Exercise Attenuates the Major Hallmarks of Aging

Abstract

Regular exercise has multi-system anti-aging effects. Here we summarize how exercise impacts the major hallmarks of aging. We propose that, besides searching for novel pharmaceutical targets of the aging process, more research efforts should be devoted to gaining insights into the molecular mediators of the benefits of exercise and to implement effective exercise interventions for elderly people.

Introduction

The number of people aged ≥60 years worldwide is expected to nearly triple by 2050, with the “oldest old” group (≥85 years) being the most rapidly expanding segment. A growing challenge is to maintain elderly people independently until end of life. In them, functional independence is directly dependent on physical fitness, i.e. “the ability to carry out daily tasks with vigor and alertness, without undue fatigue and with ample energy to enjoy [leisure] pursuits and to meet unforeseen emergencies.”¹ In turn, physical fitness is determined by several measurable health-related phenotypes, including mainly cardiorespiratory fitness and muscle function.¹ Among the physiological changes associated with aging, those affecting the cardiorespiratory and vascular system and skeletal muscles most affect physical fitness, whereas exercise can attenuate multi-system age declines, as explained below.

Summary of the Impact of Aging on the Main Physical Fitness—Related Phenotypes

Aging and cardiorespiratory fitness

Maximal oxygen uptake (VO₂max; sometimes referred to as maximal aerobic capacity or simply aerobic capacity or aerobic endurance) is a main indicator of cardiorespiratory fitness. There is variability among studies,^2–4 but the average rate of VO₂max decline in old people is ≥4–5 mL·kg⁻¹·min⁻¹/decade.⁵ Mostly reduced maximal cardiac output^6–10 but also a decline in maximal arteriovenous oxygen difference (a-vO₂ diff) (minus ∼3%/decade)^7,11 contribute to age reductions in VO₂max.^12,13 Aging skeletal muscles have a low capacity to use O₂ due to several factors, such as decreased muscle mass (see below),^14,15 increased peripheral resistance,¹⁶ reduced muscle capillary density,¹⁷ endothelial dysfunction,¹⁸ changes in skeletal muscle microcirculation,¹⁹ and reduced muscle oxidative capacity.²⁰

Aging and muscle function

Muscle mass usually starts to decline after 25–30 years of age,^21,22 such that on average 40% of muscle mass is lost by 80 years.^22,23 In turn, a quantitative loss in muscle cross-sectional area is a major contributor to the decrease in muscle strength seen with advancing age, i.e., after 60–70 years of age.²⁴ The term “sarcopenia” was originally created to refer to age-related loss of muscle mass with consequent loss of strength.²⁵ There are now four international definitions of sarcopenia.^26–29 In essence they all agree, requiring a measure of impaired walking capability (either low gait speed or a limited endurance [distance] in a 6-min walk), together with an appendicular lean mass of less than 2 standard deviations of a sex- and ethnically-corrected normal level for individuals 20–30 years old. Sarcopenia (see below for details on signaling pathways involved) occurs due to several age-related factors, such as gradual muscle denervation,^23,30 diminished satellite cells,³¹ low muscle protein synthesis,³² low anabolic hormone levels,³³ malnutrition,³⁴ increased pro-inflammatory cytokines,³⁵ oxidative stress,³⁶ mitochondrial dysfunction,^37–40 and physical inactivity.⁴¹ Although there are differences between studies, on average, 5%–13% and 11%–50% of people aged 60–70 years and ≥80 years, respectively, suffer sarcopenia,^42–47 with a higher prevalence (68%) reported in nursing home residents ≥70 years.⁴⁸ Sarcopenia needs to be differentiated from “cachexia,” a combination of both muscle and fat loss that is usually attributable to an excess of catabolic cytokines associated with a disease process, e.g., cancer.^49–51 Sarcopenia is linked with increased disability, falls, hospitalization, nursing home admission, and mortality.^48,52–54

Frailty and disability

A consequence of the aforementioned effects of aging on cardiorespiratory and muscle fitness (especially sarcopenia), alone or in combination with co-morbidities such as neurodegeneration, is the “frailty syndrome.”⁵⁵ Although no single operational definition of the frailty syndrome has been agreed upon, two major definitions with proposed assessment tools have predominated over the past decade—the frailty phenotype, also known as Fried's definition,⁵⁶ and the frailty index.⁵⁷ The most widely cited is the frailty phenotype, which is operationalized as a syndrome meeting three or more of the following five phenotypic criteria: Weakness as measured by low grip strength, slowness (low walking speed), low level of physical activity, low energy or self-reported exhaustion, and unintentional weight loss.⁵⁶ A prefrail stage, in which one or two criteria are present, identifies a subset at high risk of progressing to frailty. Older individuals with none of the above five criteria are classified as nonfrail. The frailty index was developed on the basis of a comprehensive geriatric assessment by counting the number of deficits accumulated, including diseases, physical and cognitive impairments, psychosocial risk factors, and common geriatric syndromes other than frailty.^57,58 Whereas the frailty index may have clinical utility in risk assessment and stratification, it is less clear that it adds significant value to comprehensive geriatric assessment.⁵⁹

Frailty is an important and growing problem in aging western populations, because it potentially affects 20%–30% of adults older than 75 years.⁶⁰ For instance, the prevalence of prefrail and frail individuals is high in the Spanish population (41.8% and 8.4%, respectively) and increases with age, according to a recent a population-based study conducted on 2488 individuals aged 65 years and older in a central area of the country.⁶¹ On the other hand, frailty could eventually result in disability,⁶², i.e., “difficulty or dependency in carrying out activities necessary for independent living, including roles, tasks needed for self-care and household chores, and other activities important for a person's quality of life.”⁶³

For the above-mentioned reasons, it is of paramount medical importance to attenuate age-related declines in physical fitness. Yet no single drug can reverse the age-related loss of physical fitness because none benefits all of the systems involved, whereas regular exercise has multi-system anti-aging effects (see below and Fig. 1, left column for a summary).

FIG. 1.

Summary of the main anti-aging effects of regular exercise vs. aging effects at the multi-systemic (left; see Chodzko-Zajko et al.⁶⁴ for an in-depth review) and cellular level (right; see text). AKT, protein kinase B; AMPK, AMP-activated protein kinase; ASC, apoptosis-associated speck-like protein caspase; AUF1, AU-binding factor 1; BDNF, brain-derived neurotrophic factor; FoxO3a, human protein encoded by the FOXO3 gene; Glut 4, glucose transporter type 4; HATs, histone acetyltransferases; HDACs, histone deacetylases; HRV, heart rate variability; IGF-1, insulin-like growth factor 1; IL, interleukin; jmjC, jumonji C proteins; LSD, lysine-specific histone demethylase; miR, micro-RNA; mtDNA, mitochondrial DNA; mTOR, mammalian target of rapamycin; NK cell, natural killer cell; NLRP3, NOD-like receptor protein 3; PBMCs, peripheral blood mononucleated cells; PDK4, pyruvate dehydrogenase kinase isoenzyme 4; PGC-1, peroxisome proliferator-activated receptor gamma coactivator 1; PPAR-δ, peroxisome proliferator-activated receptor δ; PTMs, post-translational modifications; Qmax, maximal cardiac output; ROM, range of motion; SIRT, sirtuin; TERT, human telomerase reverse transcriptase. Color images available online at www.liebertpub.com/rej

Exercise Attenuates Fitness and Multisystem Age-Related Declines

Benefits of exercise (particularly aerobic exercise) in cardiorespiratory fitness/cardiovascular disease

Regular exercise, particularly dynamic exercise of moderate intensity (≤70% of VO₂max or ≤80% of maximum heart rate) involving mostly the aerobic energy pathway and large muscle mass (e.g., brisk walking, bicycling) attenuates age declines in cardiorespiratory fitness (see Chodzko-Zajko et al.⁶⁴ for an in-depth review and experts' recommendations). This type of exercise, commonly referred to as “aerobic exercise” (or “endurance exercise”), has a restoring effect on an important risk factor of cardiovascular disease (CVD), i.e., endothelial dysfunction.^65–67 It also increases endothelial nitric oxide (NO^•) production and thus vascular tone regulation. Regular bouts of exercise-increased laminar flow activate (through phosphorylation via protein kinase B, Akt) endothelial NO^• synthase while attenuating NO^• degradation into reactive oxygen species (ROS) and reactive nitrogen species.⁶⁸ Together with increased angiogenesis (see further below), an additional benefit of aerobic exercise in endothelial health is stimulation of macrophage-mediated reverse cholesterol transport through activation of peroxisome proliferator-activated receptor gamma (PPARγ).^69,70 Yet aging autonomic dysfunction has a synergistic effect together with endothelial dysfunction in increasing CVD risk⁷¹ and raises the risk of a leading cause of death in most industrialized countries, sudden death due to ventricular fibrillation.⁷²

During aging, the sympathetic nervous system (SNS) outflow to several peripheral tissues increases to stimulate thermogenesis and thus to prevent increasing adiposity.⁷³ Chronically activated SNS has deleterious effects on the cardiovascular system, i.e., reduced leg blood flow, increased arterial blood pressure, impaired baroreflex function and hypertrophy of large arteries; it can also increase insulin resistance, thereby raising the risk of metabolic syndrome.^74,75 Importantly, aerobic exercise training (e.g., brisk walking) has a beneficial, dose–response effect in attenuating aging autonomic system dysfunction, with trained elderly individuals showing similar baroreflex function compared with their moderately active younger peers.⁷⁶ Heart rate variability (HRV), a marker of autonomic function and a powerful predictor of CVD outcome (high HRV is associated with a better prognosis), increases with aerobic exercise training in old people.⁷⁷ Reduced angiotensin II, increased NO^•, and mainly improved vagal modulation and decreased sympathetic tone are implicated in the beneficial exercise effects on HRV.⁷⁸

Benefits of exercise (particularly resistance exercise) in the aging muscle function

Training programs, especially if including resistance (strength) exercises (i.e., movements, such as weightlifting or exercises with resistance bands, performed against a specific external force that is regularly increased during training) are especially useful for improving muscle mass and/or strength in the elderly,⁷⁹ including in the oldest old⁸⁰ (see Table 1 for a summary of controlled exercise interventions [mostly based on resistance exercises] in the oldest old). In general, published resistance exercise interventions in old people had a duration ranging from 4 to 48 weeks and were in agreement with accepted or “traditional” exercise recommendations for older people,⁸¹ with the usual weekly schedule including two to three nonconsecutive sessions with one to three sets of 10–15 repetitions of classic weightlifting exercises, such as leg presses.

