Reversing age-associated arterial dysfunction: insight from preclinical models

Abstract

Cardiovascular diseases (CVDs) remain the leading causes of death in the United States, and advancing age is a primary risk factor. Impaired endothelium-dependent dilation and increased stiffening of the arteries with aging are independent predictors of CVD. Increased tissue and systemic oxidative stress and inflammation underlie this age-associated arterial dysfunction. Calorie restriction (CR) is the most powerful intervention known to increase life span and improve age-related phenotypes, including arterial dysfunction. However, the translatability of long-term CR to clinical populations is limited, stimulating interest in the pursuit of pharmacological CR mimetics to reproduce the beneficial effects of CR. The energy-sensing pathways, mammalian target of rapamycin, AMPK, and sirtuin-1 have all been implicated in the beneficial effects of CR on longevity and/or physiological function and, as such, have emerged as potential targets for therapeutic intervention as CR mimetics. Although manipulation of each of these pathways has CR-like benefits on arterial function, the magnitude and/or mechanisms can be disparate from that of CR. Nevertheless, targeting these pathways in older individuals may provide some benefits against arterial dysfunction and CVD. The goal of this review is to provide a brief discussion of the mechanisms and pathways underlying age-associated dysfunction in large arteries, explain how these are impacted by CR, and to present the available evidence, suggesting that targets for energy-sensing pathways may act as vascular CR mimetics.

CARDIOVASCULAR DISEASE, VASCULAR FUNCTION, AND AGING

Cardiovascular disease (CVD) is the leading cause of morbidity and mortality in the United States, with an estimated 92 million US adults presenting with at least one form of CVD. Among these, ~47 million are estimated to be over 60 yr of age (9). It is estimated that by 2030, almost 44% of the US population will have some form of CVD, and the projected direct medical costs is estimated to reach over $800 billion (9). The staggering health care and economic burden associated with CVD makes the identification of therapeutic strategies to treat CVD critical. Importantly, advanced age is the major risk factor for development of CVD (31, 74), with the prevalence of CVD increasing with age in both men and women (9). Impaired endothelial function and increased large artery stiffness (35, 42) characterize the vascular aging phenotype and represent strong predictors of CV events and clinical CVD.

ARTERIAL DYSFUNCTION IN AGING

While once thought to be an inert barrier, the endothelium is now known to play an important role in vascular homeostasis (39). In addition to its role in determining vascular permeability, the endothelium also contributes to the control of arterial tone via production of vasodilators and vasoconstrictor substances, as well as to angiogenesis and to the oxidative, inflammatory, and thrombotic phenotype of the artery (105). Nitric oxide (NO) is a key endothelium-derived vasodilator in arteries that can broadly impact endothelial function (129). With aging, endothelium-dependent dilation (EDD) is impaired in both animal models, assessed in vitro in excised arteries and human subjects, assessed by flow-mediated dilation (36). This impairment in EDD is associated with a reduction in NO bioavailability (36). Beyond enhancing vascular tone leading to increased vascular resistance, reduced NO bioavailability may also impair angiogenic responses (97), as well as contribute to the prooxidant (137), proinflammatory (139), and prothrombotic (46) arterial phenotype that is associated with aging (131), thus, contributing to the increased prevalence of CVD in older adults (44).

In addition to impaired endothelial function, stiffening of the large elastic arteries with aging is also an important contributor to increased CVD risk and is associated with pathophysiological conditions, such as hypertension, left ventricular hypertrophy, subendocardial ischemia, and cardiac fibrosis (103). Stiffening of the large arteries also leads to a widening of the pulse pressure and a greater central forward pressure wave, a significant predictor of a major CVD event (24). The large elastic arteries function as a conduit for blood to the periphery and as a capacitance organ to dampen the pulse pressure caused by the ejection of blood from the heart during systole. Clinically and experimentally in both human subjects and animal models, arterial stiffness can be assessed in vivo by aortic pulse wave velocity (PWV). This is accomplished using Doppler ultrasound to assess the time it takes for the pulse wave to travel a known distance across the large arteries, with higher PWV, indicative of stiffer large elastic arteries. With aging, increases in collagen abundance (114) and crosslinking (115) combined with decreases in elastin density (48) within the arterial wall contribute to increased stiffness. In addition, age-associated increases in vascular resistance may also contribute to changes in large artery stiffness in vivo (62).

