Delayed molecular aging, preservation of energy metabolism and enhanced exercise response in exercise-trained human muscle

Delayed molecular aging, preservation of energy metabolism and enhanced exercise response in exercise-trained human muscle | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Biological Sciences - Article Delayed molecular aging, preservation of energy metabolism and enhanced exercise response in exercise-trained human muscle Georges Janssens, Marit Kotte, Lotte Grevendonk, Angelique Scantlebery, and 12 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6074097/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Exercise is fundamental to healthy aging, yet the degree to which it mitigates age-related molecular changes and how varying physical fitness levels influence the molecular response to exercise with age remain unclear. To address this, we performed transcriptomics, lipidomics, and metabolomics on skeletal muscle of young and older adults with differing physical function, both before and after an acute bout of sub-maximal exercise. At baseline, older adults exhibited reduced expression of genes associated with cellular respiration and energy metabolism compared to young adults with comparable activity levels. Remarkably, in trained older adults, 50% of these age-related differences were absent, resulting in transcriptomic profiles for cellular respiration that closely aligned with those of young adults. Following acute exercise, trained older adults demonstrated molecular responses that more closely resembled those of younger individuals. While all participants displayed transcriptional immune and stress responses upon acute exercise, the magnitude of these responses in older adults correlated positively with their physical fitness. These findings underscore the capacity of sustained physical training to transform age-related molecular profiles, highlight a positive link between physical fitness level and exercise-induced inflammation in older adults, and provide a multi-omic molecular atlas for examining aging and fitness regulatory networks. Biological sciences/Molecular biology Biological sciences/Physiology/Ageing Biological sciences/Physiology/Metabolism Aging exercise muscle multi-omics lifestyle Figures Figure 1 Figure 2 Figure 3 Figure 4 Main Aging is associated with widespread molecular and functional changes, contributing to a decline in physical resilience and metabolic health 1,2 . Exercise is widely recognized as a powerful intervention to promote healthy aging, improving mobility, cardiovascular function, and metabolic efficiency 3–5 . Exercise training represents the cumulative adaptation of the body to sequential bouts of acute exercise, during which a dynamic molecular choreography unfolds 6,7 . Despite its established benefits, the molecular mechanisms by which exercise interacts with aging remain incompletely understood. In particular, it is unclear to what extent prolonged exercise training can offset the molecular hallmarks of aging in skeletal muscle, one of the most adaptable tissues in the body. Moreover, how older individuals respond to acute exercise at the molecular level is poorly characterized, let alone if this response depends on current physical fitness. While acute exercise is known to elicit transcriptional, metabolic, and lipidomic changes 6,7 , whether these responses vary across different fitness levels in older adults—and how they compare to younger individuals—remains an open question. Here, we address these gaps to understand how physical fitness shapes the aging trajectory which is essential for optimizing interventions to preserve muscle function and resilience across the lifespan. Aging is marked by decreased energy metabolism independent of habitual physical activity To investigate the molecular effects of an acute exercise bout across physical fitness levels during aging, we studied a cohort of 11 young and 36 older adults. The older group was further categorized as exercise-trained (n=16), normally active (n=15), or physically impaired (n=5) based on self-reported physical activity levels and a short physical performance battery test (Fig. 1a, Extended Data Fig. 1a-c). Importantly, young and normally active older adults displayed similar daily step counts and time spent in high-intensity activity (Fig. 1a, Extended Data Fig. 1b,c)—which was determined after group inclusion—suggesting that differences observed between these groups are likely due to aging rather than sedentary behavior that often accompanies aging. Trained and physically impaired older adults showed distinct activity patterns, also reflected in BMI (Fig. 1a, Extended Data Fig. 1a-c). All participants completed a 1-hour exercise bout at 50% of their maximum capacity on a stationary bike 8–10 . Muscle biopsies were collected before (baseline) and directly after exercise and analyzed using multimodal omics: (i) RNAseq for >24,000 transcripts, (ii) UPLC-HRMS metabolomics for 135 annotated metabolites, and (iii) UPLC-HRMS lipidomics for 1,383 lipids spanning 37 lipid classes (Extended Data Fig. 1d, Supplemental Tables 1-4). Statistical analyses took individual and sex into account. To explore this molecular compendium of data, we first compared young and normally active older adults at baseline across transcriptomics, metabolomics, and lipidomics. Partial Least Squares Discriminant Analysis (PLS-DA) revealed clear age-related differences in all three datasets (Fig. 1b-d). Among the 1,628 upregulated genes with aging, we observed enrichment of histamine response, microtubule organization, and splicing factors. Strikingly, in the 1,106 downregulated genes we identified reduced expression of genes strongly linked to cellular respiration and energy metabolism (Extended Data Fig. 2a,b, Fig. 1e, Supplemental Table 5). This is consistent with prior findings of reduced mitochondrial mass, respiration, and NAD + levels observed with aging in this cohort 8,9 . Metabolomic analysis confirmed a decline in NAD + pathway metabolites among the top 10 age-altered compounds (Extended Data Fig. 2c). Lipidomic analysis similarly highlighted energy storage increases, with several triglycerides accumulating among the top 20 age-altered lipid species (Extended Data Fig. 2d), aligning with previous age-related plasma 10 and muscle 11 studies from this cohort. Most normal age-related differences are absent in physically trained older adults We next examined how the molecular differences between young and old adults compared to exercise-trained and physically impaired older adults. To do this, we ranked transcripts based on how their expression followed a "muscle health” trend across the four groups: young adults, trained older adults, normally active older adults, and physically impaired older adults. By comparing these trends to age-related changes, we assessed how closely health and aging intersect (Fig. 2a). Strikingly, transcriptional changes with normal aging were strongly correlated with those following the health trend (Pearson’s r = 0.68, p < 2.2e-16), suggesting that exercise-trained older adults did not display the transcriptional shifts typically associated with aging in muscle (Fig. 2a). To explore this further, we categorized age-related transcriptional changes into two groups: (i) genes altered with aging but that maintained youthful expression in trained older adults (Fig. 2b, yellow/green), and (ii) genes consistently altered across all older adults compared to young individuals (Fig. 2b, blues). Remarkably, over half of the transcriptional differences seen in aging—both upregulated (Fig. 2c) and downregulated (Fig. 2d)—were absent in trained older adults, underscoring the benefits of exercise training. Sensitivity analyses confirmed that 45–62% of aging-related changes were absent in trained individuals, even when varying significance thresholds or excluding the physically impaired group (Supplemental Table 6). Gene Ontology analysis (Supplemental Table 7) revealed the strongest enrichment was in genes depleted in older adults but not in the exercise-trained older adults (Fig. 2e, green). These were particularly linked to cellular respiration and energy production, e.g. NDUFS1 (Fig. 2f, upper panel) and COX5A (Fig. 2f, lower panel). In contrast, the aging-related changes that occurred in all age groups, irrespective of training status, included the upregulation of synaptic transmission genes such as PCDH8 (Fig. 2g, upper panel) and UNC13C (Fig. 2g, lower panel) and the downregulation of WNT signaling pathway genes such as DAAM2 (Fig. 2h, upper panel) and CTR9 (Fig. 2h, lower panel), pointing to altered developmental processes. Nonetheless, these findings suggest that the most prominent transcriptional change in aging—reduced expression of respiration genes (Fig. 1e)—is also the most ‘preventable’ change in the sense that it is absent in trained older adults (Fig. 2e). We next investigated transcriptional changes unique to either a physically active lifestyle (trained) or physically impaired aging (Fig. 3a). In trained older adults, enriched gene pathways were predominantly metabolic, including increased lipid metabolism and decreased energy metabolism genes (Fig. 3b, yellow and orange, respectively; Supplemental Table 7). In contrast, physically impaired older adults exhibited increased immune response pathways and depletion of mitochondrial and respiration-related genes (Fig. 3b, mauve and purple, respectively; Supplemental Table 7). Notably, the mitochondrial gene depletion encompassed large and small mitochondrial ribosomal proteins, such as MRPS16, MRPL39, MRPL35, MRPL34, and MRPS18C (Fig. 3c-g), which we have previously linked to aging 12,13 . These findings highlight distinct molecular pathways associated with a physically active lifestyle (trained) and physical impairment, and further underscore the critical role of energy metabolism and mitochondrial function in healthy aging 14 . Molecular responses to exercise across age and fitness levels Exercise interventions and the stress of aging itself are known to uncover latent molecular responses and adaptive pathways 6,15 . Building on our understanding of the cohort at baseline, we next examined how an acute, one-hour bout of sub-maximal exercise rewires the molecular landscape across the four groups. Using multi-omics profiling post-exercise, PLS-DA revealed robust molecular responses across transcriptomic, lipidomic, and metabolomic levels in all groups (Fig. 4a) allowing to explore group responses across the omics levels (Fig. 4b). Older and young individuals with similar activity levels showed comparable proportions of upregulated (8.6% vs. 8.3%) and downregulated (6.6% vs. 5.0%) transcripts in response to exercise (Fig. 4c). However, older adults exhibited a stronger metabolomic and lipidomic response compared to young with similar activity levels, with greater depletions (4.4% vs. 0.7% metabolites; 1.4% vs. 0.5% lipids) and accumulations (14.8% vs. 3.0% metabolites; 21.9% vs. 5.4% lipids) of molecules (Fig. 4c). Exploring these metabolomic and lipidomic responses to acute exercise further revealed distinct patterns across age and fitness groups. In the metabolome, all groups showed typical exercise-related changes, such as acylcarnitine accumulation 16 , as well as glutathione and AICAR depletion, amongst others (Extended Data Fig. 3a-e, Extended Data Fig. 3f-h). In the lipidome, significant species-level changes were observed in all groups (Extended Data Fig. 4a-d). Unlike young individuals, however, older groups (trained, normally active, and physically impaired) showed enrichment of specific lipid classes (Supplemental Table 4), including dihexosylceramide (Hex2Cer[d]) accumulation (Extended Data Fig. 4e,f). Together these findings suggest that both metabolomic and lipidomic changes in response to exercise are more pronounced in older adults, irrespective of physical fitness level, contrasting with a more dampened response in young individuals. Following this, we turned to the transcriptomic responses and cross-referenced younger individuals with the three older adult groups (Fig. 4d-f). Remarkably, a stepwise pattern emerged, with trained older adults exhibiting exercise responses more similar to young adults (Pearson’s r=0.451, p<2e-16), more so than older adults with normal physical activity (Pearson’s r=0.394, p<2e-16), while physically impaired older adults had the most divergent response (Pearson’s r=0.263, p<2e-16) (Fig. 4g). Observing this, we next explored in more detail the factors explaining the dissimilarity in exercise response of normally active older individuals compared to young. We dissected the responses into quadrants, including co-upregulated (Q1), co-downregulated (Q3) or discordant (Q2 and Q4) (Fig. 4h, scatterplot). Here, we found co-upregulated genes to encompass stimulus and stress responses such as an upregulation of interleukin genes IL -6, ILB , and IL1RN , and increased TNF , SELE , and FOS (Fig. 4h Q1 , Fig. 4i, Extended Data Figure 5a-c, Supplemental Table 8). The stimulus and stress response activation was also clearly observed in both trained and physically impaired older adults, with fold changes of these genes following the same directionality (Fig. 4i, Extended Data Figure 5a-c, Supplemental Table 8). Co-downregulated genes included the suppression of calcium channel and ion transport (Fig. 4h, Q3, Supplemental Table 8). Intriguingly, the discordant quadrants revealed that older individuals possessed an additional immune-related stress response compared to young following exercise illustrated by the activation of IL7R and VTN (Fig. 4h, Q2, Extended Data Figure 5d-e, Supplemental Table 8), and a suppression of genes related to collagen binding (e.g. ITGA2 ) and molecular transport pathways (Fig. 4h, Q4, Extended Data Figure 5f, Supplemental Table 8). Investigating the transcriptomic response to acute exercise in trained and physically impaired older adults in a similar manner, i.e. assessing the concordant and discordant quadrant’s GO term enrichments, revealed additional insights (Extended Data Figure 6a,b). Indeed, trained older adults displayed a suppression—rather than activation—of certain immune response genes such as mast cell related CD226 gene and chemotaxis genes NMUR1 and RRH , while physically impaired older adults increased nitric oxide transport gene expression such as with the HBA1 gene, and underwent a suppression of cell-cell signaling transcription such as with FGF12 and MERTK (Extended Data Figure 6c-h, Supplemental Table 8). However, these changes were not as pronounced as the marked immune and cellular stress response displayed by all four groups following exercise (Figure 4h,i, Extended Data Figure 5a-c, Extended Data Figure 6a,b). Notably, the ability to elicit this stress response following acute exercise was lessened in impaired older