Increasing the Health Span: Unique Role for Exercise

Abstract

Health span, that period between birth and onset of major disease(s), when adequate physical and cognitive function permit those daily living activities essential to life quality, is lower in the United States than other developed countries. Physical inactivity and excessive calorie intake occupy dominant roles both in the problem, and by redressing them, in the solution. Consequently, this review focuses on evidence that appropriate exercise engagement and calorie restriction can improve physical and mental health with a view to extending the health span. Humanity, writ large, has grasped these underlying concepts for Millennia but has been largely intransigent to them. Thus, the final section proposes a novel Monty Python-esque approach that encompasses humanity’s inimical sense of humor to increase physical fitness and mental health, restore energy balance, sustain better cognitive function and extend the health span.

Graphical Abstract

Introduction

“If exercise could be packed into a pill, it would be the single most widely prescribed, and beneficial, medicine in the nation” (1). This quote, by Robert N. Butler, is supported today by the compelling weight of scientific evidence that physical exercise and/or dietary intervention have the greatest therapeutic potential to decrease the risk of most chronic disease (rev. 2) and thus increase the health span.

The problem:

One of humanity’s cardinal achievements over the past century or so is the 3-decade increase in life expectancy at birth (3), pushing humanity’s average age well beyond historical longevity (4). Unfortunately, saving the lives of patients with heart disease, cancer and diabetes has enhanced morbidity and extended the years living with chronic disease. This failure of success means that the average health span is 66.1 years with 16–20% of a 77–85-year lifespan spent in aged morbidity (3,5,6), contributing disproportionately to the $4 trillion+ annual health care costs (7) in the United States (U.S.).

One principal basis for this stark reality is that, in the 21^st Century, over 80% of adolescents and adults worldwide do not meet the physical activity guidelines for aerobic and muscle strengthening activities (8). Close to 90% of American adults are metabolically unhealthy, with 60% enduring at least 1 chronic disease and close to 50% having multiple conditions (rev. 7). A sedentary lifestyle accelerates the age-related decline of cardiorespiratory fitness (CRF, best measured as the maximum oxygen uptake, $\dot{\text{V}}{\text{O}}_2\text{max}$) by 3 decades (9), with physical inactivity initiating a cluster of diseases or “diseasome” (10) of at least 35 pathological/clinical conditions (see Table 1, ref. 11). The public health scientist Steven Blair regarded physical inactivity as this century’s biggest health problem and it is worsening (12). For instance, the U.S. Centers for Disease Control and Prevention (CDC) predicts that Type 2 diabetes (T2D) will afflict one in three U.S. adults by 2050 (rev. 2).

Table 1.

List of pathological and clinical conditions initiated/exacerbated by physical inactivity (compiled from Booth et al. ref. 11).

Accelerated biological aging /premature death Arterial dyslipidemia Bone fracture/falls Breast cancer Chronic pain Cognitive dysfunction Colon cancer Congestive heart failure Constipation Coronary heart disease Deep vein thrombosis Depression and Anxiety Diverticulitis Endometrial cancer Endothelial dysfunction Erectile dysfunction Gallbladder diseases Gestational diabetes Hemostasis Hypertension Insulin resistance Low cardiorespiratory fitness ($\dot{\text{V}}{\text{O}}_2\text{max}$) Obesity Osteoporosis Osteoarthritis Metabolic syndrome Nonalcoholic fatty liver disease Peripheral artery disease Prediabetes Pre-eclampsia Polycystic ovary syndrome Poor balance Rheumatoid arthritis Sarcopenia Stroke Type 2 diabetes

Sadly, so-called Precision Medicine initiatives have singularly downplayed the role of exercise as medicine despite the fact that it fits its tagline “…the right drug at the right dose to the right patient” (13). In contrast to the goal of modern medicine, which is to extend longevity, that of aging science seeks to restrict time spent in morbidity by expanding the health span (14). Hunter-gatherers, perhaps 40 millennia ago, who survived infancy and childhood, lived on average, seven decades (15). Today the Hadza and other hunter-gatherers, in contrast to modern Westerners, have relatively little decline in their walking speeds, grip strength and $\dot{\text{V}}{\text{O}}_2\text{max}$ as they age (rev. 16); supporting that prolongation of morbidity in old age is largely the creation of modern medicine and lifestyle (3,17).

Health span is not just the absence of major diseases, it is sustaining adequate physical and cognitive function to enjoy our environment and engage in those activities of daily living essential to life quality (18). Thus, by combining elements of both morbidity and mortality, the concept of health span embodies population health (19) but presently lacks rigorously validated metrics (20). Nevertheless, the goal of extending and improving the health span should be a primary public health initiative. Accordingly, for the optimal ‘ideal survivorship curve’ hypothesis (Figure 1; ref. 21) physically active lifestyles would telescope morbidity into the very last years/months of life (2,22,23).

Figure 1.

Schematic depicting the current U.S. health span which, at 66.1 years, lags behind that of most developed countries and sentences the majority of Americans to well over a decade of morbidity presaging the increased likelihood of dependent care when the maximal oxygen uptake ($\dot{\text{V}}{\text{O}}_2\text{max}$) falls below ~20 ml · kg⁻¹ · min⁻¹ (black curve). This review explores the evidence for, and potential of, exercise and calorie restriction to extend the health span (grey arrow, brown curve) such that the period of morbidity is reduced to just a few months or years immediately preceding death. Please note that this is a hypothetical schematic where age should not be necessarily regarded as a continuous variable. Moreover, it is to be expected that substantial variation exists across sexes, ethnic groups, genetic predisposition and disparate lifestyle factors (e.g., dietary patterns, exercise type and characteristics, family income, smoking, drug use) of the individual, most of which have yet to be defined in the context of health span and, as such, are beyond the scope of this review. See text for more details.

Thus, especially for older adults, it is most important to embrace a lifestyle that permits sustained independence, for which a minimal $\dot{\text{V}}{\text{O}}_2\text{max}$ in the range of 15 −20 ml · kg⁻¹ · min⁻¹ has been proposed (e.g., 24; rev. 25). In modern society, unlike the Hadza hunter-gatherer lifestyle that involves routine physical activity, prescribed exercise training is required to preserve aerobic capacity and strength.

The solution and associated evidence:

In keeping with the central role of exercise in extending health span, $\dot{\text{V}}{\text{O}}_2\text{max}$ is a strong and independent predictor of all-cause mortality (26). A high $\dot{\text{V}}{\text{O}}_2\text{max}$ is the most important factor delaying all-cause mortality and the onset of chronic diseases, in particular cardiovascular, metabolic, and cancer (27). In lifelong runners/exercisers, disability was postponed by 14–16 years compared with controls (rev. 28), with the most crucial risk factor being lack of exercise. In addition, adults who exercise 3 or more times per week had a 32% lower risk of dementia (29). Thus, increased physical activity may extend life expectancy several years for men and women (rev. 11) but, germane to this review, more active individuals spend fewer years disabled (30).

Exercise adaptations deliver a greater therapeutic index (i.e., benefit/side effect ratio) than afforded by any drug therapy (11) and the Physical Activity Guidelines for Americans (31) estimates that physical activity decreases mortality by 30–40%, equating to 720,000 deaths annually in the U.S. The evidence is clear that physical activity is required to maximize health span (and lifespan), with 3–5 times the recommended physical activity (450–750 min/week) reaching the maximum achievable health span benefit from exercise (32).

Mechanisms:

Understanding the molecular bases for how regular exercise promotes healthy aging and prolongs the health span among healthy adults is fundamental to public health (33). However, the system is complex: An acute bout of exercise alters the expression of ~9800 systemic circulatory analytes, including transcripts, metabolites, proteins and lipids (34). Moreover, inactivity-related diseases are not necessarily caused by the same mechanisms by which exercise prevents those very diseases, indicating that no singular molecular paradigm controls adaptations to inactivity and exercise training (rev. 11).

Importantly, exercise and caloric restriction may provide additive effects for improving mitochondrial function/health and insulin sensitivity (35,36). One putative mechanism for this effect involves re-allocation of energy from harmful oversupply to reproductive tissues and fat storage to maintenance and repair functions, synergistically retarding senescence and decreasing susceptibility to many chronic diseases (16). Touted as “perhaps the most powerful anti-aging nutritional intervention” in the absence of malnutrition and certain clinically significant diseases (e.g., amyotrophic lateral sclerosis), decreasing daily caloric intake increases health span and lifespan (Section 2, rev. 37).

Caloric restriction inhibits primary molecular indices of aging, promoting better insulin sensitivity and homeostasis and improved mitochondrial health (38,39). In pre-clinical studies, predominantly using mice, long-term caloric restriction opposed the changes in gene expression driving DNA-replication and cell cycle defects, oxidative stress, macromolecular damage, tumorigenesis, age-related inflammation, and compromised stress responses (rev. 37).

In assessing the role of exercise (and caloric restriction) in extending the health span, this review follows the logic underpinning the eponymous symposium at the American Physiological Society’s 2024 Summit with Glenn A. Gaesser, Susan B. Racette, Stephanie E. Hall, and Siddhartha S. Angadi. The following sections present contemporary evidence for exercise (especially in individuals with obesity, Section 1) and caloric restriction (Section 2) impacting physiological function, and brain health (Section 3) directly germane to extending the health span. Lastly, given the broad failures of the scientific, medical, and public health communities to translate research findings into broadly achievable exercise and dietary practices that would extend the health span, Section 4 provides one exemplar of how comedic measures, such as Monty Python’s epic Silly Walks skit, might be recruited to this critical mission.

