The Effect of Exercise on Glucoregulatory Hormones: A Countermeasure to Human Aging: Insights from a Comprehensive Review of the Literature

Hormones are secreted in a circadian rhythm, but also follow larger-scale timetables, such as monthly (hormones of the menstrual cycle), seasonal (i.e., winter, summer), and, ultimately, lifespan-related patterns. Several contexts modulate their secretion, such as genetics, lifestyle, environment, diet, and exercise. They play significant roles in human physiology, influencing growth of muscle, bone, and regulating metabolism. Exercise training alters hormone secretion, depending on the frequency, duration, intensity, and mode of training which has an impact on the magnitude of the secretion. However, there remains ambiguity over the effects of exercise training on certain hormones such as glucoregulatory hormones in aging adults. With advancing age, there are many alterations with the endocrine system, which may ultimately alter human physiology. Some recent studies have reported an anti-aging effect of exercise training on the endocrine system and especially cortisol, growth hormone and insulin. As such, this review examines the effects of endurance, interval, resistance and combined training on hormones (i.e., at rest and after) exercise in older individuals. We summarize the influence of age on glucoregulatory hormones, the influence of exercise training, and where possible, examine masters’ athletes’ endocrinological profile.


Introduction
The aging process is accompanied by one or more changes in biological functions (affecting nervous system, cardiovascular and respiratory systems, or renal function, amongst others), often associated with an increasing susceptibility to co-morbidities and mortality [1,2].
The anatomical and physiological changes associated with aging start several years before the appearance of external signs. Many of these alterations gradually manifest in the third decade and continue until death. These changes are also accompanied by a gradual decline in physical fitness and physical activity. This alteration of the cardiovascular and respiratory systems during the aging process can be mainly explained by a decline in maximum oxygen uptake (~10% per decade) starting from the age of 20 [4][5][6][7].
Advancing age is also associated with a decline in anaerobic performance, which can be mainly explained by changes involving the neuromuscular system and a major loss in type II fibers. Indeed, advanced age is accompanied by muscular wasting, a decrease in the rate of contraction, and maximum force.
According to Korhonen et al. [8], the first decline in muscle strength and mass occurs around the age of 30 and the loss is around 15% per decade from the age of 50 to 30% at the age of 70. Moreover, available scholarly literature suggests that starting from the 4th decade of life, both skeletal muscle mass and strength decline in a linear fashion and within the 8th decade of life, 50% of mass will be lost [9]. Since muscle mass amounts to 60% of body mass, its pathological changes can have deep consequences in the elderly.
One hypothesis for the reduction in physical performance and muscle weakness associated with age is an alteration of the endocrine system [10][11][12][13][14]. In particular, the glucoregulatory system that is characterized by important bio-molecules such as glucagon and insulin is critical to maintain the constancy of glucose in the internal milieu. While it is clear that exercise training improves fitness and physical capacity in older adults [15][16][17][18][19], whether exercise can improve the hormonal profiles of older adults remains contentious [20][21][22][23][24][25][26][27].
Therefore, this review will summarize the existing literature concerning the influence of age, and the effects of each mode of exercise (endurance, sprint, and resistance training) on relevant (basal) hormones belonging to the glucoregulatory system.
Where possible, we will provide evidence from masters athletes involving the influence of lifelong exercise on these hormones, but also report findings from interventional studies providing information on the training effect on these hormones.

Materials and Methods
The present review was designed as a comprehensive review of the literature. Search strategy adopted in the present review is summarized in Table 1. Table 1. Search strategy adopted in the present comprehensive review of the literature for retrieving studies investigating the effects of physical activity and exercise on glucoregulatory hormones in elderly subjects.