Loss of elasticity of the large arteries diminishes diastolic pressure, widening pulse pressure, and exposing the microvasculature to increased fluctuations in pressure (103). The downstream effect can be end-organ dysfunction, particularly to vulnerable regions, such as the cerebral and renal circulations (103), but, likely, loss of elasticity has effects in multiple other tissues and organs. This occurs through alterations in the components of the forward and reflected pulse wave, such that there is an increased amplitude of the reflected wave, and this has been associated with end-organ damage and increased cardiac load (133). In addition, studies in cultured endothelial cells suggest that a stiffer matrix mimicking that in older age promotes proatherosclerotic responses to shear stress. For example, endothelial cells cultured to a stiffer matrix demonstrate decreased alignment, a reduction in barrier function, and reduced endothelial nitric oxide synthase (eNOS) activation in response to athero-protective laminar shear stress (72). Thus, age-related dysfunction of the vascular endothelium and increases in arterial stiffness act in concert to increase the risk of CVD in older adults. As such, a better understanding of the mechanisms underlying these vascular changes may inform therapeutics to combat CVD in this population.

ROLE OF OXIDATIVE STRESS AND INFLAMMATION IN AGE-ASSOCIATED ARTERIAL DYSFUNCTION

A key mechanism underlying age-associated reductions in EDD and NO bioavailability, as well as increases in large artery stiffness, is the development of vascular oxidative stress (42, 93). Age-related vascular oxidative stress is associated with increased production of reactive oxygen species (ROS), including superoxide anion (110) and hydrogen peroxide (H₂O₂) (140). Enzymatic ROS production through NADPH oxidase (NOX) has been implicated in age-related vascular oxidative stress, and NOX exists in seven isoforms, with different expression patterns across tissues. While NOX4 is the most abundant NADPH oxidase isoform in endothelial cells (2) and is a constitutively active form of the enzyme (80), this isoform produces hydrogen peroxide rather than superoxide (120). The age-associated increase in arterial ROS is associated with increased expression of the cytosolic subunits, p47phox and p67phox (35, 40), components of the NOX2 isoform (17) and is mediated by increased activity of this oxidant enzyme (35, 40), mitochondrial dysfunction, and reductions in the antioxidant enzyme manganese (Mn) superoxide dismutase (SOD) (35, 40, 77). Superoxide mediates the age-associated reduction in NO bioavailability by reacting with NO to form peroxynitrite. This reaction quenches bioavailable NO and leads to the nitration of tyrosine residues on proteins (nitrotyrosine) in arteries of rodents and humans (35, 40, 76). Functionally, these oxidative stress-driven biochemical events impair EDD in both mice and humans (40, 42, 76, 93) and contribute to stiffer large elastic arteries in rodent models (43, 124) and older women (94). Thus, oxidative stress-associated suppression of NO is a cause of dysfunction in older arteries.

Aging is also associated with chronic, low-grade systemic inflammation that is characterized by increases in circulating C-reactive protein (CRP), proinflammatory cytokines, including TNF-α and interleukin-6 (IL-6), as well as adhesion molecules, such as vascular cell adhesion molecule-1 (VCAM-1) (34, 35). Among older adults, circulating markers of inflammation, such as CRP and IL-6, are positively related to aortic PWV (116) and inversely related to EDD (128). In addition, the proinflammatory transcription factor NF-κB is increased in endothelial cells of older humans (34, 35), and genetic blockade of NF-κB activation in endothelial cells extends life span (57). NF-κB is a proinflammatory transcription factor that resides in the cytoplasm through its interaction with the inhibitory protein IκB-α (69). In response to inflammatory stimuli (69) or ROS (117), IKKβ is activated and subsequently phosphorylates IκB-α, releasing its inhibition and allowing NF-κB to translocate into the nucleus, where it can activate gene transcription of its downstream proinflammatory target genes, such as TNF-α and IL-6 (69). Importantly, this age-associated increase in proinflammatory signaling has been implicated in arterial dysfunction in older rodents and humans. For example, NF-κB inhibition by salicylate (salsalate in humans) improves EDD in aged mice (77) and in overweight/obese middle-aged and older adults (101), and both salsalate (63) and inhibition of TNF-α (3) improve aortic PWV in older adults, supporting a causative role for inflammation in age-related arterial dysfunction.

