The role of the immunosuppressive PD-1/PD-L1 checkpoint pathway in the aging process and age-related diseases

There are certain traits typically encountered in the aging process which are associated with functional disturbances of the immune system. For instance, a chronic low-grade inflammation, commonly called inflammaging, is a challenge for the maintenance of homeostasis in aged tissues [1]. The low-grade inflammatory state with aging has been linked with an accumulation of pro-inflammatory senescent cells into tissues [2]. The senescence-associated secretory phenotype (SASP) is not only associated with aging but also cells of the tissues with several aged-related diseases display the specific markers of the SASP state [3]. An acute or chronic inflammatory process without a counteracting anti-inflammatory response leads to destructive effects in affected tissues. Thus, there exist diverse anti-inflammatory/immunosuppressive mechanisms which are not only able to suppress inflammatory responses but can also enhance the resolution of the inflammatory state and trigger tissue repair processes. The NF-κB signaling pathway is a major inducer of pro-inflammatory responses [4] and its inhibition, e.g., via endotoxin tolerance [5] or an increased level of the heat-shock proteins HSP70 and HSP90 [6, 7], can suppress inflammatory responses and accordingly enhance immunosuppression. On the other hand, the immune system possesses an armoury of immunosuppressive cells, e.g., myeloid-derived suppressor cells (MDSC), regulatory T cells (Treg), and M2 macrophages, which secrete many immunosuppressive cytokines, such as IL-6, IL-10, IFN-γ, and TGF-β [8–10]. These immunosuppressive mediators act to impair the functions of many immune effector cells and thus promote immunosuppression which subsequently induces the immunosenescence state, i.e., a decline in the activity of the immune system is evident not only in cancer and chronic inflammatory diseases but also in the aging process [11–13].

Moreover, there are many inhibitory immune checkpoint proteins which provide a crucial mechanism for the maintenance of self-tolerance and the regulation of immune responses, e.g., in chronic inflammatory conditions. For example, the programmed cell death protein-1 (PD-1) and programmed death-ligand 1 (PD-L1) establish an important inhibitory checkpoint axis which has a crucial role in the suppression of T cell activation and thus it can alleviate many immune responses [14] (Fig. 1). In fact, the PD-1/PD-L1 pathway represents a co-inhibitory mechanism which inhibits the activation of the T cell receptor (TCR), thus suppressing antigen presentation via the major histocompatibility complex class II (MHC-II) proteins. Mizuno et al. [15] demonstrated that the binding of the PD-L1 protein to the PD-1 receptor on T cells primarily suppressed the TCR signaling pathway and thus it inhibited cytokine production by mouse CD4⁺ and CD8⁺ T cells. Currently, the inhibitory mechanisms still need to be clarified. Interestingly, the presence of immunosenescence with aging and during age-related diseases is associated with a significant decline in the numbers and functions of CD4⁺ and CD8⁺ T cells as well as in the activity of B, NK, and dendritic cells and tissue macrophages [13, 16, 17]. These age-related changes indicate that the function of the PD-1/PD-L1 signaling pathway might be enhanced during aging and in age-related diseases. In fact, it is known that the expression of the PD-L1 protein is robustly increased in aged tissues and senescent cells [18–20]. This may represent an important mechanism of underpinning both immunosenescence and many age-related immune deficiencies. Here, I will briefly elucidate the properties of the PD-1/PD-L1 checkpoint pathway and then examine in detail the age-related mechanisms promoting the activation of the PD-L1 checkpoint and its role in the aging process and age-related diseases.

