Abstract
Physical inactivity is a pandemic that contributes to several chronic diseases and poses a significant burden on health care systems worldwide. The search for effective strategies to combat sedentary behavior has led to an intensification of the research efforts to unravel the biological substrate controlling activity. A wide body of preclinical evidence makes a strong case for sex steroids regulating physical activity in both genders, albeit the mechanisms implicated remain unclear. The beneficial effects of androgens on muscle as well as on other peripheral functions might play a role in favoring adaptation to exercise. Alternatively or in addition, sex steroids could act on specific brain circuitries to boost physical activity. This review critically discusses the evidence supporting a role for androgens and estrogens stimulating male physical activity, with special emphasis on the possible role of peripheral and/or central mechanisms. Finally, the potential translation of these findings to humans is briefly discussed.
Introduction
Physical inactivity has emerged as a global health crisis, being the fourth leading cause of death worldwide according to the World Health Organization (WHO) (WHO 2009). Not meeting the minimum physical activity recommendations (WHO 2010) is estimated to cause 6% of the burden of disease from coronary heart disease, 7% of type 2 diabetes and 10% of both breast and colon cancer (Lee et al. 2012). In contrast, 15 min of daily moderate-intensity exercise reduces all-cause mortality by 14% and extends by 3 years the life expectancy (Wen et al. 2011). Notwithstanding, one-third of adults worldwide are physically inactive and the prevalence increases dramatically among adolescents, with four out of five not reaching the minimum recommended levels according to self-reported measures (Hallal et al. 2012). Accelerometer data show even lower rates of adherence among adolescents, with only 8% of 12- to 15-year-old individuals attaining sufficient physical activity (Troiano et al. 2008). Taking into consideration the prevalence and risks associated with physical inactivity, as well as the fact that it can be easily and effectively prevented (Heath et al. 2012), it is without surprise that the promotion of physical activity has become a public health priority (Kohl et al. 2012).
The predisposition to engage in physical activity is a highly complex trait in humans, probably determined by the interaction of biological, psychological, social and environmental factors (Bauman et al. 2012). The role for a biological substrate influencing the levels of physical activity is a relatively recent concept supported by several lines of evidence, including human studies but mostly preclinical investigations (Bouchard & Rankinen 2006). Based on a cohort of 37,000 twin pairs, Stubble et al. described that the heritability of exercise participation ranged between 48% and 71% (Stubbe et al. 2006). Nevertheless, the extent to which genetic factors explain adult exercise behavior is highly heterogenic between studies, most likely due to differences in the methodology used to assess physical activity (Simonen et al. 2002, Eriksson et al. 2006). In addition, candidate gene studies have pointed out several hormonal- and neurotransmitter-related genes associated with physical activity (Stefan et al. 2002, Salmen et al. 2003, Simonen et al. 2003, Loos et al. 2005). However, the two genome-wide association studies (GWAS) published so far do not support a prominent role for most of these candidate genes (De Moor et al. 2009, Kim et al. 2014). This inconsistency might be explained in part by the limitations of physical activity data based on questionnaires or diaries, which show poor accuracy in detecting light-to-moderate activities and are prone to error because of individual biases (Strath et al. 2013). The latter caveat is overcome by recent efforts to implement the use of automated monitors in physical activity research. Nonetheless, the strongest body of evidence supporting that physical activity is biologically regulated still comes from preclinical studies.
Androgens and estrogens are key regulators of reproductive organs but also exert multiple effects on non-reproductive tissues, including brain, muscle, adipose tissue and bone. Testosterone, the principal circulating androgen, may stimulate the androgen receptor (AR) either directly, or indirectly following 5 alpha-reduction to dihydrotestosterone (DHT), as well as the estrogen receptors (ERα and ERβ) after testosterone aromatization into estradiol (E2) (Callewaert et al. 2010). Even though research into the biological control of physical activity has intensified in recent years, early findings in the 1920s already showed that wheel running in male rats was abolished after orchiectomy (ORX) (Gans 1927). Later studies pinpointed testosterone as the gonadal product stimulating wheel running, albeit the extent to which prior aromatization into E2 was required remained debated (see discussion in the ‘Replacement and pharmacological studies in adult rodents’ section). Physical activity is a sexually dimorphic trait in rodents (Eikelboom & Mills 1988), with females being more active than males and with fluctuations according to the estrous cycle (Wollnik & Turek 1988). Similar to what is observed in males, wheel running in females is abolished upon gonadectomy and can be restored following treatment with E2 (Berchtold et al. 2001). Based on these and other preclinical data, sex steroids have been suggested as strong candidates for contributing to the biological substrate influencing physical activity (Lightfoot 2008). Androgens are well known for their potent anabolic actions on muscle and we have previously reviewed the cellular targets and molecular pathways implicated (Dubois et al. 2012). Therefore, it is conceivable that testosterone increases activity in males by favoring adaptation to exercise (peripheral mechanisms). In addition or alternatively thereto, testosterone could act on the specific brain neural circuitries that have been identified to control physical activity levels (central mechanisms). Our understanding of the neural basis underpinning the effects of sex steroids on activity has advanced markedly over the last decade. Nevertheless, most of these studies have focused on the female brain (reviewed in Xu et al. 2017), whilst the central mechanisms implicated in male activity are less documented.
The recent publication of large clinical trials of testosterone therapy in elderly men has reignited the debate on the risks and benefits of testosterone replacement (Basaria et al. 2010, Snyder et al. 2016, Storer et al. 2016). In this context, the positive impact of testosterone stimulating physical activity and the underlying mechanisms of action represent an additional consideration for testosterone replacement in elderly, mobility-impaired, frail or sarcopenic men. Here, we review the data supporting a role for androgens and estrogens regulating male physical activity and provide an overview of the potential mechanisms implicated. How sex steroids control physical activity in females is reviewed in Lightfoot (2008) and Xu et al. (2017). We will first review some basic concepts regarding physical activity and then discuss several lines of preclinical evidence supporting the notion. In the following sections, we will review the impact of androgens in the physical ability to partake in exercise and examine the specific brain circuits that could be underlying the regulation of activity by testosterone. In the last section, we will briefly discuss what is known regarding the effects of androgens and estrogens on physical activity in men.
Categories of physical activity
Physical activity is defined as any bodily movement produced by skeletal muscles that results in energy expenditure (Caspersen et al. 1985). Although the term ‘exercise’ is often interchanged with physical activity, exercise is indeed a subset of physical activity that is planned and structured and implies a purpose to achieve health or fitness benefits (Caspersen et al. 1985). In our daily life, we also engage in incidental physical activity when performing home- or work-related activities or during transportation (Strath et al. 2013). Nevertheless, the latter activities can be performed in an exercise manner and thus it is not always straightforward to discern between the voluntary (planned) and incidental (spontaneous) domains of physical activity (Garland et al. 2011). Notably, an increase in energy expenditure associated with non-exercise activity has been shown to prevent fat gain in humans (Levine 1999). Therefore, physical activity derived from daily tasks may also play a role in disease prevention and should be included together with exercise when assessing physical activity.
With regard to preclinical animal studies, wheel running in rodents has been extensively used as a model of human exercise (Garland et al. 2011, Novak et al. 2012). Similar to the ‘runner’s high’ experienced during endurance exercise in humans, wheel running is rewarding for rodents and both rats and mice show preference for environments previously associated with running (Lett et al. 2001, Fernandes et al. 2015). When given free access to running wheels, mice reduce the time spent in cage floor activity at the expense of wheel running time, the net result being an increase in their total activity levels (De Visser et al. 2005, Silvennoinen et al. 2014). Spontaneous physical activity (SPA) refers to the non-exercise component of activity and therefore should not be used as interchangeable for wheel running, although very often these two parameters correlate reasonably well. There are several methods to assess SPA in rodents, including infrared photobeams, radiotelemetry, video tracking systems or force plates (Silvennoinen et al. 2014, Teske et al. 2014, Chabert et al. 2016). The main difference between these approaches is their accuracy in detecting ambulation and climbing as well as more subtle stationary movements, such as grooming or feeding. Regardless of the method used, a long (i.e., hours) period of acclimatization to the activity chamber is required to discharge the novelty effects on locomotion (Teske et al. 2014). Studies failing to meet the former methodological requirement were excluded from this review. It is important to note that we focus here on volitional activity, and therefore do not include preclinical studies using forced exercise capacity tests. Forced treadmill running or swimming require the use of aversive stimuli to motivate the animals, thus inducing a substantial psychological stress (Sasaki et al. 2016). In addition, rodents do not always display consistent swimming behaviors and some animals will just float, bob and/or dive (American Physiological Association 2007). Therefore, we focused on voluntary wheel running, in which animals have a complete control on the amount and intensity of exercise.
Preclinical evidence supporting a role for androgens and estrogens regulating male physical activity
Most of our knowledge regarding the role of biological factors controlling physical activity derives from rodent studies and thus the overall significance to human health remains to be determined. Nevertheless, evidence from preclinical studies in rats and mice clearly shows the importance of androgens and normal AR and ERα signaling in the regulation of male activity.
Replacement and pharmacological studies in adult rodents
The first observations showing that ORX abrogates wheel running activity in rats date from almost a century ago (Gans 1927). Subsequent studies extended these findings to mice as well as other rodent species (Daan et al. 1975, Morin & Cummings 1981, Dark & Zucker 1984, Jechura et al. 2000) and demonstrated the implication of testicular testosterone production (Daan et al. 1975, Roy & Wade 1975). It is now well established that replacement with testosterone rescues wheel running activity in a gradual dose-response manner following ORX (Butler et al. 2012). For instance, the average daily distance run by ORX mice increases from 1 to almost 4 km following 2 weeks of treatment with physiological doses of testosterone (Jardí et al. 2018). Consequently, the combination of testosterone and wheel running mediates an adaptational shift to a more oxidative fiber profile in the skeletal muscle of ORX (Allen et al. 2001, Ibebunjo et al. 2011). In gonadally intact male rats, administration of supraphysiological doses of testosterone also increases running distances (McGinnis et al. 2007). Similar to humans, aging in male mice is associated with a progressive decline in activity (Fig. 1; Hamrick et al. 2006) and this can be partly attenuated by the combination of exogenous testosterone and low-intensity physical training (Guo et al. 2012). It is important to note, however, that the concentrations of testosterone in male mice appear to remain constant with aging (Hamrick et al. 2006), in contrast to what is observed in elderly men (Wu et al. 2008). Thus, it is unlikely that a reduced androgen bioactivity is the main driver of the decreased SPA in old mice, even when this is responsive to testosterone supplementation in the setting of physical training.
Changes in SPA with age in male C57BL/6 mice. Mean values are shown with standard error means as vertical bars. (Data from Hamrick et al. 2006.)
