Abstract: Objective Mendelian randomization was used to analyze the causal association between 20 serum metabolites and the risk of ovarian cancer. Methods The gene variations of 20 serum metabolites were obtained from the MRC IEU database as tool variables reflecting exposure levels,while gene variations of ovarian cancer patients were used as instrumental variables reflecting outcome levels.The ovarian cancer dataset ieu-a-1120 included 66,450 European women samples(of which 25,509 were ovarian cancer),and the dataset ieu-a-1228 included 54,990 European women samples(of which 14,049 were ovarian cancer).Two-sample two-way Mendelian randomization analysis was performed on both datasets.This study used inverse variance-weighted(IVW),weighted median method,MR-Egger regression,and combined various analysis methods such as simple mode,and weighted mode.The causal effects of 20 metabolites and the risk of ovarian cancer were analyzed.Cochran′s Q test was used to perform sensitivity analysis and to verify the reliability of the results.MR Egger intercept test was used to assess the horizontal pleiotropy of tool variables,and use the leave-one-out method to assess whether there were single nucleotide polymorphisms(SNPs)in the results that might have a potential impact on the incidence of ovarian cancer.Finally,the effects of uridine on ovarian cancer cells were verified through cell proliferation and apoptosis experiments. Results The results showed a negative correlation between uridine and the occurrence of ovarian cancer,with statistically significance(ieu-a-1228:P=0.025;ieu-a-1120:P=0.017).MR-Egger regression analysis confirmed the sensitivity and robustness of the analysis results.The CCK8 assay confirmed that uridine inhibited the proliferation of ovarian cancer cells in a concentration-dependent manner,and uridine at a concentration of 10 mM significantly promoted apoptosis in SKOV3 cells(P＜0.001)and A2780 cells(P＜0.001).Flow cytometry analysis showed that after treatment for 24 hours,uridine at a concentration of 10 mM had the strongest pro-apoptotic effect on ovarian cancer cells,which was significant in SKOV3 and A2780 cells(P＜0.05 and P＜0.001,respectively). Conclusion Uridine is negatively correlated with the risk of ovarian cancer,which lays a theoretical foundation for further understanding the pathogenesis of ovarian cancer and optimizing clinical treatment strategies.

Key words: Mendelian randomization, Ovarian cancer, Uridine, Cell proliferation, Apoptosis