Abstract
Abstract
Characterizing temporal dynamics during aging is crucial to understanding the course of aging-related events. The transcriptome is extensively remodeled with age, but the temporal trajectories of these changes, particularly through midlife, remain poorly defined in many tissues. Here, we address this gap in the male mouse colon. Using a dataset of 83 RNA-Seq samples from five time points during adult mouse lifespan, we show that colon aging is characterized by multiple gene-expression trajectories unfolding in parallel, with a major shift occurring during a mid- to late-life transition marked by a collapse of the unfolded protein response (UPR). Moreover, we report widespread non-monotonic trajectories, including genes that peak or dip in midlife and would be missed by comparisons limited to young versus old ages. We further describe age-dependent dynamics of transposable element-containing transcripts and a nonlinear course of transcriptomic entropy across the lifespan. Importantly, we identify a small set of genes that track chronological age and use only three of them to construct a compact transcriptomic clock, which we validate using an independent RNA-seq dataset and qPCR data. Together, our results indicate that nonlinear patterns dominate the aging colon transcriptome and underscore the value of high temporal resolution for identifying key transitions during aging.