adults and heightened in the trained (Fig. 4i). These findings reveal that physical fitness level significantly shapes the molecular response to exercise with aging, and that a stronger stress response following acute exercise is associated with healthier aging. Discussion Our study disentangles the molecular interplay between physical fitness, aging and the response to acute exercise in trained, normally active, and physically impaired older adults. By performing transcriptomics, lipidomics, and metabolomics on skeletal muscle from young and older adults with varying fitness levels, we found that aging is marked by a loss of transcripts encoding for cellular respiration, even in older individuals with daily physical activity levels comparable to young adults. Remarkably, exercise-trained older adults did not exhibit this decline, with over half of the transcriptional changes seen in aging being absent. Specifically, exercise trained older adults retained a youthful transcriptional profile with key energy metabolism and cellular respiration genes maintaining expression levels similar to young individuals, and responded to an acute bout of exercise also similarly to young adults. Taken together, our findings underscore the remarkable capacity of exercise training to mitigate age-associated molecular changes in skeletal muscle, and provides the molecular basis for previous studies that have highlighted physical fitness level as a potent modulator of health, capable of preserving energy metabolism levels in aging 8–10 . Our findings offer novel insights into aging biology in humans. Dissecting baseline molecular differences between young and older adults revealed two distinct categories of age-related changes: those related to physical activity and "preventable" by exercise training, defined as those absent in trained older adults and "unavoidable" aging changes defined as those shared across all older groups. Notably, it should be the ambition of the geroscience field to focus development of therapeutics on “unavoidable” age-related changes, while promoting lifestyle alterations to address the “preventable” age-related changes. In this regard, we aimed to map out the unavoidable molecular changes occurring with aging. Interestingly, these changes did not possess a single dominant pathway or biological process as enriched, suggesting stochastic forces. Finally, our results demonstrate that an enhanced immune and stress response to acute exercise correlates with improved resilience to physical challenges. This raises concerns that longevity interventions targeting immune suppression—such as recently described IL-11 inhibitors or well-known anti-inflammatory agents 17,18 —might inadvertently reduce functional resilience in humans. Our findings illuminate the intricate relationship between aging and exercise in human muscle, advocating for exercise as a foundational strategy to promote healthy aging and resilience, and provides a robust resource to further explore molecular regulators of aging and exercise. Methods Human subjects and procedures Forty-seven participants, including 11 young and 36 older adults were recruited in the community of Maastricht and its surroundings through advertisements at Maastricht University, in local newspapers, supermarkets, and at sports clubs, to collect both –before and –after measures of an exercise intervention. The study protocol was approved by the institutional Medical Ethical Committee and conducted in agreement with the declaration of Helsinki. All participants provided their written informed consent, and the study was registered at clinicaltrials.gov with identifier NCT03666013. Physiological data from this clinical trial has been reported in our previous study as part of a different analysis 8 , and the current study uses the same individuals for whom a pre and post-exercise biopsy was available. Prior to inclusion, all subjects underwent a medical screening that included a physical examination by a physician and an assessment of physical function using the Short Physical Performance Battery (SPPB), comprised of a standing balance test, a 4-m walk test, and a chair-stand test. After the screening procedure, participants were assigned to the following study groups: Young individuals with normal physical activity (20 – 30 years), older adults with normal physical activity (65 – 80 years), physically trained older adults (65 – 80 years) and physically impaired older adults (65 – 80 years). Participants were considered normal, physically active if they completed no more than one structured exercise session per week. Participants were considered trained if they engaged in at least 3 structured exercise sessions of at least 1 hour each per week for an uninterrupted period of more than one year. Participants were classified as older adults with impaired physical function in case of an SPPB score of ≤ 9. The SPPB score was calculated according to the cut-off points determined by (Guralnik et al. 1994) 19 . Exercise intervention Participants performed a 1-h submaximal exercise bout in the fasted state on an electronically braked cycle ergometer, at 50% of their W max as measured during a maximal aerobic cycling test 8 . Participants were instructed to pedal at a controlled cadence between 60 and 70 revolutions per minute. Muscle biopsy At 9 AM, after an overnight fast from 10 PM the preceding evening, and immediately following the acute exercise bout, muscle biopsies were taken from the m. vastus lateralis under local anesthesia (1.0% lidocaine without epinephrine) according to the Bergström method 20 . The muscle biopsies were immediately frozen in melting isopentane and stored at –80°C until further analysis. Habitual physical activity Habitual physical activity was determined in all participants using an ActivPAL monitor (PAL Technologies, Glasgow, Scotland) for a consecutive period of 5 days, including two weekend days. Besides the total amount of steps per day, the total stepping time was calculated in proportion to waking time, determined according to (van der Berg et al. 2016) 21 . Stepping time (i. e., physical activity) was then further classified into high-intensity physical activity (HPA; minutes with a step frequency > 110 steps/min in proportion to waking time) and lower-intensity physical activity (LPA; minutes with a step frequency ≤ 110 steps/min in proportion to waking time) 22 . RNA sequencing: Isolation of mRNA Human muscle tissues were homogenized with a 5 mm steel bead using a TissueLyser II (QIAGEN) for 5 min at frequency of 30 times/second. RNA was extracted according to the instructions of the RNaesy Mini Kit (QIAGEN). Contaminating genomic DNA was removed using RNase-Free DNase (QIAGEN). RNA was quantified with a NanoDrop 2000 spectrophotometer (Thermo Scientific; Breda, The Netherlands) and stored at -80°C until use. RNA sequencing: Library Preparation RNA libraries were prepared and sequenced with the Illumina platform by Genome Scan (Leiden, The Netherlands). The NEBNext Ultra II Directional RNA Library Prep Kit for Illumina was used to process the sample(s). The sample preparation was performed according to the protocol "NEBNext Ultra II Directional RNA Library Prep Kit for Illumina" (NEB #E7760S/L). Briefly, mRNA was isolated from total RNA using the oligo-dT magnetic beads. After fragmentation of the mRNA, cDNA synthesis was performed. This was used for ligation with the sequencing adapters and PCR amplification of the resulting product. The quality and yield after sample preparation was measured with the Fragment Analyzer. The size of the resulting products was consistent with the expected size distribution (a broad peak between 300-500 bp). Clustering and DNA sequencing using the NovaSeq6000 was performed according to manufacturer's protocols. A concentration of 1.1 nM of DNA was used. NovaSeq control software NCS v1.6 was used. RNA sequencing: Read Mapping Analysis for read mapping was performed in R v4.1.0 and Bioconductor v3.13. Reads were subjected to quality control FastQC 23 v0.11.15 and trimmed using Trimmomatic v0.36 (Bolger et al., 2014) and aligned using HISAT2 v2.1.0 (Kim et al., 2015) to the GRCh38 (v94) human genome reference assembly. Counts were obtained using HTSeq (v0.11.0, default parameters) (Anders et al., 2015) using the corresponding GTF taking into account the directions of the reads. Statistical analyses were performed using the edgeR v3.34.1 (Robinson et al., 2010) and limma/voom v 3.48.3 28 R packages. All genes with more than 2 counts in at least 4 of the samples were kept. Count data were transformed to log2-counts per million (logCPM), normalized by applying the trimmed mean of M-values method 27 and precision weighted using voom (Law et. al., 2014). Genes were reannotated using the Ensembl genome database (v104) and the biomaRt package 30,31 . Metabolite and Lipid extraction Metabolomics and lipidomics were performed at the Core Facility Metabolomics at the Amsterdam UMC, essentially as described 13 , 32 , 33 . In a 2 mL tube, the following amounts of internal standard dissolved in MilliQ were added to each sample of approximately 3-5 mg of freeze-dried muscle tissue: adenosine- 15 N 5 -monophosphate (5 nmol), adenosine- 15 N 5 -triphosphate (5 nmol), D 4 -alanine (0.5 nmol), D 7 -arginine (0.5 nmol), D 3 -aspartic acid (0.5 nmol), D 3 -carnitine (0.5 nmol), D 4 -citric acid (0.5 nmol), 13 C 1 -citrulline (0.5 nmol), 13 C 6 -fructose-1,6-diphosphate (1 nmol), guanosine- 15 N 5 -monophosphate (5 nmol), guanosine- 15 N 5 -triphosphate (5 nmol), 13 C 6 -glucose (10 nmol), 13 C 6 -glucose-6-phosphate (1 nmol), D 3 -glutamic acid (0.5 nmol), D 5 -glutamine (0.5 nmol), D 5 -glutathione (1 nmol), 13 C 6 -isoleucine (0.5 nmol), D 3 -lactic acid (1 nmol), D 3 -leucine (0.5 nmol), D 4 -lysine (0.5 nmol), D 3 -methionine (0.5 nmol), D 6 -ornithine (0.5 nmol), D 5 -phenylalanine (0.5 nmol), D 7 -proline (0.5 nmol), 13 C 3 -pyruvate (0.5 nmol), D 3 -serine (0.5 nmol), D 6 -succinic acid (0.5 nmol), D5-tryptophan (0.5 nmol), D 4 -tyrosine (0.5 nmol), D 8 -valine (0.5 nmol). For lipidomics, the following internal standards dissolved in 50:50 MeOH:CHCl 3 were added: DG(14:0) 2 , TG(14:0) 3 , CE(16:0) -d7 , PC(14:0) 2 , PS(14:0) 2 , PE(14:0) 2 , PA(14:0) 2 , ST(17:0), PI(8:0) 2 , LPE(14:0), LPC(14:0), LPA(14:0), SPH(d17:1), SM(12:0), SPH(d17:0), S1P(d17:1), S1P(d17:0), LacCer(d18:1/12:0), GlcCer(d18:1/12:0), Cer(d18:1/12:0), C1P(d18:1/12:0), Cer(d18:1/25:0). After adding the internal standard mixtures, a 5 mm stainless-steel bead and polar phase solvents (for a total of 500 µL MilliQ and 500 µL MeOH) were added and samples were homogenized using a TissueLyser II (Qiagen, Hilden, Germany) for 5 min at a frequency of 30 times/sec. Chloroform was added for a total of 1 mL to each sample before thorough mixing. Samples were then centrifuged for 10 minutes at 18.000 g . The top and bottom layer were each transferred to a new 1.5 mL tube for separate processing. Metabolomics analysis The top layer, containing polar metabolites, was dried using a vacuum concentrator at 60°C. Dried samples were reconstituted in 100 µL 3:2 (v/v) MeOH:MilliQ. Metabolites were analyzed using a Waters Acquity ultra-high performance liquid chromatography system coupled to a Bruker Impact II™ Ultra-High Resolution Qq-Time-Of-Flight mass spectrometer. Samples were kept at 12°C during analysis and 5 µL of each sample was injected. Chromatographic separation was achieved using a Merck Millipore SeQuant ZIC-cHILIC column (PEEK 100 x 2.1 mm, 3 µm particle size). Column temperature was held at 30°C. Mobile phase consisted of (A) 1:9 (v/v) ACN:MilliQ and (B) 9:1 (v/v) ACN:MilliQ, both containing 5 mmol/L ammonium acetate. Using a flow rate of 0.25 mL/min, the LC gradient consisted of: 100% B for 0-2 min, reach 0% B at 28 min, 0% B for 28-30 min, reach 100% B at 31 min, 100% B for 31-32 min. Column re-equilibration is achieved at a flow rate of 0.4 mL/min at 100% B for 32-35 min. MS data were acquired using negative and positive ionization in full scan mode over the range of m/z 50-1200. Data were analyzed using Bruker TASQ software version 2.1.22.3. All reported metabolite intensities were normalized to dry tissue weight, as well as to internal standards with comparable retention times and response in the MS. General repeatability of metabolite analysis was assessed for each metabolite using repeated measurements of a pooled quality control (QC) sample. Metabolite identification has been based on a combination of accurate mass, (relative) retention times and fragmentation spectra, compared to the analysis of a library of standards in separate experiments not described here. Lipidomics analysis The bottom layer of the extraction, containing lipids, was dried under a stream of nitrogen at 30°C and reconstituted in 100 µL 50:50 MeOH:CHCl 3 . The UPLC system consisted of an Ultimate 3000 binary HPLC pump, a vacuum degasser, a column temperature controller, and an auto sampler (Thermo Scientific, Waltham, MA, USA). Samples were run in normal and reverse phase and positive and negative ionization. For normal phase, 2μL of each sample was injected into the system. The normal phase system consisted of a Lichrospher Si 60, 2 x 250 mm silica 100 Å column, 5 µm particle diameter (Merck, Germany), the column temperature was maintained at 25°C. Lipids were separated using a linear gradient between solution B (CHCl 3 /MeOH, 97:3 v/v) and solution A (MeOH/MilliQ, 85:15, v/v). Solution A contained 0.0125% formic acid and 3.35 mmol/l ammonia per liter of eluent. Solution B contained 0.0125% formic acid per liter. The gradient (0.3 ml/min) was as follows: 0-1 min 10%A, 1–4 min 10%A–20%A, 4–12 min 20%A–85% A, 12–12.1 min, 85%A–100% A, 12.1–14.0 min 100% A, 14-14.1 min 100%A–10%A and 14.1–15 min equilibration with 10% A. All gradient steps were linear, and the total analysis time, including the equilibration, was 15 min. For reversed phase separation, 5 μL of each sample was injected onto a Waters HSS T3 column (150 x 2.1 mm, 1.8 μm particle size). Column temperature was held at 60°C. Mobile phase consisted of (A) 4:6 (v/v) MeOH:MilliQ and B 1:9 (v/v) MeOH:IPA, both containing 0.1% formic acid and 10 mmol/L ammonia. Using a flow rate of 0.4 mL/min, the LC gradient consisted of: Dwell at 100% A at 0 min, ramp to 80% A at 1 min, ramp to 0% A at 16 min, dwell at 0% A for 16-20 min, ramp to 100% A at 20.1 min, dwell at 100% A for 20.1-21 min. A Q Exactive Plus (Thermo Scientific) mass spectrometer was used in the negative and positive electrospray ionization mode. In both ionization modes, mass spectra of the lipid species were obtained by continuous scanning from m/z 150 to m/z 2000 with a resolution of 280.000. Nitrogen was used as the nebulizing gas. The spray voltage used was 2500 V (-) and 3500 V (+), and the capillary temperature was 256°C. S-lens RF level: 50, Auxiliary gas: 