Identification of Therapeutic Targets for Premature Ovarian Failure Through Mendelian Randomization and Colocalization Analysis Using Human Plasma Proteomics.

Premature ovarian failure (POF) accelerates ovarian aging, leading to menstrual irregularities, reduced fertility, and decreased estrogen levels. Current hormone replacement therapy (HRT) cannot reverse the aging effects, highlighting the need for more targeted treatments. Genome-wide association studies (GWAS) and protein quantitative trait loci (pQTL) analyses can identify genetic variants and protein level changes associated with POF. Mendelian randomization (MR) evaluates the causal relationships between genetic variants and POF. pQTL analysis was conducted using plasma proteomics data from 54,219 participants and baseline cohort data of 34,557 individuals of European ancestry from the UK Biobank. GWAS data comprising 542 POF cases and 218,970 controls were obtained from the FinnGen database. MR analysis utilized inverse variance weighted (IVW), MR Egger, weighted median, and weighted mode methods. Colocalization analysis was performed using the "coloc" R package, and pathway enrichment analysis was conducted using the "clusterProfiler" package. Additionally, reverse MR analysis, molecular docking predictions, and summary data-based Mendelian randomization (SMR) analysis were performed. Finally, based on the scRNA-seq data of POF, analyses such as cell type annotation, gene set scoring, and cell-cell communication were performed. MR analysis identified significant causal relationships between specific proteins (BSG, CCL23, CTSC, FAP, IGSF21, LCN15, LILRB2, MUC16, PTN, SPINK1, TNFRSF1B, TNFRSF8, TNXB, and YJU2) and POF. Colocalization analysis indicated that key proteins (BSG, CCL23, FAP, and TNXB) share causal variants with POF traits. SMR analysis confirmed TNXB as a risk factor for POF. Finally, using the scRNA-seq data of POF, the expression of key gene sets was used to evaluate the scoring of different cell populations. Cell-cell communication analysis identified multiple communication pathways between high-scoring cell populations and other cell groups. The expression trend of key proteins was further verified by western blot assay. These findings are preliminary and require significant validation before clinical application. Combining pQTL and GWAS data, MR and colocalization analyses identified key proteins and genetic variants associated with POF, providing deeper insights into POF mechanisms and potential therapeutic targets.