Effects of regular exercise on vascular function with aging: Does sex matter?

Abstract

Vascular aging, featuring endothelial dysfunction and large elastic artery stiffening, is a major risk factor for the development of age-associated cardiovascular diseases (CVDs). Vascular aging is largely mediated by an excessive production of reactive oxygen species (ROS) and increased inflammation leading to reduced bioavailability of the vasodilatory molecule nitric oxide and remodeling of the arterial wall. Other cellular mechanisms (i.e., mitochondrial dysfunction, impaired stress response, deregulated nutrient sensing, cellular senescence), termed “hallmarks” or “pillars” of aging, may also contribute to vascular aging. Gonadal aging, which largely impacts women but also impacts some men, modulates the vascular aging process. Regular physical activity, including both aerobic and resistance exercise, is a first-line strategy for reducing CVD risk with aging. Although exercise is an effective intervention to counter vascular aging, there is considerable variation in the vascular response to exercise training with aging. Aerobic exercise improves large elastic artery stiffening in both middle-aged/older men and women and enhances endothelial function in middle-aged/older men by reducing oxidative stress and inflammation and preserving nitric oxide bioavailability; however, similar aerobic exercise training improvements are not consistently observed in estrogen-deficient postmenopausal women. Sex differences in adaptations to exercise may be related to gonadal aging and declines in estrogen in women that influence cellular-molecular mechanisms, disconnecting favorable signaling in the vasculature induced by exercise training. The present review will summarize the current state of knowledge on vascular adaptations to regular aerobic and resistance exercise with aging, the underlying mechanisms involved, and the moderating role of biological sex.

INTRODUCTION

After a 40-year decline in cardiovascular disease (CVD) mortality, deaths from CVD have increased over the past decade and remain the leading cause of death in the United States and other Western societies (1). Aging is a major risk factor for CVD, as the greatest burden of CVD exists in midlife and older adults (1). According to the US Census Bureau 2019 Estimates, the number of US adults aged 65 and older grew by over one third during the past decade and this number is projected to increase exponentially (https://www.census.gov/programs-surveys/popest.html). As such, effective interventions and therapeutic strategies are needed to minimize the ensuing societal and economic burden of CVD in midlife and older adults.

Behavioral interventions are commonly recommended before the initiation of pharmacological interventions for reducing CVD risk with aging (2). Regular exercise, including both aerobic and resistance exercise, is a well-endorsed prophylactic behavioral intervention to slow or prevent the development of age-associated chronic diseases, including CVD (2). Regular exercise improves modifiable CVD risk factors (e.g., high blood pressure and dyslipidemia); however, reductions in CVD risk factors only explains ≈50% of the lower CVD risk with regular exercise, suggesting other mechanisms are involved (3). Vascular aging, characterized by endothelial dysfunction and stiffening of large elastic arteries (e.g., aorta and common carotid), is an independent predictor of incident CVD (4, 5). Adverse changes in endothelial function and large elastic artery stiffness associated with aging are key antecedents in the pathogenesis of both subclinical and clinical CVD, underscoring vascular aging as an important therapeutic target to reduce CVD risk in aging humans (6). Moreover, these vascular aging processes are not mutually exclusive, such that endothelial dysfunction can drive the progression of large elastic artery stiffening and vice versa (7). Both aerobic and resistance exercise counter vascular aging, but there is considerable variation in the vascular response to exercise training with aging. Several intrinsic (e.g., sex, genetics, and epigenetics) and extrinsic (e.g., chronic disease, medication use and diet) factors have been identified that contribute to the heterogeneity of benefits associated with exercise in midlife and older adults (8). Indeed, prior reviews by our group have highlighted sex differences in the magnitude of vascular benefits in response to regular exercise in midlife and older adults (9–11). The reason(s) for the sex disparity are not completely understood but may relate to differences in gonadal aging, specifically declines in estrogen in women that influence cellular-molecular mechanisms (i.e., pillars of aging) disconnecting favorable signaling in the vasculature induced by exercise training. Understanding how biological sex and sex hormones contribute to the heterogeneity in exercise responses between men and women may optimize exercise interventions to reduce chronic disease risk with aging (8). In the spirit of the American Journal of Physiology-Heart and Circulatory Physiology expectations that authors consider sex as a biological variable (SABV) (12), this review will summarize the current state of knowledge on vascular adaptations to regular aerobic and resistance exercise with aging and the moderating role of biological sex. We will also discuss several mechanisms by which regular exercise supports healthy vascular aging, and how sex hormones, or lack thereof, may influence the ability of the aging vasculature to respond to exercise. Where mechanistic insight is lacking in humans, we will discuss findings from vascular aging preclinical models investigating mechanisms by which exercise transduces its benefits on the vasculature and whether there are potential differences by sex, such that these mechanisms may serve as viable therapeutic targets for optimizing exercise prescription for countering vascular aging and reducing CVD risk in both aging men and women. Note, the effects of acute exercise on vascular function with aging will not be discussed in the present review, but we refer the reader to several other reviews on this topic (13, 14).

