Ageing, cellular senescence and chronic kidney disease: experimental evidence

Abstract

Purpose of review

Chronic kidney disease (CKD) is often viewed as an accelerated and premature ageing of the kidney, as they share common pathological features characterized by cellular senescence. In this review, we summarize the experimental evidence linking cellular senescence to the pathobiology of kidney ageing and CKD, and discuss the strategies for targeting senescent cells in developing therapeutics for ageing-related kidney disorders.

Recent findings

Kidney ageing and CKD are featured with increased cellular senescence, an irreversible state of cell cycle arrest and the cessation of cell division. Senescent cells secrete a diverse array of proinflammatory and profibrotic factors known as senescence-associated secretory phenotype (SASP). Secondary senescence can be induced by primary senescent cells via a mechanism involving direct contact or the SASP. Various senolytic therapies aiming to selectively remove senescent cells in vivo have been developed. Senostatic approaches to suppress senescence or inhibit SASP, as well as nutrient signaling regulators are also be validated in animal models of ageing.

Summary

These recent studies provide experimental evidence supporting the notion that accumulation of senescent cells and their associated SASP is a main driver leading to structural and functional organ degeneration in CKD and other ageing-related disorder.

INTRODUCTION

In the past several decades, the elderly population has significantly increased worldwide, thanks to the rise of average life expectancy. This imposes an enormous challenge on healthcare systems and has unprecedented health, economic and social implications on a global scale [1,2]. Ageing is associated with structural and functional degeneration in major organs of the body including the kidneys. The incidence and prevalence of chronic kidney disease (CKD) are exceedingly higher in the elderly[3]. Extensive studies have shown that ageing and CKD share many common stressors and have similar underlying mechanisms, such as cellular senescence, oxidative stress, DNA damage, mitochondrial dysfunction, telomere shortening and hyperactive Wnt/β-catenin signaling [4]*.

On cellular level, ageing is characterized by cellular senescence, which is an irreversible state of cell cycle arrest and the cessation of cell division [5]*. Cellular senescence plays a critical role in regulating diverse biological events under different circumstances [6,7]. In general, acute and transient cellular senescence could be beneficial by making positive contributions to many biological processes such as embryogenesis, tissue repair and tumor suppression. However, chronic and sustained senescence after injury inevitably leads to the accumulation of senescent cells and the release of a host of proinflammatory and profibrotic factors known as the senescence-associated secretory phenotype (SASP) [8]. Cellular senescence often precede the manifestation of morphologic lesions and kidney insufficiency, indicating that it is involved in the onset of kidney ageing and the pathogenesis of CKD [9]. In this review, we concisely summarize the experimental evidence linking cellular senescence to the pathobiology of renal ageing and CKD, and discuss recent advances on understanding of the mechanism underlying cellular senescence. We also highlight potential strategies for targeting senescent cells in developing therapeutics for age-related kidney disorders.

MAJOR FEATURES IN THE AGEING KIDNEY

Ageing kidneys manifest significant changes in their morphology and function. On the macrostructural scale, kidney cortical volume decreases with age. Surface roughness develops and grows, and the frequency of forming renal cysts increases during the ageing process. On the microstructural scale, ageing kidneys are characterized by a series of pathological features resembling that found in CKD, such as tubular atrophy, inflammation, interstitial fibrosis, glomerulosclerosis, vascular rarefaction and arteriosclerosis. Kidney ageing is also accompanied by decreased number of the nephrons, with the remaining nephrons being hypertrophic [10]. Estimated glomerular filtration rate (eGFR), approximated by urinary creatinine clearance, declines at approximately 0.75 ml/min/1.73 m² per year in a longitudinal study of relatively healthy individuals [10,11]. In many aspects, ageing not only leads to renal lesions and functional deficiency but also causes an increased susceptibility to kidney injury [12].

MOUSE MODELS USED IN KIDNEY AGEING

For experimental studies, a suitable animal model is essential for interrogating the key pathways of ageing. The mouse models used in ageing studies could be classified into natural ageing and accelerated ageing (Table 1). Mice with age older than 20 months are often regarded as naturally aged, and they typically present with age-related phenotypes such as muscle weakness, cardiac hypertrophy, reduced cognitive function. Kidneys of the naturally aged mice are manifested with mild glomerulosclerosis, tubular atrophy, inflammation and renal fibrosis, and these lesions get worsened progressively with age (Table 1).

