Poster #11 - Childhood Environmental Unpredictability and Adolescent Epigenetic Aging

• In Event: Poster Session 09

Abstract

Childhood adversity takes a toll on individuals across the lifespan and is associated with health disparities. There is great interest in identifying biomarkers of biological weathering that accounts for variation in health and disease. Recent research has identified DNA methylation markers that track with the aging process (e.g., epigenetic clocks). Higher than expected epigenetic age relative to chronological age is posited as a measure of accelerated aging and has been associated with greater risk for disease and mortality later in life. However, deviations in epigenetic age from chronological age may begin much earlier in life and in part be due to childhood adversity. We build on recent calls by developmental scientists to investigate the role of environmental unpredictability as a form of childhood adversity and examine environmental unpredictability in predicting adolescent epigenetic aging.

Data are from the Fragile Families and Child Wellbeing Study, a nationally representative longitudinal study of 4898 youth born in large cities (52% Male; 48% Black, 27% Hispanic, 21% White). The study includes a large sample of children from low-income, single-parent, and racially diverse families. To create an unpredictability index, the surveys at ages 1, 3, 5 and 9 (all available childhood data) were examined for unpredictability across multiple ecological levels (e.g., family, employment) and multiple timescales (e.g., daily experiences, major transitions; see Table 1). The core feature of unpredictability we use is the lack of consistency or variation in the occurrence of environmental experiences. Saliva samples obtained at age 15 were assayed for DNA methylation (Infinium 450K and EPIC arrays n=852 and n=1114, respectively). As research on epigenetics has expanded, multiple clocks have been developed to assess epigenetic aging based on multiple aspects of aging (e.g., chronological age, mortality, biomarkers). Epigenetic age was calculated using 8 different epigenetic clocks (see Table 1). Chronological age was removed to create measures of biological weathering.

We examined childhood environmental unpredictability as a predictor of age 15 accelerated epigenetic aging (Table 2; results presented by array type). Preliminary findings varied across the 8 epigenetic clocks. Findings from 450K array subsample show few associations with adversity; there is some evidence that unpredictability may be associated with less epigenetic aging for some clocks trained on chronological age. Findings from EPIC array subsample demonstrate that childhood environmental unpredictability is associated with accelerated epigenetic aging at age 15 based on the GrimAge clock (β=.05, p<.05), PhenoAge clock (β=.07, p<.05), and DunedinPACE clock (β=.03, p=.054). Rather than chronological age, these clocks were trained on mortality risk (GrimAge), clinical markers of aging (PhenoAge), and longitudinal change in biomarkers (DunedinPACE). These findings suggest that clocks designed to assess broader aspects of aging, may be more sensitive to being shaped by adversity. Additional analyses will explore sample characteristics that may distinguish the two subsamples.

Preliminary findings demonstrate that environmental unpredictability may be associated with accelerated epigenetic aging consistent with biological weathering hypotheses. However, results varied across clocks used to calculate epigenetic aging suggesting careful consideration is needed for methodological choices in studies of childhood adversity and epigenetic aging.

Authors

• Kalsea J Koss, University of Georgia
  Presenting Author

• Rachel Ashley Brown, University of Georgia

• Michael V Cataluna, University of Georgia

• Michelle Barth, University of Georgia

• Lisa Schneper, Princeton University

• Colter Mitchell, University of Michigan - Ann Arbor

• Jeanne Brooks-Gunn, Columbia University

• Daniel Notterman, Princeton University