Section 1: Weight Loss Versus Increasing Fitness and Physical Activity for Reducing Health Risks in Obesity

The increased morbidity and mortality associated with physical inactivity and low levels of CRF and muscular fitness (40–42) are especially relevant in individuals with obesity who characteristically have lower levels of physical activity (43,44) and poorer CRF (45). Consequently, obesity is associated with reduced active life expectancy (46) and health span.

This issue has urgent public health significance because the global prevalence of obesity over the past ~50 years has doubled in 70 countries (47) and has increased 3-fold in the U.S. (48). Obesity prevalence in the U.S. is projected to reach nearly 50% by 2030, with severe obesity (body mass index [BMI] ≥ 40.0 kg/m²) expected to become the most common BMI category among women, non-Hispanic black adults, and low-income adults (49). The increase in the rampancy of obesity is temporally aligned with that of weight loss attempts (50), with > 40% of adults in the U.S. and worldwide attempting weight loss annually (51,52). The high and increasing prevalence of weight loss attempts has been largely unsuccessful at reducing the prevalence of obesity. Because weight loss is invariably transient for the majority of individuals who lose weight, weight cycling is common, affecting 20% to 55% of the general population (50). Weight cycling, as discussed below, is associated with numerous adverse health outcomes (50, 53).

Obesity is generally associated with increased mortality risk and reduced physical function (thus reducing health span), and weight loss remains the cornerstone of treatment. However, while weight loss is generally associated with improvements in physical function (54), the impact of weight loss on mortality risk is ambiguous (see below). Whether weight loss should be the primary focus for treating obesity-related health conditions has been the subject of considerable debate (53, 55–58). The relationship between BMI and mortality risk is influenced by multiple factors such as age and lifestyle. For example, meta-analyses of cohort studies demonstrate that even moderate levels of CRF attenuate or eliminate all-cause and cardiovascular disease (CVD) mortality risk associated with high BMI and/or body fat (59–61). The impact of CRF is so robust that individuals with obesity and moderate-to-high CRF have lower all-cause mortality risk compared to normal-weight individuals with low CRF (59,61)

Weight loss via energy restriction may elevate CRF, but the increase is due to an increase in relative $\dot{\text{V}}{\text{O}}_2\text{max}$ (ml · kg⁻¹ · min⁻¹) with no change, or a decrease, in absolute $\dot{\text{V}}{\text{O}}_2\text{max}$ (L/min) (54, 62, 63). The decrease in absolute $\dot{\text{V}}{\text{O}}_2\text{max}$ may be due to a decrease in lean body mass with weight loss achieved via energy restriction (54, 62–65). A decrease in $\dot{\text{V}}{\text{O}}_2\text{max}$ relative to lean body mass may also occur with energy restriction (63), suggesting that muscle quality is not improved with weight loss. This may have additional relevance because $\dot{\text{V}}{\text{O}}_2\text{max}$ in ml · kg fat-free mass⁻¹ · min⁻¹ is a stronger predictor of all-cause mortality than $\dot{\text{V}}{\text{O}}_2\text{max}$ expressed in ml · kg⁻¹ · min⁻¹ (66). Despite decreases in muscle mass with weight loss (64), muscle strength may be preserved (62, 63). Patients having undergone bariatric surgery typically increase physical functioning, but this may be due to improved efficiency of performing activities of daily living without any change in CRF or muscular fitness (54).

Despite an increase in relative CRF with weight loss, evidence for a significant mortality risk reduction associated with intentional weight loss is tenuous. Several meta-analyses have indicated that intentional weight loss is not consistently associated with lower all-cause mortality (67–73). Additionally, a systematic review of the effect of weight loss medications on mortality found no advantage of anti-obesity medications over placebo (74). Glucagon-like peptide-1 (GLP-1) receptor agonists have been reported to reduce mortality risk, but these reports are restricted to studies of patients with severe health conditions (75–77). Moreover, GLP-1 receptor agonists can cause significant loss of muscle mass, which could adversely affect skeletal muscle’s vital role in maintaining metabolic health (78).

Most meta-analyses show no statistically significantly lower risk of all-cause mortality with intentional weight loss (Figure 2). Interpretation of cohort studies included in these meta-analyses is limited due to their observational design. Meta-analyses of randomized controlled trials (RCTs) show weight loss to be associated with ~20% lower risk of all-cause mortality (68,69,71,72), but there are serious concerns with these meta-analyses. The major weakness of the RCTs is the limited number of mortality events used to calculate mortality rates, which is likely attributable to the age of the participants and the short follow-up duration for mortality ascertainment. The majority of RCTs included in these meta-analyses had <5 deaths reported in either the intervention or control groups. In one meta-analysis (69), 23 of the 34 RCTs included in the meta-analysis had ≤1 death reported in either the control or intervention group. Also, many of the studies included in these RCTs involved changes in either diet quality or physical activity, or both. An increase in physical activity and improved diet quality can reduce mortality risk independently of weight loss (58). Thus, results of these meta-analyses of RCTs must be viewed with caution. Moreover, if relative CRF was increased in any of these RCTs, it is not apparent if the higher relative CRF after weight loss provides the same mortality benefit as exercise training-induced increases in CRF, as discussed below.

Figure 2.

All-cause mortality risk associated with increasing cardiorespiratory fitness (CRF), increasing physical activity (PA), and intentional weight loss. For CRF and PA, cohorts consisted of middle-aged and older adults with no reported major health conditions at baseline. For intentional weight loss, the meta-analyses of randomized controlled trials generally consisted of adults with overweight or obesity and with 1 or more chronic health conditions (e.g., hypertension, impaired glucose tolerance, T2D, CVD). For Harrington, “healthy” and “unhealthy” were defined as “with” or “without” obesity-related risk factors. Sample sizes for meta-analyses: Harrington (2009; 10 studies; n = 28,672); Chen (2018; 3 studies; n = 6,875); Pack (2014; 1 study; n = 377); Singh (2019; 33 studies; n = 19,379); Kritchevsky (2015; 12 studies; n = 15,306; Ma (2017; 34 studies; n = 22,779); Schellenberg (2013; 2 studies; n = 5,305). The horizontal lines represent the upper and lower 95% confidence intervals (CI) for the respective risk ratio for each study. The mortality risk ratios for increasing CRF and PA are generally much lower than those for intentional weight loss, with the upper boundary of the 95% CI more consistently < 1.0.

Cohort studies have consistently shown that increasing either physical activity or CRF is associated with reductions in all-cause mortality risk that are far greater than those associated with weight loss (Figure 2). Across multiple cohort studies, increasing physical activity is associated with a reduction of ~10–50% in all-cause mortality (79–88) and ~15–40% in CVD mortality (79–82, 84, 86, 88, 89), whereas even greater reductions are observed with increases in CRF (90–94). When initially unfit individuals (generally defined as being in the bottom quintile or quartile of age-adjusted CRF level) move into a more fit category, reductions in all-cause mortality are in the range of 30% to 60% (Figure 2). Increasing CRF also is associated with a ~40–60% lower risk of CVD mortality (90,94,95). As a continuous variable, each 1-MET (1 MET = metabolic equivalent of resting metabolism, ~3.5 ml · kg⁻¹ · min⁻¹) increase in CRF is associated with a 14–29% reduction in all-cause mortality risk (91,94,96,97) and each 1 ml · kg⁻¹ · min⁻¹ increase in CRF is associated with a 7–13% reduction in all-cause mortality risk (98–100). Similar reductions are observed for CVD risk (94,99,100). For perspective, meta-analyses demonstrate that exercise training improves $\dot{\text{V}}{\text{O}}_2\text{max}$ in adults by ~3.5–5.0 ml · kg⁻¹ · min⁻¹ (101,102). In adults with obesity, high-intensity interval training and moderate-intensity continuous training have been shown to increase $\dot{\text{V}}{\text{O}}_2\text{max}$ by 3–4 ml · kg⁻¹ · min⁻¹ after just 8 weeks of training despite no loss of body weight or body fat (103).

Weight loss is unlikely to explain the reductions in mortality risk associated with increasing physical activity or CRF in the cohort studies presented in Figure 2. Exercise training rarely results in significant weight loss (104). Also, many studies that included changes in body weight or BMI in their analyses indicated that weight loss was not a factor that could explain the reduced mortality risk associated with an increase in CRF (84,85,94,95). In the Aerobics Center Longitudinal Study (94) and Oslo Ischemia Study (95), the reduced all-cause mortality risk associated with increased CRF was not related to a decrease in BMI. Among Danish adults who took up cycling, during a 10–13-year follow-up period all-cause mortality risk was reduced by 22%, even though cycling was not associated with reductions in body weight (85).

Thus, increasing CRF via physical activity is likely to increase health span to a greater extent than weight loss (Figure 1). This benefit is attributable to the greater mortality risk reduction (Figure 2) as well as greater physical function accompanying aerobic and resistance exercise training (as discussed above, absolute CRF and muscular strength are not increased with weight loss, and in some instances have been shown to decrease). The studies depicted in Figure 2 that reported on changes in physical activity did not differentiate between aerobic and resistance exercise. Also, we are not aware of any studies that specifically examined the association between changes in muscular strength and mortality risk. However, a systematic review and meta-analysis of nearly 2 million apparently healthy adults demonstrated that, compared with the lowest tertile/quartile of handgrip strength and knee extension strength, the highest tertile/quartile of handgrip strength and knee extension strength were associated with 31% and 14% lower risk of all-cause mortality, respectively (105). Another systematic review and meta-analysis indicated that self-reported engagement in resistance exercise was associated with a 21% lower risk of all-cause mortality (106). When combined with aerobic exercise, the all-cause mortality risk reduction was 40%, supporting the public health recommendation to engage in both aerobic and resistance exercise regularly.

