Regulation of T Cell Senescence in Health and Diseases Authors: Sewon Mun, Seungho Choi, Jaemin Kim, Jihyun Kim, Yeonseok Chung Published: 2026 ABSTRACT AbstractAging is accompanied by a progressive decline in immune function, a process termed immunosenescence, which affects both innate and adaptive immune compartments. Among these, the adaptive immune system—and particularly T cells—undergoes the most profound functional and phenotypic alterations, critically impairing host defense against infections, cancer, and vaccination responses. The age-associated decline of adaptive immunity is shaped by the divergent senescence pathways of CD4+ helper and CD8+ cytotoxic T cells. While both lineages enter a state of cell-cycle arrest, their distinct immunological roles dictate fundamentally different molecular triggers, metabolic adaptations, and functional outcomes. This review synthesizes these subset-specific differences, highlighting how DNA damage-dependent mechanisms and DNA damage-independent processes drive distinct senescence phenotypes. Furthermore, we discuss how these differences contribute to immune system remodeling during aging and explore emerging therapeutic strategies targeting metabolic and signaling pathways to mitigate T cell senescence.Keywords: Aging, T cells, Immunosenescence, Mitochondrial dysfunction, Metabolic reprogramming INTRODUCTION Aging exerts a systemic impact on the immune system, leading to a gradual decline in both innate and adaptive immune responses. This process, collectively termed immunosenescence, is characterized by thymic involution, progressive contraction of the naïve T-cell repertoire, and the accumulation of terminally differentiated and dysfunctional T cells (1). As a consequence, aging individuals exhibit increased susceptibility to infections, reduced vaccine efficacy, and a higher incidence of cancer and chronic inflammatory diseases. Notably, these age-associated changes are not uniformly distributed across immune cell types; rather, T cells are particularly vulnerable and are considered central drivers of immunosenescence in both mice (2) and humans (3). A major contributing factor to this decline is thymic involution, a process characterized by the progressive shrinkage and functional decline of the thymus with age. This results in a reduced output of naïve T cells, leading to a contraction of the TCR repertoire and impaired responses to novel Ags. Beyond thymic involution, peripheral T cells themselves undergo intrinsic age-associated functional and phenotypic alterations, a process referred to as T cell senescence. T cell senescence is a biological process characterized by a state of cell-cycle arrest, occurring in response to various forms of cellular stress (4). As a primary driver of immunosenescence, this process has significant implications for aging-associated immune dysfunction. Senescent T cells undergo profound functional alterations, including loss of proliferative capacity, diminished responsiveness to antigenic stimulation, and acquisition of a senescence-associated secretory phenotype (SASP). Although T cell senescence shares certain features with T cell exhaustion, such as reduced proliferative capacity, these states are mechanistically distinct. A comparative overview highlighting key differences in molecular markers, functional characteristics, and reversibility is provided in Table 1 (5,6). Senescence is widely considered irreversible and is driven by cumulative cellular stress, including DNA damage and metabolic dysfunction. Yet, emerging evidence challenges this strict irreversibility. Rather than a permanent endpoint or a simple binary switch, senescence is increasingly understood as a dynamic state that requires continuous active maintenance, such as ongoing epigenetic remodeling. If these maintenance mechanisms are interrupted, cells may potentially escape the arrest and progress to highly altered post-senescent states (7). Table 1Comparative features between T cell senescence and exhaustionAspectT cell senescenceT cell exhaustionPrimary triggersAging, chronic inflammation, metabolic stress, DNA damageChronic antigen stimulation (e.g., persistent infection, cancer)ProliferationIrreversibly arrested or severely impairedReduced but partially recoverableReversibilityGenerally irreversible (but dynamically maintained)Reversible upon removal of antigen of checkpoint blockadeKey surface markersCD57, KLRG1, loss of CD27/CD28PD-1, Tim-3, LAG-3, TIGITEffector functionAltered; may retain or acquire non-classical functionsGradual loss of effector functionsCytokine productionSASP-like pro-inflammatory secretionReduced cytokine production (e.g., IL-2, TNF)Metabolic stateMitochondrial dysfunction, altered bioenergeticsMetabolic insufficiency, reversible dysfunctionEpigenetic stateStable chromatin remodelingDistinct but more plastic epigenetic stateTim-3, T cell immunoglobulin and mucin-domain containing-3; LAG-3, lymphocyte activation gene-3; TIGIT, T cell immunoreceptor with Ig and ITIM domains. Comparative features between T cell senescence and exhaustion Tim-3, T cell immunoglobulin and mucin-domain containing-3; LAG-3, lymphocyte activation gene-3; TIGIT, T cell immunoreceptor with Ig and ITIM domains. Importantly, the pathways leading to senescence and their functional consequences are not uniform but are shaped by the distinct biological roles of T cell subsets. CD4+ helper T cells, which orchestrate immune responses, and CD8+ cytotoxic T cells, which directly eliminate infected or malignant cells (8,9), follow fundamentally different senescence trajectories. These divergent lifelong demands give rise to distinct molecular mechanisms, metabolic adaptations, and functional endpoints during aging (Fig. 1). Hierarchical overview of immunosenescence and subset-specific trajectories of T cell aging. Schematic illustration of age-associated immunosenescence across innate and adaptive immune compartments. Aging impairs innate immune cells, including macrophages, neutrophils, NK cells, and DCs, leading to reduced pathogen