A multi- tissue atlas of age-associated differential transcript usage in mouse reveals predominantly regulatory isoform switching across brain, liver, and adipose

Alternative splicing is increasingly recognized as dysregulated in the aging process across tissues and affects proteome and other regulatory changes which can be biomarkers of aging. In this study, we aimed to identify differential transcript usage (DTU) and isoforms switch accompanied with functional annotation and structural analysis in mouse brain, liver and adipose tissue during natural aging and Ercc1 knock out mice. Age associated isoform switches affected both coding and non-coding transcript features. Across tissues, 36.5-42.6% of switches altered the predicted coding sequence, 31.2-36.5% involved non-coding alternative arms, and 11-18% involved transcripts encoding an identical protein. We observed overall 3'-UTR lengthening in the brain and shortening in the adipose. A single pan-organ isoform switch event in the gene Ighm, recurs in seven separate tissue samples. BioEMU modeling confirms feasibility for a minority of switches following a seven-fold eligibility cascade; the Cyth1 exemplar shows an intrinsically disordered N-terminal extension that does not change the shared SEC7 and PH domains. Together, these results distinguish age-associated isoform changes as predominantly regulatory in mechanism, with structural consequences restricted to tissue-specific events that form a tractable target for conformational ensemble modeling.