Weight loss caveat - weight cycling

In a 1998 New England Journal of Medicine editorial on losing weight, editor-in-chief Jerome Kassirer and executive editor Marcia Angell cautioned that, “until we have better data about the risks of being overweight and the benefits and risks of trying to lose weight, we should remember that the cure for obesity may be worse than the condition” (107). The “cure”—weight loss—can be highly problematic due to its transient nature (50). Weight loss typically improves cardiometabolic disease risk factors (53), although improvements tend to revert during weight regain (50). More importantly, repeated weight loss attempts may adversely affect health due to risk factor variability (108, 109). Several meta-analyses have shown that weight cycling, defined as the repeated pattern of losing and regaining weight over time (also known as yo-yo dieting), is associated with ~35–50% increased risk of all-cause mortality (110–112), and a 36% higher risk of CVD mortality (112). The 36% higher CVD mortality risk is supported by a 35% higher risk of hypertension associated with weight cycling (112). The magnitude of these risks linked to weight cycling is similar to that associated with obesity (without consideration of CRF). Because individuals with obesity are more likely to experience weight cycling due to more frequent weight loss attempts (52), risks associated with obesity could be attributable in part to weight cycling.

The increased risk associated with weight cycling may be due to variability in cardiometabolic risk factors that fluctuate with changes in body weight (50, 108,109). Variability in CVD risk factors is associated with increased risk of all-cause mortality, stroke, coronary heart disease, and myocardial infarction (109). Risk increases in proportion to the magnitude of variability assessed by the individual standard deviation of multiple measurements over time (109), as well as in proportion to the magnitude of weight cycling (113). Elevated risk factor variability is associated with increased risk for developing CVD, even if the levels of the risk factors are trending in a positive direction (108).

Data from studies on patients with T2D is instructive. Variability in risk factors and body weight is especially detrimental to those with T2D (108) because cardiovascular complications are the major cause of mortality in T2D. High variability in body weight and CVD risk factors, including elevated blood pressure and hemoglobin A1c (HbA1c), are associated with increased risk of cardiovascular events (myocardial infarction, stroke) and all-cause mortality (108). Among Swedish adults with T2D, the combination of high variability in body weight and systolic blood pressure was associated with an 80% higher risk of cardiovascular events (108). High weight variability is consistently associated with a 50% to 58% increased risk of all-cause mortality in T2D (114). This coincides with the consistent finding that body weight variability is associated with increased risk of cardiovascular events in T2D (114).

Exercise may mitigate adverse effects of weight cycling

Data from the Look AHEAD trial suggest that exercise may mitigate potentially harmful effects of weight cycling in adults with T2D and overweight or obesity. Among participants randomized to the control group, those in the highest quartile of coefficient of BMI variability had a 4-fold higher risk of all-cause mortality, a 15.3-fold higher risk of CVD mortality, and a 2.2-fold higher risk of cardiovascular events during a median 6.7-year follow-up (115). Those in the intensive lifestyle intervention group had no increase in CVD risk when comparing highest vs. lowest quartiles of BMI variability. The authors speculated that the lack of association in the intervention group could be attributed to the exercise component of the intervention. Both aerobic (116) and resistance (117) exercise during a period of controlled weight regain can counter the adverse effects of weight regain on cardiometabolic risk markers. Thus, exercise could be crucial for mitigating the risk associated with weight cycling, especially since severe weight cyclers (defined as having lost ≥ 20 lb. intentionally on at least 3 occasions) are less likely to use exercise to control body weight (118).

Because weight cycling is more prevalent among individuals with high BMI (119, 120), it is plausible that some of the morbidity and mortality risks associated with high BMI may be due to weight cycling. Of note, weight cycling is associated with increased risk for sarcopenic obesity (121), as weight regain following weight loss is typically characterized by greater gain of fat mass compared with fat-free mass (122,123). This may have adverse consequences for weight cyclers because body composition changes accompanying weight cycling predict mobility disability in older adults (124), and sarcopenic obesity is associated with a 24% higher risk of all-cause mortality (125).

Weight loss vs. exercise for reducing cardiometabolic risk factors

It is well established that weight loss is associated with improvements in obesity-related health conditions, including cardiometabolic risk markers for CVD and T2D (53, 114). The improvements in cardiometabolic risk markers associated with weight loss interventions are generally no greater than with exercise training interventions without a specific weight loss target, with the possible exception of fasting plasma triglycerides, which tend to be reduced more with weight loss interventions (Table 2). When comparing weight loss interventions with exercise training interventions that do not focus on weight loss, there are similar increases in high-density lipoprotein cholesterol (~1–5 mg/dl) and flow-mediated dilation (~1–4%) and decreases in blood pressures (~1–5 mmHg), low-density lipoprotein cholesterol (~1–15 mg/dl), and HbA1c (~−0.2–0.9%).

Table 2.

Comparison of meta-analysis results of randomized controlled trials of weight loss interventions and exercise training interventions for improving cardiometabolic risk factors.

Variable	Weight Loss Interventions	Exercise Training Interventions
Blood pressure	↓ ~1–5 mmHg	↓ ~2–5 mmHg
HbA1c	↓ ~ 0.2–0.9%	↓ ~ 0.2–0.8%
Triglycerides	↓ ~11–58 mg/dl	↓ ~5–25 mg/dl
LDL cholesterol	↓ ~1–15 mg/dl	↓ ~3–10 mg/dl
HDL cholesterol	↑ ~1–4 mg/dl	↑ ~2–5 mg/dl
Flow-mediated dilation	↑ ~1–3 %	↑ ~1–4 %

Ranges represent results of multiple meta-analyses reviewed by Gaesser and Angadi (53).

Again, results from the Look AHEAD trial are illustrative (126). After the first year of the comprehensive lifestyle intervention in adults with T2D, a mean weight loss of 8–10 kg (8.6% of initial body weight) was associated with a 0.64% reduction in HbA1c, consistent with the general findings from multiple weight loss studies. However, the association between weight loss and HbA1c improvement may not be causal. A meta-analysis of 27 exercise training studies in patients with T2D demonstrated that exercise training decreased HbA1c by an average of 0.8% (127). The authors noted that the magnitude of HbA1c reduction was close to that observed with dietary and pharmaceutical treatments (and is also marginally greater than that observed in the Look AHEAD trial). It is not likely that weight loss explained the reduction in HbA1c in the exercise training studies because reductions in HbA1c after both aerobic exercise training and combined aerobic and resistance exercise training were the same (~0.7–0.8%), despite the fact that weight loss after combined exercise training (−5.1%) was much greater than after aerobic exercise training alone (−1.5%).

Even when weight loss occurs, correlations between change in body weight and improvement in cardiometabolic disease risk factors are weak (53). Moreover, the results for weight loss studies must be interpreted with caution because many of the weight loss studies included changes in diet quality and/or changes in physical activity, both of which can improve cardiometabolic disease risk markers independently of weight loss (53, 57, 58).

As reviewed elsewhere, exercise training has a direct impact on adipose tissue that is likely superior to that of weight loss (53). Specifically, exercise training may enhance the “fitness” of white adipose tissue by increasing mitochondrial biogenesis and insulin sensitivity independently of weight loss. Exercise training is also more effective than weight loss for reducing visceral abdominal fat, e.g., for a given amount of weight loss the reduction in visceral abdominal fat is ~50% greater after exercise training than after diet-induced weight loss. The benefits of exercise training extend well beyond improvement in traditional cardiometabolic risk factors (128), and likely constitute a “polypill” effect of exercise that positively impacts virtually every cell, organ, and system in the body (129), regardless of BMI and independent of weight loss.

Section 1 Summary

Increases in physical activity and CRF are superior to intentional weight loss for reducing morbidity and mortality, and likely will result in a greater impact on health span. A weight-neutral, fitness-focused approach to treating obesity-related health conditions may be preferred to a weight-centric approach on empirical grounds, while avoiding the pitfalls associated with weight cycling (Figure 3). This approach should not be interpreted as a categorical rejection of calorie restriction as a means to improve health span, as the following section describes the benefits of calorie restriction among healthy individuals without obesity.

Figure 3.

Comparison of weight loss vs. increasing physical activity (PA) and cardiorespiratory fitness (CRF) for obesity treatment. Increasing PA, especially via exercise of sufficient stimulus to improve both CRF and muscular fitness, may or may not change body weight or total body fat, but health improvements are largely independent of weight loss. By contrast, a weight-loss approach may increase the risk of weight cycling, which is associated with numerous adverse health outcomes, including increased mortality risk and worsening of some cardiometabolic risk factors. For cardiometabolic risk factors, “+” denotes positive change (i.e., either ↑ or ↓ depending on the risk marker). See text for more information.

