{"paper_id":"55edc229-2066-47f6-9243-d69d68fa1d5e","body_text":"Rapamycin exerts its geroprotective effects in the ageing human immune \nsystem by enhancing resilience against DNA damage \nAuthors:  Loren Kell1,2,3, Eleanor J Jones4, Nima Gharahdaghi4, Daniel J Wilkinson4, Kenneth Smith4, \nPhilip J Atherton4,5*, Anna K Simon3,6*, Lynne S Cox1*, Ghada Alsaleh2,3* \n \nAffiliations: \n1Department of Biochemistry, University of Oxford; Oxford, United Kingdom. \n2Botnar Institute for Musculoskeletal Sciences, Nuffield Department of Orthopaedics, Rheumatology and \nMusculoskeletal Sciences (NDORMS), University of Oxford; Oxford, United Kingdom. \n3The Kennedy Institute of Rheumatology, Nuffield Department of Orthopaedics, Rheumatology and \nMusculoskeletal Sciences (NDORMS), University of Oxford; Oxford, United Kingdom.  \n4MRC-Versus Arthritis Centre for Musculoskeletal Ageing Research, Centre of Metabolism, Ageing and Physiology \n(COMAP), Academic Unit of Injury, Recovery and Inflammation Sciences (IRIS), School of Medicine, University of \nNottingham, Royal Derby Hospital; Derby, United Kingdom. \n5Ritsumeikan University; Kyoto, Japan. \n6Max Delbrück Center for Molecular Medicine; Berlin, Germany. \n \n*Co-corresponding authors. Emails:  ghada.alsaleh@ndorms.ox.ac.uk; lynne.cox@bioch.ox.ac.uk; \nphilip.atherton@nottingham.ac.uk; katja.simon@mdc-berlin.de.   \n \nORCiD: Loren Kell (0000-0003-1322-2027); Eleanor J Jones (0000-0003-3261-3787); Nima Gharahdaghi (0000-0002-4650-\n244X); Daniel J Wilkinson (0000-0001-8808-8243); Kenneth Smith (0000-0001-8971-6635); Philip J Atherton (0000-0002-7286-\n046X); Anna K Simon (0000-0002-4077-7995); Lynne S Cox (0000-0002-5306-285X); Ghada Alsaleh (0000-0002-4211-3420).  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 2 \nAbstract \nmTOR inhibitors such as rapamycin are among the most robust life-extending interventions known, \nyet the mechanisms underlying their geroprotective effects in humans remain incompletely \nunderstood. At non -immunosuppressive doses, these drugs are senomorphic, i.e. they mitigate \ncellular senescence, but whether they protect genome stability itself has been unclear. Given that \nDNA damage is a major driver of immune ageing, and immune decline accelerates whole-organism \nageing, we tested whether mTOR inhibition enhances genome stability. In human T cells exposed \nto acute genotoxic stre ss, we found that rapamycin and other mTOR inhibitors suppressed \nsenescence not by slowing protein synthesis, halting cell division, or stimulating autophagy, but by \ndirectly reducing DNA lesional burden and improving cell survival. Ex-vivo analysis of aged \nimmune cells from healthy donors revealed a stark enrichment of markers for DNA damage, \nsenescence, and mTORC hyperactivation, suggesting that human immune ageing may be \namenable to  intervention by  low-dose mTOR inhibition. To test this in vivo , we condu cted a \nplacebo-controlled experimental medicine trial in older adults administered with low -dose \nrapamycin. p21, a marker of DNA damage -induced senescence, was significantly reduced  in \nimmune cells from the rapamycin compared to placebo group. These findings reveal a previously \nunrecognised role for mTOR inhibition: direct genoprotection. This mechanism may help explain \nrapamycin’s exceptional geroprotective profile and opens new avenues for its use in contexts \nwhere genome instability drives pathology, ranging from healthy ageing, clinical radiation \nexposure, and even the hazards of cosmic radiation in space travel. \n  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 3 \nINTRODUCTION \nRapamycin and other mTOR inhibitors used at low doses increase lifespan in all species \ntested to date (Bjedov et al., 2010, Ha and Huh, 2011, Harrison et al., 2009) . Importantly, this \nlifespan extension corresponds with increased healthspan, as rapamycin has been shown to \nimprove health across multiple domains (Wilkinson et al., 2012). Further gains in lifespan extension \nhave been reported when rapamycin is administered in combination with other geroprotectors such \nas trametinib (Gkioni et al., 2025) .Though mTOR inhibitors have shown remarkable anti -ageing \npotential, the exact hallmarks of ageing on which they impact are not fully understood (Weichhart, \n2018). One explanation is that mTOR inhibitors such as rapamycin are senomorphic, in that they \nlimit cellular senescence, a physiological process by which highly damaged cells exit the cell cycle \nand assume a pro -inflammatory, tissue-remodelling phenotype (Walters et al., 2016, Rolt et al., \n2019, Park et al., 2020, Walters and Cox, 2018) . mTOR activity increases during the in vitro  \nsenescence of primary human fibroblasts and in human muscle ageing in vivo ((Walters et al., \n2016, Carroll et al., 2017, Markofski et al., 2015) . Further to this correlative data, cells with \nconstitutive mTOR activation  enter premature replicative cell senescence in vitro , suggesting \nmTOR hyperactivity is sufficient to drive cellular ageing  (Zhang et al., 2003) . Consistent with a \nrole of mTOR in ageing and senescence, mTOR inhibitors attenuate a variety of senescence \nphenotypes and extend replicative lifespan in vitro (Walters et al., 2016, Rolt et al., 2019, Park et \nal., 2020). In humans, rapamycin reduced the presence of dermal cells expressing the senescence \nbiomarker, p16, when administered in a topical skin cream  (Chung et al., 2019) . The primary \ncellular mechanism underlying these senomorphic properties of mTOR inhibition are not  fully \nunderstood, though impacts on slowing protein synthesis, the cell cycle, or supporting the removal \nof dysfunctional organelles and protein aggregates through enhanced autophagy have been \nsuggested (Weichhart, 2018). Furthermore, there is a gap in our understanding of how mTOR \nactivity is associated with the ageing of cells which drive the ageing process – namely, those of \nthe immune system, for which there is increasing evidence that DNA damage is a key driver (Kell \net al., 2023). \nRecent studies have demonstrated how ageing of the immune system \n(immunosenescence) can precipitate whole-organism ageing (Yousefzadeh et al., 2021b, Desdin-\nMico et al., 2020), highlighting how strategies which target immunosenescence are at the frontiers \nof geriatric medicine. Since aged T cells drive tissue destruction and multimorbidity during ageing, \nthey further provide a cellular target for therapeutic anti-ageing intervention (Soto-Heredero et al., \n2023). At high doses, rapamycin is immunosuppressant and causes side effects such as poor \nwound healing, ulcers , and loss of metabolic control leading to diabetes (Knight et al., 2007, \nAltomare et al., 2006, Houde et al., 2010) . On the other hand, a t low doses, mTOR inhibition is \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 4 \none of the few interventions which has been shown actually to improve immunity in older people – \ni.e., to attenuate immunosenescence (Mannick and Lamming, 2023).  In humans, low-dose mTOR \ninhibitor RAD001 (everolimus) improved B and T cell responses to influenza vaccination  in older \nadults (Mannick et al., 2018, Mannick et al., 2014). A second generation mTOR inhibitor RTB101 \nsignificantly reduced respiratory tract infections (RTIs) in older adults in a Phase 2b clinical trial in \n652 study participants (Mannick et al., 2021). While a larger Phase 3 trial did not reach significance \nfor reduction in mild RTIs, there was a clear trend to improved immune function (Mannick et al., \n2021). mTOR inhibitors therefore offer a therapeutic route to enhance ageing immune responses \nagainst viral pathogens for which we currently lack effective pharmacological interventions. \nHowever, there is a gap in our understanding of how they impact on cellular processes such as \nimmune cell ageing, which underpins immunosenescence and subsequent organismal ageing, in \nan immune-unchallenged steady state. \nThere is accumulating evidence that DNA damage is a central driver of immune cell ageing, \nimmunosenescence, and whole-organism ageing (Kell et al., 2023, Yousefzadeh et al., 2021a, \nYousefzadeh et al., 2021b) . In this study,  we aimed to determine whether low -dose mTOR \ninhibition could enhance DNA stability in human T cells, a key immune cell type affected by age -\nrelated DNA damage. Using a combination of in vitro DNA damage assays, ex vivo profiling of \nage-related immune cells, and a  placebo-controlled in vivo intervention with rapamycin in older \npeople, we sought to explore mTOR inhibitors as a potential strategy to protect cells from DNA \ndamage and limit senescence . Our findings have implications for geriatric medicine, \nradioprotection during cancer therapy, and safeguarding astronauts from cosmic radiation. \n  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 5 \nRESULTS \nDNA damage in T cells is associated with elevated mTORC signalling \nIn order to develop an in vitro model for DNA damage and a reliable read-out in primary \nhuman immune cells, we cultured isolated human peripheral blood mononuclear cells (PBMCs)  \nfrom healthy donors with T cell activating antibodies against CD3 and CD28 for 3 days, followed \nby treatment with zeocin, a double-strand break (DSB) inducer, for 2 hours (DSBs in circulating \nleukocytes are predictive of increased mortality in in humans (Bonassi et al., 2021) ). Acute 15-\nminute exposure to hydrogen peroxide was used as a positive control for DNA damage induction \n(Figure 1a). After recovery, cells were analysed by flow cytometry to identify CD4+ and CD8+ T \ncells, and assessed for levels of the DNA damage marker, γH2AX (gating strategy in Figure S1). \nAs seen in Figure 1b, zeocin treatment led to a marked increase in T cells positive for  \nγH2AX, with a similar though more extensive shift to γH2AX-positivity in the peroxide -treated \ncontrols. This increased γH2AX signal was associated with a large increase in the percentage of \nT cells staining positive for γH2AX, from ~10% untreated control cells to ~30% zeocin-treated both \nCD4+ and CD8+ cells (Figure 1c). We note that the small percentage of the untreated control cells \nshowing γH2AX positivity potentially indicates DSB formation during activation. Levels of γH2AX \npositivity peaked at 4 hours post zeocin treatment, reducing by 24 hours of recovery (Figure 1d). \nConsistent with elevated γH2AX, zeocin-treated cells also showed elevate d DNA damage \nresponse signalling, including phosphorylation of checkpoint kinases Chk1 and Chk2, as well as \nincreased protein levels of tumour suppressor p53, which is stabilised by phosphorylation during \nthe DDR, and its transcriptional target, the cyclin-kinase inhibitor p21 (Figure 1e). To determine \nwhether DNA damage correlates with changes in mTOR activity, we further analysed T cells with \nlow and high levels of γH2AX for their level of phosphorylated mTORC1 target S6 (p-S6) and \nmTORC2 target Akt (p-Akt) (Figure 1f). Notably, both CD4 + and CD8+ cells with high γH2AX \nsignals showed significant increases in phospho rylated S6 (Figure 1g ), an indirect target of \nmTORC1, but no change in levels of in mTORC2 target p -Akt (Figure 1h), suggesting that DNA \ndamage in T cells is associated with elevated mTORC1 activity. \n \nSuppression of mTORC signalling reduces markers of DNA damage in human T cells in vitro \nTo test the association between high levels of DNA damage markers and elevated mTORC \nsignalling, we assessed the impact of mTORC inhibitors on the DDR in the zeocin-induced DNA \ndamage model. T cells were  incubated throughout their 3-day activation, 2-hour zeocin-treatment, \nand 4-hour recovery periods with low dose mTORC1 inhibitor rapamycin ( 10nM), pan -mTOR \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 6 \ninhibitor AZD8055 ( 100nM) or DMSO vehicle control (Figure 2a). Exposure to rapamycin and \nAZD8055 over this 3 -day activation significantly suppressed p -S6 levels, apparent as early as 6 \nhours (Figure S2a-b). CD25 upregulation, a marker of T cell activation, was not impacted by \nmTOR inhibition at this low dose (Figure S2c). As before (Figure 1b-c), zeocin treatment resulted \nin a significant increase in overall γH2AX levels. However, this surge in γH2AX was greatly \nattenuated by treatment with the mTOR inhibitors rapamycin or AZD8055 (Figure 2b), reflected \nby the percentage of CD4 + T cells staining positive for γH2AX after zeocin treatment being \nsignificantly reduced by both mTOR inhibitors (Figure 2b, middle). By contrast, in CD8+ cells, this \nreduction was only significant on rapamycin treatment (Figure 2b, right).  \nTo investigate further whether mTOR inhibition affected signalling within the DNA damage \nresponse, we assessed levels of phosphorylated ( i.e. activated) checkpoint kinases Chk1 and \nChk2. Both rapamycin and AZD8055 treatment prevented the zeocin-induced increase in levels \nfor both p-Chk1 and p-Chk2 in both CD4+ and CD8+ cells (Figure 2c-d). We additionally assessed \nlevels of the DDR proteins p53 and p21 at both 4 hours recovery from zeocin, and at a later 24 -\nhour timepoint, to assess longer -term effects on resolution of the DDR ( Figure 2e). Notably, in \ncontrol cells without zeocin -induced DNA damage, rapamycin treatment led to reduction in p53 \nand p21 levels, compared with DMSO vehicle controls (Figure 2f). p21 levels increased by 4 hours \nrecovery following zeocin treatment, remaining elevated at 24 hours; this response was completely \nablated on mTOR inhibition by rapamycin treatment (Figure 2f). Similarly, the elevated p53 signal \nseen at 4 hours was significantly reduced on rapamycin treatment. By 24 hours, the p53 signal \nwas reduced in zeocin-treated cells (with and without rapamycin treatment) compared with levels \nat 4 hours post damage in both CD4+ and CD8+ T cells, though mTORC inhibition led to a further \nsignificant drop in p53 levels (Figure 2f).  \nDirect association between high levels of damage and elevated mTORC signalling \nHaving identified that continuous mTOR inhibition suppressed DDR upregulation, we next \ninvestigated the temporal nature of this effect, by incubating cells with rapamycin either before, \nduring or after DNA damage by zeocin exposure (Figure 3a). To do this, T cells within PBMC \ncultures from healthy donors were activated for 3 days with anti -CD3 and anti-CD28 antibodies, \nthen sequentially split into aliquots and incubated with rapamycin or DMSO ± zeocin as shown in \nFigure 3 a. Following the recovery period  also in the presenc e or absence of rapamycin , the \npercentage of cells staining positive for  DNA damage marker γH2AX was assessed by flow \ncytometry.  