10, Auxiliary gas temperature 300°C, Sheath gas: 50, Sweep cone gas: 2. Lipidomics: bioinformatics for lipid identification Lipid identification was performed at the Core Facility Metabolomics at the Amsterdam UMC, essentially as described 33 . The raw LC/MS data were converted to mzXML format using MSConvert 34 . The dataset was processed using an in-house developed metabolomics pipeline written in the R programming language (http://www.r-project.org) 35 . In brief, it consisted of the following steps: (1) pre-processing using the R package XCMS 36 with minor changes to some functions in order to better suit the Q Exactive data; notably, the definition of noise level in centWave was adjusted and the stepsize in fillPeaks (2) identification of metabolites using an in-house database of (phospho)lipids, with known internal standards indicating the position of most of the lipid clusters, matching m/z values within 3 ppm deviation, (3) isotope correction to obtain deconvoluted intensities for overlapping peak groups, (4) normalization on the intensity of the internal standard for lipid classes for which an internal standard was available (with normalization on the intensity of PE(14:0) 2 for lipid classes for which no internal standard was present) and dry tissue weight. For quantifying abundances of lipid classes, the summed abundances of the individual lipid species from the relevant class were used. General repeatability of lipid analysis was assessed for each lipid using repeated measurements of a pooled quality control (QC) sample. Lipid class identification has been based on accurate mass, fragmentation analysis, relative retention times and ion mobility, compared to the analysis of relevant standards in separate experiments not described here. Exercise response quadrants Exercise response quadrants (log2 fold changes) of each of the three older adult groups were compared to young adults and dissected into quadrants as follows: Q1 (co-upregulated); log2 fold change exercise response >1 for both young and old). Q3 (co-downregulated); log2 fold change exercise response 1 old, <0 young. Q4 (discordant response); log2 fold change exercise response 0 young). Statistics and reproducibility Sample size determination for the human studies are described in the original publications 8,37 and were done to accommodate the goals of the clinical trials (clinicaltrials.gov identifier NCT03666013). No data was intentionally excluded from the analyses (if mRNA, metabolites, or lipids are missing this is due to them not being detected in the tissue/sample and if samples are omitted this is due to the sample being lost during sample preparation). Sample collection was not blinded due to obvious differences in subject groups, however, the downstream processing of all –omics samples was blinded and randomized. Un-blinding and de-randomization occurred at the final step for statistical analysis and interpretation. Unless otherwise noted, analyses were done with R 35 version 3.5.1 and Bioconductor 38 version 3.7. For RNAseq, metabolomics and lipidomics data, dfferential expression was assessed using an empirical Bayes moderated t test within limma’s linear model framework using log2 transformed data for the metabolome and lipidome and for RNAseq precision weights estimated by voom 28,29 . Individuals, sex, and before-after exercise were included as covariates in the model. Resulting p values were corrected for multiple testing using the Benjamini-Hochberg false discovery rate. Data was processed in part with the R package dplyr version 1.0.2 39 . Partial least-squares discriminant analysis (PLS-DA) was performed using the R package MixOmics version 6.6.2 40 . Networks were constructed and visualized using igraph version 1.2.4.2 41 . Unless implemented through an aforementioned R package or base R graphics, visualization of data was performed using ggplot2 version 3.2.1 42 , ggpubr v 0.2.5 43 , ggrepel version 0.8.1 44 , with colors from RColorBrewer version 1.1-2 45 . For pathway enrichment analyses: Gene ontology (GO) term enrichments were calculated for RNAseq gene lists with a hypergeometric test using the GOstats package (version 2.48.0) in R, or by using the DAVID gene enrichment online resource (https://david.ncifcrf.gov/). Metabolite enrichment analysis was performed using the MetabAnalyst v6.0 online tool (https://www.metaboanalyst.ca/) using a global test 46 of significance on KEGG human metabolic pathways (Dec. 2023). Declarations Data availability Data supporting the conclusions of our study are available as supplementary materials accompanying this article as summary statistics comparing groups (see Supplemental Data Tables). Physiological data from the cohort has been reported in our previous study as part of a different analysis 8 . All other data supporting the findings of this study are available either in additional Supplemental Data Tables or from the corresponding authors upon reasonable request. Code availability Code supporting the findings of this study are available from the corresponding authors upon reasonable request. Acknowledgements The project was supported by a Longevity Impetus Grant from Norn Group (to GEJ and RHH). Work in the Houtkooper group is financially supported by funding from the European Union’s Horizon Europe research and innovation programme through the MSCA-Doctoral Network NADIS (no. 101073251), and by the Velux Stiftung (no. 1063). GEJ is supported by an AGEM Talent grant. Human interventions were further financed by the TIFN research program Mitochondrial Health (ALWTF.2015.5) and the Netherlands Organization for Scientific Research (NWO). Author contributions G.E.J., P.S., J.H., R.H.H. conceived the study. L.G. performed the human clinical trial and experiments. G.E.J designed and performed the bioinformatics analyses. A.S. and S.W.D. performed molecular extractions for omics preparation. A.J. performed RNAseq mapping and statistical analyses. B.V.S and M.v.W., M.A.T.V., and E.J.M.W. performed the metabolomics and lipidomics analyses. G.S. and F.M.V. reviewed the manuscript. G.E.J., M.K. P.S. R.H.H and J.H. interpreted the results and wrote the manuscript with contributions from all other authors. Competing Interests The authors declare that they have no competing interests related to this work. The funders had no role in data collection, analysis, or decision to publish or in preparation of the manuscript. References López-Otín, C., Blasco, M. A., Partridge, L., Serrano, M. & Kroemer, G. The hallmarks of aging. Cell 153 , (2013). López-Otín, C., Blasco, M. A., Partridge, L., Serrano, M. & Kroemer, G. Hallmarks of aging: An expanding universe. Cell 186 , 243–278 (2023). Viña, J., Rodriguez-Mañas, L., Salvador-Pascual, A., Tarazona-Santabalbina, F. J. & Gomez-Cabrera, M. C. Exercise: The lifelong supplement for healthy ageing and slowing down the onset of frailty. Journal of Physiology (2016). doi:10.1113/JP270536 Tarazona-Santabalbina, F. J. et al. A Multicomponent Exercise Intervention that Reverses Frailty and Improves Cognition, Emotion, and Social Networking in the Community-Dwelling Frail Elderly: A Randomized Clinical Trial. J. Am. Med. Dir. Assoc. 17 , 426–433 (2016). Viña, J., Sanchis-Gomar, F., Martinez-Bello, V. & Gomez-Cabrera, M. C. Exercise acts as a drug; The pharmacological benefits of exercise. Br. J. Pharmacol. 167 , 1–12 (2012). Contrepois, K. et al. Molecular Choreography of Acute Exercise. Cell 181 , 1112-1130.e16 (2020). Hornburg, D. et al. Dynamic lipidome alterations associated with human health, disease and ageing. Nat. Metab. 5 , 1578–1594 (2023). Grevendonk, L. et al. Impact of aging and exercise on skeletal muscle mitochondrial capacity, energy metabolism, and physical function. Nat. Commun. 12 , 1–17 (2021). Janssens, G. E. et al. Healthy aging and muscle function are positively associated with NAD+ abundance in humans. Nat. Aging (2022). doi:10.1038/s43587-022-00174-3 Janssens, G. E. et al. A metabolomic signature of decelerated physiological aging in human plasma. GeroScience 45 , 3147–3164 (2023). Janssens, G. E. et al. A conserved complex lipid signature marks human muscle aging and responds to short-term exercise. Nat. Aging (2024). doi:10.1038/s43587-024-00595-2 Houtkooper, R. H. et al. Mitonuclear protein imbalance as a conserved longevity mechanism. Nature 497 , 451–457 (2013). Molenaars, M. et al. A Conserved Mito-Cytosolic Translational Balance Links Two Longevity Pathways. Cell Metab. (2020). doi:10.1016/j.cmet.2020.01.011 Houtkooper, R. H. et al. Mitonuclear protein imbalance as a conserved longevity mechanism. Nature 497 , 451–457 (2013). Reitzner, S. M. et al. Molecular profiling of high-level athlete skeletal muscle after acute endurance or resistance exercise – A systems biology approach. Mol. Metab. 79 , 101857 (2024). Hiatt, W. R., Regensteiner, J. G., Wolfel, E. E., Ruff, L. & Brass, E. P. Carnitine and acylcarnitine metabolism during exercise in humans. Dependence on skeletal muscle metabolic state. J. Clin. Invest. 84 , 1167–1173 (1989). Widjaja, A. A. et al. Inhibition of IL-11 signalling extends mammalian healthspan and lifespan. Nature 632 , (2024). He, C. et al. Enhanced Longevity by Ibuprofen, Conserved in Multiple Species, Occurs in Yeast through Inhibition of Tryptophan Import. PLoS Genet. 10 , e1004860 (2014). Guralnik, J. M. et al. A short physical performance battery assessing lower extremity function: Association with self-reported disability and prediction of mortality and nursing home admission. Journals Gerontol. (1994). doi:10.1093/geronj/49.2.M85 Bergström, J., Hermansen, L., Hultman, E. & Saltin, B. Diet, Muscle Glycogen and Physical Performance. Acta Physiol. Scand. (1967). doi:10.1111/j.1748-1716.1967.tb03720.x van der Berg, J. D. et al. Identifying waking time in 24-h accelerometry data in adults using an automated algorithm. J. Sports Sci. (2016). doi:10.1080/02640414.2016.1140908 Tudor-Locke, C. et al. How many steps/day are enough? for adults. Int. J. Behav. Nutr. Phys. Act. (2011). doi:10.1186/1479-5868-8-79 Andrews, S. & others. FastQC: a quality control tool for high throughput sequence data. 2010. Https://Www.Bioinformatics.Babraham.Ac.Uk/Projects/Fastqc/ http://www.bioinformatics.babraham.ac.uk/projects/ (2019). Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30 , 2114–2120 (2014). Kim, D., Langmead, B. & Salzberg, S. L. HISAT: a fast spliced aligner with low memory requirements. Nat. Methods 12 , 357–360 (2015). Anders, S., Pyl, P. T. & Huber, W. HTSeq—a Python framework to work with high-throughput sequencing data. Bioinformatics 31 , 166–169 (2015). Robinson, M. D., McCarthy, D. J. & Smyth, G. K. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 26 , 139–140 (2010). Ritchie, M. E. et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 43 , e47 (2015). Law, C. W., Chen, Y., Shi, W. & Smyth, G. K. voom: Precision weights unlock linear model analysis tools for RNA-seq read counts. Genome Biol. 15 , R29 (2014). Durinck, S. et al. BioMart and Bioconductor: A powerful link between biological databases and microarray data analysis. Bioinformatics 21 , 3439–3440 (2005). Durinck, S., Spellman, P. T., Birney, E. & Huber, W. Mapping identifiers for the integration of genomic datasets with the R/Bioconductor package biomaRt. Nat. Protoc. 4 , 1184–1191 (2009). Molenaars, M. et al. Metabolomics and lipidomics in C. elegans using a single sample preparation. Dis. Model. Mech. (2021). doi:10.1242/dmm.047746 Herzog, K. et al. Lipidomic analysis of fibroblasts from Zellweger spectrum disorder patients identifies disease-specific phospholipid ratios. 57 , 1447–1454 (2016). Chambers, M. C. et al. A cross-platform toolkit for mass spectrometry and proteomics. Nat. Biotechnol. 30 , 918–920 (2012). Core R Team. A Language and Environment for Statistical Computing. R Foundation for Statistical Computing 2 , https://www.R--project.org (2019). Smith, C. A., Want, E. J., O’Maille, G., Abagyan, R. & Siuzdak, G. XCMS: Processing mass spectrometry data for metabolite profiling using nonlinear peak alignment, matching, and identification. Anal. Chem. 78 , 779–787 (2006). Remie, C. M. E. et al. Sitting less elicits metabolic responses similar to exercise and enhances insulin sensitivity in postmenopausal women. Diabetologia 64 , 2817–2828 (2021). Gentleman, R. C. et al. Bioconductor: open software development for computational biology and bioinformatics. Genome Biol. 5 , (2004). Wickham, H., François, R., Henry, L. & Müller, K. dplyr: A Grammar of Data Manipulation. R package version. Media (2019). Rohart, F., Gautier, B., Singh, A. & Lê Cao, K. A. mixOmics: An R package for ‘omics feature selection and multiple data integration. PLoS Comput. Biol. 13 , e1005752 (2017). Csardi, G. & Nepusz, T. The igraph software package for complex network research. InterJournal Complex Sy , 1695 (2006). Wickham, H. Ggplot2. Wiley Interdiscip. Rev. Comput. Stat. 3 , 180–185 (2011). Kassambara, A. Package ‘ggpubr’: ‘ggplot2’ Based Publication Ready Plots. R Packag. version 0.4.0 (2020). Slowikowski, K. ggrepel: Automatically Position Non-Overlapping Text Labels with ‘ggplot2’. R package version 0.8.2 (2020). Neuwirth, E. RColorBrewer: ColorBrewer palettes. R Packag. version 1.1-2 https://cran.R-project.org/package=RColorBrewer (2014). Goeman, J. J., Van de Geer, S., De Kort, F. & van Houwellingen, H. C. A global test for groups fo genes: Testing association with a clinical outcome. Bioinformatics 20 , 93–99 (2004). Additional Declarations There is NO Competing Interest. Supplementary Files TableS1RNAseq.csv Supplemental Table 1: RNAseq transcriptomics TableS2Metabolomics.csv Supplemental Table 2: MS metabolomics TableS3LipidSpecies.csv Supplemental Table 3: MS lipidomics (lipid species) TableS4LipidClass.csv Supplemental Table 4: MS lipidomics (lipid class) TableS5GOtermsNormalOldvsYoung.csv Supplemental Table 5: RNAseq GO terms normal old vs young TableS6SensitivityAnalysis.csv Supplemental Table 6: Sensitivity analysis results TableS7GOtermsavoidableorunavoidableagingchanges.csv Supplemental Table 7: RNAseq GO terms absent/present age related changes TableS8GOtermexercisequadrants.csv Supplemental Table 8: RNAseq GO terms exercise response quadrants SupplementalFiguresv5.docx Extended Data Figures Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6074097","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Biological Sciences - Article","associatedPublications":[],"authors":[{"id":421241611,"identity":"333c1009-eeb2-4348-8a85-7acdeb5e6c7f","order_by":0,"name":"Georges Janssens","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0003-2104-8145","institution":"Academic Medical Centre","correspondingAuthor":true,"prefix":"","firstName":"Georges","middleName":"","lastName":"Janssens","suffix":""},{"id":421241612,"identity":"0de3e4bd-33fb-4dc3-ba20-4665d812e945","order_by":1,"name":"Marit Kotte","email":"","orcid":"","institution":"Maastricht