Section 2. Calorie Restriction and Health span: Highlights from the CALERIE^™ Trial

Calorie restriction (CR) is characterized by a sustained reduction in daily energy intake without compromising nutritional adequacy or diet quality. The excitement around this topic is that CR has shown great promise as a robust nutritional strategy to impact biological aging favorably and thereby increase health span (130–133). Efforts to optimize health span, which reflects the years of life lived in good health without disease or disability, are critical as our population ages. The ability of CR to prevent or delay disease onset and to extend lifespan has been demonstrated in many non-human species, providing strong support for the geroprotective role of CR. Observational data from humans provide additional compelling evidence that a CR lifestyle has merit for optimizing cardiometabolic health indices. The first randomized controlled trial of CR in humans without obesity was CALERIE^™ (Comprehensive Assessment of Long-term Effects of Reducing Intake of Energy), which was designed specifically to uncover the effects of 2 years of CR on various biomarkers of aging and health span in humans. The CALERIE^™ trial results provide compelling evidence that CR is a promising strategy to attenuate biological aging. The ongoing CALERIE^™ Legacy study and other CR initiatives funded by the National Institute on Aging are designed to answer important questions about the potential of long-term CR to increase health span.

Primary and secondary aging

Biological aging can be described in terms of primary aging and secondary aging (134). Primary aging is the progressive deterioration in tissue structure and function that occurs due to the aging process itself; this process is considered to be inevitable. Primary aging is a strong determinant of maximal lifespan. Secondary aging, in contrast, is the deterioration in tissue structure and function that is attributable to disease pathology, adverse lifestyle behaviors, and harmful environmental factors. Therefore, secondary aging is considered to be modifiable (to some extent) and a major determinant of mean lifespan. There are many strategies to impact secondary aging and prevent or mitigate disease, such as healthy dietary patterns, a physically active lifestyle, engagement in formal exercise, avoidance of tobacco, healthy sleep patterns, and stress management. An important scientific question is whether primary aging can be altered and, if so, which strategies are effective.

Calorie restriction and lifespan

Benjamin Franklin wisely and eloquently stated “To lengthen thy life, lessen thy meals.” In support of this position, CR consistently has been shown to extend both mean lifespan and maximal lifespan in many pre-clinical models, including yeast, drosophila, worms, rats, mice, dogs, and some non-human primates (135–139). Elegant rodent experiments performed by Dr. Clive McCay at Cornell University in the 1930s tested the hypothesis that growth restriction would increase lifespan (140,141). Their results were very exciting, demonstrating that in comparison to rats fed ad libitum, the survival curves of rats fed a 30% CR diet were shifted to the right, signifying longer mean and maximal lifespan.

An important question is whether exercise can enhance the effects of a CR diet. The significant benefits of exercise on most organ systems and in preventing or mitigating disease processes are well established (11,142). Therefore, it is logical to surmise that exercise may have complementary or additive benefits with CR on the aging process. Landmark studies conducted in the laboratory of Dr. John Holloszy at Washington University School of Medicine in St. Louis, MO, compared CR alone, exercise alone, and CR plus exercise to a control condition in rats (143). The survival curves reveal clearly that rats engaged in exercise in the form of wheel running had greater mean lifespan, but not greater maximal lifespan, than control rats. Rats in the CR group, however, had greater mean and maximal lifespan, indicated by a shifting of the curve to the right for both average lifespan and for the longest-lived rats. The addition of exercise to a CR diet did not have additive effects and did not change the results compared to CR alone. Therefore, CR elicited a survival benefit that was distinct from exercise. The hypothesis is that physiological effects observed in the CR group but not the exercise group may be impacting primary aging.

Calorie restriction and health span

Pursuant to extending the health span and lifespan, cross-sectional studies in humans conducted by Dr. Luigi Fontana while he was at Washington University School of Medicine in St. Louis revealed that adults who voluntarily followed a CR diet for several years (mean: 6 yr, range: 3–15 years) had very favorable cardiovascular and metabolic health indices that were vastly superior to those of an age-matched comparison group of individuals who had not been following a CR diet (144,145). The CR individuals were not in a controlled research trial with a prescribed intervention, but rather had chosen a lifestyle referred to as calorie restriction with optimal nutrition. Their high adherence to this lifestyle and striking health metrics provide convincing support for the potential impact of a CR diet on promoting healthy aging.

The CALERIE^™ trial

The promising results of CR interventions in pre-clinical models and of a CR lifestyle in cross-sectional human studies provided strong rationale for exploring the effects of long-term CR in human trials. The National Institute on Aging (NIA) of the National Institutes of Health (NIH) has been supporting the series of human CR studies referred to as CALERIE^™ since 2001 (Figure 4). The CALERIE^™ trial has the distinction of being the first RCT of long-term CR in humans without obesity and was designed to explore the effects of long-term CR on biomarkers of aging and health span. This is a very important distinction from the extensive research that has been conducted in the field of obesity treatment (see Section 1 above). Individuals who were eligible for the CALERIE^™ trial were quite healthy. Obesity, T2D, and other CVD risk factors were exclusionary because the aim of the CALERIE^™ trial was to evaluate the potential benefits of CR on the aging process among healthy adults, rather than to study the effects of CR on ameliorating metabolic dysfunction.

Figure 4.

CALERIE^™ studies of sustained calorie restriction (CR) in healthy adults without obesity. CALERIE^™ Phase 1 was designed as three single-site, randomized controlled trials (RCTs) conducted at Pennington Biomedical Research Center in Baton Rouge, LA; the Jean Mayer USDA Human Nutrition Research Center on Aging at Tufts University in Boston, MA; and Washington University School of Medicine in St. Louis, MO. Duke Clinical Research Institute in Durham, NC served as the Coordinating Center. The interventions ranged from 6 months to 1 year and the prescribed level of CR was 20–30%, reflecting a reduction in energy intake of 20–30% below weight-maintenance energy needs. A total of 141 participants were enrolled. The favorable results across sites demonstrated the feasibility of CR and important benefits on numerous well-established cardiometabolic health indices (146–149), supporting advancement to CALERIE^™ 2 (referred to as the CALERIE^™ trial), a multi-site, RCT with a 2-year CR intervention involving a 25% CR diet or and ad libitum control condition in 220 participants (150,151). The current CALERIE^™ Legacy study is an observation follow-up study of the same participants who participated in the CALERIE trial 10–15 years earlier and also includes a comparison group from the Baltimore Longitudinal Study of Aging.

CALERIE^™ Phase 1:

The first CALERIE^™ studies were designed as three single-site, RCTs conducted between 2001 and 2005 at the following centers: Pennington Biomedical Research Center (PBRC) in Baton Rouge, LA; the Jean Mayer USDA Human Nutrition Research Center on Aging at Tufts University in Boston, MA; and Washington University School of Medicine (WUSM) in St. Louis, MO. Duke Clinical Research Institute in Durham, NC served as the Coordinating Center. The CR interventions ranged from 20% to 30% CR, reflecting a reduction in energy intake of 20–30% below weight-maintenance energy needs. The intervention durations were 6 months (PBRC) and 1 year (Tufts, WUSM). A total of 141 participants were enrolled. The favorable results across sites demonstrated the feasibility of CR and important benefits on numerous well-established cardiometabolic health indices (146–149), supporting advancement to a larger, longer-term trial of CR.

CALERIE^™ Phase 2:

The second phase of the CALERIE^™ studies, which is referred to as the CALERIE^™ trial, was a multi-site, RCT conducted between 2006 and 2012 at the same sites as in CALERIE^™ 1 (150). Eligible individuals were healthy premenopausal women aged 21–47 years and men aged 21–50 years with a BMI of 22.0 to <28.0 kg/m², non-smokers, and not taking prescription medications other than oral contraceptives. A total of 220 participants were randomized to either the CR group (25% CR) or an ad libitum (AL) control group in a 2:1 allocation, stratified by study site, biological sex, and BMI (<25.0 or ≥25.0 kg/m²) (151). The intervention and control periods were 2 years in duration. CR participants received extensive education and counseling by registered dietitians and clinical psychologists to promote adoption of, and adherence to, the CR regimen through individual and group sessions throughout the 2-year intervention. AL participants did not receive an intervention. Assessments were conducted at baseline, 12 months, and 24 months in all participants; additional assessments were completed at 6 and 18 months in CR participants.

Establishing a calorie restriction prescription

The precise calorie reduction and calorie intake to meet the 25% CR prescription was individualized based on careful quantification of each participant’s baseline energy expenditure and energy intake. This was accomplished using the doubly labeled water (DLW) method during 4 continuous weeks at baseline while participants were weight stable. The DLW method is the most accurate method of quantifying energy expenditure among free-living individuals as they lead their typical lives (152). Briefly, the stable (i.e., non-radioactive) isotopes deuterium (²H) and oxygen-18 (¹⁸O) are consumed orally, mix with and enrich the total body water pool, and then are eliminated from this pool through exhaled CO₂ and water losses (i.e., urine, saliva, sweat, respiratory water, and fecal water). Because ²H is released only through water losses, whereas ¹⁸O is released through water and CO₂, the differential elimination rates of these isotopes reflect CO₂ production, from which O₂ consumption and energy expenditure (kcal/day) can be computed. Isotopic enrichment of total body water is determined by sampling urine before and 7 and 14 days after DLW administration and analyzing the samples using isotope ratio mass spectrometry.

Assessing adherence during a calorie restriction intervention

In addition to facilitating an accurate calorie prescription, the DLW method has the distinct advantage of providing an objective and reasonably accurate assessment of calorie intake during a CR intervention if changes in body composition are measured using a valid method (153,154). By determining total daily energy expenditure with DLW and measuring changes in fat mass and fat-free mass with dual-energy x-ray absorptiometry (DXA), the investigators were able to calculate energy intake (kcal/day) and %CR achieved every 6 months throughout the 2-year CR intervention.