Table 1.

Summary of Controlled Exercise Interventions (Mostly Based on Resistance Exercises) in the Oldest Old

				Training
Reference	Group	n (% women)	Mean age (SD or range in years)	Length (weeks)	Type	Frequency (sessions/week)	Sets	Exercises/session (repetitions per (×) exercise)	Intensity/duration	Effects
356	Exercise	46 (72%)	80 (7)	24	Low intensities+	3		Flexibility, balance, coordination, movement speed and strength of all major muscle groups	5–15 min (0–12 weeks)	↑ Knee extension strength ↑ Fast walking speed ↑ Total free mass
					Strength	3	1–3	6–12× knee extension	65–100% 1RM (12–24 weeks)
							1–3	6–12× knee flexion	65–100% 1RM (12–24 weeks)
							1–3	6–12× seated bench press	65–100% 1RM (12–24 weeks)
							1–3	6–12× seated row	65–100% 1RM (12–24 weeks)
							1–3	6–12× leg press	65–100% 1RM (12–24 weeks)
							1–3	6–12× biceps curl	65–100% 1RM (12–24 weeks)
	Control	44 (77%)	81 (8)	24		3		Flexibility, balance, coordination, movement speed and strength of all major muscle groups		↑ Knee extension strength ↑ Fast walking speed ↑ Total free mass (Significantly less pronounced than in the exercise group)
357	Exercise	11 (91%)	88.6 (86–95)	12	Strength	3	8 8	8x 90° knee flexion 8x 90° knee extension	50–80% 1RM 50–80% 1RM	↑ Extensor and flexor muscle strength
	Control	12 (92%)	90.6 (86–95)	12	No PA	2		Social activities, including group gatherings		↑ Extensor and flexor muscle strength (Significantly less pronounced than in the exercise group)
358	Exercise	11 (NA)	93.4 (3.2)	12	Strength+	2		8–10x upper body (1 exercise: seated bench press)	40–60% 1RM High velocity	↑ Quadriceps femoris CSA ↑ Hand grip ↑ Hip flexion strength ↑ Knee extension strength ↑ Upper-body 1RM ↑ Lower-body 1RM ↑ Maximal power at 30% 1RM ↑ Maximal power at 60% 1 RM ↓ Falls incidence
					Balance +  Gait +  Stretching (cold down)			8–10x lower body (bilateral leg extension and bilateral knee extension)	40–60% 1RM High velocity 10 min 5 min
	Control	13 (NA)	90.1 (1.1)	12	Mobility	4		Active and passive movements	30 min	= Quadriceps femoris CSA ↓ Hand grip strength ↓ Hip flexion strength ↓ Knee extension strength
359	Lifestyle integrated functional exercise	107 (55%)	82.8 (4.48)	48	Functional exercise	Several		Set of exercise		Ankle strength, knee strength and function effects were significantly better than in control group
	Exercise	105 (54%)	84.0 (4.38)	48	Strength +  Balance	3		6× lower limb strength 7 exercise	“Hard zone”	Ankle strength, knee strength and function effects were significantly better than in control group
	Control	105 (55%)	83.5 (3.81)	48	Gentle +			2 sessions, 1 booster session and 6 follow-up phone calls		Ankle strength, knee strength and function effects were significantly lower than in exercise or functional exercise group
					Flexibility			e.g., hip rotations
360	Exercise	15 (NA)	86.7 (6.1)	10	Warm-up+	3			10 min	↑ Muscle-strength ↑ BMI ↑ Mean lean body mass (no significant)
			(77–98)		Strength+		1	10–15× muscle group	Elastic therabands and soft weights
					Balance+			Balls, balance disc, and blocks	10 min
					Cool-down				5 min
	Control	15 (NA)	86.9 (5.7) (77–94)	10						No effect ↓ Mean lean body mass (no significant)
361	Exercise	25 (64%)	86.2 (1.0)	10	Strength	3	3	8× lhip extensor	80% 1RM	↑ Muscle-strength
			(72–95)				3	8× knee extensor	80% 1RM	↑ Thigh-muscle CSA
										↑ Stair-climbing power
	Exercise+	25 (64%)	87.2 (1.2)	10	Strength+	3	3	8× hip extensor	80% 1RM	Idem exercise group
	Nutritional supplement		(76–98)		240 mL liquid + CHO (60%) + fat (23%) + soy-based protein (17%)		3	8× knee extensor	80% 1RM
					Aerobic or flexibility activities			3 activities (e.g., walking)
	Nutritional supplement +Placebo activities	24 (71%)	85.7(1.2) (75–97)	10	240 ml liquid + CHO (60%) + fat (23%) + soy-based protein (17%) Aerobic or flexibility activities	3				No effect
								3 activities (e.g., walking)
	Placebo activities +Placebo	26 (54%)	89.2 (0.8) (78–98)	10	Aerobic or flexibility activities	3		3 activities (e.g., walking)		No effect
	supplement				Minimally nutritive (4 kcal)
362	Exercise + Traditional balance training	12 (50%)	81.1 (6.0) (69–95)	12	Strength+	2		Larger over extremity muscles (pulley stations)	10–15 RM	↑ Strength knee extension ↑ Sitting to standing test ↑ Static Balance
								Large under extremity muscles (leg press and pulley stations)	10–15 RM
								Step machine	100 steps level 2 150 steps level 2 100 steps level 3 150 steps level 3
					Aerobic +  Balance			Static cycle Standing, one-legged balance and walking on different surfaces (open and closed eyes)	> 15 min and > 3km
	Exercise  +  Visual computer feedback	15 (80%)	81.5 (7.7) (69–95)	12	Strength+			Larger over extremity muscles (pulley stations)	10–15 RM	↑ Strength knee extension ↑ 6 minute walk test ↑ Training-specific performance
								Large under extremity muscles (leg press and pulley stations)	10–15 RM
							Step machine	100 steps level 2 150 steps level 2 100 steps level 3 150 steps level 3
					Aerobic+			Static cycle	> 15 min and > 3km
					Balance			Visual computer feedback
363	Exercise	31 (NA)	82.2 (4.1) (75–90)	12	Strength	3	3	10x knee extensions	70–90% of maximal	↑ Leg extension strength ↑ Knee extension strength ↑ Knee flexion strength ↑ Ankle plantar flexion strength ↑ Handgrip strength ↑ Functional motor performance ↑ Balance Improvements persisted during 3-month follow-up with only moderate losses
							3	10× hip extension	70–90% of maximal
							2	10× left lifts	70–90% of maximal
							2	10× right lifts	70–90% of maximal
							2	15 lifts of bilateral plantar flexion (2–4 cm support)
					Functional-Balance training +  Physiotherapy	2		Multifactorial activities e.g., massaging or stretching	45 min per training session 25 min
	Control	26 (NA)	82.1 (4.8) (75–90)	12	Placebo activities +  Physiotherapy	3		Motor activities (e.g., flexibility, calisthenics, ball games and memory tasks)	1 hr	No effect during intervention and follow-up
364	Exercise	62 (74%)	82.3 (6.6)	12	Strength	3		Relevant muscle group	70–80% 1RM	↑ Knee extension strength ↑ Knee flexion strength ↑ Ankle flexion strength ↑ Handgrip strength ↑ Functional performance
	Control	60 (73%)	82.9 (7.0)	12	Placebo activities	2		Flexibility exercise, calisthenics, low-intensity training with hand-held weights and ball games		No effect or ↓
365	Exercise	15 (80%)	88.5 (1.0) (85–98)	12	Strength	3	1	Upper and lower extremities	5–10 min	↑ 1RM
							3	8x knee flexion+		↑ Isokinetic knee extension
								8x knee extension	50% 1RM (0–2 weeks) 80% 1RM (2–12 weeks)	↑ Isometric knee extension
	Control	15 (73%)	90.9 (1.1) (85–98)	12	No PA					= 1RM  = Isokinetic knee extension  = Isometric knee extension
366	Exercise in long-term care-high mobility	22 (NA)	81.3 (5.3)*	4	Warm-up/stretching+				5 min	↑ Mobility ↑ Balance ↑ Flexibility ↑ Knee strength ↑ Hip strength
					Aerobic +  Strength			Walking	10–15 min
					Balance Cool-down/stretching		1–2 1–2	2–10× lower body 2–10× upper body	Soft-weights and Theraband
									10 min 5 min
	Exercise in long-term care-low mobility	14 (NA)	81.3 (5.3)*	4	Warm-up/stretching +  Aerobic+				5 min	↑ Mobility ↑ Balance ↑ Flexibility ↑ Knee strength ↑ Hip strength
					Strength			Walking	< 1 min − 15 min
					Balance Cool-down/stretching		1–2 1–2	2–10× lower body 2–10× upper body	Soft-weights and Theraband 10 min 5 min
	Range of motion	32 (94%)	81.3 (5.3)*	4	Introductions				10 min	↑ Shoulder strength  ↓ Hip Strength  ↓ Mobility  ↓ Functional ability
					Vocal exercises				5 min
					Memory games				5 min
					Range of motions			Fingers, hands, arms, knees, ankles
					Relaxation exercises				5 min
367	Exercise	19 (95%)	91.7 (6.3)	12	Aerobic	5		Walking	↑ 10% weekly (maximum 30 min) (Baseline intensity defined by each participant)	↑ Maximal walk endurance ↑ Maximal walk distance (significant)  = Hand-grip strength  = BMI
	Control	12 (100%)	89.3 (4.4)	12	Social control	1			30 min	= Maximal walk endurance ↑ Maximal walk distance (no significant)  = Hand-grip strength  = BMI
368	Exercise	36 (81%)	83.7 (8) (67–98)*	24	Warm-up+	2			10 min	↑ Quadriceps strength
					Strength +  Cool-down			Isometric exercises major muscle groups	25 min (Intensity was graduated by the n° of repetitions and the gravity-resistance) 10 min
	Control	29 (90%)	82 (9.6) (67–98)*	24	Reminiscence	2		Reminiscence therapy	45 min	↓ Quadriceps strength
369	Exercise	21 (90%)	81.4 (3.4) (75–96)*	24	Stretching +  Range of movements exercises		1	5–10× each exercise	Support of a chair or work surface	No significant differences between the groups with regard to changes in any of the outcome variables (grip strength, ‘up & go’ test, ‘sit to stands' test, functional reach, spinal mobility, Barthel index, Philadelphia Geriatric Center Morale Scale) Trend toward improvement in comparison with control group (‘sit to stand ‘tests and ‘up & go’ tests)
					Strength			5–10× yellow-blue	Therabands
	Exercise	20 (85%)	82.9 (4.4) (75–96)*	24	Mobility					No significant differences between the groups with regard to changes in any of the outcome variables (grip strength, ‘up & go’ test, ‘sit to stand’ test, Functional reach, Spinal mobility, Barthel Index, Philadelphia Geriatric Center Morale Scale) Trend towards improvement in comparison with control group (‘sit to stand’ and ‘up & go’ tests)
	Control	28 (89%)	81.9 (4.7) (75–96)*	24	Health education				30 min	No significant differences between the groups with regard to changes in any of the outcome variables (grip strength, ‘up & go’ test, ‘sit to stand’ test, functional reach, spinal mobility, Barthel index, Philadelphia Geriatric Center Morale Scale)
370	Exercise	39 (67%)	82.2 (1.3)	8	Strength +  Balance +  Functional mobility	3		Combined progressive exercises	1 hour	↑ 6-min walk test
	WBV	38 (68%)	80 (1.4)	8	Vibration +  Strength +  Balance +  Functional mobility	3	5	Whole body vibration	1 min 15–30 Hz 2–8 mm peak-to-peak	↑ 6-min walk test (greater improvements than exercise group)
371	Exercise	94 (81%)	87 (8)	32	Strength	5		Upper body resistance training (arm curls or arm raises)	Implement every 2 hr for a possible total of 4 care episodes per day	↑ Arm raise ↑ Arm curl ↑ Stands 30 sec
					Aerobic			Walking or wheeling	75–90% of the maximal distance
	Control	96 (86%)	88 (7)	32	UCG					↓ Meter walked or wheeled
92	Exercise	20 (80%)	92 (2) (90–96)	8	Mobility exercise +  Stretching+	2	1	Major muscle groups	40.45 min	↑ 1RM leg press ↑ FAC (in participants with FAC > 3) ↓ Falls (1.2 fewer per participant than in the control group)
					Aerobic + Strength			Rest periods
						3 3	1 2-3 1-8	Cycling (cycle ergometer) Leg press exercise 10× biceps curls, arm extension, arm side lifts, shoulder elevations, seated bench presses and calf raises	10–15 min; RPE:12–13 30–70% 1RM ( ↑ 5%weekly) 1–3 kg or low-medium resistance bands
	Control	20 (80%)	92 (2) (90–97)	8	UCG	5	1	Mobility exercises in major muscle groups	40–45 min	14 of 18 participants showed ↑ RM leg press
372	Exercise	34 (100%)	83.5 (4.1)	10	Warm-up+	2			15 min	= Abilities to carry out instrumental activities of daily living
					Strength+		1-2	8–30× knee extension 8–30× knee flexion 8–30× hip abduction 8–30× hip adduction	Gradually increased
					Functional exercises+		2	15× exercise (e.g., rising from a chair or elbow flexion using light loads (1–2 kg))	30 min
					Relaxation				15 min
	Control	34 (100%)	82.6 (3.7)	10	Home exercise program	2–3	2	15× functional exercises		= Abilities to carry out instrumental activities of daily living

