Cellular Senescence, Inflammaging and Cardiovascular Disease

Cell cycle arrest is triggered either via the DNA damage response, inducing p53 and p21^Waf1 (p21) or via direct induction of p16^INK4A (p16). While p16 exerts its effects primarily through inhibition of cyclin‐dependent kinase (CDK) 4 and 6, p21 is a broad CDK inhibitor (CDKi). Both mechanisms result in hypophosphorylation of Retinoblastoma (Rb), reducing the transcriptional activity of E2F transcription factors, which is required for cell cycle progression and arrests the cell cycle at G1/S and G2/M (reviewed in [23]).

The widely used markers for cell cycle arrest in senescence p16 and p21 can compensate for each other to induce senescence [24], but show distinct dynamics with differential effects on cell cycle arrest and apoptosis induction. In the time course of senescence, p21 shows early induction, which is followed by p16 expression [25]. p21 dynamics play a pivotal role in cell fate decision and the potential cell cycle re‐entry [26]. As p21 and p53 may also induce apoptosis and quiescence, the commonly used marker for cell cycle arrest in cellular senescence is p16^INK4a expression, which can be directly quantified via gene reporters, that is, in the INK‐ATTAC or p16‐3MR mice [8, 27], however p21 reporter mice are also available [28].

An increased activity at pH 6 of the SA‐β‐Gal is considered the gold standard to detect senescent cells. It is a lysosomal hydrolase, which cleaves β‐d‐galactosides [29]. While being expressed in most cells, higher expression in senescent cells results in detectable activity at the suboptimal pH 6, which is distinct from the typical optimum at pH 4 in non‐senescent cells [3]. Its up‐regulation is part of both increased lysosomal biogenesis and lysosomal accumulation of the enzyme and SA‐β‐Gal up‐regulation can be caused by the CDKi cascade [29]. However, increased activity is not specific to senescent cells, as many tumors show similar activity levels at pH 6 [30, 31]. Additionally, macrophages, which have a high lysosomal β‐galactosidase activity per se, can show positive activity at pH 6 while not showing other signs of senescence [32].

Senescent cells significantly change their secretome. It encompasses pro‐inflammatory, pro‐fibrotic and growth‐inducing proteins, nucleic acids, lipids and extracellular vesicles and thereby alters the senescence niche and exerts systemic effects (the broader composition is reviewed in [33]). The SASP was first described in 2008 by three independent groups as a primarily pro‐inflammatory and protein‐based secretome and our understanding of the composition and effects has vastly improved over the past years [4, 5, 6, 33]. The SASP plays pivotal roles in local anti‐tumor effects and autocrine and paracrine induction and maintenance of the senescence state [34, 35]. The best characterized part of the SASP is proteins involved in pro‐inflammatory signaling, especially chemokines and cytokines, as well as extracellular matrix remodeling proteins, that is, matrix metalloproteinases. The pro‐inflammatory components of the SASP are mainly induced via the transcription factors nuclear factor of kappa light polypeptide gene enhancer in B‐cells (NF‐kB) and CCAAT enhancer binding protein beta (C/EBPb), inducing the expression of cytokines such as interleukins (IL) 6 and 8 in senescent cells. This effect has been reported to be primarily regulated by IL‐1α [36], which is particularly important in oncogene‐induced senescence [37]. While these effectors maintain the senescence state and hence contribute to tumor suppression of the senescent cell [6], they also contribute to systemic pro‐inflammatory effects and hence systemic tumorigenic effects [37]. C‐X‐C motif ligands (CXCL) are chemokine components of the SASP, which mediate immune cell infiltration and activation, induce cancer cell proliferation and contribute to senescence maintenance via binding to their respective receptors, with a particularly important role for CXCR2 [5, 38, 39].