In all cases, zeocin treatment resulted in an increase in γH2AX-positive cells (Figure \n3b-c), but CD4 + T cells showed a significant reduction of γH2AX positivity when treated with \nrapamycin before, during or after zeocin treatment compared with the DM SO-only controls with \nzeocin (Figure 3b). Furthermore, the fold increase in γH2AX+ cells significantly correlated with \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 7 \nlevels of both p-S6 and p-Akt, consistent with a role for mTOR activity in a highly DNA-damaged \nphenotype (Figure 3d -e). The exception to this was p -Akt in CD8 + T cells, which negatively \ncorrelated with levels of γH2AX (Figure 3e, right), consistent with the result that treatment with \nrapamycin only during or after zeocin treatment (at times when p -Akt was not suppressed) could \nminimise γH2AX induction.  In summary, these data suggest that rapamycin treatment either \nbefore, during, or after exposure to a genotoxin, i.e. even short-term treatment could prevent \nzeocin-induced γH2AX levels. \n \nReduction in DNA damage markers by mTOR inhibition is not due to impacts on cell cycle or  \nprotein synthesis \nProgression through the cell cycle is halted during the initial stages of the DDR to allow for \nrepair of DNA lesions. Thus, one possible explanation for our observation that mTOR inhibition \nlimits γH2AX+ DNA damage in T cells is that it promotes cell cycle arrest to support DNA repair . \nWe therefore first measured the proportion of cells in  each phase of the cell cycle (G0/G1, S, or \nG2/M) in untreated and zeocin-treated T cells using flow cytometry (Figure 4a). Zeocin treatment \nof both CD4+ and CD8+ T cells halved the proportion of cells in S-phase (from 52% without zeocin \nto 26% with zeocin treatment) with a concomitant increase in the proportion of cells in G2/M phase \n(from 1.3% to 15%) ( Figure 4b), suggesting that DNA-damaged cells proceed to G 2 but then \nactivate cell cycle checkpoints to prevent cell division.  \nTo test whether mTOR inhibi tion affected cell cycle progression, cells were treated with \nrapamycin continuously (RRR), during (DRD), and/or after treatment (DRR, DDR) with zeocin. \nUnder these conditions previously, rapamycin limited zeocin-induced γH2AX levels in CD4+ and \nCD8+ T cells (Figure 3b-c). We observed an increase in the G0/G1-phase population on continuous \nrapamycin treatment (RRR) in cells without overt DNA damage (i .e. γH2AX negative), though it \ndid not affect the proportion of G0/G1-phase cells in the γH2AX-positive population, indicating that \ncontinuous rapamycin treatment did not change cell cycle phase distribution in the context of DNA \ndamage (Figure 4c). Since rapamycin treatment before, during, after zeocin treatment, or \ncontinuous exposure (DDR, DRD, DRR and RRR) effectively limited the induction of γH2AX in T \ncells (Figure 3b-c), but did not affect cell cycle phase distribution in DNA-damaged γH2AX+ cells \n(Figure 4c), we concluded that the effect of rapamycin on γH2AX was not due to effects on the \ncell cycle. \nmTOR is a master anabolic regulator of protein synthesis (e.g. by activating ribosomal S6 \nprotein through S6K-dependent phosphorylation), so it is conceivable that the reduced levels of \nDNA damage proteins we detect by flow cytometry may be a consequence of blockade of their de \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 8 \nnovo synthesis (albeit th at the  acute DDR is predominantly mediated post -translationally). To \nevaluate the effects of rapamycin on nascent protein synthesis in the DNA damage assay, cells \nwere treated with rapamycin or DMSO vehicle control before, during and/or after zeocin treatment \nand then incubated for the final 30 minutes of their 4-hour recovery from zeocin with O-propargyl-\npuromycin (OPP), an alkyne analogue of puromycin that is incorporated into nascent polypeptides \nand halts further translation (Figure 4d). The mean fluorescence of labelled OPP in cells thus \nreports short term total de novo protein synthesis. One-hour treatment with 50 μg/ml cycloheximide \n(CHX) served as a positive control for inhibition of protein synthesis. As expected, CHX -treated \ncells incorporated significantly less OPP than the DMSO untreated controls ( Figure 4e). \nRapamycin treatment both during and after zeocin exposure (DRR) did not significantly affect OPP \nincorporation, though continuous rapamycin treatment (RRR) showed a non -significant trend \ntowards lower OPP fluorescence (Figure 4e). Since there was no consistent effect of rapamycin \nin decreasing OPP levels, this suggests that the effect of rapamycin on limiting zeocin -induced \nDDR signalling levels was not due to decreasing global protein synthesis. In summary so far, our \ndata indicates that the rapamycin -mediated protection from upregulation of the DDR following \nzeocin exposure is likely independent of effects on the cell cycle and protein synthesis. \n \nAutophagy is required to limit DNA damage in T cells, but rapamycin’s protective effect is \nautophagy-independent.  \nAutophagy is a cytoprotective cell recycling process that is repressed by mTORC1 activity; \nnotably, autophagy is also involved in regulation of the DNA damage response  (Vessoni et al., \n2013). We therefore probed whether the mechanism by which rapamycin treatment reduces \nmarkers of DNA damage  signalling following zeocin exposure, could be due to enhancement of \nautophagic flux, as measured by  a flow cytometry-based LC3 assay (Figure S3a-b) (Alsaleh et \nal., 2020). In the presence of zeocin -induced damage (Figure 5a), activated T cells with a high \nDNA damage load (i.e. positive for γH2AX) showed significantly lower autophagic flux than those \ncells negative for γH2AX (Figure 5b). This suggests either that cells bearing a heavy DNA lesional \nload are less able to undergo autophagy, or that those with effective autophagy rapidly resolve \nDNA damage leading to low levels of damage markers such as γH2AX.  \nTo distinguish between these possibilities, we used the drug chloroquine to  inhibit \nautophagy, which effectively halved autophagic flux in activated T cells (Figure S3c). In zeocin-\ntreated cells, autophagy blockade increased γH2AX-positive cells, confirming that autophagy does \nlimit DNA damage in human T cells ( Figure 5c). Next, we asked whether rapamycin enhanced \nautophagy in zeocin-exposed cells ( Figure 5d) and found that this was the case , regardless of \nγH2AX levels or timing of rapamycin administration  (Figure 5e ). We then asked whether \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 9 \nrapamycin’s protective effect on DNA damage depended on autophagy  by co-treating cells with \nrapamycin and chloroquine (Figure 5f). As expected, chloroquine inhibited autophag ic flux and \nincreased γH2AX positivity, but rapamycin still markedly reduced DNA damage despite strong \nautophagy inhibition in the context of chloroquine co-treatment (Figure S3d, Figure 5 h). These \nfindings indicate that while autophagy supports DNA damage resoluti on, rapamycin’s protective \neffect is independent of cell cycle pausing, protein synthesis, and autophagy. \n \nRapamycin decreases overall DNA lesional burden and reduces T cell death following DNA \ndamage  \nDDR signalling requires activation of several PI3 -like kinases (e.g. DNA-PKcs, ATM and \nATR). It is therefore possible that mTOR inhibitors reduce apparent DNA damage by inhibiting \ncritical DDR enzyme signalling, in a manner that would be highly detrimental to cell health and \nsurvival. Alternatively, reduced levels of DDR signalling may instead reflect a lower DNA lesional \nburden. To distinguish between these two possibilities, we assessed the extent of DNA breaks \nafter 4 hours of recovery from zeocin exposure, using the alkaline comet assay, in isolated CD4+ \nT cells treated with or without rapamycin (Figure 6a).  Treatment with hydrogen peroxide was used \nas a positive control for DNA breakage. Both zeocin and hydrogen peroxide treatment significantly \nincreased DNA lesions (both DSBs and SSBs) compared to untreated controls ( Figure 6b-c). \nNotably, DNA lesion burden was markedly reduced in CD4+ T cells treated with rapamycin at this \n4-hour recovery timepoint from zeocin. This suggests that the reduction in DDR signalling afforded \nby rapamycin is due to enhance genome stability rather than downstream inhibition of DDR \nenzymes (Figure 6b-c). \nTo assess further the kinetics of DNA lesional burden and potential resolution of damage, \nwe then assessed comet Olive moment in cells incubated continuously with rapamycin over a time \ncourse of up to 24 hours after exposure to zeocin. The peak of DNA lesions manifesting as comet \ntails was found to occur at 4 hours after zeocin treatment ( Figure 6d). Continuous rapamycin \ntreatment significantly limited the DNA lesion burden at all timepoints tested (Figure 6d). Notably, \nrapamycin reduced comet tails even at 0 hours post-zeocin exposure, i.e. directly after genotoxin \ntreatment, suggesting a stark enhancement of resilience from DNA  damage that may reflect \nprevention of DNA lesion formation. In summary, we can rule out a negative effect of rapamycin \ninhibiting PI3-like kinases in the DDR, and instead propose that rapamycin positively protects cells \nfrom DNA damage. \nTo explore whether this effect of rapamycin on reducing lesional load has  an impact on \noverall cell physiology, we measured cell viability by assessing fluorescence of a membrane-\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 10 \nimpermeable dye that is taken up only by dead cells  (Figure 6e). Consistent with the increase in \ncells with major DNA lesions following zeocin exposure, we observed a decrease in the percentage \nof live CD4+ T cells at 4 hours, leading to a severe reduction to only 20% live cells by 24 hours \nrecovery from zeocin in DMSO vehicle control cells i.e. the high lesional burden induced by zeocin \ntreatment is lethal to the majority of T cells ( Figure 6e). Remarkably, continuous treatment with \nlow-dose rapamycin (10 nM) supported much greater cell survival with over 60% cells still viable \n24 hours after zeocin treatment  (Figure 6e), strongly suggesting that rapamycin does indeed \nenhance DNA repair responses i.e. it may act as a genoprotector. \n \nAge-related immune subsets show elevated markers of DNA damage, cell senescence, and mTOR \nactivity \nThese findings suggest a genoprotective role for rapamycin in the immune system, but are \nbased on an in vitro model of acute DNA damage. In humans, immunosenescence involves the \nexpansion of terminally-differentiated immune subsets linked to dysfunction (Table 1), though their \nexact phenotype remains unclear. Since chronic DNA damage accumulation is a key hallmark of \nageing, that in immune cells contributes to whole -body ageing (Yousefzadeh et al., 2021b), we \nexamined whether aged human immune cells show signs of DNA damage and cell senescence, \nand whether this correlates with their mTOR activity. \nUsing 27-colour spectral flow cytometry, we analysed age -related immune cell subsets \nfrom healthy donor blood, including TEMRA T cells (CD4+ and CD8+), IgD-CD27- (double-negative) \nB cells, CD16+CD57+ NK cells, and non-classical monocytes (Figure 7a, Figure S4, Table 1). We \nthen assessed senescence- and DNA damage-associated markers (p21, p16, p53, γH2AX) and \ncell size (measured by forward scatter, FSC) to determine whether these subsets were enriched \nfor ageing biomarkers compared with their naïve counterparts  (Stein et al., 1999, Passos et al., \n2007, van Deursen, 2014, Tsai et al., 2021). \nWe observed that senescence markers were significantly enriched in age-related immune \ncells compared to their naïve equivalents, with each of the 6 subsets assessed showing significant \nelevation of at least 3/5 senescence markers  (Figure 7b-c). In particular, age -related CD4+ and \nCD8+ TEMRAs and non-classical monocytes showed the greatest number of elevated senescence \nmarkers compared to early-differentiated cells of the same lineage (4/5 each, Figure 7c). Double-\nnegative (DN) B cells and CD57 +CD16+ (double-positive, DP) NK cells all exhibited significant \nupregulation of 3/5 senescence markers (Figure 7c). Notably, DNA damage marker γH2AX was \nelevated only in age-associated T and B cells i.e. immune cell types that undergo double strand \nbreaks during V(D)J recombination to form T cell and B cell receptors (Figure 7b-c). Intriguingly, \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 11 \nthe most common senescence feature tha t was increased across age -related subsets was p21, \npresent in all 6 age -related subsets compared to their early -differentiated controls ( Figure 7c). \nThis was followed by high p53, p16, and cell size/FSC (present in 5/6 subsets each , Figure 7c). \nOverall, these data suggest that age -related subsets display several features of cellular \nsenescence, and particularly overexpress the p21 and p53 pathway, suggesting  a DNA-damage \ninduced senescence phenotype. Importantly, these data demonstrate that age-related immune \ncells may be targetable with genoprotective senotherapeutics. \nNext, we asked whether these age-related subsets showed changes in their activation of \nmTORC1 and mTORC2 by measuring their levels of p -S6 and p-Akt respectively. We observed \nthat age-related CD4+ TEMRAs, CD8+ TEMRAs, and non-classical monocytes all showed elevated \np-S6, with CD4 + TEMRAs and non -classical monocytes additionally displaying increased p -Akt \nlevels (Figure 7b-c). These findings indicate that while senescence markers were present in all \nage-related immune subsets, mTOR hyperactivation occurs only in T cell and monocyte ageing. \nGiven that diverse age-related subsets within healthy donors showed elevated senescence \nand mTORC1/2 markers, we next asked whether immune cells from older people (average age 62 \nyears-old, n=9) exhibited increased mTORC activation compared to those from younger donors  \n(<50 years old, n=8). We first verified that, compared to the younger group, older donors had an \nincreased percentage of CD8+ T cells expressing CD57 and KLRG1, and loss-of-expression of \nCD28 (Figure 7d),  all of which are markers of immunosenescence (Kell et al., 2023). Comparison \nof immune subsets between these two groups showed that, in addition to these markers of T cell \nsenescence, immune ageing corresponded with an increase in p-S6 levels across all immune cell \ntypes analysed (Figure 7e). This suggests that mTORC1 activity is a broad biomarker of human \nimmune ageing shared by cell types from diverse lineages. \n \nLow-dose rapamycin reduces markers of senescence and DNA damage in humans in vivo \nTaken together, our data so far show that age-related immune subsets exhibit features of \nDNA damage, cell senescence, and mTOR hyperactivation, and that human ageing is \naccompanied by increased mTOR activity across all immune cell types  (Figure 7d-e). We have \nfurther demonstrated that treatment with low-dose mTOR inhibitors improve survival and reduce \nmarkers of senescence and DNA damage in human T cells treated with a genotoxic agent outside \nof the body. Such findings are important but require in vivo data before they support further clinical \naction. We therefore assessed whether rapamycin treatment impacts on immune cell DNA damage \nand senescence in vivo in humans, analysing PBMCs from participants of a single-blind, placebo-\ncontrolled trial (NCT05414292), in which older male volunteers received either 1 mg/day rapamycin \n(n=4) or placebo (n=5) for 4 months ( Figure 8a). While the primary endpoint was to assess \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 12 \nchanges in muscle mass and protein synthesis, we aimed to assess  features of \nimmunosenescence in PBMCs isolated at several timepoints throughout the trial. \nParticipants in the rapamycin and placebo groups were well -matched for age and BMI  \n(Table 2). After 8 weeks of intervention, the concentration of rapamycin in the blood reached an \naverage of 3.24 ± 1.81 nM in the treatment gro up (Figure S5a), i.e., within the same order of \nmagnitude as the doses used in our in vitro  experiments (10  nM). To address concerns of \nimmunosuppression by rapamycin, the white blood cell count was assessed at 8 weeks; there \nwere no significant differences in leukocyte counts in the blood over the initial 8-week rapamycin \ntreatment period, and between rapamycin treated and placebo controls, suggesting that this low-\ndose rapamycin treatment regimen was not immunosuppressive (Figure S5b).  