Of note, the mean age of the subjects in all the studies was ≥ 80 years.

^*Data provided by the total number of participants.

BMI, Body mass index; CHO, carbohydrate; CSA, cross-sectional area; FAC, functional ambulation classification; NA, not available; PA, physical activity; RPE, rating of perceived exertion (Borg's conventional scale, 6–20 points); 1RM, one repetition maximum; SD, standard deviation; SO, social activities; TG, training group; UCG, usual care group; WBV, whole-body vibration.

Other interventions feasible in old people include “high-velocity resistance training,” i.e., focusing on speed of movement,^82,83 or even explosive-type heavy-resistance training, e.g., weightlifting exercises with a load equivalent to 75%–80% of one repetition maximum (1RM) and performed with maximal intentional acceleration of the training load during the concentric movement phase.⁸⁴ In general, besides being feasible and well tolerated even by the oldest old,⁸⁴ this alternative type of intervention would elicit similar⁸⁵ or even higher improvements in functional performance and disability compared with more traditional, lower-velocity resistance training.⁸⁶ Another approach is the use of a weighted vest, which has proved effective to improve perceived health in old people,⁸⁷ as well as lateral stability, lower-body muscular strength, muscular power, and leg lean mass (and thus likely to reduce risk fall) in postmenopausal women aged 50–75 years.⁸⁸ Preliminary findings also support the potential effectiveness of including a weighted stair climbing exercise in the training programs of old people.⁸⁹ Others, however, found no significant benefits of a home-based intervention using a weighted vest.⁹⁰

Benefits of exercise in the frail elderly

Although numerous studies have shown the benefits of resistance training in old people in general, less research has focused specifically on the frail elderly. This is an important issue because frailty status might influence the potential applicability and effectiveness of the different training programs. For instance, Faber et al. found that fall-preventive, moderate-intensity, group exercise programs (mostly based on walking and balance exercises) had positive effects on fall risk and physical performance in pre-frail but not in frail elderly (mean age 85 years).⁹¹

Fiatarone et al. reported that physically frail, oldest old men and women improved their leg muscle strength outcomes by 220% after a 10-week strength training program (three sets of eight repetitions at 80% of 1RM, three times a week).⁸⁰ Serra-Serra-Rexach et al. found an increment of 17% in the leg press strength of nonagenarians after 8 weeks of progressive strength training with lower loads (two to three sets of eight to 10 repetitions at 30% of 1RM in the initial phase progressing to 70% of 1RM at the end of the intervention).⁹² Lustosa et al. found significant improvements in the knee extensor muscle power of pre-frail elderly women after 12 weeks of strength exercises of the lower extremities at 70% of 1RM, three times a week.⁹³ Although more research is needed, higher-intensity exercises seem to elicit higher gains. Thus, Sullivan et al.⁹⁴ found greater increases in the muscle strength of frail elderly after a 12-week resistance training program of progressive intensity (starting at 20% and ending at 80% of 1RM) compared with a continuous, lower intensity protocol (20% of 1RM during the entire 12-week period).

The American College of Sports Medicine recommends a multi-component (strength, endurance, flexibility, and balance) exercise program to maintain physical fitness in old adults.⁹⁵ Villareal et al. studied the effect of such a multi-component exercise program on physical fitness in frail, obese older adults during 3 months, finding beneficial effects on muscle mass and physical function.⁹⁶ Lord et al. found significant improvements in choice stepping reaction time test, 6-min walking distance, and simple reaction time requiring a hand press, after training (aerobic exercises, specific strengthening exercises, and activities for balance, hand-eye and foot-eye coordination, and flexibility) during 12 months in frail older people living in retirement villages.⁹⁷ Eshani et al.⁹⁸ assessed the effect of a 6-month aerobic training program on frail elderly (consisting of walking at 70%–75% of maximal heart rate during 20 min at the initial phase and progressing to 60 min at the end of the program) and found a 14% of increase in VO₂max.

Evidence on the benefits of exercise interventions in elderly with frailty, co-morbidities, and subsequent physical disability comes from a recent meta-analysis by deVries et al.⁹⁹ This study included data on 18 randomized controlled trials (2,580 participants in total) of physical training interventions in community-dwelling adults aged 60–85 years who were physically frail and/or had physical disability and/or multi-morbidity. Half of the included studies used multi-component training programs, and intervention duration ranged from 5 weeks to 18 months. There were statistically significant benefits with physical exercise therapy compared to no exercise in mobility and physical functioning. There were no differences in effectiveness with regard to the duration of the program (short vs. longer interventions), whereas high-intensity programs elicited greater gains compared to low-intensity ones. The interventions that showed the largest effect sizes were those using resistance training components.