Citation: Journal of Endocrinology 238, 1; 10.1530/JOE-18-0125
Many of the actions of testosterone in male behavior are mediated by estrogen receptors (ERs). For instance, testosterone acts predominantly as a prohormone for E2 in the programming of male-typical and territorial behavior in mice, although execution of this behavior relies on the AR (Wu et al. 2009). It remains debated whether this dual androgenic and estrogenic action of testosterone also applies to physical activity. Initially, testosterone was suggested to stimulate wheel running exclusively through aromatase-dependent mechanisms (Stern & Murphy 1971, Roy & Wade 1975). Treatment with E2 was more effective than testosterone in stimulating wheel running in ORX rats and the non-aromatizable androgen DHT showed no effects (Roy & Wade 1975). Also in ORX rats, biologically effective doses of the antiandrogen cyproterone acetate failed to inhibit the stimulatory actions of testosterone on wheel running (Stern & Murphy 1971). In accordance, treatment with the 5α-reductase inhibitor dutasteride did not alter the levels of wheel running in male Zucker obese rats (Sato et al. 2013). In the latter study, however, the lack of effects could be potentially explained by the already low levels of wheel running of these animals (Stern & Johnson 1977) and/or by a compensatory effect related to the secondary increases in testosterone concentrations (Amory et al. 2007). Later studies challenged the notion of aromatase being indispensable for testosterone effects on wheel running. In male mice, treatment with DHT partly restored the drop in wheel running following ORX (Karatsoreos et al. 2007, Iwahana et al. 2008), although to a lesser extent compared with testosterone (2.3 km vs. 3.8 km; Jardí et al. 2018) and both testosterone and DHT were equally effective restoring SPA (Jardí et al. 2018). It is noteworthy that DHT metabolites 5α-androstane-3α, 17β-diol (3α-diol) and 5α-androstane-3β, 17β-diol (3β -diol) may initiate behavioral processes through AR-independent mechanisms, namely by acting on gamma-aminobutyric acid (GABA) type A (Frye et al. 1996) or ERs (Frye et al. 2008a ). Nevertheless, we showed that DHT failed to rescue wheel running in AR knockout (ARKO) mice, ruling out the implication of AR-independent mechanisms (Jardí et al. 2018). Additionally, the activity of ORX+testosterone-treated mice was partially lowered by the AR antagonist enzalutamide (Wu et al. 2016), further supporting the involvement of AR signaling in the stimulation of wheel running (Jardí et al. 2018). Taken together, these data make a strong case for the contribution of AR to the effects of androgens on activity, at least in male mice, though its importance is somewhat less compared to aromatization and ER signaling.
In an attempt to investigate the estrogenic component of testosterone actions, ORX rats and mice have been treated with systemic E2 (Roy & Wade 1975, Ogawa et al. 2003, Bowen et al. 2011, 2012, Blattner & Mahoney 2015, Royston et al. 2016). It should be noted, however, that the physiological relevance of this approach is uncertain and deserves special consideration. Serum E2 concentrations in both gonadally intact and ORX+testosterone-, but not E2-treated mice, are below the limit of detection of sensitive analysis methods using mass spectrometry (McNamara et al. 2010, Nilsson et al. 2015, Laurent et al. 2016b ), the ‘gold standard’ for sex steroids measurements. Circulating levels of E2 are also extremely low in male rats, at the level of a few picograms per milliliter (Quignot et al. 2012). Therefore, a ‘physiological’ E2 dose is virtually impossible to define in male rodents, and most ‘replacement’ doses are almost by definition supraphysiological. Remarkably, aromatase is highly expressed in several brain regions of male rodents, including the bed nucleus of the stria terminalis, the olfactory tubercle, the medial amygdala (MeA) and the hypothalamus, suggesting a role for the autocrine and paracrine actions of estrogens (Zhao et al. 2007, Stanić et al. 2014). Overall, we cannot draw conclusions about the physiological role of E2 on physical activity levels based on studies treating male rodents with systemic E2, even when pharmacological doses of E2 restore wheel running following ORX (Roy & Wade 1975, Ogawa et al. 2003, Bowen et al. 2011, 2012, Blattner & Mahoney 2015, Royston et al. 2016). Although treatment with the aromatase inhibitors letrozole and exemestane did not affect wheel running either in gonadally intact or ORX+testosterone-treated mice (Bowen et al. 2011), this study needs further technical validations. The ultimate proof of the role of ERs on male physical activity comes from studies in knockout mice lacking ERα or the aromatase enzyme (see discussion in the ‘Global KO mouse models’ section).
Androgenic-anabolic steroids (AAS) and selective androgen receptor modulators (SARMS)
AAS are synthetic derivatives of testosterone that were developed in an attempt to maximize the anabolic activity of testosterone. The prolonged abuse of AAS in men is related to several psychiatric adverse effects, including irritability, mood swings, mania (‘steroid rage’ at supraphysiological doses) and depression (Pope & Katz 1994, Bjørnebekk et al. 2017). Although scarce, a few preclinical studies have explored whether doses of AAS that model human abuse patterns affect physical activity levels. In contrast to testosterone, both nandrolone (McGinnis et al. 2007, Tanehkar et al. 2013) and stanozolol (McGinnis et al. 2007) reduced wheel running activity in gonadally intact male rats. Similarly, the combination of nandrolone and methandrostenolone decreased voluntary wheel running in mice (Onakomaiya et al. 2014). Paradoxically, despite the detrimental effects of nandrolone on wheel running in male rats (McGinnis et al. 2007, Tanehkar et al. 2013), the combination of both interventions improved their running endurance (Vanzyl et al. 1995). As for the mechanisms of action, AAS and endogenous testosterone differ in both the affinity to bind AR and the ability to be aromatized (Bergink et al. 1985, Fragkaki et al. 2009). Possibly, AAS regimes reduce wheel running by interfering with the endogenous testosterone production and hence E2 (Barone et al. 2017). Alternatively, AAS might act directly on several neurotransmitter systems in the brain that affect physical activity (Mhillaj et al. 2015).
The need to improve both the benefit/risk profile and pharmacokinetics of androgens spurred the development of selective androgen receptor modulators (SARMs) (Mohler et al. 2009). SARMs received initially a great deal of clinical attention for their potential use in the treatment of muscle wasting conditions. To date, yet, none of these compounds has shown unequivocal efficacy in clinical trials (Almeida et al. 2017). Preclinical data show that SARMs display a better myotrophic index compared to their steroidal analogues. In ORX mice, treatment with the SARM MK-4541 led to an increase in muscle mass and function comparable to DHT but without the sparing effects on seminal vesicles (Chisamore et al. 2016). The reduction of SPA following ORX, however, was only partially rescued by MK-4541 (Chisamore et al. 2016). In a similar pattern, treatment with SARM-2f completely restored muscle mass in ORX mice but was less effective increasing their wheel running activity (Morimoto et al. 2018). In a preclinical mouse model of muscular dystrophy, an enobosarm analog extended the survival of mice but failed to increase their low SPA (Ponnusamy et al. 2017). Nevertheless, the severity of the pathology of these mice may have impeded an increase in their SPA levels. The effects of SARMs on physical activity are worthy of further study and could represent an additional benefit in the treatment of muscle wasting conditions.
Perinatal studies
So far, we discussed how androgens stimulate physical activity in adult rodents. It is noteworthy that behavioral responses to sex steroids during adulthood are influenced by sex steroids acting on the nervous system during early critical periods. In their seminal work in 1959, Phoenix et al. were the first to dichotomize hormonal effects in two groups: organizational vs. activational (Phoenix et al. 1959). The organizational-activational theory states that hormones sculpt neural tissues during development so that they respond differentially to hormonal activation in adulthood (Phoenix et al. 1959). The study of the organizational effects of sex steroids has mainly focused on the programming of mating behavior (Arnold 2009). Although less evidence is available for non-sexual behavior, the results from several preclinical studies suggest that the perinatal sex steroid milieu might also determine physical activity in adulthood.
The perinatal surges of testosterone in male rodents during the late embryogenic period and the first hours of life are critical for the male brain to become capable of producing male sexual behavior (masculinization) whilst losing its ability to respond in a female-like pattern (defeminization; Lenz & McCarthy 2010 for review on this topic). Apart from the perinatal period, a growing body of evidence shows that during puberty, testosterone is also required for the full masculinization and defeminization of male brain (Schulz et al. 2004, De Lorme & Sisk 2016, Sano et al. 2016). Paradoxically, the organizational effects of testosterone on the developing male brain are actually driven by ERs, suggesting the implication of brain aromatase activity (Ogawa et al. 2000, Wu et al. 2009, Juntti et al. 2010). In females, prepubertal estrogen levels are very low, and the traditional dogma states that sexual behavior is merely the result of the absence of perinatal testosterone organizational actions. Recent data challenge the latter hypothesis demonstrating that E2 actions outside the perinatal period are required for the normal development of female sexual behavior (Brock et al. 2011). The neural substrate underlying the persistent and irreversible effects of sex steroid actions on the developing brain remains mostly unknown. Although not an exhaustive list, some suggested mechanisms include epigenetic modulation (Nugent et al. 2015), regulation of cell death and proliferation (Wu et al. 2009) and changes in synaptic patterning (Schwarz et al. 2008).
Activity is a sexually dimorphic behavior in rodents, with females showing higher levels of wheel running and SPA compared to males (Eikelboom & Mills 1988). Gender differences in activity remain after gonadectomy and thus cannot be explained by hormonal dimorphism in adulthood (Gentry & Wade 1976, Blizard 1983, Broida & Svare 1984, Bowen et al. 2012, Kuljis et al. 2013). Differences in sex chromosomes as such are also insufficient to explain the dimorphism in activity, as evidenced by studies using the mouse model of 4 core genotypes, in which sex chromosome complement is independent of gonadal phenotype (De Vries et al. 2002, Kuljis et al. 2013). In a similar pattern to XX females, ovariectomy of mice with a deletion of the testis-determining gene Sry results in running longer distances than ORX males (Kuljis et al. 2013).
Accumulating literature suggests a role for sex steroid actions during the perinatal period determining gender differences in activity. ORX at postnatal day 0 (P0) results in higher levels of SPA than that at P5, 10 or 25 in male mice (Broida & Svare 1984). Also in male mice, estrogen deficiency during early life leads to a higher wheel running response in adulthood following E2 treatment (Royston et al. 2016). From these observations, it is reasonable to assume that the neonatal testosterone surge programs male physical activity and that this might involve aromatase-dependent processes (Fig. 2). In support of this premise, female rats treated neonatally with androgens develop a resistance to E2 stimulation of wheel running in adulthood, which can be prevented by treating pups with antisense oligodeoxynucleotides to ER mRNA (Blizard 1983, McCarthy et al. 1993, Royston et al. 2016). Whether the neural mechanisms underlying the organizational effects of testosterone on activity are similar to those described for mating behavior remains to be investigated. Similarly, the organizational effects of testosterone outside the perinatal window on physical activity (i.e. puberty) await confirmation.
Schematic overview of the overall proposed mechanisms of action of testosterone on male activity (see text). A, androgens; CNS, central nervous system; E2, estradiol; PNS, peripheral nervous system; +, stimulatory effect; ++, dominant stimulatory effect.
Citation: Journal of Endocrinology 238, 1; 10.1530/JOE-18-0125
Global KO mouse models
Studies in transgenic mice lacking either steroid hormone receptors or the aromatase enzyme provide further insights into the role of sex steroids regulating physical activity. However, the interpretation of the results obtained from these mouse models is hampered by alterations of the hypothalamic–pituitary–gonadal (HPG) axis. In addition, it is reasonable to expect changes in the sex-steroid-dependent brain organization (see previous section) caused by the gene deletion during development.