University","correspondingAuthor":false,"prefix":"","firstName":"Marit","middleName":"","lastName":"Kotte","suffix":""},{"id":421241613,"identity":"a0ebf19c-8d49-4ec7-9cdc-c49909dbe918","order_by":2,"name":"Lotte Grevendonk","email":"","orcid":"https://orcid.org/0000-0001-9540-1832","institution":"Maastricht University Medical Centre","correspondingAuthor":false,"prefix":"","firstName":"Lotte","middleName":"","lastName":"Grevendonk","suffix":""},{"id":421241614,"identity":"c255d17c-7f0e-49f7-a6bc-a98a9d8173cf","order_by":3,"name":"Angelique Scantlebery","email":"","orcid":"","institution":"University of Amsterdam","correspondingAuthor":false,"prefix":"","firstName":"Angelique","middleName":"","lastName":"Scantlebery","suffix":""},{"id":421241615,"identity":"29cc2e2b-e4c4-4bf3-93c9-c492f867c113","order_by":4,"name":"Bauke Schomakers","email":"","orcid":"","institution":"Amsterdam UMC, University of Amsterdam","correspondingAuthor":false,"prefix":"","firstName":"Bauke","middleName":"","lastName":"Schomakers","suffix":""},{"id":421241616,"identity":"f8aff95a-d058-4eb7-aee9-8a950c4e2f30","order_by":5,"name":"Michel van Weeghel","email":"","orcid":"https://orcid.org/0000-0002-4916-2866","institution":"Academic Medical Centre","correspondingAuthor":false,"prefix":"","firstName":"Michel","middleName":"van","lastName":"Weeghel","suffix":""},{"id":421241617,"identity":"4459cda2-c58c-484f-badc-a884e7cd71ee","order_by":6,"name":"Jill Hermans","email":"","orcid":"","institution":"University of Amsterdam","correspondingAuthor":false,"prefix":"","firstName":"Jill","middleName":"","lastName":"Hermans","suffix":""},{"id":421241618,"identity":"cfd50668-c073-4303-9525-aa9526e888a3","order_by":7,"name":"Martin Vervaart","email":"","orcid":"","institution":"Academic Medical Centre","correspondingAuthor":false,"prefix":"","firstName":"Martin","middleName":"","lastName":"Vervaart","suffix":""},{"id":421241619,"identity":"7e5180bb-74ad-45b6-90f3-ad5a82a98af1","order_by":8,"name":"Eric Wever","email":"","orcid":"https://orcid.org/0000-0002-2116-507X","institution":"Amsterdam UMC","correspondingAuthor":false,"prefix":"","firstName":"Eric","middleName":"","lastName":"Wever","suffix":""},{"id":421241620,"identity":"8203431c-949b-49fd-a6a2-942196e65d99","order_by":9,"name":"Simone Denis","email":"","orcid":"","institution":"Academic Medical Centre","correspondingAuthor":false,"prefix":"","firstName":"Simone","middleName":"","lastName":"Denis","suffix":""},{"id":421241621,"identity":"2d8a5f36-8f7b-4507-9731-06f037957c96","order_by":10,"name":"Aldo Jongejan","email":"","orcid":"https://orcid.org/0000-0002-8948-2549","institution":"Amsterdam University Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Aldo","middleName":"","lastName":"Jongejan","suffix":""},{"id":421241622,"identity":"76bb98f3-99f7-404c-9bd6-38a794225bb8","order_by":11,"name":"Gajja Salomons","email":"","orcid":"","institution":"University of Amsterdam","correspondingAuthor":false,"prefix":"","firstName":"Gajja","middleName":"","lastName":"Salomons","suffix":""},{"id":421241623,"identity":"ee7b402a-60ca-4af4-862c-18541c323534","order_by":12,"name":"Frederic Vaz","email":"","orcid":"https://orcid.org/0000-0002-9048-1041","institution":"Amsterdam UMC","correspondingAuthor":false,"prefix":"","firstName":"Frederic","middleName":"","lastName":"Vaz","suffix":""},{"id":421241624,"identity":"8e35a6b3-2c35-44a5-8af7-9b84f5ebbac9","order_by":13,"name":"Patrick Schrauwen","email":"","orcid":"https://orcid.org/0000-0002-0973-847X","institution":"Maastricht University","correspondingAuthor":false,"prefix":"","firstName":"Patrick","middleName":"","lastName":"Schrauwen","suffix":""},{"id":421241625,"identity":"2c204ca9-2740-4cbd-b2eb-c5cb67fa9e8c","order_by":14,"name":"Joris Hoeks","email":"","orcid":"https://orcid.org/0000-0002-0265-0870","institution":"Maastricht University Medical Center+","correspondingAuthor":false,"prefix":"","firstName":"Joris","middleName":"","lastName":"Hoeks","suffix":""},{"id":421241626,"identity":"06e41889-5577-43aa-90fe-d14da7fbe491","order_by":15,"name":"Riekelt Houtkooper","email":"","orcid":"https://orcid.org/0000-0001-9961-0842","institution":"Amsterdam UMC","correspondingAuthor":false,"prefix":"","firstName":"Riekelt","middleName":"","lastName":"Houtkooper","suffix":""}],"badges":[],"createdAt":"2025-02-20 18:20:13","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6074097/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6074097/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":78735585,"identity":"8b4dbdb2-8fd4-4d35-9534-4934d340c87e","added_by":"auto","created_at":"2025-03-18 08:08:58","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":196252,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCohort description and muscle molecular profiling of older vs young adults. a,\u003c/strong\u003e Cohort characteristics, whereby data distributions can be seen in Extended Data Figure 1. Age range (represented in years), body mass index (BMI), average steps per day (Steps/day) and time spent in high activity (%) are expressed as mean values of all participants in the group with standard deviation (sd) presented adjacently (±) or in a separate row. b-d PLS-DA comparing young and normally active older adults of \u003cstrong\u003eb,\u003c/strong\u003e transcriptome, \u003cstrong\u003ec,\u003c/strong\u003e metabolome, and \u003cstrong\u003ed,\u003c/strong\u003elipidome. \u003cstrong\u003ee,\u003c/strong\u003e Gene Ontology (GO) term enrichment of significantly downregulate genes (p\u0026lt;0.05, n=1106) in normally active older adults compared to young. n=11 young, n=15 normally active older adults. Enrichments were calculated using a hypergeometric test relative to the entire transcriptome as background. P values were corrected for multiple hypothesis testing (adj pvalue) using the Benjamini method.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6074097/v1/ffef4d68b709cf91173e533f.png"},{"id":78735592,"identity":"4a554607-0d36-4171-962f-e5940ebd72b7","added_by":"auto","created_at":"2025-03-18 08:09:02","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":352731,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTranscriptomic comparisons between baseline young and old health groups. a,\u003c/strong\u003e Comparison of: (x-axis) the correlation an mRNA has to the four muscle health groups (young adults, trained older adults, older adults with normal physical activity levels, and physically impaired older adults), where the color of the point on the graph indicates p-value of the correlation (-log10 scale, with directionality represented as either yellow, positive, or green, negative), and correlation and significance were determined using Pearson’s product-moment correlation coefficient, and (y-axis) the fold change the mRNA undergoes in older vs. young adults with normal physical activity, where significance was determined using an empirical Bayes moderated t-test (two-sided, p values adjusted for multiple comparisons between groups). The graph reveals a strong correlation (Pearson’s r=0.69, p\u0026lt;2.2e-16) between age-related changes and the ‘muscle health’ trend of the four groups.\u003cstrong\u003e b,\u003c/strong\u003e schematic of age-related changes which can be absent in trained older adults (yellow, green) or occur regardless of a physically active lifestyle in older adults (lighter blue, darker blue). c-d, \u003cstrong\u003ec,\u003c/strong\u003eupregulated absent (red) or present (blue) changes with a physically active lifestyle and \u003cstrong\u003ed,\u003c/strong\u003e downregulated absent (dark blue) or present (green) changes with a physically active lifestyle reveal \u0026gt; 50% are absent. \u003cstrong\u003ee,\u003c/strong\u003e GO term enrichments for each category of c-d. Count and size refer to the number of genes counted as differentially expressed considering the size of the GO term. f-h, examples of differentially expressed genes matching criteria of schematic of panel b; \u003cstrong\u003ef\u003c/strong\u003e, \u003cem\u003eNDUFS1\u003c/em\u003e and \u003cem\u003eCOX5A\u003c/em\u003e are downregulated in normally active/physically impaired older adults and the difference is absent in trained older adults, \u003cstrong\u003eg\u003c/strong\u003e, \u003cem\u003ePCDH8\u003c/em\u003e and \u003cem\u003eUNC13C\u003c/em\u003e are upregulated in all three older adult groups and \u003cstrong\u003eh\u003c/strong\u003e, \u003cem\u003eDAAM2\u003c/em\u003e and \u003cem\u003eCTR9\u003c/em\u003e are downregulated in all three older adult groups. n = 11, 16, 15, 5 in young, and trained, normally active, and physically impaired older adults, respectively. Y-axis values expressed as counts per million (log2) with normalization and weighted with voom. Statistics calculated using a hypergeometric test (GO terms) or empirical Bayes moderated t-test (two-sided, p values, adjusted for four-group comparison).\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6074097/v1/8ac2d001251e4fc719dea921.png"},{"id":78735586,"identity":"694357a7-b186-4690-83a4-b40894eb49c7","added_by":"auto","created_at":"2025-03-18 08:08:58","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":244231,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eBaseline comparisons between trained and physically impaired aging patterns. a,\u003c/strong\u003e schematic of age-related changes which can be present only either in exercise-trained older adults (yellow, orange) or physically impaired older adults (mauve, purple). \u003cstrong\u003eb,\u003c/strong\u003eGO term enrichments for each category of schematic. Count and size refer to the number of genes counted as differentially expressed considering the size of the GO term. c-h, examples of genes downregulated in impaired aging related to mitochondrial ribosomal proteins including \u003cstrong\u003ec,\u003c/strong\u003e \u003cem\u003eMRPS16\u003c/em\u003e, \u003cstrong\u003ed,\u003c/strong\u003e \u003cem\u003eMRPL39\u003c/em\u003e, \u003cstrong\u003ee,\u003c/strong\u003e \u003cem\u003eMRPL35\u003c/em\u003e, \u003cstrong\u003ef,\u003c/strong\u003e \u003cem\u003eMRPL34\u003c/em\u003e, \u003cstrong\u003eg,\u003c/strong\u003e \u003cem\u003eMRPS18C\u003c/em\u003e. n = 11, 16, 15, 5 in young, and trained, normally active, and physically impaired older adults, respectively. Y-axis values expressed as counts per million (log2) with normalization and weighted with voom. Statistics calculated using a hypergeometric test (GO terms) or empirical Bayes moderated t-test (two-sided, p values, adjusted for four-group comparison).\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6074097/v1/f07cf6500be4b6e8b2e568be.png"},{"id":78735593,"identity":"8896a9aa-1cd9-4927-8553-76bc916ee648","added_by":"auto","created_at":"2025-03-18 08:09:02","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":515190,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAge-associated molecular exercise responses. a,\u003c/strong\u003e PLS-DA of before and after exercise in either young adults or trained, normally active, or physically impaired older adults, for the transcriptome (red), metabolome (blue) or lipidome (yellow), x-axis is % explained variance from x-variate 1, y-axis is % explained variance from x-variate 2 of PLS-DA. \u003cstrong\u003eb,\u003c/strong\u003e Before-after exercise responses for the transcriptome, metabolome, or lipidome for each individual. \u003cstrong\u003ec,\u003c/strong\u003e comparison of total changes either up or down in either young or normally active older adults at the transcript (red), metabolite (blue), or lipid (yellow) levels. \u003cstrong\u003ed-f\u003c/strong\u003e, Transcriptomic responses. Binned hexagonal scatterplots (color indicating point density) comparing log2 fold changes of before-after exercise in young (x-axis) to (y-axis) of: \u003cstrong\u003ed,\u003c/strong\u003e normally active older adults, \u003cstrong\u003ee,\u003c/strong\u003e trained older adults, or \u003cstrong\u003ef,\u003c/strong\u003e physically impaired older adults.\u003cstrong\u003e g,\u003c/strong\u003e Summary line graph of Pearson’s correlation coefficient from d-f. \u003cstrong\u003eh,\u003c/strong\u003e quadrants comparing log2 fold changes of before-after exercise in young (x-axis) or normally active older adults (y-axis) and respective GO terms for each quadrant (right panel). \u003cstrong\u003ei,\u003c/strong\u003ePairwise comparisons of transcript abundance before (left of pair) or after (right of pair) a short bout of exercise, for each of the four groups, for \u003cem\u003eIL-6\u003c/em\u003e and \u003cem\u003eSELE\u003c/em\u003e transcripts. n = 11, 15, 15 and 5 in young, and trained, normally active, and physically impaired older adults, respectively, before exercise. Y-axis values expressed as counts per million (log2). Statistics calculated using a hypergeometric test (GO terms) or empirical Bayes moderated t-test (two-sided, p values, adjusted for four-group comparison).