Associations Between Salivary DNA Methylation and War Exposure in Syrian Refugee Children and Adolescents

• In Event: Challenges to developing biology and mental health

Abstract

Background: War is a source of adverse or traumatic experiences which can have long-term impacts on physical and mental health. Epigenetic modifications such as DNA methylation (DNAm) are a possible mechanism to mediate the biological effect of such exposures, including war. However, no studies have investigated postnatal war-induced changes to DNAm in childhood and adolescence. Furthermore, most studies aiming to identify DNAm associated with exposure to events such as war tend to take a candidate gene approach, focussing on biological systems associated with stress reactivity or mental health. While some of these findings are replicated, others are not and are rarely identified within epigenome-wide association studies (EWASs). EWASs have identified additional DNAm loci associated with adverse or traumatic experiences, but to date no EWAS has been conducted on war-exposed civilian populations. More recently, epigenetic ageing has been studied in relation to adverse environments. This multi-site DNAm measure of biological ageing is frequently accelerated following adversity, even in childhood and adolescence. However, epigenetic ageing has not yet been investigated in relation to war exposure.
Aims: Overall, we aim to comprehensively assess the DNAm differences that exist following exposure to war-related events in children and adolescents. We firstly aim to identify DNAm differences associated with war exposure in a hypothesis-free manner using an EWAS design. Secondly, we aim to replicate previously published DNAm differences associated with traumatic or adverse experiences. Finally, we test whether epigenetic age acceleration (derived from epigenetic clocks) is associated with war exposure.
Study population: A sample of 1,468 Syrian refugee children and adolescents (6-19 years) living in informal tented settlements in Lebanon for which we have DNAm data. 896 were successfully followed up one year later with DNAm data.
Methods: Repeated interview data and saliva samples were collected across the two study waves. Interview data were collected face-to-face with trained local fieldworkers within the family home and included reporting of war-related events using the War Events Questionnaire. Salivary DNA was assessed for genome wide DNAm using the Illumina MethylationEPIC BeadChip array. Epigenetic age acceleration was calculated using DNAm clocks chosen due to their prevalence in the literature in addition to those designed for paediatric populations. For all analyses, linear mixed models will be fitted with DNAm proportion, or epigenetic age acceleration as the outcome variable and war exposure as a predictor. Where available we will utilise the repeated samples available for each individual, assigning the individual as a random effect within the model, to account for intra-individual variation in DNAm that may exist.
Results: Analyses indicate that while the number of different war-related events was not associated with accelerated epigenetic ageing (β=-0.0046, p=0.60), exposure to war-related violence in the home was associated with accelerated ageing estimated from some DNAm clocks. For example, age acceleration estimated from Horvath’s multi-tissue clock was elevated in exposed individuals (β=0.21, p=0.049), and this was increased when controlling for the number of different war-related events (β=0.35, p=0.0066). Results for the EWAS and replication study will be presented.

Authors

• Demelza Smeeth, Queen Mary University of London
  Presenting Author

• Stephan Beck

• Elisabeth B Binder, Max Planck Institute of Psychiatry

• Elie Karam, Institute for Development, Research, Advocacy and Applied Care

• Michael Pluess, Queen Mary University of London