Zhao Xunying, Wu Xueyao, He Lin, Xiao Jinyu, Xiang Rong, Sha Linna, Tang Mingshuang, Hao Yu, Qu Yang, Xiao Changfeng, Qin Chenjiarui, Hou Jiaojiao, Deng Qin, Zhu Jiangbo, Zheng Sirui, Zhou Jinyu, Yu Ting, Yang Bin, Song Xin, Han Tao, Liao Jiaqiang, Zhang Tao, Fan Mengyu, Li Jiayuan, Jiang Xia
Department of Epidemiology and Health Statistics, West China School of Public Health and West China Fourth Hospital, Sichuan University, Chengdu, Sichuan, China.
Division of Cancer Epidemiology and Genetics, National Cancer Institute, Rockville, Maryland, USA.
Scand J Med Sci Sports. 2025 Jan;35(1):e70014. doi: 10.1111/sms.70014.
Physical inactivity and sedentary behavior are associated with higher risks of age-related morbidity and mortality. However, whether they causally contribute to accelerating biological aging has not been fully elucidated. Utilizing the largest available genome-wide association study (GWAS) summary data, we implemented a comprehensive analytical framework to investigate the associations between genetically predicted moderate-to-vigorous leisure-time physical activity (MVPA), leisure screen time (LST), and four epigenetic age acceleration (EAA) measures: HannumAgeAccel, intrinsic HorvathAgeAccel, PhenoAgeAccel, and GrimAgeAccel. Shared genetic backgrounds across these traits were quantified through genetic correlation analysis. Overall and independent associations were assessed through univariable and multivariable Mendelian randomization (MR). A recently developed tissue-partitioned MR approach was further adopted to explore potential tissue-specific pathways that contribute to the observed associations. Among the four EAA measures investigated, consistent results were identified for PhenoAgeAccel and GrimAgeAccel. These two measures were negatively genetically correlated with MVPA (r = -0.18 to -0.29) and positively genetically correlated with LST (r = 0.22-0.37). Univariable MR yielded a robust effect of genetically predicted LST on GrimAgeAccel (β = 0.69, p = 1.10 × 10), while genetically predicted MVPA (β = -1.02, p = 1.50 × 10) and LST (β = 0.37, p = 1.90 × 10) showed marginal effects on PhenoAgeAccel. Multivariable MR suggested an independent association between genetically predicted LST and GrimAgeAccel after accounting for MVPA and other important confounders. Tissue-partitioned MR suggested skeletal muscle tissue associated variants to be predominantly responsible for driving the effect of LST on GrimAgeAccel. Findings support sedentary lifestyles as a modifiable risk factor in accelerating epigenetic aging, emphasizing the need for preventive strategies to reduce sedentary screen time for healthy aging.
身体活动不足和久坐行为与年龄相关的发病和死亡风险较高有关。然而，它们是否会导致生物衰老加速尚未完全阐明。利用现有的最大规模全基因组关联研究（GWAS）汇总数据，我们实施了一个综合分析框架，以研究基因预测的中度至剧烈休闲时间身体活动（MVPA）、休闲屏幕时间（LST）与四种表观遗传年龄加速（EAA）指标之间的关联：HannumAgeAccel、内在的HorvathAgeAccel、PhenoAgeAccel和GrimAgeAccel。通过遗传相关性分析量化了这些性状之间共享的遗传背景。通过单变量和多变量孟德尔随机化（MR）评估总体和独立关联。最近开发的组织分区MR方法被进一步采用，以探索导致观察到的关联的潜在组织特异性途径。在所研究的四种EAA指标中，PhenoAgeAccel和GrimAgeAccel得到了一致的结果。这两个指标与MVPA呈负遗传相关（r = -0.18至-0.29），与LST呈正遗传相关（r = 0.22 - 0.37）。单变量MR显示基因预测的LST对GrimAgeAccel有显著影响（β = 0.69，p = 1.10 × 10），而基因预测的MVPA（β = -1.02，p = 1.50 × 10）和LST（β = 0.37，p = 1.90 × 10）对PhenoAgeAccel有边际影响。多变量MR表明，在考虑MVPA和其他重要混杂因素后，基因预测的LST与GrimAgeAccel之间存在独立关联。组织分区MR表明，骨骼肌组织相关变异主要是导致LST对GrimAgeAccel产生影响的原因。研究结果支持久坐的生活方式是加速表观遗传衰老的一个可改变的风险因素，强调需要采取预防策略来减少久坐屏幕时间以实现健康衰老。
