Nature Aging

1Manual frailty index (FI) assessment in mice is a strong predictor of morbidity and mortality, and is frequently used in mechanistic and translational geroscience. However, it is labor-intensive, requires expert training, and is vulnerable to scorer variability. We previously developed a visual frailty index (vFI) that objectively predicts age and frailty using expert-defined, supervised behavioral features extracted from open-field videos. However, relying solely on human-defined features may miss subtle, latent behavioral signatures of aging. Here, we test whether unsupervised behavioral discovery using Keypoint-MoSeq (KPMS) could uncover these hidden signatures and improve the prediction of aging-related outcomes. Using a large dataset of isogenic C57BL/6J (B6J) and genetically diverse Diversity Outbred (DO) mice, we find that unsupervised features are highly predictive of chronological age, biological frailty, and the proportion of life lived. Notably, while supervised features overall outperformed unsupervised features in these tasks, combining both feature sets yielded the highest predictive accuracy across all outcomes. Despite these improvements, models trained on either feature set failed to generalize across strains, confirming that behavioral manifestations of aging are strongly population-specific. These findings demonstrate that supervised and unsupervised machine vision provide complementary information, establishing a highly sensitive, scalable, and non-invasive framework for objective and scalable geroscience in rodents.

Multimorbidity is highly prevalent in ageing populations, yet its shared molecular basis remains poorly defined, limiting the development of therapies that target multiple conditions. We systematically integrated measurements of 1,954 circulating proteins from 54,219 individuals in discovery and 35,559 in replication, focusing on ten common age-related diseases: coronary artery disease, chronic kidney disease, chronic obstructive pulmonary disease, dementia, heart failure, major depressive disorder, osteoarthritis, Parkinson's disease, stroke, and type 2 diabetes. Coronary artery disease emerged as a central condition in the multimorbidity network, sharing circulating protein signatures with seven other diseases. Through genetic causal-inference analyses, we identified 40 circulating proteins with cross-disease relevance, of which four were further supported by colocalization of genetic variant associations. Among these, complement C1r, encoded by C1R, emerged as a key link between coronary artery disease and dementia, supported by independent colocalization evidence (PP.H4 = 0.86). Phenome-wide association analyses of C1R variants suggested that this signal was not driven by widespread unrelated genetic effects, but instead may reflect a more specific contribution to coronary artery disease-dementia pathogenesis. In vitro experiments further suggested that fibroblast-derived C1R promotes endothelial inflammation and neuronal apoptosis, providing mechanistic plausibility. Together, these findings position C1R as a biologically plausible and therapeutically relevant molecular link between coronary artery disease and dementia.

Aging is a complex process manifesting at the molecular, cell, organ and organismal levels. It leads to functional decline, disease and ultimately death, but the relationship between these fundamental biomedical features remains elusive. By applying machine learning to plasma proteome data of over fifty thousand human subjects in the UK Biobank and other cohorts, we report organ-specific and conventional aging models trained on chronological age, mortality and longitudinal proteome data. We show how these tools predict organ/systems-specific disease through numerous phenotypes. We find that men are biologically older and age faster than women, that accelerated aging of organs leads to diseases in these organs, and that specific diets, lifestyles, professions and medications are associated with accelerated and decelerated aging of specific organs and systems. Altogether, our analyses reveal that age-related chronic diseases epitomize accelerated organ- and system-specific aging, modifiable through environmental factors, advocating for both universal whole-organism and personalized organ/system-specific anti-aging interventions.

Frailty is a clinically relevant phenotype with significant gaps in our understanding of its etiology. We performed a genome-wide association study of frailty in FinnGen (N=500,737) and replicated the signals in the UK Biobank (N=429,463) using polygenic risk scores (PRSs). We prioritized genes through proteomics integration (N[~]45,000; UK Biobank) and colocalization of protein quantitative trait loci. Frailty was measured using the Hospital Frailty Risk Score (HFRS). We observed 1,588 variants associated with frailty (p<5x10-8) of which 1,242 were novel, i.e., previously unreported for any trait. The associations mapped to 106 genes of which 31 were novel. PRS replication validated the signals ({beta}=0.074, p<2x10-16). Cell type enrichment analysis indicated expression in neuronal cells. Protein levels of KHK, CGREF1, MET, ATXN2, ALDH2, NECTIN2, APOC1, APOE and FOSB were associated with HFRS, whereas colocalized signals were observed within APOE and BRAP. Our results reveal novel genetic contributions and causal candidate genes for frailty.

