A rat epigenetic clock recapitulates phenotypic aging and co-localizes with heterochromatin-associated histone modifications

Aging has been shown to be a strong driver of DNA methylation changes, leading to the development of robust biomarkers in humans and more recently, in mice. This study aimed to generate a novel epigenetic clock in rats—a model with unique physical, physiological, and biochemical advantages for studying mammalian aging. Additionally, we incorporated behavioral data, unsupervised machine learning, and network analysis to identify epigenetic signals that not only track with age, but also relate to phenotypic aging and reflect higher-order molecular aging changes. We used DNAm data from reduced representation bisulfite sequencing (RRBS) to train an epigenetic age (DNAmAge) measure in Fischer 344 CDF (F344) rats. In an independent sample of n=32 F344 rats, we found that this measure correlated with age at (r=0.93), and related to physical functioning (5.9e-3), after adjusting for age and differential cell counts. DNAmAge was also found to correlate with age in C57BL/6 mice (r=0.79), and was decreased in response to caloric restriction (CR), such that the longer the animal was on a CR diet, the greater the decrease in DNAm. We also observed resetting of DNAm when kidney and lung fibroblasts when converted to induced pluripotent stem cells (iPSCs). Using weighted gene correlation network analysis (WGCNA) we identified two modules that appeared to drive our DNAmAge measure. These two modules contained CpGs in intergenic regions that showed substantial overlap with histone marks H3K9me3, H3K27me3, and E2F1 transcriptional factor binding. In moving forward, our ability to unravel the complex signals linking DNA methylation changes to functional aging would require experimental studies in model systems in which longitudinal epigenetic changes can be related to other molecular and physiological hallmarks of aging.研究表明，衰老可显著驱动DNA甲基化（DNA methylation）改变，进而在人类及近期研究中的小鼠体内催生可靠的生物标志物。本研究旨在构建一种全新的大鼠表观遗传时钟（epigenetic clock）——大鼠作为研究哺乳动物衰老的模型，具备独特的物理、生理与生化优势。此外，本研究整合了行为学数据、无监督机器学习与网络分析手段，以识别不仅随年龄变化，还与表型衰老相关并反映高阶分子衰老进程的表观遗传信号。本研究利用简化代表性亚硫酸氢盐测序（reduced representation bisulfite sequencing, RRBS）获得的DNA甲基化数据，在费希尔344 CDF（F344）大鼠中训练得到表观遗传年龄（DNAmAge）测算模型。在n=32只F344大鼠的独立样本中，经年龄与细胞差异计数校正后，该测算模型与年龄的相关系数达0.93，同时与躯体功能存在显著关联（P=5.9×10^-3）。研究同时发现，DNA甲基化年龄（DNAmAge）在C57BL/6小鼠中也与年龄呈显著相关（相关系数r=0.79），且可因热量限制（caloric restriction, CR）干预而降低：动物接受CR饮食的时间越长，DNA甲基化年龄的降幅便越大。本研究还观察到，当肾脏与肺成纤维细胞被重编程为诱导多能干细胞（induced pluripotent stem cells, iPSCs）时，DNA甲基化模式会发生重置。本研究采用加权基因共表达网络分析（weighted gene correlation network analysis, WGCNA），识别出两个可驱动DNA甲基化年龄测算模型的共表达模块。这两个模块包含位于基因间区的CpG位点，这些位点与组蛋白修饰标记H3K9me3、H3K27me3以及转录因子E2F1的结合区域存在显著重叠。未来，若要阐明连接DNA甲基化改变与功能性衰老的复杂信号通路，我们需要在模型系统中开展实验研究，在该系统中纵向表观遗传变化可与衰老的其他分子与生理标志性特征建立关联。