Senescent T cells: Beneficial and detrimental roles

Summary As the thymus involutes during aging, the T‐cell pool has to be maintained by the periodic expansion of preexisting T cells during adulthood. A conundrum is that repeated episodes of activation and proliferation drive the differentiation of T cells toward replicative senescence, due to telomere erosion. This review discusses mechanisms that regulate the end‐stage differentiation (senescence) of T cells. Although these cells, within both CD4 and CD8 compartments, lose proliferative activity after antigen‐specific challenge, they acquire innate‐like immune function. While this may confer broad immune protection during aging, these senescent T cells may also cause immunopathology, especially in the context of excessive inflammation in tissue microenvironments.


| INTRODUC TI ON
Aging is a dynamic process that progresses with concomitant deterioration of organ function that leads to loss of quality of life.
Immunity also declines during aging. This is highlighted by the increased incidence of malignancy, loss of immunity to previously encountered pathogens, for example, varicella-zoster virus (VZV) causing shingles, 1,2 decreased vaccine efficacy, 3 and decreased ability to respond to new pathogens. 4,5 As the aging of cells in organs and of those within the immune system occur simultaneously, decreased immunity in an older individual may not result from a defect with a particular cell type instead, the problem may arise from altered interactions between aged immune and nonimmune cells in tissues. 6 The term "cellular senescence" was established to describe aging of cell types and this process results in growth arrest that prevents the proliferation of cells harboring DNA damage and is therefore considered to be an anticancer mechanism. 7 Senescent cells can be recognized by phenotypic and functional characteristics, 7,8 and these cells accumulate within both leukocytes (especially T cells) and nonimmune-cell populations during aging. 9,10 Although senescent cells share many functional and phenotypic features, there are also cell-specific differences. 11,12 One shared feature of different senescent cell types is their secretion of multiple cytokines, chemokines, enzymes, etc. (so-called senescence-associated secretory phenotype; SASP). 7,8,10,13 This may be a mechanism for altered communication between immune and nonimmune cells during aging. 8 In this review, we focus on the development of senescent T cells and their phenotypic and functional features. We will also summarize recent information on how these cells may interact within an aged tissue environment. An emerging concept is that instead of being dysfunctional, senescent T cells begin to express the molecular machinery enabling them to behave more like natural killer (NK) cells. 6 While this may confer them with broad nonantigen-specific protection especially during aging, this may also result in immunopathology, where NK ligands are expressed by nonlymphoid cells within inflamed tissue microenvironments.

| T CELL D IFFERENTIATI ON AND THE IDENTIFI C ATI ON OF S ENE SCENT T CELL S
A big advance in identifying T cells at early and end stages of differentiation was the combination of technology to identify telomeres together with cell surface in T-cell subsets by multi-parameter flow cytometry. [14][15][16] This coordinate analysis enabled a pathway of differentiation of human T cells to be defined ( Figure 1). Hence, while naïve T cells in both CD4 + and CD8 + compartments express the co-stimulatory receptors CD28 and CD27 and have long telomeres, end-stage T cells lose expression of these molecules and shorten their telomeres considerably. 17,18 Furthermore, some effector memory (EM) T cells that are CD27 − CD28 − re-express cell surface CD45RA and are known as the EMRA subset. 19,20 Once the definition of the differentiation states of T cells was made, it enabled early and end-stage cells to be isolated and their functional characteristics to be assessed. Thus naïve, central memory and effector cells have different functions and homing potential and tissue localization. 21 In human peripheral blood, CD27 − CD28 − T cells within both CD4 + and CD8 + compartments express multiple cell surface markers of T cells senescence including KLRG1 and CD57 and also intracellular molecules associated with cell cycle arrest and senescence (p16 and p21) 20,22,23 (Figure 1). This was confirmed by single-cell RNAseq analyses of human CD8 + T cells. 24 We henceforth will refer to the CD27 − CD28 − T-cell population as senescent T cells. Of note, murine CD4 + and CD8 + T-cell populations do not lose the expression of CD28 as they undergo repeated rounds of proliferation indicating that there are key differences in differentiation-related phenotypic changes between humans and mice. 25 It is recognized widely that while the proportion of naïve T cells decreases during aging, the proportion of senescent cells increases, due in part to activation by persistent viruses such as cytomegalovirus (CMV) in vivo. [26][27][28]

