P8 Investigating epigenetic differences in response to shift work: findings from understanding society (UK longitudinal household survey). (September 2018)

• Record Type: Journal Article
• Authors: Richmond, RC; Bao, Y; Smart, M; Gorrie-Stone, T; Schalkwyk, L; Mill, J; Davey Smith, G; Benzeval, M; Relton, C; Kumari, M
• Is Part Of: Journal of epidemiology and community health. Volume 72(2018)Supplement 1
• Page Start: A64
• Page End: A65
• Publication Date: 2018-09
• Subjects: Shift work -- epigenetic -- epidemiology; Public health -- Periodicals; Epidemiology -- Periodicals; 614.4
• DOI: 10.1136/jech-2018-SSMabstracts.134
• Languages: English
• ISSNs: 0143-005X

Abstract

Abstract : Background: Shift work is a feature of many occupations and has been associated with a range of adverse health outcomes, including obesity, diabetes, depression and cancer. The main proposed mechanism linking them is a disruption in circadian rhythms, particularly among night shift workers. DNA methylation may serve as a biomarker for circadian disruption and a potential mechanism by which shift work influences disease risk. In the context of a longitudinal study, we aimed to investigate whether shift work is associated with DNA methylation. Methods: Methylation profiling was performed using Illumina EPIC micro-arrays on whole-blood DNA samples, obtained from British Household Panel Survey (BHPS) participants of Understanding Society from 2010–2012. BHPS comprises a clustered random sample of households recruited in 1991, with all members followed annually. After pre-processing, 1175 samples and 857, 071 CpG sites remained for investigation. Shift work variables were derived from 17 time points between 1991 and 2009: ever (n=359), current (n=88) and long-term ( 3 3 years) shift work (n=154) (night and rotating). Epigenome-wide association analysis was performed using multivariable regression with adjustment for age, sex, batch and blood processing day. Further models were adjusted for cell-type composition and socio-economic variables. Methylation age was also estimated based on the Horvath epigenetic clock and the impact of shift work on 'epigenetic age acceleration' (EAA) was investigated. Results: In epigenome-wide association analysis, 50 CpG sites were associated with shift work with a nominal p-value<1 × 10 -5 across the 9 main analyses, with the strongest signal of association at cg12880856 ( PPARG ) identified in relation to ever working night shifts (p=7.3 × 10 −8 ). There was nominal evidence for an inverse association between current shift and rotating work with accelerated ageing (EAA) (mean difference=-0.98 (-1.93, -0.39); p=0.041 and −1.11 (−2.19, –0.04); p=0.041, respectively). Conclusion: Shift work was associated with differential methylation in blood, including a CpG site in PPARG . This CpG site is specific to the Illumina EPIC array, not being present on the predecessor 450 K array, and represents a potentially novel finding. PPAR-gamma has been implicated in the pathology of obesity, diabetes and cancer, although further work is required to appraise the causal effect of methylation on health outcomes. Furthermore, there was some evidence that current shift workers had decelerated methylation age compared with non-shift workers. However, the number of shift workers in this study was relatively small and further validation of findings is required.