Androgen receptor knockout (ARKO) mice
Several ARKO mouse models targeting either exon 1, 2 or 3 have been generated using the Cre-LoxP system (Fan et al. 2005, Ophoff et al. 2009a , Rana et al. 2011). Our group reported an almost complete abrogation of wheel running as well as a reduced SPA in exon 2-deleted male ARKO mice (Ophoff et al. 2009a ). A similar low activity phenotype was observed in mice with a targeted disruption of AR exon 1 (Fan et al. 2005) and in a DNA-binding domain ARKO mouse model that only retains non-genomic actions (Rana et al. 2011). The fact that global ARKO mice are less physically active might predispose them to develop late-onset adiposity (Fan et al. 2005, Rana et al. 2011). Global ARKO mice appear to suffer from an impaired gonadal testosterone production (Callewaert et al. 2009) and might therefore not be useful to discern between purely AR-mediated effects and those secondary to E2 deficiency. Indeed, treatment with testosterone completely restored wheel running in male ARKO mice, an effect associated with a normalization of their E2 concentrations in brain (Jardí et al. 2018). Thus, it is reasonable to assume that the low activity phenotype of ARKO mice is derived in large part from a lack of testosterone substrate for aromatization into E2.
Aromatase knockout mice
Mice in which E2 biosynthesis is disrupted by deletion of the Cyp19a1 gene represent a useful tool to study the implication of E2 in the regulation of male physiological processes (Cooke et al. 2017). Most studies in male aromatase KO mice agree that both wheel running (Watai et al. 2007, Brockman et al. 2011) and SPA (Hill et al. 2007) are diminished in the absence of E2. Nevertheless, one study reported increased levels of wheel running in male aromatase KO mice, as part of a broader spectrum of compulsive behaviors (Hill et al. 2007). It should be noted that the authors analyzed pre-plateau phases of activity, which could have led to a mistaken estimation of the ‘true’ levels of wheel running. Be that as it may, aromatase KO mice also show an impaired negative feedback of the HPG axis, with higher levels of serum testosterone and gonadotropins (Fisher et al. 1998, Jones et al. 2000, Öz et al. 2000). The fact that aromatase KO mice display a low activity phenotype even in the presence of excess testosterone indicates a crucial role for aromatization in the testosterone-induced stimulation of male physical activity. Supporting this notion, male aromatase KO mice do not show an ORX-induced reduction of wheel running (Brockman et al. 2011).
Estrogen receptor α (ERα-) and β knockout (ERβKO) mice
The generation of ERα and ERβKO mice has advanced our understanding of the mechanisms implicated in the estrogenic regulation of male physiology (Cooke et al. 2017). Similar to male aromatase KO mice, ERαKO mice show increased levels of serum testosterone (Callewaert et al. 2009), are less active than their wild-type (WT) littermates when given access to running wheels (Blattner & Mahoney 2012), and develop elevated adiposity in late-life (Heine et al. 2000). As shown first in the seminal work of Ogawa et al. and later confirmed by others, the stimulatory actions of E2 treatment on wheel running activity are abrogated in ERαKO mice (Ogawa et al. 2003, Dworatzek et al. 2014, Blattner & Mahoney 2015). This is not the case for ERβKO mice in which the response to E2 is unaffected (Ogawa et al. 2003). Therefore, the estrogenic regulation of wheel running in males is primarily mediated by ERα, though this does not exclude the possibility that ERβ plays a role in modulating the circadian-rhythm patterns of activity, as shown for females (Royston et al. 2014).
‘Non-classical’ ER knock-in (NERKI) mice
Transgenic mice with a knock-in mutation in the ERα selectively abolishing estrogen response element (ERE) binding were originally generated to distinguish between classical and non-classical ERα actions in vivo (Jakacka et al. 2002). Male ‘non-classical’ ER knock-in mice (NERKI) mice show a reduced wheel running activity that does not increase following challenge with E2 (Blattner & Mahoney 2012, 2015). Based on these observations, it is tempting to speculate that ERα actions on wheel running activity are dependent on ERα binding to ERE elements.
G-protein-coupled receptor 30 knockout (GPR30KO) mice
The G-protein-coupled receptor 30 (GPR30) has been proposed as a membrane ER (Langer et al. 2010). There are important phenotypic differences between the four independent GPR30KO mouse models available, most likely reflecting that other factors apart from deletion of GPR30 are contributing to the observed effects (Langer et al. 2010). In two different GPR30KO mouse models, males exhibited normal SPA levels (Sharma et al. 2013, Kastenberger & Schwarzer 2014), but another study reported that GPR30KO male mice displayed a mild reduction in wheel running as well as an impaired left-ventricular cardiac function (Delbeck et al. 2011). It therefore remains uncertain whether GPR30 is physiologically relevant for physical activity, and if so, if it contributes to E2-induced responses and/or initiates independent mechanisms.
Sex-hormone-binding globulin transgenic (SHBG-tg) mice
Sex-hormone-binding globulin (SHBG) is a high-affinity binding protein for testosterone, DHT and E2 that is found circulating in humans but is almost completely lacking in rodent serum. We recently showed in a humanized mouse model expressing human SHBG that testosterone needs to be biologically available (i.e. not bound to SHBG) to produce its stimulatory effects on wheel running (Jänne et al. 1998, Jardí et al. 2018). In SHBG-tg mice, the majority of testosterone is bound by SHBG and cannot enter target tissues by the canonical way (Laurent et al. 2016a , b ). Male SHBG-tg mice display features of a mild androgen deficiency phenotype, including a 50% decrease in wheel running activity (Jardí et al. 2018). As the decline of androgen bioactivity in elderly men is also only mild (Wu et al. 2008), SHBG-tg mice may be a more suitable animal model to study late-onset hypogonadism than ORX.
The interpretation of studies using constitutive KO mice for steroid hormone receptors or the aromatase enzyme is subject to several limitations, including alterations of the HPG axis, developmental compensatory mechanisms and the impossibility to distinguish tissue-specific contributions. Nonetheless, findings in both aromatase and ERKO mice indicate that testosterone actions on physical activity require normal aromatase and ERα functions. In contrast, we cannot draw conclusions about the implication of AR from studies in global ARKO mice because of their severe hypogonadism. However, replacement and pharmacological studies made clear that the contribution of AR to testosterone effects cannot be dismissed as minimal or trivial (see discussion in the ‘Replacement and pharmacological studies in adult rodents’ section).
Mechanisms implicated in the effects of androgens and estrogens on male physical activity
Despite the large amount of preclinical data showing that androgens and estrogens control physical activity in males, the mechanisms involved are not well understood. Due to their pleiotropic nature, sex steroids could stimulate physical activity by acting on multiple organs or systems. Of particular interest, the use of genetic techniques to manipulate neuronal pools in a selective manner has allowed the identification of several sex-steroid-responsive neural circuitries implicated in the drive to engage in physical activity (Musatov et al. 2007, Xu et al. 2011, 2015, Correa et al. 2015). Besides these central mechanisms, androgens could also increase activity levels by improving muscle function (Dubois et al. 2012) and, in turn, the capacity to exercise. In other words, testosterone might influence both the motivation and/or the ability to partake in activity, referred here as central and peripheral actions, respectively (Fig. 2). Note that the term ‘peripheral’ as applied to this classification does not preclude the implication of the central nervous system (CNS) in the observed effects.
Peripheral actions
In the next section, we will examine whether androgens provide greater tolerance to exercise by improving muscle strength, motor skills, metabolism and/or other peripheral parameters in male rodents.
Muscle mass and strength
Both the mass and the ex vivo contractility of hindlimb muscles show a mild reduction following ORX or AR deletion in male mice (Axell et al. 2006, MacLean et al. 2008, De Naeyer et al. 2014, Dubois et al. 2014). The differences in strength disappear when normalized for muscle weight, suggesting that the intrinsic contractile properties of muscle remain unaltered (Axell et al. 2006, MacLean et al. 2008, Hourdé et al. 2009, De Naeyer et al. 2014). However, in five different myoblast or satellite-cell-specific ARKO models, the mass and contractility of hindlimb muscles were not, or marginally, affected (Ophoff et al. 2009b , Chambon et al. 2010, Dubois et al. 2014, Ferry et al. 2014, Rana et al. 2016), whilst all five models showed a partial decrease in the weight of the androgen-sensitive perineal muscles. Nevertheless, the expression of AR in the muscle cell lineage was required for the full gain in muscle mass induced by mechanical loading (Ferry et al. 2014), an effect that could be related to the modulation of a reserve pool of adult muscle stem cells by androgens (Kim et al. 2016). An overview of the main features of the different muscle-specific ARKO models is presented in Table 1. Overall, the former studies suggest a role for non-myocytic AR mediating the modest actions of androgens on murine muscle. In this regard, a recent study points toward neuronal AR contributing to maintain hindlimb muscle mass but not contractility in male mice (Davey et al. 2017). The cellular targets underlying the anabolic actions of androgens on muscle have been reviewed by Dubois et al. (2012).
Overview of muscle- and neuronal-specific ARKO mouse models.
| Cell type | Mouse strain | Ref | Serum testosterone/SV weight | Muscle mass | Grip strength | Contractile function | Motor performance | Wheel running | SPA |
| Myofibers | ARskm;HSACre(a,b) | Rana et al. (2016) | = | ↓ | ND | ND | ND | = | ND |
| ARskm;MCKCre(a) | Rana et al. (2016) | ↑ | = | ND | ND | ND | = | ND | |
| ARskm;HSACre | Chambon et al. (2010) | ND | = | ↓ | ↓ | ND | ND | ND | |
| ARskm;HSACreER | Chambon et al. (2010) | ND | = | ↓ | ND | ND | ND | ND | |
| ARskm;MCKCre | Ophoff et al. (2009b) | =c | ↓ | ND | (=) | =d | ND | ND | |
| ARsat+skm;MyoD | Dubois et al. (2014, 2015), Jardí et al. (2018) | =c | = | ↓ | (=) | =d | = | ND | |
| Neurons | ARNeu+glia;NesCre | Raskin et al. (2009) | ↑ | =e | ND | ND | ND | ND | (↑) |
| ARNeu+glia;NesCre | Juntti et al. (2010) | ((↑)) | ND | ND | ND | =f | ND | ND | |
| ARNeu;SynapsinICre | Yu et al. (2013) | ↑c | ND | ND | ND | ND | ND | ND | |
| ARNeu;CamKIIαCre | Davey et al. (2017) | ↑ | ↓ | ND | = | ND | ↓ | ↓ | |
| ARNeu;Thy1CreER(g) | Jardí et al. (2017b) | ↑c | = | = | ND | ND | = | ND |
aAR lox hemizygous mice show a phenotype of hyperandrogenization; bAR expression also reduced in brain and fat; cas assessed by SV weight; dtreadmill running; edetermined by DXA; frotarod test; gAR expression retained in the hypothalamus.
↑, increased; (↑), SPA restored to control values following normalization of testosterone levels; ((↑)), not statistically significant; ↓, decreased; =, unchanged; (=), minor impact on fatigue but no changes in maximal muscle tension; ND, not determined; neu, neurons; sat, satellite cells; skm, skeletal muscle myofibers; SPA, spontaneous physical activity; SV, seminal vesicles; testosterone, testosterone.