CALERIE^™ trial results

Of the 220 participants randomized to CR (n=145) or the AL control condition (n=75), 218 began their allocated assignment (n=143 CR, n=75 AL) and 188 completed the trial (n=117 CR, n=71 AL) (151). The CR and AL groups were well-matched at baseline on age (mean±SD 38.0±7.3 and 37.9±7.0 years, respectively), sex (69.0 and 70.7% women), BMI (25.2±1.8 and 25.1±1.6 kg/m²), cardiometabolic health indices (blood pressure, lipid profile, fasting glucose and insulin), and total daily energy expenditure (2467±403 and 2390±348 kcal/d). The prescribed energy intake to achieve 25% CR in the CR group averaged 1832 kcal/d, with a range of 1227 to 2719 kcal/d. CR participants achieved an average of 11.9±7.2% CR throughout the 2-year intervention, with the highest adherence observed during the first 6 months (19.5±8.7% CR). It is noteworthy that the achieved levels of CR represent a much more modest level of restriction than is typically advocated with many commercial diet plans that strive for more rapid weight loss. While the %CR achieved was significantly less than the prescribed level of 25%, several beneficial effects were observed, as described below.

Calorie restriction and body composition:

After the 2-year CR intervention, participants exhibited significant decreases in body weight (−10.4%, −7.6±0.3 [SE] kg, P<0.001) and adiposity (−5.4±0.3 kg whole-body fat mass, −2.8±0.2 kg trunk fat mass, −6.2±0.4 cm waist circumference, all P<0.001) (155), with 71% of the weight loss attributable to fat mass loss (156). Reductions also were observed in fat-free mass (−2.0±0.2 kg, P<0.001) and bone mineral density, which was not surprising in the absence of an exercise component during CR.

Calorie restriction and cardiometabolic health indices:

Numerous favorable changes were observed among markers of cardiovascular health (systolic and diastolic blood pressure, LDL-cholesterol, total cholesterol, HDL-cholesterol, triglycerides), metabolic health (fasting insulin, insulin sensitivity index, glucose tolerance), inflammation (C-reactive protein, tumor necrosis factor alpha), oxidative stress (urinary F2-isoprostanes), and immune function (white blood cell count, lymphocyte count, monocyte count) (157,158). Psychological benefits were observed for mood and quality of life (159). Cardiorespiratory fitness, expressed in relative terms, improved during 2 years of CR (baseline $\dot{\text{V}}{\text{O}}_2\text{max}$: 34.8±0.6 [SE] ml · kg⁻¹ · min⁻¹; change at 2y:+1.9±0.5 ml · kg⁻¹ · min⁻¹, P<0.001) (62). Importantly, CR participants were able to exercise longer on the treadmill during the Cornell graded exercise test after 2 years (baseline: 13.0±0.2 min; change at 2y: +2.9±0.2 min, P<0.001) and reached a higher exercise stage and correspondingly higher work rate (baseline: stage 7, 3.8 mph, 16% grade; 2y: stage 8, 4.2 mph, 16% grade) (62). These improvements in exercise capacity occurred despite a decrease in lean body mass (baseline: 46.0±0.8 kg; change at 2y: −2.3±0.2 kg, P<0.0001), a decrease in absolute $\dot{\text{V}}{\text{O}}_2\text{max}$ (baseline: 2.53±0.06 L/min; change at 2y: −0.15±0.03 L/min, P<0.001), and no change in physical activity level from baseline to 2 years. Collectively, these predominantly advantageous effects of CR on cardiometabolic health indices suggest that modest CR is a promising nutritional strategy that may increase health span.

Calorie restriction and biological aging:

Two well-established phenotypic aging algorithms are the Klemera-Doubal Method (KDM) Bioage (160) and the Homeostatic Dysregulation Index (HDI) (161), both of which are based on the following 10 biomarkers: serum albumin, alkaline phosphatase, C-reactive protein, total cholesterol, creatinine, HbA1c, systolic blood pressure, urea nitrogen, uric acid, and white blood cell count. The KDM algorithm includes chronological age, whereas HDI does not. Favorable results of 2 years of CR on biological aging were observed with both KDM & HDI indices. In comparison to the increase of 0.71 years per calendar year (95% CI: 0.41, 1.01) observed in the AL group, the CR group had a slower rate of biological aging of 0.11 years per calendar year (95% CI: −0.13, 0.36) (162). Interestingly, the effect of CR did not appear to be dependent on weight loss, based on a model that included weight as a time-varying covariate.

Other metrics of biological aging include aging clocks and a Pace of Aging algorithm that are based on DNA methylation. Using DNA extracted from white blood cells, Waziry et al (163) quantified the effects of CR relative to the AL control condition on the PhenoAge clock, the GrimAge clock, and the DunedinPACE measure. Whereas the two clocks did not show an effect of CR on the pace of aging over 2 years, the Dunedin PACE metric showed a statistically slower rate of biological aging in the CR group compared to the AL group at both 1 year and 2 years. A potential explanation, per the authors, is that DunedinPACE is more of a dynamic estimator (similar to a speedometer) that reflects the rate of biological aging over time, whereas the clocks represent static measures of age at single time points. Future work will be important to explore further and clarify the effects of long-term CR on various metrics of biological aging.

Calorie restriction and cellular senescence:

Aging at the cellular level can be explored with biomarkers of cellular senescence, which reflects a state in which dysfunctional cells are resistant to apoptosis and may adversely affect nearby healthy cells. Investigators in the laboratory of Dr. Nathan LeBrasseur at Mayo Clinic in Rochester, MN, studied several senescence biomarkers in plasma samples from CALERIE^™ participants and observed favorable effects in the CR group compared to the AL control group in eleven senescence biomarkers at one or more time point: 12 months (IL7, MMP1, MPO, RANTES, VEGF), 24 months (ICAM1, TNFR2), and both 12 and 24 months (PAI1, PARC, TARC, TNFR1) (164). These results provide additional evidence that CR may impact aging at the cellular level.

Ongoing initiatives and future directions

The CALERIE^™ initiatives began in 2001 and continue robustly today. The success of the original CALERIE^™ grants led to the CALERIE^™ Biorepository grant, the ongoing CALERIE^™ Network grant, and the ongoing CALERIE^™ Legacy study. The R33 Network grant promotes use of the CALERIE^™ biorepository and data repository through ancillary studies and new grant applications, particularly among early-stage investigators. The CALERIE^™ Legacy study is an observational follow-up study of CALERIE^™ 2 trial participants to address the important question of whether a CR intervention during early to mid-adulthood has long-term effects in attenuating biological aging. The CALERIE^™ Legacy study involves a comprehensive assessment battery designed to assess cellular, phenotypic, and functional aging biomarkers and includes a comparison group of age- and sex-matched adults from the Baltimore Longitudinal Study of Aging (165). Future directions include larger randomized trials of longer-term CR, also funded by the NIA, to explore the effects of 5 years of CR among young and older adults (U01AG073204, U01AG073240).

Section 2 Summary

Calorie restriction is a promising nutritional strategy that targets the biology of aging and therefore has the potential to increase health span. Results from human cross-sectional studies of voluntary CR and from the multi-site, randomized controlled CALERIE^™ trial overwhelmingly support modest CR as a safe and promising strategy to attenuate biological aging and enhance health span.

Section 3. Exercise can Significantly Improve the Brain Health Span

Brain health is the state of brain functioning spanning cognitive, sensory, social-emotional, behavioral, and motor functions. Cognitive health includes memory, attention, problem-solving, and language, which are essential for daily functioning. With age, the brain undergoes structural and functional changes that can impact cognitive health, which is often the most recognized aspect of brain aging. This age-related change in cognition affecting memory and learning is referred to as cognitive decline. Severe cognitive decline results from neurodegenerative diseases such as Alzheimer’s disease. Understanding both normal cognitive aging and neurodegenerative cognitive decline is essential to develop and implement strategies to improve brain health span.

Exercise has been recognized as one of the most effective strategies for maintaining cognitive health and preventing cognitive decline. Recent advances in exercise neuroscience have begun to unveil the underlying mechanisms of exercise-induced neuroprotection in both natural aging decline and the significant decline seen with neurodegenerative disease. Following a brief overview of brain aging and neurodegenerative diseases, this section will explore exercise’s impact on expanding the brain health span as indicated by cognition and discuss the possible mechanisms in natural aging and neurodegenerative disease.

Brain aging: Impaired neurogenesis

Neurogenesis, the creation of new neurons from neural stem cells, is vital in the maintenance of cognitive function. Two brain regions contain the greatest neural stem/progenitor cells for neurogenesis: the ventricular-subventricular zone of the lateral ventricles and the subgranular zone in the hippocampus dentate gyrus (166). The ventricular-subventricular zone provides new interneurons to the olfactory bulb and in mammals olfactory bulb neurogenesis continues throughout aging (167). The subgranular zone supplies neurons for the dentate gyrus in the hippocampus, important for learning and memory (168). For various and complex reasons (inflammation, transcriptional changes, and changes in cellular environment), age induces a decline in neural stem cell proliferation and neurogenesis (168).