| THE PROLIFER ATIVE P OTENTIAL OF T CELL S
The loss of thymic activity in early life results in a considerable reduction in the production of new naïve T cells. 29 Thus, in maturity, the T cell repertoire is fairly rigid as there is decreased production of T cells with novel specificities. 30 This may explain why older subjects become very susceptible to pathogens that have not been encountered previously. However, this also raises a conceptual problem, that is, how are T cell numbers maintained throughout life without any new input of thymically derived naïve cells? The repeated rechallenge by antigen can maintain the high precursor frequency of antigen-specific T cells in vivo. 31 In addition, the homeostatic proliferation driven by cytokines such as IL-7 (for CD4 + T cells) and IL-15 (for CD8 + T cells) can expand T cell populations irrespective of their antigen specificity. [32][33][34][35] However, the continuous proliferation of T cells through life may lead to loss of proliferative potential called replicative senescence that is due to telomere erosion 36 raising the question of whether this could lead to a loss of immune memory to frequently encountered antigens. 37,38 However, the loss of telomeric DNA can be mitigated if cells are able to upregulate the enzyme telomerase, which can replenish telomeres. 39 Nevertheless, T cells lose the capacity to upregulate this enzyme after repeated rounds of proliferation. 25 This raises the question of F I G U R E 1 Schematic model for T cell differentiation during aging. Upon activation, naive T cells differentiate into several stages of memory and effector cells. The proliferative potential, telomere length, and telomerase activity are reduced upon differentiation across the lifetime. As age increase, the number of terminally differentiated cell TEMRA accumulates compared with other subsets. T cells lose the expression of costimulatory receptors CD27, CD28, and CD45RA as they differentiate from naïve to terminally differentiated cell. However, they regain expression of CD45RA when they reach an end-stage and these (senescent) cells have limit replicative capacity but are highly cytotoxic and acquire NK-related function.
whether the loss of telomerase activity and the reduced capacity to divide after activation is a fixed feature of end-stage differentiation (senescence) of T cells or if this process is actively controlled by cell-signaling pathways and may be reversed, with implications for enhancing immunity during aging.
A novel mechanism for telomere maintenance that is independent of telomerase was described recently when antigen-presenting cells donate telomeres to T cells, enabling them to maintain their replicative capacity and responsiveness to antigen stimulation. 40 This telomere elongation mechanism occurred in humans and mice and involved the transfer of the telomeres together with DNA recombinant factor Rad51 via extracellular vesicles from antigen-presenting cells, enabling the donated telomeres to recombine with the ends of T-cell chromosomes. This resulted in the elongation of recipient T cell telomeres by ~3000 base pairs. Interestingly, this transfer mainly occurred in naïve and central memory T cells and to a considerably reduced extent in effector T-cell populations. 40 This indicates the existence of different cell fates after activation, whereby telomerase activation and telomere transfer may preserve the proliferative potential of some cells, whereas others are destined to reach an end stage (senescence) and lose their ability to proliferate. 20,37 End-stage (senescent) T cells that have shorter telomeres in the steady-state accumulate in vivo in healthy individuals during aging 18 and in patients with genetic defects that induce an exuberant T-cell proliferative response after stimulation, for example, Xlinked lymphoproliferative syndrome. 41 In addition, many diseases are associated with the accumulation of end-stage T cells, and it is not clear if these cells are causative or arise as a consequence of the disease state (Table 1). T cells that are specific for certain persistent infections such as that induced by cytomegalovirus, exhibit disproportionate telomere erosion compared with T cells that are specific for other persistent viruses (Epstein-Barr virus, varicellazoster virus, herpes simplex virus). 42 Thus, after a single episode of immune stimulation, a dichotomy of T cell fates may be induced with a proportion of cells retain replicative potential, whereas others differentiate toward senescence and lose their replicative capacity but are able to persist in vivo. Both types of T cells may be important for immune protection, and they coexist in healthy individuals and patients with various diseases.
In a recent study, T cell proliferation was determined in mice that were primed and boosted twice in vivo, then these cells were isolated from spleens and transferred to new mice; this sequential process of priming and boosting was repeated 51 times over 10 years. 43 Despite the tremendous expansion that the antigen-specific T cells occurred as a result of this repeated challenge, they retained their capacity to respond to the original antigen. 43 Therefore, previous models of the constrained replicative life span of T cells after repeated challenge considerably underestimated their true capacity for expansion. [44][45][46] T cells required periods of rest (~30 days) between restimulations to maintain this replicative capacity, and shorter periods of time between rechallenge (7 days) led to the loss of replicative potential. In these experiments, the antigen-specific T cells maintained their telomere length despite the substantial expansion in vivo. 43 Mice have 10-fold longer chromosomal telomeres than humans. 47 This indicates that the impact of telomere erosion on the expansion of T cells over a mouse life span will be less than that observed in humans. 47 The maintenance of long-term proliferative capacity of T cells 43 may therefore be due to long initial telomeres, the novel telomere transfer mechanism in vivo 40 and possibly continued activation of telomerase after activation (not assessed in the study of Soerens et al. 43 ) or a combination of all three. While the T-cell transfer experiments showed the impressive ability of these cells to expand beyond the life span of an individual mouse, the generation of end-stage like T cells after each round of stimulation was not assessed. 43 As the T cells were repeatedly transferred to young mice, the impact of an aged tissue environment on T cell fate after activation was not determined. Therefore, the maintenance of replicative capacity of T cells in these experiments does not reflect T cell expansion in vivo during aging. In humans and mice, end-stage effector cells are generated during an immune response, and the majority of these cells are cleared upon immune resolution. 48 Nevertheless, T cells with phenotypic and end-stage characteristics (discussed below) accumulate in both humans and mice during aging. 9,49 A precise molecular identification of end-stage human T cells and their functional attributes has not been made and the consequences of the accumulation of these cells during healthy aging and in a range of diseases is not known.

