A comprehensive study of four mammals shows that aging leaves a common molecular fingerprint, opening new ways to measure biological decline, compare interventions, and uncover pathways that may shape a healthier lifespan.
Research: Universal transcriptomic signatures of mammalian aging and mortality. Image credit: Igor Kyrlytsya / Shutterstock
In a recent study published in the journal naturean international research team has identified universal transcriptomic signatures of aging and mortality across mammalian species and developed a molecular clock that can predict lifespan, associations with chronic diseases and human outcomes, and transcriptomic markers of biological aging.
Molecular aging across species
As the population ages, age-related diseases such as dementia, cardiovascular pathology, and metabolic disorders are recognized as a significant health burden. Researchers have discovered that gene activity changes as organisms age, but many existing biological aging markers focus only on specific tissues or species.
Previous molecular clocks established using deoxyribonucleic acid (DNA) methylation have also been difficult to interpret from a biological perspective. By investigating how aging-related gene activity varies across organs and species, scientists may be able to develop treatments that slow disease, extend lifespan, and promote healthy aging.
transcriptome clock
Researchers analyzed more than 11,000 transcriptomes from humans, mice, rats, and cynomolgus monkeys and identified common molecular patterns associated with aging and mortality.
This study combined publicly available datasets with newly generated ribonucleic acid sequence (RNA-seq) data from genetically diverse UM-HET3 mice exposed to 20 pharmacological intervention trial program treatments including rapamycin, canagliflozin, captopril, 17α-estradiol, rapamycin and acarbose.
Other lifespan-altering models, such as calorie restriction and high-fat diets, were also incorporated through more broadly aggregated datasets.
Rather than using tissue type per se as the sole survival predictor, the researchers used the Gompertz survival model to estimate expected mortality and lifespan from survival data that accounted for cohort, sex, site, strain, and intervention. We also used machine learning techniques (such as elastic nets and Bayesian ridge regression) to generate transcriptome clocks to estimate chronological age, normalized age, transcriptome age, and mortality risk.
Finally, they conducted one-tissue-excluding validation tests and one-excluding-dataset validation tests to ensure that the model was accurate across different organ types and datasets.
We analyzed single-cell RNA-seq (scRNA-seq) and single-nuclear RNA-seq (snRNA-seq) datasets to determine whether senescence signals are consistently present in specific cell types. To better understand how other biological factors influence aging-related molecular changes, researchers examined endogenous and exogenous stimuli, including inflammatory stress, caloric restriction, and a Klotho knockout mouse model.
Aging markers and death clock
This study found similar patterns of aging-related gene expression in mice, rats, macaques, and humans by identifying transcriptomic markers of aging that are highly conserved across mammals.
Genes involved in inflammation, immune activation, and cellular stress pathways showed increased expression with age, whereas genes involved in mitochondrial energy production, wound healing, and extracellular matrix pathways tended to decrease.
Researchers have created a transcriptomic aging and mortality clock that accurately estimates chronological age, transcriptomic age, and expected mortality risk across tissues and species. Transcriptomic aging clocks accurately predicted chronological age and were validated in multiple ways, allowing evaluation of both positive and negative effects on the aging process.
Lifespan-extending treatments such as caloric restriction and rapamycin reduced transcriptomic age, whereas progeroid conditions, high-fat diets, and inflammatory stress accelerated molecular aging.
The death clock outperformed traditional chronological age measurements because it captured the deterioration of molecules associated with mortality risk, rather than simply the passage of time.
Researchers also demonstrated that aging affects cellular function in many tissues. Using single-cell analysis, we demonstrated age-related changes in molecular pathways in immune cells, endothelial cells, hepatocytes (hepatocytes), stem cells, and muscle-related cells.
Inflammation and mitochondrial aging pathways
Inflammation has been found to be an important factor contributing to age-related molecular changes. Researchers found that key signaling pathways such as interferon, tumor necrosis factor, interleukin, and p53 signaling become increasingly active with age and are associated with increased risk of death.
Furthermore, several cellular processes related to oxidative phosphorylation, mitochondrial protein production, lipid metabolism, and cellular respiration showed decreased activity with age.
The researchers also identified a modular network containing genes that contribute to the aging process. Some of these modules were dominated by immune response and inflammatory genes. The remainder primarily reflected genes regulating chromatin, mitochondrial activity, extracellular matrix organization, or metabolism.
Experiments on Klotho knockout mice also showed a relationship between metabolism and aging. Mice showed accelerated molecular aging, particularly in kidney and muscle tissue.
The expression of genes involved in mitochondrial respiration and energy metabolism was suppressed, while aging-related genes such as cyclin-dependent kinase inhibitor 1A were significantly upregulated.
Interestingly, inflammation-based pathways are not the main factor in this model, suggesting that different biological systems may govern molecular aging in different contexts.
Reversible aging signals and human outcomes
This study also demonstrated that the hallmarks of molecular aging can be partially reversed. Rejuvenation-related interventions, such as cellular reprogramming, metachronous parabiosis, and early embryonic development, reduced aging-associated transcriptomic patterns.
The authors also linked several conserved biomarkers to human outcomes. At UK Biobank, protein levels of genes such as CDKN1A, LGALS3, glycoprotein, and non-metastatic melanoma protein B”>GPNMB were associated with mortality and multimorbidity, supporting the relevance of these transcriptomic signatures across animal models.
Transcriptomic clock for healthy aging
The results of this study demonstrate that aging and mortality share universal transcriptomic features across mammalian species, tissues, and cell types. Biological aging is closely associated with inflammation, mitochondrial dysfunction, metabolic disorders, wound healing and decreased extracellular matrix activity.
A newly developed transcriptome clock precisely measures transcriptome age and predicted mortality-related molecular changes, capturing the effects of interventions that accelerate or slow aging.
In the future, these tools may support the study of early molecular markers of age-related functional decline, before disease symptoms appear. This study also shows that biological aging is determined by multiple interconnected biological pathways.
Understanding these pathways could enable researchers to develop treatments that extend healthy lifespans and reduce the global burden of age-related chronic diseases. Further research is needed to determine how to safely target these molecular pathways in humans.
Click here to download your PDF copy.
Reference magazines:
- Tishkovsky, A., Khordina, D., Davitadze, M., Moliere, A., Moldakozayev, A., Tong, Y., Kasahara, T., Glubokov, D., Eames, A., Katz, L. M., Vladimirova, A., Yin, K., Liu, H., Zhang, B., Kasak, U., Kasak, U., Monova. M., Van Raamsdonk, J.M., Harrison, D.E., Strong, R., Abe, T., Dmitriev, S.E., and Gladyshev, V.N. (2026). Universal transcriptomic signatures of mammalian aging and mortality. Nature. Doi: 10.1038/s41586-026-10542-3, https://www.nature.com/articles/s41586-026-10542-3
