Differences in heart DNA methylation between sudden infant death syndrome and other sudden deaths

Abstract Background Sudden Infant Death Syndrome (SIDS) remains a leading cause of infant mortality worldwide, yet its underlying mechanisms are largely unknown. SIDS etiology is complex and involves the interplay between a vulnerable infant, a critical developmental period, and external stressors. Cardiac function, including cardiorespiratory autonomic control, is considered a major contributor to SIDS. Emerging evidence suggests that epigenetic modifications of heart tissue, particularly DNA methylation (DNAm), may contribute to SIDS risk. We conducted an epigenome-wide association study (EWAS) to determine if DNA methylation patterns of post-mortem heart tissue differ between SIDS cases and non-SIDS controls, and to evaluate the role of DNAm in timing of SIDS events. Methods Using a case–control study design, we generated DNA methylation data using Illumina EPIC Methylation V2 arrays and performed single site EWAS, differentially methylated region (DMR), and epigenome network cluster analyses to assess differential DNA methylation between SIDS cases and non-SIDS controls. Post-mortem heart tissues from 324 infants (202 SIDS cases) were obtained from the Chicago Infant Mortality Study (CIMS) and the NIH NeuroBioBank (NBB). Inclusion criteria required SIDS cases to meet standardized SIDS definitions, as determined from a through case investigation (i.e., autopsy, death scene examination, and clinical history). Results Seven CpG sites were differentially methylated between SIDS cases and non-SIDS control (adjusted (adj.) p < 0.05). Two were located within DTNB (cg1703363; adj. p = 0.02) and NDUFS8 (cg12070987; adj. p = 0.02). There were two differentially methylated regions overlapping the KRTCAP3 and RUNX3 genes. Of the top ten most significant CpG-SIDS diagnosis interactions, two of them were located within mitochondrial genes, ACSL1 and AK2. Conclusions This study identified differentially methylated CpG sites and regions from heart tissue in SIDS cases versus controls, notably in DTNB, NDUFS8, KRTCAP3, and RUNX3. Our findings highlight pathways enriched for mitochondrial function, immune cell regulation, and cardiac dysfunction, which indicate the heterogenous nature of SIDS and supports the need for larger multi-tissue collaborative studies to identify SIDS subtypes, underlying pathologies, and determine the causal nature of such findings.