| HOW IS THE FUN C TI ON OF S ENE SCENT T CELL S REG UL ATED?
Senescent T cells within both CD4 and CD8 compartments exhibit low proliferative activity and telomerase induction after T-cell receptor activation. 18,22 However, these cells are highly secretory, expressing high constitutive levels of inflammatory cytokines including IFN-ɣ and TNF-and also the cytotoxic proteins perforin and granzyme. 34,50,51 This is reminiscent of the senescence-associated secretory phenotype (SASP) that is exhibited by nonsenescent lymphoid cells although with shared but also distinct secreted mediators. 7 In healthy humans, these senescent cells coexist with naïve and central memory T-cell populations that retain their proliferative potential. 38,43 It is unclear if the reduced proliferative capacity and telomerase activity of senescent cells were a permanent or a reversible functional feature of human T cells. The proliferative activity of senescent CD8 + T cells could be enhanced after T-cell stimulation either by blocking the PD-1 exhaustion-related inhibitory receptor signaling pathway, using antibodies against the ligand (PDL-1/PDL-2) that is found on antigen-presenting cells or by blocking p38 MAP kinase signaling using a small-molecule inhibitor. 22,52 However, only p38 MAP kinase but not PD-1 inhibition could reconstitute telomerase activity. 52 This indicated that key functional changes in senescent T cells were mediated by active cell-signaling processes rather than a passive loss of function and this loss of activity was reversible.
Senescent CD4 + and CD8 + T cells showed constitutive expression of p38 MAP kinase and this was induced by a novel mechanism involving AMP kinase and TAB1. 54 Furthermore, this non-canonical pathway was activated by either DNA damage-induced senescence or glucose deprivation. This showed a convergence of senescence signaling and nutrient sensing pathways in inducing the inhibition of the function of senescent T cells. 54