Oxidative stress and inflammation act in a feedforward manner to negatively impact arterial function with aging. For example, proinflammatory signaling in the vasculature can lead to the local recruitment of immune cells that produce ROS and contribute to a prooxidant environment. Furthermore, inflammatory signaling can stimulate superoxide production and oxidative stress by inducing transcription of redox-sensitive genes like those encoding subunits of NADPH oxidase (87). Thus, oxidative and inflammatory pathways interact with advancing age to induce and perpetuate arterial dysfunction.

BENEFICIAL EFFECTS OF CALORIE RESTRICTION ON ARTERIAL FUNCTION IN AGING

Calorie restriction (CR) is typically a lifelong intervention, initiated after sexual maturity, in which caloric intake is restricted by 40% compared with ad libitum (AL) intake without malnutrition. CR has been demonstrated to improve maximal and/or median life span, as well as physiological function in rodents and nonhuman primates (134). Although CR is one of the most effective antiaging interventions, CR is not universally effective, with some studies demonstrating no improvement in age-associated loss of body fat, bone mass, and density in nonhuman primates, as well as no effect or even a shortened life span in some inbred mouse strains (81, 82, 112). Nevertheless, CR has been associated with a dramatic reduction in CVD in nonhuman primates (23), a protection against cardiac ischemia (41), an improvement in EDD and NO bioavailability (26, 130), as well as a reduction in large artery stiffness (37) and reduced wall thickness in small cerebral arteries (37, 130) in aged rodents (Fig. 1). Importantly, the effects of CR to improve EDD and NO bioavailability can be, at least partially, recapitulated when mice are exposed to short-term CR (3–6 wk) that is initiated in old age (110). Furthermore, recent evidence from the CALERIE trial, in which nonobese adults were randomized to 25% CR for 2 yr, indicates the CR also slowed biological aging assessed by published biomarker algorithms, i.e., the Klemera-Doubal and homeostatic dysregulation methods (8) and reduced CVD risk factors (106), such as serum lipids and blood pressure (96) in middle-aged adults.

Fig. 1.

Effects of aging and caloric restriction (CR) on arterial function. Age-associated increases in oxidative stress and inflammation lead to endothelial dysfunction and large artery stiffening that is prevented by lifelong CR. With aging, oxidative stress and inflammation are the consequence of increases in oxidant production, NF-κB activation, and proinflammatory cytokine/adhesion molecule expression in arteries that contribute to reduced vasodilation and nitric oxide bioavailability, as well as to alterations in the structural components of the artery, i.e., collagen and elastin, favoring increased arterial stiffness. Life-long CR prevents these age-associated arterial changes.

The vascular benefits of CR appear to be, at least partially, mediated by an attenuation of age-related oxidative stress (Fig. 1), such that in vitro treatment of arteries with the superoxide scavenger TEMPOL improves EDD and NO bioavailability in old AL-fed (40), but not in old mice after CR (37, 110). These results are indicative of an attenuation of the superoxide-mediated suppression of EDD/NO after CR (37, 110, 130). Furthermore, both long-term (37, 130) and short-term (110) CR reduces arterial oxidative stress, as evidenced by a reduced abundance of nitrotyrosine (37, 110). Furthermore, age-related elevations in arterial superoxide production, measured by electron paramagnetic resonance spectroscopy (37, 110, 130) or lucigenin chemiluminescence assay (26), are reduced after CR. While increases in superoxide production with aging are associated with increased expression and activity of the oxidant enzyme, NADPH oxidase (40), CR appears to constrain this source of ROS, with the expression of NOX4 and the p67 subunit of NOX2, as well as the activity of NADPH oxidase reduced in old mice after CR (25, 37, 110). Mitochondrial ROS production, another important source of arterial oxidative stress, is also elevated in aged primary cerebral microvascular endothelial cells. In addition, there is functional evidence of Mn SODʼs role in mitochondrial ROS production. Mice with haplodeficiency in Mn SOD increase vascular superoxide production with advancing age (18, 136), and this source of ROS production is reduced in cells isolated from mice after life-long CR (25).

