Autophagy-related biomarkers in non-obstructive azoospermia: insights from transcriptomics and single-cell sequencing.

Abstract

Objective: Autophagy, by modulating cellular degradation and recycling processes, affects sperm cell survival and differentiation and may be linked to the pathophysiology of non-obstructive azoospermia (NOA). However, the role of autophagy-related genes (ARGs) in NOA has not yet been fully explored.

Methods: Transcriptomic datasets for NOA from public databases were used. By combining differential expression analysis between NOA and obstructive azoospermia (OA), weighted gene co-expression network analysis (WGCNA), and machine learning. algorithms, biomarkers were identified. Their performance was evaluated with ROC curves. Enrichment and immune analyses explored mechanisms, and validation was done via datasets (GSE9210 and GSE145467) and RT-qPCR. Single-cell level mechanisms were also studied.

Results: A total of 321 differentially expressed genes (DEGs) were screened, related to germ cell development and nucleocytoplasmic transport. Four biomarkers (ATP6V1E2, UBQLN2, FYCO1, and ITPR1) from machine learning had good diagnostic performance. FYCO1 positively correlated with T follicular helper cells and negatively with regulatory T cells. Single-gene GSEA showed it was enriched in gametogenesis and cell adhesion molecules. Single-cell analysis revealed ATP6V1E2 had the highest expression in testicular gamete cells, while UBQLIN2, FYCO1, and ITPR1 in smooth muscle cells RT-qPCR results matched the dataset trends.

Conclusion: This study screened four ARGs (ATP6V1E2, UBQLN2, FYCO1, and ITPR1) as biomarkers for NOA, which could serve as potential diagnostic tools and therapeutic targets, potentially providing new treatment strategies.