The inhibitory PD-1/PD-L1 checkpoint pathway confers a significant immunosuppressive protection in autoimmune diseases and many chronic inflammatory states although in cancers and fibrotic diseases it is associated with major detrimental effects [14, 21, 22]. The PD-1 receptor (CD279) is expressed on the surface of the activated T and B cells as well as on macrophages, monocytes, and innate lymphoid cells (ILC) including natural killer (NK) cells [14, 23–25] (Fig. 1). The PD-1 protein is a membrane-bound glycoprotein containing an extracellular IgV domain, a transmembrane sequence, and a cytoplasmic tail which contains a tyrosine-based inhibitory motif (ITIM) and a tyrosine-based switch motif (ITSM). The binding of the PD-L1 protein to the PD-1 receptor induces the phosphorylation of these two inhibitory motifs providing docking sites for the Src-homology-2 (SH2) protein phosphatases. Subsequently, the SHP-1 and SHP-2 phosphatases dephosphorylate the signaling components of the TCR and the costimulatory CD28 pathway, thus suppressing the activation, proliferation, and cytokine production of T cells [23, 26–28]. Stanford et al. [26] have reviewed the function of diverse protein tyrosine phosphatases (PTP) which are involved in the activation and inhibition of the TCR pathway thus regulating the immune homeastasis of T cells. The cytosolic SHP-1, SHP-2, and SHIP1 are major protein phosphatases suppressing TCR signaling associated with the inhibitory function of the PD-1/PD-L1 axis. It is known that PD-1/SHP-1 signaling was able to attenuate the cytotoxic activity of mouse CD8⁺ T cells [29]. Given that NK cells also display PD-1 signaling [25], it seems plausible that cancer cells with their extensive expression of PD-L1 proteins can exploit the PD-1/PD-L1 pathway to suppress the anticancer armoury of the CD4⁺ and CD8⁺ T cells and NK cells, thus escaping the surveillance by the immune system. For instance, Niu et al. [30] demonstrated that the expression of PD-1 protein was significantly increased in the NK cells of human lung cancer. These NK cells exhibited a decline in antitumor activity as compared to the PD-1-negative NK cells. Moreover, it is known that the PD-1 receptor robustly inhibited the activation of human B cells [31]. In 2001, Nishimura et al. [32] demonstrated that the PD-1 receptor-deficient mice developed dilated cardiomyopathy, an autoimmune disorder. Subsequent studies have revealed that many human immunodeficiency diseases are associated with single nucleotide polymorphism (SNP) on human PD-1 gene [33, 34] although the role of the CD-1 receptor in autoimmune diseases still needs to be clarified.

The PD-L1 (CD274 or B7-H1) protein is a type 1 integral membrane glycoprotein and it is expressed in several immune cells, especially in antigen presenting and immunosuppressive cells, as well as in many stromal cells, such as fibroblasts, epithelial and endothelial cells, pericytes, and smooth muscle cells [14, Human Protein Atlas]. There is also an extensive expression of PD-L1 protein in diverse cancer cells and in cancer-associated fibroblasts (CAF), i.e., the PD-1/PD-L1 axis is a major player in the maintenance of the immunosuppressive microenvironment in tumors [35, 36]. Dong et al. [37] cloned the human B7-H1/CD274 gene (PD-L1) from the human placenta cDNA library. They revealed that the B7-H1/PD-L1 protein did not bind to the CD28 protein, a crucial co-stimulator of T cells, but IL-2 interaction was required to induce T cell proliferation and secretion of immunosuppressive IL-10 cytokine. There also exists the PD-L2 protein (CD273 or B7-DC), mostly expressed in dendritic cells, but the regulation of the PD-1/PD-L2 axis has been less extensively studied than that of the PD-1/PD-L1 counterpart [38]. As discussed above, the PD-L1 protein acts as a ligand which activates PD-1 signaling in T cells and consequently leads to their exhaustion and even apoptosis (Fig. 1). In human NK cells, there are also several other checkpoint proteins, e.g., the HLA-E:CD94-NKG2A complex, which inhibit the immune surveillance by NK cells [39]. It is known that many inflammatory mediators, such as IFNγ, IL-6, IL-10, and TGF-β, are potent inducers of the expression of the PD-L1 protein, e.g., via epigenetic regulation [40, 41]. In inflammatory conditions, the PD-L1-positive cells can activate PD-1 signaling not only through the contact-dependent manner but also by secreting PD-L1-containing extracellular vesicles (EV) which subsequently activate PD-1 signaling in T cells [42]. This EV-mediated signaling is an especially important inducer of immunosuppression in tumors and chronic inflammatory diseases. The regulation of PD-L1 expression will be reviewed more thoroughly in the following sections with respect to cellular senescence and the aging process.