To assess whether \nmTOR activity was inhibited at this low dose of rapamycin, we analysed p-S6 levels across immune \nsubsets. We observed  a significant decrease in p-S6 levels in most immune subsets in the \nrapamycin-treated participants compared to those in the placebo group at 4-5 weeks, suggesting \nsuccessful inhibition of mTORC1 (Figure 8d). Taken together, low-dose rapamycin treatment led \nto detectable stable blood rapamycin concentrations at a level well below that used therapeutically \nfor immunosuppression  with no evidence of leukocyte suppression, plus  reduced markers of \nmTORC1 activity in peripheral immune cells after 4-5 weeks. \nTo determine whether features of immunosenescence were impacted by in vivo rapamycin \ntreatment, we analysed PBMCs from the trial using 27-colour spectral flow cytometry (Figure S4). \nSimple linear regression analyses in circulating CD4 + T cells revealed a strong and highly \nsignificant positive correlation between p-S6 and γH2AX levels in both treatment groups, indicating \nthat mTOR activity and DNA damage are positively linked in vivo (Figure 8b-c). Notably, T cells \nfrom the rapamycin group show ed lower p-S6 levels than those from the placebo group, w hich \ncorresponded with decreased γH2AX levels (Figure 8c). Overall, rapamycin treatment led to  a \ntrend towards lower γH2AX levels in immune subsets, particularly in age-related CD4 TEMRA and \ndouble-negative B cells, which with higher participant numbers might show significance  (Figure \nS5c). Consistent with these positive effects on γH2AX-marked DNA damage, 4-month rapamycin \ntreatment caused a robust and significant decrease in p21 expression across most immune cell \nsubsets studied, reflecting the attenuation of DNA damage -induced p21 with rapamycin we \nobserved in vitro  (Figure 8e). p53 expression was eleva ted at 4 months in PBMCs from the \nrapamycin-treated compared to the placebo groups (Figure 8f). A previous study demonstrated \nthat in vivo mTOR inhibition decreased the percentage of circulating PD-1+ T cells (Mannick et al., \n2014). Though we did not observe changes in PD-1 in the current trial (Figure S5d), the proportion \nof T cells expressing other immune co-inhibitory molecules, such as KLRG1 (Figure 8g), NKG2A \n(Figure 8h), and LAG3 ( Figure 8i), was reduced in the T cells from rapamycin compared to \nplacebo groups. Overall, these results suggest that rapamycin reduces the expression of immune \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 13 \nexhaustion markers, and p21, a marker of cell senescence . While participant numbers in  the \nrapamycin in vivo  study are low, the changes in  DNA damage and senescence markers is \nsignificant.  \n \n \nDISCUSSION \nWhile mTOR inhibit ion is a  well-known and potent anti -ageing intervention in animal \nmodels, an explanation for its ability to extend health - and lifespan so reproducibly has been \nlacking (Weichhart, 2018). Furthermore, our understanding around why mTOR inhibitors have \nshown benefit in boosting immune resilience in older people is incomplete. In this study, we have \ndemonstrated for the first time that mTOR inhibitors can protect T cells from DNA damage and \nsenescence marker upregulation after exposure to a genotoxic agent. We show that this is through \na mechanism independent of autophagy, cell cycle progression, and protein synthesis. Rather, we \nshow that this is through a mitigation of DNA lesional burden, affording a greater survival following \nexposure to DNA-damaging treatment. This enhancement of protection from DNA damage, which \nwe call genoprotection, offers a new explanation for previous studies that have demonstrated an \nattenuation of replicative senescence with mTOR inhibitors in 2D cell culture (Walters et al., 2016, \nPark et al., 2020, Iglesias-Bartolome et al., 2012) and in vivo in human skin (Chung et al., 2019), \nand its potent geroprotective ability . Our study also provides a novel explanation for previous \nreports which show that rapamycin improves aged antigen-specific immunity in mouse models of \nimmunosenescence bearing immune-specific knockout of DNA repair (Yousefzadeh et al., 2021b). \nFurthermore, our findings expand on  previous research showing a reduction in DNA damage \nmarkers with rapamycin in irradiated normal oral keratinocytes (Iglesias-Bartolome et al., 2012) , \nDNA repair-deficient fibroblasts  (Saha et al., 2014), human oocytes undergoing in vitro maturation \n(Yang et al., 2022), DNA repair-deficient mouse podocytes (Braun et al., 2025), and lymphocytes \nof kidney transplant patients (Chebel et al., 2016). This enhanced resilience to DNA damage with \nmTOR inhibition was shown to arise from several sources, including heightened expression of \nantioxidant enzymes, such as mitochondrial superoxide dismutase, that limit ROS and genotoxic \nstress (Iglesias-Bartolome et al., 2012) , and increased protein expression of the DNA repair \nfactors, MGMT and NDRG1, via  a post-transcriptional mechanism (Dominick et al., 2017).  Our \ndata from human immune cells may therefore reflect a universal impact of rapamycin on promoting \ngenome integrity in eukaryotes. \nIn the present study, we asked whether age-related whether age-related immune cells from \ndiverse haematopoietic lineages exhibited DNA damage -induced senescence, by \ncomprehensively profiling senescence markers in human immune subsets using high-dimensional \nspectral cytometry. Our data are the first to show that age-related immune subsets from diverse \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 14 \nimmune lineages, in CD4+ and CD8+ T cells, B cells, NK cells, and monocytes , are uniformly \nenriched for senescence biomarkers. In particular, the DNA damage -induced cyclin kinase \ninhibitor, p21, was a senescence marker upregulated in most age -related immune subsets. This \nsuggests that the form of senescence which immune cells undergo with ageing may be p53- and \np21- driven, hinting – importantly – towards a more DNA damage -induced type of senescence, \nconsistent with other evidence that DNA damage play s a central role in the decline of immune \nsystem function (Kell et al., 2023). Immune cells from older donors exhibited higher levels of p-S6, \nindicating mTORC1 activity, suggesting that, like ageing of other human tissues (Markofski et al., \n2015), immune ageing is associated with mTORC1 hyperactivity.  \nMost importantly, our findings translate to the in vivo condition in humans. Through a small \npilot trial with limited participant numbers, ours is the first study demonstrating a significant \nreduction in p21-marked cellular senescence upon 4-month, low-dose rapamycin treatment vs \nplacebo in immune cells in the blood of older people.  We also found that rapamycin increased p53 \nlevels in circulating immune cells. p53 serves multiple physiological roles in vivo; for example, in \naddition to its well -known role in signal transduction of acute DNA damage, it also regulates \nmitochondrial respiration – indeed, mice null for p53 have very poor exercise tolerance with early \nfatigue onset (Bartlett et al., 2014) . Though at  this stage highly speculative, it is possible that \nelevated p53 in immune cells from  rapamycin-treated participants may indicate better overall \nmetabolic health. In addition, enhancement of p53 expression has been shown recently to improve \nDNA repair after irradiation-induced senescence of human dermal fibroblasts (Miller et al., 2025). \nTherefore, while p53 was suppressed by rapamycin following acute DNA damage in vitro, our \nobservation that longer-term rapamycin administration in older individuals increases p53 levels \nmay reflect improved genome integrity.  \nOur findings allow us to speculate that the positive effect of  6-week treatment with the \nrapalogue RAD001 (everolimus) on boosting flu vaccine responses and respiratory infections may \nbe through an attenuation of immune cell DNA damage and subsequent senescence (Mannick et \nal., 2018, Mannick et al., 2014) . In the cited studies, e verolimus caused a reduction in the \nproportion of circulating PD -1+ CD4+ and CD8+ T cells (Mannick et al., 2014) . In our study, we \nobserved a significant reduction in both KLRG1+ and NKG2A+ CD4+ T cells, and near-significantly \nLAG3+ CD4+ T cells in the rapamycin compared to placebo groups. Like PD -1, these three cell -\nsurface proteins are all immune checkpoint inhibitors, each with roles in limiting T cell activation. \nTherefore, we observed similar functional effects of rapamycin as perhaps potentiating a less-\nexhausted T cell phenotype. Overall, based on our in vitro effects mTOR inhibitors and in vivo \neffects of rapamycin, we suggest that rapamycin positively enhances genome stability, a nd \ntherefore targets a central hallmark of ageing (Lopez-Otin et al., 2023). \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 15 \nOur discovery that mTOR inhibitors are genoprotective make them amenable for use in a \nwide range of clinical scenarios where the induction of DNA damage leads to pathology. For \nexample, cancer treatments such as radio- or chemotherapy lead to widespread DNA damage of \nhealthy tissue; thus, treatment with a genoprotector, such as low-dose rapamycin, after remission \nfrom the original tumour may attenuate the accelerated ageing associated with such cancer \ntherapies (Wang et al., 2024) . Likewise, exposure to the space environment, and especially the \nDNA instability caused by cosmic radiation, is of increasing concern as space travel becomes more \ncommonplace (Beheshti et al., 2021). Our study, through its novel identification of rapamycin as a \ngenoprotector, suggests potential avenues for mitigating these harmful DNA-damaging effects of \nspace travel.  \nGenoprotectors such as rapamycin present a new and exciting therapeutic approach for \nthe treatment of age -related diseases, both infectious and chronic in nature.  SARS-CoV-2, the \nvirus behind the COVID-19 pandemic, induces DNA damage and senescence by degrading DDR \nenzymes (Gioia et al., 2023); heightened virus-induced senescence in this way strongly contributes \nto disease mortality (Camell et al., 2021, Lee et al., 2021). Perhaps prophylactic treatment of older \ncare home residents with genoprotectors such as low-dose mTOR inhibitors may provide a much-\nneeded boost to genome stability and immune resilience in this vulnerable population during future \npandemics (Cox et al., 2020). Likewise, since mTOR inhibitors improve vaccine responses in older \npeople (Mannick et al., 2018, Mannick et al., 2021, Mannick et al., 2014) , future vaccine drives \ncould consider administering short-term mTOR inhibition treatments prior to immunisations against \npathogens that particularly affect the older population, such as influenza, coronaviruses, and VZV. \nInfections by other pathogens, such as Salmonella Typhi, Leishmania, and some Gram-negative \nbacteria (which release cytolethal distending toxin), all drive pathology through the induction of \nDNA damage and senescence (Ibler et al., 2019, Mathiasen et al., 2021, Covre et al., 2018); low-\ndose mTOR inhibition in these contexts of infection could possibly limit genome instability and \ndisease progression. Though largely untested in humans, research in mice at least suggests that \nrapamycin improves immune control of Leishmaniasis (Khadir et al., 2018). Finally, in addition to \nprogeroid diseases resulting from a DNA repair deficiency (Werner syndrome, Rothmund -\nThomson syndrome, Bloom syndrome, Cockayne’s syndrome, Fanconi’s anaemia, and Ataxia -\nTelangiectasia), chronic viral infection and rheumatic diseases exhibit decreased DNA repair factor \nexpression in immune cells (Zhao et al., 2018, Shao et al., 2009, Li et al., 2016) . It is possible, \nthough unexplored, that rapamycin could limit DNA damage and attenuate pathology in these \ndiseases.  \nGiven the known physiological roles of senescent cells and DNA damage, caution must be \ntaken in administering genoprotective low -dose mTOR inhibitors (de Magalhaes, 2024) . For \nexample, inhibiting seno -conversion of virus -infected cells may disrupt their removal by the \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 16 \nimmune system. Additionally, genoprotectors may disrupt the intentional induction of DNA damage \nby adaptive immune cells during VDJ recombination, potentially leading to immunodeficiency; \nhowever, as the thymus (where T cell VDJ recombination occurs) atrop hies with age, it is likely \nthat late-life administration of low-dose rapamycin would not impact on T cells, but possibly B cell \nmaturation. Finally, senescent cells play critical roles in tissue regeneration in response to damage \n(Chen et al., 2023) ; thus, genoprotectors may have unforeseen consequences such as by \nhindering wound healing, a process which is already impaired in older people (Demaria et al., \n2014, Wicke et al., 2009). \nTaken together, our findings of immune cell benefit on rapamycin treatment in vitro and \nfrom analysis of PBMCs from an in vivo  low-dose rapamycin trial, lead us to conclude that \nrapamycin at 1 mg/day enhances the resilience of the ageing immune system to DN A damage. \nOur findings support the initiation of phase 2 double-blind placebo-controlled studies of rapamycin \nto support healthy immunity and reduce immunosenescence in at-risk older adults. \n \n  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 17 \nMATERIALS AND METHODS \nEthical approval for study \nHealthy control blood was taken with fully informed consent under ethical approval from \nthe Local Research Ethics Committee (REC) in Oxford, reference 11/H0711/7, to cover the use of \nhuman blood products purchased from National Health Services Blood and Tr ansplant service \n(NHS England). PBMCs from participants of the Rapamune Trial (registered on Clinical Trials.gov, \nnumber NCT05414292) were obtained under ethical approval by the University of Nottingham \nFMHS REC, reference FMHS 90-0820. \n \nIn vivo rapamycin treatment \nIn the ongoing single-blind, placebo-controlled Rapamune trial (NCT05414292), older male \nparticipants (between 50-90 years old) are randomised into two groups and receive either 1 mg/day \nrapamycin (Pfizer, Belgium) or a placebo sucrose/lactose tablet (Hsconline). The primary outcome \nis change in muscle mass in response to unilateral leg extension resistance exercise training (RET) \n3 times per week for 14 weeks at 75% of the participant’s 1 repetition maximum. Secondary \noutcomes include changes in muscle str ength, power, function, neuromuscular function, and \nassessment of muscle biopsies on protein synthesis and degradation. In this study, PBMCs were \nisolated at 5 time points over 4 months of intervention (rapamycin and placebo). After 8 weeks of \nintervention, the blood concentration of rapamycin and white blood cell counts were measured by \nProf. Dan Wilkinson at the University of Nottingham. Inclusion criteria for the trial were that the \nparticipant could give informed consent to participate in the study, and  that they were able to \ncomplete the RET. Exclusion criteria included having a BMI <18 or >35 kg/m 2, active \ncardiovascular, cerebrovascular, respiratory, or metabolic disease, clotting dysfunction, having a \nhistory of neurological or musculoskeletal conditions, having taken part in a recent study in the last \n3 months, or contraindications either to MRI scanning or rapamycin.  White blood cell \nconcentrations were quantified in the Royal Derby Hospital Pathology laboratory. \n \nMeasurement of blood rapamycin concentration using LC-MS \n50 µl of D3-labelled rapamycin was added to 50 µl of whole blood, before adding 100 µl of \nprecipitation reagent (70:30, Methanol:0.3M Zinc Sulfate). Samples were vortexed for 30 s and \nmixed on a Vibrax shaker at RT (1000 rpm) for 10 mins. Samples were then centrifuged at 10000g \nfor 10 mins at 4 °C. Supernatant was aliquoted into a 2 ml screw top autosampler vial with low \nvolume insert, before injection into the LC-MS. Samples were quantified against a standard curve \nof known rapamycin concentrations ranging from 50 ng/ml to 0.39ng/ml prepared in the same way \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 18 \nas the samples. Analysis was performed using a Waters ACQUITY UHPLC attached to a Thermo \nScientific TSQ Quantum Ultra MS. Rapamycin was isolated using an Agilent Zorbax SB-Aq Narrow \nBore RR Column (2.1mm x 100mm x 3.5µm) and a binary buffer system of 2 mM Ammonium \nAcetate in Water (Buffer A) and 2mM Ammonium Acetate in Methanol (Buffer B) at a flow rate of \n0.3 ml/min. Gradient conditions were as follows: 80%B for 0.5 mins, 80%B to 90% B 0.5 -2 mins, \n99% B 2 – 6 mins, 99%B to 80% 6-6.5 mins, 80% B 6.5-10 mins. Rapamycin was detected using \nsingle reaction monitoring (SRM) for m/z transitions of 931.6 m/z – 864.66m/z for unlabelled \nrapamycin and 934.520 – 864.660 for D3-labelled rapamycin. \n \nPBMC isolation and culture \nFresh blood was either collected in EDTA tubes (9 ml) or in blood cones (10 ml) as \nconcentrated by-products of the apheresis process, supplied by the National Health Service Blood \nand Transplant service (NHS England). PBMCs were isolated using standard Fic oll density \ngradient centrifugation. Briefly, blood was diluted 1:1 in sterile Dulbecco’s PBS (DPBS, 1:5 for \nblood cones) (Sigma -Aldrich) and 15 ml gently pipetted over 20 ml Histopaque -1077 (Sigma) \nbefore centrifugation at 500g for 30 mins at room tempera ture with minimum deceleration. The \nPBMC layer was collected by aspiration and washed twice in DPBS. PBMC number was \ndetermined by mixing 1:1 (v/v) with Trypan Blue (Sigma) and counting using a haemocytometer. \nFor cryopreservation, PBMCs were resuspended at 5x106 cells/ml in freezing medium (50% FBS, \n40% RPMI 1640, 10% DMSO (all Sigma)) and placed at -80°C before transferral to a liquid nitrogen \nfacility (-196°C) for long-term storage. Peripheral blood mononuclear cells (PBMCs) were cultured \nin R10 (RPMI 16 40 (Gibco) containing penicillin (100 U/ml) and streptomycin (100 µg/ml) (both \nSigma) and 10% FBS), in a humidified incubator with 5% CO2 at 37°C. Details of the drugs used \nfor in vitro assays are provided in Table S1. \n \nDNA damage assay in PBMCs \nCryopreserved PBMCs were thawed in R10 (10 ml per 1 cryovial of cells) and centrifuged \nat 500g for 5 mins. The supernatant was removed, and cells resuspended in R10 at a concentration \nof 1x106 cells/ml and allowed to recover overnight in R10. The next day, PBMCs were subjected \nto T cell-specific activation using antibodies targeting CD3 (clone OKT3) and CD28 (clone CD28.2) \nat 1 µg/ml final concentration (both BioLegend) in the presence of drug treatment or vehicle, where \nappropriate, for 3 days. Where specified for individual experiments, CD4+ T cells were isolated by \nnegative selection on a magnetic column using a kit (Miltenyi) and subsequently activated in wells \nof a 24-well plate (pre-coated with 1 µg/ml anti-CD3 in PBS for 2 h at 37°C) with 1 µg/ml anti-CD28 \nin the culture media, for 3 day s (0.5 -1x106 cells/ml). After 3 days, cells were harvested by \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 19 \ntrituration, centrifuged at 500g for 5 mins, and resuspended and incubated in R10 with 200 µg/ml \nzeocin for 2 h or 25 µM H2O2 for 15 mins (Table S1), in a 24-well plate. As both zeocin and H2O2 \nare soluble directly in water and cell culture medium, negative controls were not treated with \nsolvent. Cells were then washed in R10 and allowed to recover for a defined period (see individual \nfigures) before being collected for further downstream analysis. Where specified, the DNA damage \nassay was performed in the presence of rapamycin  (10 nM), AZD8055 (100 nM), or chloroquine \n(10 µM) (drug details provided in Table S1). \n \nStaining for conventional flow cytometry \nCells were pelleted at 500g for 5 mins and resuspended in 50 µl of a master mix of cell -\nsurface-staining antibodies diluted in FACS buffer (0.2% BSA (w/v), 2 mM EDTA in PBS) and \nZombie NIR Live/Dead viability dye (1/1000 dilution, BioLegend) with incubation for 30 minutes at \n4°C. Cells were washed in FACS buffer and fixed for 20 minutes at 4°C in BD Cytofix Fixation \nBuffer (BD Biosciences). Permeabilisation of cells was performed by washing cells in 1X BD \nPhosflow Perm/Wash Buffer I (BD Biosciences), follow ed by incubation in the permeabilisation \nbuffer for 10 minutes at RT in the dark. Intracellular antibody staining was performed overnight at \n4°C in the dark in BD Phosflow Perm/Wash Buffer I. When necessary, staining of unconjugated \nprimary antibodies with  a fluorescence -conjugated secondary antibody was performed in BD \nPhosflow Perm/Wash Buffer I for 1 h at RT in the dark. Cells were washed once more and stored \nin FACS buffer prior to acquisition on a flow cytometer (BD LSRFortessa)  and analysis using \nFlowJo version 10.9.0 (gating strategy in Figure S1). Compensation analysis was performed using \nsingle-stained compensation beads (Thermo Fisher or BioLegend). Details of antibodies used for \nflow cytometry staining are provided in Table S2. \n \nStaining for spectral flow cytometry \nPBMCs were stained for analysis for spectral flow cytometry as above, with some \nmodifications. First, cells were harvested and pelleted (500g for 5 mins) and incubated in 50 µl \nsolution containing LIVE/DEAD™ Fixable Blue Dead Cell Stain Kit (1/400 dilution, Invitrogen) and \nFcR blocking reagent (1/400 dilution, Miltenyi) in PBS for 15 minutes at 4°C. Surface and \nintracellular antigen staining was then performed as above. After staining, cells were filtered \nthrough a 70 µm Flowmi cell strainer (Sigma) to remove aggregates, before analysis on a 5-laser \n(R/B/V/YG/UV) Aurora spectral flow cytometer (Cytek Biosciences). For each experiment, all \nsamples were processed in one batch and single -colour cell - and bead -based controls were \ngenerated in parallel alongside the sample staining for spectral unmixing.  Details of antibodies \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 20 \nused for spectral flow cytometry staining are provided in Table S2. The gating strategy is provided \nin Figure S4. \n \nAutophagic flux analysis \nMeasurement of autophagic flux in cells was performed using an antibody -based LC3 \nassay kit (Cytek Biosciences) with the following modifications  (Figure S3). 2 hours prior to LC3 \nstaining, each sample of cells was divided for treatment either with 10 nM Bafilomycin A 1 (Table \nS1) or vehicle (0.1% DMSO final concentration) in R10. Cells were harvested and incubated in \nLive/Dead viability dye (1/1000 dilution, BioLegend) and FcR blocking reagent (1/400 dilution, \nMiltenyi) in PBS for 15 minutes at 4°C, then sur face staining with antibodies was performed as \ndescribed above, followed by washing in 1X assay buffer (Cytek Biosciences) diluted in dH2O, and \nsubsequently permeabilised using 0.05% saponin (w/v) in PBS for 3 minutes at RT. Cells were \nthen incubated with an LC3-FITC conjugated antibody (Table S2) in 1X assay buffer for 30 mins \nat 4°C. For co-staining of LC3 with an anti -γH2AX antibody (Table S2), the incubation time was \nincrease to 1 hour in 1X assay buffer. After staining, cells were then washed in 1X assay buffer \nand fixed in 2% PFA for 10 minutes at RT, before being finally resuspended in FACS buffer and \nanalysed on an LSRFortessa cytometer (B D) or 5 -laser Aurora spectral flow cytometer (Cytek)  \n(Figure S3). Autophagic flux was calculated from the mean fluorescence intensity (MFI) of LC3 in \nBafilomycin A1- and vehicle-treated conditions (LC3BafA and LC3Veh respectively) using the formula: \n𝐴𝑢𝑡𝑜𝑝ℎ𝑎𝑔𝑖𝑐\t𝑓𝑙𝑢𝑥 = 𝐿𝐶3!\"#$ − 𝐿𝐶3%&'\n𝐿𝐶3%&'\n \n \nCell cycle phase distribution analysis \nTo assess cell cycle phase distribution using flow cytometry, cells were incubated for 4 h \nwith 10 µM  of 5-ethynyl-2′-deoxyuridine (EdU) (Invitrogen). Cells were then harvested and \nincubated in Live/Dead viability dye (1/1000 dilution, BioLegend) and FcR blocking reagent (1/400 \ndilution, Miltenyi) in PBS for 15 minutes at 4°C. After this, surface antigen stai ning for flow \ncytometry was performed as described above. DNA -incorporated EdU was identified via a click \nchemistry reaction according to the manufacturer’s instructions (Invitrogen), followed by overnight \nstaining for intracellular γH2AX as described above. Cells were then washed and incubated with \nFxCycle™ Violet Stain for DNA (1/1000 dilution, Invitrogen) in permeabilisation buffer (from EdU \nkit, Invitrogen) and analysed on an LSR Fortessa X-20 cytometer (BD). \n \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 21 \nProtein synthesis assay \nNascent protein synthesis was measured using the Click-iT™ Plus OPP Alexa Fluor™ \n594 Protein Synthesis Assay Kit (Invitrogen). Cells were incubated with 20 µM O-propargyl-\npuromycin (OPP, from the kit) for 30 mins, then collected and stained for surface antigens. Cells \ntreated for 1 h with 50 µg/ml cycloheximide (Table S1) served as a positive control for protein \nsynthesis inhibition. Cells were subsequently fixed, permeabilised, and polypeptide-incorporated \nOPP was detected, according to the manufacturer’s instructions. Cells were analysed on a \nFortessa X-20 flow cytometer (BD).  \n \nAlkaline comet assay \nSuperFrost Plus Adhesion slides (VWR) were first pre -coated with normal melting point \nagarose (NMPA) (1% (w/v) in dH 2O, Sigma) and allowed to dry overnight. Following treatments, \ncells were harvested by trituration and resuspended in PBS at a concentration of 2x10 5 cells/ml. \n250 µl of the cell suspension was mixed with 1 ml of low melting point agarose (1% w/v in PBS, \nSigma) pre-warmed to 37°C, and 1 ml of the mixture was pipetted onto one NMPA -coated slide. \nCoverslips were then placed on the slides and lef t to gel on ice. For the lysis of cells, coverslips \nwere removed, and slides were incubated overnight at 4°C in fresh, ice -cold lysis buffer (2.5 M \nNaCl, 100 mM EDTA, 10 mM Tris base containing 1% DMSO (v/v) and 1% Triton X-100 (v/v), pH \n10.5). The next da y, slides were incubated for 30 mins in the dark with fresh alkaline \nelectrophoresis buffer (300 mM NaOH, 1mM EDTA and 1% DMSO (v/v), pH > 13). Slides were \nthen electrophoresed in the dark for 25 mins at 1 V/cm (distance between electrodes), at a constant \ncurrent of 300 mA. Slides were then neutralised with 3 x 5 mins incubations in neutralisation buffer \n(500 mM Tris -HCl, pH 8.1) and subsequently left to dry overnight. The next day, slides were \nrehydrated for 30 mins in dH2O, and DNA was stained for 30 mins with 1X SYBR™ Gold Nucleic \nAcid Gel Stain (Invitrogen) in dH2O. Comets were visualised with a Zeiss Axio Imager and analysed \nusing the OpenComet plugin in Fiji version 2.3.0. \n \nStatistical tests and figures \nAll statistical and (log)normality testing of data was performed using GraphPad Prism \nversion 10.0.0. P-values indicating statistical significance are either indicated exactly or \nrepresented as: ns (not significant) p > 0.05, * p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001. \nUnless otherwise stated, bar graph data are always represented as mean ± SEM. Box-and-whisker \nplots always show minimum to maximum values, with the median, 25 th, and 75 th percentiles \nindicated. Figures were made using Microsoft PowerPoint version 16.88. \n  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 22 \nREFERENCES AND NOTES \n \nALSALEH, G., PANSE, I., SWADLING, L., ZHANG, H., RICHTER, F. C., MEYER, A., LORD, J., \nBARNES, E., KLENERMAN, P., GREEN, C. & SIMON, A. K. 2020. Autophagy in T cells \nfrom aged donors is maintained by spermidine and correlates with function and vaccine \nresponses. Elife, 9. \nALTOMARE, J. F., SMITH, R. E., POTDAR, S. & MITCHELL, S. H. 2006. Delayed gastric ulcer \nhealing associated with sirolimus. Transplantation, 82, 437-8. \nBARTLETT, J. D., CLOSE, G. L., DRUST, B. & MORTON, J. P. 2014. The emerging role of p53 \nin exercise metabolism. Sports Med, 44, 303-9. \nBEHESHTI, A., MCDONALD, J. T., HADA, M., TAKAHASHI, A., MASON, C. E. & MOGNATO, \nM. 2021. Genomic Changes Driven by Radiation-Induced DNA Damage and Microgravity \nin Human Cells. Int J Mol Sci, 22. \nBJEDOV, I., TOIVONEN, J. M., KERR, F., SLACK, C., JACOBSON, J., FOLEY, A. & \nPARTRIDGE, L. 2010. Mechanisms of life span extension by rapamycin in the fruit fly \nDrosophila melanogaster. Cell Metab, 11, 35-46. \nBONASSI, S., CEPPI, M., MOLLER, P., AZQUETA, A., MILIC, M., NERI, M., BRUNBORG, G., \nGODSCHALK, R., KOPPEN, G., LANGIE, S. A. S., TEIXEIRA, J. P., BRUZZONE, M., \nDA SILVA, J., BENEDETTI, D., CAVALLO, D., URSINI, C. L., GIOVANNELLI, L., \nMORETTI, S., RISO, P., DEL BO, C., RUSSO, P., DOBRZYNSKA, M., \nGOROSHINSKAYA, I. A., SURIKOVA, E. I., STARUCHOVA, M., BARANCOKOVA, M., \nVOLKOVOVA, K., KAZIMIROVA, A., SMOLKOVA, B., LAFFON, B., VALDIGLESIAS, V., \nPASTOR, S., MARCOS, R., HERNANDEZ, A., GAJSKI, G., SPREMO-POTPAREVIC, \nB., ZIVKOVIC, L., BOUTET-ROBINET, E., PERDRY, H., LEBAILLY, P., PEREZ, C. L., \nBASARAN, N., NEMETH, Z., SAFAR, A., DUSINSKA, M., COLLINS, A. & H, C. P. 2021. \nDNA damage in circulating leukocytes measured with the comet assay may predict the \nrisk of death. Sci Rep, 11, 16793. \nBRAUN, F., MANDEL, A. M., BLOMBERG, L., WONG, M. N., CHATZINIKOLAOU, G., MEYER, \nD. H., REINELT, A., NAIR, V., AKBAR-HAASE, R., MCCOWN, P. J., HAAS, F., CHEN, \nH., RAHMATOLLAHI, M., FERMIN, D., EBBESTAD, R., SLAATS, G. G., BORK, T., \nSCHELL, C., KOEHLER, S., BRINKKOETTER, P. T., LINDENMEYER, M. T., COHEN, \nC. D., KANN, M., UNNERSJO-JESS, D., BLOCH, W., SAMPSON, M. G., DOLLE, M. E., \nPUELLES, V. G., KRETZLER, M., GARINIS, G. A., HUBER, T. B., SCHERMER, B., \nBENZING, T., SCHUMACHER, B. & KURSCHAT, C. E. 2025. Loss of genome \nmaintenance is linked to mTOR complex 1 signaling and accelerates podocyte damage. \nJCI Insight, 10. \nCALLENDER, L. A., CARROLL, E. C., BOBER, E. A., AKBAR, A. N., SOLITO, E. & HENSON, S. \nM. 2020. Mitochondrial mass governs the extent of human T cell senescence. Aging Cell, \n19, e13067. \nCAMELL, C. D., YOUSEFZADEH, M. J., ZHU, Y., LANGHI PRATA, L. G. P., HUGGINS, M. A., \nPIERSON, M., ZHANG, L., O'KELLY, R. D., PIRTSKHALAVA, T., XUN, P., EJIMA, K., \nXUE, A., TRIPATHI, U., MACHADO ESPINDOLA-NETTO, J., GIORGADZE, N., \nATKINSON, E. J., INMAN, C. L., JOHNSON, K. O., CHOLENSKY, S. H., CARLSON, T. \nW., LEBRASSEUR, N. K., KHOSLA, S., O'SULLIVAN, M. G., ALLISON, D. B., \nJAMESON, S. C., MEVES, A., LI, M., PRAKASH, Y. S., CHIARELLA, S. E., HAMILTON, \nS. E., TCHKONIA, T., NIEDERNHOFER, L. J., KIRKLAND, J. L. & ROBBINS, P. D. \n2021. Senolytics reduce coronavirus-related mortality in old mice. Science. \nCARROLL, B., NELSON, G., RABANAL-RUIZ, Y., KUCHERYAVENKO, O., DUNHILL-TURNER, \nN. A., CHESTERMAN, C. C., ZAHARI, Q., ZHANG, T., CONDUIT, S. E., MITCHELL, C. \nA., MADDOCKS, O. D. K., LOVAT, P., VON ZGLINICKI, T. & KOROLCHUK, V. I. 2017. \nPersistent mTORC1 signaling in cell senescence results from defects in amino acid and \ngrowth factor sensing. J Cell Biol, 216, 1949-1957. \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 23 \nCHEBEL, A., CATALLO, R., MABON, C., BACHY, E., WENNER, T., SALLES, G., POUTEIL-\nNOBLE, C. & FFRENCH, M. 2016. Rapamycin safeguards lymphocytes from DNA \ndamage accumulation in vivo. Eur J Cell Biol, 95, 331-41. \nCHEN, C., SACLIER, M., CHANTREL, J., MELLA, S., CHICHE, A. & LI, H. 2023. Single-cell \nRNAseq reveals the pro-regenerative role of senescent FAPs in muscle regeneration. \nbioRxiv. \nCHUNG, C. L., LAWRENCE, I., HOFFMAN, M., ELGINDI, D., NADHAN, K., POTNIS, M., JIN, \nA., SERSHON, C., BINNEBOSE, R., LORENZINI, A. & SELL, C. 2019. Topical \nrapamycin reduces markers of senescence and aging in human skin: an exploratory, \nprospective, randomized trial. Geroscience, 41, 861-869. \nCOLONNA-ROMANO, G., BULATI, M., AQUINO, A., PELLICANO, M., VITELLO, S., LIO, D., \nCANDORE, G. & CARUSO, C. 2009. A double-negative (IgD-CD27-) B cell population is \nincreased in the peripheral blood of elderly people. Mech Ageing Dev, 130, 681-90. \nCOVRE, L. P., MARTINS, R. F., DEVINE, O. P., CHAMBERS, E. S., VUKMANOVIC-STEJIC, \nM., SILVA, J. A., DIETZE, R., RODRIGUES, R. R., DE MATOS GUEDES, H. L., \nFALQUETO, A., AKBAR, A. N. & GOMES, D. C. O. 2018. Circulating Senescent T Cells \nAre Linked to Systemic Inflammation and Lesion Size During Human Cutaneous \nLeishmaniasis. Front Immunol, 9, 3001. \nCOX, L. S., BELLANTUONO, I., LORD, J. M., SAPEY, E., MANNICK, J. B., PARTRIDGE, L., \nGORDON, A. L., STEVES, C. J. & WITHAM, M. D. 2020. Tackling immunosenescence to \nimprove COVID-19 outcomes and vaccine response in older adults. Lancet Healthy \nLongev, 1, e55-e57. \nCZESNIKIEWICZ-GUZIK, M., LEE, W. W., CUI, D., HIRUMA, Y., LAMAR, D. L., YANG, Z. Z., \nOUSLANDER, J. G., WEYAND, C. M. & GORONZY, J. J. 2008. T cell subset-specific \nsusceptibility to aging. Clin Immunol, 127, 107-18. \nDE MAGALHAES, J. P. 2024. Cellular senescence in normal physiology. Science, 384, 1300-\n1301. \nDEMARIA, M., OHTANI, N., YOUSSEF, S. A., RODIER, F., TOUSSAINT, W., MITCHELL, J. R., \nLABERGE, R. M., VIJG, J., VAN STEEG, H., DOLLE, M. E., HOEIJMAKERS, J. H., DE \nBRUIN, A., HARA, E. & CAMPISI, J. 2014. An essential role for senescent cells in \noptimal wound healing through secretion of PDGF-AA. Dev Cell, 31, 722-33. \nDESDIN-MICO, G., SOTO-HEREDERO, G., ARANDA, J. F., OLLER, J., CARRASCO, E., \nGABANDE-RODRIGUEZ, E., BLANCO, E. M., ALFRANCA, A., CUSSO, L., DESCO, M., \nIBANEZ, B., GORTAZAR, A. R., FERNANDEZ-MARCOS, P., NAVARRO, M. N., \nHERNAEZ, B., ALCAMI, A., BAIXAULI, F. & MITTELBRUNN, M. 2020. T cells with \ndysfunctional mitochondria induce multimorbidity and premature senescence. Science, \n368, 1371-1376. \nFRASCA, D., DIAZ, A., ROMERO, M. & BLOMBERG, B. B. 2017. Human peripheral \nlate/exhausted memory B cells express a senescent-associated secretory phenotype and \npreferentially utilize metabolic signaling pathways. Exp Gerontol, 87, 113-120. \nGIOIA, U., TAVELLA, S., MARTINEZ-ORELLANA, P., CICIO, G., COLLIVA, A., CECCON, M., \nCABRINI, M., HENRIQUES, A. C., FUMAGALLI, V., PALDINO, A., PRESOT, E., \nRAJASEKHARAN, S., IACOMINO, N., PISATI, F., MATTI, V., SEPE, S., CONTE, M. I., \nBAROZZI, S., LAVAGNINO, Z., CARLETTI, T., VOLPE, M. C., CAVALCANTE, P., \nIANNACONE, M., RAMPAZZO, C., BUSSANI, R., TRIPODO, C., ZACCHIGNA, S., \nMARCELLO, A. & D'ADDA DI FAGAGNA, F. 2023. SARS-CoV-2 infection induces DNA \ndamage, through CHK1 degradation and impaired 53BP1 recruitment, and cellular \nsenescence. Nat Cell Biol. \nGKIONI, L., NESPITAL, T., BAGHDADI, M., MONZÓ, C., BALI, J., NASSR, T., CREMER, A. L., \nBEYER, A., DEELEN, J., BACKES, H., GRÖNKE, S. & PARTRIDGE, L. 2025. The \ngeroprotectors trametinib and rapamycin combine additively to extend mouse healthspan \nand lifespan. Nat Aging, 5, 1249-1265. \nHA, C. W. & HUH, W. K. 2011. Rapamycin increases rDNA stability by enhancing association of \nSir2 with rDNA in Saccharomyces cerevisiae. Nucleic Acids Res, 39, 1336-50. \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 24 \nHARRISON, D. E., STRONG, R., SHARP, Z. D., NELSON, J. F., ASTLE, C. M., FLURKEY, K., \nNADON, N. L., WILKINSON, J. E., FRENKEL, K., CARTER, C. S., PAHOR, M., \nJAVORS, M. A., FERNANDEZ, E. & MILLER, R. A. 2009. Rapamycin fed late in life \nextends lifespan in genetically heterogeneous mice. Nature, 460, 392-5. \nHAZELDINE, J., HAMPSON, P. & LORD, J. M. 2012. Reduced release and binding of perforin at \nthe immunological synapse underlies the age-related decline in natural killer cell \ncytotoxicity. Aging Cell, 11, 751-9. \nHEARPS, A. C., MARTIN, G. E., ANGELOVICH, T. A., CHENG, W. J., MAISA, A., LANDAY, A. \nL., JAWOROWSKI, A. & CROWE, S. M. 2012. Aging is associated with chronic innate \nimmune activation and dysregulation of monocyte phenotype and function. Aging Cell, \n11, 867-75. \nHOUDE, V. P., BRULE, S., FESTUCCIA, W. T., BLANCHARD, P. G., BELLMANN, K., \nDESHAIES, Y. & MARETTE, A. 2010. Chronic rapamycin treatment causes glucose \nintolerance and hyperlipidemia by upregulating hepatic gluconeogenesis and impairing \nlipid deposition in adipose tissue. Diabetes, 59, 1338-48. \nIBLER, A. E. M., ELGHAZALY, M., NAYLOR, K. L., BULGAKOVA, N. A., S, F. E.-K. & \nHUMPHREYS, D. 2019. Typhoid toxin exhausts the RPA response to DNA replication \nstress driving senescence and Salmonella infection. Nat Commun, 10, 4040. \nIGLESIAS-BARTOLOME, R., PATEL, V., COTRIM, A., LEELAHAVANICHKUL, K., MOLINOLO, \nA. A., MITCHELL, J. B. & GUTKIND, J. S. 2012. mTOR inhibition prevents epithelial stem \ncell senescence and protects from radiation-induced mucositis. Cell Stem Cell, 11, 401-\n14. \nKELL, L., SIMON, A. K., ALSALEH, G. & COX, L. S. 2023. The central role of DNA damage in \nimmunosenescence. Front Aging, 4, 1202152. \nKHADIR, F., SHALER, C. R., ORYAN, A., RUDAK, P. T., MAZZUCA, D. M., TAHERI, T., \nDIKEAKOS, J. D., HAERYFAR, S. M. M. & RAFATI, S. 2018. Therapeutic control of \nleishmaniasis by inhibitors of the mammalian target of rapamycin. PLoS Negl Trop Dis, \n12, e0006701. \nKNIGHT, R. J., VILLA, M., LASKEY, R., BENAVIDES, C., SCHOENBERG, L., WELSH, M., \nKERMAN, R. H., PODDER, H., VAN BUREN, C. T., KATZ, S. M. & KAHAN, B. D. 2007. \nRisk factors for impaired wound healing in sirolimus-treated renal transplant recipients. \nClin Transplant, 21, 460-5. \nLEE, S., YU, Y., TRIMPERT, J., BENTHANI, F., MAIRHOFER, M., RICHTER-PECHANSKA, P., \nWYLER, E., BELENKI, D., KALTENBRUNNER, S., PAMMER, M., KAUSCHE, L., \nFIRSCHING, T. C., DIETERT, K., SCHOTSAERT, M., MARTINEZ-ROMERO, C., \nSINGH, G., KUNZ, S., NIEMEYER, D., GHANEM, R., SALZER, H. J. F., PAAR, C., \nMULLEDER, M., UCCELLINI, M., MICHAELIS, E. G., KHAN, A., LAU, A., SCHONLEIN, \nM., HABRINGER, A., TOMASITS, J., ADLER, J. M., KIMESWENGER, S., GRUBER, A. \nD., HOETZENECKER, W., STEINKELLNER, H., PURFURST, B., MOTZ, R., DI \nPIERRO, F., LAMPRECHT, B., OSTERRIEDER, N., LANDTHALER, M., DROSTEN, C., \nGARCIA-SASTRE, A., LANGER, R., RALSER, M., EILS, R., REIMANN, M., FAN, D. N. \nY. & SCHMITT, C. A. 2021. Virus-induced senescence is a driver and therapeutic target \nin COVID-19. Nature, 599, 283-289. \nLI, Y., SHEN, Y., HOHENSINNER, P., JU, J., WEN, Z., GOODMAN, S. B., ZHANG, H., \nGORONZY, J. J. & WEYAND, C. M. 2016. Deficient Activity of the Nuclease MRE11A \nInduces T Cell Aging and Promotes Arthritogenic Effector Functions in Patients with \nRheumatoid Arthritis. Immunity, 45, 903-916. \nLIBRI, V., AZEVEDO, R. I., JACKSON, S. E., DI MITRI, D., LACHMANN, R., FUHRMANN, S., \nVUKMANOVIC-STEJIC, M., YONG, K., BATTISTINI, L., KERN, F., SOARES, M. V. & \nAKBAR, A. N. 2011. Cytomegalovirus infection induces the accumulation of short-lived, \nmultifunctional CD4+CD45RA+CD27+ T cells: the potential involvement of interleukin-7 \nin this process. Immunology, 132, 326-39. \nLIGOTTI, M. E., ACCARDI, G., AIELLO, A., APRILE, S., CALABRO, A., CALDARELLA, R., \nCARUSO, C., CIACCIO, M., CORSALE, A. M., DIELI, F., DI SIMONE, M., GIAMMANCO, \nG. M., MASCARELLA, C., AKBAR, A. N., MERAVIGLIA, S. & CANDORE, G. 2023. \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 25 \nSicilian semi- and supercentenarians: identification of age-related T-cell \nimmunophenotype to define longevity trait. Clin Exp Immunol, 214, 61-78. \nLOPEZ-OTIN, C., BLASCO, M. A., PARTRIDGE, L., SERRANO, M. & KROEMER, G. 2023. \nHallmarks of aging: An expanding universe. Cell, 186, 243-278. \nLOPEZ-VERGES, S., MILUSH, J. M., PANDEY, S., YORK, V. A., ARAKAWA-HOYT, J., \nPIRCHER, H., NORRIS, P. J., NIXON, D. F. & LANIER, L. L. 2010. CD57 defines a \nfunctionally distinct population of mature NK cells in the human CD56dimCD16+ NK-cell \nsubset. Blood, 116, 3865-74. \nMANNICK, J. B., DEL GIUDICE, G., LATTANZI, M., VALIANTE, N. M., PRAESTGAARD, J., \nHUANG, B., LONETTO, M. A., MAECKER, H. T., KOVARIK, J., CARSON, S., GLASS, \nD. J. & KLICKSTEIN, L. B. 2014. mTOR inhibition improves immune function in the \nelderly. Sci Transl Med, 6, 268ra179. \nMANNICK, J. B. & LAMMING, D. W. 2023. Targeting the biology of aging with mTOR inhibitors. \nNat Aging, 3, 642-660. \nMANNICK, J. B., MORRIS, M., HOCKEY, H. P., ROMA, G., BEIBEL, M., KULMATYCKI, K., \nWATKINS, M., SHAVLAKADZE, T., ZHOU, W., QUINN, D., GLASS, D. J. & \nKLICKSTEIN, L. B. 2018. TORC1 inhibition enhances immune function and reduces \ninfections in the elderly. Sci Transl Med, 10. \nMANNICK, J. B., TEO, G., BERNARDO, P., QUINN, D., RUSSELL, K., KLICKSTEIN, L., \nMARSHALL, W. & SHERGILL, S. 2021. Targeting the biology of ageing with mTOR \ninhibitors to improve immune function in older adults: phase 2b and phase 3 randomised \ntrials. Lancet Healthy Longev, 2, e250-e262. \nMARKOFSKI, M. M., DICKINSON, J. M., DRUMMOND, M. J., FRY, C. S., FUJITA, S., \nGUNDERMANN, D. M., GLYNN, E. L., JENNINGS, K., PADDON-JONES, D., REIDY, P. \nT., SHEFFIELD-MOORE, M., TIMMERMAN, K. L., RASMUSSEN, B. B. & VOLPI, E. \n2015. Effect of age on basal muscle protein synthesis and mTORC1 signaling in a large \ncohort of young and older men and women. Exp Gerontol, 65, 1-7. \nMATHIASEN, S. L., GALL-MAS, L., PATERAS, I. S., THEODOROU, S. D. P., NAMINI, M. R. J., \nHANSEN, M. B., MARTIN, O. C. B., VADIVEL, C. K., NTOSTOGLOU, K., BUTTER, D., \nGIVSKOV, M., GEISLER, C., AKBAR, A. N., GORGOULIS, V. G., FRISAN, T., ODUM, \nN. & KREJSGAARD, T. 2021. Bacterial genotoxins induce T cell senescence. Cell Rep, \n35, 109220. \nMILLER, K. N., LI, B., PIERCE-HOFFMAN, H. R., PATEL, S., LEI, X., RAJESH, A., TENECHE, \nM. G., HAVAS, A. P., GANDHI, A., MACIP, C. C., LYU, J., VICTORELLI, S. G., WOO, S. \nH., LAGNADO, A. B., LAPORTA, M. A., LIU, T., DASGUPTA, N., LI, S., DAVIS, A., \nKOROTKOV, A., HULTENIUS, E., GAO, Z., ALTMAN, Y., PORRITT, R. A., GARCIA, G., \nMOGLER, C., SELUANOV, A., GORBUNOVA, V., KAECH, S. M., TIAN, X., DOU, Z., \nCHEN, C., PASSOS, J. F. & ADAMS, P. D. 2025. p53 enhances DNA repair and \nsuppresses cytoplasmic