The ex vivo contractility studies referred to in the earlier sections do not support a role for an intrinsic muscle dysfunction impairing the physical ability of ORX and ARKO mice. Yet, these findings do not exclude the possibility that androgens regulate the neuromuscular function in vivo. Exercise training potentiates the stimulatory effects of androgens on muscle mass and strength (Guo et al. 2012, Cozzoli et al. 2013). Indeed, the combination of both interventions benefits endurance capacity in rodents (Vanzyl et al. 1995, Georgieva & Boyadjiev 2004, Cozzoli et al. 2013), presumably through increasing muscle function. However, ORX per se diminishes grip strength by only about 5–15% in some but not all studies (Borst et al. 2007, Windahl et al. 2011, White et al. 2013, Chisamore et al. 2016). The discrepancies between studies might be due to the lack of sensitivity of the conventional forelimb grip strength test (Takeshita et al. 2017). Additionally, the motivation to hold onto the grid may vary between mice, causing inconsistencies in the results. Be that as it may, a modest reduction (~10%) in grip strength was also found in two muscle-specific ARKO mouse models (Chambon et al. 2010, Dubois et al. 2014) (Table 1). However, this did not impede muscle-specific ARKO mice from displaying similar performances to their WT littermates in both the voluntary wheel running and the forced treadmill tests (Table 1). Overall, we conclude that it is unlikely that the mild loss of muscle function following androgen deficiency in male mice is the main driver of their low activity phenotype.
Motor skills
Preclinical data show that testosterone deficiency exerts detrimental effects on male CNS health (Khasnavis et al. 2013, Jayaraman et al. 2014, Atallah et al. 2017). In mice, prepubertal ORX led to a poor performance in the rotarod test, which the authors associated with a loss of midbrain dopaminergic neurons (Khasnavis et al. 2013). Conversely, also in the rotarod test, male testicular-feminized (Tfm) mice expressing a non-functional AR performed better than their WT littermates (Rizk et al. 2005). Although this might seem unexpected at first, the increased rotarod performance of Tfm mice could be due to their lower body weight, since these two variables share an inverse correlation (Mao et al. 2015). Confirming the latter results, male ARKO mice showed normal motor skills despite their loss in striatal dopamine concentrations, a fact that might be explained by compensatory adaptations during ontogeny (Jardí et al. 2018). In contrast to prepubertal castration, ORX adult rodents showed an altered performance in neither the rotarod test nor a nigrostriatal pathology (Kritzer et al. 2001, Frye et al. 2008b , Khasnavis et al. 2013). In the beam-walking test, which detects more subtle motor coordination deficits, one study showed that ORX mice performed worse than sham-operated controls (McDermott et al. 1994). However, the performance in the latter test is in part influenced by motivational factors (McDermott et al. 1994, Curzon et al. 2009, Deacon 2013).
Regarding peripheral nerve health, both ORX and treatment with the antiandrogen flutamide reduced the synthesis of myelin proteins in the sciatic nerve of male rodents (Magnaghi et al. 1999, 2004, Jayaraman et al. 2014). In male mice, the decrease in myelin sheath was associated with macrophage infiltration (Jayaraman et al. 2014). Also in male mice, however, nerve function as assessed by sciatic nerve conductance studies was not altered following either ORX or ARKO (Jardí et al. 2018).
Taken together, we conclude that the adverse actions of androgen deficiency on male neural health do not seem to translate into a deterioration of motor performance that would limit physical activity, at least in adult rodents. This does not preclude a role for androgens improving motor functional recovery in animal models of neural disease (Uchida et al. 2009, Yoo & Ko 2012, Ponnusamy et al. 2017).
Other peripheral actions
Skeletal muscle metabolism is critical to maintain fuel homeostasis during exercise. In both male rats and mice, androgen deficiency decreases muscle glycogen stores (Ramamani et al. 1999, Dubois et al. 2016), a suggested risk factor for fatigue development (Xirouchaki et al. 2016). Conversely, testosterone promotes glucose uptake in skeletal muscle, although it is not clear to what extent these effects are dependent on local AR actions (Sato et al. 2008, Ibebunjo et al. 2011, Dubois et al. 2016, Kelly et al. 2016). In male mice, deletion of myogenic AR did not affect either muscle glycogen stores (Ophoff et al. 2009b ) or glucose tolerance (Dubois et al. 2016), in contrast to that observed in both liver- (Lin et al. 2008) and adipose-tissue-specific ARKO mice (McInnes et al. 2012). The mechanisms by which testosterone acts on metabolic pathways have been reviewed elsewhere (Kelly & Jones 2013). Besides metabolism, long-term ORX might affect exercise capacity in rodents by attenuating cardiac contractility, as suggested by both in vivo echocardiographic and ex vivo studies (Sebag et al. 2011, Eleawa et al. 2013). These alterations in cardiac contractile function could arise from testosterone modulating Ca2+ handling in ventricular myocytes, as recently reviewed (Ayaz & Howlett 2015). Additionally, androgens might influence adaptation to exercise by affecting hemoglobin and hematocrit levels (Bhasin et al. 2012), bone (Vanderschueren et al. 2014) and/or lung function (Townsend et al. 2012). However, overall, the actions of androgens on these parameters or on muscle do not appear to play an indispensable role in maintaining the capacity of male rodents for endurance activities. In contrast to what is observed in the voluntary wheel running test, both ARKO mice and ORX rats perform as well as control animals when forced to run in a treadmill (Ophoff et al. 2009a , Petroianu et al. 2010). Therefore, we conclude that androgens must act through other mechanisms beyond muscle to stimulate physical activity. Nevertheless, it is still possible that the increase in activity over time potentiates the peripheral actions of androgens so that they acquire a greater importance in sustaining activity (Fig. 2).
Central actions
In the next section, we will review the available findings supporting the notion that testosterone is mainly acting centrally to boost physical activity in male rodents.
Central-nervous-system-specific KOs
In the mouse brain, AR and ERα display a moderate-to-high regionalized expression profile, both showing a high expression in the hypothalamus and brainstem, whilst AR levels are also abundant in the hippocampus and cerebral cortex (Gofflot et al. 2007, Lein et al. 2007, Mahfouz et al. 2016). Four different whole brain ARKO mouse models have been generated so far, allowing to explore the impact of AR signaling in the nervous tissue without interfering with its peripheral functions (Raskin et al. 2009, Juntti et al. 2010, Yu et al. 2013, Davey et al. 2017). An overview of the main features of the different neuronal ARKO models is presented in Table 1. All four models show an intact peripheral masculinization but unfortunately have an impaired negative feedback of the HPG axis in males, resulting in higher serum levels of testosterone and increased seminal vesicle weights (Raskin et al. 2009, Juntti et al. 2010, Yu et al. 2013, Davey et al. 2017). It is nevertheless clear that the neuronal AR is required for the full expression of both male-typical sexual and aggressive behaviors (Raskin et al. 2009, Juntti et al. 2010, Marie-Luce et al. 2013). Regarding physical activity, one study reported an increase in SPA in Nestin-Cre neuronal ARKO mice, although levels were restored to control values following normalization of testosterone levels, pointing to an implication of the altered HPG axis in the observed effects (Raskin et al. 2009). Contrarily, both SPA and wheel running were reduced by 60% in CamKII-Cre neuronal ARKO mice, which retains only non-DNA-binding-dependent AR actions (Davey et al. 2017). The latter results support the interpretation that AR exerts a predominant role in the regulation of male physical activity by testosterone. Nonetheless, the above-mentioned studies did not include littermates carrying the Cre transgene as controls. This is particularly important since Cre-driven transgenes might exert independent effects on behavioral outcomes (Harno et al. 2013, Chen et al. 2016). There is also the possibility that brain AR deletion at early stages may have indirectly interfered with the organizational actions of testosterone by altering the HPG axis. To circumvent this caveat, we recently generated a tamoxifen-inducible neuronal ARKO mouse model by using the Thy1-CreER, which drives gene deletion in extrahypothalamic regions of the brain (Jardí et al. 2017a ). In contrast to CamKII-Cre neuronal ARKO mice, deletion of AR in the CNS of pubertal male Thy1-CreER neuronal ARKO mice did not affect wheel running (Jardí et al. 2017b ). The differences between studies could be attributed to the residual expression of hypothalamic AR in our mice (Jardí et al. 2017a ). All in all, the study of neuronal ARKO mice has not provided clear answers so far regarding the role of brain androgenic signaling in male physical activity. To the best of our knowledge, only one study reported a whole-brain ERαKO in male mice, but the authors did not assess physical activity in these animals (Xu et al. 2011).
Hypothalamic circuitries
The hypothalamus is a key brain region that integrates nutritional, hormonal and neural information to orchestrate adaptive physiological responses aimed to maintain the whole-body energy balance (Lenard & Berthoud 2008, Schneeberger et al. 2014). Technical advances in viral-vector-mediated gene manipulation as well as in opto- and chemogenetics have allowed the identification of hypothalamic circuitries implicated in the control of energy homeostasis by sex steroids. In particular, hypothalamic ERα signaling in female rodents has been shown to exert a holistic control of their overall energy balance, influencing food intake as well as energy expenditure (Xu et al. 2017 for review on this topic). Silencing of ERα in the ventromedial nucleus of the hypothalamus (VMH) by small hairpin RNA knockdown led to obesity in both female rats and mice, an effect associated with a decline in physical activity and thermogenesis (Musatov et al. 2007). Remarkably, the effects of estrogens on these two parameters of energy expenditure seem to be driven by independent ERα modules of the VMH (Xu et al. 2011, Correa et al. 2015). In male mice, global deletion of either AR or ERα results in a decreased energy expenditure that might predispose them to increased adiposity (Heine et al. 2000, Fan et al. 2005). It is not clear whether androgens, similar to estrogens in females, regulate male energy expenditure by acting on AR and/or ER-positive hypothalamic neurons. Hypothalamic testosterone implants restored wheel running activity in ORX mice (Model et al. 2015). However, since testosterone can diffuse to adjacent brain areas, these might also be implicated in the observed effects. According to unpublished data of Ogawa et al. (Sano et al. 2013), site-specific ERα knockdown in the VMH completely abolished estrogen-induced facilitation of wheel running in male mice, similar to that described in females (Fig. 3). Notwithstanding the aforementioned limitation of treating male mice with estrogens, these results suggest that the VMH forms part of the neural circuitries mediating testosterone-induced wheel running in male mice and highlights the implication of local aromatization in this brain region (Roselli et al. 1985). In contrast to females, however, the regulation of male physical activity by ERα expression in the VMH may involve a distinct module than the ventrolateral region since stimulation of this neuronal cluster did not increase SPA in male mice (Correa et al. 2015).
Schematic overview of the proposed neuronal mechanisms by which testosterone increases male physical activity (see text). A, androgens; DA, dopamine; E2, estradiol; MeA, medial amygdala; NAcc, nucleus accumbens; SPA, spontaneous physical activity; VMH, ventromedial hypothalamus.
Citation: Journal of Endocrinology 238, 1; 10.1530/JOE-18-0125
Medial amygdala
The neuronal basis regulating physical activity likely implies other neuronal circuitries beyond the hypothalamic energy balance systems, such as brain regions involved in emotional and motivational processing (Fig. 3). MeA neurons projecting to the hypothalamus play a central role in generating emotional responses to chemosensory signals (Keshavarzi et al. 2014, Takahashi 2014). Compared to other amygdalar nuclei, the expression of sex steroid receptors is enriched in the MeA (Gofflot et al. 2007). In addition, a high proportion of MeA neurons are aromatase positive, and therefore this brain region might be exposed to relatively high levels of E2 (Wu et al. 2009, Unger et al. 2015). In male mice, ERα signaling in the MeA is required for the pubertal organizational actions of testosterone on social and sexual behaviors (Sano et al. 2016). In addition, Xu et al. provided compelling evidence that ERα signaling in the MeA stimulated SPA in males and secondarily prevented body weight gain in the setting of high-fat diet (Xu et al. 2015). Particularly, deletion of ERα in the MeA of adult male mice resulted in a low activity phenotype, whilst the other components of energy expenditure, namely thermogenesis and resting metabolic rate, remained unaltered (Xu et al. 2015). Together with the VMH (unpublished data of Ogawa et al.), the MeA is the only brain region identified so far that could be mediating the stimulatory effects of testosterone on male physical activity (Fig. 3). In addition, these findings also imply a role for brain aromatization in testosterone regulation of activity, similar to that described for male-typical aggressive behavior (Unger et al. 2015).