VASCULAR ENDOTHELIAL FUNCTION

The vascular endothelium is a single layer of cells that exerts regulatory control on vascular tone and blood flow, inflammation, and thrombosis through the release of vasoprotective molecules such as nitric oxide (NO). Damage to endothelial cells results in a vasoconstrictive, proinflammatory, procoagulation, and proliferative phenotype. Endothelial dysfunction is characterized by an impairment of vascular endothelium-dependent dilation (EDD), related in part to a decrease in NO bioavailability. In humans, the reference standard technique for assessing endothelial function of conduit arteries (i.e., macrovascular) is brachial artery flow-mediated dilation (FMD), which involves ultrasound measurement of brachial artery diameter in response to reactive hyperemia induced via lower or upper arm occlusion (15). Techniques commonly used to assess resistance artery (i.e., microvascular) endothelial function include the measurement of forearm blood flow (FBF) response to pharmacological administration of endothelium-dependent agents [e.g., acetylcholine (ACh)] into an artery to promote EDD via activation of endothelial nitric oxide synthase (eNOS), which catalyzes the conversion of l-arginine to NO. Resistance vessel endothelial function can also be assessed by measuring cutaneous blood flow response to local heating, reactive hyperemia, or intradermal microdialysis delivery of ACh via laser Doppler flowmetry (16). Importantly, brachial artery FMD (17), FBF_ACh (18), and cutaneous microvascular responses to local heating (19) are predictors of future CVD events in asymptomatic adults (17), and patients with established CVD (17, 18) and end-stage renal disease (19), demonstrating their utility for assessing CVD risk and preclinical disease. In rodents, EDD is commonly assessed via pressure myography, in which an artery (e.g., common carotid) is exposed to increasing doses of ACh (20). Endothelial function can also be assessed using wire myography, in which aortic rings are suspended in a warmed physiological saline solution and passive tension is assessed throughout exposure to increasing concentrations of eNOS-activating compounds (20).

Aging Effects on Vascular Endothelial Function

Aging is associated with a decline in both macro- and microvascular endothelial function (21–23); however, there are sex differences in the rate of this decline (21, 22). Endothelial function declines steadily in men after the fourth decade of life, whereas in women, the decline occurs approximately a decade later, where it declines much more rapidly (21, 22). This rapid decline in endothelial function during midlife in women has been attributed to the loss of estrogens during menopause (21, 22, 24, 25). Although men do not experience the female equivalent of menopause, andropause or late-onset hypogonadism may contribute to age-associated endothelial dysfunction in men. We showed that healthy middle-aged/older men with low testosterone (<300 ng/dL) have greater age-associated endothelial dysfunction compared with men who have testosterone levels >400 ng/dL (26).

Preclinical and human studies have implicated oxidative stress, characterized as excessive reactive oxygen species (ROS) production relative to antioxidant defense capacity, and inflammation as key mechanisms contributing to endothelial dysfunction with aging (Fig. 1) (25, 27–34). Excess ROS impairs vascular endothelial function by scavenging NO and inhibiting its biosynthesis by disrupting eNOS function (35). Inflammatory cytokines (i.e., TNF-α) inactivate eNOS (36) and initiate production of other inflammatory cytokines, ROS, and vasoconstrictors (e.g., angiotensin II) (37, 38), thereby inducing endothelial dysfunction (39).

Figure 1.

Mechanisms of vascular aging. Endothelial function, i.e., reduced nitric oxide (NO)-mediated vasodilation, and large elastic arterial stiffening with aging are associated with increased oxidative stress, characterized as excessive reactive oxygen species (ROS) production relative to antioxidant defense capacity and low-grade inflammation. Several hallmarks or pillars of vascular aging including mitochondrial dysfunction, deregulated nutrient sensing, and cellular senescence contribute to vascular aging in part by propagating these mechanisms.

In both women and men, the decline in gonadal function and sex hormone levels appear to contribute to oxidative stress- and inflammation-mediated endothelial dysfunction with aging (Fig. 1) (25, 26, 30). Preclinical data demonstrate that physiological levels of both estrogen and testosterone mitigate ROS bioactivity (40–42) at least in part by increasing mitochondrial and cytosolic antioxidants (43–45). Estrogen and testosterone also antagonize the proinflammatory effects of cytokines (e.g., TNF-α) (46, 47), and estrogen may also inhibit inflammation and ROS by controlling the proinflammatory transcription factor nuclear factor-κB (NF-κB) (48, 49). Estrogen and testosterone may also exert their vascular-protective effects through binding to estrogen receptors and androgen receptors, respectively, in endothelial cells to initiate both genomic and nongenomic pathways to induce vasodilation (50–52).

Aerobic Exercise Effects on Endothelial Function

Cross-sectional studies of highly endurance-trained (i.e., those who participated in at least 3 days/wk of vigorous aerobic exercise) (28, 53, 54) and competitive master athletes (55–63) and longitudinal intervention studies in previously sedentary adults (54, 56, 60) have consistently shown that regular aerobic exercise attenuates or ameliorates the age-related macro- (28, 53, 54, 58, 60–63) and microvascular (55–57, 59) endothelial dysfunction in healthy midlife and older men. In contrast, the beneficial effects of aerobic exercise training on endothelial function in postmenopausal women are more variable. We (60, 64, 65) and others (66–71) have reported no significant effects of regular aerobic exercise on conduit and resistance artery endothelial function in healthy postmenopausal women, whereas others have reported beneficial effects (72–79). The inconsistent findings in postmenopausal women may be attributed to differences in methodology and blood vessel assessment, and/or characteristics of the population (80). For example, studies examining lower limb microvascular endothelial function show beneficial effects of regular aerobic exercise in postmenopausal women (76, 81), whereas studies that examined upper limb endothelial function did not (60, 64, 65). In addition, endothelial function is significantly improved with aerobic exercise training in postmenopausal women with CVD risk factors (e.g., obesity, prehypertension/hypertension) (74, 77, 82) and/or with severe endothelial dysfunction (69, 74).

