Association of prenatal alcohol exposure with offspring DNA methylation in mammals: a systematic review of the evidence

Abstract Background Prenatal alcohol exposure (PAE) is associated with a range of adverse offspring neurodevelopmental outcomes. Several studies suggest that PAE modifies DNA methylation in offspring cells and tissues, providing evidence for a potential mechanistic link to Fetal Alcohol Spectrum Disorder (FASD). We systematically reviewed existing evidence on the extent to which maternal alcohol use during pregnancy is associated with offspring DNA methylation. Methods A systematic literature search was conducted across five online databases according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. PubMed, Web of Science, EMBASE, Google Scholar and CINAHL Databases were searched for articles relating to PAE in placental mammals. Data were extracted from each study and the Risk of Bias in Non-Randomized Studies of Interventions (ROBINS-I) was used to assess the potential for bias in human studies. Results Forty-three articles were identified for inclusion. Twenty-six animal studies and 16 human studies measured offspring DNA methylation in various tissues using candidate gene analysis, methylome-wide association studies (MWAS), or total nuclear DNA methylation content. PAE dose and timing varied between studies. Risk of bias was deemed high in nearly all human studies. There was insufficient evidence in human and animal studies to support global disruption of DNA methylation from PAE. Inconclusive evidence was found for hypomethylation at IGF2/H19 regions within somatic tissues. MWAS assessing PAE effects on offspring DNA methylation showed inconsistent evidence. There was some consistency in the relatively small number of MWAS conducted in populations with FASD. Meta-analyses could not be conducted due to significant heterogeneity between studies. Conclusion Considering heterogeneity in study design and potential for bias, evidence for an association between PAE and offspring DNA methylation was inconclusive. Some reproducible associations were observed in populations with FASD although the limited number of these studies warrants further research. Trail Registration: This review is registered with PROSPERO (registration number: CRD42020167686).摘要 背景 产前酒精暴露（Prenatal Alcohol Exposure, PAE）与子代一系列不良神经发育结局密切相关。多项研究显示，PAE可改变子代细胞及组织的DNA甲基化状态，为其与胎儿酒精谱系障碍（Fetal Alcohol Spectrum Disorder, FASD）之间存在潜在机制关联提供了实证支持。本研究系统回顾了现有证据，旨在明确妊娠期母体饮酒与子代DNA甲基化之间的关联程度。 方法 本研究遵循系统综述与荟萃分析优先报告条目（Preferred Reporting Items for Systematic Reviews and Meta-Analyses, PRISMA）指南，在5个在线数据库中开展系统性文献检索。检索覆盖PubMed、Web of Science、EMBASE、Google Scholar及CINAHL数据库，筛选与胎盘类哺乳动物PAE相关的研究文献。提取各项研究的核心数据，并采用非随机干预研究偏倚风险工具（Risk of Bias in Non-Randomized Studies of Interventions, ROBINS-I）评估人体研究的潜在偏倚风险。 结果 本研究共筛选纳入43篇相关文献，其中26项动物研究与16项人体研究，通过候选基因分析、甲基化组全基因组关联研究（methylome-wide association studies, MWAS）或总核DNA甲基化含量检测，对子代不同组织的DNA甲基化水平进行了测定。各项研究中PAE的暴露剂量与暴露时机均存在差异。几乎所有人体研究的偏倚风险均被评定为高风险。现有人体及动物研究证据均不足以支持PAE会导致DNA甲基化发生全局性紊乱。在体细胞组织的IGF2/H19区域出现低甲基化的相关证据尚无定论。针对PAE对子代DNA甲基化影响的MWAS研究结果并不一致。在少量针对FASD人群开展的MWAS研究中，观察到了一定的一致性结果。由于各项研究间异质性显著，无法开展荟萃分析。 结论 鉴于研究设计存在异质性且存在潜在偏倚风险，现有证据无法明确PAE与子代DNA甲基化之间存在明确关联。在FASD人群中观察到了部分可重复的关联，但此类研究数量有限，仍需开展进一步研究。 试验注册：本综述已在PROSPERO注册平台注册（注册编号：CRD42020167686）。