Benefits of exercise (particularly, resistance exercise) in the aging muscle tissue—main signaling pathways involved in exercise adaptations

The molecular mechanisms involved in the activation of signal transduction cascades regulating the adaptations to exercise are dependent on specific signaling pathways activated or repressed by numerous stimuli such as alterations in metabolites concentrations, adenosine triphosphate (ATP)-to-adenosine diphosphate (ADP) ratio, calcium flux, intracellular pH, redox balance, ROS production, and intracellular oxygen pressure.^100–103 Post-exercise changes in gene transcription involve immediate early genes, myogenic regulators, genes that regulate carbohydrate metabolism and lipid mobilization, transport and oxidation, mitochondrial metabolism, and oxidative phosphorylation, as well as transcriptional regulators of gene expression and mitochondrial biogenesis,^104–107 or alterations in the DNA-binding activity of a variety of transcription factors, such as myocyte enhancer factor 2 (MEF2),¹⁰⁷ histone deacetylases (HDACs),¹⁰⁸ and nuclear respiratory factors (NRFs).^109,110

The most relevant signaling pathways modulated by exercise include calcium/calmodulin-dependent protein kinase (CaMKs) signaling, mitogen-activated protein kinases (MAPKs) signaling, ATP turnover and adenosine monophosphate (AMP)-activated protein kinase (AMPK) signaling, oxidized nicotinamide adenine dinucleotide (NAD⁺)/reduced nicotinamide adenine dinucleotide (NADH) ratio and sirtuins (SIRTs), activation of mammalian target of rapamycin (mTOR), oxygen sensing and the hypoxia-inducible factor 1 (HIF-1), mitochondrial biogenesis pathway, and the PPARγ co-activator-1α (PGC-1α) and −1β(PGC-1β).¹¹¹ Accordingly, several important adaptations in skeletal muscle, such as mitochondrial biogenesis, anti-oxidant defense, hypertrophy, cytoprotection, and fiber transformation, are regulated primarily by these pathways. Below we summarize the main biological mechanisms and pathways through which exercise may attenuate sarcopenia (see also Fig. 2).

FIG. 2.

Main signaling pathways involved in the exercise effects in the skeletal muscle tissue. 4E-BP1, eukaryotic translation initiation factor 4E (eIF4E) binding protein; AKT, protein kinase B; AMP, adenosine monophosphate; AMPK, AMP activated protein kinase; ATF2, activating transcription factor 2; ATP, adenosine triphosphate; CaMKII, calmodulin-dependent protein kinase II; CREB, cAMP response-element-binding protein; ERK1/2, extracellular signal-regulated kinase 1 and 2; FAK, focal adhesion kinase; FoxO1, human protein encoded by the FOXO gene; FOXOs, Forkhead box-O transcription factors; HDACs, histone deacetylases; HIF, hypoxia-inducible factor; JNK, c-Jun NH₂-terminal kinase; MAFbx, or Atrogin-1; MuRF-1, muscle RING-finger protein-1; mTOR, mammalian target of rapamycin; mTORC1, mTOR complex 1; NAD, nicotinamide adenine dinucleotide; PA, phosphatidic acid; p70S6K, ribosomal protein S6K; PDH, prolyl hydroxylase; PGC-1α, peroxisome proliferator-activated receptor-γ coactivator-1α; PI13K, phosphatidylinositol 3-kinase; Rheb, Ras homolog enriched in brain gene; ROS, reactive oxygen species; SIRT, sirtuin; Tsc1, tuberous sclerosis complex 1; Tsc2, tuberous sclerosis complex 2. Color images available online at www.liebertpub.com/rej

Calcium is implicated in the regulation of numerous intracellular proteins, such as protein kinase C, calcineurin, and CaMKs that mediate cellular signal transduction.¹¹² Exercise increases CaMKII phosphorylation in an intensity-dependent manner. CaMKs and calcium signaling affect glucose transport,¹¹³ lipid uptake and oxidation,¹¹⁴ and skeletal muscle plasticity.¹¹⁵ In addition, the transcription factors cyclic AMP response element-binding protein (CREB), MEF2, and HDACs are CaMK targets involved in the regulation of skeletal muscle gene expression. Exercise also stimulates MAPK-related pathways, including extracellular signal-regulated kinase 1 and 2 (ERK1/2),¹¹⁶ p38,¹¹⁷ and c-Jun NH₂-terminal kinase (JNK).¹¹⁶ For instance, p38 MAPK can stimulate upstream transcription factors of the PGC-1α gene through skeletal muscle contraction.¹¹⁸ Moreover, MAPKs regulate a wide range of physiological processes, such as differentiation, hypertrophy, inflammation, and gene expression.¹¹⁹

The role of ROS in the exercise-induced adaptations of skeletal muscles has been studied extensively, particularly with regard to aerobic exercise.^120,121 Contracting skeletal muscles produce ROS, activating MAPK signaling and transcription factor nuclear factor-kappa B (NF-κB), thereby linking signal transduction to transcriptional processes.¹²² Acute exercise activates JNK signaling in a ROS-dependent manner, as evidenced by attenuated JNK signaling during exercise with infusion of the anti-oxidant N-acetylcysteine.¹²³ Contraction-induced increases in interleukin-6 (IL-6) secretion, an exercise-associated cytokine with potent multi-organ metabolic effects¹²⁴ (see further below), is JNK dependent,¹²⁵ which attests to the likely importance of JNK signaling in mediating metabolic adaptations to exercise. AMPK is a serine/threonine kinase that modulates cellular metabolism acutely through phosphorylation of metabolic enzymes¹²⁶ and, over time, via transcriptional regulation.^127,128 Given the rate of ATP turnover during muscle contraction, AMPK acts as a signal transducer for metabolic adaptations by responding to an altered cellular energy status. Overall, AMPK activation preserves ATP by inhibiting both biosynthetic and anabolic pathways, while simultaneously stimulating catabolic pathways to re-establish cellular energy stores.¹²⁹ Chronic AMPK activation modifies metabolic gene expression and stimulates mitochondrial biogenesis,¹²⁷ partly via AMPK-induced modulation of the DNA-binding activity of transcription factors, including NRF-1, MEF2, and HDACs.^127,130

Sestrins are a recently discovered hallmark of aging sarcopenia. Mammalian cells express sestrins in response to stress including DNA damage, oxidative stress, and hypoxia. Sestrins can inhibit the activity of the mTOR complex 1 (mTORC1) through activation of AMPK.¹³¹ Sestrins prevent sarcopenia, insulin resistance, diabetes, and obesity. They also extend life span and health span through activation of AMPK, suppression of mTORC1, and stimulation of autophagic signaling.¹³¹ Recently, we proposed a possible role of the AMPK-modulating functions of sestrins in the benefits produced by exercise in older subjects.¹³²

The regulation of the SIRT family of protein deacetylases is NAD⁺ dependent.¹³³ Both deacetylases SIRT1 and SIRT3 respond to elevations in [NAD⁺] and the NAD⁺/NADH ratio. Increased SIRT activity is associated with positive adaptations in skeletal muscle metabolism, including improved mitochondrial function and exercise performance.^134,135 Likewise, the adaptive muscle growth consequent to mechanical loads induced by resistance exercise is largely determined by the enhanced skeletal muscle protein synthesis due to the activation of mTOR, ribosomal protein S6K (p70S6K), and downstream targets.¹³⁶ p70S6K is a major regulator of muscle protein synthesis through pathways of protein translation and ribosome biogenesis involving eukaryotic translation initiation factor 4E (eIF4E), 4E binding protein 1 (4E-BP1), and elongation factor 2 (eEF2). Phosphorylation of 4E-BP1 by mTOR suppresses binding and inhibition of eIF4E by 4E-BP1. Phosphorylation of S6K leads to the phosphorylation of the 40S ribosomal protein S6 (rpS6) and eukaryotic translation initiation factor 4B (eIF4B). Collectively these events lead to the formation of the translation initiation complex and activate protein synthesis inducing cellular hypertrophy.¹³⁷ Mechanosensory regulation of muscle protein synthesis also involves other signaling proteins, such as focal adhesion kinases (FAK), a class of transmembrane receptors that act as protein tyrosine kinases. FAK proteins are pivotal points for the transmission of contractile force throughout the skeletal muscle structure and a central component of integrin signaling. The grade of expression and activity of FAK in skeletal muscle is loading dependent,^138,139 and contraction results in conformational modifications and activation of FAK phosphotransferase,^138,140 which can trigger muscle protein synthesis through mTOR-dependent or -independent mechanisms.¹⁴¹

Oxygen sensing is also involved in the adaptations of skeletal muscle fibers to exercise, particularly aerobic exercise. Hypoxia-inducible factor (HIF), a heterodimeric transcription factor composed of two subunits (HIF-1α and HIF-1β), regulates the major signal transduction pathway sensitive to the intracellular partial pressure of oxygen. Activation of HIF-1 by many stimuli, including aerobic exercise, induces transcription of target genes involved in erythropoiesis, angiogenesis, glycolysis, and energy metabolism.^105,142,143

Exercise benefits in neurodegeneration—main signaling pathways involved

Physical exercise produces important benefits in several neurodegenerative diseases.^144,145 Here we focus on the effects of exercise in Alzheimer's disease, which comprises per se 50%–56% of all causes of aging dementia (an additional 13%–17% is caused by Alzheimer's disease combined with vascular diseases).¹⁴⁶ Exercise, especially aerobic exercise, is beneficial for patients with Alzheimer's disease,¹⁴⁷ not only because it attenuates patients' physical and psychosocial dependence¹⁴⁸ but it also because it improves several features of the pathophysiology of this disorder,^149,150 including oxidative stress regulation, autophagy systems, neurotrophic signaling, mitochondrial biogenesis, angiogenesis, neurogenesis, and the modulation of specific amyloid-β (Aβ)-degrading enzymes (see Table 2 for a summary of intervention studies in humans and Fig. 3 for a summary of the putative molecular pathways involved).^149,150

Table 2.