Rewarding and motivational pathways
Besides the metabolic-regulatory circuitries of the hypothalamus and the MeA, brain regions controlling rewarding and motivational functions might also be involved in the regulation of physical activity (Garland et al. 2011) (Fig. 3). Endurance running generates a rewarding aftereffect in both humans and rodents (Boecker et al. 2008, Fernandes et al. 2015). The traditional view attributes reward functions to dopamine signaling in the nucleus accumbens (Wise 2004, Palmiter 2008). However, recent findings show that other dopamine terminal fields are also components of the brain reward system, including the dorsal striatum, amygdala and frontal cortex (Wise 2004, Palmiter 2008). Following the withdrawal of running wheels, mice selectively bred for high wheel running showed a more pronounced increase in striatal activation compared to controls (Rhodes et al. 2003). Also in male mice, the relative volume of the striatum predicted the levels of subsequent wheel running activity on in vivo MRI (Cahill et al. 2015).
Testosterone might act directly on male midbrain dopamine neurons by binding to their AR and/or ERs receptors but also indirectly through the afferent neurons projecting to them (Kritzer & Creutz 2008, Aubele & Kritzer 2012, Purves-Tyson et al. 2012, Locklear et al. 2017). However, it is not clear if the main contribution of androgens to the overall dopaminergic tone is at the level of dopamine synthesis, transport, release, metabolization and/or signaling (Aubele & Kritzer 2011, Khasnavis et al. 2013, Purves-Tyson et al. 2014, Wang et al. 2016). Testosterone restoration of basal extracellular dopamine levels in the mesocortical dopaminergic pathway implies AR-dependent mechanisms, though this might not apply to the rest of dopaminergic systems (Kritzer 2003, Aubele & Kritzer 2011, Locklear et al. 2017). Amphetamine injections induce a hyperlocomotion response in rodents, which has been ascribed to the drug’s dopamine-releasing actions (Salahpour et al. 2008). Notably, ORX increases, whilst testosterone restores the responsiveness to the locomotor effects of amphetamine in both rats and mice (Purves-Tyson et al. 2015, Jardí et al. 2018). Overall, there is enough evidence to conclude that androgens influence the dopamine function in vivo. In addition, we recently showed that systemic administration of the dopamine receptor 2 antagonist L-741,626 reduced by almost 50% the distance ran by ORX+testosterone-treated mice, whilst it had no effect on castrated mice receiving vehicle (Jardí et al. 2018). Further research is required to narrow down the relevant androgen-responsive components of the dopamine system that might be mediating the effects of testosterone on physical activity (Jardí et al. 2018).
To summarize, the control of activity and metabolism are regulated in a coupled manner by estrogen-responsive hypothalamic pathways in females. The implication of similar mechanisms in males remains to be proven. Extrahypothalamic inputs coming from brain regions related to emotional and rewarding processes might modulate the homeostatic control exerted by the hypothalamus. Testosterone might act on the MeA to increase physical activity in males following aromatization into E2 (Fig. 3). The dopamine system is another potential candidate mediating the effects of testosterone on activity, but further research is required to determine the mechanisms implicated (Fig. 3).
Human evidence
After reviewing the available data, it is clear that androgens regulate physical activity in male rodents, but in men, it remains a question. The etiology of physical activity in humans appears to be much more complex than in animals, being influenced by the social and cultural environments (Bauman et al. 2012). Heritability studies in twins and families suggest that our drive to engage in exercise is also determined by intrinsic biological processes (Simonen et al. 2002, Bouchard & Rankinen 2006, Stubbe et al. 2006). Nevertheless, the genetics of physical activity remain currently unknown, with only limited genotype–phenotype associations found so far. A candidate gene study in postmenopausal women showed an association between aromatase gene polymorphisms and time spent in physical activity (Salmen et al. 2003). This association for the aromatase gene was replicated in a GWAS for leisure-time exercise behavior but only in one out of the two populations of adult men included (De Moor et al. 2009). Polymorphisms in the AR or ERα gene were not associated with the amount of physical activity per week in either young or middle-aged men or women (Lorentzon et al. 1999, Salmén et al. 2000, Okura et al. 2003, Walsh et al. 2005). Aging is associated with a decline in physical activity in both genders (Jones et al. 2011) as well as with a gradual reduction in serum testosterone, especially its free fraction, in men (Wu et al. 2008). So far, studies of the association between late-onset hypogonadism and the age-related decline in activity have produced inconsistent results. A small cross-sectional study in community-dwelling men showed an association between low testosterone levels and a decrease in pedometer-recorded step counts (Cobo et al. 2017). Nevertheless, other studies using larger cohorts and activity-based questionnaires yielded conflicting observations (Amin et al. 2000, Beutel et al. 2005, Tajar et al. 2012). Similarly, there is no clear evidence from interventional studies that supports a causal effect of sex steroids on physical activity. Several double-blind, placebo-controlled clinical trials have examined the effects of testosterone replacement on both muscle as well as functional mobility in large samples of community-dwelling men with late-onset hypogonadism (Emmelot-Vonk et al. 2008, Srinivas-Shankar et al. 2010, Snyder et al. 2016, Storer et al. 2016). Overall, testosterone supplementation in hypogonadal men results in modest gains in muscle mass and strength, but the benefits on their overall physical function are inconsistent (Emmelot-Vonk et al. 2008, Srinivas-Shankar et al. 2010, Snyder et al. 2016, Storer et al. 2016). In addition, testosterone therapy in men is hampered by its potential adverse effects on the cardiovascular system and the prostate (Basaria et al. 2010). Therefore, the clinical meaningfulness of testosterone replacement in healthy elderly men with late-onset hypogonadism remains debated. Remarkably, treatment with testosterone for 6 months did not affect the self-reported levels of activity in a cohort of 274 community-dwelling frail elderly men with low levels of testosterone (Srinivas-Shankar et al. 2010). Smaller studies in similar patient populations also showed a lack of testosterone effect on activity scores (Kenny et al. 2001, 2010). Contrarily, a positive association between serum testosterone levels and activity scores was found in a cohort of elderly men receiving a combined treatment of testosterone and recombinant human growth hormone (Sattler et al. 2011). The authors hypothesized that this association might reflect a central action of testosterone on the individual’s mood (Sattler et al. 2011). To date, there is no real evidence supporting the former premise, with the effects of testosterone on vitality and mood parameters in elderly men being unclear (Snyder et al. 1999, Srinivas-Shankar et al. 2010, Snyder et al. 2016). More importantly, all previous studies were based on activity questionnaires, which show limited reliability and accuracy (Strath et al. 2013). Thus, caution should be taken when interpreting the data. In conclusion, it is premature to support or reject convincingly the implication of androgens in physical activity in men.
Conclusion
A wide body of preclinical evidence makes a strong case for androgens as potential candidates contributing to the biological basis regulating male physical activity. Nonetheless, reliable human studies are lacking and our present knowledge of the implicated processes is very limited. This is explained in part by the fact that the interest in the biological factors determining activity levels is relatively recent. Measurement of physical activity in humans most often relies on self-reported instruments due to their practicality and low cost. Nevertheless, these measures have severe limitations in terms of reliability and validity, being subjected to several biases arising from both the respondent and/or the investigator. In addition, the terminology as well as the analysis and interpretation of the data differ markedly between studies. The current challenge is to validate affordable objective measures that can be implemented in large cohorts. In comparison, the mechanisms underlying the anabolic effects of androgens on muscle are better defined. Androgens increase muscle power in both young and elderly men, particularly when combined with exercise. In male mice, myogenic AR signaling drives in part the mild effects of testosterone stimulating appendicular muscle strength. Also in male mice, the modest and slow gains in muscle mass induced by androgens contrast with their dramatic and fast actions increasing wheel running. Taken together, preclinical data do not support the notion that androgens stimulate physical activity primarily by favoring muscle function. On the other hand, the central nature of testosterone effects on activity is supported by the recent identification of sex-steroid-responsive neuronal circuitries modulating physical activity. Although data in males are still scarce, the former findings indicate that brain aromatization is required for the full action of testosterone on activity. In conclusion, the CNS appears as the primary target of action for androgens stimulating physical activity, whilst the multiple beneficial effects of testosterone on muscle and other peripheral organs do not drive the increase in activity but might contribute to the adaptation to exercise. A better understanding of the underlying biological mechanisms could have implications for the design of more effective interventions to promote physical activity.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this review.
Funding
This work was supported by the Research Foundation Flanders (FWO; grant G0D2217N) and the KU Leuven Research Council (grant GOA/15/017). F J is supported by a post-doctoral grant from FWO.
