DNA methylation signature of passive smoke exposure is less pronounced than active smoking: the Understanding Society study

Introduction The extent of the biological impact of passive smoke exposure is unclear. We sought to investigate the association between passive smoke exposure and DNA methylation, which could serve as a biomarker of health risk. Materials and methods We derived passive smoke exposure from self-reported questionnaire data among smoking and non-smoking partners of participants enrolled in the UK Household Longitudinal Study ‘Understanding Society’ (n=769). We performed an epigenome-wide association study (EWAS) of passive smoke exposure with DNA methylation in peripheral blood measured using the Illumina Infinium Methylation EPIC array. Results No CpG sites surpassed the epigenome-wide significance threshold of p<5.97 × 10−8 in relation to partner smoking, compared with 10 CpG sites identified in relation to own smoking. However, 10 CpG sites surpassed a less stringent threshold of p<1 × 10−5 in a model of partner smoking adjusted for own smoking (model 1), 7 CpG sites in a model of partner smoking restricted to non-smokers (model 2) and 16 CpGs in a model restricted to regular smokers (model 3). In addition, there was evidence for an interaction between own smoking status and partners’ smoking status on DNA methylation levels at the majority of CpG sites identified in models 2 and 3. There was a clear lack of enrichment for previously identified smoking signals in the EWAS of passive smoke exposure compared with the EWAS of own smoking. Conclusion The DNA methylation signature associated with passive smoke exposure is much less pronounced than that of own smoking, with no positive findings for ‘expected’ signals. It is unlikely that changes to DNA methylation serve as an important mechanism underlying the health risks of passive smoke exposure.