Interestingly, emerging studies have revealed that the PD-L1 protein can also trigger intrinsic effects not only in cancer cells but also in other PD-L1-containing cells, such as in fibroblasts [43–45]. The cell-intrinsic (either cytoplasmic or nuclear) signaling of the PD-L1 protein can regulate many cellular activities, e.g., cellular differentiation, autophagic flux, energy metabolism, and inflammatory responses [45, 46]. For instance, Guo et al. [44] demonstrated that the PD-L1 protein interacted with the Smad3 transcription factor and via β-catenin signaling induced the differentiation of human pulmonary fibroblasts into fibrogenic myofibroblasts. These investigators also reported that the expression of PD-L1 was clearly increased in the lungs of patients with idiopathic pulmonary fibrosis (IPF). It is known that the PD-1/PD-L1 signaling pathway has a crucial profibrotic role in the development of IPF disease [22, 47], although it is not known whether it is caused by intrinsic or extrinsic regulation. Interestingly, Gao et al. [48] demonstrated that the acetylation of the PD-L1 protein at Lys263 within its cytoplasmic tail regulated the translocation of the PD-L1 protein from the cell-surface into the nucleus. The acetylation of the PD-L1 protein by the p300 acetyltransferase blocked its translocation into the nucleus, whereas the deacetylation by histone deacetylase 2 (HDAC2) enhanced its nuclear translocation. The relocation of the PD-L1 protein into the nucleus was driven by clathrin-dependent endocytosis. Gao et al. [48] also revealed results indicating that the nuclear PD-L1 might trigger the expression of those genes which are involved in the immune-response pathways and thus could facilitate the immune evasion of tumor cells. They also reported that the inhibition of the PD-L1 nuclear translocation enhanced the efficacy of the PD-1 blockade therapy.

The PD-1/PD-L1 signaling pathway is also able to enhance the differentiation of immune cells towards the immunosuppressive state (Fig. 1). For instance, there is robust evidence that an increase in PD-1 signaling can promote the differentiation of CD4⁺ T cells into immunosuppressive Treg cells in diverse experimental models [49, 50]. It seems that cancer chemotherapy with the anti-PD-L1 antibody can alleviate both the inhibition of T cells and suppress the differentiation of Tregs. Interestingly, the expression level of PD-1 receptor controls the polarization of macrophages, i.e., a deficiency of PD-1 expression promoted the development of the pro-inflammatory M1 phenotype, whereas its overexpression enhanced the appearance of the immunosuppressive M2 phenotype, especially in tumor-associated macrophages (TAM) [51, 52]. Wei et al. [52] reported that the M2 polarization was induced via the activation of the ERK/AKT/mTOR signaling pathway in human macrophages. TAMs have an important role in the maintenance of an immunosuppressive state in the tumor microenvironment. In addition, it is known that PD-1 signaling enhanced the proliferation and immunosuppressive activity of MDSC, especially in the microenvironment of tumors [53] and sepsis [54]. It seems that the PD-L1-induced activation of PD-1 enhances immunosuppression either by suppressing effector immune cells, such as T, B, and NK cells, or by stimulating the differentiation of immunosuppressive Tregs, MDSCs, and M2 macrophages (Fig. 1). On the other hand, immunosuppressive cells secrete many cytokines, such as IL-6, IL-10 and TGF-β [55, 56], which are potent inducers of the expression of PD-L1, i.e., the PD-1/PD-L1 checkpoint axis acts in cooperation with immunosuppressive cells (Fig. 2).