Table1.

Experimental models of ageing

Model	Gross condition	Renal manifestation
Naturally aged mice
Naturally aged mice [45,72]	Muscle weakness, cardiac hypertrophy, reduced cognitive function.	Glomerulosclerosis, tubular atrophy, inflammation, renal fibrosis.
Accelerated aged mice (chemically and radiation-induced)
D-galactose-induced aged mice [48]	Neurodegeneration, osteoporosis, hearing loss, muscle weakness, weight loss.	Tubular atrophy, inflammation, renal fibrosis.
γ-Ray irradiation-induced aging mice [73]	Decreased white blood cell, red blood cell and platelet counts.	Renal fib rosis.
Accelerated aged mice (genetical modified mice)
SAMP6 [14]	Osteoporosis, increased adipose tissue, hepatic lipidosis.	Renal fibrosis.
SAMP8 [14]	Neurodegeneration, retinal degeneration, cardiac fibrosis.	Renal fibrosis.
Klotho null mice [74]	Short lifespan, infertility, growth retardation, reduced cognitive function, arteriosclerosis, osteoporosis, ectopic calcification, skin atrophy and emphysema.	Renal interstitial fibrosis, renal vascular wall thickening, renal ectopic calcification.
Zmpste24 deficient mice [75]	Growth retardation, alopecia, muscle weakness, rib fractures, dilated cardiomyopathy, muscular dystrophy, lipodystrophy.	Unknow
Ercc1-/Δ mice [76]	Growth retardation, liver failure, increased mortality, spleen hypoplasia, dystonia, ataxia, sarcopenia, kyphosis, uraemic encephalopathy.	Renal failure, glomerulosclerosis, proteinuria, renal interstitial fibrosis, tubular atrophy.
Xpd TTD/ TTD mice [77]	Osteoporosis, kyphosis, osteosclerosis, cachexia, infertility, short lifespan.	Impaired renal function, inflammation.

Abbreviation: SAMP6, senescence-accelerated mouse prone 6; Zmpste24, zinc metalloproteinase Ste24; Ercc1-/Δ, hypomorphic mice with a reduced level of the ERCC1-XPF DNA repair endonuclease; XPD, xeroderma pigmentosum group D.

There are also numerous models with accelerated ageing (Table 1). Mouse models of accelerated ageing could be either induced chemically by D-galactose or irradiation-induced by γ-ray. The D-galactose-induced aged mice display tubular atrophy, renal inflammation and interstitial fibrosis in the kidneys. They also present with premature ageing with neurodegeneration, osteoporosis, hearing loss, muscle weakness and weight loss. The γ-ray irradiation-induced aged mice present with decreased counts of white blood cell, red blood cell and platelets in the circulation, as well as fibrotic lesions in the kidney. These models are characterized by an increased cellular senescence, which undoubtedly contributes to the premature ageing phenotypes in these models.

Another type of accelerated ageing models is induced by gene mutations in the genetically modified mice (Table 1). Of them, Klotho-null mice are the most well characterized and widely used model. Klotho is anti-ageing protein, and its loss in mice is associated with short lifespan, infertility, growth retardation, reduced cognitive function, arteriosclerosis, osteoporosis and ectopic calcification. Renal manifestations of Klotho^−/− mice include interstitial fibrosis, renal vascular wall thickening, renal ectopic calcification. Mouse models of progeroid syndrome, a group of rare genetic disorders that mimic physiological aging, are also used in ageing studies [13]. These mice have increased numbers of senescent cells that contribute to premature ageing. For example, the senescence-accelerated mouse prone 6 and 8 (SAMP6 and SAMP8) mice develop a variety of age-associated diseases, such as retinal degeneration, renal fibrosis, immune dysfunction and degenerative joint disorder [14]. Excision repair cross complementing 1 (Ercc1) -/Δ mice, another model of human progeroid syndrome, are impaired in several DNA repair systems and develop age-dependent, all kinds of features of ageing such as motor abnormalities, sarcopenia, cachexia, overall frailty, liver failure, proteinuria and renal fibrosis, and have a shortened life span of 6–7 months. Mice with deficiency of Sirtuin 6 (Sirt6), a NAD⁺-dependent histone deacetylase and ADP-ribose transferase enzyme, also exhibit premature ageing and suffer from ageing-associated CKD and other degenerative disorders [15]. These studies support the notion that accumulation of DNA damage and genotoxic stress, as well as loss of anti-ageing proteins, contributes to kidney aging and other age-related disorders.