sensing, phagocytic activity, and dysregulated cytokine production. In the adaptive immune system, both B and T cells undergo progressive remodeling, characterized by reduced naïve cell output and diminished antigen responsiveness. Within the T cell compartment, aging-associated changes are subset-specific. CD4+ T cells show relative resilience, maintaining metabolic flexibility while shifting toward pro-inflammatory phenotypes and contributing to age-associated inflammatory conditions. In contrast, CD8+ T cells are more susceptible to replicative and metabolic stress, resulting in the accumulation of TEMRA, reduced proliferative capacity, and impaired antigen-specific responses. These distinct features reflect divergent senescence trajectories between CD4+ and CD8+ T cells, highlighting that immunosenescence is shaped by lineage-specific vulnerabilities and functional adaptations. DC, dendritic cell. Recent advances in single-cell transcriptomics, multi-omics profiling, and epigenetic analyses have substantially refined our understanding of T cell senescence. These approaches have revealed age-associated clonal remodeling, loss of naïve T cell diversity, and stable chromatin remodeling that reinforces subset-specific dysfunctional states (10,11). Together, these findings underscore the need for a framework that integrates the divergent senescence trajectories of CD4+ and CD8+ T cells. In this review, we synthesize current evidence on the mechanistic, metabolic, and functional distinctions between these subsets, with the aim of informing future therapeutic strategies targeting age-associated immune dysfunction. SENESCENCE-INDUCING FACTORS AND MECHANISMS T cell senescence is induced by a complex interplay of intrinsic and extrinsic stressors that accumulate over the lifespan. Unlike acute activation-induced differentiation, senescence arises from chronic and unresolved cellular stress, including persistent antigenic stimulation, DNA damage, oxidative stress, mitochondrial dysfunction, and inflammatory cytokine exposure. Importantly, these triggers do not operate independently but converge on shared signaling networks that enforce stable cell-cycle arrest and functional remodeling. In the aging immune system, both DNA damage—dependent and DNA damage—independent pathways contribute to this process, and their relative contributions vary depending on the nature and chronicity of the inducing stress. DNA damage-dependent pathway T cells can enter a state of senescence independent of extensive cell division, with chronic inflammation being one of the key triggers. Persistent exposure to pro-inflammatory cytokines such as TNF-α and IL-6, characteristic of chronic inflammatory states, induces high levels of intracellular stress and ROS. This inflammation-driven stress directly causes DNA damage, activating the DNA damage response (DDR) and leading to cellular senescence without significant telomere shortening (12). Furthermore, senescent T cells themselves adopt SASP characterized by proinflammatory mediators. The expression of these SASP factors is sustained through NF-κB activation driven both by persistent DDR signaling and by autocrine and paracrine SASP mediators themselves, forming a self-reinforcing feedback loop that perpetuates chronic inflammation (13). A second route to DNA damage-dependent senescence operates through replicative attrition rather than exogenous inflammatory stress. This process, in particular, reveals distinct differences between CD4+ and CD8+ T cell subsets, stemming from fundamental differences in how they manage telomere biology. Telomeres, the repetitive nucleotide sequences at the ends of linear chromosomes, act as a 'mitotic clock,' shortening with each cell division and triggering replicative senescence upon reaching a critical length. While both T cell subsets experience telomere attrition as they differentiate from naïve to memory populations (14), this process is significantly more pronounced in the CD8+ lineage (15). Memory CD8+ T cells often possess shorter telomeres than even their naïve counterparts, a direct reflection of the extensive clonal expansion required to control infections. This accelerated telomere shortening is mechanistically linked to the differential regulation of telomerase, the reverse transcriptase enzyme responsible for elongating telomeres. While CD4+ T cells can sustain telomerase activity over multiple rounds of stimulation, CD8+ T cells exhibit a rapid decline in this capacity. By the fourth stimulation, telomerase activity in CD8+ T cells can become virtually undetectable in vitro (16). This premature loss of telomere maintenance is further exacerbated by the earlier downregulation of the co-stimulatory molecule CD28 in CD8+ T cells, a key signal for telomerase induction (17). This intrinsic vulnerability is illustrated in the context of chronic viral infections, such as with Epstein-Barr virus. During the acute phase of infection, Ag-specific CD8+ T cells can preserve their telomere length to support the necessary expansion. However, once viral latency is established and the cells enter a state of chronic surveillance, they exhibit marked telomere shortening, progressively driving them toward senescence (18). This transition is thought to be driven by multiple converging mechanisms. During acute infection, Ag-driven CD28 costimulation transiently induces telomerase activity, thereby preserving telomere length during clonal expansion (19). In contrast, repeated antigenic stimulation under chronic or latent conditions leads to progressive loss of CD28 expression and reduced telomerase inducibility. Sustained type I interferon signaling may further suppress human telomerase reverse transcriptase (hTERT) expression through activation of the p38 MAPK pathway (20). In addition, oxidative stress at telomeric regions accelerates telomere attrition, and critically shortened telomeres can activate an ataxia telangiectasia mutated (ATM)/p53-dependent DDR, ultimately promoting replicative senescence (21). Despite these insights, the precise