chromatin fragments and inflammation in senescent cells. Nat \nCommun, 16, 2229. \nNEVALAINEN, T., AUTIO, A., KUMMOLA, L., SALOMAA, T., JUNTTILA, I., JYLHA, M. & \nHURME, M. 2019. CD27- IgD- B cell memory subset associates with inflammation and \nfrailty in elderly individuals but only in males. Immun Ageing, 16, 19. \nNYUGEN, J., AGRAWAL, S., GOLLAPUDI, S. & GUPTA, S. 2010. Impaired functions of \nperipheral blood monocyte subpopulations in aged humans. J Clin Immunol, 30, 806-13. \nPALACIOS-PEDRERO, M. A., JANSEN, J. M., BLUME, C., STANISLAWSKI, N., JONCZYK, R., \nMOLLE, A., HERNANDEZ, M. G., KAISER, F. K., JUNG, K., OSTERHAUS, A., \nRIMMELZWAAN, G. F. & SALETTI, G. 2022. Signs of immunosenescence correlate with \npoor outcome of mRNA COVID-19 vaccination in older adults. Nat Aging, 2, 896-905. \nPARK, J. H., LEE, N. K., LIM, H. J., JI, S. T., KIM, Y. J., JANG, W. B., KIM, D. Y., KANG, S., \nYUN, J., HA, J. S., KIM, H., LEE, D., BAEK, S. H. & KWON, S. M. 2020. Pharmacological \ninhibition of mTOR attenuates replicative cell senescence and improves cellular function \nvia regulating the STAT3-PIM1 axis in human cardiac progenitor cells. Exp Mol Med, 52, \n615-628. \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 26 \nPASSOS, J. F., SARETZKI, G., AHMED, S., NELSON, G., RICHTER, T., PETERS, H., \nWAPPLER, I., BIRKET, M. J., HAROLD, G., SCHAEUBLE, K., BIRCH-MACHIN, M. A., \nKIRKWOOD, T. B. & VON ZGLINICKI, T. 2007. Mitochondrial dysfunction accounts for \nthe stochastic heterogeneity in telomere-dependent senescence. PLoS Biol, 5, e110. \nRIDDELL, N. E., GRIFFITHS, S. J., RIVINO, L., KING, D. C., TEO, G. H., HENSON, S. M., \nCANTISAN, S., SOLANA, R., KEMENY, D. M., MACARY, P. A., LARBI, A. & AKBAR, A. \nN. 2015. Multifunctional cytomegalovirus (CMV)-specific CD8(+) T cells are not restricted \nby telomere-related senescence in young or old adults. Immunology, 144, 549-60. \nROLT, A., NAIR, A. & COX, L. S. 2019. Optimisation of a screening platform for determining IL-6 \ninflammatory signalling in the senescence-associated secretory phenotype (SASP). \nBiogerontology, 20, 359-371. \nSAHA, B., CYPRO, A., MARTIN, G. M. & OSHIMA, J. 2014. Rapamycin decreases DNA damage \naccumulation and enhances cell growth of WRN-deficient human fibroblasts. Aging Cell, \n13, 573-5. \nSEIDLER, S., ZIMMERMANN, H. W., BARTNECK, M., TRAUTWEIN, C. & TACKE, F. 2010. \nAge-dependent alterations of monocyte subsets and monocyte-related chemokine \npathways in healthy adults. BMC Immunol, 11, 30. \nSHAO, L., FUJII, H., COLMEGNA, I., OISHI, H., GORONZY, J. J. & WEYAND, C. M. 2009. \nDeficiency of the DNA repair enzyme ATM in rheumatoid arthritis. J Exp Med, 206, 1435-\n49. \nSOTO-HEREDERO, G., GOMEZ DE LAS HERAS, M. M., ESCRIG-LARENA, J. I. & \nMITTELBRUNN, M. 2023. Extremely Differentiated T Cell Subsets Contribute to Tissue \nDeterioration During Aging. Annu Rev Immunol, 41, 181-205. \nSTEIN, G. H., DRULLINGER, L. F., SOULARD, A. & DULIC, V. 1999. Differential roles for cyclin-\ndependent kinase inhibitors p21 and p16 in the mechanisms of senescence and \ndifferentiation in human fibroblasts. Mol Cell Biol, 19, 2109-17. \nSTERN, A. D., RAHMAN, A. H. & BIRTWISTLE, M. R. 2017. Cell size assays for mass \ncytometry. Cytometry A, 91, 14-24. \nTSAI, C. H., CHANG, C. Y., LIN, B. Z., WU, Y. L., WU, M. H., LIN, L. T., HUANG, W. C., HOLZ, \nJ. D., SHEU, T. J., LEE, J. S., KITSIS, R. N., TAI, P. H. & LEE, Y. J. 2021. Up-regulation \nof cofilin-1 in cell senescence associates with morphological change and p27(kip1) -\nmediated growth delay. Aging Cell, 20, e13288. \nVAN DEURSEN, J. M. 2014. The role of senescent cells in ageing. Nature, 509, 439-46. \nVESSONI, A. T., FILIPPI-CHIELA, E. C., MENCK, C. F. & LENZ, G. 2013. Autophagy and \ngenomic integrity. Cell Death Differ, 20, 1444-54. \nWALTERS, H. E. & COX, L. S. 2018. mTORC Inhibitors as Broad-Spectrum Therapeutics for \nAge-Related Diseases. Int J Mol Sci, 19. \nWALTERS, H. E., DENEKA-HANNEMANN, S. & COX, L. S. 2016. Reversal of phenotypes of \ncellular senescence by pan-mTOR inhibition. Aging (Albany NY), 8, 231-44. \nWANG, S., EL JURDI, N., THYAGARAJAN, B., PRIZMENT, A. & BLAES, A. H. 2024. \nAccelerated Aging in Cancer Survivors: Cellular Senescence, Frailty, and Possible \nOpportunities for Interventions. Int J Mol Sci, 25. \nWEICHHART, T. 2018. mTOR as Regulator of Lifespan, Aging, and Cellular Senescence: A \nMini-Review. Gerontology, 64, 127-134. \nWICKE, C., BACHINGER, A., COERPER, S., BECKERT, S., WITTE, M. B. & KONIGSRAINER, \nA. 2009. Aging influences wound healing in patients with chronic lower extremity wounds \ntreated in a specialized Wound Care Center. Wound Repair Regen, 17, 25-33. \nWILKINSON, J. E., BURMEISTER, L., BROOKS, S. V., CHAN, C. C., FRIEDLINE, S., \nHARRISON, D. E., HEJTMANCIK, J. F., NADON, N., STRONG, R., WOOD, L. K., \nWOODWARD, M. A. & MILLER, R. A. 2012. Rapamycin slows aging in mice. Aging Cell, \n11, 675-82. \nYANG, Q., XI, Q., WANG, M., LIU, J., LI, Z., HU, J., JIN, L. & ZHU, L. 2022. Rapamycin \nimproves the developmental competence of human oocytes by alleviating DNA damage \nduring IVM. Hum Reprod Open, 2022, hoac050. \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 27 \nYOUSEFZADEH, M., HENPITA, C., VYAS, R., SOTO-PALMA, C., ROBBINS, P. & \nNIEDERNHOFER, L. 2021a. DNA damage-how and why we age? Elife, 10. \nYOUSEFZADEH, M. J., FLORES, R. R., ZHU, Y., SCHMIECHEN, Z. C., BROOKS, R. W., \nTRUSSONI, C. E., CUI, Y., ANGELINI, L., LEE, K. A., MCGOWAN, S. J., BURRACK, A. \nL., WANG, D., DONG, Q., LU, A., SANO, T., O'KELLY, R. D., MCGUCKIAN, C. A., \nKATO, J. I., BANK, M. P., WADE, E. A., PILLAI, S. P. S., KLUG, J., LADIGES, W. C., \nBURD, C. E., LEWIS, S. E., LARUSSO, N. F., VO, N. V., WANG, Y., KELLEY, E. E., \nHUARD, J., STROMNES, I. M., ROBBINS, P. D. & NIEDERNHOFER, L. J. 2021b. An \naged immune system drives senescence and ageing of solid organs. Nature, 594, 100-\n105. \nZHANG, H., CICCHETTI, G., ONDA, H., KOON, H. B., ASRICAN, K., BAJRASZEWSKI, N., \nVAZQUEZ, F., CARPENTER, C. L. & KWIATKOWSKI, D. J. 2003. Loss of Tsc1/Tsc2 \nactivates mTOR and disrupts PI3K-Akt signaling through downregulation of PDGFR. J \nClin Invest, 112, 1223-33. \nZHAO, J., DANG, X., ZHANG, P., NGUYEN, L. N., CAO, D., WANG, L., WU, X., MORRISON, Z. \nD., ZHANG, Y., JIA, Z., XIE, Q., WANG, L., NING, S., EL GAZZAR, M., MOORMAN, J. \nP. & YAO, Z. Q. 2018. Insufficiency of DNA repair enzyme ATM promotes naive CD4 T-\ncell loss in chronic hepatitis C virus infection. Cell Discov, 4, 16. \n \n \n  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 28 \nACKNOWLEDGEMENTS \nGeneral: Investigations into the  in vivo effects of rapamycin on immune cell senescence were \nperformed on samples from the “Impacts of Mechanistic Target of Rapamycin (mTOR) Inhibition \non Aged Human Muscle (Rapamune)”  trial, registered on ClinicalTrials.gov, ID NCT05414292: \nhttps://clinicaltrials.gov/study/NCT05414292.  \n \nFunding: This work was supported by the following grants: the Mellon Longevity Graduate \nProgramme at Oriel College, University of Oxford (LK, LSC); UK SPINE (Research England) proof-\nof-concept grant (LSC, PJA, DJW, AKS); Wellcome Trust (AKS); Helmholtz Society (AKS); Versus \nArthritis grant 22617 (GA); BBSRC (the Biotechnology and Biological Sciences Research Council) \ngrant BB/W01825X/1 (LSC); MRC (Medical Research Council) (LSC); MRC grant MR/P021220/1 \nas part of the MRC-Versus Arthritis Centre for Musculoskeletal Ageing Research awarded to the \nUniversities of Nottingham and Birmingham (PJA, DJW) ; Public Health England (now UK Health \nSecurity Agency) (LSC); Diabetes UK/BIRAX (LSC). \n \nAuthor contributions: Author contributions were as follows:  conceptualization (LK, LSC, AKS, \nGA, PJA , KS ); m ethodology (LK, EJJ, D JW, NG, GA ); i nvestigation (LK, EJJ, D JW, NG ); \nvisualization (LK); funding acquisition (LSC, AKS, GA, PJA ); project administration (LSC, AKS, \nGA, PJA, EJJ, LK, DJW, KS); supervision (GA, LSC, AKS); writing – original draft (LK); writing – \nreview & editing (LK, LSC, GA, AKS). \n \nCompeting interests: AKS consults for Oxford Healthspan, The Longevity Labs, and Calico. LSC \nis Program Director (Dynamic Resilience) for non -profit Wellcome Leap and is co -director of UK \nAgeing Research Networks and BLAST ageing network (UKRI funded). LSC holds voluntary roles \nwith the All Party Parliamentary Group for Longevity (UK); Europea n Geriatric Medicine Society \nspecial interest group in Ageing Biology; Clinical and Translational Theme panel, Biochemical \nSociety (UK); and Medical Research Council A geing Research Steering Group. GA , EJJ, NG, \nDJW, PJA, KS, and LK report no conflicts of interest. \n \nData and materials availability : All data are available in the main text or the supplementary  \nmaterials. \n \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 29 \nList of Supplemental Materials \n• Figure S1: Gating strategy for CD4+ and CD8+ T cells using conventional flow cytometry after \nin vitro T-cell-specific activation of PBMCs \n• Figure S2: Effects of mTOR inhibitors on human T cell activation over 3 days \n• Figure S3: Flow cytometry-based measurement of autophagic flux  \n• Figure S4: Gating strategy for PBMCs using 27-colour spectral flow cytometry \n• Figure S5: In vivo rapamycin treatment in older humans \n• Table S1: Details of drugs used in cell culture experiments \n• Table S2: Details of antibodies used in flow cytometry \n  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 30 \n \nFigure 1 T cell DNA damage is associated with elevated mTORC1 activity \n(a) DNA damage assay design. (b) Representative histograms of γH2AX levels by flow cytometry in \nuntreated, zeocin-treated (200 µg/ml), and H2O2-treated (25 µM) PBMCs gated on CD4+ T cells. (c-\nd) Proportion of γH2AX+ of CD4+ (left) and CD8+ (right) T cells after (c) 4 hours recovery from zeocin \ntreatment, n=6 healthy donors, or (d) across different recovery times after zeocin treatment, from 4 \nindependent experiments using PBMCs from 1 donor. (e) Heatmaps for levels of DDR signalling \nmolecules expressed as fold change from untreated (UT) cells. (f) Representative gating of γH2AX+ \nand γH2AX- cells based on untreated (grey), zeocin-treated (red) cells, and fluorescence minus one \n(FMO, dotted grey) control. (g-h) gMFI of (g) p -S6 and (h) p-Akt in zeocin-treated CD4+ or CD8+ T \ncells gated as either positive or negative for γH2AX, n=6 healthy donors. P-values are derived from \na two-way ANOVA with Šídák's multiple comparisons test (d), and a Wilcoxon matched-pairs signed \nrank test (c, g, h). \n \n \n  \nStaining \nanalysis\nCD4+ T cells CD8+ T cells\nCD4+ CD8+\nγH2AX+γH2AX-\nUntreated\nZeocin\nγH2AX\nCD4+ T cells CD8+ T cells\nc\nf g h\nFigure 1 T cell DNA damage is associated with elevated mTORC1 activity\n(a) DNA damage assay design. (b) Representative histograms of γH2AX levels by flow cytometry in untreated, zeocin-treated (200 µg/ml), \nand H2O2-treated (25 µM) PBMCs gated on CD4+ T cells. (c-d) Proportion of γH2AX+ of CD4+ (left) and CD8+ (right) T cells after (c) 4 hours \nrecovery from zeocin treatment, n=6 healthy donors, or (d) across different recovery times after zeocin treatment, from 4 independent \nexperiments using PBMCs from 1 donor. (e) Heatmaps for levels of DDR signalling molecules expressed as fold change from untreated (UT) \ncells. (f) Representative gating of γH2AX+ and γH2AX- cells based on untreated (grey), zeocin-treated (red) cells, and fluorescence minus \none (FMO, dotted grey) control. (g-h) gMFI of (g) p-S6 and (h) p-Akt in zeocin-treated CD4+ or CD8+ T cells gated as either positive or \nnegative for γH2AX, n=6 healthy donors. P-values are derived from a two-way ANOVA with Šídák's multiple comparisons test (d), and a \nWilcoxon matched-pairs signed rank test (c, g, h).\nCD4+ CD8+\nUT ZEO\nγH2AX\np-Chk1\np-Chk2\np21\np53\n1.0\n1.5\n2.0\n2.5\n3.0\nUT ZEO\nγH2AX\np-Chk1\np-Chk2\np21\np53\n1.0\n1.5\n2.0\n2.5\n3.0\nCD4+ T cells CD8+ T cells\nNormalised count\nb\nd e\n0 2 4 24\n0\n5\n10\n15\n20\n25\nTime after zeocin (hours)\n%γH2AX+ (of CD4+ T cells)\nUT\nZEO\np = 0.1128\np = 0.1962\np = 0.0119\np = 0.0734\n0 2 4 24\n0\n5\n10\n15\n20\nTime after zeocin (hours)\n%γH2AX+ (of CD8+ T cells)\nUT\nZEO\np = 0.0421\np = 0.0017\np = 0.0003\np = 0.0285\nUntreated\nZeocin\nH2O2\nγH2AX-PerCP-Cy5.5\nγH2AX+\nγH2AX-PerCP-Cy5.5\na\nUT ZEO\n0\n20\n40\n60%γH2AX+ (of CD4+ T cells)\np = 0.0312\nUT ZEO\n0\n20\n40\n60%γH2AX+ (of CD8+ T cells)\np = 0.0312\n- +\n0\n2000\n4000\n6000\n8000p-S6 gMFI within subset\np = 0.0312\nγH2AX\n- +\n0\n2000\n4000\n6000\n8000\n10000p-S6 gMFI within subset\np = 0.0312\nγH2AX\n- +\n0\n1000\n2000\n3000\n4000p-Akt gMFI within subset\np = 0.3125\nγH2AX\n- +\n0\n1000\n2000\n3000p-Akt gMFI within subset\np = 0.8438\nγH2AX\nActivate\nα-CD3/28\nActivate\n3 d\nRecovery\nUT\nZEO\n2 h\nUT vs ZEO\nRecover\nRecover\n0, 2, 4, 24 h\nStaining \nand \nanalysis\n15 \nmins \nH2O2\nRecover\nFold \nchange \nfrom UT\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 31 \n \n \nFigure 2 mTOR inhibition reduces the DNA damage response in genotoxin -exposed \nhuman T cells from healthy donors \n(a) Experimental design. (b) Representative flow cytometry fluorescence histograms of γH2AX levels \nin CD4+ T cells with fluorescence minus one (FMO) control (left), with quantification of the proportion \nof γH2AX+ cells across conditions in CD4 + and CD8 + T cells, n=3 healthy donors. (c -d) \nRepresentative fluorescence histograms of p -Chk1 (c) or p -Chk2 (d) levels in CD4 + T cells, with \nquantification of fluorescence relative to average of DMSO untreated control in CD4 + and CD8+ T \ncells, n=3 healthy donors. (e) Experimental design for (f). (f) Heatmaps of gMFI of p21 and p53 \nassessed by flow cytometry, in CD4 + and CD8+ T cells at 4 and 24  hours recovery from zeocin, \nexpressed as fold change from the DMSO untreated (UT) condition for each recovery time point. P-\nvalues represent comparisons to the DMSO zeocin (ZEO) condition, n=4 healthy donors. P -values \nare determined from a two-way ANOVA with Šídák’s (b-d) or Dunnett’s (f) multiple comparisons test. \n \n \n  \nActivate\nα-CD3/28\nActivate\n3 d\nRecovery\nUT\nZEO\n2 h\nUT vs ZEO\nRecover\nRecover\n4 h\nStaining \nand \nanalysis\nDMSO\nRAPA 10 nM\nAZD8055 100 nM\na b\nc\nd f\np-Chk1\np-Chk2\nFigure 2 mTOR inhibition reduces the DNA damage response in genotoxin-exposed human T cells \nfrom healthy donors \n(a) Experimental design. (b) Representative flow cytometry fluorescence histograms of γH2AX levels in CD4+ T cells with fluorescence minus \none (FMO) control (left), with quantification of the proportion of γH2AX+ cells across conditions in CD4+ and CD8+ T cells, n=3 healthy donors. \n(c-d) Representative fluorescence histograms of p-Chk1 (c) or p-Chk2 (d) levels in CD4+ T cells, with quantification of fluorescence relative to \naverage of DMSO untreated control in CD4+ and CD8+ T cells, n=3 healthy donors. (e) Experimental design for (f). (f) Heatmaps of gMFI of p21 \nand p53 assessed by flow cytometry, in CD4+ and CD8+ T cells at 4 and 24  hours recovery from zeocin, expressed as fold change from the \nDMSO untreated (UT) condition for each recovery time point. P-values represent comparisons with the DMSO zeocin (ZEO) condition, n=4 \nhealthy donors. P-values in graphs are derived from a two-way ANOVA with Šídák’s (b-d) or Dunnett’s (f) multiple comparisons test.