Discordance in the literature may also be attributed to generalizing findings from studies that included a single sex. We are aware of only four studies that investigated the effects of regular aerobic exercise on endothelial function in healthy middle-aged/older adults that enrolled both men and postmenopausal women (60, 72, 83, 84). Two of these studies evaluated sex differences in aerobic exercise training responses and reported different findings (60, 72). Black et al. (72) reported preserved brachial artery FMD in postmenopausal women who participated in regular moderate-intensity aerobic exercise and a nonstatistically significant trend for improvements in FMD following 24 wk of moderate-intensity aerobic exercise in previously sedentary postmenopausal women. There was an absence of an exercise effect on FMD in midlife and older men (72). However, it is worth noting that there was no age-related impairment in endothelial function in these men (72). In contrast to these findings, Pierce et al. (60) found that after 8 wk of moderate-intensity aerobic exercise training (“brisk walking”), brachial artery FMD increased ≈50% in previously sedentary midlife and older men, but did not change in postmenopausal women. There were no differences between the men and women in exercise training dose or adherence. To determine whether the exercise training stimulus was insufficient to invoke improvements in endothelial function in postmenopausal women, Pierce et al. (60) assessed brachial artery FMD in a large cohort of Master endurance athletes and sedentary middle-aged/older men and postmenopausal women. Consistent with findings from their aerobic exercise intervention study, endurance-trained middle-aged/older men had higher FMD compared with age-matched sedentary peers, whereas there was no difference in FMD between endurance-trained and sedentary postmenopausal women despite having similar training volume and aerobic fitness as the men (60). Results of these studies are summarized in Fig. 2.

Figure 2.

Effects of exercise modality on vascular aging. Regular aerobic exercise enhances endothelial function with aging in men but not consistently in estrogen-deficient postmenopausal women. Regular aerobic exercise reduces large elastic artery stiffness in middle-aged and older men and postmenopausal women. The effects of resistance exercise training on endothelial function in middle-aged/older adults are not completely clear. Large elastic artery stiffness may increase with resistance exercise training in middle-aged/older men but may decrease in middle-aged/older women.

Resistance Exercise Effects on Endothelial Function

Regular resistance exercise is an important component of exercise guidelines in midlife and older adults, yet relative to aerobic exercise training, the influence of resistance training on age-related endothelial dysfunction is largely understudied. A recent meta-analysis assessing the effectiveness of resistance training for improving brachial artery FMD in adults of all ages found that, overall, resistance training is an effective approach for improving endothelial function (85). However, it is worth noting that the majority of these studies were in young adults and in midlife and older adults with cardiovascular and metabolic diseases (85). We are aware of only four studies that have examined the effects of dynamic resistance exercise on endothelial function in healthy midlife and older adults (66, 86–88). In studies where the resistance training program included both upper and lower body exercises, there were no significant improvements in brachial artery FMD (66, 87, 88) or popliteal artery FMD (87). Interestingly, in a study conducted in postmenopausal women, brachial artery FMD was improved following resistance training that exclusively used the legs despite the women having normal endothelial function (i.e., mean baseline FMD was 9.2%) before starting the resistance training program (86). Whether the type of resistance exercises (e.g., upper vs. lower body, isokinetic machine based, low intensity, progressive, etc.) impose different adaptive stimuli on endothelial function warrants further investigation.

Further investigation is also needed on whether there are sex differences in resistance exercise training responses on endothelial function. In the only study that enrolled both midlife/older men and postmenopausal women, there was no effect of resistance exercise training on FMD in either the men or the women (87). However, because SABV was not evaluated in this study, it is unknown whether the resistance training responses were greater in one sex versus the other. Results of these studies are summarized in Fig. 2.

LARGE ELASTIC ARTERY STIFFNESS

The major functions of the large elastic arteries (i.e., aorta, carotid) are to buffer the rise in systolic pressure generated by the left ventricle during cardiac contraction by storing a portion of the ejected stroke volume and slowly conducting continuous blood flow to systemic circulation (89). The buffering capacity of the large arteries is facilitated by the intricately organized extracellular matrix proteins (e.g., elastin, collagen, etc.) and arterial geometry (90). In humans, the reference standard noninvasive method of measuring large elastic artery stiffness is carotid-femoral pulse wave velocity (PWV), which is an indicator of the speed of the pulse wave generated by left ventricular contraction through the aorta, with higher values indicating greater stiffness (91, 92). Importantly, carotid-femoral PWV is an independent predictor of cardiovascular events (4, 93). Local carotid artery stiffness can also be measured using a combination of ultrasonography of carotid artery diameter changes with systole and diastole and changes in central arterial pressure (91, 94). Similar to humans, in preclinical models, large elastic artery (e.g., aorta) stiffness can be assessed in vivo via aortic PWV (i.e., PWV between the aortic arch and abdominal aorta) (95), as well as by measuring carotid artery diameter in response to incremental increases in luminal pressure (96). Aortic stiffness can also be assessed ex vivo in isolated aortic segments, in which aorta rings undergo stress-strain testing to determine the intrinsic mechanical stiffness of the arterial wall (i.e., elastic modulus) (95).

