Immune-related hub genes in intrauterine adhesions: a bioinformatics approach

Intrauterine adhesions (IUA) are a leading cause of acquired female infertility that predominantly arises following surgical intrauterine interventions. Clinical strategies are available for managing IUA, however, the molecular pathogenesis of IUA, particularly the role of immune dysregulation in endometrial repair processes, has not been fully characterized, necessitating comprehensive mechanistic studies.

We used computational biology methods to determine the molecular pathogenesis of IUA, the results of which were experimentally validated. (i) We systematically reanalyzed GSE224093, a publicly available endometrial transcriptomic dataset, using GEO2R. Differential gene expression was analyzed with stringent statistical thresholds; the results were immunologically contextualized via intersection with the ImmPort database. (ii) Multilayer functional annotation was conducted using Gene Ontology (GO) enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway topology analyses. (iii) Weighted gene co-expression network analysis and scale-free topology optimization were used to identify conserved coexpression modules. (iv) Seven hub genes underwent quantitative real-time polymerase chain reaction (qPCR) validation in human and murine models, with four verified using immunohistochemistry (IHC).

Our integrated multiomics analysis identified seven hub genes associated with IUA pathogenesis through GO, KEGG, and GSEA enrichment analyses. The expression levels TUBB3, WNT5A, GDF7, IGF1, and BIRC5 were downregulated, and those of PTGDS and CCL14 were upregulated. The qPCR results confirmed these expression patterns in human and murine endometria (p < 0.05), with TUBB3 and PTGDS species-specifically diverging from the computational predictions. The IHC results provided the corresponding protein expression changes for IGF1, WNT5A, BIRC5, and GDF7 in IUA (p < 0.01). CCL14 could not be amplified in murine models due to technical constraints.

We identified seven immune-related gene signatures through integrative bioinformatics. We experimentally validated three genes (TUBB3, PTGDS, and CCL14) demonstrating species-specific expression patterns. We proposed four mechanistically plausible biomarkers (WNT5A, IGF1, BIRC5, and GDF7) for developing IUA diagnostic tools. The conserved dysregulation of WNT5A- and TGF-β-associated genes (GDF7 and IGF1) suggest therapeutic targets for preventing adhesion recurrence. This study advances our understanding of IUA pathogenesis. Single-cell transcriptomics should be examined in future studies to determine the cellular-subtype-specific contributions to IUA.