HALLMARKS OF CELLULAR SENESCENCE

There are several hallmarks that could be used to identify cellular senescence. Senescent cells often possess characteristic features of an enlarged size, increased expression of β-galactosidase, accumulation of lipofuscin granules and senescence-associated heterochromatin, among others [16]. Senescent cells also produce and secrete a series of soluble factors known as SASP, which consist of a diverse array of proinflammatory cytokines, chemokines, growth factors and extracellular matrix remodeling factors, such as interleukin-1β (IL-1β), IL-6, IL-8, C-X-C motif chemokine ligand 1 (CXCL1), transforming growth factor-β1 (TGF-β1), plasminogen activator inhibitor-1 (PAI-1) and monocyte chemoattractant protein-1 (MCP-1) [17]. The SASP components could be harnessed for assessing senescence state [18].

The acidic senescence-associated β-galactosidase (SA-β-gal) activity reflects an increased β-galactosidase activity in lysosomes, which becomes the best described and most used marker to identify senescent cells [19,20]. The expression levels of cyclin-dependent kinase (CDK) inhibitors, including p16^INK4a (hereafter referred as p16), p21^CIP1 (hereafter referred as p21), p19^ARF (in mouse), p14^ARF (in human), p27^KIP1 and p15^INK4b are prominently elevated in the senescent cells [4]. Furthermore, accumulation of senescence-associated DNA damage foci (SADF) and senescence-associated heterochromatic foci (SAHF) in the nucleus is also described as senescence biomarkers [21]. Overexpression of SAHF-associated proteins can induce cellular senescence and repress those genes related to cell proliferation [22]. The γ-H2AX, generated by phosphorylation of the histone H2AX, is also defined to reflect the magnitude of senescence [23]. Besides, down-regulation of cell proliferation marker Ki67 and structural nuclear Lamin B1 protein (LMNB1) is also utilized as senescence-associated markers [24].

Recent studies have shown that early growth response 2 (Egr2) is a direct transcriptional activator of the p16/pRb (the retinoblastoma tumor suppressor protein) and p53/p21 pathways, and serves as a novel marker of cellular senescence [25]. In addition, tRNA-derived fragments is a potential biomarker for evaluating senescent cells [26]. The human positive cofactor 4 (PC4) is a sensitive marker and can disrupt proteostasis resulting in age-related organ dysfunction [27]. The histone acetyltransferase KAT7, a genetic and epigenetic marker of senescence, is upregulated in murine models of ageing [28]. Cdkn1a transcript variant 2 is accumulated and activated during ageing and could be used for the diagnosis of ageing-related diseases [29].

It should be stressed that all these biomarkers are neither specific nor unique for senescent cells. In addition, not all cells with these markers exhibit senescent pathology and not all senescent cells express p16 [30]. A single-cell RNA sequencing reveals that cells in response to the same stressors display distinct senescent subtypes [31]. Therefore, multiple biomarkers are recommended to be used in concert for identifying senescent cells in aged kidneys.

DIFFERENT TYPES OF CELL SENESCENCE

Cellular senescence could be divided into two main subtypes, namely acute senescence and chronic senescence. Acute senescence occurs after extracellular cues target a specific group of cells in the tissue. Acute senescence can positively exert beneficial effects by constituting a programmed mechanism that regulates embryogenesis, limits tumorigenesis, promotes wound healing and improves tissue repair [32]. Acute senescent cells are subsequently cleaned by infiltrated macrophage in a timely fashion. When the immune system cannot remove senescent cells at the rate at which senescent cells are being produced, an inevitable outcome is that the accumulation of these cells leads to a disruption in tissue homeostasis. Chronic senescence is typically induced by tissue injury, extracellular stressors or macromolecular damage. As such, chronic senescence usually play a critical role in reducing cell renewal and repopulation, triggering inflammation and fibrotic responses via secreting SASP, thereby promoting CKD progression [8].