References
Allen DL, Harrison BC, Maass A, Bell ML, Byrnes WC & Leinwand LA 2001 Cardiac and skeletal muscle adaptations to voluntary wheel running in the mouse. Journal of Applied Physiology 90 1900–1908. (https://doi.org/10.1530/eje.0.1330361)
Almeida M, Laurent MR, Dubois V, Claessens F, O’Brien CA, Bouillon R, Vanderschueren D & Manolagas SC. 2017 Estrogens and androgens in skeletal physiology and pathophysiology. Physiological Reviews 97 135–187. (https://doi.org/10.1152/physrev.00033.2015)
American Physiological Association 2007 Exercise protocols using rats and mice. Resource Book for the Design of Animal Exercise Protocols, ch 3, pp 53–56. (https://doi.org/10.2460/ajvr.68.6.583)
Amin S, Zhang Y, Sawin CT, Evans SR, Hannan MT, Kiel DP, Wilson PWF & Felson DT 2000 Association of hypogonadism and estradiol levels with bone mineral density in elderly men from the Framingham study. Annals of Internal Medicine 133 951–963. (https://doi.org/200012190-00010 [pii])
Amory JK, Wang C, Swerdloff RS, Anawalt BD, Matsumoto AM, Bremner WJ, Walker SE, Haberer LJ & Clark RV 2007 The effect of 5α-reductase inhibition with dutasteride and finasteride on semen parameters and serum hormones in healthy men. Journal of Clinical Endocrinology and Metabolism 92 1659–1665. (https://doi.org/10.1210/jc.2006-2203)
Arnold AP 2009 The organizational-activational hypothesis as the foundation for a unified theory of sexual differentiation of all mammalian tissues. Hormones and Behavior 55 570–578. (https://doi.org/10.1016/j.yhbeh.2009.03.011)
Atallah A, Mhaouty-Kodja S & Grange-Messent V 2017 Chronic depletion of gonadal testosterone leads to blood-brain barrier dysfunction and inflammation in male mice. Journal of Cerebral Blood Flow and Metabolism 37 3161–3175. (https://doi.org/10.1177/0271678X16683961)
Aubele T & Kritzer MF 2011 Gonadectomy and hormone replacement affects in vivo basal extracellular dopamine levels in the prefrontal cortex but not motor cortex of adult male rats. Cerebral Cortex 21 222–232. (https://doi.org/10.1093/cercor/bhq083)
Aubele T & Kritzer MF 2012 Androgen influence on prefrontal dopamine systems in adult male rats: localization of cognate intracellular receptors in medial prefrontal projections to the ventral tegmental area and effects of gonadectomy and hormone replacement on glutamate-stimulated. Cerebral Cortex 22 1799–1812. (https://doi.org/10.1093/cercor/bhr258)
Axell A-M, MacLean HE, Plant DR, Harcourt LJ, Davis JA, Jimenez M, Handelsman DJ, Lynch GS & Zajac JD 2006 Continuous testosterone administration prevents skeletal muscle atrophy and enhances resistance to fatigue in orchidectomized male mice. American Journal of Physiology: Endocrinology and Metabolism 291 E506–E516. (https://doi.org/10.1152/ajpendo.00058.2006)
Ayaz O & Howlett SE 2015 Testosterone modulates cardiac contraction and calcium homeostasis: cellular and molecular mechanisms. Biology of Sex Differences 6 9. (https://doi.org/10.1186/s13293-015-0027-9)
Barone R, Pitruzzella A, Marino Gammazza A, Rappa F, Salerno M, Barone F, Sangiorgi C, D’Amico D, Locorotondo N, Di Gaudio F, et al.2017 Nandrolone decanoate interferes with testosterone biosynthesis altering blood–testis barrier components. Journal of Cellular and Molecular Medicine 21 1636–1647. (https://doi.org/10.1111/jcmm.13092)
Basaria S, Coviello AD, Travison TG, Storer TW, Farwell WR, Jette AM, Eder R, Tennstedt S, Ulloor J, Zhang A, et al.2010 Adverse events associated with testosterone administration. New England Journal of Medicine 363 109–122. (https://doi.org/10.1056/NEJMoa1000485)
Bauman AE, Reis RS, Sallis JF, Wells JC, Loos RJF & Martin BW 2012 Correlates of physical activity: why are some people physically active and others not? Lancet 380 258–271. (https://doi.org/10.1016/S0140-6736(12)60735-1)
Berchtold NC, Kesslak JP, Pike CJ, Adlard PA & Cotman CW 2001 Estrogen and exercise interact to regulate brain-derived neurotrophic factor mRNA and protein expression in the hippocampus. European Journal of Neuroscience 14 1992–2002. (https://doi.org/10.1046/j.0953-816x.2001.01825.x)
Bergink EW, Janssen PSL, Turpun EW & Van Der Vies J 1985 Comparison of the receptor binding properties of nandrolone and testosterone under in vitro and in vivo conditions. Journal of Steroid Biochemistry 22 831–836. (https://doi.org/10.1016/0022-4731(85)90293-6)
Beutel ME, Wiltink J, Hauck EW, Auch D, Behre HM, Brähler E & Weidner W 2005 Correlations between hormones, physical, and affective parameters in aging urologic outpatients. European Urology 47 749–755. (https://doi.org/10.1016/j.eururo.2005.02.015)
Bhasin S, Travison TG, Storer TW, Lakshman K, Kaushik M, Mazer NA, Ngyuen AH, Davda MN, Jara H, Aakil A, et al.2012 Effect of testosterone supplementation with and without a dual 5α-reductase inhibitor on fat-free mass in men with suppressed testosterone production: a randomized controlled trial. JAMA 307 931–939. (https://doi.org/10.1001/jama.2012.227)
Bjørnebekk A, Walhovd KB, Jørstad ML, Due-Tønnessen P, Hullstein IR & Fjell AM 2017 Structural brain imaging of long-term anabolic-androgenic steroid users and nonusing weightlifters. Biological Psychiatry 82 294–302. (https://doi.org/10.1016/j.biopsych.2016.06.017)
Blattner MS & Mahoney MM 2012 Circadian parameters are altered in two strains of mice with transgenic modifications of estrogen receptor subtype 1. Genes, Brain and Behavior 11 828–836. (https://doi.org/10.1111/j.1601-183X.2012.00831.x)
Blattner MS & Mahoney MM 2015 Changes in estrogen receptor signaling alters the timekeeping system in male mice. Behavioural Brain Research 294 43–49. (https://doi.org/10.1016/j.bbr.2015.07.060)
Blizard DA 1983 Sex differences in running-wheel behaviour in the rat: the inductive and activational effects of gonadal hormones. Animal Behaviour 31 378–384. (https://doi.org/10.1016/S0003-3472(83)80056-6)
Boecker H, Sprenger T, Spilker ME, Henriksen G, Koppenhoefer M, Wagner KJ, Valet M, Berthele A & Tolle TR 2008 The runner’s high: opioidergic mechanisms in the human brain. Cerebral Cortex 18 2523–2531. (https://doi.org/10.1093/cercor/bhn013)
Borst SE, Conover CF, Carter CS, Gregory CM, Marzetti E, Leeuwenburgh C, Vandenborne K & Wronski TJ 2007 Anabolic effects of testosterone are preserved during inhibition of 5-reductase. AJP: Endocrinology and Metabolism 293 E507–E514. (https://doi.org/10.1152/ajpendo.00130.2007)
Bouchard C & Rankinen T 2006 Are people physically inactive because of their genes? President’s Council on Physical Fitness and Sports Research Digest 7 1–8.
Bowen RS, Ferguson DP & Lightfoot JT 2011 Effects of aromatase inhibition on the physical activity levels of male mice. Journal of Steroids and Hormonal Science 1 1–7. (https://doi.org/10.4172/2157-7536.S1-001)
Bowen RS, Knab AM, Hamilton AT, McCall JR, Moore-Harrison TL & Lightfoot JT 2012 Effects of supraphysiological doses of sex steroids on wheel running activity in mice. Journal of Steroids and Hormonal Science 3 110. (https://doi.org/10.4172/2157-7536.1000110)
Brock O, Baum MJ & Bakker J 2011 The development of female sexual behavior requires prepubertal estradiol. Journal of Neuroscience 31 5574–5578. (https://doi.org/10.1523/JNEUROSCI.0209-11.2011)
Brockman R, Bunick D & Mahoney MM 2011 Estradiol deficiency during development modulates the expression of circadian and daily rhythms in male and female aromatase knockout mice. Hormones and Behavior 60 439–447. (https://doi.org/10.1016/j.yhbeh.2011.07.011)
Broida J & Svare B 1984 Sex differences in the activity of mice: modulation by postnatal gonadal hormones. Hormones and Behavior 18 65–78. (https://doi.org/10.1016/0018-506X(84)90051-5)
Butler MP, Karatsoreos IN, Le Sauter J & Silver R 2012 Dose-dependent effects of androgens on the circadian timing system and its response to light. Endocrinology 153 2344–2352. (https://doi.org/10.1210/en.2011-1842)
Cahill LS, Steadman PE, Jones CE, Laliberté CL, Dazai J, Lerch JP, Stefanovic B & Sled JG 2015 MRI-detectable changes in mouse brain structure induced by voluntary exercise. NeuroImage 113 175–183. (https://doi.org/10.1016/j.neuroimage.2015.03.036)
Callewaert F, Venken K, Ophoff J, De Gendt K, Torcasio A, van Lenthe GH, Van Oosterwyck H, Boonen S, Bouillon R, Verhoeven G, et al.2009 Differential regulation of bone and body composition in male mice with combined inactivation of androgen and estrogen receptor-alpha. FASEB Journal 23 232–240. (https://doi.org/10.1096/fj.08-113456)
Callewaert F, Boonen S & Vanderschueren D 2010 Sex steroids and the male skeleton: a tale of two hormones. Trends in Endocrinology and Metabolism 21 89–95. (https://doi.org/10.1016/j.tem.2009.09.002)
Caspersen CJ, Powell KE & Christenson GM 1985 Physical activity, exercise, and physical fitness: definitions and distinctions for health-related research. Public Health Reports 100 126–131. (https://doi.org/10.2307/20056429)
Chabert C, Bottelin P, Pison C & Dubouchaud H 2016 A low-cost system to easily measure spontaneous physical activity in rodents. Journal of Applied Physiology 120 1097–1103. (https://doi.org/10.1152/japplphysiol.00888.2015)
Chambon C, Duteil D, Vignaud A, Ferry A, Messaddeq N, Malivindi R, Kato S, Chambon P & Metzger D 2010 Myocytic androgen receptor controls the strength but not the mass of limb muscles. PNAS 107 14327–14332. (https://doi.org/10.1073/pnas.1009536107)
Chen CV, Brummet JL, Jordan CL & Breedlove SM 2016 Down, but not out: partial elimination of androgen receptors in the male mouse brain does not affect androgenic regulation of anxiety or HPA activity. Endocrinology 157 764–773. (https://doi.org/10.1210/en.2015-1417)
Chisamore MJ, Gentile MA, Dillon GM, Baran M, Gambone C, Riley S, Schmidt A, Flores O, Wilkinson H & Alves SE 2016 A novel selective androgen receptor modulator (SARM) MK-4541 exerts anti-androgenic activity in the prostate cancer xenograft R–3327G and anabolic activity on skeletal muscle mass & function in castrated mice. Journal of Steroid Biochemistry and Molecular Biology 163 88–97. (https://doi.org/10.1016/j.jsbmb.2016.04.007)
Cobo G, Gallar P, Di Gioia C, García Lacalle C, Camacho R, Rodriguez I, Ortega O, Mon C, Vigil A, Lindholm B, et al. 2017 Hypogonadism associated with muscle atrophy, physical inactivity and ESA hyporesponsiveness in men undergoing haemodialysis. Nefrologia : Publicacion Oficial de La Sociedad Espanola Nefrologia 37 54–60. (https://doi.org/10.1016/j.nefro.2016.04.009)
Cooke PS, Nanjappa MK, Ko C, Prins GS & Hess RA 2017 Estrogens in male physiology. Physiological Reviews 97 995–1043. (https://doi.org/10.1152/physrev.00018.2016)
Correa SM, Newstrom DW, Warne JP, Flandin P, Cheung CC, Lin-Moore AT, Pierce AA, Xu AW, Rubenstein JL & Ingraham HA 2015 An estrogen-responsive module in the ventromedial hypothalamus selectively drives sex-specific activity in females. Cell Reports 10 62–74. (https://doi.org/10.1016/j.celrep.2014.12.011)
Cozzoli A, Capogrosso RF, Sblendorio VT, Dinardo MM, Jagerschmidt C, Namour F, Camerino GM & De Luca A 2013 GLPG0492, a novel selective androgen receptor modulator, improves muscle performance in the exercised-mdx mouse model of muscular dystrophy. Pharmacological Research 72 9–24. (https://doi.org/10.1016/j.phrs.2013.03.003)
Curzon P, Zhang M, Radek RJ & Fox GB 2009 The behavioral assessment of sensorimotor processes in the mouse: acoustic startle, sensory gating, locomotor activity, rotarod, and beam walking. In Methods of Behavior Analysis in Neuroscience, pp 1–28. Boca Raton, FL, USA: Taylor & Francis.
Daan S, Damassa D, Pittendrigh CS & Smith ER 1975 An effect of castration and testosterone replacement on a circadian pacemaker in mice (Mus musculus). PNAS 72 3744–3747. (https://doi.org/10.1073/pnas.72.9.3744)
Dark J & Zucker I 1984 Gonadal and photoperiodic control of seasonal body weight changes in male voles. American Journal of Physiology 247 R84–R88.