molecular events governing this transition—particularly the relative contributions of CD28 loss, hTERT regulation, and chronic inflammatory signaling—remain incompletely understood. Collectively, these findings suggest that the intrinsic capacity for telomere maintenance is a key determinant of replicative lifespan, rendering the highly proliferative CD8+ T cell compartment particularly vulnerable to senescence. DNA damage-independent pathway Beyond DNA damage-driven mechanisms, T cell senescence can also be induced through pathways that operate independently of overt genomic instability. Among these, metabolic dysregulation has emerged as a particularly critical determinant, as mitochondrial integrity is intrinsically linked to T cell fate and longevity. Disruption of mitochondrial homeostasis caused by altered nutrient sensing, impaired biogenesis, or loss of membrane potential can trigger senescence-associated programs even in the absence of canonical DDR activation (22). Importantly, the vulnerability to such mitochondrial deterioration differs markedly between CD4+ and CD8+ T cell subsets, revealing a layer of subset-specific regulation that shapes divergent senescence trajectories. Mitochondrial integrity is a critical determinant of cellular longevity and function, and its decline is a central, causal driver of T cell senescence (23). While the broader population of senescent T cells, particularly terminally differentiated effector memory cells re-expressing CD45RA (TEMRA), generally displays reduced mitochondrial mass, significant subset-specific differences exist (24). The CD4+ TEMRA cells maintain a comparatively higher mitochondrial mass and exhibit augmented biogenetic capacity relative to their CD8+ TEMRA counterparts. This enhanced capacity is largely mediated by the sustained activity of peroxisome proliferator-activated receptor-gamma coactivator 1-alpha (PGC-1α), a master regulator of mitochondrial biogenesis (25,26). This mitochondrial robustness endows senescent CD4+ T cells with superior metabolic plasticity, allowing them to efficiently utilize both oxidative phosphorylation (OXPHOS) and glycolysis. Functionally, this translates into higher basal oxygen consumption rates, greater maximal respiratory capacity, and preserved ATP generation, which collectively confer a metabolic stability that mitigates age-associated functional decline. In contrast, CD8+ T cells undergoing senescence undergo a pronounced and progressive collapse of their mitochondrial network. This is characterized by a sharp decline in mitochondrial content, reduced membrane potential, and defective biogenesis, which has been attributed to diminished signaling by nuclear respiratory factor 1 and impaired transcriptional regulation of respiratory chain components (27). These defects precipitate a metabolic crisis, forcing a shift toward glycolytic dependence. This inefficient metabolic state is accompanied by excessive production of mitochondrial ROS and the accumulation of oxidative damage, which further accelerates the progression toward a senescent phenotype. The causal link between mitochondrial health and subset-specific senescence is not merely correlational. Experimental evidence demonstrates that inducing mitochondrial impairment in healthy CD4+ T cells, for example by using pharmacological inhibitors of mitochondrial complex I, is sufficient to drive them into a state that phenocopies CD8+ T cell senescence. These manipulated CD4+ T cells exhibit increased p53 phosphorylation, fragmented mitochondrial morphology, and an accelerated entry into senescence (25). This mechanistic validation reveals that the inherent mitochondrial robustness of CD4+ T cells is a primary factor distinguishing their senescence susceptibility from that of CD8+ T cells, positioning mitochondrial integrity as a central determinant of T cell aging. Recent evidence suggests that age-associated gut dysbiosis may act as an extrinsic driver of T cell senescence by promoting systemic inflammaging and barrier dysfunction. Notably, senescent mouse models exhibit altered microbiota composition and increased gut permeability, while adoptive transfer of functional CD4+ T cells partially restores microbial homeostasis and barrier integrity, highlighting a bidirectional and potentially reversible gut-immune axis (28). PHENOTYPIC AND FUNCTIONAL HALLMARKS Beyond the initiating mechanisms, senescent T cells exhibit a distinct set of phenotypic, molecular, and functional hallmarks that collectively define their biological identity. These include structural alterations, changes in surface marker expression, differentiation-state skewing, and profound functional reprogramming. Importantly, these characteristics differ substantially between CD4+ and CD8+ T-cell subsets and serve as key indicators of age-associated immune remodeling. Morphological and lysosomal features In addition to the molecular pathways described above, T cell senescence is also accompanied by distinct morphological and intracellular structural changes. Although cellular enlargement and flattening have been extensively characterized in fibroblast models of senescence, similar phenotypic features have also been observed in senescent lymphocytes, although they remain less systematically defined (29). Senescent lymphocytes undergo significant morphological transformations compared to their non-senescent counterparts. These cells become markedly larger and adopt a flattened appearance (30). A prominent feature of this phenotype is lysosomal expansion. Senescent T cells exhibit increased lysosomal accumulation and content, a phenomenon commonly referred to as lysosomal hypertrophy. This structural remodeling is closely associated with increased senescence-associated β-galactosidase (SA-β-gal) activity, one of the most widely used hallmarks of cellular senescence (31). In T cells, elevated SA-β-gal activity likely reflects increased lysosomal biogenesis and proteostatic burden driven by chronic activation, replicative stress, and age-associated metabolic dysregulation (22). Thus, lysosomal remodeling represents an important cellular manifestation of T cell senescence and may contribute to impaired immune fitness during aging. Molecular markers The molecular signature of senescent T cells is characterized by the progressive downregulation of key co-stimulatory molecules, particularly CD27 and CD28, together with the upregulation of senescence-associated markers such as killer cell lectin-like receptor subfamily G member 1 (KLRG1) and CD57, reflecting repeated Ag exposure, chronic stimulation, and age-associated cellular stress (32). These changes fundamentally impair T cell activation and proliferative capacity. These phenotypic shifts follow distinct trajectories depending on the T-cell subset. As senescence progresses, both CD4+ and CD8+ T cells eventually converge on a terminal phenotype characterized by KLRG1 and CD57 co-expression, although the sequence of co-stimulatory molecule loss differs between subsets. CD4+ T cells typically lose CD27 before CD28, whereas CD8+ T cells generally lose CD28 first (33). CD57 is a well-established marker of T cell senescence and is associated with shortened telomeres and reduced proliferative capacity (34). Senescent T cells also express elevated levels of CD107a (lysosomal-associated membrane protein 1), granzyme B (GzmB), and perforin, even in the absence of stimulation, and may display TCR-independent cytotoxic activity resembling that of NK cells (32). Additional surface alterations include increased CD45RA and reduced CCR7 and CD45RO, consistent with terminal differentiation (34). Moreover, emerging markers such as T cell immunoreceptor with Ig and ITIM domains, T cell immunoglobulin and mucin-domain containing-3, and various NK-like receptors have been associated with senescent phenotypes, expanding the molecular framework used to identify these cells (35,36). At the intracellular level, senescent T cells exhibit increased expression of cell-cycle regulators, including p53, p21, and p16, together with activation of DDR pathways marked by phosphorylated histone H2A.X and ATM kinase (37,38). Recent single-cell transcriptomic analyses across diverse human and murine aging tissues further suggest that p16-positive and p21-positive cells are rarely co-expressed within the same cell, but instead define distinct senescent populations, or “senotypes,” with separate developmental trajectories, divergent SASP programs, and different intercellular communication patterns (37). These findings underscore that cellular senescence is not a single uniform state, but rather a heterogeneous process shaped by cell type and tissue context. Recent epigenomic studies further indicate that aging and senescence in T cells are accompanied by stable chromatin remodeling programs that reinforce long-term dysfunction, including HELIOS-associated chromatin changes in aged naïve CD4+ T cells and enhancer remodeling networks in senescent CD8+ T cells (39,40). Differentiation states and memory/effector cell composition The cellular processes of senescence do not occur in isolation; they profoundly reshape the composition and functional balance of the entire T cell compartment, leading to the well-documented hallmarks of immunosenescence. These changes are particularly pronounced in the CD8+ lineage. With advancing age, the pool of naïve T cells declines in both subsets, but this reduction is significantly sharper for CD8+ T cells (41). This is accompanied by a preferential expansion and accumulation of TEMRA cells within the CD8+ compartment (42). This dramatic skewing of the CD8+ population is a major contributor to the age-associated inversion of the CD4:CD8 ratio, a widely recognized biomarker of immune dysfunction and inflammaging (43). Recent single-cell transcriptomic and paired TCR-sequencing studies have further shown that aging is accompanied by marked clonal remodeling of the T cell compartment, including expansion of highly differentiated CD8+ subsets and reshaping of TCR repertoire structure (44,45). The stability of the most primitive memory populations also differs with age. T stem cell memory (TSCM) cells, which possess the greatest capacity for self-renewal and multilineage differentiation (46), are relatively well preserved within the CD4+ lineage during aging, potentially supporting long-term maintenance of immunological memory. In contrast, the CD8+ TSCM pool declines substantially with age. Despite this absolute decline, the ratio of CD8+ TSCM to naïve cells increases, suggesting a compensatory adaptation to sustain the peripheral T-cell pool as thymic output wanes (47). Importantly, preservation of CD4+ TSCM cells should not be interpreted as uniformly beneficial. Recent work suggests that age-inappropriate maintenance of CD4+ TSCM cells may also contribute to pathology, particularly in the context of autoimmune disease (48). Sustained CD4+ TSCM populations, supported by specialized tissue niches through Notch receptor-Jagged1 signaling, can continuously replenish pathogenic effector T cells and thereby promote chronic autoimmune inflammation in aged hosts. These findings suggest that progressive loss of T-cell stemness during aging may represent a host-protective adaptation against autoimmunity, highlighting the potential risks associated with indiscriminate immune rejuvenation strategies. Within the CD4+ lineage, aging biases differentiation toward short-lived effector cells at the expense of long-lived memory and T follicular helper subsets. Senescent CD4+ T cells preferentially adopt pro-inflammatory Th1 and Th17 phenotypes, contributing to the chronic, low-grade inflammation of aging, often referred to as "inflammaging” (49). This shift is driven by transcriptional reprogramming, characterized by the upregulation of B lymphocyte-induced maturation protein 1 and runt-related transcription factor 3 (50,51), along with the concomitant loss of memory-associated factors, including T cell factor 1 and lymphoid enhancer-binding factor 1 (52). Meanwhile, recent studies further indicate that the age-expanded CDT8+ TEMRA