白介素信号通路与痛风性关节炎风险的因果关系——基于孟德尔随机化的多因子分析

Causal Relationship between Interleukin Signaling Pathway and Risk of Gouty Arthritis—Multivariate Analysis Based on Mendelian Randomization

摘要: 目的：本研究旨在利用两样本孟德尔随机化方法，系统地评估循环炎症蛋白与痛风性关节炎之间的因果关系，以阐明炎症通路在痛风发病机制中的潜在因果作用，并为识别新的治疗靶点提供遗传学证据。方法：本研究采用两样本孟德尔随机化设计。暴露(132种循环炎症蛋白)与结局(痛风性关节炎)的遗传工具变量均来源于公开的全基因组关联研究汇总数据。痛风数据包含3576例病例和147,221例对照。我们选取在全基因组显著性水平上相关的单核苷酸多态性作为工具变量，并进行了连锁不平衡聚类。主要采用逆方差加权法进行因果估计，并通过MR-Egger、加权中位数等多种方法进行补充。通过MR-Egger截距检验、Cochran’s Q统计量、留一法分析等进行敏感性分析。此外，还进行了多变量孟德尔随机化分析以评估血清尿酸与炎症通路的独立因果效应。结果：正向MR分析发现，基因预测的较高水平的成纤维细胞生长因子-21、基质金属蛋白酶-1和粒细胞集落刺激因子是痛风的风险因素，而较高水平的γ-干扰素是保护因素。反向MR分析表明，痛风疾病状态会因果性地提高C-X-C基序趋化因子配体1和肿瘤坏死因子-α的水平，同时降低白介素-1受体拮抗剂的水平。多变量MR分析显示，在相互校正后，血清尿酸与白介素相关炎症通路对痛风风险均具有显著的独立直接因果效应。敏感性分析未发现明显的水平多效性，证实了结果的稳健性。结论：本研究从遗传学角度证实了多种循环炎症蛋白与痛风性关节炎之间存在因果关系，揭示了炎症通路独立于血清尿酸在痛风发病中的直接驱动作用。

Abstract: Objective: To systematically evaluate the causal relationship between circulating inflammatory proteins and gouty arthritis by two-sample Mendelian randomization, to elucidate the potential causal role of inflammatory pathways in the pathogenesis of gout, and to provide genetic evidence for the identification of new therapeutic targets. Methods: This study employed a two-sample Mendelian randomization design. Genetic instrumental variables for both exposure (132 circulating inflammatory proteins) and outcome (gouty arthritis) were derived from publicly available genome-wide association study meta-data. The gout dataset comprised 3576 cases and 147,221 controls. We selected single-nucleotide polymorphisms (SNPs) showing genome-wide significance as instrumental variables and performed linkage disequilibrium clustering. Causal estimation was primarily conducted using inverse variance weighting, supplemented by methods including MR-Egger and weighted median. Sensitivity analyses were performed via MR-Egger intercept testing, Cochran’s Q statistic, and one-leave-one-out analysis. Additionally, multivariate Mendelian randomization was conducted to evaluate the independent causal effects of serum uric acid on inflammatory pathways. Results: Forward multiple regression (MR) analysis revealed that elevated gene-predicted levels of fibroblast growth factor-21, matrix metalloproteinase-1, and granulocyte colony-stimulating factor were risk factors for gout, while higher γ-interferon levels served as a protective factor. Reverse MR analysis demonstrated that gout pathologically elevates the levels of C-X-C motif chemokine ligand 1 and tumor necrosis factor-α, while simultaneously reducing interleukin-1 receptor antagonist levels. Multivariate MR analysis showed that, after mutual adjustment, serum uric acid and interleukin-associated inflammatory pathways exhibited significant independent direct causal effects on gout risk. Sensitivity analysis confirmed the robustness of the results by showing no significant level of heterogeneity. Conclusion: This study confirmed the causal relationship between the multiple circulating inflammatory proteins and gouty arthritis from the genetic angle, and revealed that the inflammatory pathway is independent of serum uric acid in the direct driving of gout.