\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6074097/v1/d8114ae8304d64ffcda8bcc8.png"},{"id":78736165,"identity":"3cf60f9e-6f70-4626-a924-2d58ab43e8e7","added_by":"auto","created_at":"2025-03-18 08:16:47","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2137816,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6074097/v1/3fef5b9c-58a7-4a8d-b5a6-26b27550021d.pdf"},{"id":78735589,"identity":"4b759224-bab7-4e0b-9c14-39e8445676c6","added_by":"auto","created_at":"2025-03-18 08:08:59","extension":"csv","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":27219777,"visible":true,"origin":"","legend":"\u003cp\u003eSupplemental Table 1: RNAseq transcriptomics\u003c/p\u003e","description":"","filename":"TableS1RNAseq.csv","url":"https://assets-eu.researchsquare.com/files/rs-6074097/v1/0ce18a893221c63e17462d07.csv"},{"id":78735590,"identity":"43797dad-12ba-4d72-a924-d90fb46a4374","added_by":"auto","created_at":"2025-03-18 08:09:00","extension":"csv","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":202124,"visible":true,"origin":"","legend":"\u003cp\u003eSupplemental Table 2: MS metabolomics\u003c/p\u003e","description":"","filename":"TableS2Metabolomics.csv","url":"https://assets-eu.researchsquare.com/files/rs-6074097/v1/42dc877d42286cefef8caf3d.csv"},{"id":78735594,"identity":"3d9a905f-0538-4cd6-b3ec-1e025d734c0f","added_by":"auto","created_at":"2025-03-18 08:09:02","extension":"csv","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":1625512,"visible":true,"origin":"","legend":"\u003cp\u003eSupplemental Table 3: MS lipidomics (lipid species)\u003c/p\u003e","description":"","filename":"TableS3LipidSpecies.csv","url":"https://assets-eu.researchsquare.com/files/rs-6074097/v1/120335597fc439815b6706e0.csv"},{"id":78735595,"identity":"44ad4c5e-b0b9-4925-9778-61d61ad624c7","added_by":"auto","created_at":"2025-03-18 08:09:02","extension":"csv","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":55201,"visible":true,"origin":"","legend":"\u003cp\u003eSupplemental Table 4: MS lipidomics (lipid class)\u003c/p\u003e","description":"","filename":"TableS4LipidClass.csv","url":"https://assets-eu.researchsquare.com/files/rs-6074097/v1/b7465023da298adc968c6124.csv"},{"id":78735596,"identity":"fa9cffb7-cf36-489c-81bd-498c100bd494","added_by":"auto","created_at":"2025-03-18 08:09:02","extension":"csv","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":44466,"visible":true,"origin":"","legend":"\u003cp\u003eSupplemental Table 5: RNAseq GO terms normal old vs young\u003c/p\u003e","description":"","filename":"TableS5GOtermsNormalOldvsYoung.csv","url":"https://assets-eu.researchsquare.com/files/rs-6074097/v1/66cdb50f4f07248f40d092b4.csv"},{"id":78735598,"identity":"4e24d1e6-7f28-4e9b-a176-94f3fc45fc13","added_by":"auto","created_at":"2025-03-18 08:09:02","extension":"csv","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":2412,"visible":true,"origin":"","legend":"\u003cp\u003eSupplemental Table 6: Sensitivity analysis results\u003c/p\u003e","description":"","filename":"TableS6SensitivityAnalysis.csv","url":"https://assets-eu.researchsquare.com/files/rs-6074097/v1/6c874a09f570864e313d26bd.csv"},{"id":78735583,"identity":"6a316c91-d2ec-433b-9363-bc11f8a7f14d","added_by":"auto","created_at":"2025-03-18 08:08:57","extension":"csv","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":553367,"visible":true,"origin":"","legend":"\u003cp\u003eSupplemental Table 7: RNAseq GO terms absent/present age related changes\u003c/p\u003e","description":"","filename":"TableS7GOtermsavoidableorunavoidableagingchanges.csv","url":"https://assets-eu.researchsquare.com/files/rs-6074097/v1/040693a396276e9ffd108b5b.csv"},{"id":78735591,"identity":"868bc7d7-4887-459a-902f-b8efd635da36","added_by":"auto","created_at":"2025-03-18 08:09:01","extension":"csv","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":1025631,"visible":true,"origin":"","legend":"\u003cp\u003eSupplemental Table 8: RNAseq GO terms exercise response quadrants\u003c/p\u003e","description":"","filename":"TableS8GOtermexercisequadrants.csv","url":"https://assets-eu.researchsquare.com/files/rs-6074097/v1/b39009296bf459c315dee5f0.csv"},{"id":78735597,"identity":"ad7282c0-d731-41af-bc59-f5034ac1a4e2","added_by":"auto","created_at":"2025-03-18 08:09:02","extension":"docx","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":1842098,"visible":true,"origin":"","legend":"\u003cp\u003eExtended Data Figures\u003c/p\u003e","description":"","filename":"SupplementalFiguresv5.docx","url":"https://assets-eu.researchsquare.com/files/rs-6074097/v1/21bffc4e6bc1d0131cd6d6aa.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Delayed molecular aging, preservation of energy metabolism and enhanced exercise response in exercise-trained human muscle","fulltext":[{"header":"Main","content":"\u003cp\u003eAging is associated with widespread molecular and functional changes, contributing to a decline in physical resilience and metabolic health\u003csup\u003e1,2\u003c/sup\u003e. Exercise is widely recognized as a powerful intervention to promote healthy aging, improving mobility, cardiovascular function, and metabolic efficiency\u003csup\u003e3\u0026ndash;5\u003c/sup\u003e. Exercise training represents the cumulative adaptation of the body to sequential bouts of acute exercise, during which a dynamic molecular choreography unfolds\u003csup\u003e6,7\u003c/sup\u003e. Despite its established benefits, the molecular mechanisms by which exercise interacts with aging remain incompletely understood. In particular, it is unclear to what extent prolonged exercise training can offset the molecular hallmarks of aging in skeletal muscle, one of the most adaptable tissues in the body. Moreover, how older individuals respond to acute exercise at the molecular level is poorly characterized, let alone if this response depends on current physical fitness. While acute exercise is known to elicit transcriptional, metabolic, and lipidomic changes\u003csup\u003e6,7\u003c/sup\u003e, whether these responses vary across different fitness levels in older adults\u0026mdash;and how they compare to younger individuals\u0026mdash;remains an open question. Here, we address these gaps to understand how physical fitness shapes the aging trajectory which is essential for optimizing interventions to preserve muscle function and resilience across the lifespan.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAging is marked by decreased energy metabolism independent of habitual physical activity\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo investigate the molecular effects of an acute exercise bout across physical fitness levels during aging, we studied a cohort of 11 young and 36 older adults. The older group was further categorized as exercise-trained (n=16), normally active (n=15), or physically impaired (n=5) based on self-reported physical activity levels and a short physical performance battery test (Fig. 1a, Extended Data Fig. 1a-c). Importantly, young and normally active older adults displayed similar daily step counts and time spent in high-intensity activity (Fig. 1a, Extended Data Fig. 1b,c)\u0026mdash;which was determined after group inclusion\u0026mdash;suggesting that differences observed between these groups are likely due to aging rather than sedentary behavior that often accompanies aging. Trained and physically impaired older adults showed distinct activity patterns, also reflected in BMI (Fig. 1a, Extended Data Fig. 1a-c). All participants completed a 1-hour exercise bout at 50% of their maximum capacity on a stationary bike\u003csup\u003e8\u0026ndash;10\u003c/sup\u003e. Muscle biopsies were collected before (baseline) and directly after exercise and analyzed using multimodal omics: (i) RNAseq for \u0026gt;24,000 transcripts, (ii) UPLC-HRMS metabolomics for 135 annotated metabolites, and (iii) UPLC-HRMS lipidomics for 1,383 lipids spanning 37 lipid classes (Extended Data Fig. 1d, Supplemental Tables 1-4). Statistical analyses took individual and sex into account.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo explore this molecular compendium of data, we first compared young and normally active older adults at baseline across transcriptomics, metabolomics, and lipidomics. Partial Least Squares Discriminant Analysis (PLS-DA) revealed clear age-related differences in all three datasets (Fig. 1b-d). Among the 1,628 upregulated genes with aging, we observed enrichment of histamine response, microtubule organization, and splicing factors. Strikingly, in the 1,106 downregulated genes we identified reduced expression of genes strongly linked to cellular respiration and energy metabolism (Extended Data Fig. 2a,b, Fig. 1e, Supplemental Table 5). This is consistent with prior findings of reduced mitochondrial mass, respiration, and NAD\u003csup\u003e+\u003c/sup\u003e levels observed with aging in this cohort\u003csup\u003e8,9\u003c/sup\u003e. Metabolomic analysis confirmed a decline in NAD\u003csup\u003e+\u003c/sup\u003e pathway metabolites among the top 10 age-altered compounds (Extended Data Fig. 2c). Lipidomic analysis similarly highlighted energy storage increases, with several triglycerides accumulating among the top 20 age-altered lipid species (Extended Data Fig. 2d), aligning with previous age-related plasma\u003csup\u003e10\u003c/sup\u003e and muscle\u003csup\u003e11\u003c/sup\u003e studies from this cohort.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMost normal age-related differences are absent in physically trained older adults\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe next examined how the molecular differences between young and old adults compared to exercise-trained and physically impaired older adults. To do this, we ranked transcripts based on how their expression followed a \u0026quot;muscle health\u0026rdquo; trend across the four groups: young adults, trained older adults, normally active older adults, and physically impaired older adults. By comparing these trends to age-related changes, we assessed how closely health and aging intersect (Fig. 2a). Strikingly, transcriptional changes with normal aging were strongly correlated with those following the health trend (Pearson\u0026rsquo;s r = 0.68, p \u0026lt; 2.2e-16), suggesting that exercise-trained older adults did not display the transcriptional shifts typically associated with aging in muscle (Fig. 2a).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo explore this further, we categorized age-related transcriptional changes into two groups: (i) genes altered with aging but that maintained youthful expression in trained older adults (Fig. 2b, yellow/green), and (ii) genes consistently altered across all older adults compared to young individuals (Fig. 2b, blues). Remarkably, over half of the transcriptional differences seen in aging\u0026mdash;both upregulated (Fig. 2c) and downregulated (Fig. 2d)\u0026mdash;were absent in trained older adults, underscoring the benefits of exercise training. Sensitivity analyses confirmed that 45\u0026ndash;62% of aging-related changes were absent in trained individuals, even when varying significance thresholds or excluding the physically impaired group (Supplemental Table 6). Gene Ontology analysis (Supplemental Table 7) revealed the strongest enrichment was in genes depleted in older adults but not in the exercise-trained older adults (Fig. 2e, green). These were particularly linked to cellular respiration and energy production, e.g. \u003cem\u003eNDUFS1\u003c/em\u003e (Fig. 2f, upper panel) and \u003cem\u003eCOX5A\u003c/em\u003e (Fig. 2f, lower panel). In contrast, the aging-related changes that occurred in all age groups, irrespective of training status, included the upregulation of synaptic transmission genes such as \u003cem\u003ePCDH8\u003c/em\u003e (Fig. 2g, upper panel) and \u003cem\u003eUNC13C\u003c/em\u003e (Fig. 2g, lower panel) and the downregulation of WNT signaling pathway genes such as \u003cem\u003eDAAM2\u003c/em\u003e (Fig. 2h, upper panel) and \u003cem\u003eCTR9\u003c/em\u003e (Fig. 2h, lower panel), pointing to altered developmental processes. Nonetheless, these findings suggest that the most prominent transcriptional change in aging\u0026mdash;reduced expression of respiration genes (Fig. 1e)\u0026mdash;is also the most \u0026lsquo;preventable\u0026rsquo; change in the sense that it is absent in trained older adults (Fig. 2e).\u003c/p\u003e\n\u003cp\u003eWe next investigated transcriptional changes unique to either a physically active lifestyle (trained) or physically impaired aging (Fig. 3a). In trained older adults, enriched gene pathways were predominantly metabolic, including increased lipid metabolism and decreased energy metabolism genes (Fig. 3b, yellow and orange, respectively; Supplemental Table 7). In contrast, physically impaired older adults exhibited increased immune response pathways and depletion of mitochondrial and respiration-related genes (Fig. 3b, mauve and purple, respectively; Supplemental Table 7). Notably, the mitochondrial gene depletion encompassed large and small mitochondrial ribosomal proteins, such as \u003cem\u003eMRPS16, MRPL39, MRPL35, MRPL34,\u003c/em\u003e and \u003cem\u003eMRPS18C\u003c/em\u003e (Fig. 3c-g), which we have previously linked to aging\u003csup\u003e12,13\u003c/sup\u003e. These findings highlight distinct molecular pathways associated with a physically active \u0026nbsp;lifestyle (trained) and physical impairment, and further underscore the critical role of energy metabolism and mitochondrial function in healthy aging\u003csup\u003e14\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMolecular responses to exercise across age and fitness levels\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eExercise interventions and the stress of aging itself are known to uncover latent molecular responses and adaptive pathways\u003csup\u003e6,15\u003c/sup\u003e. Building on our understanding of the cohort at baseline, we next examined how an acute, one-hour bout of sub-maximal exercise rewires the molecular landscape across the four groups. Using multi-omics profiling post-exercise, PLS-DA revealed robust molecular responses across transcriptomic, lipidomic, and metabolomic levels in all groups (Fig. 4a) allowing to explore group responses across the omics levels (Fig. 4b). Older and young individuals with similar activity levels showed comparable proportions of upregulated (8.6% vs. 8.3%) and downregulated (6.6% vs. 5.0%) transcripts in response to exercise (Fig. 4c). However, older adults exhibited a stronger metabolomic and lipidomic response compared to young with similar activity levels, with greater depletions (4.4% vs. 0.7% metabolites; 1.4% vs. 0.5% lipids) and accumulations (14.8% vs. 3.0% metabolites; 21.9% vs. 5.4% lipids) of molecules (Fig. 4c).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eExploring these metabolomic and lipidomic responses to acute exercise further revealed distinct patterns across age and fitness groups. In the metabolome, all groups showed typical exercise-related changes, such as acylcarnitine accumulation\u003csup\u003e16\u003c/sup\u003e, as well as glutathione and AICAR depletion, amongst others (Extended Data Fig. 3a-e, Extended Data Fig. 3f-h). In the lipidome, significant species-level changes were observed in all groups (Extended Data Fig. 4a-d). Unlike young individuals, however, older groups (trained, normally active, and physically impaired) showed enrichment of specific lipid classes (Supplemental Table 4), including dihexosylceramide (Hex2Cer[d]) accumulation (Extended Data Fig. 4e,f). Together these findings suggest that both metabolomic and lipidomic changes in response to exercise are more pronounced in older adults, irrespective of physical fitness level, contrasting with a more dampened response in young individuals.