Aging Effects on Large Elastic Artery Stiffness

Advancing age is associated with stiffening of the large elastic arteries mediated by structural changes in the arterial wall and functional changes that increase vascular smooth muscle cell tone (6, 97). Specifically, the extracellular matrix of the artery wall remodels (i.e., increased collagen deposition and degradation and fragmentation of elastin) and there is an upregulation of advanced glycation end products (AGEs) that cross-link collagen fibers to increase stiffness (6, 96, 97). Moreover, vessel wall stiffening impairs the mechanical properties of endothelial cells that are sensitive to changes in cyclic stretch and increases nonlaminar flow that upregulates expression of inflammatory cytokines to increase the generation of ROS and reduce NO bioavailability, consequently increasing vascular smooth muscle cell tone (Fig. 1) (98). An increase in the intrinsic stiffness of vascular smooth muscle cells may also contribute to age-related aortic stiffening (99). Alterations in key signaling pathways, including the renin-angiotensin II, aldosterone, endothelin-1, and sympathetic nervous system activity, also contribute to age-associated arterial stiffening (98, 100, 101).

The effect of biological sex on the stiffness of large elastic arteries with aging remains unclear. Although some cross-sectional studies demonstrate no sex differences in carotid-femoral PWV with advancing age (102–104), sex differences in local (i.e., carotid) artery stiffness emerge when lumen diameter is considered, with carotid artery stiffening increasing more rapidly around age 45 yr in women compared with men, likely reflecting sex differences in arterial geometry (103, 105). Gonadal aging and changes in sex hormones (10, 90) contribute to age-associated large elastic artery stiffening in both women and men, and at least in women, the increase in arterial stiffening is related, in part, to increased oxidative stress and inflammation (i.e., TNF-α) (30, 106, 107). In women, carotid-femoral PWV increases during the menopause transition (108) and age-associated reductions in carotid artery compliance (inverse of stiffness) are attenuated in postmenopausal using estrogen-based hormone therapy (109). In midlife and older men, low testosterone is associated with greater carotid artery stiffening (110).

Aerobic Exercise Effects on Large Elastic Artery Stiffness

Aerobic exercise is an effective strategy for slowing or improving age-associated large artery stiffening in both midlife/older women and men (92, 94, 109, 111–113). In cross-sectional studies, the age-related increase in large artery stiffening, measured by carotid-femoral PWV, is prevented in habitual endurance-trained older men (114) and postmenopausal women (92). Similarly, endurance-trained midlife and older men and postmenopausal women have lower carotid artery stiffening and greater carotid artery compliance compared with their age-matched sedentary peers (94, 112, 113, 115). The lifelong (>25 yr) “dose” (i.e., frequency of exercise sessions >30 min) of aerobic exercise needed to attenuate age-associated carotid artery stiffness in older adults was shown to be 2 to 3 bouts/wk (113); however, higher doses (i.e., ≥4 bouts/wk of >30 min/bout) may be required to prevent age-related aortic stiffening in older adults (113). Whether the exercise dose of lifelong exercise or exercise training prescription is the same for men and women is unclear. We have shown that age-associated carotid artery stiffness is attenuated in estrogen-deficient postmenopausal women who were performing regular aerobic exercise (primarily running) an average of 5 days/wk for at least 5 years (109). Importantly, the benefits of aerobic exercise can be observed even when initiated at midlife or later. Intervention studies of moderate- to high-intensity aerobic exercise improve large artery stiffness in previously sedentary middle-aged/older men and postmenopausal women (94, 109, 116–118). Results of these studies are summarized in Fig. 2.

Resistance Exercise Effects on Large Elastic Artery Stiffness

In contrast to the beneficial effects of aerobic exercise on large artery stiffness, habitual resistance exercise may not exert beneficial effects in middle-aged/older adults and may even augment age-associated large artery stiffening. In a cross-sectional study of healthy middle-aged/older men, age-associated reductions in carotid artery compliance were greater in those who had performed high-intensity resistance training for >2 years compared with age-matched sedentary men (119). The degree of carotid artery stiffening was inversely correlated with the number of years of resistance training and independent of carotid systolic blood pressure (119). These data are consistent with the findings of a systematic review and meta-analysis of randomized controlled trials demonstrating that high-intensity resistance training alone is associated with increased large elastic arterial stiffness (120). However, this association was only observed in young adults (120). Moreover, moderate-intensity resistance training was not associated with changes in arterial stiffness in middle-aged/older adults (120). Interestingly, in a systematic review and meta-analysis in postmenopausal women, low-to-moderate intensity resistance training was associated with an overall decrease in aortic PWV (111); however, most of the studies evaluated were conducted in postmenopausal women with hypertension. In studies that examined the effects of resistance training on arterial stiffness in normotensive postmenopausal women, there were no changes in large artery stiffness as measured by carotid-femoral PWV (86). The only study that we are aware of that examined sex differences in changes in arterial stiffness following resistance training was conducted in middle-aged/older men with prehypertension/hypertension and postmenopausal women (121). Resistance training produced no significant changes in carotid-femoral PWV in the postmenopausal women but by contrast, increased PWV in men (121). Results of these studies are summarized in Fig. 2.