Cellular senescence, a hallmark of ageing, drives tissue dysfunction by promoting inflammation and fuelling disease. Yet, the dynamics of senescent cell accumulation across tissues and their cell type identity remain poorly understood. Here, we introduce the first, single-cell, protein-level approach, combining multiple senescence markers for the identification and quantification of senescent cells across multiple tissues in mice and in human PBMCs. Applying this method, we reveal widespread but heterogeneous changes in senescence marker expression across cell types and tissues. The cells we identify as senescent displayed transcriptomic senescence signatures, providing a direct molecular link between protein- and mRNA-level detection of senescence. Importantly, senescence accumulation was strongly coordinated within organs but showed little correlation across them, supporting the idea of a tissue specific progression of ageing. These findings refine our understanding of the tissue-specific dynamics of senescence accumulation with age, and provide a framework for evaluating diverse therapeutic interventions.

BackgroundThe relationship between hip osteoarthritis (hip OA) and Alzheimers disease (AD) presents a critical paradox within the emerging "bone-brain axis": widespread phenotypic comorbidity sharply contradicts evolutionary theories of biological antagonism. This study integrates longitudinal and multi-omic analyses to determine whether this clinical overlap masks an underlying genetic neuroprotection. MethodsWe analyzed longitudinal phenotypic data from 261,767 UK Biobank participants using Cox proportional hazards and Fine-Gray competing risk models. To investigate the shared genetic architecture, we applied MiXeR modeling to genome-wide association study summary statistics. Causal relationships were evaluated using global and cell-type-stratified Mendelian randomization across eight distinct brain cell types. Shared genomic loci were identified via conjunctional/conditional false discovery rate and fine-mapping. Single-nucleus RNA-sequencing (snRNA-seq) data from the ROSMAP cohort validated the disease-associated transcriptional dynamics of prioritized target genes. ResultsObservational survival analyses initially suggested an increased AD risk in patients with hip OA; however, this association was fully attenuated after adjusting for a history of depression, revealing a "phenotypic illusion" driven by the pain-depression axis. Conversely, cell-type-stratified genomic analyses uncovered a profound biological antagonism: genetic liability for hip OA confers robust neuroprotection specifically localized to the neurovascular unit (NVU), primarily driven by astrocytes and pericytes. Mechanistically, this NVU fortification is orchestrated by the MAPT locus and PI3K/AKT signaling, with snRNA-seq confirming the active transcriptional remodeling of these core effectors in the AD brain. ConclusionWe demonstrate that genetic liability to hip OA confers robust neurovascular protection against AD, a profound biological antagonism that is clinically masked by depression-mediated phenotypic comorbidity. These findings propose an evolutionary trade-off model within the bone-brain axis, underscoring the urgency of active hip OA pain management to mitigate depressive symptoms and decelerate cognitive aging, while cautioning against the uncritical repurposing of anabolic inhibitors across these interconnected systems.

Earlier menopause is a risk factor for several age-related diseases, including dementia. The biological pathways linking menopause timing to later-life brain aging are not understood. Leveraging large-scale plasma proteomics in postmenopausal women from the UK Biobank (N=15,012), earlier menopause was associated with upregulation of pro-inflammatory and extracellular matrix degradation pathways, plus accelerated aging across proteomic clocks of organ and cellular aging, including brain and oligodendrocyte aging. Elevated GDF15, a canonical aging marker, was the top protein correlate of earlier menopause. We observed robust replication of menopause timing proteomic shifts in the Womens Health Initiative Long Life Study (N=1,210). In UKB, proteins associated with earlier menopause, including GDF15, exhibited concordant associations with incident dementia risk and brain atrophy, cerebral small vessel disease burden, and white matter microstructural integrity. Collectively, our findings identify proteomic signatures linking ovarian aging to brain aging, providing a framework to inform interventions to reduce dementia risk.

Mosaic loss of Y chromosome (mLOY) in blood cells is an age-related somatic mutation, but its relationship with pulmonary health remains undercharacterized. Leveraging mLOY assessment in over 12,000 men, including 5,097 from the COPDGene Study and 7,235 from six additional cohorts in Trans-Omics for Precision Medicine program, we investigated its association with respiratory outcomes and epigenetic aging. Cross-sectionally, mLOY was associated with airflow obstruction with prevalence increasing with age, particularly in men with a former smoking history. Longitudinally, mLOY associated with lung function decline. Notably, mLOY was also associated with greater CT-quantified lung emphysema and faster pace epigenetic aging. Prospectively, in participants with normal lung function at baseline, mLOY was associated with lower future lung function and faster pace of epigenetic aging. These associations remained robust after adjusting for clonal hematopoiesis and telomere length. Collectively, these findings position mLOY as a potential biomarker of respiratory aging and obstructive lung disease. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=200 SRC="FIGDIR/small/25332379v1_ufig1.gif" ALT="Figure 1"> View larger version (45K): org.highwire.dtl.DTLVardef@109babcorg.highwire.dtl.DTLVardef@1c2bc5corg.highwire.dtl.DTLVardef@1dc516forg.highwire.dtl.DTLVardef@16666f5_HPS_FORMAT_FIGEXP M_FIG C_FIG