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFollowing this, we turned to the transcriptomic responses and cross-referenced younger individuals with the three older adult groups (Fig. 4d-f). Remarkably, a stepwise pattern emerged, with trained older adults exhibiting exercise responses more similar to young adults (Pearson\u0026rsquo;s r=0.451, p\u0026lt;2e-16), more so than older adults with normal physical activity (Pearson\u0026rsquo;s r=0.394, p\u0026lt;2e-16), while physically impaired older adults had the most divergent response (Pearson\u0026rsquo;s r=0.263, p\u0026lt;2e-16) (Fig. 4g). Observing this, we next explored in more detail the factors explaining the dissimilarity in exercise response of normally active older individuals compared to young. We dissected the responses into quadrants, including co-upregulated (Q1), co-downregulated (Q3) or discordant (Q2 and Q4) (Fig. 4h, scatterplot). Here, we found co-upregulated genes to encompass stimulus and stress responses such as an upregulation of interleukin genes \u003cem\u003eIL\u003c/em\u003e-6, \u003cem\u003eILB\u003c/em\u003e, and \u003cem\u003eIL1RN\u003c/em\u003e, and increased \u003cem\u003eTNF\u003c/em\u003e, \u003cem\u003eSELE\u003c/em\u003e, and \u003cem\u003eFOS\u003c/em\u003e (Fig. 4h Q1\u003cstrong\u003e,\u0026nbsp;\u003c/strong\u003eFig. 4i,\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eExtended Data Figure 5a-c,\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eSupplemental Table 8). The stimulus and stress response activation was also clearly observed in both trained and physically impaired older adults, with fold changes of these genes following the same directionality (Fig. 4i, Extended Data Figure 5a-c,\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eSupplemental Table 8). Co-downregulated genes included the suppression of calcium channel and ion transport (Fig. 4h, Q3,\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eSupplemental Table 8). Intriguingly, the discordant quadrants revealed that older individuals possessed an additional immune-related stress response compared to young following exercise illustrated by the activation of \u003cem\u003eIL7R\u003c/em\u003e and \u003cem\u003eVTN\u003c/em\u003e (Fig. 4h, Q2,\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eExtended Data Figure 5d-e,\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eSupplemental Table 8), and a suppression of genes related to collagen binding (e.g. \u003cem\u003eITGA2\u003c/em\u003e) and molecular transport pathways (Fig. 4h, Q4,\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eExtended Data Figure 5f,\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eSupplemental Table 8).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eInvestigating the transcriptomic response to acute exercise in trained and physically impaired older adults in a similar manner, i.e. assessing the concordant and discordant quadrant\u0026rsquo;s GO term enrichments, revealed additional insights (Extended Data Figure 6a,b). Indeed, trained older adults displayed a suppression\u0026mdash;rather than activation\u0026mdash;of certain immune response genes such as mast cell related \u003cem\u003eCD226\u003c/em\u003e gene and chemotaxis genes \u003cem\u003eNMUR1\u003c/em\u003e and \u003cem\u003eRRH\u003c/em\u003e, while physically impaired older adults increased nitric oxide transport gene expression such as with the \u003cem\u003eHBA1\u003c/em\u003e gene, and underwent a suppression of cell-cell signaling transcription such as with \u003cem\u003eFGF12\u003c/em\u003e and \u003cem\u003eMERTK\u003c/em\u003e (Extended Data Figure 6c-h, Supplemental Table 8). However, these changes were not as pronounced as the marked immune and cellular stress response displayed by all four groups following exercise (Figure 4h,i, Extended Data Figure 5a-c, Extended Data Figure 6a,b). Notably, the ability to elicit this stress response following acute exercise was lessened in impaired older adults and heightened in the trained (Fig. 4i). These findings reveal that physical fitness level significantly shapes the molecular response to exercise with aging, and that a stronger stress response following acute exercise is associated with healthier aging.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eOur study disentangles the molecular interplay between physical fitness, aging and the response to acute exercise in trained, normally active, and physically impaired older adults. By performing transcriptomics, lipidomics, and metabolomics on skeletal muscle from young and older adults with varying fitness levels, we found that aging is marked by a loss of transcripts encoding for cellular respiration, even in older individuals with daily physical activity levels comparable to young adults. Remarkably, exercise-trained older adults did not exhibit this decline, with over half of the transcriptional changes seen in aging being absent. Specifically, exercise trained older adults retained a youthful transcriptional profile with key energy metabolism and cellular respiration genes maintaining expression levels similar to young individuals, and responded to an acute bout of exercise also similarly to young adults. Taken together, our findings underscore the remarkable capacity of exercise training to mitigate age-associated molecular changes in skeletal muscle, and provides the molecular basis for previous studies that have highlighted physical fitness level as a potent modulator of health, capable of preserving energy metabolism levels in aging\u003csup\u003e8\u0026ndash;10\u003c/sup\u003e. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOur findings offer novel insights into aging biology in humans. Dissecting baseline molecular differences between young and older adults revealed two distinct categories of age-related changes: those related to physical activity and \u0026quot;preventable\u0026quot; by exercise training, defined as those absent in trained older adults and \u0026quot;unavoidable\u0026quot; aging changes defined as those shared across all older groups. Notably, it should be the ambition of the geroscience field to focus development of therapeutics on \u0026ldquo;unavoidable\u0026rdquo; age-related changes, while promoting lifestyle alterations to address the \u0026ldquo;preventable\u0026rdquo; age-related changes. In this regard, we aimed to map out the unavoidable molecular changes occurring with aging. Interestingly, these changes did not possess a single dominant pathway or biological process as enriched, suggesting stochastic forces. Finally, our results demonstrate that an enhanced immune and stress response to acute exercise correlates with improved resilience to physical challenges. This raises concerns that longevity interventions targeting immune suppression\u0026mdash;such as recently described \u003cem\u003eIL-11\u003c/em\u003e inhibitors or well-known anti-inflammatory agents\u003csup\u003e17,18\u003c/sup\u003e\u0026mdash;might inadvertently reduce functional resilience in humans. Our findings illuminate the intricate relationship between aging and exercise in human muscle, advocating for exercise as a foundational strategy to promote healthy aging and resilience, and provides a robust resource to further explore molecular regulators of aging and exercise.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cem\u003eHuman subjects and procedures\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eForty-seven participants, including 11 young and 36 older adults were recruited in the community of Maastricht and its surroundings through advertisements at Maastricht University, in local newspapers, supermarkets, and at sports clubs, to collect both \u0026ndash;before and \u0026ndash;after measures of an exercise intervention. The study protocol was approved by the institutional Medical Ethical Committee and conducted in agreement with the declaration of Helsinki. All participants provided their written informed consent, and the study was registered at clinicaltrials.gov with identifier NCT03666013. Physiological data from this clinical trial has been reported in our previous study as part of a different analysis\u003csup\u003e8\u003c/sup\u003e, and the current study uses the same individuals for whom a pre and post-exercise biopsy was available.\u0026nbsp;Prior to inclusion, all subjects underwent a medical screening that included a physical examination by a physician and an assessment of physical function using the Short Physical Performance Battery (SPPB), comprised of a standing balance test, a 4-m walk test, and a chair-stand test. After the screening procedure, participants were assigned to the following study groups: Young individuals with normal physical activity (20 \u0026ndash; 30 years), older adults with normal physical activity (65 \u0026ndash; 80 years), physically trained older adults (65 \u0026ndash; 80 years) and physically impaired older adults (65 \u0026ndash; 80 years). Participants were considered normal, physically active if they completed no more than one structured exercise session per week. Participants were considered trained if they engaged in at least 3 structured exercise sessions of at least 1 hour each per week for an uninterrupted period of more than one year. Participants were classified as older adults with impaired physical function in case of an SPPB score of \u0026le; 9. The SPPB score was calculated according to the cut-off points determined by (Guralnik et al. 1994)\u003csup\u003e19\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eExercise intervention\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eParticipants performed a 1-h submaximal exercise bout in the fasted state on an electronically braked cycle ergometer, at 50% of their W\u003csub\u003emax\u003c/sub\u003e as measured during a maximal aerobic cycling test\u003csup\u003e8\u003c/sup\u003e. Participants were instructed to pedal at a controlled cadence between 60 and 70 revolutions per minute.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMuscle biopsy\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAt 9 AM, after an overnight fast from 10 PM the preceding evening, and immediately following the acute exercise bout, muscle biopsies were taken from the \u003cem\u003em. vastus lateralis\u003c/em\u003e under local anesthesia (1.0% lidocaine without epinephrine) according to the Bergstr\u0026ouml;m method\u003csup\u003e20\u003c/sup\u003e. The muscle biopsies were immediately frozen in melting isopentane and stored at \u0026ndash;80\u0026deg;C until further analysis.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eHabitual physical activity\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eHabitual physical activity was determined in all participants using an ActivPAL monitor (PAL Technologies, Glasgow, Scotland) for a consecutive period of 5 days, including two weekend days. Besides the total amount of steps per day, the total stepping time was calculated in proportion to waking time, determined according to (van der Berg \u003cem\u003eet al.\u003c/em\u003e 2016)\u003csup\u003e21\u003c/sup\u003e. Stepping time (i. e., physical activity) was then further classified into high-intensity physical activity (HPA; minutes with a step frequency \u0026gt; 110 steps/min in proportion to waking time) and lower-intensity physical activity (LPA; minutes with a step frequency \u0026le; 110 steps/min in proportion to waking time)\u003csup\u003e22\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eRNA sequencing: Isolation of mRNA\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eHuman muscle tissues were homogenized with a 5 mm steel bead using a TissueLyser II (QIAGEN) for 5 min at frequency of 30 times/second. RNA was extracted according to the instructions of the RNaesy Mini Kit (QIAGEN). Contaminating genomic DNA was removed using RNase-Free DNase (QIAGEN). RNA was quantified with a NanoDrop 2000 spectrophotometer (Thermo Scientific; Breda, The Netherlands) and stored at -80\u0026deg;C until use. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eRNA sequencing: Library Preparation\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eRNA libraries were prepared and sequenced with the Illumina platform by Genome Scan (Leiden, The Netherlands). The NEBNext Ultra II Directional RNA Library Prep Kit for Illumina was used to process the sample(s). The sample preparation was performed according to the protocol \u0026quot;NEBNext Ultra II Directional RNA Library Prep Kit for Illumina\u0026quot; (NEB #E7760S/L). Briefly, mRNA was isolated from total RNA using the oligo-dT magnetic beads. After fragmentation of the mRNA, cDNA synthesis was performed. This was used for ligation with the sequencing adapters and PCR amplification of the resulting product. The quality and yield after sample preparation was measured with the Fragment Analyzer. The size of the resulting products was consistent with the expected size distribution (a broad peak between 300-500 bp). Clustering and DNA sequencing using the NovaSeq6000 was performed according to manufacturer\u0026apos;s protocols. A concentration of 1.1 nM of DNA was used. NovaSeq control software NCS v1.6 was used.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eRNA sequencing: Read Mapping\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAnalysis for read mapping was performed in R v4.1.0 and Bioconductor v3.13. Reads were subjected to quality control FastQC\u003csup\u003e23\u003c/sup\u003e v0.11.15 and trimmed using Trimmomatic v0.36 (Bolger et al., 2014) and aligned using HISAT2 v2.1.0 (Kim et al., 2015) to the GRCh38 (v94) human genome reference assembly. Counts were obtained using HTSeq (v0.11.0, default parameters) (Anders et al., 2015) using the corresponding GTF taking into account the directions of the reads. Statistical analyses were performed using the edgeR v3.34.1 (Robinson et al., 2010) and limma/voom v 3.48.3\u003csup\u003e28\u003c/sup\u003e R packages. All genes with more than 2 counts in at least 4 of the samples were kept. Count data were transformed to log2-counts per million (logCPM), normalized by applying the trimmed mean of M-values method \u003csup\u003e27\u003c/sup\u003e and precision weighted using voom (Law et. al., 2014). Genes were reannotated using the Ensembl genome database (v104) and the biomaRt package\u003csup\u003e30,31\u003c/sup\u003e\u003cstrong\u003e.