Role of Cellular Senescence in the Pathogenesis of Systemic Sclerosis

Abstract

Purpose of review:

Systemic sclerosis (SSc) is a chronic rheumatic disease that is characterized by immune activation, vasculopathy, and fibrosis of the skin and internal organs. It has been proposed that premature onset of aging pathways and associated senescent changes in cells contribute to the clinical and pathological features of SSc. The aim of this review is to critically review recent insights into the involvement of cellular senescence in SSc.

Recent findings:

Cellular senescence plays a critical role in SSc pathogenesis, particularly involving endothelial cells and fibroblasts. Immunosenescence could also contribute to SSc pathogenesis by direct alteration of cellular functions or indirect promotion of defective immune surveillance. Molecular studies reveal experimental evidence through which cellular senescence contributes to fibrosis. The few proof-of-concept trials using senotherapeutics showed promising results in various fibrotic diseases, including SSc.

Summary:

There is increasing evidence implicating cellular senescence in SSc. The mechanisms underlying premature cellular senescence in SSc, and its potential role in pathogenesis, merit further investigation. Emerging drugs targeting senescence-related pathways might be potential therapeutic options for SSc.

Introduction

Systemic sclerosis (SSc) is a systemic autoimmune disease of unknown etiology and incompletely understood pathogenesis. SSc is characterized by progressive fibrosis, vascular involvement, and inflammation in the skin and multiple internal organs. The disease shows substantial patient-to-patient heterogeneity, follows a chronic and often progressive course, and is associated with substantial disability and high mortality. There is currently no cure or disease-modifying therapy for SSc, and management is suboptimal and largely symptom-based.

The peak incidence of SSc is reported to occur between 45 and 64 years of age [1]. Notably, age at disease onset was one of the characteristics that best predicted patient survival; advanced age was associated with worse survival [2]. Similarly, in the Scleroderma Lung Study I and II, older age was also associated with increased mortality [3]. In addition, pulmonary, renal, and cardiac complications are more prevalent in older SSc patients [4, 5]. Could the generally greater age at disease onset, higher frequency of SSc diagnosis in patients older than 45 years, and higher mortality in older patients imply that SSc represents a disease of accelerated aging? Here we will review recent experimental data supporting the involvement of cellular senescence, a hallmark of aging, in SSc. The potential of using drugs targeting cellular senescence, so-called senolytic therapies, in treating SSc will also be discussed.

Hallmarks of cellular senescence

Originally dismissed as a cell culture artifact, cellular senescence is now viewed as a fundamental cellular mechanism of normal tissue homeostasis, disease, and aging. Cellular senescence is involved not only in embryonic development but also tissue remodeling such as wound healing [6, 7]. In certain circumstances cellular senescence acts as a protective mechanism. For instance, cellular senescence suppresses malignant transformation by preventing cancer cell proliferation [8]. However, cellular senescence can also promote cancer development by altering the cellular microenvironment, thus acting as a double-edge sword. Indeed, persistent tissue accumulation of senescent cells could negatively impact homeostasis, leading to inflammation and age-related diseases.

Depending on how it is triggered, cellular senescence can be classified as replicative senescence, physiological senescence, drug-induced senescence, and stress-induced senescence (Figure 1). In addition to irreversible growth arrest and inability to replicate DNA, senescent cells undergo a series of morphological and molecular alterations that distinguish them from multiplying cells. Cells undergoing senescence have enlarged, flattened, and irregular cell shapes. This is typically coupled with increased numbers of lysosomes, which express β-galactosidase and contain lipofuscin granules. The mitochondria in senescent cells produce reactive oxygen species (ROS) and express high levels of anti-apoptotic proteins. Senescent cells are also characterized by nuclear changes, including enlarged nuclei, impaired nuclear integrity, global epigenetic changes of the chromatin landscape, formation of senescence-associated heterochromatin foci (SAHF), and DNA segments with chromatin alterations reinforcing senescence (DNA SCARS). Senescence is associated with overexpression of cell cycle inhibitors, particularly p21 and p16, which block cyclin-dependent kinases (CDKs)-cyclin complexes from phosphorylating retinoblastoma protein (Rb), leading to cell cycle arrest. Another striking feature of cellular senescence is the expression of a suite of proteins termed the senescence-associated secretory phenotype (SASP). The SASP comprises a variety of biologically active factors including growth factors, cytokines, chemokines, extracellular matrix (ECM) proteins and enzymes. These secreted factors signal in both autocrine and paracrine fashions, and can themselves induce senescence in neighboring cells, allowing senescence to spread. This cellular senescence cascade is depicted in Figure 1.