ROLE OF GONADAL AGING AS A MODULATOR OF VASCULAR ADAPTATIONS TO EXERCISE TRAINING WITH AGING

As described above, the loss of gonadal function appears to influence vascular aging in both women and men in a sex-specific manner (24–26); however, gonadal failure (i.e., menopause) is inevitable in women in midlife but rare in men until much later in life. Therefore, because gonadal failure and subsequent estrogen deficiency has been shown to be the triggering event for the development of endothelial dysfunction with aging in women (25), the inconsistent benefits of aerobic exercise on endothelial function in postmenopausal women may be related, in part, to estrogen deficiency. In support of this, we previously reported that brachial artery FMD increased following 12 wk of moderate-intensity aerobic exercise in previously sedentary postmenopausal women who were treated with either oral or transdermal estradiol, but not in women treated with placebo (64), consistent with our previous observations in estrogen-deficient postmenopausal women (60). These findings provided the first direct evidence that estrogen may be necessary to induce the beneficial effects of aerobic exercise on endothelial function in postmenopausal women. Other evidence to support that estrogen plays a permissive role in endothelial adaptations to aerobic exercise in women comes from studies conducted in amenorrheic premenopausal athletes and acute aerobic exercise studies from our laboratory (122) and others (123–125). First, highly trained premenopausal amenorrheic athletes have reduced FMD compared with eumenorrheic athletes and sedentary controls, and FMD is restored to eumenorrheic athletic levels with recovery of menses or treatment with oral contraceptives (123, 124). Second, we showed that FMD is enhanced following a single bout of moderate-intensity aerobic exercise in postmenopausal women treated with acute transdermal estradiol but not in women treated with placebo (122).

Although the loss of estrogen does not appear to be the triggering event in age-associated arterial stiffening in women, estrogen may be important for transducing aerobic exercise benefits on arterial stiffness. We previously demonstrated that carotid artery compliance can be restored to premenopausal levels with moderate-intensity aerobic exercise in postmenopausal women who were chronically using estrogen-based hormone therapy (109). Collectively, these findings provide support for the importance of estrogen for transducing aerobic exercise vascular adaptations in women.

The impact of gonadal aging and exercise training benefits in postmenopausal women is also the basis for the “exercise timing hypothesis” that suggests that regular aerobic exercise is only beneficial if initiated soon after menopause rather than many years later (126). Leg microvascular function was improved following intense aerobic exercise training in early (<5 years since final menstrual period) postmenopausal women compared with older postmenopausal women (76, 126). The reasons for this are not completely understood, but it was speculated that changes in estrogen levels and estrogen receptors with time after menopause could influence the effect of exercise in postmenopausal women (126). In our intervention studies (60, 64), controlling for the time since the onset of menopause did not influence our findings of a lack of exercise training adaptations on endothelial function in healthy postmenopausal women. Moreover, findings from our cross-sectional studies of Masters endurance athletes described above, suggest that any benefits of habitual aerobic exercise observed in the early postmenopausal years are not retained into the late postmenopausal period (60, 64, 65). Future investigations are needed to understand if the effects of exercise training on vascular function decline gradually over time following the onset of menopause or cease more abruptly at an unknown menopausal age (126).

Whether sex hormones play a similar permissive role in aerobic exercise training benefits on the aging vasculature in men is unclear. We have shown that the improvement in brachial artery FMD with 12 mo of progressive resistance training in healthy older men with low-normal testosterone levels (average, ≈290 ng/dL) is only observed in men randomized to testosterone supplementation and not those receiving placebo (88). In contrast, Chasland et al. (127) reported improvements in brachial artery FMD following 12 wk of circuit exercise training consisting of resistance and aerobic (cycling) exercise in middle-aged/older men with low-to-normal testosterone levels (average, ≈320 ng/dL) treated with placebo but not in men treated with testosterone. The discrepancy in findings could be related to the type of exercise training (i.e., resistance exercise alone vs. combined resistance and aerobic exercise) or differences in gonadal status of the study populations. In our study (88), the average baseline testosterone level was consistent with testosterone deficiency (i.e., serum testosterone <300 ng/dL) as defined by the American Urology Association (128), whereas the average baseline testosterone of men enrolled in the study by Chasland et al. (127) was within the normal range (128). Future studies should examine whether there is a threshold level of endogenous testosterone at which vascular adaptations to exercise training disappear. Results of these studies are summarized in Fig. 3.

Figure 3.

Influence of sex hormones on vascular adaptations to regular aerobic and resistance exercise in men and women. Sex hormones regulate vascular aging in men and women and thus may modulate the effects of regular exercise on vascular function in aging men and women. Regular aerobic exercise consistently improves vascular function in aging men by ameliorating oxidative stress; however, if testosterone is required to transduce exercise signals to the vasculature is unclear. In contrast, estradiol appears to be required to transduce exercise signals to improve oxidative stress-related suppression of endothelial function, as aerobic exercise has no obvious effect on oxidative stress-mediated endothelial dysfunction in estrogen-deficient postmenopausal women. Estradiol may also enhance exercise-training benefits on large elastic artery stiffening in women but this warrants further study. It is unclear if sex hormones modulate the effects of resistance exercise training on vascular function in both men and women; however, testosterone therapy may improve endothelial function following regular resistance training in middle-aged/older men with low-to-normal testosterone levels. +/?, weak evidence that sex hormones may modulate exercise effects but is unclear; ?, no studies have examined.

MECHANISMS BY WHICH REGULAR EXERCISE EXERTS BENEFICIAL EFFECTS ON VASCULAR FUNCTION WITH AGING

The mechanism by which regular exercise counters vascular aging is not completely understood but is likely related to modulating oxidative stress, inflammation, and other key molecular pillars (e.g., mitochondrial dysfunction and dysregulated stress response, nutrient sensing, cellular senescence) that have been identified as drivers of vascular aging (see Figs. 1 and 4) (130, 131). Because of the interconnectedness between these pillars, exercise amelioration of one pillar (e.g., dysregulated stress response) may depend on amelioration or dampening of others (e.g., mitochondrial dysfunction) (130). Below, we discuss potential pillars of aging that appear to be involved in improvements in age-related vascular dysfunction with exercise training, and note where sex differences may exist. As data on the molecular mechanisms underlying resistance exercise training effects on vascular aging are limited, particularly in humans, the discussion will focus primarily on molecular mechanisms associated with aerobic exercise training effects.