| THE IDENTIFIC ATI ON OF A NOVEL MULTI -PROTEIN INHIB ITORY COMPLE X IN S ENE SCENT CD 4 + T CELL S
In addition to p38, there are two other subgroups of MAPKs namely Erk and Jnk. 55,56 Collectively, these signal-transducing enzymes are involved in a wide range of mammalian physiology, including senescence, aging, and metabolism. 57 It was considered that each MAPK was regulated independently in different cell types. [58][59][60] Since p38 MAPkinase could regulate the function of senescent T cells, it raised the question of whether Erk and Jnk also had a role in regulating the function of this population. It was therefore of considerable interest that the p38, Erk, and Jnk MAPK subgroups were colocalized within an individual inhibitory signaling complex in senescent CD4 + T cell-containing AMPK, that also contained the stress proteins known as sestrins. 56 Sestrins are the mammalian products of the Sesn1, Sesn2, and Sesn3 genes. 61 Sestrin expression is induced upstream of AMPK activation in senescent CD4 + T cells. 56 This novel sestrin-activated MAPkinase activation complex (called an sMAC) was identical in senescent humans and in old (20 months) murine CD4 + T cells. 56 Furthermore, this complex operated independently of mTOR activity. 56 The expression of sestrins was increased in both older humans and mice and the blocking of this complex enhanced antigen-specific proliferation of senescent CD4 + T cells in humans in vitro and in CD4 + T cells mice in vivo. 54,56 Although this sestrin complex coordinates the simultaneous activation of all 3 MAP kinases in senescent T cells, each MAPkinase, once activated, controlled a distinct functional response. However, the disruption of global sMAC signaling restored antigen-specific proliferation and cytokine production in CD4 + T cells from old humans and enhanced responsiveness to influenza vaccination in old mice. 56 This raises the question of whether the sestrins also regulated the function of senescent CD8 + T cells.

| THE REG UL ATI ON OF S ENE SCENT CD8 + T CELL FUN C TI ON BY S E S TRIN S
Single-cell RNAseq analyses of human CD8 + T cells confirm that loss of costimulatory molecules (CD27 and CD28) and acquisition of KLRG1 and CD57 receptors are found in the same cells that express senescence characteristics. 24 Furthermore, the senescent CD8 population also expresses high levels of sestrins compared with the nonsenescent population. 24 Thus, senescent T cells in both CD4 and CD8 compartments can be identified by the same phenotypic characteristics. A significant increase in senescent-like CD8 + compared with CD4 + T cells has consistently been observed in healthy old subjects. 53,62,63 Single-cell analyses also showed that senescent CD8 + T cells express a wide range of activating and inhibitory NK receptors and their adaptor molecules, 24 confirming previous reports. 64 Human senescent CD8 + T cells can kill NK target cells in vitro and whole CD8 + T cells from old mice, that also express NK receptors can kill NK target cells in vivo. 24 Natural killer-like senescent CD8 + T cells also lose TCR signaling-related molecules and reduced TCR signaling-induced proliferative activity. A key observation was that the transition from TCR to NKR-related function is reversible by blocking the sestrins. 24 Therefore, senescent CD8 + T cells do not lose functionality, instead, they have altered functional capabilities that may enable them to provide broad rather than antigen-specific immune protection, for example, against tumors or infected cells. In this context, the accumulation of these cells in older individuals, perhaps TA B L E 1 List of putative beneficial roles of senescent T cell and their senescence-associated marker expression.