In addition to pro‐inflammatory components, various growth factors are contained in the SASP. Within this group, Transforming Growth Factor beta (TGF‐β) and Growth Differentiation Factor 15 (GDF‐15) play pivotal roles. GDF‐15 is a core SASP component which associates with aging, frailty and age‐related diseases [15, 40, 41, 42]. It belongs to the TGF‐β cytokine superfamily and plays roles in embryonic development, tissue repair but also cancer progression and metastasis [43, 44]. Moreover, it has been associated with poor outcomes in various cardiovascular diseases, such as coronary artery disease [45], atrial fibrillation [46] and heart failure with reduced and preserved ejection fraction [47]. TGF‐β is a prototypical pro‐fibrotic growth factor, which induces fibroblast proliferation and activation [48] and is required for endothelial to mesenchymal transition and activation in the heart [49]. TGF‐β plays a central role in the SASP‐mediated induction of senescence in embryogenic senescence, as well as postnatal senescence in the liver and the tumor microenvironment [50, 51].

Furthermore, the SASP changes with different senescence states, with higher expression levels of typical components, such as IL1A, CCL2 and TNFRSF11B observed in deep senescence states than in early states. More recent data showed that late senescence induces the retrotransposable element LINE1, inducing type 1 interferon responses, which are required for the maintenance of senescence, suggesting a time‐dependent program of SASP composition [52].

Interestingly, high SASP expression can contribute to an immune‐suppressive signature, with increased T regulatory cells, macrophages with immune‐suppressive properties and exhausted T cells which is dependent on TGF‐β [50]. Additionally, senescent niches express high levels of immune inhibitory signals, potentially facilitating this phenotype [13]. This might indicate that insufficient clearance and accumulation of senescent cells in chronic senescence is a consequence of TGF‐β and SASP‐mediated effects.

The primary biomarkers p16 and SA‐β‐Gal have allowed for identification and a thorough mechanistic understanding of senescence in vitro. Given the relevant heterogeneity in context and niche‐dependent functions, phenotypes and paracrine effects of senescence in vivo, "omics" approaches, especially single‐cell and single‐nucleus sequencing, seem promising modalities to investigate senescence in vivo. However, the identification of senescent cells is hampered by the usually low expression of CDKN1A/Cdkn1a (encoding for p21) and CDKN2A/Cdkn2a (encoding for p16), which are hence prone to drop‐out effects. Therefore, gene sets, databanks and scoring approaches are required to characterize cellular senescence in development, aging and disease.

The SeneQuest [53], CellAge [54] and SenMayo [55] databases, which are curated sets of senescence‐associated genes, may facilitate such approaches. However, as these datasets lack cell type specificity, they seem limited in in vivo settings. There is an urgent need for organ and cell type‐specific characterization of cellular senescence to facilitate therapeutically targeting the detrimental effects of cellular senescence in general and in the cardiovascular system specifically.

Apart from genetic predisposition, such as familial hypercholesterinemia or hyperlipoproteinemia (a), atherosclerosis is a typical age‐associated disease. Aged a therosclerosis‐prone Ldlr^−/− mice show a myeloid skewing in the blood with an overall decreased abundance of lymphoid cells. On the other hand, within atherosclerotic plaques, aging associates with an increase in CD4+, CD8+ and CD4 + CD8+ T‐cells. Aging induced a shift in the macrophage population with an increase in Il1b^hi macrophages and a higher expression of MMPs [61] (see also Chapter 2 for further details on macrophage‐related aging phenotypes).

Atherogenesis involves senescence of various cell types, including endothelial cells (EC), vascular smooth muscle cells (VSMC), macrophages and foam cells. Endothelial cell senescence contributes to risk factors associated with atherosclerosis, especially insulin resistance [62] and endothelial dysfunction [63], which are reversible through senolysis [64]. Senescent endothelial cells act as a feed‐forward mediator of senescence via SASP components, including IL‐1α, IL‐6, IL‐8 and CCL2, but also proteins associated with matrix remodeling and growth factor signaling, hence forming a pro‐inflammatory and pro‐atherogenic milieu [62]. Mature atherosclerotic plaques show a high burden of senescent endothelial cells in humans [65] and a model of endothelial cell senescence in an Apoe^−/− background shows accelerated plaque progression and inflammatory responsiveness via enhanced VCAM1 promoter accessibility for NF‐B [66]. Importantly, senescent ECs show increased monocyte adhesion [67] which in combination with increased endothelial permeability [68] may facilitate tissue infiltration with bone marrow derived macrophages.