Insulin, Aging and Physical Activity
Insulin plays a key role in glucose uptake by muscle, fat, and liver cells. Moreover, insulin inhibits both the liver glucose production and its secretion in blood.
Recent reports suggest that the insulin/insulin-like growth factor-1 (IGF-1) signaling pathways and molecular cascades have an important, evolutionarily conserved influence over rate of aging and, thus, longevity [28]. The most important effects of advancing age on this hormone are the increase of fasting insulin and decrease in insulin sensitivity [29,30].
Many studies examined the effect of different training modes, volumes and intensities on insulin levels in older adults. From the available investigations, it appears that short-term (two weeks) training was unable to reduce fasting insulin level in a group of 28 healthy middle-aged (40-55 years) sedentary men, as shown by Heiskanen and coauthors [31]. More in detail, a program of six supervised cycle ergometer training sessions, characterized either by high-intensity (n = 14; 4-6 × 30 s all-out cycling/4-min recovery) or continuous moderate-intensity (n = 14; 40-60 min at 60% peak O 2 uptake) training did not affect fasting insulin concentration.
In contrast, Kirwan et al. [32] reported that nine months of endurance training reduced fasting insulin and improved insulin action. Seals and colleagues [33] (12 months of endurance training program), Kahn and coworkers [34] (six months of intensive endurance exercise program), Evans and coauthors [35] (10-12 months of endurance training program) reported similar results. Therefore, it appears that an intervention with a longer duration (e.g., from six up to 9-12 months) is required to observe significant changes in fasting insulin in older adults. On the other hand, some studies investigating the effects of 6/9-month training programs, such as the investigation by Goulet et al. [36], Dipietro and coworkers [37] or Ihalainen and collaborators [38] failed to report beneficial changes in insulin concentration.
The length of the training program seems to have an impact on insulin (in terms of levels or activity) depending on the age group in which the intervention is carried out. Herbert et al. [23] reported a moderate decrease in basal insulin following six weeks of high-intensity interval training (HIIT) in sedentary older males, suggesting that sprint training can reduce fasting insulin in older adults. Guezennec et al. [39,40] have investigated the impact of four months of weight lifting in athletes aged 35 years old. After maximal sessions, the level of insulin did not change significantly.
Other studies examined the effect of resistance training in insulin sensitivity in elderly subjects and reported that strength training induced improvement in insulin-stimulated glucose uptake promoted by glucose transporter type 4 (GLUT-4) in elderly [41]. Further studies investigated the influence of 12 weeks of high resistance training (weight lifting program) in the elderly and observed decreased insulin response [42].
Furthermore, when comparing young and middle-aged men, Sellami et al. [43] investigated the impact of 13 weeks of combined sprint and strength training on insulin concentration in blood. They reported a significant decrease in fasting insulin in both groups. Interestingly, the effect of age that was evident at baseline was no longer present post-training, suggesting that combined sprint and strength training can prevent the negative effects of aging in trained men [43].
From a molecular standpoint, it seems that lifelong regular physical activity leads to epigenetic mechanisms in terms of global DNA methylation patterns positively impacting on skeletal muscles' functioning in aged healthy individuals. One study has recently found that DNA methylation was statistically significantly lower in 714 promoters of genes involved in glycogen metabolism, glycolysis, oxidative stress resistance and muscle contraction, activity and myogenesis, whereas, methylation of introns, exons and CpG islands was apparently independent of physical activity practice [38]. Other cellular mechanisms that can explain how exercise can mitigate the mandatory age-related change in insulin levels include GLUT expression and translocation, skeletal muscle capillarization, improving insulin activity and sensitivity and favoring glucose uptake [42,[44][45][46][47][48][49][50][51][52].
Even if short-term training cannot affect insulin levels, it seems to be sufficient in improving or at least preserving insulin secretion pattern and response to oral glucose load. Some studies have, indeed, shown that a single bout of high intensity intermittent exercise [53], a couple of bouts of exercise [54,55] or light/moderate-intensity physical activity [56][57][58][59][60] can be sufficient in preserving insulin activity and response to oral glucose tolerance test (OGTT).
Summarizing (Table 2), based on the available studies, it appears that aging is associated with an increase of insulin level; a major part of this increase can be counteracted by exercise training. Exercise is, indeed, a full mediator of the relationship between inactivity time sedentary behaviors and insulin resistance [69]. Exercise, especially long-term (i.e., 12-24 weeks and not less than 8-10 weeks) [70][71][72] endurance, resistance and multimodal/combined training [73,74] or short-term (i.e., bouts of six weeks of HIIT) [75][76][77] training program, can positively impact on insulin levels [78], even though existing scholarly findings are not so clear-cut and warrant further investigations.   Whole-body strength training using 2-5 sets and 4-12 repetitions per exercise and 7-9 exercises per session for 6 mo No differences between groups and between before and after the intervention Abbreviations: AUC (area under the curve); CI (confidence interval); d (day); MET (metabolic equivalent task); min (minute); mo (month); NR (not reported); OGTT (oral glucose tolerance test); OR (odds-ratio); w (week); y (years).