Summary of Controlled Exercise Interventions in Biomarkers of Neurodegeneration

Reference	Group	n (Age, years)	Health status	Sex	Effects
373	High-intensity ET	Women: 10 (65.3 ± 9.4 years) Men: 9 (70.9 ± 6.7 years)	Amnestic mild cognitive impairment	Both	BDNF increased in men but decreased in women. ET group improved cognitive function compared to controls.
	Control (stretching)	Women: 5 (74.6 ± 11.1 years) Men: 5 (70.6 ± 6.1 years)
374	High-intensity ET Control (stretching)	29 (74.3 ± 2.8 years)	Glucose tolerance criteria for pre-diabetes or newly diagnosed	Both	Although BDNF did not change, ET group improved cognitive functions compared to controls.
375	ET Control	20 (62 years) 19 (63 years)	Coronary artery disease	Both	No changes were observed in VEGF concentrations for both groups.
376	High-intensity ST Moderate-intensity ST Control (stretch)	62 (65–75 years)	Healthy	Men	IGF-1 increased in both ST groups.
377	ST (non-frail) ST (pre-frail)	24 (70.5 ± 4.6 years) 24 (72.5 ± 4.3 years)	Pre-frail and non- frail	Women	BDNF increased in both groups. No changes in GDNF or NGF.
378	ET Control (stretching)	60 (67.6 ± 5.8 years) 60 (65.5 ± 5.4 years)	Healthy	Both	ET increased the size of the hippocampus and BDNF.
379	ST	21 (85.0 ± 4.5y)	Healthy	Women	No changes on VEGF after ST
380	Combined Control	10 (66.1 ± 3.1 years) 10 (67.7 ± 5.2 years)	Obese	Women	VEGF increased after 12 weeks of combined training.
381	ET ST	181 (70.3 ± 4.5 years) 167 (71.0 ± 4.5 years)	Healthy	Women	Only ST increased BDNF
382	Combined (aerobic and resistance)	20 (92.3 ± 2.3 years)	Healthy	Both	No changes in BDNF, ACE, APP, EGF, and TNFα
	Control	20 (92.1 ± 2.3 years)
383	Exercise + medical treatment Medical treatment	20 (69 ± 8 years) 20 (70 ± 11 years)	Peripheral artery disease	Both	Exercise increased circulating EPC counts and decreased ADMA levels. No changes in VEGF and SDF-1.
384	Multi-modal (aerobic, strength and motor fitness)	25 (69.0 ± 3.1 years)	Healthy	Women	Exercise increased BDNF and cognitive performance.
	Control	24 (68.8 ± 3.5 years)
385	ET Control (flexibility, toning and balance)	30 (67.3 ± 5.8 years) 35 (65.4 ± 5.2 years)	Healthy	Both	No changes in BDNF, IGF-1, VEGF
386	Acute exercise ET Control	18 (62 ± 10 yeasrs) 7 (64 ± 6 years) 6 (56 ± 9)	intermittent claudication	Both	No change was observed in VEGF concentrations in response to acute exercise and to the training

ACE, angiotensin-converting enzyme; APP, amyloid precursor protein; DMA, asymmetric dimethylarginine; BDNF, brain-derived neurotrophic factor; EGF, epidermal growth factor; EPC, endothelial progenitor cells; ET, endurance training; GDNF, glial-derived neurotrophic factor; IGF-1, insulin like growth factor 1; NGF, nerve growth factor; SDF-1, stromal cell-derived factor-1; ST, strength training; TNFα, tumor necrosis factor-α; VEGF, vascular endothelial growth factor.

FIG. 3.

Main signaling pathways involved in the exercise effects in neurodegeneration, especially with regard to Alzheimer's disease. 4-HNE, 4-hydroxynonenal; 8-OHdG, 8-hydroxy-2′-deoxyguanosine; Aβ, amyloid-pβ; BACE, β-secretase; BDNF, brain-derived-neurotrophic factor; IGF-1, insulin-like growth factor 1; LTP, long-term potentiation; MDA, malondialdehyde; RCD, reactive carbonyl derivative; ROS, reactive oxygen species; VEGF, vascular endothelial growth factor. Color images available online at www.liebertpub.com/rej

Oxidative stress plays an important role in the etiology of Alzheimer's disease.^151–154 The brain is especially sensitive to oxidative stress compared to other organs owing to its high metabolic rate (i.e., high O₂ consumption) together with its relatively low anti-oxidant defense capacity and its high levels of polyunsaturated fatty acids and metals.^151,155 Elevated levels of brain oxidative stress are found with normal aging,^156,157 and this phenomenon is exacerbated in Alzheimer's disease due to additional sources of ROS such as Aβ accumulation and mitochondrial dysfunction.^{151,153,154,158} Although there is some controversy, aerobic exercise enhances anti-oxidant defense and mitigates oxidative damage in the brain of rodent models.¹⁵⁰ Thus, exercise increases glutathione peroxidase (GPx) activity in whole brain,¹⁵⁹ as well as superoxide dismutase (SOD) and GPx activity in the brainstem and corpus striatum.¹⁶⁰ Using a triple transgenic mouse model of Alzheimer's disease, we recently showed that aerobic exercise training can increase hippocampal catalase mRNA levels.¹⁶¹

Exercise attenuates aging neurodegeneration partly by up-regulating neurotrophic factors, such as the brain-derived neurotrophic factor (BDNF).^162,163 Circulating BDNF levels increase with aerobic exercise, especially when intensity is high.^164–166 Exercise-produced BDNF can help maintain brain function and promote neuroplasticity^167,168 as well as repairing motor neurons.¹⁶⁹ Increased BDNF transcripts in exercised rodents' brains are well documented, which provides a biological explanation for the beneficial effect that exercise has in cognitive function, with tropomyosin receptor kinase (trkB), CREB, or synapsin I signaling been involved.¹⁷⁰ These pathways are involved in synaptogenesis¹⁷¹ and long-term memory formation.¹⁷² Furthermore, the activation of the transcription factor CREB leads to induction of several genes that regulate neurotrophic effects, including those encoding PGC-1α,¹⁷³ dynorphin, and BDNF.¹⁷⁴ In addition, when the exercise induction of the BDNF pathway is blocked, aerobic exercise is unable to activate CREB and stimulate cognition.^170,175 The hippocampal regulation of BDNF induced by exercise is mediated by neurotransmitters,^167,176 neuroendocrine mechanisms,¹⁶⁷ and insulin-like growth factor 1 (IGF-1) modulation.^177,178 Thus, hippocampal IGF-1 levels increase with exercise training¹⁷⁷ and have neurotrophic effects because IGF-1 activates BDNF signaling¹⁷⁷ and increases trkB levels.¹⁷⁹ Aerobic exercise induces other neurotrophic factors, such as vascular endothelial growth factor (VEGF),¹⁸⁰ nerve growth factor (NGF),¹⁸¹ glial-derived neurotrophic factor (GDNF),¹⁸² neurotrophin (NT)-3,¹⁸³ and NT-4/5,¹⁸⁴ all of which act synergistically to induce neurogenesis and neuroplasticity.^167,180

Although neurons are post-mitotic cells, neurogenesis can still occur in specific areas of the adult hippocampus¹⁸⁵ with some stimuli such as ischemia/reperfusion, aging, metabolic pathology, or physical exercise being able to change the rate of neurogenesis.¹⁵⁰ Van Praag and collaborators have extensively studied the effects of exercise, particularly of the aerobic type, on adult neurogenesis¹⁸⁶ and have demonstrated that the newly formed neurons are associated with the cognitive and synaptic effects induced by exercise.¹⁸⁷ This phenomenon is especially important in age-related neurodegeneration because attenuation of accelerated neuronal loss can prevent several age-related disorders, including Alzheimer's disease. Several signaling pathways induced by aerobic exercise have been suggested to mediate this process such as BDNF,^188,189 VEGF,¹⁸⁰ and IGF-1.¹⁹⁰

Both age and disease-induced neurodegeneration are partially produced by a dysregulation of protein homeostasis (see further below on the exercise effects in aging “proteostasis”). Aerobic exercise increases proteolytic degradation by proteasomes and neprilysin, a specific Aβ-degrading enzyme.^167,191 Activation of the brain proteasome is important in preventing Alzheimer's disease because proteasome inhibition produces Aβ accumulation, a hallmark of this disorder.¹⁹² The neurofibrillary tangles produced by an accumulation of hyperphosphorylated tau proteins is also an important hallmark of Alzheimer's disease. Lysine residues of tau are susceptible to ubiquitination, indicating interaction of tau aggregation by oligomerization and ubiquitination-mediated degradation through the proteasome system.¹⁹³ The proteasome might also be involved in the learning process, because its inhibition in the hippocampus blocks long-term memory.^194–195

Exercise and the Cellular Hallmarks of Aging

In a recent state-of-the-art review, López-Otín et al.¹⁹⁶ nicely postulated nine hallmarks of aging that might be targeted in future pharmacological interventions—genomic instability, telomere attrition, epigenetic alterations, loss of protein homeostasis (proteostasis), deregulated nutrient sensing, mitochondrial dysfunction, cellular senescence, stem cell exhaustion, and altered intercellular communication. Although more research is needed, exercise, which is available at low cost and largely free of adverse effects,¹¹¹ can influence, at least partly, most of these hallmarks (see Fig. 1, right column, for a summary).