Senescent foamy macrophages, as indicated by p16‐induction, are enriched in atherosclerotic plaque early in the course of atherogenesis in Ldlr^−/− mice and genetic senescence clearance attenuates plaque progression. Senescent cells are major contributors to the pro‐inflammatory plaque milieu and plaque remodeling via the expression of matrix metalloproteinases [69]. As part of the vascular niche, VSMC undergo age‐associated senescence [13]. Senescent VSMC exert a typical pro‐inflammatory SASP profile, including IL1α, IL6, IL8 and CCL2 [25]. A comprehensive review on cellular senescence in atherosclerosis is provided in [70].

The most frequent forms of heart failure, heart failure with preserved ejection fraction (HFpEF), as well as ischemic cardiomyopathy, the most common etiology of heart failure with reduced ejection fraction (HFrEF), occur strongly age‐associated. Early work on senescence in human heart failure showed increased senescence in presumed (c‐kit positive) cardiac stem cells [71, 72]. However, with the increasing evidence against the existence of relevant cell populations with stem cell properties in the adult human heart, it is likely, that the increased abundance of senescent c‐kit positive cells in the failing heart represent EC [73]. This is in line with the finding that senescence‐prone mice on western diet develop endothelial senescence and heart failure with preserved ejection fraction [74].

In human idiopathic dilated cardiomyopathy, senescence spatially overlaps with perivascular fibrosis. Similarly, in mouse models of fibrotic heart disease, fibrosis, a key feature of chronic myocardial disorders, was also associated with cellular senescence, suggesting that fibroblast senescence may contribute to the progression of heart failure. However, the role of fibroblast senescence in myocardial fibrosis remains controversial. On the one hand, cellular senescence correlates with myofibroblast‐driven perivascular fibrosis in the Angiotensin‐II hypertension model and left ventricular pressure overload via transaortic constriction (TAC) [75]. This observation is, however, not informative on the consequences of fibroblast senescence, which are controversial. Using genetic depletion of senescence via deletion of p53 and p16 in TAC‐treated mice results in increased perivascular fibrosis [75]. Treatment with the senolytic agent navitoclax 4 weeks after initiation of AngII treatment in WT mice, however, reduces fibrosis and improves diastolic dysfunction [76]. The relevant difference between these models may lie in the timing of senolysis. While global deletion of p53 and p16 prevents the development of cellular senescence from the beginning of LV pressure overload, treatment with navitoclax 4 weeks after induction of AngII treatment clears senescence in the later course of fibrosis. This may indicate anti‐fibrotic effects of acute myofibroblast senescence, while chronic fibroblast senescence may be pro‐fibrotic, which is in line with findings from wound‐healing studies [27].

Increased fibroblast‐derived TGF‐β induces cardiomyocyte senescence and dysfunction in LV‐pressure overload [77]. Cardiomyocyte senescence induced by acute myocardial infarction is associated with impaired regeneration and angiogenesis [78, 79], which is in accordance with the atypical SASP of senescent cardiomyocytes which is mainly characterized by growth factors, which induce paracrine senescence in other myocardial cell populations [80].

As it is a well‐defined characteristic of age‐associated cardiovascular disease, targeting cellular senescence represents a promising therapeutic target.

The survival and accumulation of senescent cells rely on pathways to evade apoptosis. These pathways are targeted by so‐called senolytic agents or senolytics. By inhibiting the anti‐apoptotic pathways, senolytics induce apoptosis in senescent cells and hence induce senescence clearance [81]. The best characterized mechanism for apoptosis evasion is via anti‐apoptotic BCL‐2 family member signaling. This complex regulatory network contains pro‐ and anti‐apoptotic mediators, regulating intrinsic apoptosis via mitochondrial outer membrane permeabilization (reviewed in [82]). Upstream regulators of the BCL‐2 family, which are activated in senescent cells and exert anti‐apoptotic effects are ephrins, Src kinases, PI3KCD, plasminogen activator inhibitors (i.e., PAI‐1/Serpine1).

Additionally, senescent cells can evade extrinsic apoptosis, that is, viaPAI, which inhibits death receptors and caspase 3 [83]. Importantly, there is relevant heterogeneity between the induced pathways and not all senescent cells induce similarmechanisms, which also results in heterogeneous responses to different senolytic agents [81].