IGF-1, Aging and Physical Activity
The IGF1 gene is situated on the long arm of chromosome 12. IGF-I is an endocrine and autocrine/paracrine growth factor expressed by multiple cell types. It plays a key role in the growth of cells, muscle, cartilage, bone, skin, and controls cell growth. The concentration of IGF-1 in blood peaks around adolescence and then declines after middle-age. This reduction in anabolic hormones has been termed the 'somatopause', and is suggested as a mechanism for the process of aging.
However, other studies failed to observe any change in IGF-I following six months' endurance training in~67-year-old males. Herbert et al. [23] investigated the difference between endurance-trained master athletes (~60 years) and lifelong sedentary older adults (~62 years) and observed greater serum IGF-1 concentration in the trained compared to the sedentary subjects (~18.4 vs.~13.1 µg/dL, respectively). Moreover, when exposing sedentary individuals to an endurance training program of 150 min/week, there was a small, non-significant increase in IGF-1 (~8% increase).
In addition, few studies explored the influence of sprint training on IGF-1 in older adults. Herbert et al. [23] observed that old (~62 years) sedentary subjects experienced a large increase in IGF-1 following 12 weeks' preconditioning and HIIT (~13.1 to~16.9 µg/dL). Although six weeks of preconditioning of 150 min/week accounted for 8% of the change in IGF-1, HIIT was responsible for a further 21% increase (28% greater than baseline). Findings from the same study suggest a trivial change in IGF-1 post-HIIT in age-matched master athletes. Therefore, post-HIIT, the sedentary individuals and master athletes had IGF-1 concentrations that were not significantly different.
Furthermore, when looking at the alteration of IGF-1 after resistance training in older adults, Parkhouse et al. [80] observed an increase in~68 year old females' circulating IGF-1. However, a recent investigation reported decreased systemic IGF-1 following 12 weeks' resistance training in older adults (74 ± 6 years) with an increase in lean mass [81].
As such, Arnason et al. [81] hypothesized that IGF-I was redistributed from circulation into tissue during periods of anabolism. As a result of the ambiguity in the findings, the role of IGF-I in the adaptive process to exercise during middle and older age remains unclear. The majority of studies reported that resistance training can increase the concentration of IGF-1 in blood and increase muscle mass and function [82][83][84][85][86][87][88][89][90][91][92][93][94][95][96][97]. Yet, more longitudinal studies are needed to explore the influence of resistance training on IGF-1 in older adults, given the presence of discrepancies among the findings.
In a recent study, Sellami et al. [43] investigated the influence of age on somatotropic hormones. They observed that young males had greater serum IGF-1 concentration than middle-aged men. Moreover, Sellami et al. [43] reported that 13 weeks of combined sprint and resistance training increased circulating IGF-1 in middle-aged participants. Furthermore, the effect of age that was apparent at the study commencement was abrogated post-training, suggesting that exercise can counteract the effect of age on IGF-1 in middle-aged men.
Taken together (Table 3), these data suggest that HIIT, resistance and combined training may be a countermeasure to the age and lifestyle-related reduction in IGF-1, activating some gene pathways and protein cascades [98,99]. Randomized allocation to a resistance followed by endurance training (n = 12), endurance training followed by resistance training (n = 12, interval resistance-endurance training (n = 12) and a control (n = 12) groups No differences among the groups and no difference between before and after the intervention

