Integrated transcriptomic and co-expression network analysis identifies HIF1A as a key immune regulator in biliary...

Biliary atresia (BA) is a rapidly progressive neonatal cholangiopathy with unclear pathogenesis. This study aimed to elucidate the key molecular mechanisms and regulatory networks underlying BA through integrative transcriptomic analysis, providing insights for early diagnosis and targeted therapy.

Three BA-related liver microarray datasets (GSE46960, GSE159720, and GSE15235) were retrieved from the Gene Expression Omnibus (GEO) database and analyzed using a unified pipeline. Differentially expressed genes (DEGs) in the training cohort (GSE46960; 64 BA vs. 14 controls) were identified with limma under Benjamini-Hochberg false discovery rate (BH-FDR) control. Functional enrichment [Gene Ontology (GO)/Kyoto Encyclopedia of Genes and Genomes (KEGG)], xCell-based immune infiltration analysis, and gene-set intersection with Hedgehog and NOTCH developmental pathways were performed. HIF1A-stratified transcriptional programs were evaluated by DEG analysis and Gene Set Enrichment Analysis (GSEA). In the validation cohorts, we assessed expression trends and effect sizes across BA subtypes and HIF1A-defined groups. Weighted gene co-expression network analysis (WGCNA) was then applied to DEGs from the inflammatory subtype (GSE15235) to identify co-expression modules associated with HIF1A.

We identified 343 DEGs between BA and controls, predominantly enriched in cytokine signaling, immune activation, and fibrosis-related pathways, whereas downregulated genes were linked to erythrocyte function and lipid metabolism. Intersection of DEGs with Hedgehog and NOTCH gene sets highlighted six development-related genes; among them, only HIF1A showed consistent and significant upregulation in independent BA samples. Intersection with immune-related gene sets yielded 29 immune-related DEGs centered on HIF1A in the protein-protein interaction (PPI) network. Across cohorts, high HIF1A expression was associated with enhanced enrichment of neutrophils, monocytes, and T cells, as well as positive correlations with neutrophil markers and major histocompatibility complex (MHC) genes. HIF1A-stratified DEGs were enriched in tumor necrosis factor (TNF) and hypoxia-inducible factor-1 (HIF-1) signaling, inflammatory and cytokine-response pathways, and perturbations of bile acid and adenosine triphosphate (ATP)-binding cassette (ABC) transporter pathways. WGCNA of HIF1A-associated DEGs identified a key co-expression module (MEblue) strongly correlated with HIF1A and enriched for leukocyte chemotaxis and cytokine signaling.

HIF1A may play a central role in the pathogenesis of BA by linking hypoxia responses with immune and fibrotic remodeling, particularly via neutrophil- and cytokine-driven inflammation. These findings suggest that HIF1A and its downstream networks represent promising candidates for early detection and targeted intervention in BA, warranting validation in larger, clinically annotated cohorts and experimental models.