Paternal Age-Related Changes in DNA Methylation from an Autism-Enriched Cohort

Friday, May 16, 2014: 5:06 PM
Marquis A (Marriott Marquis Atlanta)
J. I. Feinberg1, K. M. Bakulski2, R. Tryggvadottir1, S. C. Brown3, A. E. Jaffe4, L. R. Goldman5, L. A. Croen6, I. Hertz-Picciotto7, C. J. Newschaffer8, M. D. Fallin9 and A. P. Feinberg1, (1)Medicine, Johns Hopkins University, Baltimore, MD, (2)Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, (3)Mental Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, (4)Lieber Institute for Brain Development, Baltimore, MD, (5)George Washington University School of Public Health and Health Services, Washington, DC, (6)Division of Research, Kaiser Permanente Northern California, Oakland, CA, (7)UC Davis MIND Institute, Sacramento, CA, (8)Drexel University School of Public Health, Philadelphia, PA, (9)Johns Hopkins Bloomberg School of Public Health, Baltimore, MD
Background: Age-related epigenetic changes have been proposed to contribute to common human disease. Autism has been suspected to be associated with increased paternal age but the exact biological mechanisms are yet unknown.

Objectives: We examined the hypothesis that DNA methylation (DNAm) in semen-derived DNA samples might show age-specific alterations in the paternal genome with the intention that these might contribute to autism risk in offspring.

Methods: We performed comprehensive genome-wide DNAm analyses on DNA derived from semen samples contributed by fathers enrolled in the Early Autism Risk Longitudinal Investigation (EARLI) cohort.  We analyzed methylation data from 50 samples run on the CHARM 3.0 array, which measures DNAm at over 7 million CpG sites, in addition to data from 42 samples run on the Illumina Infinium 450k platform (~480,000 CpG sites).  A total of 32 samples were analyzed on both platforms.  Next, we used a well-tested statistical algorithm to identify regions in the genome with age-dependent changes in DNAm.  

Results: We identified differentially methylated regions (DMRs) of the paternal genome with significant age-dependent changes. Intriguingly, many of these regions contained genes enriched for neurodevelopmental pathways.

Conclusions:  This is the first evidence for an age-related change in DNAm in semen, which may be a plausible potential mechanism for the paternal age association with ASDs. While autism outcomes in the offspring of this cohort will not be available for several years, the EARLI cohort provided a unique biological resource of fathers of high-risk siblings, and such ASD-related analyses from these age-associated epigenetic findings are planned.

See more of: Genetics
See more of: Genetics