Genomic instability

A 5-month aerobic exercise program prevented mitochondrial DNA (mtDNA) instability in multiple tissues of mtDNA mutator (progeroid) mice, thereby reducing multisystem pathology and preventing premature mortality.¹⁹⁷ Oxidative damage to DNA occurs during the aging process.¹⁹⁸ Resistance exercise decreases such damage in old people, as indicated by 8-hydroxy-2′-deoxyguanosine (8-OHdG) determination, through stimulation of endogenous anti-oxidant defense,¹⁹⁹ whereas in rodent models aerobic exercise improves DNA repair mechanisms (e.g., proteasome complex),²⁰⁰ as well as NF-κB and PGC-1α signaling.^121,201

Telomere attrition and telomerase activity

Accelerated telomere shortening is linked with numerous age-related chronic diseases and risk factors.^202–210 On the other hand, there is increasing evidence supporting an association between habitual physical exercise, particularly aerobic exercise, and longer leukocyte telomere length.^211–217 Leukocyte telomere length is also positively associated with cardiorespiratory fitness (expressed as VO₂max),^213,218,219 which, in turn, is associated with lower CVD and all-cause mortality.²²⁰ Long-term aerobic exercise can modulate leukocyte telomere length as well as the network of proteins that interact with telomeres, through activation and induction of telomerase enzyme activity (mediated by human telomerase reverse transcriptase [TERT]) and the shelterin complex (or telosome).²¹⁷ In effect, TERT mRNA expression is up-regulated in leukocytes after exercise.²²¹ Exercise also regulates the microRNAs (miRNAs) that control the downstream expression of genes involved in telomere homeostasis.²²¹ The association between physical exercise and telomere length could also be due to lower oxidative stress and inflammation, exercise-induced regulation of telomeric genes, or a complex interplay between these processes.^221,222 The effects of exercise on skeletal muscle tissue telomeres have been less studied compared with leukocytes, and the results are less conclusive.^223,224

Epigenetic adaptations

The relationship between epigenetics regulation (e.g., DNA methylation) and aging is complex and controversial, i.e., hypomethylation or hypermethylation might be either beneficial or detrimental depending on the different cell types; and, whether manipulations of histone-modifying enzymes can influence aging through purely epigenetic mechanisms remains to be clarified.¹⁹⁶ While keeping in mind the above-mentioned controversy, exercise seems to induce epigenetic modifications that can help attenuate age-deregulations,²²⁵ and several mechanisms, such as metabolic adaptations and transient hypoxia conditions, have been proposed recently.²²⁶ Regular aerobic exercise can modify genome-wide DNA methylation in humans.²²⁷ Animal^216,228,229 and human research²¹⁶ suggests that aerobic exercise induces, through epigenetic mechanisms, telomerase activity and the transcription of genes encoding telomere-stabilizing proteins. Both resistance and aerobic exercise can increase DNA methylation, cause histone modifications, and induce miRNAs in a wide range of tissues, including among others, muscle, brain, and cardiovascular system.^225,226 Transient DNA hypomethylation of gene-specific promoter regions precedes increases in mRNA expression in response to acute exercise.²³⁰ In turn, these pulses of elevated mRNA during recovery from acute exercise facilitate protein synthesis and induce gradual structural remodeling and long-term functional adjustments.²³¹ In general, these adaptations are intrinsic to the working skeletal muscle and collectively contribute to maximize substrate delivery, mitochondrial respiratory capacity, and contractile function during exercise. The net effect is promotion of optimal performance during a future exercise challenge, resulting in a robust defense of homeostasis in the face of metabolic perturbation and, consequently, enhanced resistance to fatigue.^232,233 Several epigenetic mechanisms, including histone H4 deacetylation and loss of promoter methylation, have been implicated in the modified gene expression profile that occurs as an adaptation to aerobic exercise.²³⁴

Epigenetic mechanisms are not restricted to early stages of human development but are broad dynamic controllers of genomic plasticity in response to environmental factors such as exercise.²³⁵ For instance, in young adults, the class II HDACs 4 and 5 (transcriptional repressors) can translocate from the nucleus to the sarcoplasm of muscle fibers in response to aerobic exercise.¹⁰⁸ Over-expression of HDAC5 in transgenic mice blocks the effects of exercise training, further suggesting a contribution of histone modifications in the transcriptomic response to muscle contraction.²³⁶ In human and mouse muscles, methylation of PGC-1α, mitochondrial transcription factor (TFAM), MEF2A, citrate synthase (CS), and pyruvate dehydrogenase kinase isozyme 4 (PDK4) gene promoters decreases after an acute bout of aerobic exercise.²³⁰ The degree of DNA methylation of a large number of genes changes in response to exercise in both skeletal muscles and adipose tissue.^237,238 In addition, aerobic exercise-induced SIRT-1 regulates the tumor suppressor p53, PGC-1α, NF-κB as well as other transcription factors via its deacetylase activity.^225,239

Chronic moderate aerobic exercise increases the methylation levels of the pro-inflammatory apoptosis-associated speck-like protein caspase (ASC) gene, which modulates IL-1β and IL-18 in the leukocytes of old people, thereby contributing to attenuation of age-related increases in pro-inflammatory cytokines.²⁴⁰ Aerobic exercise training also alters DNA methylation in a chronic manner.²⁴¹ Thus, 48 hr after a bout of aerobic exercise, the DNA methylation profile of genes involved in diverse metabolic pathways, as well as in calcium and insulin signaling, was recently found to be differentially methylated in skeletal muscle.²⁴¹ A majority of the detected genes in this study were chronically hypomethylated after exercise training. Both aerobic and resistance exercise also can help combating aging sarcopenia and frailty by modulating, through epigenetic mechanisms, several myogenic regulatory factors, e.g., myogenin, myoblast determination protein 1 (MyoD), or myogenic factors 5 (Myf5) and 6 (Myf6, also known as myogenic regulatory factor 4 [Mrf4] or herculin).^242–244 Exercise also helps to attenuate the aging-related epigenetic deregulation of growth factors in neurodegenerative diseases, not only by up-regulating BDNF induction as mentioned above, but also by promoting remodeling of the chromatin containing the BDNF gene.²⁴⁵

Overall, the study of exercise-induced epigenetic modifications is just in its infancy. Yet the studies available have already provided new insights into the potential tissue-specific alterations in DNA methylation induced by exercise and into some of the mechanisms explaining the beneficial effects of regular exercise.

Loss of proteostasis

Proteostasis is defined as the protein homeostasis that is responsible for refolding or degrading altered proteins by several mechanisms, such as autophagy, proteasomal degradation, or chaperone-mediated folding. A loss of function in these processes leads to an aggregation of damaged proteins and thereby proteotoxic effects that have been associated with aging^196,246 and age-related conditions such as Alzheimer's or Parkinson's disease.²⁴⁷

The autophagy–lysosomal and the ubiquitin–proteasome systems, two important proteostatic mechanisms, are impaired by aging.^248,249 Exercise has a beneficial effect in autophagy.²⁵⁰ In aging mouse models, aerobic exercise induces autophagy in: (1) The brain, supporting its potential to promote elimination of damaging proteins causing neurodegeneration²⁵¹; (2) heart²⁵²; or (3) muscle (besides preventing apoptosis), by modulating IGF-1, Akt/mTOR, and Akt/Forkhead box O3A (FoxO3a) signaling, thereby preventing loss of muscle mass/strength.^253,254 Although data are still scarce in aging humans, autophagy muscle markers are up-regulated after combined exercise training (walking and moderate-intensity leg resistance exercises) in old women.²⁵⁵

Aerobic exercise induces autophagy in mice through activation of the BCL-2–beclin-1 complex,²⁵⁶ whereas beclin-1 disruption in transgenic mice reduces autophagy leading to neurodegeneration.²⁵⁷ Moreover, the aging human brain shows a down-regulation of beclin-1,²⁵⁸ whereas healthy centenarians have higher serum levels of beclin-1 compared with young controls, suggesting that elevated basal levels of autophagy may be related to healthy human exceptional longevity.²⁵⁹ MacKenzie et al. showed that acute high-resistance exercise evoked increased muscle protein synthesis and decreased protein degradation in rats, through activation of the class 3 phosphatidylinositol 3OH-kinase (PI3K) Vps34 mVps34,²⁶⁰ which is known to regulate autophagy by forming a complex with beclin-1.²⁶¹ Using an atrogin-1 (also known as MAFbx) knockout mouse model, Zaglia et al. demonstrated that autophagy dysfunction promotes cardiomyopathy and premature death.²⁶² Atrogin-1 is a muscle-specific ubiquitin ligase involved in muscle atrophy through FoxO signaling.²⁶³ Similar to skeletal muscle, atrogin-1 up-regulation in the heart leads to atrophy.²⁶⁴ Interestingly, aged atrogin-1 knockout mice have reduced tolerance to treadmill exercise and shortened life span compared with age-matched controls.²⁶² In this animal model, muscle age-related increases in oxidative damage and apoptosis are attenuated by regular aerobic exercise, whereas both mechanisms are negatively correlated with autophagy.²⁶¹

Deregulated nutrient sensing

Exercise exerts protective effects against age declines in the glucose-sensing somatotrophic axis,²⁶⁵ and also activates at the muscle level the three main interconnected nutrient-sensing systems, i.e., the amino acid–sensing mTOR pathway^266,267 and the low energy–sensing AMPK and SIRT pathways,^132,268 thereby promoting a beneficial muscle anabolic state. On the other hand, exercise improves insulin sensitivity through increased production of the glucose transporter type 4 (Glut 4).²⁶⁹

In addition, resistance exercise acutely increases the circulating levels of testosterone, growth hormone (GH), and IGF-1, with the magnitude of the effect usually increasing with higher exercise intensity²⁷⁰ or duration,²⁷¹ shorter rest intervals,²⁷² and higher exercising muscle mass.^273,274 Thus, resistance exercise is a useful approach to prevent sarcopenia by virtue of its ability to increase protein synthesis^275–277 and skeletal muscle mass.^278,279 However, the rate of muscle protein synthesis in response to exercise training is lower in the elderly than in younger people,^280,281 leading to a lower capacity to improve skeletal muscle strength and fiber size.²⁸² On the other hand, supplementation with essential amino acids together with resistance training increases muscle protein synthesis both in young and old individuals, although such an effect is also attenuated in the latter, owing, at least partly, to lower ERK1/2 and mTOR signaling.²⁸⁰ (For a more extensive description about nutrient-sensing modulation by exercise, see above.)