Brain aging: Neuroinflammation

Aging results in impaired immune function throughout the body, both innate and adaptive, resulting in increased susceptibility to infections and reduced vaccine effectiveness. The brain is an immune-privileged site (adaptive immunity and inflammation are highly controlled) largely due to the anatomy of the blood-brain barrier (169). Microglia serve as the innate immune cells of the brain. A “resting” microglia is in constant movement, surveilling for bacteria/viruses and proteins and DNA/RNA released by damaged cells. When activated, microglia signal a pro-inflammatory response and subsequent release of cytokines, such as IL-1, IL-6, and TNF-α. In an aged brain, microglia maintain an active pro-inflammatory state and become less effective at clearing misfolded proteins (170). Inflammaging is the phenomenon of low-grade, chronic damage resulting from increased inflammation levels in the body and is a significant risk factor in morbidity and mortality in the older population (171). Both reduced neurogenesis and neuroinflammation play significant roles in the decrements in brain health over the lifespan.

Neurodegenerative disease

Age is the single greatest risk factor in the development of Alzheimer’s disease (AD). AD is the 6^th leading cause of death in the U.S. and 1 in 3 seniors dies with AD or another form of dementia. The hallmarks of AD pathology include amyloid-β (Aβ) plaques and neurofibrillary tangles of hyperphosphorylated tau proteins. These result from genetic mutations in amyloid precursor protein, presenilin 1, and presenilin 2 for those with familial AD, and while sporadic AD does not present with these genetic mutations, the AD progression follows a similar cascade leading to neurodegeneration. As a result of failed Aβ-targeted drug trials, emerging preclinical data highlight the potential role of microglial activation, neuroinflammation, and oxidative stress in disease development and progression (172,173).

The role of exercise in brain health

Aerobic exercise utilizes large muscle groups in a rhythmic cadence and is the most studied form of exercise in brain research. Aerobic exercise has been strongly associated with improved cognition in middle-aged and older healthy adults (174) and in older adults with mild cognitive impairment (175). The benefits of aerobic exercise are thought to be mediated by increased cerebral blood flow (176), enhanced neurogenesis (177), and the upregulation of exerkines discussed below.

Resistance training or strength training utilizing free weights, resistance machines, or bodyweight exercises has cognitive benefits for both healthy and cognitively impaired individuals. Indeed, there is solid evidence linking both muscle mass and muscular strength (both of which undergo an age-related decline without intervention) to cognitive function. Specifically, decreased muscle strength is associated with both elevated risk of mortality (178) and poorer brain functioning (179).

Flexibility and balance exercises stress the neuromuscular demands in exercise and can lead to cognitive improvement through the complexity of movement. Tai Chi has gained a lot of interest in its effects on brain health. In healthy older adults, Tai Chi training with its low-impact, slow, intentional movements showed greater cognition improvements compared to brisk walking (180). In contrast, static stretching lacks effect on cognition (181).

Brain health-promoting exerkines

Exercise is a certified promoter of brain health and while the mechanisms continue to be studied, several exerkines have been identified as key players in exercise neuroscience research. Exerkines are signaling molecules released in response to exercise that help mediate the systemic adaptations to exercise. Several exerkines have been found to impact brain health and are discussed below (Figure 5).

Figure 5.

Exercise in both human and animal models has identified potential makers for exercise neuroprotection: Brain-derived neurotrophic factor (BDNF), irisin cathepsin B (CatB), vascular endothelial growth factor (VEFG), ketone bodies, and lactate have been documented to increase following exercise and are associated with increased neurogenesis, decreased neuroinflammation and improved cognition.

Brain-derived neurotrophic factor (BDNF) is a neuronal growth factor important to synaptic plasticity, neuronal survival, development, and differentiation. BDNF binds to its TrkB receptor which autophosphorylates and activates several downstream pathways, including phosphoinositide 3 kinase (PI3K)/Akt, Ras/extracellular signal-regulated kinase (ERK) and phospholipase C_y/cAMP responsive element binding protein (CREB) pathways, all of which play a role in neuronal growth, plasticity, and survival (182).

Irisin is the cleaved and secreted portion of membrane protein, fibronectin type III domain-containing protein 5 (FNDC5). Irisin was first discovered as the peptide responsible for the conversion of white adipose tissue to brown adipose tissue (183). Recently, irisin has been proposed as the critical step in mediating the connection between physical activity and BDNF production in the brain. Simply put, exercise effects are mediated by peroxisome proliferator-activated receptor-γ coactivator-1α (PGC-1α) which in turn increases the expression of membrane protein, FNDC5. With exercise, FNDC5 is cleaved by PGC-1α to form irisin (184). This key paper by Wrann et al. (184) found that exercise increases gene expression of FNDC5 in the hippocampus and this has been confirmed by others; interestingly, the hippocampus is the only area in the brain that consistently shows an exercise-induced increase in FNDC5. In addition, peripheral delivery of FNDC5 via adenoviral vectors leads to increased BDNF gene expression in the hippocampus (185).

It is notable that CSF irisin levels show a positive correlation with age in the healthy age-matched controls but not in patients with AD (185). Peripheral administration of FNDC5/irisin via adenoviral vector rescued memory impairment in AβO-infused mice and lead to increases of FNDC5/irisin in the hippocampus (185). In addition, there is a strong correlation between CSF irisin and BDNF, Aβ_42, and cognition (Mini-Mental State Examination, MMSE) in human patients with AD (186). FNDC5 is predominantly expressed in the skeletal muscle tissue with additional high expression in the pericardium, intracranial artery, and rectum (187).

In addition to FNDC5/irisin, Cathepsin B (CTSB) is a myokine regulated by exercise that has been shown to increase BDNF in the brain (188). Treadmill training in rhesus monkeys and humans resulted in increases in plasma CTSB and hippocampal Ctsb gene expression (188). In CTSB knock-out mice, no protection was afforded with exercise (188). Cathepsin B is a lysosomal cysteine protease that acts to degrade proteins, thereby managing protein turnover in the body. Systemic CTSB can cross the blood-brain barrier and promote the expression of BDNF in the hippocampus, stimulating neurogenesis. However, Moon et al. also found that exercise directly induced an increase in cathepsin B gene expression in the hippocampus (188). Running led to an increased muscular expression of the CTSB gene in mice and an increase in CTSB in plasma from mice, rhesus monkeys, and humans following treadmill running training over 4 months. A recent study in older adults at risk for AD due to family history and genetic factors similarly found CTSB plasma levels to be increased following a 26-week aerobic exercise intervention and the changes in CTSB were positively associated with changes in cognitive function (189). Moon et al. (188) also performed studies in CTSB knockout mice and showed that mice lacking CTSB were resistant to the effects of voluntary exercise, specifically as regards hippocampal growth and improved cognition. It is not known if the myokine CTSB mediates enhanced cognitive function in response to exercise training. In CTSB knockout mice, Aβ levels and amyloid plaque deposition were increased when compared to wild-type mice (190).

Vascular endothelial growth factor (VEGF) is a key regulator of blood vessel development and is stimulated in conditions of hypoxia. VEGF is produced by many cell types, including tumor cells, macrophages, platelets, keratinocytes, and renal mesangial cells. VEGF is produced in skeletal muscles under the regulation of PGC-1α (183). Exercise-induced skeletal muscle release of VEGF has been associated with increased hippocampal cerebral blood flow (191). While it is still unclear if skeletal muscle-derived VEGF can cross the blood-brain barrier, it is clear that skeletal muscle VEGF is essential for neurogenesis in the hippocampus (191,192).

Ketone bodies are utilized as an energy source under conditions of fasting or intense exercise, they can be taken up directly into neurons by monocarboxylate transporters. Beta-hydroxybutyrate (BHB) is a ketone body produced in the liver and released into systematic circulation, it can cross the blood-brain barrier and stimulate the production of BDNF in the hippocampus via histone acetylation at BDNF promoters (193). When BHB is injected into the rat brain it has been shown to increase BDNF expression (193) and in neurons exposed to Aβ1–42, BHB protected the culture from Aβ toxicity (194).

Lactate. Long gone are the days of lactate being considered as a waste product of metabolism; it has a vast role in metabolism and as a sensing and signaling molecule in both health and disease. Lactate is a carbon vehicle utilized at all times by nearly all tissues (195–198). During exercise, the brain consumes lactate as a substitute for glucose (199). High-intensity interval exercise training leads to greater executive functioning benefits compared to moderate continuous exercise in healthy adults and is more effective at improving impaired cognition (200,201). While causality studies have not been conducted, correlative human studies indicate a role for lactate in exercise-induced neuroprotection (202). In an animal model of moderate brain injury, lactate administration 30 minutes following brain injury and continuing for 3 hours has been found to improve cognition compared to a sham injection (203). While the mechanisms of brain metabolic shifts with aging and neurodegeneration are further investigated there is significant evidence that lactate plays a role in neuroprotection. In addition to lactate serving as an important metabolic substrate in the brain, it can also act as a signaling molecule regulating angiogenesis and inflammation. Both high intensity (~90% $\dot{\text{V}}{\text{O}}_2\text{max}$) interval exercise and subcutaneous lactate injections lead to increases in brain vascular endothelial growth factor A (VEGFA) and capillary density in mice, which was absent in HCAR1 (hydroxycarboxylic acid receptor 1, lactate receptor) knockout mice (204). Long-term exogenous lactate in macrophages triggers a shift from the pro-inflammatory M1 phenotype to the anti-inflammatory M2 phenotype and may prevent an excessive inflammatory response (205–207).