In addition to increased oxidant production, an inadequate antioxidant defense also contributes to increased tissue oxidative stress in aging. SODs are critical antioxidant enzymes that convert superoxide to hydrogen peroxide (H₂O₂), playing a primary role in ROS removal (47). Although at high concentrations, H₂O₂ is, itself, a ROS that can impair vascular function (113), the H₂O₂ produced through the actions of SOD can also promote vasodilation via direct actions as an endothelium-derived hyperpolarizing factor (EDHF) (66), as well as indirectly by inducing the expression of eNOS, the enzyme responsible for the production of NO (38). Interestingly, the importance of H₂O₂ as a vasodilator differs across the life span and with the emergence of coronary artery disease (CAD). Indeed, recent evidence demonstrates that the primary mediator of flow-induced dilation in coronary arteries shifts from prostacyclin in youth to NO in healthy adulthood and is again altered in the presence of CAD to H₂O₂ (10).

Although elevated superoxide production should signal an increase in antioxidant defenses, total antioxidant capacity is reduced (65), and the expression and activity of critical antioxidant enzymes fail to increase with aging. For example, neither copper zinc (CuZn)—the intracellular isoform—nor manganese (Mn)—the mitochondrial isoform of SOD (37, 40, 110)—is increased in arteries of old mice (37, 40). Although the expression of extracellular (ec) SOD has been shown to be either unchanged (37, 78) or increased (40) in arteries of old mice, there is no increase in activity of any of these SOD isoforms in arteries with advancing age (37, 40, 110). Thus, an inadequate response in SOD expression/activity with aging may contribute to both an excess of superoxide leading to reduced NO bioavailability, as well as to an inadequate H₂O₂ production that both may contribute to impaired vasodilation (66). Further, there is direct functional evidence in CuZn SOD heterozygous mice that reductions in CuZn SOD increase superoxide and limit vasodilation and that these effects are exacerbated with advancing age (33). In addition to reducing the production of ROS, CR-mediated improvements in oxidative stress are also associated with improved antioxidant defenses. Indeed, in contrast to AL aging, total antioxidant capacity of the serum of old CR rats does not differ from that of young rats (65). In addition, arterial expression of the antioxidants, Mn, CuZn, and ec SOD (37, 110), as well as total SOD activity (110), are increased in aged mice after CR. Thus, a primary mechanism underlying the beneficial effects on the vasculature appears to be through a reduction in oxidative stress that is mediated by both a reduction in oxidant production and an improvement in antioxidant defenses.

Although less is known regarding the effects of CR on arterial inflammation, there is evidence that CR induces an anti-inflammatory effect (Fig. 1). With aging, inflammation impairs EDD by contributing to superoxide-mediated suppression of NO bioavailability (77) and contributes to the stiffening of large elastic arteries by increasing vascular smooth muscle tone, which results from reduced NO bioavailability and/or increased concentrations of local or circulating constrictors (73), as well as by stimulating increased collagen production, a structural component of the arterial wall that confers stiffness. Inflammation-associated increases in collagen production are downstream of activation of NADPH oxidase-derived ROS production (127), highlighting the complex interactions of oxidative stress and inflammation in the aged vasculature. CR can act directly to reduce inflammation by lowering expression of proinflammatory adhesion molecules and cytokines, as well as indirectly by reducing the vascular inflammation associated with ROS. Indeed, CR reduces serum abundance of the adhesion molecules E-selectin, P-selectin, and VCAM-1 (141), as well as arterial gene expression for ICAM-1 in old rats (26). In addition, age-associated increases in serum cytokine expression, e.g., CRP (65), IL-6, and TNF-α (125), as well as endothelial cell secretion of TNF-α, IL-6, and IL-1β (25) are all reduced in rodents after CR. Although the mechanisms are incompletely understood, these anti-inflammatory effects may be downstream of attenuated NF-κB activity, as both gene expression and activity of NF-κB are reduced in nuclear extracts of arteries and endothelial cells from CR compared with AL-fed old mice (25, 26). Likewise, it appears that the age-associated decrease in activity of the NF-κB inhibitor, IκB-α (77, 135), is attenuated after CR (135). These anti-inflammatory effects may translate clinically, as serum CRP is reduced in human subjects after CR (45). Taken together, it appears that a reduction in NF-κB-associated inflammatory signaling may also contribute to improved vascular dysfunction in aged mice after CR.