Senescent cells, if not be removed appropriately and timely, can propagate and expand via a process causing secondary senescence. Primary senescent cells are induced by detrimental factors such as defective DNA repair, oncogene activation, telomere shortening and others. Primary senescent cells produce distinct paracrine and endocrine signals, leading to other non-senescent cells to undergo secondary senescence. Secondary senescence has been demonstrated to occur in response to SASP secretion through two broad types of mechanisms [33]*. Firstly, SASP factors have been shown to be involved in spreading senescence via a paracrine fashion, which is also called paracrine senescence. Secondly, primary senescent cells are capable of inducing senescence by a direct cell-to-cell contacts, which is defined as juxtacrine senescence. Secondary senescence allows a small number of senescent cells to expand and make it possible to spread to distant locations, leading to age-related diseases.

MECHANISM OF CELLULAR SENESCENCE

The mechanism of cellular senescence remains incompletely understood. There are a wide variety of extracellular stressors and triggers that can induce cellular senescence. These stressors act through diverse mechanisms, eventually leading to permanent cell cycle arrest (Fig. 1).

Figure 1. Major signaling pathways that regulate cellular senescence.

A wide variety of stressors induce cellular senescence via diverse signaling pathways. Many of these stressors such as DNA damage, oxidative stress and nephrotoxins trigger DNA damage response (DDR), leading to activation of ATM/ATR kinases, which induces G2/M phase cell cycle arrest via checkpoint kinase 1 and 2 (CHK1/CHK2) or G1/S phase arrest via p53/p2-mediated inhibition of CDK2. Wnt/β-catenin stimulates p53 and p16 to promote cellular senescence, while anti-ageing proteins Klotho and Sirt1 hamper senescence by blocking these signal pathways. RAAS, renin angiotensin aldosterone system; ATR, ATM and Rad3-related.

Increased expression of cyclin-dependent kinase inhibitors

Numerous signal pathways are involved in cellular senescence, primarily including the p53/p21 and p16 pathways (Fig. 1). The ATM (ataxia telangiectasia mutated) protein kinase, p53, p21 and p16 are activated and induced in response to stress-triggered DNA damage [4]. This cascade of events eventually leads to inhibition of the phosphorylation of cyclin-dependent kinase (CDK) complexes and Rb, resulting in the cessation of cell proliferation and cellular senescence. P53/p21 pathway is thought to be implicated in the onset of cellular senescence, whereas p16 signaling is largely involved in the initiation and maintenance of senescent phenotype [18].

Down-regulation of anti-ageing proteins

There are some intrinsic anti-ageing factors that prevent cellular senescence and premature ageing. Among them, Klotho and Sirt1 are the two most studied proteins. Klotho is an interesting anti-ageing protein that halts cellular senescence. It is a single-pass transmembrane protein and expressed primarily in the kidney [34]. It has been shown that serum level of Klotho decreases with physiological ageing, indicating its potential as a marker of ageing [34]. Mice with Klotho deficiency manifest an accelerated ageing phenotype, characterized by short lifespan, infertility, growth retardation, reduced cognitive function and ectopic calcification. On the contrary, overexpression of Klotho in transgenic mice leads to an increased longevity and prevents ageing-related disorders [35]. Klotho exerts suppressive effects on the p53/p21 and Wnt/β-catenin signaling [36,37]. Klotho also protects against mitochondrial dysfunction in aged cells and CKD by attenuating Wnt/β-catenin signaling [38].