Identification of senescent cells within T cell compartment Mechanism and impact on protective immunity Species References
Tumor immunity (disease) Advanced melanoma Cytotoxic activity mediated by CD4 + T cells against human melanoma cell line Human [86,169,170] Bladder cancer CD4 + CCR7 − CD45RA + Cytotoxic activity mediated by EMRA CD4 + T cells Human

| INTER AC TI ON B E T WEEN S ENE SCENT S TROMAL VER SUS IMMUNE CELL S
Senescent nonlymphoid cells in tissues can be recognized and eliminated by the immune system. 65 by senescent cells from recognition by the immune system as previously described in cancer and in virally infected cells in vivo. [79][80][81] Aging is associated with the accumulation of senescent structural cells within tissues that secrete a wide range of pro-inflammatory mediators. Therefore, immune responses that take place in the tissue environments of older individuals will involve the interaction between old (senescent) leukocytes and old tissue cells that will be considerably different from immune responses in tissues of younger There is increased expression of the inhibitory NK-ligand HLA-E in the skin during aging and this has been suggested to be one reason that senescent cells accumulate in the tissues of older individuals. 6

| IS THE ACCUMUL ATI ON OF S ENE SCENT T CELL S DURING AG ING OF B ENEFIT TO THE HOS T ?
Both NK cells and CD8 + T cells have been shown to kill autologous senescent fibroblasts in vitro. 6 As the accumulation of senescent tissue cells has a detrimental impact on organ function and removal of senescent cells in vivo reverses age-associated pathology, the accumulation of senescent T cells during aging may be beneficial as they can eliminate senescent nonlymphoid tissue cells. Therefore, in addition to the broad nonantigen-specific protection against malignancy and infection that senescent CD8 + T cells may provide, they may also reduce organ-specific dysfunction in older individuals by clearing senescent cells from tissues 84,85 ( Figure 2).
Senescent-like CD8 + T cells can kill tumor targets in vivo. 24 A literature search has identified a wide range of diseases where senescent T cells accumulate (Table 1). It has been suggested that senescent CD8 + T cells play an essential role in the immunosurveillance of malignant cells. 24 Cytotoxic CD8 + T cells are the majority of tumor-infiltrating lymphocytes (TILs) 84 and these cells express a CCR7 − CD45RA + (or TEMRA) phenotype in nonsmall-cell lung cancer ( shown in murine models, where CD4 + -CTL populations were associated with protection against lethal influenza virus challenge. 98 In addition, they contribute to B-cell maturation and antibody production and perforin-mediated cytolytic activity, which was associated with antiviral protective immunity. 98 The expansion of CMV-specific CD4 T cells with an effector memory phenotype and increased inflammatory capacity as well as GrzB have been associated with better clinical outcome for patients. 99 An interesting observation is that fol-

| A PUTATIVE ROLE FOR S ENE SCENT T CELL S IN INDUCING PATHOLOGY
There is a wide range of diseases where senescent T cells accumulate (Table 2). These cells may directly or indirectly contribute to the disease process. The question is whether the expanded CMV-specific effector CD8 + T cells that are found in older subjects 28  Leishmania. 112 CD8 + T cell activity increases as the infection progresses. [113][114][115] In addition, the severity of the disease and the number of CD8 + T cells present in the lesions are linked significantly, 113,116 and this is independent of the parasite burden within the lesions. 117 This raises the question about the cause of the lesions in the skin, suggesting the possibility that chronic inflammation and nonspecific cytotoxic responses may lead to nonspecific tissue destruction.

| The accumulation of senescent CD8 + T cells in patients with cutaneous leishmaniasis
Individuals infected with L. braziliensis have elevated numbers of circulating CD4 + and CD8 + T cells that have characteristics of TA B L E 2 List of detrimental roles of senescent T cell and their senescence-associated marker expression.  CD4 + T-cell clones also produced gamma interferon, tumor necrosis factor alpha (TNF-α) and TNF- C. O. Gomes, unpublished). Therefore, we propose for the first time that senescent CD4-CLT may be associated with CL immunopathology. The mechanism involved in the tissue damage within the cutaneous lesions of infected individuals comprises cytokine production and direct cytotoxicity.