Growth Hormone, Aging and Physical Activity
Growth hormone (GH) secretion decreases with age, resulting in a downstream reduction in IGF-1 levels. This change, termed as the somatopause, is associated with loss of vitality, muscle mass, physical function, and an increased risk of frailty, cardiovascular disease, and adiposity, amongst others [100].
Veldhuis et al. [101] showed that GH secretion during puberty varied between 1-1.5 mg/day, while elderly people can produce only 50 µg/day. Several factors may be responsible for this decline, such as physical inactivity, poor nutrition, and subsequent changes in body composition. Moreover, Khan et al. [102] found that GH pulse decreased, and this decline was related to the alteration of hypothalamic and somatostatin hormones.
Moreover, GH has a beneficial neuroprotective effect [103] mainly due to the activation of anti-apoptotic pathway [104], this one particularly studied in literature. GH is also able to act on brain derived neurotrophic factor (BDNF) and neurotrophin-3 (NT3) [103] which, in turn, are particularly sensitive to physical activity.
Until now, there have been no studies that have explored the impact of endurance training on GH in older adults. Deuschle et al. [105] studied 11 elderly male marathon runners compared to 10 age-matched male sedentary people (controls), in order to study plasma concentration of GH, total and free IGF-I/II and IGF-binding protein-1, 2, and 3 and insulin. In particular, authors did not find any differences between runner vs controls, except for IGF-binding protein-1 and 2 increased in runners.
Moreover, Vanhelder al. [106] found similar results with a group of men aged 24-54 years who participated in resistance training for one year. The program was composed of two exercises (exercise protocol 1: vertical leg lifts at 85% of the subjects seven repetition maximum (SRM)/exercise protocol 2: vertical leg lifts with one third of the previously used load). The results showed that GH increased immediately after 5, 10, 25 min of exercise protocol 1. However, there was no significant increase after exercise protocol 2. These findings suggest that the frequency, duration of exercise play an important role in the regulation of GH secretion. Generally, the available studies showed that the frequency and intensity of resistance training are important factors in the regulation of GH secretion.
Recently, Sellami et al. [43] reported that younger adults had greater GH at rest and in response to sprint exercise than middle-aged participants. However, 13 weeks of combined sprint and resistance training abrogated this age effect and increased GH at rest and post-exercise in both young and middle-aged participants.
Summarizing (Table 4), very few studies have investigated the effect of physical activity and training on GH levels in elderly subjects [74,107,108], generally reporting negative findings. Further studies are needed to elucidate the mechanism of exercise on GH. Randomized allocation to a resistance followed by endurance training program (n = 12), endurance training followed by resistance training (n = 12), interval resistance-endurance training (n = 12) groups

Glucagon, Aging and Physical Activity
Glucagon is a peptide hormone, belonging to the secretin family of hormones, produced and released by the alpha cells of the pancreas. Being the major catabolic hormone of the human body, it increases blood glucose and fatty acids concentration, differently from insulin [109].
Of note, no studies investigated the effect of physical activity on glucagon concentration in elderly subjects, with the exception of Hagberg and coworkers [82], who found no changes in trained older subjects, whereas untrained individuals reported increases in glucagon levels.

Cortisol, Aging and Physical Activity
Cortisol, the primary stress hormone, is a steroid belonging to the glucocorticoid family, produced and released by the zona fasciculata of the adrenal cortex. This hormone plays a key role in controlling blood glucose and metabolism in general. Studies exploring the impact of age on cortisol have shown that cortisol increases with human aging. Seaton [110] reported that there was an elevation of night time cortisol levels in elderly individuals and this increase could be caused by stressful factors such as insomnia. Our laboratory has demonstrated that middle-aged men have higher basal cortisol concentrations than young men [111].
There are only a few studies that have examined the effect of exercise training on cortisol in elderly subjects. Herbert et al. [23] investigated the difference between lifelong sedentary and endurance-trained master athletes and observed no difference in basal cortisol. Moreover, when exposing sedentary individuals to an endurance training program of 150 min/week, there was no alteration to basal cortisol. Similarly, De Souza Vale et al. [112] investigated the effect of three months of water aerobics training in elderly women and reported no alteration to basal cortisol. However, an increase in cortisol following six-week HIIT in master athletes has been observed, with a concomitant increase in peak power output [23].
In middle-aged men, we have previously observed no alteration to basal cortisol following combined sprint and resistance training, however the acute cortisol response to a supramaximal sprint was elevated post-training [111].
Regarding the aging-related changes in the effect of exercise training on cortisol level, Kraemer et al. [113] compared the level of cortisol in young and older men after heavy resistance training three times per week for 10 weeks. Results showed a decline in resting cortisol at three and 10 weeks in the older group. However, Häkkinen et al. [114] reported that elderly subjects and middle-aged subjects did not experience any change in cortisol after six months' progressive resistance training. Similarly, Izquierdo et al. [115] investigated the effect of 16 weeks of progressive resistance training in older and middle-aged participants and observed no change in cortisol in the middle-aged group, but a decrease in the elderly group.
In summary, given the ambiguity of cortisol adaptation to resistance training, more research is required to determine the effect of training variables (duration, intensity, volume, frequency) and participant characteristics (age, training status, sex) on cortisol level.