Davey RA, Clarke MV, Russell PK, Rana K, Seto J, Roeszler KN, How JMY, Chia LY, North K & Zajac JD 2017 Androgen action via the androgen receptor in neurons within the brain positively regulates muscle mass in male mice. Endocrinology 158 3684–3695. (https://doi.org/10.1210/en.2017-00470)
Deacon RMJ 2013 Measuring motor coordination in mice. Journal of Visualized Experiments 75 e2609. (https://doi.org/10.3791/2609)
Delbeck M, Golz S, Vonk R, Janssen W, Hucho T, Isensee J, Schäfer S & Otto C 2011 Impaired left-ventricular cardiac function in male GPR30-deficient mice. Molecular Medicine Reports 4 37–40. (https://doi.org/10.3892/mmr.2010.402)
De Lorme KC & Sisk CL 2016 The organizational effects of pubertal testosterone on sexual proficiency in adult male Syrian hamsters. Physiology and Behavior 165 273–277. (https://doi.org/10.1016/j.physbeh.2016.08.008)
De Moor MHM, Liu Y-J, Boomsma DI, Li J, Hamilton JJ, Hottenga J, Levy S, Liu X, Pei Y, Posthuma D, et al.2009 Genome-wide association study of exercise behavior in Dutch and American adults. Medicine and Science in Sports and Exercise 41 1887–1895. (https://doi.org/10.1249/MSS.0b013e3181a2f646)
De Naeyer H, Lamon S, Russell AP, Everaert I, De Spaey A, Vanheel B, Taes Y & Derave W 2014 Androgenic and estrogenic regulation of Atrogin-1, MuRF1 and myostatin expression in different muscle types of male mice. European Journal of Applied Physiology 114 751–761. (https://doi.org/10.1007/s00421-013-2800-y)
De Visser L, Van Den Bos R & Spruijt BM 2005 Automated home cage observations as a tool to measure the effects of wheel running on cage floor locomotion. Behavioural Brain Research 160 382–388. (https://doi.org/10.1016/j.bbr.2004.12.004)
De Vries GJ, Rissman EF, Simerly RB, Yang L-Y, Scordalakes EM, Auger CJ, Swain A, Lovell-Badge R, Burgoyne PS & Arnold AP 2002 A model system for study of sex chromosome effects on sexually dimorphic neural and behavioral traits. Journal of Neuroscience 22 9005–9014. (https://doi.org/22/20/9005)
Dubois V, Laurent M, Boonen S, Vanderschueren D & Claessens F 2012 Androgens and skeletal muscle: cellular and molecular action mechanisms underlying the anabolic actions. Cellular and Molecular Life Sciences 69 1651–1667. (https://doi.org/10.1007/s00018-011-0883-3)
Dubois V, Laurent MR, Sinnesael M, Cielen N, Helsen C, Clinckemalie L, Spans L, Gayan-Ramirez G, Deldicque L, Hespel P, et al.2014 A satellite cell-specific knockout of the androgen receptor reveals myostatin as a direct androgen target in skeletal muscle. FASEB Journal 28 2979–2994. (https://doi.org/10.1096/fj.14-249748)
Dubois V, Simitsidellis I, Laurent MR, Jardi F, Saunders PTK, Vanderschueren D & Claessens F 2015 Enobosarm (GTx-024) modulates adult skeletal muscle mass independently of the androgen receptor in the satellite cell lineage. Endocrinology 156 4522–4533. (https://doi.org/10.1210/en.2015-1479)
Dubois V, Laurent MR, Jardi F, Antonio L, Lemaire K, Goyvaerts L, Deldicque L, Carmeliet G, Decallonne B, Vanderschueren D, et al.2016 Androgen deficiency exacerbates high-fat diet-induced metabolic alterations in male mice. Endocrinology 157 648–665. (https://doi.org/10.1210/en.2015-1713)
Dworatzek E, Mahmoodzadeh S, Schubert C, Westphal C, Leber J, Kusch A, Kararigas G, Fliegner D, Moulin M, Ventura-Clapier R, et al. 2014 Sex differences in exercise-induced physiological myocardial hypertrophy are modulated by oestrogen receptor beta. Cardiovascular Research 102 418–428. (https://doi.org/10.1093/cvr/cvu065)
Eikelboom R & Mills R 1988 A microanalysis of wheel running in male and female rats. Physiology and Behavior 43 625–630. (https://doi.org/10.1016/0031-9384(88)90217-X)
Eleawa SM, Sakr HF, Hussein AM, Assiri AS, Bayoumy NMK & Alkhateeb M 2013 Effect of testosterone replacement therapy on cardiac performance and oxidative stress in orchidectomized rats. Acta Physiologica 209 136–147. (https://doi.org/10.1111/apha.12158)
Emmelot-Vonk MH, Verhaar HJJ, Nakhai Pour HR, Aleman A, Lock TMTW, Bosch JLHR, Grobbee DE & Van Der Schouw YT 2008 Effect of testosterone supplementation on functional mobility, cognition, and other parameters in older men: a randomized controlled trial. JAMA 299 39–52. (https://doi.org/10.1001/jama.2007.51)
Eriksson M, Rasmussen F & Tynelius P 2006 Genetic factors in physical activity and the equal environment assumption – the Swedish young male twins study. Behavior Genetics 36 238–247. (https://doi.org/10.1007/s10519-005-9018-7)
Fan WQ, Yanase T, Nomura M, Okabe T, Goto K, Sato T, Kawano H, Kato S & Nawata H 2005 Androgen receptor null male mice develop late-onset obesity caused by decreased energy expenditure and lipolytic activity but show normal insulin sensitivity with high adiponectin secretion. Diabetes 54 1000–1008. (https://doi.org/10.2337/diabetes.54.4.1000)
Fernandes MFA, Matthys D, Hryhorczuk C, Sharma S, Mogra S, Alquier T & Fulton S 2015 Leptin suppresses the rewarding effects of running via STAT3 signaling in dopamine neurons. Cell Metabolism 22 741–749. (https://doi.org/10.1016/j.cmet.2015.08.003)
Ferry A, Schuh M, Parlakian A, Mgrditchian T, Valnaud N, Joanne P, Butler-Browne G, Agbulut O & Metzger D 2014 Myofiber androgen receptor promotes maximal mechanical overload-induced muscle hypertrophy and fiber type transition in male mice. Endocrinology 155 4739–4748. (https://doi.org/10.1210/en.2014-1195)
Fisher CR, Graves KH, Parlow AF & Simpson ER 1998 Characterization of mice deficient in aromatase (ArKO) because of targeted disruption of the cyp19 gene. Medical Sciences 95 6965–6970. (https://doi.org/10.1073/pnas.95.12.6965)
Fragkaki AG, Angelis YS, Koupparis M, Tsantili-Kakoulidou A, Kokotos G & Georgakopoulos C 2009 Structural characteristics of anabolic androgenic steroids contributing to binding to the androgen receptor and to their anabolic and androgenic activities. Applied modifications in the steroidal structure. Steroids 74 172–197. (https://doi.org/10.1016/j.steroids.2008.10.016)
Frye CA, Van Keuren KR & Erskine MS 1996 Behavioral effects of 3α-androstanediol I: modulation of sexual receptivity and promotion of GABA-stimulated chloride flux. Behavioural Brain Research 79 109–118. (https://doi.org/10.1016/S0166-4328(96)00004-6)
Frye CA, Koonce CJ, Edinger KL, Osborne DM & Walf AA 2008a Androgens with activity at estrogen receptor beta have anxiolytic and cognitive-enhancing effects in male rats and mice. Hormones and Behavior 54 726–734. (https://doi.org/10.1016/j.yhbeh.2008.07.013)
Frye CA, Edinger K & Sumida K 2008b Androgen administration to aged male mice increases anti-anxiety behavior and enhances cognitive performance. Neuropsychopharmacology 33 1049–1061. (https://doi.org/10.1038/sj.npp.1301498)
Gans HM 1927 Studies on vigor. Effect of fractional castration on the voluntary activity of male albino rats. Endocrinology 11 145–148. (https://doi.org/10.1210/endo-11-2-145)
Garland T, Schutz H, Chappell MA, Keeney BK, Meek TH, Copes LE, Acosta W, Drenowatz C, Maciel RC, van Dijk G, et al.2011 The biological control of voluntary exercise, spontaneous physical activity and daily energy expenditure in relation to obesity: human and rodent perspectives. Journal of Experimental Biology 214 206–229. (https://doi.org/10.1242/jeb.048397)
Gentry RT & Wade GN 1976 Sex differences in sensitivity of food intake, body weight, and running- wheel activity to ovarian steroids in rats. Journal of Comparative and Physiological Psychology 90 747–754. (https://doi.org/10.1037/h0077246)
Georgieva KN & Boyadjiev NP 2004 Effects of nandrolone decanoate on V˙O2max, running economy, and endurance in rats. Medicine and Science in Sports and Exercise 36 1336–1341. (https://doi.org/10.1249/01.MSS.0000135781.42515.17)
Gofflot F, Chartoire N, Vasseur L, Heikkinen S, Dembele D, Le Merrer J & Auwerx J 2007 Systematic gene expression mapping clusters nuclear receptors according to their function in the brain. Cell 131 405–418. (https://doi.org/10.1016/j.cell.2007.09.012)
Guo W, Wong S, Li M, Liang W, Liesa M, Serra C, Jasuja R, Bartke A, Kirkland JL, Shirihai O, et al.2012 Testosterone plus low-intensity physical training in late life improves functional performance, skeletal muscle mitochondrial biogenesis, and mitochondrial quality control in male mice. PLoS ONE 7 e51180. (https://doi.org/10.1371/journal.pone.0051180)
Hallal PC, Andersen LB, Bull FC, Guthold R, Haskell W, Ekelund U, Alkandari JR, Bauman AE, Blair SN, Brownson RC, et al.2012 Global physical activity levels: surveillance progress, pitfalls, and prospects. Lancet 380 247–257. (https://doi.org/10.1016/S0140-6736(12)60646-1)
Hamrick MW, Ding K-H, Pennington C, Chao YJ, Wu Y-D, Howard B, Immel D, Borlongan C, McNeil PL, Bollag WB, et al.2006 Age-related loss of muscle mass and bone strength in mice is associated with a decline in physical activity and serum leptin. Bone 39 845–853. (https://doi.org/10.1016/j.bone.2006.04.011)
Harno E, Cottrell EC & White A 2013 Metabolic pitfalls of CNS cre-based technology. Cell Metabolism 18 21–28. (https://doi.org/10.1016/j.cmet.2013.05.019)
Heath GW, Parra DC, Sarmiento OL, Andersen LB, Owen N, Goenka S, Montes F, Brownson RC, Alkandari JR, Bauman AE, et al.2012 Evidence-based intervention in physical activity: lessons from around the world. Lancet 380 272–281. (https://doi.org/10.1016/S0140-6736(12)60816-2)
Heine PA, Taylor JA, Iwamoto GA, Lubahn DB & Cooke PS 2000 Increased adipose tissue in male and female estrogen receptor-alpha knockout mice. PNAS 97 12729–12734. (https://doi.org/10.1073/pnas.97.23.12729)
Hill RA, McInnes KJ, Gong ECH, Jones MEE, Simpson ER & Boon WC 2007 Estrogen deficient male mice develop compulsive behavior. Biological Psychiatry 61 359–366. (https://doi.org/10.1016/j.biopsych.2006.01.012)
Hourdé C, Jagerschmidt C, Clément-Lacroix P, Vignaud A, Ammann P, Butler-Browne GS & Ferry A 2009 Androgen replacement therapy improves function in male rat muscles independently of hypertrophy and activation of the Akt/mTOR pathway. Acta Physiologica 195 471–482. (https://doi.org/10.1111/j.1748-1716.2008.01902.x)
Ibebunjo C, Eash JK, Li C, Ma Q & Glass DJ 2011 Voluntary running, skeletal muscle gene expression, and signaling inversely regulated by orchidectomy and testosterone replacement. American Journal of Physiology-Endocrinology and Metabolism 300 E327–E340. (https://doi.org/10.1152/ajpendo.00402.2010\r10.1152/ajpendo.00402.2010)
Iwahana E, Karatsoreos I, Shibata S & Silver R 2008 Gonadectomy reveals sex differences in circadian rhythms and suprachiasmatic nucleus androgen receptors in mice. Hormones and Behavior 53 422–430. (https://doi.org/10.1016/j.yhbeh.2007.11.014)
Jakacka M, Ito M, Martinson F, Ishikawa T, Lee EJ & Jameson JL 2002 An estrogen receptor (ER)α deoxyribonucleic acid-binding domain knock-in mutation provides evidence for nonclassical ER pathway signaling in vivo. Molecular Endocrinology 16 2188–2201. (https://doi.org/10.1210/me.2001-0174)
Jänne M, Deol HK, Power SGA, Yee S & Hammond GL 1998 Human sex hormone-binding globulin gene expression in transgenic mice. Molecular Endocrinology 12 123–136. (https://doi.org/10.1210/mend.12.1.0050)
Jardí F, Laurent MR, Dubois V, Khalil R, Deboel L, Schollaert D, Van Den Bosch L, Decallonne B, Carmeliet G, Claessens F, et al.2017a A shortened tamoxifen induction scheme to induce CreER recombinase without side effects on the male mouse skeleton. Molecular and Cellular Endocrinology 452 57–63. (https://doi.org/10.1016/j.mce.2017.05.012)
Jardí F, Laurent MR, Khalil R, Kim N, Dubois V, Deboel L, Schollaert D, Carmeliet G, Decallonne B, Claessens F, et al. 2017b Deletion of androgen receptor in neurons accelerates the age-related loss of cortical thickness in male mice. Journal of Bone and Mineral Research 32 1152.