Figure 1. Hallmarks of cellular senescence.

The senescent phenotypes are classified by how they are triggered: replicative senescence, physiological senescence, drug-induced senescence, and stress-induced senescence. The cellular mechanisms and alterations of senescence include chromatin remodeling, cell cycle arrest, mitochondrial and lysosomal changes, and expression of senescence-associated secretory phenotype (SASP). These secreted factors signal not only in an autocrine fashion to reenforce the senescent phenotype, but also paracrinally to affect neighboring cells.

Implications of cellular senescence for SSc pathogenesis

Do senescent changes in various cell types implicated in the pathogenesis of SSc contribute to disease development and progression? Here we summarize the phenotypic and functional changes of senescent cells and discuss the implication of these changes in the context of SSc-associated vasculopathy, inflammation, and fibrosis.

Vasculopathy-Endothelial cells

Senescent endothelial cells (ECs) show characteristic changes in gene expression and function including lower nitric oxide production due to downregulation of eNOS, and increase in expression of PAI-1, which is one of the hallmark factors of SASP (Figure 2). These endothelial alterations are associated with functional changes including lower angiogenic and proliferative capacities, as well as inflammation, atherogenesis and increased thrombosis risks [9]. Notably, senescent ECs also undergo endothelial-to-mesenchymal transition (Endo-MT), a process that contributes to organ fibrosis [10]. Recent studies reveal that these changes phenocopy the hallmarks of SSc ECs [11–13], suggesting that cellular senescence is associated with, and might play a role in, endothelial dysfunction in SSc.

Figure 2. Summary of phenotypic and functional changes of senescent cells.

Cells that play critical roles in SSc pathogenesis are listed. Senescence in stromal cells (endothelial cells, vascular smooth muscle cells, and fibroblasts) is associated with promoting SSc vasculopathy and fibrosis. Immunosenescence, which is associated with alteration of immune functions in both innate and adaptive immune subsets, could also promote autoimmunity and inflammation pertinent to SSc.

Vasculopathy- Vascular smooth muscle cells

Alterations in vascular smooth muscle cells (VSMCs) contribute to the proliferative vasculopathy noted in SSc. VSMCs isolated from SSc skin biopsies showed increased proliferation and viability, and resistance to apoptosis, compared to VSMCs isolated from healthy skin [14]. In contrast, senescent VSMCs showed lower proliferative capabilities but display a pro-inflammatory SASP characterized by secretion of MMP9 and IL-1 [15]. Production of ECM components such as collagen is downregulated in these cells. In addition, it has been shown that VSMC senescence promotes atherosclerosis and necrotic core formation [16]. Taken together, the data suggest that VCMC senescence does not directly contribute to proliferative vasculopathy in SSc, since they are unable to replicate. However, it is possible that senescent VSMCs promote expansion of a subset of VSMCs into synthetically active cells indirectly through SASP, contributing to the proliferative vasculopathy phenotype in SSc.

Fibrosis-Fibroblasts

In addition to growth arrest, senescent fibroblasts are characterized by increased ECM production, increased glycolysis and other metabolic changes, and other features of activation, including myofibroblast transformation [17]. The secretome of senescent fibroblasts comprises proinflammatory cytokines (TNF-α, TGF-β, IL-6), chemokines (MCP-1), growth factors (FGF, CTGF, PDGF), MMPs, and other bioactive factors [18]. These secreted proteins can act in a self-amplified network to affect the local microenvironment and spread the senescent signal to neighboring cells. All of these properties of senescent fibroblasts have the potential to contribute to SSc fibrosis.