Organ-specific aging clocks have shown promise as predictors of disease risk and aging trajectories; however, the underlying biological mechanisms they reflect remain largely unexplored. Here, we use large-scale proteomic and imaging data to investigate the relationships among organ-specific and modality-specific aging clocks and to uncover the biological processes they represent. By estimating paired protein-based and imaging-based aging clocks across 8 major organs, we demonstrate that these omics and structural profiles exhibit distinct phenotypic and genetic signatures, each potentially quantifying different stages and playing complementary roles within a unified biological aging process. Furthermore, context-specific aging clocks from multiple organs often converge and jointly capture established biological and disease pathways. For example, 65.7% of the KEGG Alzheimers disease pathway is enriched by at least one of 11 protein- and imaging-based aging clocks, with each clock representing different components of the pathway. These results underscore the importance of a pan-organ multi-modal perspective for quantifying the mechanisms underlying age-related diseases. Additionally, we identify modality-specific links between aging clocks and complex diseases and lifestyle factors. In summary, we uncover intricate relationships among molecular and structural aging clocks across human organs, providing novel insights into their context-specific roles in capturing consequences of aging biology and their implications for disease risk.

Biological aging is a multifactorial process involving complex interactions of cellular and biochemical processes that is reflected in omic profiles. Using common clinical laboratory measures in ~30,000 individuals from the MGB-Biobank, we developed a robust, predictive biological aging phenotype, EMRAge, that balances clinical biomarkers with overall mortality risk and can be broadly recapitulated across EMRs. We then applied elastic-net regression to model EMRAge with DNA-methylation (DNAm) and multiple omics, generating DNAmEMRAge and OMICmAge, respectively. Both biomarkers demonstrated strong associations with chronic diseases and mortality that outperform current biomarkers across our discovery (MGB-ABC, n=3,451) and validation (TruDiagnostic, n=12,666) cohorts. Through the use of epigenetic biomarker proxies, OMICmAge has the unique advantage of expanding the predictive search space to include epigenomic, proteomic, metabolomic, and clinical data while distilling this in a measure with DNAm alone, providing opportunities to identify clinically-relevant interconnections central to the aging process.

BackgroundThe global aging population underscores a critical need to tackle accompanying health and economic challenges, at all levels of society. This All-of-Society approach emphasizes the involvement of various stakeholders--governments, NGOs, researcher centers, private companies, local communities, and opinion leaders--to collectively promote healthy aging. However, how stakeholders enable healthy longevity remains unclear. ObjectiveThis study examines how global stakeholders (governments, NGOs, researcher centers, private companies, local communities, and opinion leaders) create value towards healthy longevity. We identify the healthy longevity dimension of stakeholders value propositions and examine alignment between their propositions as an indicator of shared goals. MethodsFollowing the All-of-Society approach, we analyzed the healthy longevity aspects of value propositions among the six classes of stakeholders (N=128). We (1) employed semantic topic modeling to identify the primary value proposition topics as related to healthy longevity and (2) computed proposition alignment using similarity networks. ResultsOur analysis revealed varying degrees of alignment between stakeholders healthy longevity propositions, with the lowest alignment observed for local communities and researcher centers. ConclusionsFindings underscore a key need to strengthen synergies between academic and community-based initiatives to promote translational science and highlight opportunities for strategic partnerships in the evolving healthy longevity field. What is already known on this topicThe National Academy of Medicines All-of-Society approach advocates for multi-stakeholder engagement towards healthy longevity, but specific stakeholder contributions, and their alignment toward shared goals, are poorly understood. What this study addsTo our knowledge, this study is the first to provide empirical evidence into the value propositions of healthy longevity stakeholders on a societal scale. It highlights key areas where multi-stakeholder collaboration can be strengthened--particularly between academic and local community initiatives--and proposes five strategies to strengthen collaboration. How this study might affect research, practice, or policyPrioritizing (1) community-based participatory research, (2) translating healthy aging-related research findings into accessible resources, (3) prioritizing equity in intervention delivery, (4) establishing community advisory boards, and (5) developing knowledge translation and training programs, could potentially better align academic and community efforts towards more aligned, equitable and effective healthy longevity initiatives.

