to those from brain-based results. We also used a combination of GWAS SNP data and 450K array DNAm data measured on infant cord blood from the EARLI study and SEED child blood to identify SNPs associated with DNA methylation (meQTL lists) in these tissues. We tested for enrichment of methylation-QTL (meQTL) CpG targets in our EWAS results. We also took the reverse approach of exploring the most recent PGC-AUT GWAS results for enrichment of meQTLs across cord, peripheral blood and publicly available fetal brain and lung meQTL lists. We further assessed biological pathways implicated by ASD-specific meQTLS and their CpG targets.

Results: While no single CpG met epigenome-wide statistical significance (p < 1.12x10-7) in our EWAS meta-analysis, 7 showed differences at p < 1x10-5 and 48 at 1x10-4. Of the top 7, 5 showed brain-based ASD associations, often with larger effect sizes. We also observed suggestive evidence for enrichment of CpG sites controlled by SNPs (meQTL targets) among the EWAS hits. When considering GWAS hits, we previously discovered enrichment of ASD-associated SNPs for fetal brain (OR = 3.55; P < 0.001) and peripheral blood meQTLs (OR = 1.58; P < 0.001). The CpG targets of ASD meQTLs across cord blood, peripheral blood, and brain tissues showed enrichment for immune-related pathways, consistent with other expression and DNAm results in ASD, and revealed pathways not implicated by genetic findings alone. We will update these enrichment findings with the latest PGC-AUT GWAS hits for INSAR presentation.

Conclusions: We report the largest case-control EWAS study of ASD to date. Our results suggest the potential to observe disease associations from blood-based samples, given the concordance of findings between blood and brain EWAS among suggestive hits. However, the greater current utility may be in using DNAm data to inform GWAS results, where we observed enrichment of ASD-associated SNPs for meQTLs and showed involvement of immune biology when considering the epigenetic targets of these ASD SNPs. Our joint analysis of genotype and DNAm demonstrates the potential of both brain and blood-based DNAm for insights into ASD.

See more of: Genetics
See more of: Genetics