\nDMSO RAPA DMSO RAPA\np21\np53\n0\nUT ZEO\n0.5\n1.0\n1.5\nFold change \nfrom DMSO UT\n✱✱✱✱\n✱\nCD4+\nCD8+\n4 hours 24 hours\nDMSO RAPA DMSO RAPA\np21\np53\n0\nUT ZEO\n0.5\nFold change \nfrom DMSO UT\n1.0\n1.5\np=\n0.0001\np<\n0.0001\nDMSO RAPA DMSO RAPA\np21\np53\n0\nUT ZEO\n0.5\n1.0\n1.5\nFold change \nfrom DMSO UT\np<\n0.0001\np=\n0.0367DMSO RAPA DMSO RAPA\np21\np53\n1.0\nUT ZEO\n1.5\nFold change \nfrom DMSO UT\np=\n0.0084\np<\n0.0001\nDMSO RAPA DMSO RAPA\np21\np53\n1.0\nUT ZEO\nFold change \nfrom DMSO UT\n1.5\np=\n0.0004\np=\n0.0027\nγH2AX-PerCP-Cy5.5\nγH2AX+\nFMO\nZEO\nDMSO\nRAPA\nAZD8055\n+\n-\n+\n-\n+\n-\n+\np-Chk1-AF488 p-Chk2-AF488\nActivate\nα-CD3/28\nActivate\n3 d\nRecovery\nUT\nZEO\n2 h\nUT vs ZEO\nRecover\nRecover\n0 – 24 h\nAnalysis\nDMSO\nRAPA 10 nM\nDMSO RAPA AZD8055\n0\n5\n10\n15\n20\n25\n30\n35\n40\n45%γH2AX+ (of CD4+ T cells) \nUT\nZEO\n0.0007\n0.0011\n0.0017\nDMSO RAPA AZD8055\n0\n5\n10\n15\n20\n25\n30\n35\n40\n45%γH2AX+ (of CD8+ T cells) \nUT\nZEO\n0.0031\n0.0189\n0.1628\nγH2AX-PerCP-Cy5.5\nγH2AX+\nFMO\nZEO\nDMSO\nRAPA\nAZD8055\n+\n-\n+\n-\n+\n-\n+\np-Chk1-AF488 p-Chk2-AF488\nCD4+ T cells CD8+ T cells\nγH2AX\nγH2AX-PerCP-Cy5.5\nγH2AX+\nFMO\nZEO\nDMSO\nRAPA\nAZD8055\n+\n-\n+\n-\n+\n-\n+\np-Chk1-AF488 p-Chk2-AF488\nDMSO RAPA AZD8055\n0.0\n0.5\n1.0\n1.5\n2.0\n2.5Relative p-Chk1 MFI\n0.0045\n0.0044\n0.0060\nDMSO RAPA AZD8055\n0.0\n0.5\n1.0\n1.5\n2.0\n2.5Relative p-Chk1 MFI\n0.0067\n0.01370.0058\nUT\nZEO\nCD4+ T cells CD8+ T cells\nγH2AX-PerCP-Cy5.5\nγH2AX+\nFMO\nZEO\nDMSO\nRAPA\nAZD8055\n+\n-\n+\n-\n+\n-\n+\np-Chk1-AF488 p-Chk2-AF488γH2AX-PerCP-Cy5.5\nγH2AX+\nFMO\nZEO\nDMSO\nRAPA\nAZD8055\n+\n-\n+\n-\n+\n-\n+\np-Chk1-AF488 p-Chk2-AF488\nDMSO RAPA AZD8055\n0.0\n0.5\n1.0\n1.5\n2.0Relative p-Chk2 MFI\nUT\nZEO\n0.0081\n0.02730.1572\nDMSO RAPA AZD8055\n0.0\n0.5\n1.0\n1.5\n2.0Relative p-Chk2 MFI\nUT\nZEO\n0.0077\n0.02900.0453\ne\nCD4+ T cells CD8+ T cells\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 32 \n \nFigure 3 Rapamycin inhibits γH2AX at any point with respect to genotoxic treatment and \nsuppression of mTOR activity correlates with DNA damage \n(a) Experimental design. (b-c) Proportion of γH2AX+ of (b) CD4+ and (c) CD8+ T cells, in PBMCs \ntreated with 10 nM RAPA or DMSO vehicle control pre-, co-, and/or post-treatment with zeocin. Data \nare expressed as fold change from DMSO →DMSO→DMSO (DDD) untreated controls for each \ndonor. P-values represent comparisons to the DDD ZEO condition, n=6 healthy donors. (d-e) Simple \nlinear regression analyses from experiment in (a), of γH2AX and either p-S6 (d) or p-Akt (e) in zeocin-\ntreated CD4+ or CD8+ T cells. Values are expressed as fold change from DMSO→DMSO→DMSO \nUT with line of best fit and 95% confidence intervals indicated. Data are pooled from 6 independent \nexperiments, n=6 healthy donors. P -values are derived from a two -way ANOVA with Šídák’s (b-c) \nmultiple comparisons test. \n \n \n  \nCD4+ T cells\nCD8+ T cells\na b\ne\nd\nCD4+ T cells CD8+ T cells\nFigure 3 Rapamycin inhibits γH2AX at any point with respect to genotoxic treatment and does not \nimpact on cell cycle progression\n(a) Experimental design. (b-c) Proportion of γH2AX+ of (b) CD4+ and (c) CD8+ T cells, in PBMCs treated \nwith 10 nM RAPA or DMSO vehicle control pre-, co-, and/or post-treatment with zeocin. Data are expressed \nas fold change from DMSO→DMSO→DMSO (DDD) untreated controls for each donor. P-values represent \ncomparisons to the DDD ZEO condition, n=6 healthy donors. (d-e) Simple linear regression analyses from \nexperiment in (a), of γH2AX and either p-S6 (d) or p-Akt (e) in zeocin-treated CD4+ or CD8+ T cells. Values \nare expressed as fold change from DMSO→DMSO→DMSO UT with line of best fit and 95% confidence \nintervals indicated. Data are pooled from 6 independent experiments, n=6 healthy donors. P-values are \nderived from a two-way ANOVA with Šídák’s (b-c) multiple comparisons test.\n \nDMSO\nActivate\nα-CD3/28\nDMSO\nPre\nRecovery\nUT\nZEO\nCo\nUT vs ZEO\nDMSO\nAnalysis\nRAPA\nRAPA\nDMSO\nRAPA\nDMSO\nRAPA\nDMSO\nRAPA\nRAPA UT\nZEO\nDMSO\nRAPA\nUT\nZEO\nDMSO\nRAPA\nDMSO\nRAPA\nDMSO\nRAPA\nDMSO\nRAPA\nUT\nZEO\nDDD UT\nDDR UT\nDDD ZEO\nDDR ZEO\nDRD UT\nDRR UT\nDRD ZEO\nDRR ZEO\nRDD UT\nRDR UT\nRDD ZEO\nRDR ZEO\nRRD UT\nRRR UT\nRRD ZEO\nRRR ZEO\n3 d 2 h 4 h\nPost\nDMSO-DMSO-DMSODMSO-DMSO-RAPADMSO-RAPA-DMSODMSO-RAPA-RAPARAPA-DMSO-DMSORAPA-DMSO-RAPARAPA-RAPA-DMSORAPA-RAPA-RAPA\n0\n1\n2\n3\n4\n5\n%γH2AX+ (of CD4+ T cells) \n(fold DMSO UT)\nUT\nZEO\nPre\nCo\nDMSO\nRAPA\nDMSO\nDMSO\nRAPA\nDMSO\nRAPA\nRAPA\nPost RAPADMSO DMSO RAPA\nDMSO\nDMSO\nRAPA\nDMSO\nRAPA\nDMSO\nRAPA\nDMSO\nRAPA\nRAPA\nRAPA\nDMSO\np<0.0001 p<0.0001 p<0.0001p<0.0001p<0.0001 p<0.0001p<0.0001\nPre\nCo\nDMSO\nRAPA\nDMSO\nDMSO\nRAPA\nDMSO\nRAPA\nRAPA\nPost RAPADMSO DMSO RAPA\nDMSO\nDMSO\nRAPA\nDMSO\nRAPA\nDMSO\nRAPA\nDMSO\nRAPA\nRAPA\nRAPA\nDMSO\nDMSO-DMSO-DMSODMSO-DMSO-RAPADMSO-RAPA-DMSODMSO-RAPA-RAPARAPA-DMSO-DMSORAPA-DMSO-RAPARAPA-RAPA-DMSORAPA-RAPA-RAPA\n0\n2\n4\n6\n8\n%γH2AX+ (of CD8+ T cells) \n(fold DMSO UT)\nUT\nZEOp = 0.0332 p = 0.0122p = 0.0051\np = 0.9964p = 0.9794p = 0.9545p = 0.5569\nPre\nCo\nDMSO\nRAPA\nDMSO\nDMSO\nRAPA\nDMSO\nRAPA\nRAPA\nPost RAPADMSO DMSO RAPA\nDMSO\nDMSO\nRAPA\nDMSO\nRAPA\nDMSO\nRAPA\nDMSO\nRAPA\nRAPA\nRAPA\nDMSO\nPre-treatment:\nRAPA\nDMSO\nPre-treatment:\nRAPA\nDMSO\np-S6 (mTORC1) vs γH2AX \np-Akt (mTORC2) vs γH2AX \nCD4+ T cells CD8+ T cells\nc\n0.0 0.5 1.0 1.5 2.0\n0\n2\n4\n6\n8\np-S6 gMFI \n(fold change from DMSO UT)\n%γH2AX+ (of CD4+ T cells) \n(fold change from DMSO UT)\np=0.0004\nR2=0.2434\n0.5 1.0 1.5\n0\n2\n4\n6\n8\np-Akt gMFI \n(fold change from DMSO UT)\n%γH2AX+ (of CD4+ T cells) \n(fold change from DMSO UT)\np=0.0003\nR2=0.2523\n0.0 0.5 1.0 1.5 2.0\n0\n2\n4\n6\n8\np-S6 gMFI \n(fold change from DMSO UT)\n%γH2AX+ (of CD8+ T cells) \n(fold change from DMSO UT)\np=0.0006\nR2=0.2299\n0.5 1.0 1.5\n0\n2\n4\n6\n8\np-Akt gMFI \n(fold change from DMSO UT)\n%γH2AX+ (of CD8+ T cells) \n(fold change from DMSO UT)\np=0.0312\nR2=0.09695\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 33 \n \nFigure 4 Mitigation of DNA damage markers by rapamycin is not due to modulation of the \ncell cycle or protein synthesis \n(a-b) Representative histograms for cell cycle phases in untreated (UT) and zeocin-treated conditions \nwith quantification IN (b), n=3 healthy donors. (c) Cell cycle phases in zeocin-treated T cells treated \nwith rapamycin continuously (“RRR”), during (“DRD”), and/or after (“DRR”, “DDR”) zeocin exposure, \ngated as positive or negative for γH2AX, n=3 healthy donors. Unless otherwise indicated, statistical \ncomparisons between different treatment conditions within each cell cycle phase (G 0/G1, S, G 2/M) \nwere not significant, as determined by a two-way ANOVA with Dunnett’s multiple comparisons test. \n(d) Experimental design for measuring nascent protein synthesis with 20 µM O-propargyl-puromycin \nincorporation over 30 minutes. 1 -hour treatment with 50 µg/ml cycl oheximide (CHX) was used a \npositive control for inhibition of protein synthesis. (e) OPP mean fluorescence intensity (MFI) of CD4+ \nT cells, relative the DMSO untreated (UT) control, across conditions undergoing treatment with 10 \nnM rapamycin continuously (”RRR”) or during and after zeocin exposure (“DRR”). P-values are \nderived from a one-sample t-test (theoretical mean = 1), n=3 donors. \n \n \n  \nCD4+ T cells CD8+ T cells\n0\n50\n100% cell cycle phase\nUT ZEO\nG0/G1 phase\nS phase\nG2/M phase\n0\n50\n100% cell cycle phase\nG0/G1 phase\nS phase\nG2/M phase\nUT ZEO\nCD4+ T cells CD8+ T cells\nγH2AX+c\nUT ZEO\nEdU-AF488\nS phase\nG0/G1\nphase\nG2/M \nphase\nS phase\nG0/G1\nphase\nG2/M \nphase\nFxCycleViolet\nEdU-AF488\nDNA-FxCycle Violet\nCD4+ T cells CD8+ T cells\nG0/G1 phase S phase G2/M phase\n0\n20\n40\n60\n80\n100\n% cell cycle phase\n(of γH2AX- CD8+ T cells)\nDDD\nDDR\nDRD\nDRR\nRRR\nns\nns\nns\n✱✱\nns\nns\nns\nns\nns\nns\nns\nns\nγH2AX-\nG0/G1 phase S phase G2/M phase\n0\n20\n40\n60\n80\n100\n% cell cycle phase\n(of γH2AX+ CD4+ T cells)\nDDD\nDDR\nDRD\nDRR\nRRR\nG0/G1 phase S phase G2/M phase\n0\n20\n40\n60\n80\n100\n% cell cycle phase\n(of γH2AX- CD4+ T cells)\nDDD\nDDR\nDRD\nDRR\nRRR\n✱\nG0/G1 phase S phase G2/M phase\n0\n20\n40\n60\n80\n100\n% cell cycle phase\n(of γH2AX+ CD8+ T cells)\nDDD\nDDR\nDRD\nDRR\nRRR\nG0/G1 phase S phase G2/M phase\n0\n20\n40\n60\n80\n100\n% cell cycle phase\n(of γH2AX- CD8+ T cells)\n✱✱\n0.0\n0.5\n1.0\n1.5\nOPP MFI \nrelative to DMSO UT\n✱\nns ns\nns\nDDD DDDDRR RRRCHX\nUT ZEO\nActivate\nα-CD3/28\nPre-treatment\n3 d\nRecovery\nCo-\ntreatment\n2 h\nUT vs ZEO\nPost-\ntreatment\n4 h\nStaining \nand \nanalysis\n+ 20 µM OPP\n(30 mins)\nDMSO\nDMSO\nUT\nZEO\nDMSO\nRAPA\nDMSO\nRAPARAPA ZEO\nRAPA RAPAZEO\nCHX\nDDD UT\nDDD ZEO\nDRR ZEO\nRRR ZEO\nCHX\na b\nd e\nFigure 4 Mitigation of DNA damage markers by rapamycin is not due to modulation of the cell \ncycle or protein synthesis\n(a-b) Representa.ve histograms for cell cycle phases in untreated and zeocin-treated condi.ons with \nquan.fica.on  (b), n=3 healthy donors. (c) Cell cycle phases in zeocin-treated T cells treated with rapamycin \ncon.nuously (“RRR”), during (“DRD”), and/or aIer  (“DRR”, “DDR”) zeocin exposure, gated as posi.ve  or nega.ve  \nfor γH2AX, n=3 healthy donors. Unless otherwise indicated, sta.s.cal  comparisons between different treatment \ncondi.ons within each cell cycle phase (G0/G1, S, G2/M) are not significant, as determined by a two-way ANOVA \nwith DunneX’s mul.ple  comparisons test. (d) Experimental design for measuring nascent protein synthesis with 20 \nµM O-propargyl-puromycin incorpora.on over 30 minutes. 1-hour treatment with 50 µg/ml cycloheximide (CHX) \nwas used a posi.ve  control for inhibi.on  of protein synthesis. (e) OPP mean fluorescence intensity (MFI) of CD4+ T \ncells across condi.ons, rela.ve  the DMSO untreated (UT) control, across condi.ons undergoing treatment with 10 \nnM rapamycin con.nuously (”RRR”) or during and aIer  zeocin exposure (“DRR”). P-values are derived from a one-\nsample t-test (theore.cal mean = 1), n=3 donors.\n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 34 \n \nFigure 5 Autophagy is required for the resolution of DNA damage, but is not required \nfor the mitigation of DDR upregulation by rapamycin \n(a) Experimental design for combined DNA damage and autophagic flux assay. (b) Quantification of \nautophagic flux in γH2AX+ and γH2AX- populations in flow cytometry-gated CD4+ and CD8+ T cells, \nn=5 healthy donors. (c) Representative γH2AX histograms (left) and quantified percentage (right) of \nγH2AX+ CD4+ and CD8+ T cells after zeocin treatment following 3 -day T cell -specific activation in \nchloroquine (CQ, 10 µM) or DMSO control, n=4 healthy donors. (d) Experimental design for \nmeasuring autophagic flux in CD4+ and CD8+ T cells across different rapamycin treatment conditions \nwith respect to zeocin exposure. (e) Autophagic flux in zeocin -exposed cells gated as positive or \nnegative for γH2AX, across rapamycin treatment conditions. (f) Experimental design for treatment of \nzeocin-exposed activated T cells with chloroquine (CQ), rapamycin  (RAPA), or dual treatment \n(CQ+RAPA). (g) Percentage of γH2AX+ CD4+ T cells after zeocin treatment across conditions in (f), \nn=3 healthy donors. P -values are derived from a paired t -test (b), a two -way ANOVA with \nŠídák's (d,e) or Tukey’s multiple comparisons test (g). \nCD4+ T cells CD8+ T cells\nc\nActivate\nα-CD3/28\nActivate\n3 d\nRecovery\nUT\nZEO\n2 h\nUT vs ZEO\nRecover\nRecover\n4 h\nStaining \nand \nanalysis\nAutophagic flux \nmeasurement\na\nFigure 5 Autophagy is required for the resolution of DNA damage, but does not explain the \nmitigation of DDR upregulation by rapamycin\n(a) Experimental design for combined DNA damage and autophagic flux assay. (b) Quantification of autophagic flux in \nγH2AX+ and γH2AX- populations in flow cytometry-gated CD4+ and CD8+ T cells, n=5 healthy donors. (c) Representative \nγH2AX histograms (left) and quantified percentage (right) of γH2AX+ CD4+ and CD8+ T cells after zeocin treatment \nfollowing 3-day T cell-specific activation in chloroquine (CQ, 10 µM) or DMSO control, n=4 healthy donors. (d) Experimental \ndesign for measuring autophagic flux in CD4+ and CD8+ T cells across different rapamycin treatment conditions with respect \nto zeocin exposure. (e) Autophagic flux in zeocin-exposed cells gated as positive or negative for γH2AX, across rapamycin \ntreatment conditions. (f) Experimental design for treatment of zeocin-exposed activated T cells with chloroquine (CQ), \nrapamycin (RAPA), or dual treatment (CQ+RAPA). (g) Percentage of γH2AX+ CD4+ T cells after zeocin treatment across \nconditions in (f), n=3 healthy donors. P-values are derived from a paired t-test (b), a two-way ANOVA with Šídák's (d,e) or \nTukey’s multiple comparisons test (g).