Currently available senolytics are repurposed compounds, with no available drugs being specifically developed or approved for their usage as senolytics. The best characterized senolytic agents are the tyrosine kinase inhibitor dasatinib, which is clinically used for treating chronic myeloid and acute lymphoblastic leukemia [84, 85]. The senolytic effect stems from its inhibition of Src family kinases and ephrin‐induced anti‐apoptotic effects [86, 87]. While showing good senolytic capacity in preadipocytes, given as a single drug it does not induce apoptosis in senescent endothelial cells. Therefore, it is commonly combined with the plant flavonoid quercetin, which shows a complementary profile (dasatinib + quercetin; D + Q), especially including additional targeting of senescent endothelial cells [56, 88]. In addition to quercetin, additional natural products and phytochemicals have shown the ability to act as senolytics and interfere with SASPs such as fisetin, curcumin, and procyanidin C1 [89, 90, 91]. Fisetin inhibits the PI3K/Akt/mTOR pathway and shows some specificity for senescent endothelial cells [89]. The compound navitoclax exerts effects on various cell types and is a direct inhibitor of the anti‐apoptotic Bcl‐2 family members Bcl‐2, Bcl‐xL and Bcl‐w [92].

Senolytics have demonstrated that selectively clearing cardiac senescent cells can mitigate the detrimental consequences of cardiac aging and related diseases [93]. We and others have shown that senolytic treatment in aged mice rescues aging‐induced adverse remodeling in the heart using navitoclax [80], D + Q [56, 57, 88] or fisetin [13]. Navitoclax treatment improves left ventricular remodeling and post‐myocardial infarction survival [94, 95]. D + Q has shown promising results to reduce atrial fibrillation inducibility after myocardial infarction and arrhythmia rates with aging [56, 96].

In the murine Angiotensin II infusion model, treatment with navitoclax attenuates the pro‐fibrotic response through clearance of senescent fibroblasts [76] (see also chapter 1.3.2). Interestingly, D + Q has been found to prevent mitochondrial DNA–induced inflammation and prolong the survival of cardiac allografts [97].

The effects of senolysis in late atherosclerosis remain controversial. One experimental study reported reduced lesion size in ApoE^−/− mice upon navitoclax treatment [98], while another demonstrated increased lesion size and induced mortality [99]. The latter trial used very high navitoclax doses, which might have induced toxicity. In Ldlr^−/− mice, the same dosage given for a significantly shorter period reduced atherosclerosis progression [69]. All three studies demonstrate smooth muscle cells and macrophages/foam cells to be the main senescent populations in atherosclerosis.

The pioneering development of chimeric antigen receptors (CAR‐)T cells has expanded opportunities for cell‐specific targeting in cardiovascular medicine. Beyond preclinical studies showing that engineered CAR‐T cells or CAR‐macrophages directed against activated fibroblasts can improve cardiac function in injury models [100, 101], similar strategies may be envisioned to selectively eliminate senescent cells. Indeed, natural killer group 2D (NKG2D‐) CAR T cells have been shown to eliminate senescent cells in aged mice and nonhuman primates [102], while other studies demonstrated prophylactic and durable efficacy of senolytic CAR‐T cells against age‐related metabolic dysfunction [103]. Designing targeted approaches to eliminate senescent cells within the cardiovascular system could thus provide a foundation for innovative and highly specific therapeutic strategies. Given the complexity and high treatment costs associated with these therapies, a prerequisite for clinical use will be the identification of suitable patients, who are at high risk for adverse outcomes and will very likely profit from targeted senolysis.

Table 1 gives an overview of the currently available work on the experimental use of senolytics in cardiovascular disease.