compartment is accompanied by impaired terminal differentiation of virus-specific populations, suggesting defective maintenance of effective antiviral immunity. This altered differentiation trajectory has been linked to dysregulated expression of transcription factors such as zinc finger E-box-binding homeobox 2 and Kruppel-like factor 2 (53,54). Functional reprogramming in senescent T cells Senescent T cells, while having lost their ability to proliferate due to cell cycle arrest, remain metabolically active and exert a powerful influence on their surrounding tissue microenvironment through the SASP. In this process, senescent T cells have been confirmed to secrete not only classical SASP factors like IL-6, IL-8, CXCL1, and CXCL2, but also a variety of inflammatory and immunomodulatory cytokines and chemokines such as TNF-α, IL-18, IL-29, CCL5, CCL16, CCL23, and IFN-γ (32,55,56). The expression of granzyme K (GzmK) increases significantly with age, particularly in CD8+ T cells, and is strongly associated with chronic cytomegalovirus infection, a major driver of T-cell senescence (57). Unlike classical cytotoxic granzymes such as GzmB, which primarily mediate perforin-dependent target cell apoptosis, GzmK is a serine protease with distinct functional properties and is increasingly recognized as a mediator of inflammatory and non-canonical effector responses. GzmK+ CD8+ T cells typically exhibit relatively weak direct cytotoxicity due to low expression of GzmB and perforin, yet display pronounced pro-inflammatory activity through robust production of IFN-γ and TNF (58). In addition to its role as a phenotypic marker of senescence-associated T cell subsets, GzmK may contribute functionally to chronic tissue inflammation and age-associated immune dysregulation through protease-dependent modulation of inflammatory signaling pathways. GzmK expression has also been identified in emerging cytotoxic CD4+ T cell subtypes (59). The reprogramming of cytotoxic capabilities manifests as 'functional devolution' in senescent CD8+ T cells. These cells exhibit reduced intrinsic cytotoxic capacity that normally depends on TCR-mediated Ag recognition. Instead, they acquire an innate-like killing mechanism by expressing NK cell-associated receptors such as NKG2D. This is mediated by the ITAM adapter molecule DNAX-activating protein of 12 kDa, which bypasses TCR signaling and allows the cells to eliminate stressed or malignant cells (60). In contrast, a much more dramatic functional neogenesis occurs in senescent CD4+ T cells. These cells, which classically serve helper functions, can acquire cytotoxic properties and differentiate into CD4+ CTLs through de novo expression of GzmB and perforin, driven in part by the induction of the lineage-inappropriate transcription factor eomesodermin (Eomes) (61). Recent studies have highlighted that this phenotypic conversion is not merely a marker of immune aging but may represent an adaptive response to age-associated tissue stress. Notably, Hasegawa et al. (62) demonstrated that cytotoxic CD4+ T cells selectively eliminate senescent cells by recognizing cytomegalovirus-derived Ags presented on senescent target cells, thereby contributing to endogenous senolytic immune surveillance. Building on this concept, recent work further demonstrated that age-associated CD4+ T cells acquire Eomes-dependent transcriptional programs that actively modulate cellular senescence and organismal aging (63). METABOLIC CHARACTERISTICS Senescent T cells undergo extensive metabolic reprogramming that alters their glucose, lipid, and mitochondrial metabolism. This metabolic transformation is not merely a consequence of aging but represents a driving mechanism that promotes and maintains T cell senescence (64). Glucose metabolism Despite being proliferatively arrested, senescent T cells undergo a metabolic shift toward aerobic glycolysis, driven by the sustained activation of the PI3K/protein kinase B (Akt)/mTOR pathway (65,66). This process is marked by increased expression of glucose transporter 1 and key glycolytic enzymes (67). This enhanced glycolytic flux serves not merely to support cell survival but also to provide the substantial energy and biosynthetic precursors required for the synthesis and secretion of the SASP, which is regulated by the mTOR and p38 MAPK pathways (64). Lipid metabolism The lipid metabolism of senescent T cells enters a state of dysregulation, visually hallmarked by the prominent accumulation of cytoplasmic lipid droplets (LDs) (65). This condition arises from a profound imbalance between fatty acid influx and consumption, where senescent cells upregulate both fatty acid uptake transporters, such as CD36, and key enzymes involved in de novo fatty acid synthesis, including fatty acid synthase and acetyl-CoA carboxylase. Concurrently the capacity to utilize these lipids as fuel through mitochondrial fatty acid oxidation is severely impaired, resulting in elevated LDs level (64,68). This metabolic disarray is actively exacerbated by stress signals, which activate MAPK/STAT pathways to increase levels of cytosolic phospholipase A2α. This enzyme liberates excessive free fatty acids from membrane phospholipids, overwhelming cellular processing capacity and directly fueling LD formation, thereby accelerating the onset of senescence (64,69). Excess lipid accumulation creates a vicious feedback loop, imposing further stress on already damaged mitochondria and reinforcing the inflammatory senescent state. Mitochondrial dysfunction In senescent T cells, dysfunctional mitochondria are a major driver of metabolic collapse. Rather than simply failing to produce energy, they can transform into active signaling platforms that reinforce the senescent state. Mitochondria in senescent T cells are characterized not only by impaired electron transport chain (ETC) function but also by altered morphology and membrane hyperpolarization, a combination that promotes excessive ROS generation and further exacerbates mitochondrial damage (64). In this context, emerging evidence suggests that