The age-related diseases are progressive disorders which commonly involve both inflammatory changes and fibrotic degeneration. Typically, these diseases also are associated with the accumulation of senescent cells with inflammatory and degenerative changes, such as in atherosclerosis, chronic obstructive lung disease (COPD), coronary artery disease (CAD), and Alzheimer’s disease (AD) [70–73] as well as in fibrotic diseases, e.g., in idiopathic pulmonary disease (IPF), cardiac fibrosis, and systemic sclerosis [70, 74, 75]. There exists a significant crosstalk between senescent cells and inflammatory cells in the pathogenesis of many age-related diseases, e.g., via the control of the expression of inhibitory checkpoint proteins. For instance, the PD-1/PD-L1 axis has an important role in the pathogenesis of human AD and COPD [76–78]. Kummer et al. [79] demonstrated that the expression of PD-L1 was robustly increased in the brain of transgenic AD mice with clear positive staining in the astrocytes surrounding amyloid-β plaques. Moreover, a deficiency of microglial PD-1 protein increased the inflammatory response and the deposition of amyloid-β plaques within the brain. Given that inflammatory mediators are potent inducers of the expression of PD-L1 protein, it seems that in many age-related diseases, the activation of PD-1/PD-L1 signaling is related to defence against chronic inflammation. The pro-inflammatory SASP state of senescent stromal cells maintains the inflammatory microenvironment and aggravates the pathology of age-related diseases. It seems plausible that the induction of PD-L1 expression in senescent cells has a crucial role in the pathogenesis of chronic age-related diseases since it promotes immunosenescence and prevents the clearance of senescent cells.

Fibrosis in cardiac muscle, kidney, liver, and lung as well as in cases of systemic sclerosis are common age-related fibrotic diseases. The differentiation of tissue fibroblasts into fibrogenic myofibroblasts and an increase in the numbers of senescent fibroblasts are the major hallmarks of fibrotic lesions [75, 80]. Emerging studies have revealed that the function of the PD-1/PD-L1 axis has a crucial role in the accumulation of fibrotic lesions [22, 47]. There are a number of investigations indicating that in mouse and human fibrotic tissues, there exists a robust increase in the expression of PD-L1 protein not only in fibroblasts but also in endothelial and epithelial cells [18, 22, 44, 47, 81]. Onorati et al. [18] utilized a single-cell transcriptome technique to determine in human interstitial lung fibrosis that the high expression of PD-L1 protein was localized into senescent cells which also displayed a proinflammatory phenotype. There are studies indicating that the PD-L1 protein might promote fibrosis by enhancing the differentiation of tissue fibroblasts into fibrogenic myofibroblasts. For example, Guo et al. [44] demonstrated that the PD-L1 protein was required for the TGF-β-induced differentiation of myofibroblasts both from human primary lung fibroblasts or fibroblasts isolated from the IPF patients. They revealed that TGF-β exposure induced the interaction between the PD-L1 protein and the SMAD3 transcription factor to induce the transcription of α-smooth muscle actin (α-SMA). It seems that the PD-L1 protein can enhance tissue fibrosis via some form of intrinsic regulation. There are several reports that immunosuppressive CAFs express a high level of PD-L1 protein [82, 83]. For instance, Li et al. [82] demonstrated that CAFs were able to promote the expression of PD-L1 protein in mouse cancer cells by secreting the CXCL5 chemokine. Given that the inhibitors of PD-1/PD-L1 signaling are effective in many cancers, the checkpoint inhibitors may also represent promising therapeutic choices for treating fibrotic diseases [84]. Currently, there are some animal studies in experimental disease models which have revealed that the PD-1/PD-L1 inhibitors are able to attenuate the severity of the fibrosis [85, 86].

Given that the PD-1/PD-L1 pathway has an important role in the generation of an immunosuppressive microenvironment in tumors, there has been a significant interest in examining the regulation mechanisms underpinning the expression of the PD-L1 protein [43, 87, 88]. It seems that there exists a complex network of regulatory processes from epigenetic to post-transcriptional mechanisms which not only control the expression level of PD-L1 but also the resistance to the PD-1/PD-L1 blocking therapies. Currently, the signaling mechanisms which regulate the age-related increase in the expression of PD-L1 still need to be clarified although many of them are known to be associated with the regulation of the aging process. Next, I will briefly examine some of those signaling pathways and emphasize that the upregulation of the PD-1/PD-L1 pathway seems to be involved in the regulation of the aging process (Fig. 2).