Although preclinical studies indicate that senolytics can improve cardiac function in aged animals, evidence in humans remains limited, and no therapies have yet been shown to conclusively slow or reverse age‐related decline in heart function. Nevertheless, several early‐phase clinical trials are now investigating fisetin in older adults, targeting vascular aging and arterial diseases (Table 2). For example, one (NCT06399809) is assessing whether fisetin can reduce senescent cell burden and improve functional outcomes in individuals with peripheral artery disease. Additional trials are evaluating intermittent fisetin treatment to enhance vascular endothelial function and reduce aortic stiffness, as well as its safety, pharmacokinetics, and tolerability in healthy and older adults with multiple chronic conditions (NCT06133634, NCT06431932). Most evidence on the effects of senolytics comes from studies in age‐related neurodegenerative diseases, as reviewed in [106]. As these results may also be instrumental for developing cardiovascular‐targeted strategies, and with the brain–heart interaction gaining increasing recognition, these findings are additionally summarized in Table 2. While most studies are ongoing, phase I clinical trials investigating D + Q showed feasibility and trends towards improved surrogate markers in patients at risk or with mild Alzheimer's disease [107, 108].

While these senolytic interventions show encouraging neuro and cardioprotective effects, they are not without potential side effects, including cytopenias, gastrointestinal symptoms, and off‐target toxicity. Recently, galactose‐modified senolytic prodrugs have been developed, which take advantage of SA‐β‐Gal activity in senescent cells hence increasing specificity. Examples are JHB75B, a duocarmycin derivate, which belongs to a group of highly cytotoxic antibiotics [109], and Nav‐Gal, a navitoclax prodrug, which carries a reduced risk of inducing cytopenias [110]. Overall, the currently available data call for randomized controlled trials with clinically relevant endpoints to investigate beneficial effects of senolysis in at‐risk aging populations and patients with cardiovascular disease.

Senomorphics are a class of pharmacological agents that target the detrimental effects of senescent cells without inducing apoptosis, distinguishing them from senolytics. These compounds primarily act by modulation of the SASP. They target signaling pathways involved in SASP regulation, including mTOR, AMPK, NF‐κB, p38 MAPK, and JAK/STAT (Figure 2). Rapamycin and metformin are among the most extensively studied senomorphics. Rapamycin suppresses mTOR signaling, thereby reducing oxidative stress and enhancing autophagy [111]. In preclinical studies, it has been shown to delay the age‐related decline in both systolic and diastolic cardiac function in mice [112] and dogs [113]. Improvements in cardiac performance have also been reported in a mouse model of progeria [111]. Moreover, treatment with rapamycin further alleviated oxidative stress and vascular dysfunction in aged mice [114].

Beyond its widespread use in diabetes management, metformin is now being explored for its ability to slow biological aging. The ongoing TAME trial investigates if it can reduce the burden of age‐related diseases [115]. Clinical trials suggest benefits beyond glucose control, as older individuals with impaired glucose tolerance showed improvements in both muscle and fat tissue function when treated with metformin [116]. Mechanistic work in experimental models links these effects to enhanced autophagy in vascular cells, activation of AMPK, and suppression of stress responses such as mTOR signaling and endoplasmic reticulum stress in the heart [111]. Remarkably, long‐term administration in primates slowed aging across diverse organs, such as the cardiovascular system, the brain and skin, by mitigating inflammation, senescence, and epigenetic alterations [117]. While these data are promising, clinical confirmation is still required to establish metformin as a therapy that truly modifies human aging.

Ruxolitinib, a JAK1/2 inhibitor, is another well‐studied senomorphic. By blocking JAK/STAT signaling, it reduces the secretion of pro‐inflammatory SASP factors and alleviates systemic inflammation, thereby improving tissue function in preclinical models of aging [118]. Other compounds, including p38 MAPK inhibitors (e.g., SB203580), glucocorticoids like dexamethasone, and sirtuin activators such as resveratrol, have also been shown to attenuate SASP production and restore cellular function.

Taken together, the currently available data suggest that cellular senescence is a hallmark of both cardiovascular aging and disease, which can be therapeutically targeted. As the cell types and mechanisms involved vary between diseases, future work has to identify suitable compounds to target the distinct deleterious populations and effects to guide new precision‐medicine approaches in cardiovascular aging and disease.

Overall, age‐related changes in innate immune function can contribute to heart failure through multiple mechanisms, including altered responses to infections [149] or direct effects of aging on the immune system.