\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMetabolite and Lipid extraction\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eMetabolomics and lipidomics were performed at the Core Facility Metabolomics at the Amsterdam UMC, essentially as described\u003csup\u003e13\u003c/sup\u003e\u003csup\u003e,\u003c/sup\u003e\u003csup\u003e32\u003c/sup\u003e\u003csup\u003e,\u003c/sup\u003e\u003csup\u003e33\u003c/sup\u003e. In a 2 mL tube, the following amounts of internal standard dissolved in MilliQ were added to each sample of approximately 3-5 mg of freeze-dried muscle tissue: adenosine-\u003csup\u003e15\u003c/sup\u003eN\u003csub\u003e5\u003c/sub\u003e-monophosphate (5 nmol), adenosine-\u003csup\u003e15\u003c/sup\u003eN\u003csub\u003e5\u003c/sub\u003e-triphosphate (5 nmol), D\u003csub\u003e4\u003c/sub\u003e-alanine (0.5 nmol), D\u003csub\u003e7\u003c/sub\u003e-arginine (0.5 nmol), D\u003csub\u003e3\u003c/sub\u003e-aspartic acid (0.5 nmol), D\u003csub\u003e3\u003c/sub\u003e-carnitine (0.5 nmol), D\u003csub\u003e4\u003c/sub\u003e-citric acid (0.5 nmol), \u003csup\u003e13\u003c/sup\u003eC\u003csub\u003e1\u003c/sub\u003e-citrulline (0.5 nmol), \u003csup\u003e13\u003c/sup\u003eC\u003csub\u003e6\u003c/sub\u003e-fructose-1,6-diphosphate (1 nmol), guanosine-\u003csup\u003e15\u003c/sup\u003eN\u003csub\u003e5\u003c/sub\u003e-monophosphate (5 nmol), guanosine-\u003csup\u003e15\u003c/sup\u003eN\u003csub\u003e5\u003c/sub\u003e-triphosphate (5 nmol), \u003csup\u003e13\u003c/sup\u003eC\u003csub\u003e6\u003c/sub\u003e-glucose (10 nmol), \u003csup\u003e13\u003c/sup\u003eC\u003csub\u003e6\u003c/sub\u003e-glucose-6-phosphate (1 nmol), D\u003csub\u003e3\u003c/sub\u003e-glutamic acid (0.5 nmol), D\u003csub\u003e5\u003c/sub\u003e-glutamine (0.5 nmol), D\u003csub\u003e5\u003c/sub\u003e-glutathione (1 nmol), \u003csup\u003e13\u003c/sup\u003eC\u003csub\u003e6\u003c/sub\u003e-isoleucine (0.5 nmol), D\u003csub\u003e3\u003c/sub\u003e-lactic acid (1 nmol), D\u003csub\u003e3\u003c/sub\u003e-leucine (0.5 nmol), D\u003csub\u003e4\u003c/sub\u003e-lysine (0.5 nmol), D\u003csub\u003e3\u003c/sub\u003e-methionine (0.5 nmol), D\u003csub\u003e6\u003c/sub\u003e-ornithine (0.5 nmol), D\u003csub\u003e5\u003c/sub\u003e-phenylalanine (0.5 nmol), D\u003csub\u003e7\u003c/sub\u003e-proline (0.5 nmol), \u003csup\u003e13\u003c/sup\u003eC\u003csub\u003e3\u003c/sub\u003e-pyruvate (0.5 nmol), D\u003csub\u003e3\u003c/sub\u003e-serine (0.5 nmol), D\u003csub\u003e6\u003c/sub\u003e-succinic acid (0.5 nmol), D5-tryptophan (0.5 nmol), D\u003csub\u003e4\u003c/sub\u003e-tyrosine (0.5 nmol), D\u003csub\u003e8\u003c/sub\u003e-valine (0.5 nmol). For lipidomics, the following internal standards dissolved in 50:50 MeOH:CHCl\u003csub\u003e3\u003c/sub\u003e were added: DG(14:0)\u003csub\u003e2\u003c/sub\u003e, TG(14:0)\u003csub\u003e3\u003c/sub\u003e, CE(16:0)\u003csub\u003e-d7\u003c/sub\u003e, PC(14:0)\u003csub\u003e2\u003c/sub\u003e, PS(14:0)\u003csub\u003e2\u003c/sub\u003e, PE(14:0)\u003csub\u003e2\u003c/sub\u003e, PA(14:0)\u003csub\u003e2\u003c/sub\u003e, ST(17:0), PI(8:0)\u003csub\u003e2\u003c/sub\u003e, LPE(14:0), LPC(14:0), LPA(14:0), SPH(d17:1), SM(12:0), SPH(d17:0), S1P(d17:1), S1P(d17:0), LacCer(d18:1/12:0), GlcCer(d18:1/12:0), Cer(d18:1/12:0), C1P(d18:1/12:0), Cer(d18:1/25:0). After adding the internal standard mixtures, a 5\u0026thinsp;mm stainless-steel bead and polar phase solvents (for a total of 500 \u0026micro;L MilliQ and 500 \u0026micro;L MeOH) were added and samples were homogenized using a TissueLyser II (Qiagen, Hilden, Germany) for 5\u0026thinsp;min at a frequency of 30 times/sec. Chloroform was added for a total of 1 mL to each sample before thorough mixing. Samples were then centrifuged for 10 minutes at 18.000\u003cem\u003eg\u003c/em\u003e. The top and bottom layer were each transferred to a new 1.5 mL tube for separate processing.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMetabolomics analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe top layer, containing polar metabolites, was dried using a vacuum concentrator at 60\u0026deg;C. Dried samples were reconstituted in 100 \u0026micro;L 3:2 (v/v) MeOH:MilliQ. Metabolites were analyzed using a Waters Acquity ultra-high performance liquid chromatography system coupled to a Bruker Impact II\u0026trade; Ultra-High Resolution Qq-Time-Of-Flight mass spectrometer. Samples were kept at 12\u0026deg;C during analysis and 5 \u0026micro;L of each sample was injected. Chromatographic separation was achieved using a Merck Millipore SeQuant ZIC-cHILIC column (PEEK 100 x 2.1 mm, 3 \u0026micro;m particle size). Column temperature was held at 30\u0026deg;C. Mobile phase consisted of (A) 1:9 (v/v) ACN:MilliQ and (B) 9:1 (v/v) ACN:MilliQ, both containing 5 mmol/L ammonium acetate. Using a flow rate of 0.25 mL/min, the LC gradient consisted of: 100% B for 0-2 min, reach 0% B at 28 min, 0% B for 28-30 min, reach 100% B at 31 min, 100% B for 31-32 min. Column re-equilibration is achieved at a flow rate of 0.4 mL/min at 100% B for 32-35 min. MS data were acquired using negative and positive ionization in full scan mode over the range of m/z 50-1200. Data were analyzed using Bruker TASQ software version 2.1.22.3. All reported metabolite intensities were normalized to dry tissue weight, as well as to internal standards with comparable retention times and response in the MS. General repeatability of metabolite analysis was assessed for each metabolite using repeated measurements of a pooled quality control (QC) sample. Metabolite identification has been based on a combination of accurate mass, (relative) retention times and fragmentation spectra, compared to the analysis of a library of standards in separate experiments not described here.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eLipidomics analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe bottom layer of the extraction, containing lipids, was dried under a stream of nitrogen at 30\u0026deg;C and reconstituted in 100\u0026nbsp;\u0026micro;L\u0026nbsp;50:50 MeOH:CHCl\u003csub\u003e3\u003c/sub\u003e. The UPLC system consisted of an Ultimate 3000 binary HPLC pump, a vacuum degasser, a column temperature controller, and an auto sampler (Thermo Scientific, Waltham, MA, USA). Samples were run in normal and reverse phase and positive and negative ionization. For normal phase, 2\u0026mu;L of each sample was injected into the system. The normal phase system consisted of a Lichrospher Si 60, 2 x 250 mm silica 100 \u0026Aring; column, 5 \u0026micro;m particle diameter (Merck, Germany), the column temperature was maintained at 25\u0026deg;C. Lipids were separated using a linear gradient between solution B (CHCl\u003csub\u003e3\u003c/sub\u003e/MeOH, 97:3 v/v) and solution A (MeOH/MilliQ, 85:15, v/v). Solution A contained 0.0125% formic acid and 3.35 mmol/l ammonia per liter of eluent. Solution B contained 0.0125% formic acid per liter. The gradient (0.3 ml/min) was as follows: 0-1 min 10%A, 1\u0026ndash;4 min 10%A\u0026ndash;20%A, 4\u0026ndash;12 min 20%A\u0026ndash;85% A, 12\u0026ndash;12.1 min, 85%A\u0026ndash;100% A, 12.1\u0026ndash;14.0 min 100% A, 14-14.1 min 100%A\u0026ndash;10%A and 14.1\u0026ndash;15 min equilibration with 10% A. All gradient steps were linear, and the total analysis time, including the equilibration, was 15 min. For reversed phase separation, 5 \u0026mu;L of each sample was injected onto a Waters HSS T3 column (150 x 2.1 mm, 1.8 \u0026mu;m particle size). Column temperature was held at 60\u0026deg;C. Mobile phase consisted of (A) 4:6 (v/v) MeOH:MilliQ and B 1:9 (v/v) MeOH:IPA, both containing 0.1% formic acid and 10 mmol/L ammonia. Using a flow rate of 0.4 mL/min, the LC gradient consisted of: Dwell at 100% A at 0 min, ramp to 80% A at 1 min, ramp to 0% A at 16 min, dwell at 0% A for 16-20 min, ramp to 100% A at 20.1 min, dwell at 100% A for 20.1-21 min. A Q Exactive Plus (Thermo Scientific) mass spectrometer was used in the negative and positive electrospray ionization mode. In both ionization modes, mass spectra of the lipid species were obtained by continuous scanning from m/z 150 to m/z 2000 with a resolution of 280.000. Nitrogen was used as the nebulizing gas. The spray voltage used was 2500 V (-) and 3500 V (+), and the capillary temperature was 256\u0026deg;C. S-lens RF level: 50, Auxiliary gas: 10, Auxiliary gas temperature 300\u0026deg;C, Sheath gas: 50, Sweep cone gas: 2.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eLipidomics: bioinformatics for lipid identification\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eLipid identification was performed at the Core Facility Metabolomics at the Amsterdam UMC, essentially as described\u003csup\u003e33\u003c/sup\u003e. The raw LC/MS data were converted to mzXML format using MSConvert\u003csup\u003e34\u003c/sup\u003e. The dataset was processed using an in-house developed metabolomics pipeline written in the R programming language (http://www.r-project.org)\u003csup\u003e35\u003c/sup\u003e. In brief, it consisted of the following steps: (1) pre-processing using the R package XCMS\u003csup\u003e36\u003c/sup\u003e with minor changes to some functions in order to better suit the Q Exactive data; notably, the definition of noise level in centWave was adjusted and the stepsize in fillPeaks (2) identification of metabolites using an in-house database of (phospho)lipids, with known internal standards indicating the position of most of the lipid clusters, matching m/z values within 3 ppm deviation, (3) isotope correction to obtain deconvoluted intensities for overlapping peak groups, (4) normalization on the intensity of the internal standard for lipid classes for which an internal standard was available (with normalization on the intensity of PE(14:0)\u003csub\u003e2\u003c/sub\u003e for lipid classes for which no internal standard was present) and dry tissue weight. For quantifying abundances of lipid classes, the summed abundances of the individual lipid species from the relevant class were used. General repeatability of lipid analysis was assessed for each lipid using repeated measurements of a pooled quality control (QC) sample. Lipid class identification has been based on accurate mass, fragmentation analysis, relative retention times and ion mobility, compared to the analysis of relevant standards in separate experiments not described here.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eExercise response quadrants\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eExercise response quadrants (log2 fold changes) of each of the three older adult groups were compared to young adults and dissected into quadrants as follows: Q1 (co-upregulated); log2 fold change exercise response \u0026gt;1 for both young and old). Q3 (co-downregulated); log2 fold change exercise response \u0026lt;-1 for both young and old. Q2 (discordant response); log2 fold change exercise response \u0026gt;1 old, \u0026lt;0 young. Q4 (discordant response); log2 fold change exercise response \u0026lt;-1 old, \u0026gt;0 young).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eStatistics and reproducibility\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSample size determination for the human studies are described in the original publications\u003csup\u003e8,37\u003c/sup\u003e and were done to accommodate the goals of the clinical trials (clinicaltrials.gov identifier NCT03666013). No data was intentionally excluded from the analyses (if mRNA, metabolites, or lipids are missing this is due to them not being detected in the tissue/sample and if samples are omitted this is due to the sample being lost during sample preparation). Sample collection was not blinded due to obvious differences in subject groups, however, the downstream processing of all \u0026ndash;omics samples was blinded and randomized. Un-blinding and de-randomization occurred at the final step for statistical analysis and interpretation. Unless otherwise noted, analyses were done with R\u003csup\u003e35\u003c/sup\u003e version 3.5.1 and Bioconductor\u003csup\u003e38\u003c/sup\u003e version 3.7. For RNAseq, metabolomics and lipidomics data, dfferential expression was assessed using an empirical Bayes moderated t test within limma\u0026rsquo;s linear model framework using log2 transformed data for the metabolome and lipidome and for RNAseq precision weights estimated by voom\u003csup\u003e28,29\u003c/sup\u003e. Individuals, sex, and before-after exercise were included as covariates in the model. Resulting \u003cem\u003ep\u003c/em\u003e values were corrected for multiple testing using the Benjamini-Hochberg false discovery rate. Data was processed in part with the R package dplyr version 1.0.2\u003csup\u003e39\u003c/sup\u003e. Partial least-squares discriminant analysis (PLS-DA) was performed using the R package MixOmics version 6.6.2\u003csup\u003e40\u003c/sup\u003e. Networks were constructed and visualized using igraph version 1.2.4.2\u003csup\u003e41\u003c/sup\u003e. Unless implemented through an aforementioned R package or base R graphics, visualization of data was performed using ggplot2 version 3.2.1\u003csup\u003e42\u003c/sup\u003e, ggpubr v 0.2.5\u003csup\u003e43\u003c/sup\u003e, ggrepel version 0.8.1\u003csup\u003e44\u003c/sup\u003e, with colors from RColorBrewer version 1.1-2\u003csup\u003e45\u003c/sup\u003e. For pathway enrichment analyses: Gene ontology (GO) term enrichments were calculated for RNAseq gene lists with a hypergeometric test using the GOstats package (version 2.48.0) in R, or by using the DAVID gene enrichment online resource (https://david.ncifcrf.gov/). Metabolite enrichment analysis was performed using the MetabAnalyst v6.0 online tool (https://www.metaboanalyst.ca/) using a global test\u003csup\u003e46\u003c/sup\u003e of significance on KEGG human metabolic pathways (Dec. 2023).\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cem\u003eData availability\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eData supporting the conclusions of our study are available as supplementary materials accompanying this article as summary statistics comparing groups (see Supplemental Data Tables). Physiological data from the cohort has been reported in our previous study as part of a different analysis\u003csup\u003e8\u003c/sup\u003e. All other data supporting the findings of this study are available either in additional Supplemental Data Tables or from the corresponding authors upon reasonable request.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCode availability\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eCode supporting the findings of this study are available from the corresponding authors upon reasonable request.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAcknowledgements\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe project was supported by a Longevity Impetus Grant from Norn Group (to GEJ and RHH). Work in the Houtkooper group is financially supported by funding from the European Union\u0026rsquo;s Horizon Europe research and innovation programme through the MSCA-Doctoral Network NADIS (no. 101073251), and by the Velux Stiftung (no. 1063). GEJ is supported by an AGEM Talent grant. Human interventions were further financed by the TIFN research program Mitochondrial Health (ALWTF.2015.5) and the Netherlands Organization for Scientific Research (NWO).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAuthor contributions\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eG.E.J., P.S., J.H., R.H.H. conceived the study. L.G. performed the human clinical trial and experiments. G.E.J designed and performed the bioinformatics analyses. A.S. and S.W.D. performed molecular extractions for omics preparation. A.J. performed RNAseq mapping and statistical analyses. B.V.S and M.v.W., M.A.T.V., and E.J.M.W. performed the metabolomics and lipidomics analyses. G.S. and F.M.V. reviewed the manuscript. G.E.J., M.K. P.S. R.H.H and J.H. interpreted the results and wrote the manuscript with contributions from all other authors.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCompeting Interests\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests related to this work. The funders had no role in data collection, analysis, or decision to publish or in preparation of the manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eL\u0026oacute;pez-Ot\u0026iacute;n, C., Blasco, M. A., Partridge, L., Serrano, M. \u0026amp; Kroemer, G. The hallmarks of aging. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e153\u003c/strong\u003e, (2013).\u003c/li\u003e\n\u003cli\u003eL\u0026oacute;pez-Ot\u0026iacute;n, C., Blasco, M. A., Partridge, L., Serrano, M. \u0026amp; Kroemer, G. Hallmarks of aging: An expanding universe. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e186\u003c/strong\u003e, 243\u0026ndash;278 (2023).\u003c/li\u003e\n\u003cli\u003eVi\u0026ntilde;a, J., Rodriguez-Ma\u0026ntilde;as, L., Salvador-Pascual, A., Tarazona-Santabalbina, F. J. \u0026amp; Gomez-Cabrera, M. C. Exercise: The lifelong supplement for healthy ageing and slowing down the onset of frailty. \u003cem\u003eJournal of Physiology\u003c/em\u003e (2016). doi:10.1113/JP270536\u003c/li\u003e\n\u003cli\u003eTarazona-Santabalbina, F. J. \u003cem\u003eet al.\u003c/em\u003e A Multicomponent Exercise Intervention that Reverses Frailty and Improves Cognition, Emotion, and Social Networking in the Community-Dwelling Frail Elderly: A Randomized Clinical Trial. \u003cem\u003eJ. Am. Med. Dir. Assoc.\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 426\u0026ndash;433 (2016).\u003c/li\u003e\n\u003cli\u003eVi\u0026ntilde;a, J., Sanchis-Gomar, F., Martinez-Bello, V. \u0026amp; Gomez-Cabrera, M. C. Exercise acts as a drug; The pharmacological benefits of exercise. \u003cem\u003eBr. J. Pharmacol.\u003c/em\u003e \u003cstrong\u003e167\u003c/strong\u003e, 1\u0026ndash;12 (2012).\u003c/li\u003e\n\u003cli\u003eContrepois, K. \u003cem\u003eet al.\u003c/em\u003e Molecular Choreography of Acute Exercise. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e181\u003c/strong\u003e, 1112-1130.e16 (2020).\u003c/li\u003e\n\u003cli\u003eHornburg, D. \u003cem\u003eet al.\u003c/em\u003e Dynamic lipidome alterations associated with human health, disease and ageing. \u003cem\u003eNat. Metab.\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 1578\u0026ndash;1594 (2023).\u003c/li\u003e\n\u003cli\u003eGrevendonk, L. \u003cem\u003eet al.\u003c/em\u003e Impact of aging and exercise on skeletal muscle mitochondrial capacity, energy metabolism, and physical function. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 1\u0026ndash;17 (2021).\u003c/li\u003e\n\u003cli\u003eJanssens, G. E. \u003cem\u003eet al.\u003c/em\u003e Healthy aging and muscle function are positively associated with NAD+ abundance in humans. \u003cem\u003eNat. Aging\u003c/em\u003e (2022). doi:10.1038/s43587-022-00174-3\u003c/li\u003e\n\u003cli\u003eJanssens, G. E. \u003cem\u003eet al.\u003c/em\u003e A metabolomic signature of decelerated physiological aging in human plasma. \u003cem\u003eGeroScience\u003c/em\u003e \u003cstrong\u003e45\u003c/strong\u003e, 3147\u0026ndash;3164 (2023).\u003c/li\u003e\n\u003cli\u003eJanssens, G. E. \u003cem\u003eet al.\u003c/em\u003e A conserved complex lipid signature marks human muscle aging and responds to short-term exercise. \u003cem\u003eNat. Aging\u003c/em\u003e (2024). doi:10.1038/s43587-024-00595-2\u003c/li\u003e\n\u003cli\u003eHoutkooper, R. H. \u003cem\u003eet al.\u003c/em\u003e Mitonuclear protein imbalance as a conserved longevity mechanism. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e497\u003c/strong\u003e, 451\u0026ndash;457 (2013).\u003c/li\u003e\n\u003cli\u003eMolenaars, M. \u003cem\u003eet al.\u003c/em\u003e A Conserved Mito-Cytosolic Translational Balance Links Two Longevity Pathways. \u003cem\u003eCell Metab.\u003c/em\u003e (2020). doi:10.1016/j.cmet.2020.01.011\u003c/li\u003e\n\u003cli\u003eHoutkooper, R. H. \u003cem\u003eet al.\u003c/em\u003e Mitonuclear protein imbalance as a conserved longevity mechanism. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e497\u003c/strong\u003e, 451\u0026ndash;457 (2013).\u003c/li\u003e\n\u003cli\u003eReitzner, S. M. \u003cem\u003eet al.\u003c/em\u003e Molecular profiling of high-level athlete skeletal muscle after acute endurance or resistance exercise \u0026ndash; A systems biology approach. \u003cem\u003eMol. Metab.\u003c/em\u003e \u003cstrong\u003e79\u003c/strong\u003e, 101857 (2024).\u003c/li\u003e\n\u003cli\u003eHiatt, W. R., Regensteiner, J. G., Wolfel, E. E., Ruff, L. \u0026amp; Brass, E. P. Carnitine and acylcarnitine metabolism during exercise in humans. Dependence on skeletal muscle metabolic state. \u003cem\u003eJ. Clin. Invest.\u003c/em\u003e \u003cstrong\u003e84\u003c/strong\u003e, 1167\u0026ndash;1173 (1989).\u003c/li\u003e\n\u003cli\u003eWidjaja, A. A. \u003cem\u003eet al.\u003c/em\u003e Inhibition of IL-11 signalling extends mammalian healthspan and lifespan. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e632\u003c/strong\u003e, (2024).\u003c/li\u003e\n\u003cli\u003eHe, C. \u003cem\u003eet al.\u003c/em\u003e Enhanced Longevity by Ibuprofen, Conserved in Multiple Species, Occurs in Yeast through Inhibition of Tryptophan Import. \u003cem\u003ePLoS Genet.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, e1004860 (2014).\u003c/li\u003e\n\u003cli\u003eGuralnik, J. M. \u003cem\u003eet al.\u003c/em\u003e A short physical performance battery assessing lower extremity function: Association with self-reported disability and prediction of mortality and nursing home admission. \u003cem\u003eJournals Gerontol.\u003c/em\u003e (1994). doi:10.1093/geronj/49.2.M85\u003c/li\u003e\n\u003cli\u003eBergstr\u0026ouml;m, J., Hermansen, L., Hultman, E. \u0026amp; Saltin, B. Diet, Muscle Glycogen and Physical Performance. \u003cem\u003eActa Physiol. Scand.\u003c/em\u003e (1967). doi:10.1111/j.1748-1716.1967.tb03720.x\u003c/li\u003e\n\u003cli\u003evan der Berg, J. D. \u003cem\u003eet al.\u003c/em\u003e Identifying waking time in 24-h accelerometry data in adults using an automated algorithm. \u003cem\u003eJ. Sports Sci.\u003c/em\u003e (2016). doi:10.1080/02640414.2016.1140908\u003c/li\u003e\n\u003cli\u003eTudor-Locke, C. \u003cem\u003eet al.\u003c/em\u003e How many steps/day are enough? for adults. \u003cem\u003eInt. J. Behav. Nutr. Phys. Act.\u003c/em\u003e (2011). doi:10.1186/1479-5868-8-79\u003c/li\u003e\n\u003cli\u003eAndrews, S. \u0026amp; others. FastQC: a quality control tool for high throughput sequence data. 2010. \u003cem\u003eHttps://Www.Bioinformatics.Babraham.Ac.Uk/Projects/Fastqc/\u003c/em\u003e http://www.bioinformatics.babraham.ac.uk/projects/ (2019).\u003c/li\u003e\n\u003cli\u003eBolger, A. M., Lohse, M. \u0026amp; Usadel, B. Trimmomatic: a flexible trimmer for Illumina sequence data. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 2114\u0026ndash;2120 (2014).\u003c/li\u003e\n\u003cli\u003eKim, D., Langmead, B. \u0026amp; Salzberg, S. L. HISAT: a fast spliced aligner with low memory requirements. \u003cem\u003eNat. Methods\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 357\u0026ndash;360 (2015).\u003c/li\u003e\n\u003cli\u003eAnders, S., Pyl, P. T. \u0026amp; Huber, W. HTSeq\u0026mdash;a Python framework to work with high-throughput sequencing data. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e31\u003c/strong\u003e, 166\u0026ndash;169 (2015).\u003c/li\u003e\n\u003cli\u003eRobinson, M. D., McCarthy, D. J. \u0026amp; Smyth, G. K. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 139\u0026ndash;140 (2010).\u003c/li\u003e\n\u003cli\u003eRitchie, M. E. \u003cem\u003eet al.\u003c/em\u003e limma powers differential expression analyses for RNA-sequencing and microarray studies. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cstrong\u003e43\u003c/strong\u003e, e47 (2015).\u003c/li\u003e\n\u003cli\u003eLaw, C. W., Chen, Y., Shi, W. \u0026amp; Smyth, G. K. voom: Precision weights unlock linear model analysis tools for RNA-seq read counts. \u003cem\u003eGenome Biol.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, R29 (2014).\u003c/li\u003e\n\u003cli\u003eDurinck, S. \u003cem\u003eet al.\u003c/em\u003e BioMart and Bioconductor: A powerful link between biological databases and microarray data analysis. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 3439\u0026ndash;3440 (2005).\u003c/li\u003e\n\u003cli\u003eDurinck, S., Spellman, P. T., Birney, E. \u0026amp; Huber, W. Mapping identifiers for the integration of genomic datasets with the R/Bioconductor package biomaRt. \u003cem\u003eNat. Protoc.\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 1184\u0026ndash;1191 (2009).\u003c/li\u003e\n\u003cli\u003eMolenaars, M. \u003cem\u003eet al.\u003c/em\u003e Metabolomics and lipidomics in C. elegans using a single sample preparation. \u003cem\u003eDis. Model. Mech.\u003c/em\u003e (2021). doi:10.1242/dmm.047746\u003c/li\u003e\n\u003cli\u003eHerzog, K. \u003cem\u003eet al.\u003c/em\u003e Lipidomic analysis of fibroblasts from Zellweger spectrum disorder patients identifies disease-specific phospholipid ratios. \u003cstrong\u003e57\u003c/strong\u003e, 1447\u0026ndash;1454 (2016).\u003c/li\u003e\n\u003cli\u003eChambers, M. C. \u003cem\u003eet al.\u003c/em\u003e A cross-platform toolkit for mass spectrometry and proteomics. \u003cem\u003eNat. Biotechnol.\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 918\u0026ndash;920 (2012).\u003c/li\u003e\n\u003cli\u003eCore R Team. A Language and Environment for Statistical Computing. \u003cem\u003eR Foundation for Statistical Computing\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, https://www.R--project.org (2019).\u003c/li\u003e\n\u003cli\u003eSmith, C. A., Want, E. J., O\u0026rsquo;Maille, G., Abagyan, R. \u0026amp; Siuzdak, G. XCMS: Processing mass spectrometry data for metabolite profiling using nonlinear peak alignment, matching, and identification. \u003cem\u003eAnal. Chem.\u003c/em\u003e \u003cstrong\u003e78\u003c/strong\u003e, 779\u0026ndash;787 (2006).\u003c/li\u003e\n\u003cli\u003eRemie, C. M. E. \u003cem\u003eet al.\u003c/em\u003e Sitting less elicits metabolic responses similar to exercise and enhances insulin sensitivity in postmenopausal women. \u003cem\u003eDiabetologia\u003c/em\u003e \u003cstrong\u003e64\u003c/strong\u003e, 2817\u0026ndash;2828 (2021).\u003c/li\u003e\n\u003cli\u003eGentleman, R. C. \u003cem\u003eet al.\u003c/em\u003e Bioconductor: open software development for computational biology and bioinformatics. \u003cem\u003eGenome Biol.\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, (2004).\u003c/li\u003e\n\u003cli\u003eWickham, H., Fran\u0026ccedil;ois, R., Henry, L. \u0026amp; M\u0026uuml;ller, K. dplyr: A Grammar of Data Manipulation. R package version. \u003cem\u003eMedia\u003c/em\u003e (2019).\u003c/li\u003e\n\u003cli\u003eRohart, F., Gautier, B., Singh, A. \u0026amp; L\u0026ecirc; Cao, K. A. mixOmics: An R package for \u0026lsquo;omics feature selection and multiple data integration. \u003cem\u003ePLoS Comput. Biol.\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, e1005752 (2017).\u003c/li\u003e\n\u003cli\u003eCsardi, G. \u0026amp; Nepusz, T. The igraph software package for complex network research. \u003cem\u003eInterJournal\u003c/em\u003e \u003cstrong\u003eComplex Sy\u003c/strong\u003e, 1695 (2006).\u003c/li\u003e\n\u003cli\u003eWickham, H. Ggplot2. \u003cem\u003eWiley Interdiscip. Rev. Comput. Stat.\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 180\u0026ndash;185 (2011).\u003c/li\u003e\n\u003cli\u003eKassambara, A. Package \u0026lsquo;ggpubr\u0026rsquo;: \u0026lsquo;ggplot2\u0026rsquo; Based Publication Ready Plots. \u003cem\u003eR Packag. version 0.4.0\u003c/em\u003e (2020).\u003c/li\u003e\n\u003cli\u003eSlowikowski, K. ggrepel: Automatically Position Non-Overlapping Text Labels with \u0026lsquo;ggplot2\u0026rsquo;. \u003cem\u003eR package version 0.8.2\u003c/em\u003e (2020).\u003c/li\u003e\n\u003cli\u003eNeuwirth, E. RColorBrewer: ColorBrewer palettes. \u003cem\u003eR Packag. version 1.1-2\u003c/em\u003e https://cran.R-project.org/package=RColorBrewer (2014).\u003c/li\u003e\n\u003cli\u003eGoeman, J. J., Van de Geer, S., De Kort, F. \u0026amp; van Houwellingen, H. C. A global test for groups fo genes: Testing association with a clinical outcome. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 93\u0026ndash;99 (2004).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Aging, exercise, muscle, multi-omics, lifestyle","lastPublishedDoi":"10.21203/rs.3.rs-6074097/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6074097/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Exercise is fundamental to healthy aging, yet the degree to which it mitigates age-related molecular changes and how varying physical fitness levels influence the molecular response to exercise with age remain unclear. To address this, we performed transcriptomics, lipidomics, and metabolomics on skeletal muscle of young and older adults with differing physical function, both before and after an acute bout of sub-maximal exercise. At baseline, older adults exhibited reduced expression of genes associated with cellular respiration and energy metabolism compared to young adults with comparable activity levels. Remarkably, in trained older adults, 50% of these age-related differences were absent, resulting in transcriptomic profiles for cellular respiration that closely aligned with those of young adults. Following acute exercise, trained older adults demonstrated molecular responses that more closely resembled those of younger individuals. While all participants displayed transcriptional immune and stress responses upon acute exercise, the magnitude of these responses in older adults correlated positively with their physical fitness. These findings underscore the capacity of sustained physical training to transform age-related molecular profiles, highlight a positive link between physical fitness level and exercise-induced inflammation in older adults, and provide a multi-omic molecular atlas for examining aging and fitness regulatory networks.","manuscriptTitle":"Delayed molecular aging, preservation of energy metabolism and enhanced exercise response in exercise-trained human muscle","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-03-18 08:08:19","doi":"10.21203/rs.3.rs-6074097/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-aging","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"nataging","sideBox":"Learn more about [Nature Aging](https://www.nature.com/nataging/)","snPcode":"","submissionUrl":"","title":"Nature Aging","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Research","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"988d1029-295a-4c82-8c24-04b924aad8f1","owner":[],"postedDate":"March 18th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":45819805,"name":"Biological sciences/Molecular biology"},{"id":45819806,"name":"Biological sciences/Physiology/Ageing"},{"id":45819807,"name":"Biological sciences/Physiology/Metabolism"}],"tags":[],"updatedAt":"2025-03-18T08:08:19+00:00","versionOfRecord":[],"versionCreatedAt":"2025-03-18 08:08:19","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6074097","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6074097","identity":"rs-6074097","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}