Deep learning foundation models excel at disease prediction from medical images, yet their potential to bridge tissue morphology with the genetic architecture of disease remains underexplored. Here, we present REECAP (Representation learning for Eye Embedding Contrastive Age Phenotypes), a framework that fine-tunes the RETFound retinal foundation model using a contrastive objective guided by chronological age. Applied to 87,478 fundus images from 52,742 UK Biobank participants, REECAP aligns image representations along the aging axis, yielding multivariate ageing phenotypes for genome-wide association studies (GWAS). GWAS of REECAP embeddings identifies 178 loci, including 27 that colocalize with risk loci of age-related eye diseases, 14 of which remained undetected by conventional disease-label GWAS. By enabling conditional image synthesis, REECAP further links genetic variation to interpretable anatomical changes. Benchmarking against alternative embedding models, we show that REECAP enhances both locus discovery and disease relevance of genetic associations, suggesting that aging-informed tissue embeddings represent a powerful intermediate phenotype to discover and interpret disease loci.

Little is known about tissue specific changes that occur with aging in humans. Using the description of 33 million histological samples we extract thousands of age- and mortality-associated features from text narratives that we call The Human Pathome (pathoage.com). Notably, we can broadly determine when pathological aging starts, indicating a sexual dimorphism with females aging earlier but slower and males aging later but faster. Using machine learning, we employ unsupervised topic-modelling to identify terms and themes that predict age and mortality. As a proof of principle, we cross reference these terms in PubMed to identify nintedanib as a potential aging intervention and show that nintedanib reduces markers of cellular senescence, reduces pro-fibrotic gene pathways in senescent cells and extends the lifespan of fruit flies. Our findings pave the way for expanded exploitation of population datasets towards discovery of novel aging interventions.

Epigenetic clocks can predict pathological aging associated with Alzheimers disease (AD) risk, albeit findings are mixed regarding if clocks are predictive in blood and in non-European populations. We constructed epigenetic clocks from blood methylation data in 704 older Hispanic adults and tested the association with a clinical diagnosis of AD and plasma biomarker levels. Biological age and age acceleration, the rate of biological aging, were significantly associated with sex, clinical diagnosis, and levels of eight plasma biomarkers, including P-Tau217 levels. Additionally, biomarker associations trended more significant among APOE-{varepsilon}4 non-carriers. We also identified that methylation levels in CD4 and CD8 T-cell types are associated with biological aging and showed slightly stronger associations in men. We demonstrate that biological aging, in blood, in a Hispanic cohort of both demented and non-demented individuals, can stratify AD risk, predicting plasma biomarker levels even in preclinical disease.

Turquoise killifish (Nothobranchius furzeri) are naturally short-lived vertebrates, recapitulating several aspects of human aging, including protein aggregation, telomere shortening, cellular senescence, and declined antibody diversity. The mechanistic causes of systemic aging in killifish are still poorly understood. Here we ask whether killifish undergo significant age-dependent changes in the main hematopoietic organ, which together could contribute to systemic aging. To characterize immune aging in killifish, we employed single-cell RNA sequencing, proteomics, cytometry and a functional in vitro assay. Our data indicate how old killifish display increased inflammatory markers, and while immune cells from adult killifish display increased markers of proliferation and replication-independent DNA repair in progenitor-like cell clusters, progenitors from old killifish display extensive markers of DNA double-strand breaks. In less than 10 weeks, killifish undergo several dramatic spontaneous aging-related changes in the immune niche, which could be functionally linked with its extensive systemic aging and serve as targets for anti-aging interventions.

Organ-specific proteomic clocks are promising tools for quantifying heterogeneity in biological aging, but their longitudinal behavior remains largely unexplored. Here, we analyzed paired plasma proteomic profiles with 10-year follow-up in middle-aged adults (n= 1,250) to evaluate their longitudinal properties. Cross-sectional associations of protein concentrations with age mirrored average longitudinal trajectories, validating the common cross-sectional training of clocks. Organ-specific age acceleration was moderately stable over the decade, and aging across organs progressed in parallel, with the immune and adipose systems acting as central hubs and early cardiorespiratory aging predicting downstream metabolic aging. Critically, longitudinal changes in predicted age tracked subclinical risk factor alterations. In women, the menopausal transition dominated the aging landscape and was associated with multi-organ age acceleration. Medication initiation altered clocks through specific drug-targeted proteins (such as renin and APOB) rather than generalized organ aging. Together, these findings position organ-specific proteomic clocks as interpretable, dynamic indicators of aging and organ health.