\nb\nUT ZEO\n0\n20\n40\n60%γH2AX+ (of CD4+ T cells) \nDMSO\nCQ\n0.3522 0.0205\nγH2AX-PerCP-Cy5.5\nDMSO\nCQ\nDMSO\nCQ\nUT\nZEO\nUT ZEO\n0\n20\n40\n60%γH2AX+ (of CD8+ T cells) \nDMSO\nCQ\n0.1004 0.0255\nCD4+ T cells CD8+ T cells\n- +\n0.0\n0.2\n0.4\n0.6Autophagic flux\np = 0.0036\nγH2AX\n- +\n0.0\n0.2\n0.4\n0.6Autophagic flux\np = 0.0025\nγH2AX\nDMSO\nActivate\nα-CD3/28\nDMSO\nPre-treatment\n3 d\nRecovery\nUT\nZEO\nCo-\ntreatment\n2 h\nUT vs ZEO\nDMSO\nPost-\ntreatment\n4 h\nStaining \nand \nanalysis\nRAPA\nDMSO\nRAPA\nRAPA\nRAPA UT\nZEO\nDMSO\nRAPA\nUT\nZEO\nDMSO\nDMSO\nRAPA\nRAPA\nUT\nZEO\nAutophagic flux \nmeasurement\nd\nCD4+ T cells\nPre → Co →  Post\nγH2AX- γH2AX+\n0.0\n0.2\n0.4\n0.6Autophagic flux\nDMSO→DMSO→DMSO\nDMSO→RAPA→RAPA\nRAPA→DMSO→DMSO\nRAPA→RAPA→RAPA\np = 0.0179\np = 0.0002\np = 0.0002\np = 0.1426\np < 0.0001\np = 0.0001\nγH2AX- γH2AX+\n0.0\n0.2\n0.4\n0.6Autophagic flux\nDMSO→DMSO→DMSO\nDMSO→RAPA→RAPA\nRAPA→DMSO→DMSO\nRAPA→RAPA→RAPA\np = 0.5189\np = 0.2950\np = 0.3329\np = 0.1584\np = 0.0100\np = 0.0794\ne CD8+ T cells\nγH2AX- γH2AX+\n0.0\n0.2\n0.4\n0.6Autophagic flux\nDMSO→DMSO→DMSO\nDMSO→RAPA→RAPA\nRAPA→DMSO→DMSO\nRAPA→RAPA→RAPA\np = 0.5189\np = 0.2950\np = 0.3329\np = 0.1584\np = 0.0100\np = 0.0794\nf g\nDMSO\nCQ\nRAPA\nCQ+RAPA\nDMSO\nCQ\nRAPA\nCQ+RAPA\n0\n20\n40\n60%γH2AX+ (of CD4+ T cells) \nUT\nZEO\n0.0001 0.0113\n0.0266\n0.0025\nActivate\nα-CD3/28\nActivate\n3 d\nRecovery\nUT\nZEO\n2 h\nUT vs ZEO\nRecover\nRecover\n4 h\nAnalysis\nDMSO\nCQ 10 µM\nRAPA 10 nM\nCQ+RAPA\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 35 \n \n \n  \n \nFigure 6 Rapamycin attenuates DNA lesional burden and improves survival after \nexposure to a DNA-damaging agent \n(a) Experimental design for zeocin treatment (200 µg/ml) following continuous exposure to rapamycin \n(10 nM) or DMSO vehicle control in isolated CD4 + T cells. (b) Representative images of comets, \nscalebar represents 200 µm. (c) DNA lesions as measured by the Olive moment of >250 comets \nanalysed across conditions following a 4 -hour recovery from zeocin. (d) Comet Olive moments \nthroughout a 24 -hour time course of recovery from zeocin normalised to the median value of the \nDMSO untreated (UT) condition at each time point. Data are the median ± SEM of 100 -250 nuclei \nanalysed per condition. Representative of three independent experiments, n=3 healthy donors. (e) \nLive cells across conditions, as measured by lack of fluorescence of a membrane-permeable dye, at \n4 and 24 hours recovery from zeocin exposure . P-values are derived from a one-way ANOVA with \nTukey’s multiple comparisons test (c), or a two-way ANOVA with Šídák's multiple comparisons test \n(e). \nFigure 6 Rapamycin attenuates DNA lesional burden and improves survival after exposure to a \nDNA-damaging agent\n(a) Experimental design for exposure to zeocin (200 µg/ml) or H2O2 (25 µM) following continuous \nexposure to rapamycin (10 nM) or DMSO vehicle control in isolated CD4+ T cells. (b) Representative \nimages of comets, scalebar represents 200 µm. (c) DNA lesions as measured by the Olive moment of \n>250 comets analysed across conditions following a 4-hour recovery from zeocin. (d) Comet Olive \nmoments throughout a 24-hour time course of recovery from zeocin normalised to the median value of \nthe DMSO untreated (UT) condition at each time point. Data are the median ± SEM of 100-250 nuclei \nanalysed per condition. Representative of three independent experiments, n=3 healthy donors. (e) \nLive cells across conditions, as measured by lack of fluorescence of a membrane-permeable dye. P-\nvalues are derived from a one-way ANOVA with Tukey’s multiple comparisons test (c), or a two-way \nANOVA with Šídák's multiple comparisons test (f).\nDMSO RAPA\nUT ZEO\nDMSO RAPA\na b\nc\nUT ZEO UT ZEO H2O2\n0\n20\n40\n60\n80Olive moment\n<0.0001\n<0.0001\n<0.0001\n0.6226\n0.0311\nDMSO RAPA\nd\nFSC-A\nLIVE DEAD Blue\n4 \nhours\n24 \nhours\nUT ZEO\nDMSO RAPA DMSO RAPA\n81.5 84.6 74.0 86.5Live Live Live Live\n77.4 84.7 21.7 61.9Live Live Live Live\ne\n0 5 10 15 20 25\n0\n2\n4\n6\n8\n10\n12\nTime after zeocin (hours)\nNormalised Olive Moment\nDMSO UT\nDMSO ZEO\nRAPA UT\nRAPA ZEO\nDMSO H2O2\nUT ZEO\n0\n20\n40\n60\n80\n100%live cells (of CD4+ T cells)\n4-hour recovery\nDMSO\nRAPA\n0.0649 0.0053\nUT ZEO\n0\n20\n40\n60\n80\n100%live cells (of CD4+ T cells)\nDMSO\nRAPA\n0.1274 0.0031\n24-hour recovery\nActivate\nα-CD3/28\nActivate\n3 d\nRecovery\nUT\nZEO\n2 h\nUT vs ZEO\nRecover\nRecover\n4 – 24 h\nStaining \nand \nanalysis\nDMSO\nRAPA 10 nM\nH2O2\n25 µM\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 36 \n \nFigure 7 Age-related peripheral immune subsets from healthy donors display elevated \nmarkers of cellular senescence and mTORC1/2 hyperactivity \n(a) Immune cell subsets in PBMCs from a healthy donor identified by flow cytometry. Gates \nhighlighted in bold indicate age-related immune subsets. (b) Geometric mean fluorescence intensity \n(gMFI) of biomarkers for senescence and mTORC1/2 activity as measured by spectral flow cytometry \nacross immune subsets in n=8 healthy donors. Data represent log2(fold change) from the mean of \nthe lefthand, most early -differentiated immune cell population for each cell type. FSC = forward \nscatter. (c) Table summarising significantly increased markers (black dots) in age -related immune \nsubsets compared to early -differentiated immune cell counterparts, as assessed with one -way \nANOVAs with Dunnett’s multiple comparisons test be tween age -related and early -differentiated \nsubsets for each cell type (n=8 healthy donors). (d) Proportion of CD8 + T cells positive for CD28, \nCD57, and KLRG1 in PBMCs from healthy younger donors (17-50 years old, n=8) and older donors \n(average age 62 years old, n=9). P -values are derived from unpaired t -tests. (e) Geometric mean \nfluorescence intensity (gMFI) of p-S6 in immune subsets in older donors (n=9) expressed as log2(fold \nchange) from the mean value of cells of control volunteers (n=8), assessed by flow cytometry. \n \n \n  \nCD57-BV421\na\nTEMRA\nCD45RA-BV605\nCD4+ T cells CD8+ T cells\nCD4+ T cells CD8+ T cells NK CD56bright NK CD56dimB cells\nIgD-BV480\nCD27-PE-Cy7\nB cells\nCD16-PE-Dazzle 594\nNK CD56bright NK CD56dim\nTNTCM\nTEM TEMRA\nTNTCM\nTEM TEMRA\nUnswitched \nmemory\nNaïve\nunswitched\nSwitched \nmemory\nDN\nb\nc d\nCD57+CD16+\nCD57+CD16-\nCD57-\nCD16+\nDN\nCD57+CD16+\nCD57+CD16-\nCD57-\nCD16+\nDN\nFigure 7 Age-related peripheral immune subsets from healthy donors display elevated markers of \ncellular senescence and mTORC1/2 hyperactivity\n(a) Immune cell subsets in PBMCs from a healthy donor identified by flow cytometry. Gates highlighted in bold indicate age-related \nimmune subsets. (b) Geometric mean fluorescence intensity (gMFI) of biomarkers for senescence and mTORC1/2 activity as \nmeasured by flow cytometry across immune subsets in n=8 healthy donors. Data are log2(fold change) from the mean of the \nlefthand, most early-differentiated immune cell population for each cell subset. FSC = forward scatter. (c) Table summarising \nsignificantly increased markers (black dots) in age-related immune subsets compared to early-differentiated immune cell \ncounterparts, as assessed with one-way ANOVAs with Dunnett’s multiple comparisons test between age-related and early-\ndifferentiated subsets for each cell type (n=8 healthy donors). (d) Proportion of CD8+ T cells positive for CD28, CD57, and KLRG1 \nin PBMCs from healthy younger donors (17-50 years old, n=8) and older donors (average age 62 years old, n=9). P-values are \nderived from unpaired t-tests. (e) Geometric mean fluorescence intensity (gMFI) of p-S6 in immune subsets in older (n=9) \nexpressed as log2(fold change) from the mean value of cells of control volunteers (n=8), assessed by flow cytometry.\ne\nLog2(FC) \nfrom naïve\nCD14-APC-Fire 810\nMonocytes\nNon-classical\nIntermediate\nClassical\nMonocytes\n●●●●●●\n●●●●●●\n●●●●●●\n●●●●●●\n●●●●●●\n●●●●●●\n●●●●●●\nDP NK CD56\nbright\nDN B cellCD8\n+ TEMRA\nCD4\n+ TEMRA\nDP NK CD56\ndim\np21\np53\np16\nγH2AX\nFSC\np-S6\np-Akt\nNC mono\nCD28+ CD57+ KLRG1+\nControl Older\n0\n20\n40\n60\n80\n100%CD57+ (of CD8+ T cells)\np = 0.0211\nControl Older\n0\n20\n40\n60\n80\n100%KLRG1+ (of CD8+ T cells)\np = 0.0376\nControl Older\n0\n20\n40\n60\n80\n100%CD28+ (of CD8+ T cells)\np = 0.0225\nTN TCMTEMTEMRA\np21\np53\np16\nγH2AX\nFSC\np-S6\np-Akt\n-1\n0\n1\nTN TCMTEMTEMRA\np21\np53\np16\nγH2AX\nFSC\np-S6\np-Akt\n-1\n0\n1\nNaïve unswitchedMemory unswitchedMemory switchedDouble negative\np21\np53\np16\nγH2AX\nFSC\np-S6\np-Akt\n-1\n0\n1\nCD57\n- CD16\n-\nCD57\n+ CD16\n-\nCD57\n- CD16\n+\nCD57\n+ CD16\n+\np21\np53\np16\nγH2AX\nFSC\np-S6\np-Akt\n-1\n0\n1\nCD57\n- CD16\n-\nCD57\n+ CD16\n-\nCD57\n- CD16\n+\nCD57\n+ CD16\n+\np21\np53\np16\nγH2AX\nFSC\np-S6\np-Akt\n-1\n0\n1\nClassicalIntermediateNon-classical\np21\np53\np16\nγH2AX\nFSC\np-S6\np-Akt\n-1\n0\n1\nControlOlder\nCD4 total\nCD8 total\nB total\nNK total\nCD4 TN\nCD4 TCM\nCD4 TEM\nCD4 TEMRA\nCD8 TN\nCD8 TCM\nCD8 TEM\nCD8 TEMRA\nB Naive unswitched\nB Memory unswitched\nB Memory switched\nB DN\nCD56bright CD16-CD57-\nCD56bright CD16+CD57-\nCD56bright CD16-CD57+\nCD56bright CD16+CD57+\nCD56dim CD16-CD57-\nCD56dim CD16+CD57-\nCD56dim CD16-CD57+\nCD56dim CD16+CD57+\n0\np-S6 log2(FC)\n0.5\n1.0\n1.5\n2.0\n2.5\nBold gate = age-related subset\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 37 \n \nFigure 8 Low-dose rapamycin in vivo attenuates biomarkers of immune cell senescence \nand exhaustion \n(a) Outline of the rapamycin experimental medicine trial. (b-c) Simple linear regression analyses of \nγH2AX and p-S6 geometric fluorescence intensity (gMFI) over all timepoints in CD4+ T cells from (b) \nplacebo and (c) rapamycin groups. Graphs show line of best-fit with 95% confidence intervals. (d-f) \nLog2(fold change) from baseline in gMFI of (d) p-S6, (e) p21, and (f) p53 across immune subsets. (g-\ni) Percentage of defined T cell subsets positive for (g) KLRG1, (h) NKG2A, and (i) LAG3  in \nparticipants. In (d-e), each value is expressed as log2(fold change) from baseline for each participant. \nP-values are derived from an unpaired t -test between placebo (n=5) and rapamycin (n=4) at each \ntime point. Statistically significant (*p<0.05) p-values are indicated. \n \n \n  \n2-3 4-5 8 16 2-3 4-5 8 16\nCD4 total\nCD8 total\nCD4 TN\nCD4 TCM\nCD4 TEM\nCD4 TEMRA\nCD8 TN\nCD8 TCM\nCD8 TEM\nCD8 TEMRA\n-1.0\n-0.5\n0\n*\n0.062\n0.073\n0.070\n*\nPlacebo Rapamycin\nLog2(FC) from \nbaseline\n0.5\n1.0\n2-3 4-5 8 16 2-3 4-5 8 16\nCD4 total\nCD8 total\nCD4 TN\nCD4 TCM\nCD4 TEM\nCD4 TEMRA\nCD8 TN\nCD8 TCM\nCD8 TEM\nCD8 TEMRA\n-1.0\n-0.5\n0\n*\n*\n*\nPlacebo Rapamycin\nLog2(FC) from \nbaseline\n0.5\n1.0\n2-3 4-5 8 16 2-3 4-5 8 16\nCD4 total\nCD8 total\nCD4 TN\nCD4 TCM\nCD4 TEM\nCD4 TEMRA\nCD8 TN\nCD8 TCM\nCD8 TEM\nCD8 TEMRA\n-1.0\n-0.5\n*\n*\n*\n0.051\n0.051\n0.067\n* *\n* *\n* * *\nPlacebo Rapamycin\nLog2(FC) from \nbaseline\n0\n0.5\n1.0\n2-3 4-5 8 16 2-3 4-5 8 16\nCD4 total\nCD8 total\nB total\nNK total\nNK CD56bright\nNK CD56dim\nHLA-DR+\nCD4 TN\nCD4 TCM\nCD4 TEM\nCD4 TEMRA\nCD8 TN\nCD8 TCM\nCD8 TEM\nCD8 TEMRA\nB Naïve unswitched\nB Memory unswitched\nB Memory switched\nB DN\nCD56dim CD16-CD57-\nCD56dim CD16+CD57-\nCD56dim CD16-CD57+\nCD56dim CD16+CD57+\n-1\n0\n1\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\nLog2(FC) from \nbaseline\nPlacebo Rapamycin\n2-3 4-5 8 16 2-3 4-5 8 16\nCD4 total\nCD8 total\nB total\nNK total\nNK CD56bright\nNK CD56dim\nHLA-DR+\nCD4 TN\nCD4 TCM\nCD4 TEM\nCD4 TEMRA\nCD8 TN\nCD8 TCM\nCD8 TEM\nCD8 TEMRA\nB Naïve unswitched\nB Memory unswitched\nB Memory switched\nB DN\nCD56dim CD16-CD57-\nCD56dim CD16+CD57-\nCD56dim CD16-CD57+\nCD56dim CD16+CD57+\n-1.0\n-0.5\n0\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\nLog2(FC) from \nbaseline\nPlacebo Rapamycin\n0.5\n1.0\n2-3 4-5 8 16 2-3 4-5 8 16\nCD4 total\nCD8 total\nB total\nNK total\nNK CD56bright\nNK CD56dim\nHLA-DR+\nCD4 TN\nCD4 TCM\nCD4 TEM\nCD4 TEMRA\nCD8 TN\nCD8 TCM\nCD8 TEM\nCD8 TEMRA\nB Naïve unswitched\nB Memory unswitched\nB Memory switched\nB DN\nCD56dim CD16-CD57-\nCD56dim CD16+CD57-\nCD56dim CD16-CD57+\nCD56dim CD16+CD57+\n-1.0\n-0.5\n*\n*\n*\n*\n*\n*\n*\n0.057\n*\n*\n*\n*\n*\n*\n*\n*\nLog2(FC) from \nbaseline\nPlacebo Rapamycin\n0\n0.5\n1.0\na\nFigure 8 Low-dose rapamycin in vivo attenuates biomarkers of immune cell senescence \nand exhaustion\n(a) Outline of the Rapamune trial. (b-c) Simple linear regression analyses of γH2AX and p-S6 in \nover pooled timepoints in CD4+ T cells from (b) placebo and (c) rapamycin groups. Graphs show \nline of best-fit and 95% confidence intervals. (d-f) Geometric mean fluorescence intensity of (d) \np-S6, (e) p21, and (f) p53 across immune subsets. (g-i) Percentage of defined T cell subsets \npositive for (g) KLRG1, (h) NKG2A, and (i) LAG3 across peripheral immune subsets in \nparticipants in rapamycin (n=4) and placebo (n=5) groups across four time points, as assessed \nby spectral flow cytometry. Each value is expressed as log2(fold change) from baseline for each \nparticipant. P-values are derived from an unpaired t-test between placebo and rapamycin at \neach time point. Statistically significant (*p<0.05) p-values are indicated.\np-S6\nWeek:\nd p21 p53\nKLRG1 LAG3\nWeek: Week:\nNKG2A\nWeek:\nh i\nfe\ng\nn=4\nn=5 Placebo\nRapamycin 1 mg/day\n0 2-3 4-5 8 16Week:\n16 weeks total\nAnalysis of PBMCs\nby spectral flow \ncytometry\n2-3 4-5 8 16 2-3 4-5 8 16\nCD4 total\nCD8 total\nB total\nNK total\nNK CD56bright\nNK CD56dim\nHLA-DR+\nCD4 TN\nCD4 TCM\nCD4 TEM\nCD4 TEMRA\nCD8 TN\nCD8 TCM\nCD8 TEM\nCD8 TEMRA\nB Naïve unswitched\nB Memory unswitched\nB Memory switched\nB DN\nCD56dim CD16-CD57-\nCD56dim CD16+CD57-\nCD56dim CD16-CD57+\nCD56dim CD16+CD57+\n-1\n0\n1\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\n*\nLog2(FC) from \nbaseline\nPlacebo Rapamycin\nb c\n0 1000 2000 3000 4000 5000\n0\n500\n1000\n1500\n2000\n2500\np-S6 gMFI\nγH2AX gMFI\nPlacebo\nR2 = 0.5481\np <0.0001\n0 1000 2000 3000 4000 5000\n0\n500\n1000\n1500\n2000\n2500\nRapamycin\np-S6 gMFI\nγH2AX gMFI\nR2 = 0.7758\np <0.0001\n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 38 \nTable 1 Identification of age-related peripheral immune subsets in humans \nPeripheral \nimmune \ncell \nAge-related \nsubset \nEvidence that subset \naccumulates with chronological \nage in humans \nFlow cytometry gating \nstrategy in human blood \n \nCD4+ T \ncell \n \n \nTEMRA, CD27-\nCD45RA+ \n \n(Callender et al., 2020, Libri et al., \n2011, Ligotti et al., 2023) \n \nCD3+CD19-CD4+CD8-CD27-\nCD45RA+ \nCD8+ T \ncell \nTEMRA, CD27-\nCD45RA+ \n(Callender et al., 2020, \nCzesnikiewicz-Guzik et al., 2008, \nRiddell et al., 2015, Ligotti et al., \n2023) \n \nCD3+CD19-CD4-CD8+CD27-\nCD45RA+ \nB cell Double negative \n(DN), IgD-CD27- \n(Frasca et al., 2017, Colonna-\nRomano et al., 2009, Nevalainen \net al., 2019) \nCD3-CD19+CD27-IgD- \n \nNK \nCD56bright \n \nDouble positive \n(DP), \nCD16+CD57+ \n \n \n \n(Hazeldine et al., 2012, Lopez-\nVerges et al., 2010) \n \n \nCD3-CD19-HLA-DR-\nCD56+CD56brightCD16+CD57+ \n \nNK \nCD56dim \n \nDouble positive \n(DP), \nCD16+CD57+ \nCD3-CD19-HLA-DR-\nCD56+CD56dimCD16+CD57+ \n \nMonocyte \n \nNon-classical \n(NC), CD14-\nCD16+ \n \n \n(Hearps et al., 2012, Nyugen et \nal., 2010, Seidler et al., 2010) \n \n \nCD3-CD19-HLA-DR+CD14-\nCD16+ \n \n  \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint \n\n 39 \nTable 2 Baseline characteristics of participants in the Rapamycin trial \n \nPlacebo (n=5) Rapamycin (n=4) p-value \nAge (years) 64.2 ± 5.34 60.0 ± 4.62 0.1667 (ns) \nBMI (kg/m2) 26.4 ± 3.31 27.1 ± 1.24 >0.9999 (ns) \nData are presented as mean ± SD. P-values are derived from Mann-Whitney tests. ns = not \nsignificant. \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted August 19, 2025. ; https://doi.org/10.1101/2025.08.15.670559doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}