Prenatal maternal smoke, DNA methylation, and multi-omics of tissues and child health

Cosín Tomàs, Marta; Cilleros-Portet, Ariadna; Aguilar Lacasaña, Sofía; Fernandez-Jimenez, Nora; Bustamante Pineda, Mariona

doi:http://dx.doi.org/10.1007/s40572-022-00361-9

Prenatal maternal smoke, DNA methylation, and multi-omics of tissues and child health

License

Citation

Cosin-Tomas M, Cilleros-Portet A, Aguilar-Lacasaña S, Fernandez-Jimenez N, Bustamante M. Prenatal maternal smoke, DNA methylation, and multi-omics of tissues and child health. Curr Environ Health Rep. 2022 Sep;9(3):502-12. DOI: 10.1007/s40572-022-00361-9

Abstract

Purpose of review: Maternal tobacco smoking during pregnancy is of public health concern, and understanding the biological mechanisms can help to promote smoking cessation campaigns. This non-systematic review focuses on the effects of maternal smoking during pregnancy on offspring's epigenome, consistent in chemical modifications of the genome that regulate gene expression. Recent findings: Recent meta-analyses of epigenome-wide association studies have shown that maternal smoking during pregnancy is consistently associated with offspring's DNA methylation changes, both in the placenta and blood. These studies indicate that effects on blood DNA methylation can persist for years, and that the longer the duration of the exposure and the higher the dose, the larger the effects. Hence, DNA methylation scores have been developed to estimate past exposure to maternal smoking during pregnancy as biomarkers. There is robust evidence for DNA methylation alterations associated with maternal smoking during pregnancy; however, the role of sex, ethnicity, and genetic background needs further exploration. Moreover, there are no conclusive studies about exposure to low doses or during the preconception period. Similarly, studies on tissues other than the placenta and blood are scarce, and cell-type specificity within tissues needs further investigation. In addition, biological interpretation of DNA methylation findings requires multi-omics data, poorly available in epidemiological settings. Finally, although several mediation analyses link DNA methylation changes with health outcomes, they do not allow causal inference. For this, a combination of data from multiple study designs will be essential in the future to better address this topic.