lipid peroxidation-driven cell death pathways, such as ferroptosis, are closely linked to metabolic stress and redox imbalance, further contributing to immune cell dysfunction and senescence-associated phenotypes (70). Supporting the central role of mitochondrial dysfunction in immune aging, recent evidence has shown that T cells harboring dysfunctional mitochondria are sufficient to drive systemic multimorbidity and premature senescence phenotypes, highlighting the causal contribution of T cell mitochondrial defects to organismal aging (23). Furthermore, a recent aging-associated T cell study identified KLRG1 as a marker of regulatory T cell subsets with mitochondrial alterations that progressively accumulate with age, suggesting that mitochondrial dysfunction is linked not only to effector T-cell senescence but also to age-associated remodeling of immunoregulatory populations (71). A critical downstream consequence of this mitochondrial dysfunction is the surge in mitochondrial ROS (mtROS). Elevated mtROS not only heightens cellular oxidative stress but also contributes to the induction of SASP. Mechanistically, stress-induced permeabilization of the mitochondrial outer membrane leads to leakage of mitochondrial DNA (mtDNA) into the cytosol. There, the innate immune sensor cyclic GMP-AMP synthase detects mtDNA and activates the stimulator of interferon genes (STING) pathway, driving inflammatory SASP gene expression and effectively translating a metabolic defect into a sustained inflammatory state (72,73). Taken together, the integrated phenotypic, functional, and metabolic hallmarks that define senescent T cells are summarized in Fig. 2. A comparative overview of the major differences between senescent CD4+ and CD8+ T cells, including their phenotypic, functional, and metabolic features, is provided in Table 2. Phenotypic, molecular, and functional characteristics of senescent CD4+ and CD8+ T cells. Comparative schematic of senescent CD4+ and CD8+ T cells. Senescent T cells exhibit loss of co-stimulatory molecules (CD27, CD28) and increased expression of senescence-associated markers (KLRG1, CD57), along with lysosomal expansion and SA-β-gal activity. Aging reshapes T cell composition, with a reduction in naïve T cells, more pronounced in CD8+ T cells, which accumulate TEMRA, while CD4+ T cells shift toward pro-inflammatory phenotypes. Functionally, senescent T cells display reduced proliferation and a SASP. CD8+ T cells show impaired TCR-dependent cytotoxicity and increased reliance on NK receptor pathways, whereas CD4+ T cells can acquire cytotoxic features. GLUT1, glucose transporter 1. Table 2Comparative features of senescence trajectories in CD4 and CD8 T cellsAspectCD4+ T cellsCD8+ T cellsTelomeresMaintain telomerase activity to preserve telomere lengthRapidly lose telomerase activity, leading to accelerated telomere shorteningMitochondriaMaintain relatively healthy mitochondria and stable metabolic functionUndergo mitochondrial collapse and metabolic crisisSurface marker loss patternCD27 loss followed by CD28 lossCD28 loss followed by CD27 lossFunctional reprogrammingFunctional neogenesis: acquire cytotoxic properties (CD4+ CTLs)Functional devolution: loss of TCR-dependent cytotoxicity with gain of NK-like cytotoxicityImmune impactBetter preserve the memory stem cell pool and adopt inflammatory phenotypesDepletion of naïve/stem cell pools, accumulation of TEMRA cells, and inversion of the CD4:CD8 ratio Comparative features of senescence trajectories in CD4 and CD8 T cells THERAPEUTIC STRATEGIES Recent studies have revealed that the function and aging process of T cells are closely linked to cellular metabolic pathways. Specifically, the mTOR and p38 MAPK signaling pathways play crucial roles in regulating the metabolic reprogramming and functional changes of senescent T cells. Therefore, therapeutic strategies targeting these metabolic pathways to control T cell senescence are being actively researched. This section will discuss the roles of the mTOR and p38 MAPK pathways in T cell senescence and the therapeutic approaches to inhibit them. In addition, from the perspective of T cell metabolism, enhancing mitochondrial function is being proposed as a method to control T cell senescence (Fig. 3). Therapeutic targeting of metabolic and signaling pathways in senescent T cells. Overview of therapeutic strategies targeting mTOR, p38 MAPK, mitochondrial dysfunction, and NAD+ metabolism to mitigate T-cell senescence. Schematic overview of therapeutic strategies targeting T cell senescence through metabolic and signaling pathways. Chronic activation of the PI3K-Akt-mTOR axis drives metabolic reprogramming and promotes the SASP, which can be mitigated by mTOR inhibition. The p38 MAPK pathway contributes to proliferative arrest and senescence maintenance, and its inhibition can partially restore T cell function. Strategies to improve mitochondrial function, including enhancement of mitochondrial biogenesis and restoration of NAD+ levels, further support metabolic recovery and may alleviate senescence-associated dysfunction. Together, these approaches highlight the potential of targeting metabolic vulnerability to modulate T cell aging. mTOR inhibition mTOR is a key metabolic regulatory kinase that controls cell growth, proliferation, and metabolism, functioning through two distinct complexes, mTORC1 and mTORC2 (74). In T cells, this pathway is activated by TCR and co-stimulatory signals, which triggers the PI3K-Akt signaling cascade, orchestrating a transition from the catabolic state of quiescent T cells to the highly anabolic state required for effector functions. Specifically, activated mTORC1 promotes protein synthesis by phosphorylating S6K1 and 4E-BP1, and excessively increases anabolic processes such as glycolysis and lipid synthesis (75). However, in the context of aging, chronic antigenic stimulation or cellular stress causes the persistent activation of this PI3K-Akt-mTOR axis. This chronic hypermetabolic state not only supplies the energy needed for senescent T cells to actively