ROLE OF mTOR, AMPK, AND SIRT-1 IN CR

While the molecular mechanisms underlying the beneficial effects of CR on arterial function are still being explored, reducing oxidative stress and inflammation via the modulation of energy-sensing pathways has been suggested as a possible mechanism. Elucidating these upstream pathways is critical because, despite the strong evidence for the beneficial effects of CR, the translatability of lifelong 40% CR to humans is limited. Nevertheless, evidence from the CALERIE trial suggests that shorter-term 25% CR is effective at delaying age-related phenotypes and improving CVD risk factors in adults (8, 96), suggesting that manipulation of pathways involved in the beneficial effects of CR in animal models may be also be efficacious in adults. Previous studies have identified energy-sensing pathways, including the mammalian target of rapamycin (mTOR), AMPK, and sirtuin (SIRT) pathways as critical modulators of longevity after CR (12, 52, 68, 84). As such, these pathways may hold promise as CR mimetics.

Rapamycin is an inhibitor of mTOR that was discovered more than 30 yr ago from an Easter Island soil sample (122). It is a potent antifungal metabolite produced from bacteria that, in addition to its antifungal properties, was also found to have antiproliferative and immunosuppressant properties when used in high concentrations (83). mTOR is a highly conserved serine/threonine kinase that responds to changes in energy balance or growth factors and regulates many cellular functions, including translation, transcription, protein turnover, metabolism, and stress responses (132). Early studies treating yeast and flies with rapamycin revealed that inhibition of mTOR activity increased life span (67), and a follow-up study reproduced these longevity-related findings in mice (56), demonstrating that inhibiting mTOR via dietary rapamycin could not only increase life span but may also delay age-associated phenotypes. Genetic manipulation of a downstream target of mTOR, ribosomal S6 protein kinase (S6K1), supported these findings by demonstrating that reduced mTOR signaling increased life span and ameliorated age-associated reductions in bone mineral density and insulin sensitivity (119), suggesting that inhibition of mTOR signaling may be a mechanism by which CR exerts its beneficial effects. Although most studies to date examining the role of mTOR in cellular homeostasis have focused on insulin and nutrient signaling, enhanced AMPK signaling was implicated in the blunting of age-related physiological dysfunction after mTOR inactivation in S6K1 knockout mice (119). Although, 6 mo of 35–40% CR initiated at 3 mo of age did not alter phosphorylation of mTOR, S6K, or TSC2 in the aorta, skeletal muscle, liver, brain, kidney, and lung (121), these studies were performed in young rats and as such, the effect of CR on mTOR pathway intermediates needs to be assessed in old animals in which mTOR signaling is elevated. Taken together, these studies suggest that reducing mTOR signaling can mimic many of the effects of CR, but the direct role of mTOR in the beneficial effects of CR remains to be elucidated.

AMPK is also a highly conserved heterotrimeric serine-threonine kinase that, like mTOR, is an important energy-sensing signaling protein integrating energy balance, metabolism, and stress resistance (55, 59). AMPK is activated in response to increases in the AMP:ATP ratio. Pharmacologically, AMPK activity can be increased directly after treatment with aminoimidazole carboxamide ribonucleotide (AICAR) and indirectly after metformin treatment. AICAR is an adenosine analog taken up into cells by adenosine transporters and phosphorylated by adenosine kinase, resulting in accumulation of AICA-Riboside monophosphate (ZMP), which mimics the stimulating effect of AMP on AMPK. Although the precise mechanism of action is not known, metformin is known to act by inhibiting mitochondrial complex I, altering the AMP:ATP ratio and leading to AMPK activation (7). While the effects of long-term AICAR treatment on longevity are not known, increased life span has been demonstrated after long-term metformin treatment (5). Although age-related endothelial dysfunction is associated with decreased arterial AMPK (79), AMPK activity was shown to be unchanged in heart, muscle, and liver after 4 mo of CR (51). Still, it remains to be elucidated what role, if any, AMPK activation may play in the beneficial effects of CR on vascular function per se.