Sirt1, an intrinsic inhibitor of ageing, is capable of deacetylating both histone and non-histone proteins such as FOXO, p53 and NF-κB [39,40]. Therefore, Sirt1 have a broad impact on a variety of key signaling pathways associated with cellular senescence and ageing. It has been reported that podocyte-specific Sirt1 depletion in mice manifests age-related sclerosis in glomerulus [41]. Sirt1 has anti-ageing properties and reverses senescence in endothelial cells, and ablation of Sirt1 in these cells results in accelerated senescence phenotype through acetylating p53. In contrast, overexpression of Sirt1 prevents cellular senescence via restraining p53 activity [42]. Furthermore, it has been shown that sodium tetrasulfide (Na₂S₄) directly sulfhydrated Sirt1, thereby inhibiting p65 NF-κB and STAT3 phosphorylation/acetylation and alleviating diabetic renal lesions [43].

Other regulatory mechanisms

Recent studies have uncovered several novel signal pathways that play a role in exacerbating renal ageing. For example, Wnt/β-catenin is shown to take part in mediating cellular senescence and age-related renal fibrosis [44,45]. Wnt9a/β-catenin activation is associated with renal tubular senescence and renal fibrosis in diseased kidneys, which is accompanied by an increased expression of p16, p53 and p21, and enhanced SA-β-gal activity in renal tubular epithelia [46]. Recent studies identify the calcium-activated chloride channel accessory 1 (CLCA1) as a novel mediator of kidney injury in aging through TMEM16A/Cl⁻ current pathway [47]. The IκB kinases including IKKε, IKKα and IKKβ are key regulators of cellular senescence and renal ageing, which is regulated by extracellular vesicles (EVs)³⁸. Cannabinoid receptor 2 (CB2) could accelerate mitochondrial damage in renal tubular cells and exerts a detrimental influence on renal ageing [48]. The proper functioning of the circadian clockwork is crucial to prevent ageing. Conversely, ageing reprograms the circadian clock output and disturbs sleep-wake cycles, causing a lowered capacity to synchronize circadian rhythms in peripheral tissues [49,50]. Recent evidence suggests that dysbiosis of intestinal flora and changes of microbial metabolism exhaust reparative potentials in the cell and lead to ageing [51]. Phenotypic discrepancies of senescent cells created by metabolic and epigenomic reprogram may open up new approaches for the treatment of age-related diseases [52].

THERAPEUTIC STRATEGIES FOR KIDNEY AGEING

Given the critical role of cellular senescence in kidney ageing and diseases, it is conceivable that senescence could be used as a therapeutic target for the treatment of CKD. In recent years, novel therapeutic approaches have been developed to halt ageing-associated organ dysfunction via targeting senescent cells [53]. There are three main strategies to pharmacologically regulate senescence for therapeutic intervention. Table 2 lists these three types of anti-ageing remedies that have been validated in a variety of experimental models of ageing, including senolytic, senostatic and nutrient signaling regulators.

Table 2.

Therapeutic strategies targeting cellular senescence.

Senotherapy	Agents
Senolytics	Dasatinib and quercetin [78], fisetin [79], catechins [80], uPAR-targeted CAR-T [58], ABT263 [55], A1331852 [81], A1155463 [81], ABT737 [81], alvespimycin (17-DMAG) [76], FOXO4-DRI [56], panobinostat [82], BPTES [61]
Senostatics	SS31 [83], klotho [62], mitoQ [84], SKQ1 [85], melatonin [86], ICG001 [45], Na2S4 [43], ganoderma lucidum [87], ruxolitinib [75], MSC-derived EVs [64], SIRT1 activator [65], PPAR-γ agonists [66], metformin [67], rapamycin [88]
Nutrient regulators	Calorie restriction [69,89], exercise [70], resveratrol [90], curcumin [91], spermidine [92], barbarum [93], DHA [71].