produce SASP factors that cause inflammation in surrounding tissues but also leads to mitochondrial dysfunction, increased ROS generation, and DNA damage accumulation, thereby reinforcing the senescent phenotype (76,77). Therefore, inhibiting mTOR is a promising therapeutic strategy to counteract T cell senescence. The prototypical mTOR inhibitor, rapamycin, effectively controls senescence through two primary mechanisms: fundamentally remodeling T cell metabolism and function, and blocking detrimental inflammatory signals. First, rapamycin resets the metabolic and functional programs of senescent T cells by normalizing their metabolic state through the inhibition of excessive glycolysis (78). Second, it effectively suppresses the secretion of the SASP, one of the most harmful features of senescent cells. Specifically, rapamycin not only lowers mRNA levels of inflammatory cytokines such as IL-6 but also selectively blocks the translation of the membrane-bound cytokine IL1A. This, in turn, reduces the transcriptional activity of NF-κB, a master regulator of the SASP, leading to a broad suppression of the inflammatory phenotype (79). However, the long-term use of mTOR inhibitors is associated with notable metabolic side effects, including hyperglycemia, dyslipidemia, insulin resistance, and increased risk of new-onset diabetes, largely attributed to off-target inhibition of mTORC2 signaling (80). In addition, impaired wound healing and mucocutaneous toxicity have been frequently reported in transplant recipients receiving mTOR inhibitors, along with potential reproductive dysfunction, highlighting the need for careful dose optimization and patient selection (81). p38 MAPK inhibition Similar to mTOR inhibitors, the p38 MAPK pathway represents a strategy to target senescence by modulating the metabolic pathways of T cells. Cellular senescence can be categorized into two main types. Replicative senescence is telomere-dependent and occurs after a critical number of cell divisions. In contrast, telomere-independent senescence, which can be induced by factors like chemotherapy, primarily serves to prevent tumorigenesis following oncogene activation or DNA damage (82). As a central regulator of cellular senescence, the p38 MAPK signaling pathway responds to various stressors, including telomere shortening and oncogenic stress. However, in senescent T cells, p38 is not activated through conventional signaling cascades driven by external stimuli but instead through a unique “intrasensory” pathway initiated by intracellular metabolic stress or DNA damage. Specifically, this abnormal p38 activation, mediated by the AMP-activated protein kinase-TAB1 protein, acts as a primary driver that induces and maintains the senescent state of T cells by inhibiting telomerase activity and cell proliferation. Indeed, it has been experimentally demonstrated that selectively blocking this pathway successfully reverses the proliferative defects of senescent T cells (83). Enhancing mitochondrial function Mitochondrial enhancement addresses the fundamental mitochondrial dysfunction characteristic of senescent T cells. Interventions that promote mitochondrial biogenesis, improve OXPHOS, and reduce mitochondrial-derived oxidative stress show promise for reversing senescence-associated metabolic dysfunction. As mitochondria are the primary sites of cellular iron metabolism, appropriate iron availability can help restore mitochondrial function in aging T cells (84). Evidence from model organisms supports this concept; for example, studies in yeast have shown that iron supplementation enhances the expression of genes involved in the tricarboxylic acid cycle and the ETC, leading to improved ATP production and an extended chronological lifespan (85). These findings suggest that iron availability promotes mitochondrial activity not only through enhanced energy generation but also by supporting iron-sulfur cluster biogenesis and preserving mitochondrial integrity. However, this balance becomes considerably more complex in the context of aging, as heme and iron toxicity in the aged spleen can impair T-cell immunity by inducing functional iron deprivation, suggesting that excessive iron accumulation within the aging tissue microenvironment may paradoxically reduce iron bioavailability to T cells and compromise their mitochondrial fitness (86). Nicotinamide adenine dinucleotide (NAD+) is another critical therapeutic target, as its intracellular levels decline with age, contributing to reduced oxidative capacity and immune dysfunction. As an essential cofactor for sirtuins such as SIRT1, NAD+ regulates mitochondrial biogenesis through PGC-1α deacetylation and supports mitochondrial quality control (87,88). In B16F10 melanoma-bearing mice, tumor-infiltrating T cells frequently display senescence-associated phenotypes. Supplementation with NAD+ precursors, particularly nicotinamide mononucleotide (NMN), has been shown to mitigate this process. By elevating intracellular NAD+ levels, NMN restores mitochondrial activity and alleviates STING-dependent T cell senescence (89). These findings are further supported by emerging translational evidence linking T cell senescence to immunotherapy resistance. The therapeutic relevance of this metabolic restoration is further underscored in the context of cancer immunotherapy. Interferon-induced senescent CD8+ T cells reduce the efficacy of anti-PD-1 immunotherapy, highlighting the clinical significance of strategies aimed at reversing T-cell senescence and restoring mitochondrial fitness (90). Moreover, the benefits of NAD+ restoration are significantly amplified when combined with strategies that prevent its depletion, particularly through CD38-mediated NAD+ consumption. Combination therapy using NAD+ precursors with CD38 inhibitors (e.g., compound 78c) more effectively restores NAD+ levels and improves T cell mitochondrial function than precursors alone. This strategy also holds promise in chimeric Ag receptor T cell therapy, where age-related NAD+ decline can impair antitumor activity (91). While strategies aimed at enhancing mitochondrial function, including NAD+ boosting, show promise in restoring cellular bioenergetics, their systemic effects may be context-dependent. Although NAD+ supports mitochondrial function and DNA repair in normal cells, it may also facilitate tumor cell metabolic adaptation and survival through pathways such as PARP activation and NAMPT-mediated immunosuppressive signaling (92). These findings underscore the need for a more nuanced, cell type-specific approach to metabolic interventions. More direct interventions include mitochondrial transplantation, a novel approach that has shown high efficacy in preclinical cancer immunotherapy models. T cells receiving mitochondria from bone marrow-derived mesenchymal stem cells display augmented antitumor responses, prolonged survival, and greater persistence within the tumor microenvironment (87). Moreover, mitochondrial transfer into naïve CD4+ T cells has been shown to enhance protective effector and memory T cell formation during Mycobacterium tuberculosis infection in mice. In addition, it boosts mitochondrial mass and reshapes cytokine production, leading to reduced exhaustion and senescence in elderly human T cells. Together, these results highlight mitochondrial transfer as a promising strategy to rejuvenate aged CD4+ T cells and improve immune responses in both elderly individuals and patients with tuberculosis (93). Despite its promise, significant technical and translational hurdles remain, including the efficiency of mitochondrial transfer, the optimization of ex vivo delivery methods, and safety considerations regarding mitochondrial integration and immune recognition. Finally, PGC-1α enhancement represents a promising therapeutic strategy for immunosenescence. Overexpression of PGC-1α in skeletal muscle is known to prevent age-associated muscle wasting and maintain metabolic function (94). Similar principles could apply to T cells, where sustained PGC-1α expression could preserve function and prevent senescence (88). However, therapeutic targeting requires careful consideration of tissue- and cell-type-specific isoforms, which are driven from distinct promoters and are differentially regulated during aging (95). Despite their therapeutic potential, long-term modulation of these pathways may involve important clinical limitations, including immunosuppression, altered immune homeostasis, and systemic metabolic trade-offs associated with chronic mTOR inhibition or NAD+ supplementation. CONCLUSION AND PERSPECTIVE The process of T cell senescence is not a uniform decline but a highly nuanced and divergent series of events, shaped fundamentally by the distinct biological roles of CD4+ and CD8+ T cell subsets. This review has illuminated that these lineages follow fundamentally different trajectories into senescence, driven by intrinsic differences in telomere maintenance, metabolic programming, and mitochondrial resilience. The accelerated replicative exhaustion of CD8+ T cells presents a clear contrast to the relative metabolic and mitochondrial robustness of their CD4+ counterparts. This divergence culminates in a paradoxical and profound functional reprogramming. Rather than a simple loss of function, senescence remodels T cell capabilities, leading to the 'functional devolution' of CD8+ cells towards innate-like cytotoxicity and the 'functional neogenesis' of CD4+ cells into novel senolytic effectors. These contrasting fates are central to understanding the architectural shifts in the aging immune system, including the inversion of the CD4:CD8 ratio, and contribute directly to the chronic, low-grade inflammation characteristic of aging. A clear understanding of these subset-specific mechanisms is paramount, moving beyond a generalized view of immunosenescence to one that appreciates the unique contributions of each T cell population. This refined perspective is not merely academic; it reveals a landscape of precise therapeutic targets. Strategies aimed at modulating the mTOR and p38 MAPK pathways or enhancing mitochondrial function hold significant promise for selectively intervening in these processes. Future research focused on harnessing these distinct pathways will be critical for developing effective interventions to ameliorate the course of immunosenescence and mitigate the burden of age-related disease. ACKNOWLEDGEMENTS This study was supported by the National Research Foundation of Korea grants (2022R1A6A1A03046247 and RS-2024-00444468) and from the Institute for Basic Science (IBS-R801-D2). The figures were created with BioRender.com. Abbreviations protein kinase B ataxia telangiectasia mutated DNA damage response eomesodermin electron transport chain granzyme B granzyme K human telomerase reverse transcriptase killer cell lectin-like receptor subfamily G member 1 lipid droplet mitochondrial DNA mitochondrial ROS nicotinamide adenine dinucleotide nicotinamide mononucleotide oxidative phosphorylation peroxisome proliferator-activated receptor-gamma coactivator 1-alpha senescence-associated β-galactosidase senescence-associated secretory phenotype stimulator of interferon genes terminally differentiated effector memory cells re-expressing CD45RA T stem cell memory Footnotes Conflict of Interest: The authors declare no potential conflicts of interest. Author Contributions: Conceptualization: Choi S, Mun S, Kim J.1Funding acquisition: Chung Y.Supervision: Chung Y.Visualization: Mun S, Kim J.2Writing - original draft: Choi S, Mun S, Kim J1, Kim J.2Writing - review & editing: Kim J2, Chung Y. Conceptualization: Choi S, Mun S, Kim J.1 Conceptualization: Choi S, Mun S, Kim J.1 Funding acquisition: Chung Y. Funding acquisition: Chung Y. Supervision: Chung Y. Supervision: Chung Y. Visualization: Mun S, Kim J.2 Visualization: Mun S, Kim J.2 Writing - original draft: Choi S, Mun S, Kim J1, Kim J.2 Writing - original draft: Choi S, Mun S, Kim J1, Kim J.2 Writing - review & editing: Kim J2, Chung Y. Writing - review & editing: Kim J2, Chung Y. Kim J1, Jaemin Kim; Kim J2, Jihyun Kim. References