Comprehensive microbiome and metabolome analysis revealed the changes of semen microbial characteristics and metabolic phenotypes in patients with idiopathic oligoasthenozoospermia

Background The etiology and pathogenesis of idiopathic oligoasthenospermia (IOA) remain unclear, and current treatment options yield suboptimal outcomes. Consequently, there is an urgent need to identify novel biomarkers and develop diagnostic tools to improve patient identification and clinical management. Multi-omics technologies offer a promising pathway toward achieving this goal in the future. Methods This study included 40 untreated patients with idiopathic oligoasthenospermia (IOA) and 30 healthy fertile males (HP) as controls. Semen samples were analyzed using 16S rRNA gene sequencing (microbiome) and non-targeted metabolomics (GC-MS/LC-MS coupled). A microbe-metabolite association network was integrated at the genus level based on Spearman correlation algorithms. Results Semen microbiome analysis revealed that both microbial composition and species richness differed between IOA patients and HP controls. Non-targeted metabolomics further demonstrated characteristic metabolic dysregulation in seminal plasma of IOA patients, with a metabolic signature effectively distinguishing cases from controls (VIP > 1, FDR < 0.05). KEGG pathway enrichment analysis indicated that differentially expressed metabolites primarily involved amino acid metabolism, carbohydrate metabolism, and related signaling pathways (corrected p-value < 0.05). Construction of a Spearman correlation network between microbiota and metabolites (|r| > 0.6) identified significant interactions between core bacterial genera such as Dialister , Prevotellaceae_NK3B31_group , Lawsonella , and Blautia with seminal plasma metabolites, suggesting potential involvement of the microbiota-metabolite axis in the pathological process of IOA. Conclusion The microbial community structure and metabolic profiles in the semen of IOA patients exhibit significant disruption. Diagnostic models constructed based on combined microbial-metabolite features demonstrate potential for effectively distinguishing disease phenotypes. The core dysregulated bacterial genera, associated metabolites, and related pathways may serve as early diagnostic biomarkers and therapeutic intervention targets.