Mitochondrial dysfunction

mtDNA mutations (typically deletions) accumulate with age in different tissues,^283,284 including mainly the nervous and skeletal muscle tissue.^37,285 Mutations in muscle mtDNA play a causal role in the physiological mechanisms implicated in sarcopenia,^37–40 particularly in abnormalities of the electron transport system, muscle fiber atrophy, and breakage.^37,39,40 Despite classic studies demonstrating exercise-induced mitochondrial biogenesis in young but not in aged mice,²⁸⁶ recent findings have shown that aerobic exercise training attenuates mitochondrial dysfunction and loss of mitochondrial content in the aging human skeletal muscle while increasing oxidative capacity and the activity of different electron transport chain protein complexes.²⁸⁷

The accumulated damage to the mitochondria due to the ROS generated from the electron transport chain is the base of the mitochondrial theory of aging first proposed by Harman.²⁸⁸ Oxidative damage to mtDNA increases with aging, affecting mtDNA replication and transcription, which, in turn, alters the functionality of mitochondrial proteins.^289,290 Lanza et al. demonstrated that age-related decline in mitochondrial oxidative capacity was absent in endurance-trained humans, who showed elevated expression of mitochondrial proteins, mtDNA, and mitochondrial transcription factors.²⁹¹ In mtDNA mutator mice, aerobic exercise promoted systemic mitochondrial biogenesis, prevented mtDNA depletion and mutations, increased mitochondrial oxidative capacity and respiratory chain assembly, restored mitochondrial morphology, and blunted pathological levels of apoptosis in multiple tissues. Thus, the authors concluded that a systemic “mitochondrial rejuvenation” occurred as a result of the training program.¹⁹⁷

The lower mitochondrial enzyme activity commonly shown in older compared with younger adults^17,292 is associated with a down-regulation of the mRNAs encoding mitochondrial proteins in skeletal muscle.^293–295 Aged subjects have cytochrome c oxidase (COX)-deficient muscle fibers,²⁹⁶ especially in sarcopenic muscles or in those focal regions with higher content of mtDNA mutations.^39,296–299 Yet resistance exercise training can reverse the muscle transcriptional signature of aging back to that of younger levels for most genes involved in mitochondrial function.²⁹⁹ Gomes et al.³⁰⁰ recently provided novel insights into the mechanisms responsible for the age decline of mitochondrial homeostasis by elegantly showing a regulatory pathway that is SIRT1-mediated and independent of PGC-1α and -β, with aging declining NAD⁺ levels and thereby reducing SIRT1 activity and leading to impaired oxidative phosphorylation (OXPHOS). Short treatment (1 week, or ∼8 months, when translated to the human life span) of 22-month-old mice with nicotinamide mononucleotide (NMN) (a precursor to NAD⁺ increasing NAD⁺ levels in vivo) reversed several biochemical indicators of muscle mitochondrial senescence, with increased OXPHOS transcripts in the gastrocnemius muscle. It was proposed that NMN or other compounds able to increase NAD⁺ are candidates to be included in the human anti-aging armamentarium. And yet NMN, as opposed to exercise, was unable to reverse other age-dependent whole-organism effects, such as loss of muscle strength.

Besides the above-mentioned benefits of resistance exercise on muscle strength until end of life (which were summarized in Table 1), regular exercise has a profound beneficial effect on human mitochondrial function/biogenesis,³⁰¹ with this effect being both PGC-1 and SIRT mediated.¹¹¹ An active lifestyle attenuates aging mitochondrial dysfunction, promoting longevity through pathways common to the effects of caloric restriction.²⁹¹ In addition, some “myokines” (see further below for the definition of myokine) have a mitochondrial rejuvenating effect, e.g., visfatin, a NAD⁺ biosynthetic enzyme that stimulates the SIRT-1 pathway.²⁵²

Cellular senescence

Cellular senescence is defined as a stable arrest of the cell cycle coupled with stereotyped phenotypic changes, and its regulation during aging is a complex process. Indeed, the same phenomenon, i.e., elimination of senescent cells, that is beneficial to delay age-related pathologies and thus to promote longevity, could also stimulate cancer development.¹⁹⁶ Besides inducing secretion of anti-tumorigenic myokines such as secreted protein acidic and rich in cysteine (SPARC, also known as basement membrane protein [BM]-40), calprotectin, or leukemia inhibitory factor,³⁰² exercise, mainly aerobic exercise, may decrease cancer incidence and help improve cancer prognosis through several mechanisms, including greater natural killer (NK) cell activity, enhanced antigen presentation, reduced inflammation, and prevention of functional senescent cells' accumulation.³⁰³ Telomere-associated proteins regulate cellular senescence and, as described above, are up-regulated by exercise. Moreover, aerobic exercise increases the aortic expression of telomere repeat-binding factor 2 and Ku70 and reduces the expression of apoptosis regulators, such as cell-cycle-checkpoint kinase 2, p16^INK4a, and p53 or survival regulators.²¹⁶

Telomere-associated proteins, as well as p16^INK4a/Rb and p19^ARF/p53 signaling, are considered main pathways in the control of human aging and age-associated pathologies.¹⁹⁶ p16^INK4a and p21 are cell cycle inhibitors that are up-regulated in senescent cells.^304,305 p21 is a downstream target of p53 and telomere dysfunction, whereas p16^INK4a appears to be up-regulated in a p53- and telomere-independent manner.³⁰⁶ Sousa-Victor et al. recently highlighted the importance of p16^INK4a in the modulation of cellular senescence. In a geriatric mouse model, muscle satellite cells lose their quiescent state owing to deregulation of p16^INK4a, whereas repressing p16^INK4a restores muscle regenerative capacity.³⁰⁷ Thus, we suggested the importance of p16^INK4a modulation as a new target for combating aging-related chronic diseases.³⁰⁸ Lifestyle factors, including smoking and physical aerobic exercise practice, have been associated with p16^INK4a mRNA levels in peripheral blood T ymphocytes,³⁰⁹ a biomarker of human aging.³¹⁰ Thus, physical exercise is inversely correlated with p16^INK4a mRNA levels, i.e., higher amounts of physical exercise leads to down-regulation of p16^INK4a in blood cells, which might promote protective effects against age-dependent alterations.³⁰⁹

As exposed above, cellular senescence plays a key role not only in cancer development^311,312 but also in aging.³¹³ In fact, cell senescence is one of the major paradigms of aging research through the acquisition of the senescence-associated secretory phenotype (SASP) or senescence-messaging secretome.³¹⁴ SASP is a DNA damage response, which, through production of inflammatory, growth-promoting, and remodeling factors can potentially explain how senescent cells alter tissue microenvironments.³¹⁵ Different animal model investigations have shown that exercise modulates senescence associated to aging. Thus, 12 weeks of aerobic (swimming) exercise training suppressed liver senescence markers and down-regulated inflammatory mediators by reducing gamma glutamyltranspeptidase activity and levels of p53, p21, and IL-6 in a d-galactose–induced senescence rat model.³¹⁶ Werner et al.²²⁸ studied the effect of aerobic exercise on telomere-regulating and cellular senescence mechanisms at the cardiac level in wild-type, endothelial NO^• synthase (eNOS)-deficient and TERT-deficient mice models. Their results showed that exercise up-regulated cardiac telomere-stabilizing proteins, promoted anti-senescent effects, and provided protection against doxorubicin-induced cardiomyopathy. Werner et al.²¹⁶ also studied the effects of aerobic exercise on vascular telomere biology and endothelial apoptosis in mice and the effects of long-term aerobic training on telomere biology in circulating leukocytes in humans. Besides improving telomere biology in the thoracic aorta and in mononuclear cells, exercise reduced the vascular expression of apoptosis regulators. Moreover, endurance athletes had increased telomerase activity and down-regulated cell cycle inhibitors compared with sedentary subjects. These findings are supported by Song et al.,³⁰⁹ who found that in humans aerobic exercise reduced the expression of DNA damage biomarkers and correlated positively with p16^INK4a expression and negatively with telomere length in peripheral blood T lymphocytes.

Stem cell exhaustion

The decline in the regenerative potential of tissues is a main characteristic of aging, whereas exercise is one of the most potent stimuli for the proliferation/migration of the different adult stem cell subsets from their home tissue (e.g., bone marrow) to target damaged tissues for subsequent engraftment and regeneration.²⁵² Thus, regular exercise attenuates age-associated reduction in the endothelium-reparative capacity of endothelial progenitor cells.³¹⁷ Exercise activates mesenchymal stem cells, which are pluripotent progenitors with a wide variety of therapeutic potential (e.g., as vehicles of anti-cancer genes) and promotes proliferation of neural stem cells, thereby contributing to improve brain regenerative capacity and cognitive ability.²⁵²

Arguably the most affected stem cell type during aging is the myogenic one, known as satellite cells.³¹⁸ Although the human skeletal muscle tissue maintains myofiber replacement and repair potential throughout most of life, the efficiency of this process declines with aging, owing to satellite cell alterations. Age-reduced number or functionality of these myogenic cells prevents proper maintenance of muscle mass.^318–321 Specifically, aging atrophy of type II muscle fibers is accompanied by a specific decline in the content of type II muscle fiber satellite cells.³¹⁸ Thus, since sarcopenia is associated with atrophy of type II muscle fibers, its pathophysiological mechanisms are closely related with the decline in satellite cell content with aging.³¹ Both aging reductions in muscle mass and strength are positively correlated with muscle fiber type specific cross-sectional area, myonuclear content, and satellite cell content.³²²

Animal studies have demonstrated that aerobic exercise increases myofibers that contain higher numbers of satellite cells in both young and old rats,³²³ and also promotes expansion of the satellite cell pool in young and old mice.³²⁴ The contribution of these stem cells to skeletal muscle regeneration has been well documented.^325,326 As stated by Hawke and Garry,³²⁶ because adult myofibers are post-mitotic cells, the regulation of skeletal muscle is dependent on a small population of resident cells that are the satellite cells. The regulation of satellite cells involves several mechanisms, including immune response, neurotransmitters, neurotrophic and vascular factors (among other growth factors such as IGF-1³²⁷), cytokines such as IL-6,³²⁸ testosterone, or NO^•, most of which are modulated by exercise.¹¹¹

