Epigenetic and Structural Brain Aging and Their Associations With Major Depressive Disorder

Abstract

Background

A growing body of evidence suggests that major depressive disorder (MDD) may be associated with premature biological aging. However, most studies reported to date have examined brain-based (BrainAge) and DNA methylation (DNAm)–based measures (DNAmAge) of biological age (BioAge) in isolation.

Methods

We investigated 2 well-studied BioAge measures in lifetime and current MDD: BrainAge and DNAmAge (4 separate DNAmAge measures based on Horvath, Hannum, GrimAge, and PhenoAge clocks). We used cross-sectional cohort data from GS:STRADL (Generation Scotland: STratifying Resilience and Depression Longitudinally) (BrainAge n = 833; DNAmAge n = 587; age range 26–76 years) and used UK Biobank (UKB) data to test for replication of BrainAge associations with MDD (BrainAge n = 12,018, age range 45–80 years). Premature brain and DNAm aging were operationalized as predicted age difference (PAD), and analyses controlled for age, sex, smoking, and alcohol intake. We also tested individual and additive associations of brain- and DNAm-based PADs to lifetime/current MDD using logistic regression.

Results

Individuals with lifetime MDD showed significantly higher BrainAge and DNAmAge in GS, ranging from 1.60 to 2.45 years, than individuals without MDD for all measures except for Horvath age. No differences were found for BrainAge in the UKB. In terms of PAD, lifetime MDD was significantly associated with GrimAge-PAD, PhenoAge-PAD, and Brain-PAD, ranging from odds ratio (OR) = 1.21−1.30 (and in UKB, Brain-PAD OR = 1.05). DNAm-PAD and Brain-PAD demonstrated shared and distinctive associations with lifetime MDD, where PhenoAge-PAD plus Brain-PAD explained maximum variance (area under the curve = 0.69, R² = 9%). No significant associations were found for current MDD.

Conclusions

Our findings highlight shared and distinct associations of premature brain and DNAm aging in lifetime MDD.