Epigenetic and population-based approaches to studying the association between socioeconomic status and depression risk

AIMS: Depression is a common mental disorder and the leading cause of disability world-wide (GBD 2018). With heritability measures around 37% (Sullivan, Neale, & Kendler, 2000), most of the risk for developing depression is of a likely environmental origin. Socioeconomic status has been repeatedly associated with depression and its many comorbid diseases (Gold et al., 2020) in studies that differ widely in their age range and geographical coverage. As a result, estimates of depression risk stratified by SES are imprecise and not easily generalisable. The biological mechanisms underlying the association between SES and depression are similarly poorly understood, although differences in alcohol consumption, smoking behaviour, and obesity are widely reported. Inflammatory mechanisms have also been implicated in both SES and depression. The aim of this thesis is to establish whether SES is associated with depression using the broadest possible evidence-base, to provide an estimate of depression risk linked to low SES, and to explore it’s underlying biological mechanisms. METHODS: I conducted a systematic review and meta-analysis of studies investigating the association between clinical depression and SES. There were no restrictions on country of origin, age of participants or publication date. The 140 included studies (N = 329,005, N of cases = 48,158) were meta-analysed (dichotomised into low and high SES) and reviewed, and gross national income per capita was used to interrogate differences in risk estimates and between-study heterogeneity. To investigate the biological mechanisms associated with low SES I performed a methylome wide association study (MWAS) on the Scottish Index of Multiple Deprivation (SIMD, a continuous measure) using the single largest DNA methylation (DNAm) dataset, Generation Scotland. Estimates from the SIMD MWAS were used to predict Index of Multiple Deprivation (IMD) in the Avon Longitudinal Study of Parents and Children (ALSPAC) as a replication analysis. I then performed an MWAS on the interaction of inflammatory disease and depression to further interrogate the findings from the SIMD MWAS and connect these findings to a possible low-SES subtype of depression. For both MWASs I performed differentially methylated point (DMP) and region (DMR) analysis, functional mapping and annotation of the differentially methylated genes, and gene set enrichment analysis. RESULTS: The systematic review and meta-analysis revealed an increased risk of depression associated with low SES (odds ratios from 1.57 to 1.95). The results were consistent across cultures, and SES measures. Low income was associated with the highest risk of depression. There was considerable between study effect size heterogeneity (I2 = 81-87%) in almost all meta-analysed outcomes. The MWAS of SIMD returned 15 epigenome wide significant DMPs, and 449 DMRs. Functional annotation and gene enrichment analysis revealed bone health, cardiovascular diseases, inflammatory diseases, and major depression as associated phenotypes. The inflammatory depression MWAS showed no epigenome wide significant DMP but 8 DMPs reached a less rigorous ‘discovery threshold’ and 7 epigenome wide significant DMRs were found. Genetic functional annotation and gene enrichment analysis implicated inflammatory markers, inflammatory diseases, response to SSRIs, liver and kidney function, and gluconeogenic pathways. CONCLUSION: Findings from this thesis show that low SES and risk of clinical depression are positively associated. Longitudinal studies hint at a low SES to MDD direction of causality, but the evidence is only suggestive, and it is likely to be a bidirectional relationship. Low SES is associated with many physical measures and conditions even after accounting for smoking, alcohol consumption, and BMI. Low SES is associated with a range of inflammatory markers and chronic mental and physical diseases. The biological mechanisms underlying the link between SES and depression may be related to systemic chronic inflammation. Possible sources for this systemic inflammation include several frequently comorbid chronic health problems and environmental toxins. These findings provide further impetus to investigate the mechanisms underlying the depression-SES association and develop interventions to mitigate the many health risks associated with low SES