Timing of exercise relative to the lifespan

Throughout the lifespan the brain enters different stages, i.e., developing (prenatal), peak growth (adolescent), peak volume (mid-life) and declining volume (late-life). The timing of exercise can vary its impact on brain structure and function. More research is needed to understand if the timing of exercise across the lifespan can elicit varying benefits to the brain health span. It is beyond the scope of this review but should also be recognized that the diurnal timing of exercise may affect physiological responses and be a factor in modulating its effects.

Prenatal.

Regular exercise during pregnancy has been shown to improve general intelligence, increase vocabulary, and improve early motor skills in children 1–5 years old (208–210). This holds true in the rodent models as well, with gestational exercise improving spatial learning and memory in their offspring (211,212), as well as enhanced hippocampal cell survival (213,214). In transgenic AD mice, sedentary offspring of exercised dames had reduced Aβ plaque compared to non-exercised offspring at 5 months of age (215). While the cellular and subcellular impacts of prenatal exercise are unknown, spatial transcriptomics of offspring has confirmed that prenatal exposure to a neurotoxicant impeded the development of neural progenitor and oligodendrocyte progenitor cells, impacting the subpopulation and ultimately disrupting the neural circuit function and behaviors (216).

Adolescent.

Brain plasticity is at its peak during the adolescent years, making this an important life phase to study. It is possible that exercise during this period could build a greater, long-lasting reserve in brain function. In 2011 Hopkins found that healthy rats exercising (running) in adolescence performed better in memory tests and had higher levels of BDNF than their sedentary controls (217). Additionally, Serra et al. determined that exercise initiated during adolescence in healthy rats resulted in an increased number of cortical and hippocampal neuronal cells and dendritic arborization that was maintained even after 60 days of sedentary behavior (218). While it is accepted that exercise during adolescence leads to long-term improved hippocampal function (219), to the authors’ knowledge this has not been tested in a model of neurodegeneration.

Mid-life

brain volume is at its peak and exercise during this period is important because exercise increases gray matter volume and enhances connectivity during mid-life. However, this is also the time that declines in brain volume begin, at a rate of 5% every 10 years beginning at 40 years old (220). Furthermore, research indicates that AD begins up to 20 years before diagnosis. Several studies have demonstrated strong positive correlations between mid-life fitness and reduced risk of dementia (221,222).

Late-life.

In a large meta-analysis of more than 160,000 participants, it was determined that older adults who participated in physical activity reduced their risk of developing AD by 45% (223). Also, in a population of over 700 older adults followed for 4 years, it was determined that individuals with low reported daily physical activity were 53% more likely to develop AD (224). Exercise in late life can improve cognitive function, reduce the risk of falls, and enhance overall quality of life.

Section 3 Summary

Promoting lifestyle changes is the greatest means to improve the brain health span. As reviewed above, exercise is a positive impactor of cognition by improving neurogenesis, reducing neuroinflammation and modulating key molecular pathways. Thus, exercise can extend the brain’s health span and delay the onset of neurodegenerative diseases (Figure 1). Benefits are possible at all life stages, from prenatal development to late-life and across all movement modes. Further research is needed to understand the mechanisms underlying exercise-impacts on brain health and to identify the most effective exercise regimens for different populations to fully realize the potential of exercise in expanding the brain health span. Additionally, practical recommendations for integrating exercise into daily life should be incorporated into public policy. As a consequence, we will improve not only individual brain health but also the overall health and well-being of the population.

Section 4. A Creative Monty Python-inspired Solution to Promote Physical Activity and Increase Calorie Expenditure

From the Introduction and Sections 1–3 it is clear that humanity faces a conundrum. We know that regular physical activity and potentially calorie deficits induced by dietary restriction can, via improved cardiovascular fitness and improvement across multiple systems, including metabolic control, immune regulation and brain health, help extend the health span. Tragically, however, exercise professionals as well as the medical and public health communities have failed to translate the associated health benefits of these principles into altered human behaviors on a population basis, as exemplified by global rates of physical inactivity remaining low and intransigent for at least the past 20 years (225). Certainly, in the U.S., this failing contributes substantially to the tragically low health span of ~66 years (226), which is below that of most other developed countries.

Although structural factors, including transportation infrastructure and urban planning, as well as lack of access to public spaces for recreation, present barriers to physical activity engagement, there is an underlying Darwinian neurophysiology and psychology that impacts behaviors to minimize perceived effort. This Theory of Effort Minimization in Physical Activity—TEMPA—suggests that all physical activity, including breathing, elicits an effort-related signal that is processed as a physiological cost to be minimized, and operates throughout the lifespan (227). Accordingly, evolution has adapted the anatomy, biomechanics (228,229) and physiology (227,230,231) beyond that of apes such that walking, as currently practiced, is 2–3 times more efficient in Homo sapiens versus chimpanzees (Pan troglodytes) (232,233; Figure 6A). Obesity is practically unknown in P. troglodytes. While efficiency is certainly an important goal in virtually all human endeavors, particularly during sports performance, exercise efficiency may be counterproductive from the standpoint of not invoking sufficient metabolic stress to promote healthful CRF adaptations and facilitate energy balance for obesity treatment/prevention. For example, in a study designed to identify factors predicting fat loss in an aerobic exercise intervention in premenopausal women, it was observed that the baseline walking energy cost was positively correlated with fat loss (234). Although the correlation was modest and explained only a small amount of the variance in fat loss, the results suggest that inefficiency may actually be a desirable trait from the perspective of energy balance (and also of health span—see below).

Figure 6.

A. The Problem: As humans have evolved, walking has become far more efficient, with a far lower calorie expenditure. B. In 1970 Monty Python’s Ministry of Silly walks comedic skit presented a putative solution to counter the increased efficiency of walking. At right is the actor and comedian John Cleese as Mr. Teabag demonstrating his iconic Silly Walk.

Moreover, because humans with obesity adopt their walking pace such that energetic cost per unit mass is surprisingly low and similar to their lean counterparts (235,236), their obesity (and also CRF) may be intractable to the prescription of regular walking as a countermeasure. What would clearly be advantageous from an energy expenditure and fitness perspective for modern humans is the exact opposite to TEMPA: Namely PEMPA – the Practice of Effort Maximization in Physical Activity.

An evolutionary warning

It may be hyperbolic but, across evolutionary time, the humble sea squirt, Clavelina moluccensis, offers a peek into one frightening potential future for the human race if PEMPA is not successful. C. moluccensis begins life much like a tadpole, with a muscular tail, backbone, brain and an eye, happily swimming around after food. But, when it locates a convenient rock where the current delivers that food, it affixes to said rock, and promptly absorbs its brain, backbone and muscles to become simply a mouth and digestive system (237): its health span is truncated abruptly. Far too many humans in contemporary Western cultures affix themselves to their couch and gorge on an electronically-summoned stream of high-calorie foods. Their muscles atrophy from lack of use, the television and social media streaming-devices promote cognitive dysfunction, and their fat stores hypertrophy. A future not unlike the humans on the Axiom spaceship in the Pixar blockbuster Wall-E seems within the realm of possibility.

The solution

In opposing TEMPA any PEMPA strategy must check many boxes. It must be accessible, doable, entertaining and interesting enough to be adopted and maintained and have a sufficiently high energy expenditure to increase physical fitness and redress any cumulative calorie imbalance. Ideally, it will entail performing some activity that every able-bodied person can do and, so as not to incur time constraints, replace rather than add to current activities and meet current physical activity guidelines. One strategy considered was to replace upright walking with walking on all fours like chimpanzees and capuchin monkeys (238). However, the social indignity, public health concerns and upper body weakness of modern humans render this idea untenable. Alternatively, backward walking increases energy expenditure compared to forward walking (239,240), but has obvious limitations for reasons of safety.

Because time spent in physical activity has not changed appreciably during the past half-century, and perceived lack of time is frequently cited as a primary reason to not exercise, the solution to TEMPA must be PEMPA that does not increase overall time requirement to meet recommended physical activity guidelines. Accordingly, three of the authors (GAG, DCP, SSA) conscripted the heretofore unrecognized comedic genius of Monty Python’s Ministry of Silly Walks (MoSW) to help address the inactivity (and obesity) epidemic that erodes the health span (241,242; Figures 6B & 7).

Figure 7.

A. Oxygen uptake (V̇O₂; ml · kg⁻¹ · min⁻¹) during participants’ usual walking (normal gait) and inefficient walking (Silly Walking, Mr. Teabag style) in 13 men and women (data from Gaesser et al. 242). For comparison, V̇O₂ during walking at 2.0 mph, 3.0 mph, and 4.0 mph in healthy adults (n = 40) is depicted by blue symbols (Data from Sawyer et al. 243) B. Association between energy expenditure (kcal/min; 1 kcal=4.18 kJ) and body mass (kg) for participants’ usual walking (dashed line) and the Mr. Teabag Silly Walking (solid line). Pearson correlations (r): participants’ usual walking (r=0.90); Mr. Teabag walk (r=0.81) (data from Gaesser et al. 242).

Specifically, the hypothesis was tested that John Cleese’s walk, as the lanky-legged Mr. Teabag, was sufficiently inefficient to raise the metabolic rate substantially. A cohort of men and women ranging in age from 20–70 years were trained by practicing to mimic the Mr. Teabag and other MoSW walks from the relevant videos which were displayed on computer monitors. These subjects’ energy expenditures were subsequently measured by indirect calorimetry at freely-chosen walking speeds and compared with that of their normal gait. The silly walk of John Cleese, operating within solid PEMPA principles, raised the energetic cost of ambulation by ~2.5 fold, averaging a V̇O₂ of nearly 30 ml · kg⁻¹ · min⁻¹ (Figure 7A), such that exchanging 1 min of normal walking with 1 min of Cleese silly walking increased energy expenditure by an average of 8.0 kcal/min for men and 5.2 kcal/min for women (242).