Mechanisms and impact in pathology
The initial inflammatory response after parasite infection is important to trigger the leishmanicidal mechanisms, 125 but it also can induce the expression of stress ligands by the surrounding stroma, including those that bind to NK receptors. [126][127][128] The induction of stress ligands may lead to nonspecific NK-related cytotoxic killing and tissue damage that exacerbates inflammation further in a positive feedback loop, promoting a progressive increase in the size of the skin lesions. 129,130 At this advanced stage, infiltrating senescent CD8 + T cells do not have to be specific for L. braziliensis but may be able to recognize other antigens (e.g., CMV) that can drive T-cell senescence and, therefore, the acquisition of NK characteristics. The mechanism that induces the tissue pathology in CL is unclear, however, the wide expression of NK receptors on both senescent CD4 + and CD8 + T cells in the lesions may be involved and it is essential to determine the presence of different NK ligands for senescent T cells within the tissue that may drive nonspecific pathology. Moreover, identifying the source of inflammation is also pivotal in addressing immuno-based therapies to improve the patients healing.

| The possible role of senescent T cells in different human diseases
The persistent antigenic availability found in autoimmune diseases, organ transplants, and other diseases fosters the continuous acti- Alzheimer's, 146 and cardiovascular diseases. 147,148 Furthermore, a direct immunopathological role of senescent cells has also been reported in patients with acute hepatitis A infection and this is also associated with the co-localization of CD8-NKR-expressing cells and NKG2D-ligand-expressing hepatocytes. However, the NKRexpressing CD8 + T cells required a preactivation step with IL-15, which was produced by HAV-infected cells, to induce their killing capacity and liver injury. 149 Therefore, two signals may be required

