Li · Clinical epigenetics 2020 · pooled family and twin observational study · n=4217

Genetic and environmental causes of variation in epigenetic aging across the lifespan.

Cited 69 times in the scientific literature.

Level 3 - non-randomized controlled study

Cross-sectional and familial variance-component analysis of pooled observational cohorts (analogy for non-intervention study).

PubMed 33092643 · doi:10.1186/s13148-020-00950-1 · record verified 2026-08-30

What was done

Genome-wide DNA methylation data were pooled for 4,217 individuals aged 0 to 92 years across 1,871 families. DNA methylation age was computed primarily using the Horvath epigenetic clock, alongside two other epigenetic clocks. Investigators evaluated familial correlations across monozygotic twin, dizygotic twin, sibling, parent-offspring, and spouse pairs stratified by cohabitation status, fitting genetic and environmental variance components models.

What was found

Twin pair correlations around birth ranged from -0.12 to 0.18 and were not significantly different from zero (all P > 0.29). Across all relative pairs, correlations increased with duration of living together (all P < 0.02), rising faster in monozygotic twins than in dizygotic twins/siblings, and faster in siblings than in parent-offspring pairs (P < 0.001). Correlations declined after living apart (P = 0.02). Shared environmental effects were 1.41 times greater (95% CI 1.16 to 1.66) for monozygotic pairs compared to dizygotic/sibling pairs, and 2.03 times greater (95% CI 1.13 to 9.47) for dizygotic/sibling pairs compared to parent-offspring pairs. Additive genetic factors explained an estimated 13% of variance (95% CI -10% to 35%), which was not statistically significant (P = 0.27).

Why it matters

These findings suggest that epigenetic aging clocks reflect cumulative environmental and lifestyle exposures rather than strong genetic programming. They also challenge the equal-environment assumption traditionally used in twin-based heritability estimates of epigenetic clocks.

Limits

The study relies on pooled cross-sectional family data rather than long-term longitudinal tracking of individual methylation changes across a lifetime. Specific environmental exposures responsible for shared variance were not measured or identified.

Cited by