The results of these experiments are relevant to countering two major public health issues—physical inactivity and obesity – that are directly intertwined with the health span. Specifically: 1) the metabolic equivalent (MET) value of silly walking constitutes vigorous exercise (~same as jogging/running at over 5 mph), so it is likely to be better for improving CRF and enhancing the health span, and 2) if substituted for just ~22–34% of the daily steps of adults worldwide, it has the potential to eliminate future weight gain. Indeed, had such a strategy been implemented in the 1970s, it may have prevented—or at least greatly attenuated—the international obesity pandemic of the past four decades (242).

The global obesity pandemic is a result of small imbalances in energy intake and expenditure that have occurred progressively over time. The energy gap estimated to account for the greater annual weight gain and increasing obesity prevalence observed during the past 40 years is less than 100 kcal/day and likely to be considerably lower for most individuals (244). For our men and women, who mostly fell within the normal weight BMI category of 18.5 to ≤25 kg/m², ~12–19 min/day of Cleese silly walking would expend an additional 100 kcal/day, sufficient to close the energy gap. And even less time would be necessary for individuals with higher body mass (Figure 7B). The heaviest subject, at 127.5 kg, could expend an additional 100 kcal/day with only ~8 min of Cleese silly walking per day!

The increased fitness alone would have enormous health span-related impacts because the health benefits of physical activity are independent of BMI status and weight loss (53,245). In fact, aerobically fit adults with obesity have lower all-cause mortality risks compared to unfit adults without obesity and have a mortality risk that is not much different than that of fit adults without obesity (53,245). For this reason alone, the Silly Walks initiative will have massive benefits for health span (as well as lifespan).

The recent decades during which contemporary Western populations have not endured food deprivation has been too short a period for evolution to rewire our brains to perform physical activity for the explicit purpose of energy balance (246). This likely explains why public health campaigns to increase physical activity have largely been ineffective. However, as discussed above, the Silly Walks initiative would call for replacing a currently low- with a high-energy expenditure activity. We contend that this would be less likely to result in downregulation of spontaneous physical activity and unlikely to diminish the number of steps currently being taken each day (presumably because these steps are vital to an established routine of daily living).

Silly Walking as a population wide health span initiative

Silly walking will improve the health of both silly and non-silly walkers. Using the PEMPA model based on MoSW principles has significant public health implications for both the silly and the non-silly walkers. Silly walking is likely—at least initially—to evoke laughter among non-silly walkers. This will confer significant health benefits on them as well, because laughter has been documented to have beneficial effects on blood pressure (247), vascular function (247,248), arterial stiffness (247), and diabetic complications (249), and has been shown to improve pain tolerance (250), and cardiovert arrhythmias (251)! This likely explains why frequent laughter is associated with reduced risk of CVD (252). What other public health intervention, except possibly stopping smoking, can boast of healthful effects in the user and the non-user? We recommend urgent adoption of Silly Walking as a public health measure to combat the twin pandemics of obesity and CVD in silly and non-silly walkers and extend the health span.

We cannot, at the present time, advocate extension of the Silly Walks initiative to other forms of exercise such as swimming or cycling. Not only are they largely body mass independent but we note that both may be lethally hazardous to the exponent.

One potential behavioral barrier to adopting PEMPA is the very nature of high-intensity activity exemplified in the Cleese Silly Walk. Vigorous intensity tasks produce an effort-related signal that would obviously be processed as such, thus acting against PEMPA in favor of TEMPA. The solution, however, is to perform high-intensity tasks, such as the Cleese Silly Walk, in brief bouts lasting no more than ~1–2 min so as to minimize the effort-related signal. Rating of perceived exertion (RPE) during exercise is both intensity- and time-dependent (253). In adults performing a high-intensity interval exercise (HIIE) session of 16×1-min intervals at a power output calculated to elicit 95% of maximum heart rate followed by 1 min of low-intensity active recovery, RPE during the first couple of HIIE intervals was only ~5 on the 10-point modified Borg scale (253,254). Thus, intermittent bouts of Cleese-style Silly Walking, lasting as short as ~1 min, could potentially optimize PEMPA without triggering a TEMPA override.

Limiting PEMPA-motivated activity to only 1–2 min at a time would not compromise benefits because exercise energy expenditure (255) and cardiovascular benefits (256) are essentially the same regardless of whether exercise is performed all at one time or distributed in short bouts during the day. Cardiovascular benefits of exercise are evident even with activity bouts lasting only 1–2 min (257).

Section 4 Summary

In 1970 John Cleese and the MoSW skit might have unwittingly touched on a powerful way to enhance cardiorespiratory fitness and thus increase the health span. Promoting the inefficiency of physical activity and movement (PEMPA) that we routinely undertake, and requiring no additional time commitment, might buttress the public health imperative to popularize regular physical activity in a comedic fashion. Concomitantly, Silly Walking may help redress the energetic imbalance created in evolutionary time by TEMPA and exacerbated in recent decades due to the profligacy of labor- and movement-saving devices combined with ready access to cheap calorie-dense foods.

As a necessary caveat, the potential risks for this type of intervention, especially in older populations must be considered. These risks would include the risk of falls which in-and-of themselves are a significant factor when it comes to morbidity and mortality in older populations, particularly in individuals with low fitness levels, movement disorders, joint disease, or other health conditions. However, these individuals may be able to adopt less risky ways to move their own bodies in more energetic ways to derive at least some benefit from PEMPA (i.e., Silly Walking Lite).

Overall Conclusions

In the nearly 50 years since Robert Butler, Director of the National Institute on Aging, characterized exercise as the neglected therapy by likening it to essential medicine, the science to validate his visionary statement is robust and ever-expanding. In 2007, the “Exercise is Medicine” initiative jointly launched by the American College of Sports Medicine and the American Heart Association (AHA) (258) advised that physical activity be assessed as a vital sign, and in 2016 the AHA recommended that CRF be measured in clinical practice as a vital sign (259). Despite this scientific enlightenment on the polypill-effects of exercise, physical activity participation remains low, with far too many individuals of all ages failing to meet physical activity recommendations. A major challenge is to reverse this public health crisis that negatively impacts both health span and lifespan.

One reason for the low physical activity participation rates may be that exercise has traditionally been viewed primarily as energy expenditure, i.e., to “burn calories” as a means to lose weight. This has proved futile, as obesity prevalence continues to increase despite correspondingly similar increases in weight loss attempts. Physical activity promotion without a weight loss focus may be preferred because physical activity and CRF are superior to intentional weight loss for reducing morbidity and mortality, and likely will result in a greater impact on health span while avoiding the pitfalls associated with weight cycling.

A weight-neutral approach to physical activity promotion should not be interpreted as diminishing the promising nutritional strategy of CR. Results from RCTs consistently support modest CR to attenuate biological aging and enhance health span. The CALERIE^™ trial is perhaps the most notable example. It must be emphasized, however, that weight loss was not the goal in the CALERIE^™ trial and the magnitude of CR maintained during the trial was much less than that advocated in typical commercial weight loss plans. Moreover, improvements in some metrics of cardiometabolic health and biological aging in the CALERIE^™ trial were independent of weight loss. Whether exercise can enhance the effects of CR is unknown, but considering the well-established benefits of exercise on most organ systems and in preventing or mitigating disease processes, it is reasonable to infer that exercise may have complementary or additive benefits with CR on the aging process.

A major concern is whether exercise and CR can be sustained long-term. In the CALERIE^™ trial, the level of calorie restriction achieved was only ~9% during last 18 months of the 2-year trial (151). In a subgroup of participants who were followed after the intervention, the level of CR fell to 5% and participants regained 54% of initial weight loss (260). Thus, a more modest CR goal might be advised at the outset to enhance long-term adherence. Also, CR may be more suitable for young and middle-aged adults because older adults may be at greater risk of functional decline with CR (261).

With regard to CRF, it is important to note that the dose of exercise necessary to maintain CRF is much less than the dose required to increase CRF. In an elegantly designed series of experiments in the 1980s, Hickson et al. (262) demonstrated that ~10–20% increases in CRF achieved after 10 weeks of training could be maintained during 15 weeks of two-thirds reduction in either training frequency (6 days/week → 2 days/week) or exercise session duration (40 min/session → 13 min/session), as long as the intensity of exercise training was maintained. Such a strategy might be appealing to individuals who cite “lack of time” as a reason to not exercise, or to individuals inclined to add humor to their exercise routine: 13 min/day of Mr. Teabag-style silly walking is more than sufficient to meet the current physical activity guidelines for vigorous-intensity activity (75 min/week). Breaking up those 13 min into shorter bouts could potentially optimize PEMPA without triggering a TEMPA override, while still inducing substantial health benefits.

Exercise can extend the brain’s health span and delay the onset of neurodegenerative diseases. Benefits are possible at all life stages, from prenatal development to late-life and across all movement modes. This has special relevance considering plasma proteomics data from the UK Biobank cohort showing that youthful brains promote health span and longevity (263). Because human organs age at different rates (264), exercise has the potential to enhance the health span of all organs due to its impact on virtually every cell in the body.