Figure 4.

Pharmacological approaches to target specific molecular pillars of aging to transduce exercise-mediated benefits on the vasculature. Key molecular pillars have been identified that drive the vascular aging process, including mitochondrial dysfunction and dysregulated stress response, inflammation, dysregulated energy and nutrient sensing, and cellular senescence. Although regular exercise can modulate these pillars, there may be some pillars that are resistant to exercise effects and because the pillars are interconnected, failure to elicit exercise effects in one pillar may impact the ability to target others. Thus, several pharmacological compounds may be able to target specific pillars to transduce exercise-mediated benefits on the vasculature. These hypotheses are illustrated with dashed lines. This figure was adapted and modified from Goh et al. (129) with permission.

Mitochondrial Dysfunction and Dysregulated Stress Response

Aging is associated with a decline in stress resistance, an adaptive mechanism that activates a network of cellular processes to counter various biological insults or stressors (e.g., reactive oxygen species, inflammation, radiation, chemicals, toxins) and improve cell survival (129). Mitochondria are critical for mediating the cellular response to stress and for maintaining cellular homeostasis (132). In the vasculature, mitochondrial networks play vital roles in energy metabolism, intracellular signaling, cellular redox homeostasis, and regulation of programmed cell death (133). With aging, mitochondrial networks become fragmented because of alterations in biogenesis and in the homeostatic regulation of mitochondria dynamics (e.g., reduced fusion and increased fission or vice versa) that are important for maintaining mitochondrial quality and mass (134). In endothelial cells, fragmented mitochondrial networks promote uncoupled respiration and increased ROS production (e.g., superoxide) (135). We (136–139) and others (140–143) showed in preclinical studies in male mice and clinical studies in middle-aged/older men and postmenopausal women that mitochondrial dysfunction (i.e., dysregulated mitochondrial stress response, reduced arterial protein expression of mitochondrial signaling/biogenesis and antioxidants) and excess mitochondrial-mediated oxidative stress contribute to age-related endothelial dysfunction and large elastic artery stiffening.

One of the benefits of aerobic exercise in countering oxidative stress and inflammation-associated vascular aging may be through enhanced mitochondrial function and stress resistance. During acute exercise, ROS production is increased from aerobic bioenergetic reactions in the mitochondria and cytosol of the active skeletal muscle (144). These exercise-induced ROS are crucial signaling molecules that promote adaptive mechanisms to exercise training, including enhanced mitochondrial health characterized by increased mitochondrial volume, the abundance of oxidative enzymes to increase ATP production, and antioxidant defenses (144). Chronic aerobic exercise training thus enhances mitochondrial function and antioxidant defense systems, effectively lowering ROS bioactivity and increasing resistance to oxidative damage in the vasculature (144–146).

Consistent with this, in preclinical studies, we have shown that the beneficial effects of 10–14 wk aerobic exercise training (i.e., voluntary wheel running) on NO bioavailability and endothelial function of large elastic arteries (i.e., carotid) in older male mice were associated with suppression of NADPH oxidase-derived and mitochondria-mediated excessive superoxide-related oxidative stress, inflammation and improved arterial mitochondrial health (i.e., mitochondrial biogenesis, dynamics, oxidative phosphorylation, antioxidants), and arterial resilience to acute mitochondria and metabolic stress (145, 147, 148). As an extension of these findings, we have shown that lifelong voluntary wheel running can prevent the decline in endothelial function and large elastic artery stiffening throughout the lifespan by preserving NO bioavailability and preventing excessive mitochondrial superoxide-related oxidative stress, which was related to lower vascular inflammation (149).

Observations from clinical studies conducted in sedentary and Masters endurance-trained men are consistent with these preclinical findings of enhanced mitochondrial health and antioxidant defenses, reduced ROS, and increased resistance to oxidative stress. Protein expression of nitrotyrosine (oxidative footprint), NADPH oxidase p47^phox (prooxidant enzyme), and NF-κB measured in biopsied endothelial cells from the brachial artery were higher in sedentary middle-aged/older men compared with young sedentary and endurance-trained middle-aged/older men, with the latter two groups not differing (150). The middle-aged/older sedentary men also had reduced endothelial expression of the mitochondrial antioxidant MnSOD, whereas MnSOD expression was preserved in endurance-trained middle-aged/older men (150). In addition, we have shown that infusion of vitamin C increased brachial artery FMD in middle-aged/older sedentary men but had no effect in endurance-trained men, indicating that aerobic exercise training improved endothelial function by mitigating tonic excessive ROS-related suppression of endothelial function (28).

