Yuan Shuai, Khodursky Samuel, Geng Jiawei, Sharma Pranav, Spin Joshua M, Tsao Philip, Levin Michael G, Damrauer Scott M
Department of Surgery, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA.
Corporal Michael J. Crescenz VA Medical Center, Philadelphia, PA, USA.
medRxiv. 2025 Mar 1:2025.02.27.25322973. doi: 10.1101/2025.02.27.25322973.
Smoking is a well-established risk factor for abdominal aortic aneurysm (AAA). However, the molecular pathways underlying this relationship remain poorly understood. This study aimed to identify circulating protein mediators that may explain the association between smoking and AAA.
We conducted a network Mendelian randomization (MR) study utilizing summary-level data from the largest available genome-wide association studies. Our primary smoking exposure was the lifetime smoking index, with smoking initiation and cigarettes per day included as supplementary traits. The AAA dataset comprised 39,221 cases and 1,086,107 controls. Protein data were sourced from two large cohorts: UKB-PPP, where proteins were measured using the Olink platform in 54,219 individuals, and deCODE, where proteins were measured using the SomaScan platform in 35,559 individuals. Two-sample MR was employed to estimate the association between smoking and AAA (β) and between smoking and circulating protein levels (β). Summary data-based MR was then used to assess the association between smoking-related proteins and AAA risk (β). Mediation pathways were identified based on the directionality of effect estimates, and the corresponding mediation effects were quantified.
Genetically predicted smoking traits were consistently associated with an increased risk of AAA. The lifetime smoking index was associated with the levels of 543 out of 5,764 unique circulating proteins, with 470 of these associations replicated in supplementary analyses using additional smoking traits and protein sources. Among the smoking-related proteins, genetically predicted levels of 22 were associated with AAA risk. Eight mediation pathways were identified accounting for 42.7% of the total smoking-AAA association and with mediation effects >4% for ADAMTS15, IL1RN, MMP12, PGF, PCSK9, and UXS1.
This study identified numerous circulating proteins potentially causally linked to smoking, and eight of these proteins were found to mediate the association between smoking and AAA risk.
吸烟是腹主动脉瘤(AAA)公认的危险因素。然而,这种关系背后的分子途径仍知之甚少。本研究旨在确定可能解释吸烟与AAA之间关联的循环蛋白介质。
我们利用来自现有最大规模全基因组关联研究的汇总数据进行了一项网络孟德尔随机化(MR)研究。我们主要的吸烟暴露指标是终生吸烟指数,将开始吸烟年龄和每日吸烟量作为补充特征。AAA数据集包括39221例病例和1086107例对照。蛋白质数据来自两个大型队列:UKB-PPP,其中在54219名个体中使用Olink平台测量蛋白质;以及deCODE,其中在35559名个体中使用SomaScan平台测量蛋白质。采用两样本MR来估计吸烟与AAA之间的关联(β)以及吸烟与循环蛋白水平之间的关联(β)。然后使用基于汇总数据的MR来评估吸烟相关蛋白与AAA风险之间的关联(β)。根据效应估计的方向性确定中介途径,并对相应的中介效应进行量化。
遗传预测的吸烟特征与AAA风险增加始终相关。终生吸烟指数与5764种独特循环蛋白中的543种水平相关,其中470种关联在使用其他吸烟特征和蛋白质来源的补充分析中得到重复。在与吸烟相关的蛋白中,遗传预测的22种蛋白水平与AAA风险相关。确定了八条中介途径,占吸烟与AAA总关联的42.7%,其中ADAMTS15、IL1RN、MMP12、PGF、PCSK9和UXS1的中介效应>4%。
本研究确定了许多可能与吸烟存在因果关系的循环蛋白,其中八种蛋白被发现介导了吸烟与AAA风险之间的关联。