Immune cell senescence

When senescence occurs in cells of the immune system, it alters long-term immune responses, and consequently affects the capacity of the host to fight infections and other inflammatory processes. Indeed, immunosenescence disturbs both innate and adaptive immunity, both of play significant roles in SSc pathogenesis [19]. In general, age-related immune alterations include increase in monocytes, decreased lymphocytes and naïve cells, and increased memory cells (Figure 2).

Inflammation-Innate Immunity

Dendritic cells are central in coordinating immune responses and play key roles in maintaining tolerance and immunity. In the circulation and in the skin, these cells produce inflammatory mediators to activate other immune cells, as well as fibroblasts to promote fibrosis [20, 21]. When senescent, dendritic cells show impaired antigen presentation, interferon production and phagocytosis, leading to increased risk of autoimmunity (Figure 2). The ability of neutrophils to chemotax, migrate, form NETs and phagocytize is impaired with age. Senescent neutrophils express high levels of CXCR4 and become responsive to SDF-1α [22]. Although neutrophils from SSc patients have been shown to harbor functional defects [23], other studies suggest that NET formation is augmented in these SSc [24, 25]. It is thus difficult to assess the contribution of neutrophil senescence in SSc pathogenesis. In SSc, NK cells show decreased expression of chemokine and activation receptors [26]. Moreover, CD56^bright NK cells, which produce cytokines but are weakly cytotoxic, are reduced in SSc patients [27, 28]. These resemble some of the properties of senescent NK cells (Figure 2). Macrophage senescence is associated with defective phagocytosis and reduced potency, as shown by decreased IL-10 production by M2 macrophages [29, 30]. Macrophages from SSc patients show an activated and profibrotic transcriptome and phenotype [31, 32]. It is still unclear whether senescent macrophages contribute to SSc pathogenesis, although senescent macrophages and their SASP promote inflammation and fibrosis in a mouse model of radiation-induced pulmonary fibrosis [33].

Inflammation-Adaptive Immunity

SSc is associated with significant changes in adaptive immunity. Abnormal activation of B cells is prominent in SSc [34, 35], and B cells are expanded with increased naïve population and reduced activated memory B cells [36–38]. In addition, regulatory B cells are decreased and functionally impaired in SSc, as shown by reduced IL-10 production [39]. Interesting, activated B cells in SSc produce higher levels of IL-6 and TGFβ compared to ones from healthy controls [40]. The numbers and percentages of B cells are significantly decreased with age [41]. There is also a shift in the proportion of different B cell subsets. Late memory B cells, which are defined as CD19⁺IgG⁺IgD^-CD27^- B cells, are increased in elderly compared to younger individuals [42]. In contrast, IgD^-CD27⁺ B cells are significantly decreased. Senescent B cells show intrinsic defects such as decreased somatic hypermutation, defective class switch recombination, and reduced antibody affinity and neutralization capacity. It also results in increased SASP production and autoantibody production [43]. The senescent phenotype of B cells is summarized in Figure 2. Senescent B cells, or so-called age-associated B cells (ABCs), have been implicated in autoimmune and autoinflammatory disorders including systemic lupus erythematosus [44, 45], however their role in SSc has not been fully elucidated. An early report suggested that ABCs, defined as CD11c⁺CD21^- B cells, are present in higher frequency in the blood of SSc patients [46]. Comprehensive characterization of these cells and determination of their mechanistic implications in SSc warrant further research.

A declining CD4/CD8 ratio in T cells is associated with immunosenescence [47]. Similar to B cells, aging causes shrinking of the population of naïve T cells and an increase in memory T cells [48]. These T cells lose the expression of CD27 and CD28, but express NK markers including CD57 and KLRG-1 [49]. Senescent T cells include a fraction of cells that are CD45RO⁺, termed T effector memory cells [50]. Functionally senescent T cells acquire a proinflammatory phenotype, harboring features of Th1, Th17, Tfh, and Treg cells [51]. They produce SASPs that are characterized by pro-inflammatory cytokines. In SSc, there is an increased proportion of CD27^-CD28^- in CD8 T cells in the circulation [52]. CD4⁺CD28^- T cells are also reported to be expanded [38]. In addition, CD8⁺CD28^- T cells, which produce high levels of IL-13, have been shown in SSc skin [53]. Th17 cells are expanded in SSc [54]. CD4/CD8 ratio is increased in SSc patients compared to controls. CD4⁺ T cells display an activated phenotype in SSc with elevated activation markers CD69 and GITR [54]. The apparent impact of senescent T cells in SSc is not known. It has been shown that adoptive transfer of senescent T cells into young mice accelerated angiotensin-induced cardiovascular damage and kidney fibrosis in an IFN-γ-dependent manner [55]. The expansion of CD8⁺CD28^- T cells in SSc implies that at least a subset of T cells undergoes senescence in this disease. Indeed these cells have been shown to undergo rapid replication compared to CD8⁺CD28⁺ T cells, coupled with significantly shortened telomeres, suggestive of a replicative senescent phenotype [56].