To our knowledge, no data are available regarding aerobic exercise training effects on vascular mitochondrial function or stress response in middle-aged/older females. However, clinical data from our group suggest that mitochondrial adaptations and stress response with exercise may be impaired in estrogen-deficient postmenopausal women. First, in our aforementioned studies demonstrating no beneficial effects of regular aerobic exercise on endothelial function in estrogen-deficient postmenopausal women (64), vitamin C infusion increased FMD in Masters endurance-trained postmenopausal women and in previously sedentary women treated with daily placebo after 12 wk of aerobic exercise training indicating ROS-related suppression of FMD, in contrast to the findings discussed above in Masters endurance-trained middle-aged/older men (28). Contrary to placebo-treated postmenopausal women, there was no effect of vitamin C infusion on FMD after 12 wk of aerobic exercise training in both oral and transdermal estradiol-treated postmenopausal women who observed vascular improvements with exercise training (64). These data suggest that in the absence of high circulating estrogen, aerobic exercise training has no obvious effect on vascular mitochondrial antioxidants and oxidative stress and its tonic suppression of endothelial function in postmenopausal women, whereas in the presence of high-circulating estrogen_, oxidative stress-related suppression of endothelial function is ameliorated with aerobic exercise training. It is possible that because estrogen plays a critical role in regulating vascular mitochondrial health (biogenesis, antioxidants), stress resistance, and ROS production (43, 45, 151–153), vascular antioxidants may have been enhanced following aerobic exercise training in the estradiol-treated postmenopausal women. Consistent with this theory, in Wistar rats with intact ovaries (high estrogen), 90 days of swim training increased the antioxidants SOD and glutathione peroxidase activity and glutathione content, and reduced lipoperoxidation in erythrocytes and liver tissue; there was no change in ovariectomized (low estrogen) rats (154), suggesting that estrogen deficiency hinders antioxidant adaptations to exercise training.

Collectively, these data suggest sex differences in the responsiveness and adaptability of endogenous antioxidant enzyme systems to oxidant exposures, where in contrast to middle-aged/older men, due to estrogen deficiency, postmenopausal women appear to have a redox imbalance stemming from excessive ROS that negates positive adaptations to exercise training. Whether testosterone (or estrogen) in aging men is important for exercise adaptive responses to mitochondrial health, stress response, and the mitigation of vascular oxidative stress and inflammation is unknown.

Deregulated Nutrient Sensing

Aging is associated with a deregulation in several key energy- and nutrient-sensing pathways that are important for cellular homeostasis, including reduced nicotinamide adenine dinucleotide (NAD⁺) bioavailability and expression and activation of sirtuin-1, increased expression and activation of mammalian target of rapamycin (mTOR), and reduced activity of adenosine monophosphate-activated protein kinase (AMPK). In studies in mice (155–158) and humans (156, 159), we have shown that age-related reductions in NAD⁺ bioavailability and sirtuin-1 are linked to endothelial dysfunction and large elastic artery stiffness and that targeting these pathways can counter these components of vascular aging. In old male mice, supplementation with a NAD⁺-boosting compound (nicotinamide mononucleotide) reversed age-associated endothelial dysfunction and large elastic artery stiffening by ameliorating excessive oxidative stress, restoring sirtuin-1, increasing mitochondrial antioxidants (MnSOD), and decreasing collagen and increasing elastin (155). In humans, we showed that 6 wk of oral supplementation with the NAD⁺-boosting compound nicotinamide riboside can increase circulating NAD⁺ levels and reduce PWV in older men and women; however, there was no effect on FMD, and we did not assess sex differences (159).

To date, no study that we are aware of has assessed whether the effect of aerobic exercise training on vascular function with aging is related to improvements in pathways associated with nutrient sensing. However, aerobic and resistance exercise training was shown to increase NAD⁺ bioavailability in skeletal muscle in middle-aged/older men and women (160), although sex differences were not evaluated. Future studies should determine if improvements in vascular function with aging are mediated by improvements in energy- and nutrient-sensing pathways and whether there are sex differences in this adaptation.

Cellular Senescence

Cellular senescence is a state of essentially irreversible cell cycle arrest that is characterized by proinflammatory senescence-associated secretory phenotype (SASP). Under normal physiological conditions, senescence is beneficial in maintaining tissue health (e.g., inhibiting neoplastic growth) (161); however, with aging, there is an accumulation of senescent cells in tissues leading to adverse cellular and physiological function and chronic diseases, in part by increasing inflammation and oxidative stress (162). In turn, these cells secrete a variety of proinflammatory factors through the SASP, which can induce senescence in neighboring cells, further exacerbating the physiological burden of cellular senescence (163).

Accumulating evidence suggests that endothelial and vascular smooth muscle cell senescence contribute to vascular dysfunction with aging (164, 165). Importantly, we have reported that aerobic exercise training may preserve endothelial function with aging, in part by diminishing cellular senescence (165). Sedentary middle-aged/older adults had elevated markers of cellular senescence (e.g., p53, p21, and p16) in arterial endothelial cells compared with young adults, which were inversely associated with endothelial dysfunction (i.e., greater arterial endothelial cellular senescence was associated with lower FMD), whereas habitually aerobic exercise-trained middle-aged/older men and women had levels of arterial endothelial cell senescence similar to young adults (165). As we did not assess SABV in this study, future studies are warranted to determine whether gonadal status, cellular senescence, and vascular function are related.

RESEARCH GAPS

Below, we list several important research gaps and areas of future investigation to advance our understanding of the effects of exercise on large elastic artery stiffening and vascular endothelial dysfunction with aging.

Resistance Training

Resistance exercise training improves vascular endothelial function in young adults; however, there are limited data on the effectiveness of resistance exercise training, or underlying mechanisms, on endothelial function in middle-aged/older adults. Large randomized controlled resistance training interventions in middle-aged/older men and women are needed to determine if measures of endothelial function (macro- and microvascular endothelial function) and arterial stiffness (both PWV and carotid artery compliance) are favorably or adversely changed, or unaffected.