Renowned for their regenerative abilities, axolotls also exhibit exceptional longevity, resistance to age-related diseases and apparent lack of physiological declines through lifespan, and have thus been considered organisms of negligible senescence. Whether axolotls display epigenetic hallmarks of ageing remains unknown. Here, we probe the axolotl DNA methylome throughout lifespan and present its first epigenetic clocks. Both at tissue-specific or pan-tissue levels, the clocks are biphasic, capable of predicting age during early life but not for the rest of its lifespan. We show that axolotls exhibit evolutionarily conserved features of epigenetic ageing during early life, yet their methylome is remarkably stable across lifespan, including at Polycomb Repressive Complex 2 (PRC2) target sites, suggesting that this species deviates from known patterns of epigenetic ageing. This study provides molecular insights into negligible senescence and furthers our understanding of ageing dynamics in animals capable of extreme regeneration.

Menopause is a hallmark process in biological aging that has been associated with later life neurodegenerative risk. We leveraged proteomics data from multiple cohorts (N>3,000) to identify biological changes underlying menopause and its links to brain aging. In N=80 rigorously-phenotyped pre-, peri-, and postmenopausal women with serum NULISAseq proteomics, spontaneous menopause was characterized by dysregulation in inflammatory, synaptic, metabolic, and Alzheimers disease (AD) biologic processes, which tracked with hormones and not age. Pro-inflammatory protein upregulation was especially pronounced in women with vasomotor symptoms. In two cohorts of older women (N=94; N=100), menopause-related proteomic elevations associated with poorer cognitive outcomes and plasma AD biomarkers. Finally, validation analyses in age-matched pre- and postmenopausal women with plasma Olink proteomics (N=2,814) replicated the observed proteomic shifts and revealed menopause-related upregulation of additional inflammatory and catabolic processes. The molecular signatures of menopause may inform biomarkers or therapeutic targets for brain health in women.

Aging is a complex process characterized by a progressive decline in physiological functions driven by both biological and environmental factors, with notable differences between sexes. Immune function is strongly influenced by biological sex, affecting both innate and adaptive immune responses, including macrophage behavior. In this study, we investigated the effects of age and sex on the immune cell composition within the peritoneal cavity niche and identified macrophages as the most affected cell type. Macrophages, as central components of the innate immune system, play critical roles in maintaining tissue homeostasis and responding to infections. Here, we find that aging induces sex-specific remodeling of murine peritoneal macrophage transcriptomic and epigenomic landscapes. Consistently, peritoneal macrophages undergo sex-specific functional remodeling with aging (i.e. female-specific phagocytic decline and metabolic rewiring). Modulation of gonadal hormone signaling showed that changes in circulating estrogen levels likely contribute to aspects of female-specific macrophage age-related changes. Importantly, multi-omic analysis identified candidate transcription factors whose sex-specific age-regulated expression may drive aspects of sex-specific omic remodeling with aging. Specifically, Irf2 downregulation in female macrophages recapitulates distinct transcriptomic and metabolic aspects of macrophage female aging phenotypes. These findings suggest that female-specific age-related functional remodeling arises through hormone-dependent and -independent mechanisms in peritoneal macrophages.

Oct4, Sox2, and Klf4 (OSK) Yamanaka factors induce pluripotency and reverse age-related epigenetic changes, yet the mechanisms by which they promote rejuvenation remain poorly explored. Oxidative stress contributes to CNS aging and retinal pigmented epithelium (RPE) degeneration in age-related macular degeneration. We find that OSK expression in RPE restores retinal structure and visual function in aged mice and promotes oxidative resilience through a non-canonical, Tet2-independent pathway. Integrative functional genomics identifies GSTA4, a detoxifying enzyme that clears the lipid peroxidation byproduct 4-HNE, as a necessary and sufficient OSK effector. Dynamic GSTA4 regulation by OSK recapitulates a stem cell derived stress resilience program. GSTA4 overexpression alone enhances mitochondrial resilience, rejuvenates the aged RPE transcriptome, and reverses visual decline. GSTA4 is consistently upregulated across diverse lifespan-extending interventions suggesting a broader pro-longevity role. These findings uncover a previously unrecognized protective axis driven by Yamanaka factors that circumvents reprogramming, providing therapeutic insights for age-related diseases. HIGHLIGHTSO_LIOSK-GSTA4 provides a dynamic, Tet2-independent stress-resilience axis. C_LIO_LIFunctional genomics pinpoints GSTA4 as a direct downstream effector activated by OSK. C_LIO_LIRPE aging involves progressive accumulation of 4-HNE that can be detoxified by GSTA4. C_LIO_LIEnhancing GSTA4 rejuvenates RPE cells, restores vision and is associated with lifespan-extending interventions. C_LI