基于双向两样本孟德尔随机化方法分析皮肤 菌群与乳腺炎的关联

Bidirectional Two-Sample Mendelian Randomization Analysis of the Association between Skin Microbiota and Mastitis

摘要: 目的：通过孟德尔随机化(Mendelian randomization, MR)方法从遗传学角度分析皮肤菌群与乳腺炎之间的因果关联。方法：使用双向两样本MR方法，利用皮肤菌群和FinnGen数据库中乳腺炎的全基因组关联研究(Genome-wide association studies, GWAS)的汇总数据探究皮肤菌群与乳腺炎的关系。运用5种MR方法包括逆方差加权法(Inverse variance weighted, IVW)、MR-Egger法、加权中位数法(Weighted median, WME)、简单模式法(Simple mode, SM)以及加权模式法(Weighted mode, WM)从遗传学角度评估皮肤菌群与乳腺炎的潜在因果关联，其中IVW为主方法。此外，进行了敏感性分析检验异质性和水平多效性以增强结果稳健性。结果：正向MR分析结果显示，遗传预测的人葡萄球菌和不动杆菌属与乳腺炎风险增加相关，栖水菌属和微球菌属与乳腺炎风险降低相关，反向MR分析结果显示遗传预测的乳腺炎与β-变形菌纲及奈瑟菌属水平降低相关，以上结果的敏感性分析均未显示异质性和水平多效性。结论：本研究提供了遗传学证据支持人葡萄球菌和不动杆菌属为乳腺炎的风险因素，微球菌属和不动杆菌属为乳腺炎的保护因素，乳腺炎对β-变形菌纲及奈瑟菌属具有抑制作用。鉴于数据来源存在异质性(如皮肤采样部位及乳腺炎亚型存在差异)，上述发现仅具探索性意义，后续仍需开展大规模全基因组关联研究及功能实验加以验证。

Abstract: Objective: To investigate the potential causal relationship between skin microbiota and mastitis from a genetic perspective using Mendelian randomization (MR). Methods: A bidirectional two-sample MR approach was employed to explore the causal relationships between skin microbiota and mastitis. Summary-level data from genome-wide association studies (GWAS) of skin microbiota and mastitis from the FinnGen database were utilized. Five MR methods were employed to evaluate the potential causal associations from a genetic standpoint: Inverse variance weighted (IVW), MR-Egger, Weighted median (WME), Simple mode (SM), and Weighted mode (WM), with IVW as the primary method. Sensitivity analyses were performed to assess heterogeneity and pleiotropy, thereby strengthening the robustness of the findings. Results: Forward MR analysis indicated that genetically predicted Staphylococcus hominis and Acinetobacter were associated with increased mastitis risk, while Enhydrobacter and Micrococcus were associated with decreased mastitis risk. Reverse MR analysis suggested that genetically predicted mastitis was associated with reduced levels of β-proteobacteria and Neisseriaceae. Sensitivity analyses indicated no significant heterogeneity or horizontal pleiotropy for the aforementioned results. Conclusion: This study provides genetic evidence supporting that Staphylococcus hominis and Acinetobacter may be risk factors for mastitis, while Enhydrobacter and Micrococcus may be protective factors, and that mastitis may exert an inhibitory effect on β-proteobacteria and Neisseriaceae. These findings should be interpreted as exploratory, given the heterogeneity in data sources (e.g., variation in skin sites and mastitis subtypes). Further validation through larger-scale GWAS and functional experiments is warranted.