Considering SABV in Preclinical and Clinical Studies

The majority of the available evidence on mechanisms underlying exercise training effects on vascular aging is based on preclinical studies using older male animals and clinical studies in humans using mostly middle-aged/older men, or men and women combined without addressing SABV. There is a clear need to perform studies in female rodents and postmenopausal women to better elucidate the mechanism(s) mediating the differential responses of exercise training on age-related vascular dysfunction across both sexes. To improve the cohesiveness of the literature, future exercise training studies would do well to include evaluation of SABV as part of the a priori study design.

Changes in Gonadal Function with Aging

Gonadal aging contributes to the heterogeneity in the biology of aging. Future studies (both preclinical and human) should investigate whether gonadal aging and sex hormone deficiency influence the ability of cells to mediate exercise training adaptations on vascular aging and the mechanisms mediating the effects of sex hormones (e.g., estrogen, testosterone). Moreover, the menopause transition (i.e., perimenopause) is a critical timepoint in which vascular function is influenced in women (25, 108) but currently there is no preclinical literature regarding the influence of exercise training on vascular health (and underlying mechanisms) in perimenopausal female animals. Prior studies have used OVX models to experimentally induce a menopause-like state of estrogen loss in female animals, which does not represent the human condition (i.e., loss of ovarian function coupled with advanced age). Thus, use of the 4-vinylcyclohexene-diepoxide animal model, which allows for progressive follicular exhaustion and ultimately follicular failure in female rodents, could overcome limitations of OVX animal models (166).

Targeting the Pillars of Aging

Because the pillars of aging are an interconnected network, failure to elicit exercise effects in one pillar may impact the ability to target others (130). Future studies could investigate pharmacological approaches to target specific pillars to transduce exercise-mediated benefits on the vasculature, particularly in middle-aged/older adults who may be resistant to improvements (e.g., estrogen-deficient postmenopausal women) (Fig. 4). For example, because mitochondria may act as a central hub for integrating exercise-induced signals for vascular benefits (167), the administration of a commercially available mitochondrial-targeted antioxidant (e.g., MitoQ) could restore the ability of aerobic exercise training to enhance mitochondrial health, upregulate antioxidant defense systems, reduce ROS production, oxidative stress, and inflammation, and improve NO-mediated endothelial function (136, 137). Alternative approaches to restore or optimize exercise training benefits that could be employed that target other pillars of aging that may be resistant to exercise training, include NAD⁺-boosting supplements (e.g., NR), natural senolytics (compounds that target and clear excess senescent cells) such as fisetin and/or anti-inflammatory compounds like curcumin.

Considering the Role of Estrogen

Finally, estrogen and exercise share common intracellular signaling pathways to mediate NO release, as such it is plausible that the estrogen treatment modulates exercise-induced increases in shear stress-related signaling to activate eNOS and increase NO production and vasodilation (11, 64). Exercise-mediated elevations in blood flow increase shear stress along the endothelial surface, stimulating mechanosensors [i.e., integrins, caveolae, ion channels, G-coupled protein receptors (GPCRs)] that transduce mechanical forces into biochemical signals to phosphorylate and activate eNOS and increase NO (168). Estrogen activates eNOS and increases NO via estrogen receptor (ER)α-mediated nongenomic signaling pathways involving caveloae, integrins, ion channels, and GPCRs (50). Future studies should examine if estrogen or testosterone modulates exercise-induced intracellular signaling pathways and subsequent vascular adaptations.

SUMMARY

Vascular aging, featuring endothelial dysfunction and large elastic artery stiffening, is the major risk factor for CVD. As such, effective evidence-based treatments and/or prevention strategies are needed to mitigate the harmful effects of vascular aging. Regular exercise confers pleiotropic benefits across multiple organ systems including the cardiovascular system and is a first-line strategy for primary prevention of CVD. We have reviewed evidence from preclinical and clinical studies demonstrating that regular exercise, primarily aerobic, is effective at countering vascular aging through targeting several pillars of aging. However, not all middle-aged/older adults reap the same benefits of exercise in an equivalent manner (8, 9). As discussed, some studies (60, 64–71), but not all (72–79), report diminished or an absence of vascular adaptations to exercise training in estrogen-deficient postmenopausal women, particularly endothelium-specific benefits. The reasons for the exercise response variation in age-associated vascular function between the sexes are not completely understood. We highlighted several important research gaps to address in future research studies to advance our understanding of the effects of exercise on vascular aging. For example, including factors that contribute to sex differences in exercise benefits on vascular aging, and potential ways to optimize the vascular health-promoting effects of exercise for all middle-aged/older adults, particularly those who may be resistant (i.e., nonresponders) to exercise adaptations should be considered.

GRANTS

Work from the author’s research was supported by the following National Institutes of Health Awards: R01 AG027678 (to K.L.M.), R56 AG072094 (to K.L.M.), R01 AG049762 (to K.L.M.), F32 HL151022 (to Z.S.C.), K99 HL159241 (to Z.S.C.), F32 AG071273 (to L.E.D.), K12 AR084226 (to L.E.D.), T32 AG000279 (to R.R.), R01 AG055822 (to D.R.S.), R01 AG0661514 (to D.R.S.), R01 AG071506 (to D.R.S.), R01 AG073117 (to D.R.S.), R01 AG066730 (to D.R.S.), and R21 AG078408 (to D.R.S.) and Eastern Colorado VA GRECC.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

K.L.M., Z.S.C., L.E.D., and R.R. prepared figures; K.L.M., Z.S.C., L.E.D., and R.R. drafted manuscript; K.L.M., Z.S.C., L.E.D., R.R., and D.R.S. edited and revised manuscript; K.L.M., Z.S.C., L.E.D., R.R., and D.R.S. approved final version of manuscript.