Sirtuin-1 (SIRT-1) is a nuclear NAD-dependent nuclear deacetylase (21) that acts by deacetylating histone and nonhistone proteins (102) impacting gene expression, metabolism, and aging (104). Across multiple tissues, SIRT-1 activity has been associated with reductions in oxidative stress, inflammation, and proapoptotic signaling, as well as in improvements in insulin sensitivity, DNA damage repair, and telomere stability (104). SIRT-1 also appears to play a critical role in longevity as overexpression of SIRT-1 recapitulates the CR phenotype (13), and deletion of SIRT-1 abolishes the life span extension afforded by CR (12). With aging, arterial SIRT-1 expression is reduced (36), and this effect is attenuated by CR (22). Although in vivo treatment of mice with a nonspecific SIRT-1 activator, resveratrol, failed to increase life span (100), there were improvements in the vascular aging phenotype after reseveratrol treatment (28).

Interestingly, these signaling pathways do not act independently, but rather there is a large degree of interdependence between mTOR, AMPK, and SIRT-1 activities. For example, there is a reciprocal relationship between these kinases, as AMPK activation can result in direct mTOR inhibition (53), and AMPK signaling is upregulated as a consequence of reduced mTOR signaling (119). In contrast to mTOR, which is elevated in aged arteries (78), AMPK expression and activity are reduced in arteries with aging (79). Moreover, SIRT-1 directly, and negatively, interacts with mTOR, such that deficiency or inhibition of SIRT-1 results in mTOR activation (50). Cross talk also exists between SIRT-1 and AMPK, such that decreases in AMPK activity reduce SIRT-1 responsiveness to low-energy states (19). Although the mechanisms underlying the beneficial effects of these pathways on longevity and health are incompletely understood, reductions in oxidative stress and inflammation may be key players.

OTHER POTENTIAL MEDIATORS OF THE BENEFICIAL EFFECTS OF CR ON VASCULAR AGING

The nuclear receptor peroxisome proliferator-activated receptor gamma (PPARγ) may play a role in age-associated arterial dysfunction and the beneficial effects of CR. CR and PPARs regulate the expression of many common genes, suggesting that the effects of CR are mediated by PPARs, and the role of PPARs in CR and long-lived mutant mice has been previously reviewed (90). Direct evidence for a role of PPARγ in age-associated arterial dysfunction per se comes from a study in which bradykinin (BK)-induced dilation was assessed in isolated mesenteric arteries from younger and older adults that were treated in vitro with the PPARγ agonist, GW7647. Older age was associated with a reduction in vasodilation to BK that was improved after GW7647 treatment, suggesting a role for reduced PPARγ in impaired vasodilator responses in aging (4). Although PPARγ has been demonstrated to have CR-like effects on transcription (6), suggesting that PPARγ agonists may act as CR mimetics, there is also evidence for negative feedback regulation between PPARγ and the longevity gene SIRT-1. Indeed, PPARγ binds the SIRT-1 promoter negatively regulating transcription of SIRT-1 and can also inhibit SIRT-1 activity through direct interaction (54). Thus, although PPARγ activation may have some CR-mimetic transcriptional effects, it remains unclear what role this factor plays in the beneficial effects of CR on the vasculature per se.

Increased activity of the renin-angiotensin-aldosterone system (RAAS) is also associated with both oxidative stress and inflammation, processes linked to the endothelial dysfunction and arterial stiffening associated with vascular aging, and the role of RAAS in arterial stiffness (86) and the vascular aging phenotype have previously been reviewed (98). Briefly, it appears that hyperactivity of RAAS leads to arterial dysfunction by attenuating vasodilators, such as bradykinin, and increasing arterial fibrosis (98). Although it is unclear what role RAAS plays in the beneficial effects of CR in the absence of preexisting obesity, CR-associated weight loss in obese subjects leads to lower blood pressure that was associated with lower RAAS activity mediated by a reduction in renin (60). Thus, reductions in the RAAS system may also play a role in the beneficial effects of CR, and drugs targeting components of this system may be efficacious as CR mimetics.