Together, these studies indicate that cellular senescence might play a critical role in SSc pathogenesis, particularly involving ECs and fibroblasts. Immunosenescence could contribute to SSc pathogenesis in two distinct ways: senescent immune cells can directly induce inflammation and autoimmunity through their altered cellular functions; and senescent immune cells are defective in immune surveillance, including for senescent cells. Impaired ability of senescent macrophages, NK cells, and CD8+ T cells to eliminate senescent cells might indirectly promote SSc pathogenesis. Indeed, it has been shown that senescent cells are predominantly cleared by the immune system [57].

Cellular senescence has been implicated in fibrotic diseases including SSc

Cellular senescence and fibrosis have been linked. While cellular senescence plays a beneficial role in physiologic repair and wound healing, persistence of senescent cell accumulation and their release of SASP factors promote fibrosis [58]. As mentioned earlier, cellular senescence can result from various types of stress, including telomere attrition, DNA damage, oxidative stress, and even mitochondrial dysfunction [59–62]. There is accumulating evidence that cellular senescence is elevated in fibrotic diseases. For instance, in idiopathic pulmonary fibrosis (IPF) and SSc-associated interstitial lung disease, epithelial cells and fibroblasts display a senescent transcriptome signature and significantly increased levels of the senescence marker p16INK4A [63–65].

By secreting SASP, including ECM proteins (fibronectin, various collagens, and laminins), the matrix remodeling proteases (MMP-1, 3, 10, 12, 13, and 14), and growth factors implicated directly in fibrosis (TGFβ, PDGF, IL-6) [66, 67], senescent cells promote chronic inflammation [68], epithelial-to-mesenchymal transition (EMT) [69] and profibrotic phenotype changes of fibroblasts and macrophages [70] in a paracrine fashion. Fibroblasts explanted from skin biopsies of SSc patients with early diffuse disease displayed significantly elevated levels of the senescence effectors CDKN2A (encoding p16), TP53, PAPPA, IGFBPs, PDGFB, TNF, chemokine ligands, and multiple ECM remodeling genes (unpublished). The association of senescence in SSc is also shown at the genomics level. Whole exom sequencing of microdissected areas of dermal fibrosis in skin biopsies from patients with early disease with severe skin/lung involvement revealed the presence of large number of somatic mutations [71]. The mutation pattern exhibited a clock-like “senescence” signature that resemble what is seen in cancer. The authors speculate that genomic instability and somatic hypermutation is critical in SSc pathogenesis that drives fibrosis and inflammation.

As shown in numerous studies, senescent IPF fibroblasts are less sensitive to cytotoxic and pro-apoptotic signals, resulting in their accumulation [64, 72]. Overexpression of α-smooth muscle actin and ECM components by senescent fibroblasts promotes the development of fibrosis [73]. The apoptosis-resistant phenotype of senescent lung fibroblasts and myofibroblasts has been attributed to multiple mechanisms. In IPF, senescent fibroblasts exhibit decreased levels of the pro-apoptotic proteins Bak and Bax, and increased levels of the anti-apoptotic proteins Bcl-2 family proteins Bcl-2, Bcl-W, and Bcl-XL [17, 74, 75]. This imbalance of pro- and anti-apoptotic proteins contributes to failure of senescent fibroblasts to be properly eliminated. Senescent IPF lung fibroblasts are also highly resistant to apoptosis induced by Fas ligand and TNF-associated apoptotic ligand. In these cells, levels of FasL, TRAIL, and Caveolin-1 (Cav-1) are reduced, while AKT activity is elevated [76].