Currently, there is a debate if it will be possible to develop anti-aging therapies which might increase health span and even extend the lifespan of elderly humans. Recently, Guarente et al. [166] summarized the present clinical knowledge and ongoing trials on the eight most promising drug candidates for anti-aging medicines. Two of these drugs, i.e., metformin and rapamycin, are clinically utilized drugs which have been able to extend the lifespan of rodents [167, 168]. Metformin, a drug used in the clinical treatments of type 2 diabetes, has multiple targets in the body, e.g., it activates the AMP-activated kinase (AMPK) which controls many aging-associated signaling pathways [169]. Interestingly, metformin has displayed diverse therapeutic properties in many immune-mediated diseases [170] and also alleviated many of the hallmarks of cancers [171]. For instance, metformin exposure inhibited the function of immunosuppressive MDSCs in tumor-bearing mice [172]. This is probably induced by the activation of AMPK signaling which is known to inhibit the functions of MDSCs [173]. There is abundant evidence that metformin can reduce the expression of PD-L1 protein through diverse mechanisms, although mainly via AMPK signaling [174–176]. Cha et al. [174] demonstrated that metformin via AMPK activation phosphorylated the PD-L1 protein at the Ser195 site in cancer cells. The phosphorylation of the PD-L1 protein induced its glycosylation and consequently proteasomal degradation via the endoplasmic-reticulum-associated degradation (ERAD) pathway. Cha et al. [174] also reported that the metformin-induced downregulation of the amount of PD-L1 protein present in cancer cells enhanced their surveillance by cytotoxic T lymphocytes. They also verified that metformin exposure reduced the level of PD-L1 in human tumors. It seems that there are other mechanisms which metformin can exploit to inhibit the expression of PD-L1 protein. Lu et al. [175] revealed that metformin treatment inhibited the expression of PD-L1 by suppressing the IL-6/JAK2/STAT3 pathway in human esophageal squamous carcinoma cells. This inhibition was an AMPK-independent process. However, it seems likely that many metformin-induced responses are mediated by AMPK-induced autophagy, either by inhibiting mTOR or activating ULK1 signaling [177].

Rapamycin, an inhibitor of mTOR signaling, is a potent enhancer of autophagy. As discussed above in the section ‘mTOR-related signaling’, the signaling pathways via the activation of mTOR stimulate the expression of PD-L1 protein in many experimental conditions. Given that rapamycin has been able to extend the lifespan of rodents [167], the inhibitors of mTOR protein, such as rapalogs, have been claimed to be promising candidates for anti-aging therapeutics [178]. Khedri et al. [179] demonstrated that exposure of mice to rapamycin significantly reduced the expression of PD-L1 in the brain but not in the spleen. They proposed that rapamycin might have beneficial effects on CNS autoimmune diseases, such as those encountered in multiple sclerosis [180]. Onorati et al. [18] reported that rapamycin exposure was able to downregulate the expression of PD-L1 protein, especially on the cell surface of senescent human fibroblasts. Rapamycin treatment also reduced the expression of IL-6 in senescent cells, an effect associated with a decline in the SASP state. This is an interesting observation since a decrease in the expression of PD-L1 in senescent cells might enhance their immune surveillance and thus rapamycin could possess senolytic properties.

