Identifying epigenetic and microbial biomarkers for preterm birth using DNA methylation and gut microbiome data

Preterm birth (PTB), defined as delivery before 37 weeks, is a major cause of neonatal morbidity and mortality worldwide. Evidence suggests that both epigenetic dysregulation and gut microbial imbalance contribute to the inflammatory and metabolic disturbances associated with PTB; however, few studies have examined these factors in conjunction to identify integrated predictive biomarkers.

This study aimed to identify epigenetic and microbial signatures associated with PTB by integrating maternal second-trimester genome-wide DNA methylation profiles with gut microbiome composition.

A case-control study was conducted with 120 pregnant women, grouped into two categories: preterm (≤37 weeks, n = 60) and full-term (>37 weeks, n = 60). Maternal blood and fecal samples were collected simultaneously. DNA methylation was profiled using the Illumina MethylationEPIC array, and gut microbiota were characterized through 16S rRNA sequencing. Differentially methylated regions (DMRs) and differentially abundant taxa were identified using FDR < 0.05. Sparse canonical correlation analysis and network modeling were applied to integrate the datasets and identify linked epigenetic-microbial features predictive of PTB.

Women with PTB showed distinct methylation changes in immune and inflammation-related genes, including IL6, CXCL10, TNFAIP3, and PPARGC1A. Gut microbiome analysis revealed significantly reduced α-diversity (Shannon index: 2.81 ± 0.31 vs. 3.42 ± 0.36, p = 0.002), enrichment of pro-inflammatory taxa (Prevotella, Sutterella, Veillonella), and depletion of beneficial genera, including Lachnospiraceae, Faecalibacterium, and Bifidobacterium. Integrated analysis showed strong cross-domain associations (r = 0.68, p < 0.001), linking immune-gene hypomethylation with enrichment of inflammatory taxa. The combined biomarker panel achieved high predictive performance (AUC = 0.87; sensitivity = 82%; specificity = 84%), outperforming methylation-only and microbiome-only models. Functional enrichment highlighted convergence on NF-κB signaling, cytokine interactions, and butyrate metabolism.

The study shows that coordinated epigenetic alterations and gut microbial dysbiosis contribute to PTB. These biomarkers are detectable during the second trimester (18-24 weeks), supporting their potential for mid-pregnancy risk stratification. The integrated methylation-microbiome signature offers strong potential for early, non-invasive prediction and supports development of targeted maternal interventions.