Cardiac macrophages play a key role by regulating fibrosis and thereby promoting diastolic dysfunction—a hallmark of the aging heart [150]. With age, there is a shift towards pro‐inflammatory bone marrow–derived macrophages and a decline in reparative cardiac resident macrophages, altering the cardiac microenvironment and driving fibroblast activation, which in turn increases collagen deposition. This process may be further exacerbated by cellular senescence, as senescent macrophages release TGF‐β, thereby enhancing fibroblast activation [151]. Pioneering studies by Hulsman et al. demonstrated a crucial role of macrophages in facilitating electrical stability in the heart [152]. The reduced number of cardiac resident macrophages, which mediate this effect, in aging may contribute to impaired atrioventricular conduction often observed in the elderly. In line, others reported that aging results in chronotropic incompetence and disturbed repolarization after lipopolysaccharide injection [145].

Finally, one should not forget that macrophages are also important players in the resolution of fibrosis. Recent studies have highlighted their role in reverse remodeling of fibrotic tissue [153, 154]. Thereby, macrophages contribute to the resolution and clearance of fibrotic material following the relief of cardiac stress, such as unloading after pressure overload. Since the reversal of fibrosis depends on the clearance capacity of macrophages, a function that is impaired during aging [155], it is plausible that the accumulation of aged macrophages may hinder this reparative process as discussed. Cardiac macrophages promote diastolic dysfunction via IL‐10‐mediated fibroblast activation in aged mice, an interesting finding, as IL‐10 is anti‐inflammatory, but a typical SASP component in macrophage senescence [150, 156].

Fibroblasts themselves may fuel inflammaging. In the aging mouse heart, mesenchymal fibroblasts adopt a pro‐inflammatory phenotype, producing elevated levels of chemokines, primarily monocyte chemoattractant protein‐1 (MCP‐1), which promotes leukocyte recruitment into the heart [157]. In addition, aged cardiac fibroblasts increasingly express the complement factor C3, which can activate macrophages via their C3 receptors [13]. Furthermore, the finding that cardiac fibroblasts express MHC class II molecules for antigen presentation to CD4+ T cells upon interferon gamma stimulus and contribute to myocardial fibrosis, further strengthens the suggested fibroblast–immune cell interaction [158]. Together, these processes create a self‐perpetuating cycle in which aging exacerbates immune cell–fibroblast interactions, amplifying inflammation and fibrosis [159].

Adaptive immunity, specifically T‐cells, may further aggravate cardiac inflammaging. Adoptive transfer of aged, senescent T cells to immunodeficient RAG1 KO mice accelerates Ang II‐induced cardiovascular and renal fibrosis compared with young T‐cell transfer [160]. Likewise, adoptive cell transfer of T cells from aged animals [146] or transplantation of terminally differentiated CD4+ T cells in young mice promotes myocardial inflammation and leads to impaired cardiac function. Transplantation of these T cell populations in healthy mice caused rather mild effects on cardiac function [146] suggesting that spontaneous, heart‐directed immune responses during aging may arise, but that additional innate immune and intrinsic cardiac changes may contribute to the decline of age‐related myocardial inflammation and functional decline.

Ample experimental data support the anti‐aging effects of dietary interventions, for example, omega‐3 fatty acids, polyphenols (e.g., resveratrol, curcumin) or other phytochemicals exhibiting antioxidant, anti‐inflammatory, and autophagy‐enhancing effects that target core pathways involved in cellular senescence and tissue degeneration. These phytochemicals regulate key molecular players such as sirtuins, AMPK, NF‐κB, and mTOR, offering promise in delaying age‐associated pathologies and promoting longevity. However, there is no documented benefit in clinical studies for the prevention of cardiovascular aging.

In addition, anti‐inflammatory diets (e.g., a Mediterranean diet) and lifestyle interventions such as regular moderate exercise, caloric restriction, or intermittent fasting have well‐documented beneficial effects and reduce inflammaging [105]. For example, the randomized controlled CALERIE Phase 2 trial demonstrated significant improvements in aging biomarkers and insulin resistance after 2 years of moderate calorie restriction compared with a usual diet [200]. However, most studies investigated these interventions in patients with metabolic syndrome or demonstrated diabetes‐preventive effects, making it difficult to distinguish metabolic from specific anti‐aging effects. Therefore, prospective studies specifically designed to evaluate the clinically relevant impact of such interventions on cardiovascular aging are still lacking.