Senolytic is an approach or agent that selectively induces death of senescent cells, thereby eliminating them in the affected organs (Table 2). Studies from animal experiments show that senolytic agents can reverse ageing phenotypes and improve kidney function [30,54]. For example, ABT263 (navitoclax), an inhibitor of Bcl-2 family, has been confirmed to have the ability to eliminate senescent cells in aged mice and preserve tissue regeneration [55]. The anti-senescence peptide FOXO4-DRI is potent remedy for the treatment of kidney disease by binding with FOXO4 to disrupt the p53-FOXO4 interaction, thereby releasing p53 into the cytosol and triggering p53-mediated apoptosis [56]. A peptide that removes p16^high cells can attenuate nonalcoholic steatohepatitis-related hepatic lipidosis and immune cell infiltration in p16-Cre^ERT2-tdTomato mouse model [57]. Recent studies show that urokinase-type plasminogen activator receptor (uPAR)-targeted chimeric antigen receptor (CAR) T cells can remove senescent cells. Such uPAR-targeted CAR T cells have therapeutic effect on senescence-associated organ dysfunction via selectively clearing uPAR-expressing senescent cells in murine models [58]**. Similarly, eliminating p16-positive senescent cells in the p16-3MR (tri-modal reporter) and the INK-ATTAC (apoptosis through targeted activation of caspase 8) models also promotes would healing and prevents ageing-associated disorder [59,60]. Inhibition of kidney-type glutaminase (KGA)-dependent glutaminolysis in aged mice attenuates ageing and exhibits a therapeutic benefit in age-associated tissue degeneration [61]*. These experimental studies suggest that senolysis can be harnessed as a therapeutic strategy for the treatment of ageing-related diseases.

In addition to senolytic, a related concept is the so-called senostatic, which means to suppress senescence or the SASP generated by senescent cells. Therapies inhibiting the SASP have proven to be effective in experimental animal models. For example, high phosphate induces kidney tubular epithelial cell senescence, and Klotho have the potential to hamper tubular epithelial senescence [62]. A recent study has indicated that mesenchymal stem cells (MSC)-derived EVs suppress the progression of ageing by blocking cellular senescence [63,64]. Sirt1 activator and peroxisome proliferator-activated receptor-γ (PPAR-γ) agonists have been used for prevention and therapies of ageing-related renal deficiency [65,66]. Metformin exerts an anti-ageing effect by targeting senescent MSCs in CKD [67].

There are nutrient signaling regulators, which can regulate ageing-related diseases and promote health. Numerous studies show that caloric restriction (CR) prolongs lifespan in laboratory animals [68,69]. Moderate CR can be used to halt the age-related kidney dysfunction by inhibiting the endothelin-1 (ET-1) expression [68]. In aged mice, physical exercise reduces triacylglycerol and restores plasma lipids homeostasis by regulating lipid metabolism [70]. Dehydroascorbic acid (DHA) retards or even reverses renal impairment in mouse models of ageing [71].

CONCLUSION

Over the past several years, growing evidence has shown that aged kidneys and CKD share many pathological features and move to similar destination. Accumulation of senescent cells and their associated SASP has been identified as a main driver leading to structural and functional degeneration in CKD and ageing-related disorder of other organs. We have gained a wealth of new knowledge and better understanding on the role, triggers and mechanisms of cellular senescence in ageing and ageing-associated degenerative diseases.

Future studies are warranted to explore new characteristics and delicate regulation of cellular senescence and define the specific molecular targets of senescent cells. Single cell RNA sequencing and relevant bioinformatics analyses need to be employed to facilitate our understanding of the heterogeneity and dynamics of senescent cells in the affected organs. Whether primary and secondary senescence displays any divergence in response to current senotherapy remains an open question. As emerging evidence points to senescent cells as a promising target for therapeutic intervention, this provides an unprecedented opportunity for developing more effective senotherapies to fight against CKD and other ageing-related disorders. It is hopeful that future therapeutics based on targeting cellular senescence will be translated into the clinic for improving the life of millions of patients with ageing-related diseases.

KEY POINTS.

• Aged kidney and CKD share common pathological features characterized by cellular senescence, an irreversible state of cell cycle arrest and the cessation of cell division.

• There are several models of ageing, including naturally aged and accelerated aged mice, for interrogating the signal pathways involved in cellular senescence and evaluating therapeutic efficacy.

• Senescent cells secrete a host of proinflammatory and profibrotic factors known as senescence-associated secretory phenotype (SASP), which can lead neighboring healthy cells to undergo secondary senescence.

• Key senescence signaling pathways include the p53/p21 and p16 routes, leading to inhibition of the phosphorylation of cyclin-dependent kinase complexes and cell cycle arrest.

• Senescent cells are a promising target for senotherapy, an approach to specifically eliminate or inhibit cellular senescence.