Traditional medicine and phytochemicals have long been exploited in health care for diverse diseases and even as anti-aging remedies. Many clinically used drugs, such as aspirin and metformin, have their roots in herbal medicine. Phytochemicals, mostly polyphenols, are multitarget molecules, like metformin, and they affect many age-related signaling pathways. In fact, it is known that several phytochemicals possess immunomodulatory properties [181]. For instance, there is clear evidence that phytochemicals can improve the efficacy of many cancer therapies based on PD-1/PD-L1 blockade [182, 183]. Currently, the molecular mechanisms need to be clarified before phytochemicals can be administered as checkpoint inhibitor therapies. Guo et al. [184] revealed in in silico experiments that several polyphenols, such as curcumin and resveratrol, were able directly to inhibit the interaction between the PD-1 and PD-L1 proteins. Accordingly, Jing et al. [185] reported that quercetin also inhibited the interaction between the PD-1/PD-L1 proteins and thus promoted the ability of T cells to destroy human cancer cells. Phytochemicals can also enhance the efficacy of immunotherapy with PD-1/PD-L1 inhibitors by downregulating the expression of PD-L1 in cancer cells. For instance, Liu et al. [186] demonstrated that curcumin exposure suppressed the expression of PD-L1 and PD-L2 proteins in human head and neck squamous carcinoma cells (HNSCC). They also reported that curcumin potentiated the cytotoxic activity of CD8⁺ T cells when administered as a combination treatment with PD-L1 antibody. Recently, Liu et al. [187] reviewed the molecular mechanisms describing how different phytochemicals alter the function of the PD-1/PD-L1 pathway in tumor immunotherapy.

There is convincing evidence that the expression of PD-L1 protein is significantly increased in many tissues with aging. This age-related response seems to be attributable to an accumulation of senescent cells which are known to exhibit a robust expression of the PD-L1 protein. This increased expression level of the PD-L1 protein indicates that senescent cells can evade immune clearance by inhibiting the activity of cytotoxic CD8⁺ T and NK cells. This might explain why senescent cells accumulate with aging and disturb the maintenance of tissue homeostasis. Senescent cells not only express the PD-L1 protein but they can also secrete EVs coated with PD-L1 proteins. It is known that the shedding of EVs is robustly increased from the senescent cells of diverse cell types, e.g., fibroblasts and endothelial cells [188, 189]. The secretion of EVs enhances the paracrine spreading of senescence into neighboring cells within tissues [190, 191]. Borghesan et al. [190] reported that the IFN-γ inducible transmembrane 3 protein (IFITM3) was able to increase the secretion of EVs and thus promote the paracrine senescence in aged tissues. Interestingly, Benayoun et al. [64] demonstrated that in mice, the gene transcription of the IFN-γ-driven genes was strongly enriched in different aging tissues. Moreover, Kang et al. [125] demonstrated that TGF-β exposure of human and mouse fibroblasts also induced the secretion of the PD-L1-coating EVs which inhibited the function of T cells. There is mounting evidence that the PD-L1-coated EVs, secreted either from cancer cells or CAFs, have a key role in the maintenance of immunosuppressive state in the microenvironment surrounding tumors [192, 193]. It is known that chronic inflammatory conditions also increase the secretion of immunosuppressive EVs [66]. It seems that an excessive expression of PD-L1 protein and their packaging into EVs in senescent cells not only induces an accumulation of senescent cells into aged tissues but also promotes the generation of immunosenescence with aging by inhibiting the functions of T and NK cells (Fig. 2).

The immunosenescence associated with aging involves a substantial decline in the functional properties of the immune system [13, 16, 194]. Furthermore, the immunosenescence present in tumors displays more or less similar alterations in immune cells as those encountered in the aging process [11]. Currently, it is known that the PD-1/PD-L1 checkpoint axis impairs the functions of T and B cells as well as NK cells and macrophages, inducing relatively similar changes as those observed in immunosenescence, i.e., an activation of the PD-1/PD-L1 pathway with aging might promote the aging of the immune system. However, it needs to be clarified whether other inhibitory immune checkpoints, such as CTLA-4, LAG-3, TIM-3, or VISTA, affect the age-related immune deficiency. In addition to the contact-mediated inhibition, immunosuppressive cells, such as MDSCs, Tregs, and M2 macrophages, secrete cytokines which evoke a decline in the functions of both adaptive and innate immunity [9, 13, 55, 58]. Interestingly, immunosuppressive cells are able to induce alterations in immune cells which are very reminiscent to those observed in the immunosenescent phenotype [13, 195]. In fact, it is known that the contact-dependent PD-1/PD-L1 checkpoints and immunosuppressive cells cooperate in pathological states since as described above, the activation of the PD-1 protein can enhance the differentiation of immunosuppressive cells, and vice versa, the cytokines secreted by immunosuppressive cells can stimulate the expression of the PD-1/PD-L1 proteins (Figs. 1 and 2).