Not only in young adults³²⁹ but also during aging, resistance training is able to induce skeletal muscle satellite cell proliferation and differentiation, thereby resulting in hypertrophy of type II fibers.³³⁰ The latter phenomenon, in turn, attenuates the pro-sarcopenic physiological events related to type II fiber atrophy associated with aging,^{31,318, 322} On the other hand, although resistance training in the elderly of both sexes can counteract the loss of muscle mass and strength,³³¹ a recent study reported that satellite cell induction in response to a single bout of resistance exercise is delayed in old men.³³² McKay et al.³³³ also showed that, compared to young adults, muscle levels of myostatin, a protein that inhibits muscle differentiation and growth in the myogenesis process, were two-fold higher in older individuals, who also had 35% fewer basal stem cells and a type II fiber-specific impairment in stem cell content. The authors concluded that the co-localization of myostatin with satellite cells explains the worsened myogenic capacity of the aged skeletal muscle.³³³ In fact, an aging-blunted activation of type II muscle fiber satellite cells in response to an acute bout of resistance exercise was recently shown by Snijders et al.³³² In addition, the satellite cell response to resistance exercise is related to the extent of muscular hypertrophy induced by training.³³⁴

Altered intercellular communication

Aging is associated with altered intercellular communication leading to inflammation or “inflammaging.”¹⁹⁶ Several mechanisms are responsible for this process, including accumulation of pro-inflammatory tissue damage, immune dysfunction, release of pro-inflammatory cytokines by senescent cells, higher activation of NF-κB, or impaired autophagy regulation.^196,335 These events activate the NOD-like receptor protein 3 (NLRP3) “inflammasome,” characterized by elevations in IL-1β, tumor necrosis factor-α (TNF-α), and interferons.^335,336 Interestingly, calorie restriction and exercise-mediated weight loss in obese individuals with type 2 diabetes lead to a reduction in adipose tissue expression of the NLRP3 inflammasome and IL-1β, and thus to reduced inflammation.³³⁷

The decay factor AUF1 (AU-binding factor 1, also known as heterogeneous nuclear ribonucleoprotein D or hnRNP D) is implicated in the cessation of the inflammatory response (by mediating cytokine mRNA degradation) and also in the maintenance of telomere length by modulating TERT.³³⁸ Down-regulation of AUF1 leads to accelerated cellular senescence and premature aging in mice, which is rescued by normalizing the expression of this factor.¹⁹⁶ Lai et al. found that chronic muscle contractile activity increased different AUF1 isoforms (p37, p40, and p45) in the muscle of healthy rats, resulting in improved muscle plasticity in response to subsequent contractile activity.³³⁹ Senescent cells transmit their condition to other cells through multiple mechanisms, including ROS, growth factors, and interleukins.³⁴⁰ As mentioned above, chronic physical exercise (mostly of the aerobic type) decreases ROS damage, and it does so by decreasing ROS production at the mitochondrial level while up-regulating endogenous anti-oxidant defense.¹²¹

Importantly, skeletal muscle fibers produce hundreds of secreted factors or “myokines” (including the above-mentioned neurotrophins) with a potential drug-like effect at the local and systemic levels, i.e., proteins, growth factors, cytokines, or metallopeptidases, and this secretory capacity increases during and after exercise training (see Fiuza-Luces²⁵² for an in-depth review and Table 3 for some illustrative examples). Systemic low-level inflammation and related conditions such as CVD or cancer can be attenuated by the cumulative effect of regular exercise bouts, during which the muscle can release IL-6, arguably the myokine prototype.³⁴¹ This, in turn, creates a healthy milieu by inducing the production of the anti-inflammatory cytokines IL-1 receptor antagonist (IL-1Ra), IL-10, or TNF soluble receptors (sTNF-R) while inhibiting the pro-inflammatory cytokine TNF-α.³⁴¹ The release of IL-6 from working muscles increases with exercise intensity³⁴² and duration,³⁴³ and endogenous NO^• and the interaction between Ca²⁺/nuclear factor of activated T cell (NFAT) and glycogen/p38 MAPK are the proposed upstream signals leading to its secretion.³⁴⁴ Other anti-inflammatory myokines include IL-4, IL-10, or IL-13.^345,346 Thus, life-long aerobic exercise training is associated with lower inflammation levels.^347–349 Higher levels of aerobic exercise have also been associated with lower levels of C-reactive protein (CRP), IL-6, and TNF-α in people aged 70–79 years.³⁵⁰ However, although exercise training is known to have beneficial anti-inflammatory effects across a broad spectrum of organs and systems, more research is needed in the elderly, particularly to determine if the molecular mechanisms and pathways involved are similar in old people compared with younger population segments.³⁵¹

Table 3.

Examples of Myokines Released During Exercise with a Potential Anti-Aging Effect

Name of molecule	Main tissue(s) of origin	Main type of exercise inducing its release/secretion	Main target tissue(s)	Main biological effect(s) associated with exercise-induced release/secretion	Main aging hallmark targeted	Potential future anti-aging application	Illustrative references
Brain-derived neurotrophic factor (BDNF)	Central nervous system Vascular endothelial cells, platelets, lymphocytes, eosinophils, monocytes, pituitary gland, working skeletal muscle	Moderate-intense “aerobic” exercise (e.g., brisk walking)	Brain Motor neurons	↑ Neuroplasticity ↑ Motor unit regeneration	Cellular senescence in the brain	Protection against neurodegeneration (including possibly dementia)	167–169
Interleukin-4 (IL-4) and IL-13	Lymphocytes, mast cells and neutrophils Various origins (brain, cancer cells, liver, fibroblasts, and muscle cells) Working muscles	Resistance exercise (e.g., weightlifting)	Skeletal muscle	↑ Muscle growth	Loss of muscle proteostasis	Aging muscle atrophy/sarcopenia	387–389
IL-6 (also termed interferon, beta 2)	Working muscles Immune cells Adipocytes	Intense ‘aerobic’ exercise (e.g., brisk/very brisk walking)	Skeletal muscle Adipose tissue Pituitary gland-liver Immune cells	↑ Muscle lipolysis ↑ Muscle growth ↑ Adipocyte lipolysis ↑ Liver-glucose release to blood ↓Inflammation ↑Immunomodulation	Altered inter-cellular communication (‘inflammaging’)	Age-related cardio-metabolic diseases	129, 390–393
IL-15	Working muscles Various origins (lymphoid tissues, kidney, brain, cardiac muscle, lung, pancreas, testis, liver, placenta, epithelial cells, and activated macrophages, and maybe adipocytes)	Mainly resistance exercise	Skeletal muscle Adipose tissue	Promotes muscle anabolism/inhibits catabolism Anti-obesogenic (↓mainly visceral fat) effect Insulin-sensitizing effect	Loss of muscle proteostasis	Aging muscle wasting/sarcopenia	394–397
Leukemia inhibitory factor (LIF)	Working muscles Central nervous system	Mainly resistance exercise	Skeletal muscle	Mainly local (autocrine/paracrine effect): ↑ Muscle growth (satellite cell proliferation) ↑ Muscle regeneration	Loss of muscle proteostasis	Aging muscle wasting/sarcopenia	398–401
Myostatin (also termed, growth differentiation factor 8 [GDF8])	Skeletal muscle	Both “aerobic” and resistance exercise	Skeletal muscle Adipose tissue	Main effects associated to myostatin inhibition which can be partly achieved by exercise are: ↑Muscle growth ↓Adiposity ↑Insulin sensitivity	Attenuation of disease/age muscle wasting Obesity/diabetes prevention	Use of exercise as a coadjuvant of myostatin-inhibition therapies for muscle wasting	402–406
Visfatin (also known as (nicotinamide phosphoribosyltransferase [NAMPT] or pre-B cell enhancing factor [PBEF])	Ubiquitous expression in human tissues, including adipose and skeletal muscle tissue (i.e., it is both an adypokine and a myokine), liver, bone marrow, lymphocytes, β-cells and human islets, heart	“Aerobic” exercise	Skeletal muscle and adipose tissue	AMPK activation→↑ sirtuin1 (SIRT1)→peroxisome proliferator-activated receptor γ co-activator α (PGC-1α) It provides NAD+	Mitochondrial dysfunction	Exercise as a major component of anti-aging medicine	407–410

The information provided in the table is based (and adapted from) a previous review paper by the authors.¹¹¹

AMPK AMP-activated protein kinase; NAD⁺, oxidized nicotinamide adenine dinucleotide.

Perspective

The benefits of regular exercise are such that a dose–response is usually observed in humans. Higher levels of moderate-to-vigorous exercise (≥450 min/week, clearly above the minimum international recommendations of 150 min/week) are associated with longer life expectancy.³⁵² Furthermore, elite athletes—those humans sustaining the highest possible exercise levels (e.g., former participants in the Tour de France cycling race or former Olympic marathoners)—usually live considerably longer than the general population.³⁵³

Physical exercise has a profound effect on the expression of a substantial proportion of our genome, which has evolved to optimize aerobic metabolism in an environment of food scarcity. Thus, physical inactivity is becoming a major public health problem worldwide.³⁵⁴ Exercise certainly cannot reverse the aging process, but it does attenuate many of its deleterious systemic and cellular effects. Most common age-associated chronic conditions are diseases of physiology and thus physiological interventions, of which physical exercise is arguably the best example, are largely the answer.³⁵⁵ We propose that more research efforts should be devoted to gain insights into the molecular mediators of the exercise benefits. Besides putting more anti-aging drugs on the market, it would be wise to determine which are the most effective combinations, types, and dosages (frequency, duration, intensity) of exercise for older people and to implement efficient exercise interventions for this and younger population segments. Note: For the purposes of simplicity in this review, we have mostly used the common term “exercise” instead of “physical activity.”

Acknowledgments

Research in the field of exercise and health by A. Lucia is funded by Fondo de Investigaciones Sanitarias (FIS, grant no. PI12/00914).

Author Disclosure Statement

No competing financial interests exist.

All co-authors fully reviewed the entire manuscript and participated in at least some of its parts, including literature search (all authors), figure design (F.S.-G., N.G., H.P.-G.), tables edition (A.S.-L., C.F.-L., N.G.), or writing and editing of a significant part of the manuscript (all authors).