| HOW C AN S ENE SCENT S TROMAL AND S ENE SCENT T CELL S B E TARG E TED?
The The establishment of pharmacotherapeutic approaches that abolish senescent cells selectively introduced using "senolytic" drugs.
The discovery of senolytic drugs has a long history and involved the testing of many strategies to target senescent cells in many different tissues. 158,159 Successful drug candidates targeted antiapoptotic pathways, specifically in senescent cells. The activity was demonstrated in mice when a short-term administrate combination of dasatinib and quercetin reduced a significant number of senescent cells and improved survival rate, therefore extending old mice life span. 154 Long-term treatment up to 23 months of a combination between dasatinib (D) and quercetin (Q) has shown therapeutic effect in reducing intervertebral disc degeneration by alleviating senescent cell burden in elderly wild-type C57BL/6 mice. 160 Interestingly, mice infected with Leishmania major and treated with quercetin orally for 28 consecutive days had increased lesional wound-healing potential compared with control mice. Moreover, they reduced the systemic levels of tumor necrosis factorα, interleukin-6, and adiponectin, as well as superoxide dismutase and glutathione peroxidase activities, suggesting that target senescence cells may be contribute toward reducing Leishmania pathology. Once again, the impact of senolytic drugs on senescent T cells was not determined in these experiments, but the results that emerged suggest the participation of the population in the aggravation of many diseases. Another example of using senolytic drugs was in a β-cell senescent-induced metabolic disease type 2 diabetes model, using ABT263 (Navitoclax) that reduced SASP release from senescent β-cell and improved glucose tolerance in mice. 161 Together, these data signify therapeutic potential for health benefits during aging by using senolytic drugs with welltolerated toxicity at least in preclinical models.
A caveat with the use of senolytic drugs is that it is currently unclear what impact they will have on the persistence and function of senescent T cells. If these cells have a beneficial role then removing them may have detrimental consequences (Figure 2A). However, if these cells contribute to immunopathology in certain diseases, then their removal, even temporarily, may allow the destructive inflammatory and tissue damage loop to be broken ( Figure 2B). With senolytic therapy in general, it is not clear how long it will take before senescent cells of different types reappear after their initial removal.
There are more than 300 clinical trials worldwide for senolytic drugs, especially D + Q, navitoclax, and fisetin (clini caltr ials.gov). 159 These studies are mostly focusing on age-related disorders in older subjects aged over 65 years, such as Alzheimer's disease, inflammationdriven disease, skeletal, and joint disease with the aim to improve quality of life in patients and to reduce frailty.
The first trial in humans was performed in idiopathic pulmonary fibrosis patients, by intermittent dosing of D + Q for 3 weeks. 162 The result showed improvement in physical and pulmonary function. The SASP measurement for IL6, MMP7, and TIMP2 after treatment exhibits a slightly reduce but is not statistically significant. However, the level was correlated with the patient physical activity. This might be due to short-term administration and a low number of preexisting senescent cells. 162 Interestingly, out of numerous clinical trials that were registered in clini caltr ials.gov, only a few studies have mentioned measurements of immune senescent cells (as defined by p16 expression).
Reducing the number of senescent T cells to an optimal level that would give benefit over harm is the most challenging aspect of this therapy. Excessive elimination of the senescent T-cell burden might disturb tissue homeostasis. The effect of senolytic therapy on circulating senescent T cells is unclear. The drug distributes to blood and any tissues. 162 Depletion of the cytotoxic population of terminally differentiated cells will possibly interfere with immune responses in older adults, where these populations accumulate. 163 In addition, interfering with cross talk between senescent stromal and immune cells by senolytic therapy may have an impact on the immune-mediated clearance of these cells in tissues. Of interest, it was shown that the use of the senolytic drug Quercetin in patients with Leishmania reduced the immunopathology in the skin but it was not clear if senescent cells in both lymphoid and nonlymphoid compartments were affected by the treatment. 164 It was shown recently that inhibiting baseline inflammation in the skin of healthy older humans using an oral p38 MAP kinase inhibitor (Losmapimod from GSK) enhanced the ability of these individuals to respond to an antigen-specific immune challenge in the skin. 165 The high level of inflammation was due in part to the secretion of chemokines by senescent fibroblasts, which recruited inflammatory monocytes that then inhibited resident memory T cells in the tissue. 166 The short-term p38 inhibition was shown to reduce the SASP secreted by fibroblasts but did not change the number of senescent cells that were present. 166 Therefore, the impact that senescent stromal cells have on altered immunity in tissues can be reduced by blocking the mediators they secrete without actually removing the cells that are responsible.
Recently, the possibility of using chimeric antigen receptor (CAR)-T-cell therapy for targeting senescent cells has raised interest. 168 CAR-T is approved to be used in the clinical setting for various immunogenic cancers, for example, multiple myeloma, large B-cell lymphoma, mantle cell lymphoma, etc. Utilizing its selectivity and efficacy, CAR-T cells could possibly be exploited to kill senescent cells (or senescent immune cells) more potently with fewer off-target effects than senolytic drugs. By using urokinase-type plasminogen activator receptor (uPAR) as an antigen, uPAR CAR-T cells efficiently kill senescent cells in mice with distinctive condition including chemotherapy-induced senescence of adenocarcinomabearing mice, and chemically induced liver fibrosis mice. 168 The data suggest a potential to use CAR-T in many other age-related diseases.
Identifying new surface antigens that are specific to stromal or immune cells requires further study. 168 Finally, if senescent T cells in inflamed tissue environments can cause non-specific immunopathology ( Figure 2B) then reducing their NK-like activity may alleviate this process. In this situation, the inhibition of sestrins in senescent CD8 + T cells can switch their function from NK-like to T-cell-like activity. 24 Sestrin inhibitors may therefore be useful to reduce immunopathology in various diseases.
However, inhibiting the sestrins reconstitutes the proliferative activity of senescent T cells, which harbor DNA damage and this may be associated with some risk of malignancy. Nevertheless, the temporary blockade of sestrins may break the positive feedback loop that leads to immunopathology, and this would be of benefit.

| CON CLUDING REMARK S
We live in a world where older individuals are increasing in numbers disproportionately to younger subjects. The increase in life span is not commensurate with an increase in health span. Senescent cells both in tissues and in the immune system accumulate during aging and may have a role in decreased health. Therefore, a better understanding of the biology that underpins their interactive function is essential to determine how the immune system and tissue environments can be targeted individually or together for the benefit of health during aging and in patients with various inflammatory diseases.

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors have no conflicts of interest to disclose.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sharing not applicable-no new data generated.