DNA methylation signatures of C-reactive protein associations with structural neuroimaging measures and major depressive disorder

Abstract

Background: Inflammatory processes are implicated in the aetiology of Major Depressive Disorder (MDD); however, the relationship between peripheral inflammation, brain structure and depression remains unclear. This study investigates associations between depression, structural neuroimaging measures and two measures of inflammation: serum-based C-reactive protein (CRP) and a methylation-based score for CRP (DNAm CRP) in a large community-based sample. Methods: Serum CRP and DNAm CRP were assessed for participants in Generation Scotland (nMDD cases= 240, nControls= 558) alongside structural brain scans (T1 and diffusion MRI). Linear regression was used to investigate associations between (i) both CRP measures and depressive symptoms, (ii) both CRP measures and structural imaging variables and (iii) inflammation x MDD interaction effects (for both CRP measures) with imaging measures. Results: Increased serum CRP was significantly associated with overall MDD symptom severity and specifically with somatic symptoms- general interest (β= 0.145, pFDR = 6x10-4) and energy levels (β= 0.101, pFDR = 0.027) and also reduced entorhinal cortex thickness (β= -0.095, pFDR = 0.037). DNAm CRP was significantly associated with reduced global grey matter/cortical volume and reduced integrity of 16 white matter tracts and showed larger effect sizes (βaverage= -0.15) compared to serum CRP across all measures (βaverage = 0.01). Conclusions: Acute measures of CRP were related to current depression symptoms, specifically somatic symptoms, whereas methylation signatures of inflammation demonstrated greater differences in global and regional brain structure. This study highlights the utility of combining serological and methylation markers to study chronic inflammation effects on the brain in psychiatric disorders.

A Deep Survival EWAS approach estimating risk profile based on pre-diagnostic DNA methylation: An application to breast cancer time to diagnosis

“…One is related to the lack of adjustments for clinical covariates in our analyses. However, this choice was supported by recent literature, suggesting that DNAm signature of risk factors may predict diseases better than measured risk factor itself [33][34][35]. Therefore, adjustment for clinical covariates and lifestyle related risk factors that is usually implemented in EWAS studies may remove significant variability due to genetic and metabolic profiles that can influence the response to exposures and stressors.…”