Mitochondria-localized protein-encoding genes and programmed cell death-related genes reveal potential molecular perturbations underlying spontaneous preterm birth.

Multiple programmed cell death (PCD) modalities, including apoptosis, autophagy, and ferroptosis, are closely implicated in spontaneous preterm birth (SPTB). Mitochondria serve as central regulators of various PCD pathways, playing a critical role in cellular stress responses and homeostasis. However, comprehensive studies integrating mitochondria-localized protein-encoding genes and PCD-related genes to explore the molecular mechanisms underlying SPTB remain limited. Two gene expression datasets from the Gene Expression Omnibus (GEO) were analyzed. Mitochondria-localized protein-encoding genes from MitoCarta3.0 and curated PCD-related gene sets were combined to define a mitochondrial PCD-related gene set (mtPCD-RGs). Differentially expressed genes (DEGs) between SPTB and control samples were identified, and their intersection with mtPCD-RGs was obtained. A protein-protein interaction (PPI) network was constructed, followed by least absolute shrinkage and selection operator (LASSO) regression, random forest (RF) modeling to identify hub genes. External validation of gene expression was performed. Subsequent analyses included gene set enrichment analysis (GSEA), immune infiltration analysis, and the construction of competing endogenous RNA (ceRNA) network. A total of 63 intersecting genes were identified, and 46 genes were retained after PPI network filtering. BST2 and TRIM22 were eventually identified as key hub genes for SPTB, and their incorporation into a nomogram enabled individualized risk prediction for clinical application. Immunological analysis revealed significant differences in naïve CD4 + T cells, resting NK cells, monocytes, and eosinophils in SPTB peripheral blood samples compared to the control group. BST2 and TRIM22 showed positive correlations with monocyte abundance and negative correlations with naïve CD4 + T cells and resting NK cells. BST2 and TRIM22 are identified as pivotal genes in SPTB, potentially reflecting different outcomes of pregnancy, respectively. These genes may serve as potential biomarkers and provide mechanistic insights into SPTB, highlighting promising targets for future therapeutic interventions.