Cathecolamines, Aging and Physical Activity
Catecholamine levels have been found to be different between young (20-years-old) and middle-aged men (40-years-old), with plasma noradrenaline concentrations being significantly lower (p < 0.05) in the young group when compared to the aged group. However, the precise neurobiological mechanisms leading to this difference in concentration levels are not very well-known and conflicting findings have been reported in the literature.
For instance, Hoeldtke et al. [116] showed that basal plasma noradrenaline concentration was greater in the elderly due to age-affected sympathetic nervous activity or sensitivity to sympatho-adrenal stimulation, without any difference in noradrenaline clearance. On the other hand, other authors found that clearance of noradrenaline tended to diminish with advancing age, which may contribute to the increased plasma concentrations observed.
Of note, no study has examined the effects of exercise training on catecholamines in older adults. In fact, the majority of studies investigated the impact of different types of training (sprint, endurance, resistance training) on catecholamine in young individuals [117][118][119][120][121][122][123]. Results were found to be at variance, and most of the time it was concluded that duration, intensity and type of training (aerobic and anaerobic) are the principal factors that induced alterations in catecholamine responses.
A notable exception was the investigation carried out by Poehlman and Danforth [71], who assessed the effects of 8 weeks of an endurance training program on norepinephrine kinetics in a sample of 19 older persons aged 64 ± 1.6 yr. Resting concentrations of norepinephrine were found to be increased by 24% after cycling exercise due to a 21% increase in norepinephrine appearance rate, whereas no change in norepinephrine clearance could be detected.
As such, due to the dearth of data and information, future studies are needed to determine the effect of different exercise training modes and moderator variables on catecholamine secretion and catecholamine circulating concentration in older adults.

Discussion and Future Prospects
There is an increasing interest in exercise training, as a therapeutic lifestyle strategy to attenuate the hallmarks of aging and improve health. Exercise training attenuates many markers of biological aging and one of the underlying mechanisms may be through the promotion of a more 'youthful' endocrine profile. In vitro experiments suggest that cells treated with plasma isolated from younger individuals are healthier or more 'youthful' than those treated with plasma from their older peers. Therefore, in situ cells exposed to a youthful systemic environment will likely have improved functioning compared to those exposed to an older systemic environment. Evidence cited in this review suggests that it is possible that exercise can act as a countermeasure to endocrinological aging.
Regarding this last point, it is necessary to keep in mind that both similarities and differences in aging between/within genders exist.
However, despite such an increasing body of interest, the physiological effects of physical activity and exercise on glucoregulatory hormones in elderly subjects are relatively understudied. Evidence of the impact of training is generally circumstantial and randomized studies, carried out with high methodological rigor and quality are few or lacking for some hormones. Whereas insulin has captured the attention of scholars, there is a relative dearth of data and information for other hormones.
Given the importance of the topic of counter-aging effect of sports and physical activity and considering the epidemiological and clinical burden of aging and age-related disorders, more attention in the field is needed. Longitudinal studies employing larger sample sizes are warranted.

Conflicts of Interest:
The authors declare no conflict of interest.