Intriguingly, many crucial enhancers of the aging process are inducers of PD-L1 expression. Cellular senescence might represent the final outcome since many SASP components are important enhancers of PD-L1 expression, whereas an increased expression of PD-L1 inhibits the immune surveillance of senescent cells. Moreover, many inflammatory SASP factors, such as chemokines and complements, promote the recruitment of immunosuppressive cells into aged tissues, thus enhancing the development of an immunosuppressive microenvironment as well as a more systemic immunosenescence. It is likely that many inflammatory mediators strive to stimulate the PD/PD-L1 pathway in order to prevent excessive inflammatory injuries in tissues. Another interesting observation indicates that a decline in the autophagy stimulates the accumulation of PD-L1 proteins because the turnover of PD-L1 proteins is regulated by selective autophagy. In this situation, the signaling pathways activating mTOR signaling have a crucial role since they can also stimulate the expression of the immunosuppressive PD-L1 protein in non-senescent cells. Many investigators have claimed that mTOR signaling has a key role in the aging process. For example, activation of the insulin/IGF-1 pathway stimulates mTOR signaling and consequently can activate STAT3 signaling which promotes the expression of PD-L1 and enhances immunosuppression [196–198]. There is clear evidence that insulin/IGF-1 signaling can promote the aging process although it has critical functions in protein synthesis and energy metabolism [199]. Interestingly, current anti-aging therapies, i.e., metformin and rapamycin, are inhibitors of mTOR signaling and accordingly potent stimulators of autophagy. It seems that the decline in autophagy not only disturbs the cellular proteostasis and tissue homeostasis but via the activation of PD-L1 signaling, it can enhance the accumulation of senescent cells and thus promote the aging process.

The presence of a chronic low-grade inflammation with aging remodels the properties of the immune system, inducing a permanent decline in its functional activities, a state called immunosenescence. It is not only during the aging process where one encounters immunosenescence, it also exists in many chronic age-related inflammatory diseases and cancer microenvironment. It is known that a defect in immune surveillance has a key role in allowing tumor cells to evade immune clearance. In tumors, the activation of immunosuppressive cells and many inhibitory immune checkpoints, such as the PD-1/PD-L1 pathway, disturb the function of immune cells, especially that of the cytotoxic CD8⁺ T and NK cells. Interestingly, it was recently demonstrated that the expression of the immunosuppressive PD-L1 protein was robustly upregulated in senescent cells and that this phenomenon suppressed the cytotoxicity of CD8⁺ T cells [20]. This is an important observation since it indicates that senescent cells can be protected against immune surveillance and subsequent clearance, i.e., leading to their accumulation within aged tissues. The PD-1/PD-L1 axis has a crucial role in tumor growth but it seems that it also enhances the aging process. Intriguingly, the signaling pathways associated with the promotion of the aging process are crucial inducers of the expression of the immunosuppressive PD-L1 protein in senescent cells. This seems paradoxical but currently, the exact role of senescent cells is unknown although most likely they are enhancers of the aging process. Currently, there is intense research to develop senolytic drugs aiming to destroy senescent cells, so-called senolytic therapy [200]. The PD-1/PD-L1 checkpoint inhibitor therapy might represent a promising approach, given that it is being clinically exploited in cancer immunotherapy.

The author thanks Dr. Ewen MacDonald for checking the language of the manuscript.

This article is the sole work of the author.

Open access funding provided by University of Eastern Finland (including Kuopio University Hospital).