TARGETING ENERGY-SENSING PATHWAYS TO REVERSE ARTERIAL AGING

Rapamycin

While traditionally used only as an immunosuppressant in transplant patients, rapamycin has recently been found to have many beneficial effects, including actions as a tumor suppressant in certain cancers (107), as well as being a potential therapeutic target for cardiac hypertrophy (123) and vascular restenosis (138). Inhibition of mTOR by rapamycin reduces mitochondrial ROS production in the livers of middle-aged mice (89), as well as protects the mitochondria from oxidative stress (64). In addition, rapamycin increases cardiac MnSOD expression (29) and the expression of CuZn SOD in spermatogonial stem cells (71). In addition, rapamycin also led to a downregulation of the inflammatory cytokine TNF-α in these spermatogonial stem cells from young adult mice (71). This may result from direct actions on NF-κB, as rapamycin can act on IKKβ, leading to inhibition of its phosphorylation (111, 126). Further supporting these antioxidant and anti-inflammatory effects of rapamycin, pathway analysis revealed an upregulation of free radical scavenging genes and a downregulation of NF-κB signaling genes after rapamycin treatment in adult stem cells (71).

In arteries, age-associated increases in mTOR activation were reversed by 6 wk of dietary rapamycin treatment and led to improved EDD and decreased arterial stiffness in old mice (78). Similar to CR (37), rapamycin-mediated increases in EDD were associated with increased NO bioavailability (78), and improvements in arterial stiffness were associated with a decrease in both the collagen content and abundance of advanced glycation end products, indicative of reduced collagen cross-linking. However, unlike CR, there was no effect on elastin content after rapamycin treatment in old mice (78). Furthermore, like the vascular benefits of CR that appear to be mediated, at least in part, by a reduction in oxidant production by NADPH oxidase, there was a tendency for rapamycin treatment to reduce NADPH oxidase expression that was accompanied by reduced superoxide production and a reduced abundance of arterial nitrotyrosine. Despite similarities in the reduction of oxidant production between CR and rapamycin, the role of antioxidant defenses in mediating the vascular benefits appear to differ. For example, rapamycin treatment led to an increase in arterial expression of only the CuZn isoform of SOD (78), whereas CR increased all SOD isoforms (37) (Fig. 2). Thus, although rapamycin treatment recapitulates many of the vascular benefits afforded by CR, some differences in the underlying mechanisms between these interventions do exist and may limit implications for reducing CVD risk.

Fig. 2.

Targeting energy-sensing pathways as vascular caloric restriction mimetics. Although pharmacological inhibition of mammalian target of rapamycin (mTOR) or activation of AMPK or sirtuin-1 (SIRT-1) can lead to improvements in endothelial function and/or arterial stiffness in old mice, the mechanisms underlying these beneficial effects can be disparate from that of caloric restriction. Superscripts indicate the isoform/subunit of NADPH oxidase that was examined. Arrows indicate activation, while lines ending in a circle indicate inhibition. *Denotes a tendency for decrease.

AMPK

The nonspecific AMPK activator metformin has been used extensively in the setting of diabetes (7), and recently, it was the first drug approved by the Food and Drug Administration for a clinical trial aimed at determining protective effects against a number of age-related diseases (7). In addition to the multiple beneficial effects of metformin on physiological function and disease, such as improving glucose tolerance and reducing oxidative stress and inflammation (88), direct AMPK activation by AICAR has also been shown to increase tissue antioxidant defenses, including increased skeletal muscle expression of MnSOD (16). In addition, there are multiple reports that AMPK activation can reduce inflammatory cytokines and that this is associated with blunted NF-κB signaling in a variety of tissues, including endothelial cells (58). In cultured endothelial cells, AMPK is a physiological activator of eNOS (95), suggesting that augmenting AMPK signaling may increase bioavailability of NO and enhance arterial function. Indeed, previous studies have demonstrated improvements in endothelial function in type 1 and 2 diabetic rodents (70) and reductions in aortic PWV in premenopausal women with polycystic ovary syndrome (1) in response to enhanced AMPK activity after metformin treatment. Likewise, in aged mice, augmenting AMPK activation by AICAR leads to increased EDD (79). However, increases in EDD after AICAR resulted from increased reliance on EDHF and not an increase in NO bioavailability, as occurs with CR (37) (Fig. 2). Although the effect of AICAR on arterial stiffness in an aged mouse model is not known, accelerated high-fat diet-induced arterial stiffening and elevated collagen I expression were attenuated by AICAR in a mouse model deficient in the putative antiaging gene, klotho (85). This study suggests a beneficial effect of AMPK activation on arterial stiffening in a manner consistent with changes observed in aged mice after CR, although this requires further elucidation.