Senotherapeutics: targeting cellular senescence in SSc

As summarized above, increasing evidence links cellular senescence to the pathogenesis and progression of SSc. Senescent cells accumulate in fibrotic skin and lung. Assuming that their persistent accumulation in these tissues is detrimental, could targeting cellular senescence represent a viable therapeutic strategy?

Senotherapeutics describes a field focused on therapeutically eliminating or disabling senescent cells in a variety of human conditions [77, 78]. Examples of senotherapeutics include senolytics, which are drugs that remove senescent cells, or senomorphics, which are compounds that modulate SASP expression and function. It is perhaps not surprising that there is also substantial interest in targeting cellular senescence in fibrotic diseases. Recent pilot studies provide some evidence for the efficacy of “senolytic” therapy in the treatment of fibrotic diseases, including SSc. A single-arm, open-label clinical trial in 12 SSc patients with interstitial lung disease were treated with the putative senolytic drug dasatinib (a tyrosine kinase inhibitor) for 9 months. Three patients (25%) showed clinical improvement, which correlated with a decrease in the level of a senescence-related gene set in skin biopsies [79]. In contrast, patients who failed to show clinical improvement with dasatanib therapy showed no change in the senescence signature. These intriguing pilot observations suggest that dasatanib might have targeted senescent cells in the skin, which possibly contributed to the clinical improvement in these patients. It is unclear why only a subset of treated patients demonstrated clinical improvement and a decline is senescence gene signature levels; of note, “improvers” in this trial tended to have higher senescence signatures prior to therapy compared to “non-improvers”. Another pilot study sought to evaluate senolytic treatment in IPF, using a “senolytic cocktail”. In this open-label clinical trial, intermittent combination therapy with dasatinib plus quercetin (a flavanol) for three weeks resulted in significant improvements in physical performance, although pulmonary functions and circulating levels of senescence and fibrosis markers remained unchanged [80]. In another open-label short-term phase 1 pilot study, patients with diabetic kidney disease were treated with the senolytic cocktail. In this study, once-daily oral administration of 100 mg dasatinib plus 1000 mg quercetin for three days was associated with a reduction in the epidermis and adipose tissue of p16INK4A-positive cells (38%) and p21CIP1-positive cells (30%) [81].

To date, there is a scarcity of clinical evidence on the effectiveness and safety of senotherapeutics, including systemic effects and side effects of this treatment strategy. Most senotherapeutics currently being investigated are repurposed drugs or dietary interventions. However the next generation of senotherapeutics, including navitoclax, HSP90 inhibitors, as well as senomorphic drugs including JAK inhibitors and rapamycin, are on the horizon. It is likely that effective senolytic therapy can be administered episodically rather than chronically, potentially limiting adverse effects, as opposed to senomorphics that need continuous administration.

Conclusions

Experimental evidence supports the concept that cellular senescence in present in SSc and plays a critical role in disease establishment and progression. Potential use of senotherapeutics in SSc would be expected to be an effective treatment approach. Clearly, very substantial hurdles remain before larger clinical trials of senolytic therapy for fibrosis can be considered. It will be necessary to further assess specificity of such therapies in killing senescent cells, and characterize their cytotoxic effects, to demonstrate their long-term safety, and to identify reliable markers for response [82]. Furthermore, adding senotherapeutics as combination therapy in conjunction with other drugs modulating various pathways in SSc pathogenesis may have added benefits and result in improved outcome for this complex and potentially fatal disease.

KEY POINTS.

• Cellular senescence, an irreversible exit from the cell cycle with both beneficial and harmful biological functions, is both a hallmark of aging, and contributes to age-related health decline through direct and indirect mechanisms.

• Accumulating evidence support the tissue presence and potential pathogenic roles of cellular senescence in fibrosis in a variety of conditions, including SSc.

• Cellular senescence might be directly responsible for the emergence of pathogenic myofibroblasts in SSc, and their apparent resistance to apoptosis.

• Emerging evidence suggest that senolytic therapy is feasible for fibrotic conditions, and might be associated with clinical improvement and reduction of tissue senescent cell burden.