SIRT-1

Beneficial effects of SIRT-1 activation by the nonspecific activator resveratrol have been reported in the setting of metabolic (11) and neurodegenerative diseases (99). Recent studies have also examined the efficacy of resveratrol to attenuate age-related vascular dysfunction with promising results. Indeed, In addition to a direct vasodilatory effect on arteries (20), treatment with resveratrol can improve vasodilation in response to other agonists, such as ACh, and this is achieved by increasing NO bioavailability in a manner dependent on AMPK (20). Furthermore, as observed after CR, increased NO bioavailability after resveratrol treatment is associated with decreased superoxide and increased MnSOD expression in the superior thyroid artery of patients with hypertension (20). In addition, SIRT-1 activation by resveratrol (15) has anti-inflammatory effects both in vitro and in vivo. In cultured cells, treatment with resveratrol reduces TNF-α and H₂O₂-induced endothelial activation via reductions in NF-κB signaling (27). However, resveratrol is known to have direct antioxidant effects, which may explain its beneficial effects on arteries, independent of SIRT-1 activation per se (30).

Nevertheless, direct evidence for a protective role of SIRT-1 has come from genetic models, such that cardiac-specific SIRT-1 overexpression leads to cardio-protection against ROS and delays age-related cardiac phenotypes (61). The anti-inflammatory effects of SIRT-1 have been supported in a genetic model of SIRT-1 deletion. Indeed, macrophage-specific deletion SIRT-1 leads to hyperacetylation of NF-κB and increased transcription of proinflammatory cytokines (118). Taken together, studies using resveratrol and genetic models have inspired the development of more specific SIRT-1 activators as potential therapeutics to treat age-related diseases, including CVD. One such small molecule activator of SIRT-1, SRT1720, has recently been shown to increase life span (92) and improve metabolic function in aged mice (91). Activation of SIRT-1 by SRT1720, has also been demonstrated to have vascular benefits in aged mice (49). In vivo treatment with SRT1720 increased arterial SIRT-1 expression and activity and improved EDD in aged mice. However, unlike CR, this beneficial effect was mediated by an increased reliance on cyclooxygenase vasodilators (49) rather than NO or EDHF. Similar to CR, SRT1720 treatment led to a reduction in arterial superoxide production that was associated with increased expression of all SOD isoforms as well as catalase, but SIRT-1 activation by SRT1720 did not impact NADPH oxidase, as was found after CR and rapamycin treatment (25, 37, 49, 78, 110). Although the effect of mTOR inhibition or AMPK activation on age-associated arterial inflammation is not known, treatment with SRT1720 reversed age-associated NF-κB activation and reduced arterial cytokine expression (49) consistent with the effects of CR on arterial inflammation (Fig. 2). Thus, SIRT-1 may also represent a viable target to recapitulate at least some of the vascular benefits of CR.

Conclusions

Although targeting mTOR, AMPK, and SIRT-1 can lead to improvements in various aspects of the vascular aging phenotype, the magnitude of improvement and mechanisms by which these improvements are achieved vary depending on intervention. Although most of the studies to date focus on EDD and arterial stiffness as measures of vascular function and measures of oxidative stress and inflammation as the primary macromechanistic processes that underlie arterial dysfunction with advancing age, other vascular functions, including angiogenesis, barrier function, and thrombosis, are also likely to be critical to cardiovascular health with advancing age. It is important to note that NO plays a role in all of these processes (14, 32, 75, 108, 109), and as such, the inability of AICAR or SRT1720 treatments to reverse the age-associated reduction in NO bioavailability may limit the overall efficacy of these treatment strategies as true CR mimetics. Still, evidence to date suggests that while CR remains the most effective strategy to improve life span and good health, targeting of critical energy-sensing, and perhaps other life span-extending, pathways may provide vascular protection that warrants further exploration.

GRANTS

This work was supported in part by Merit Review Award I01BX002151 from the United States Department of Veterans Affairs Biomedical Laboratory Research and Development Service and by the National Institute on Aging Award R01AG048366.

DISCLAIMERS

The contents do not represent the views of the US Department of Veterans Affairs, the National Institute on Aging or the United States Government.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

V.R.G., J.C., and L.A.L. edited and revised manuscript; V.R.G., J.C., and L.A.L. approved final version of manuscript; L.A.L. prepared figures; L.A.L. drafted manuscript.