Jardí F, Laurent MR, Kim N, Khalil R, De Bundel D, Van Eeckhaut A, Van Helleputte L, Deboel L, Dubois V, Schollaert D, et al.2018 Testosterone boosts physical activity in male mice via dopaminergic pathways. Scientific Reports 8 957.
Jayaraman A, Lent-Schochet D & Pike CJ 2014 Diet-induced obesity and low testosterone increase neuroinflammation and impair neural function. Journal of Neuroinflammation 11 162. (https://doi.org/10.1186/s12974-014-0162-y)
Jechura TJ, Walsh JM & Lee TM 2000 Testicular hormones modulate circadian rhythms of the diurnal rodent, Octodon degus. Hormones and Behavior 38 243–249. (https://doi.org/10.1006/hbeh.2000.1624)
Jones ME, Thorburn AW, Britt KL, Hewitt KN, Wreford NG, Proietto J, Oz OK, Leury BJ, Robertson KM, Yao S, et al.2000 Aromatase-deficient (ArKO) mice have a phenotype of increased adiposity. PNAS 97 12735–12740. (https://doi.org/10.1073/pnas.97.23.12735)
Jones H, Millward P & Buraimo B 2011 Adult participation in sport. Analysis of the taking part survey. London, UK: Department for Digital, Culture, Media & Sport. (available at: https://www.gov.uk/government/publications/adult-participation-in-sport-analysis-of-the-taking-part-survey)
Juntti SA, Tollkuhn J, Wu MV, Fraser EJ, Soderborg T, Tan S, Honda SI, Harada N & Shah NM 2010 The androgen receptor governs the execution, but not programming, of male sexual and territorial behaviors. Neuron 66 260–272. (https://doi.org/10.1016/j.neuron.2010.03.024)
Karatsoreos IN, Wang A, Sasanian J & Silver R 2007 A role for androgens in regulating circadian behavior and the suprachiasmatic nucleus. Endocrinology 148 5487–5495. (https://doi.org/10.1210/en.2007-0775)
Kastenberger I & Schwarzer C 2014 GPER1 (GPR30) knockout mice display reduced anxiety and altered stress response in a sex and paradigm dependent manner. Hormones and Behavior 66 628–636. (https://doi.org/10.1016/j.yhbeh.2014.09.001)
Kelly DM & Jones TH 2013 Testosterone: a metabolic hormone in health and disease. Journal of Endocrinology 217 R25–R45. (https://doi.org/10.1530/JOE-12-0455)
Kelly DM, Akhtar S, Sellers DJ, Muraleedharan V, Channer KS & Jones TH 2016 Testosterone differentially regulates targets of lipid and glucose metabolism in liver, muscle and adipose tissues of the testicular feminised mouse. Endocrine 54 504–515. (https://doi.org/10.1007/s12020-016-1019-1)
Kenny AM, Prestwood KM, Gruman CA, Marcello KM & Raisz LG 2001 Effects of transdermal testosterone on bone and muscle in older men with low bioavailable testosterone levels. Journal of Gerontology 56A M266–M277. (https://doi.org/10.1111/j.1532-5415.2010.02865.x)
Kenny AM, Kleppinger A, Annis K, Rathier M, Browner B, Judge JO & McGee D 2010 Effects of transdermal testosterone on bone and muscle in older men with low bioavailable testosterone levels, low bone mass, and physical frailty. Journal of the American Geriatrics Society 58 1134–1143. (https://doi.org/10.1111/j.1532-5415.2010.02865.x)
Keshavarzi S, Sullivan RKP, Ianno DJ & Sah P 2014 Functional properties and projections of neurons in the medial amygdala. Journal of Neuroscience 34 8699–8715. (https://doi.org/10.1523/JNEUROSCI.1176-14.2014)
Khasnavis S, Ghosh A, Roy A & Pahan K 2013 Castration induces parkinson disease pathologies in young male mice via inducible nitric-oxide synthase. Journal of Biological Chemistry 288 20843–20855. (https://doi.org/10.1074/jbc.M112.443556)
Kim J, Min H, Oh S, Kim Y, Lee AH & Park T 2014 Joint identification of genetic variants for physical activity in Korean population. International Journal of Molecular Sciences 15 12407–12421. (https://doi.org/10.3390/ijms150712407)
Kim J-H, Han G-C, Seo J-Y, Park I, Park W, Jeong H-W, Lee SH, Bae S, Seong J, Yum M-K, et al.2016 Sex hormones establish a reserve pool of adult muscle stem cells. Nature Cell Biology 18 930–940. (https://doi.org/10.1038/ncb3401)
Kohl HW, Craig CL, Lambert EV, Inoue S, Alkandari JR, Leetongin G & Kahlmeier S 2012 The pandemic of physical inactivity: global action for public health. Lancet 380 294–305. (https://doi.org/10.1016/S0140-6736(12)60898-8)
Kritzer MF 2003 Long-term gonadectomy affects the density of tyrosine hydroxylase- but not dopamine-β-hydroxylase-, choline acetyltransferase- or serotonin-immunoreactive axons in the medial prefrontal cortices of adult male rats. Cerebral Cortex 13 282–296. (https://doi.org/10.1093/cercor/13.3.282)
Kritzer MF & Creutz LM 2008 Region and sex differences in constituent dopamine neurons and immunoreactivity for intracellular estrogen and androgen receptors in mesocortical projections in rats. Journal of Neuroscience 28 9525–9535. (https://doi.org/10.1523/JNEUROSCI.2637-08.2008)
Kritzer MF, McLaughlin PJ, Smirlis T & Robinson JK 2001 Gonadectomy impairs T-maze acquisition in adult male rats. Hormones and Behavior 39 167–174. (https://doi.org/10.1006/hbeh.2001.1645)
Kuljis DA, Loh DH, Truong D, Vosko AM, Ong ML, McClusky R, Arnold AP & Colwell CS 2013 Gonadal- and sex-chromosome-dependent sex differences in the circadian system. Endocrinology 154 1501–1512. (https://doi.org/10.1210/en.2012-1921)
Langer G, Bader B, Meoli L, Isensee J, Delbeck M, Noppinger PR & Otto C 2010 A critical review of fundamental controversies in the field of GPR30 research. Steroids 75 603–610. (https://doi.org/10.1016/j.steroids.2009.12.006)
Laurent MR, Helsen C, Antonio L, Schollaert D, Joniau S, Vos MJ, Decallonne B, Hammond GL, Vanderschueren D & Claessens F 2016a Effects of sex hormone-binding globulin (SHBG) on androgen bioactivity in vitro. Molecular and Cellular Endocrinology 437 280–291. (https://doi.org/10.1016/j.mce.2016.08.041)
Laurent MR, Hammond GL, Blokland M, Jardí F, Antonio L, Dubois V, Khalil R, Sterk SS, Gielen E, Decallonne B, et al.2016b Sex hormone-binding globulin regulation of androgen bioactivity in vivo: validation of the free hormone hypothesis. Scientific Reports 6 35539. (https://doi.org/10.1038/srep35539)
Lee IM, Shiroma EJ, Lobelo F, Puska P, Blair SN, Katzmarzyk PT, Alkandari JR, Andersen LB, Bauman AE, Brownson RC, et al.2012 Effect of physical inactivity on major non-communicable diseases worldwide: an analysis of burden of disease and life expectancy. Lancet 380 219–229. (https://doi.org/10.1016/S0140-6736(12)61031-9)
Lein ES, Hawrylycz MJ, Ao N, Ayres M, Bensinger A, Bernard A, Boe AF, Boguski MS, Brockway KS, Byrnes EJ, et al.2007 Genome-wide atlas of gene expression in the adult mouse brain. Nature 445 168–176. (https://doi.org/10.1038/nature05453)
Lenard NR & Berthoud H-R 2008 Central and peripheral regulation of food intake and physical activity: pathways and genes. Obesity 16 S11–S22. (https://doi.org/10.1038/oby.2008.511)
Lenz KM & McCarthy MM 2010 Organized for sex – steroid hormones and the developing hypothalamus. European Journal of Neuroscience 32 2096–2104. (https://doi.org/10.1111/j.1460-9568.2010.07511.x)
Lett BT, Grant VL & Koh MT 2001 Naloxone attenuates the conditioned place preference induced by wheel running in rats. Physiology and Behavior 72 355–358. (https://doi.org/10.1016/S0031-9384(00)00427-3)
Levine JA 1999 Role of nonexercise activity thermogenesis in resistance to fat gain in humans. Science 283 212–214. (https://doi.org/10.1126/science.283.5399.212)
Lightfoot JT 2008 Sex hormones’ regulation of rodent physical activity: a review. International Journal of Biological Sciences 4 126–132. (https://doi.org/10.7150/ijbs.4.126)
Lin H-Y, Yu I-C, Wang R-S, Chen Y-T, Liu N-C, Altuwaijri S, Hsu C-L, Ma W-L, Jokinen J, Sparks JD, et al.2008 Increased hepatic steatosis and insulin resistance in mice lacking hepatic androgen receptor. Hepatology 47 1924–1935. (https://doi.org/10.1002/hep.22252)
Locklear MN, Michealos M, Collins WF & Kritzer MF 2017 Gonadectomy but not biological sex affects burst-firing in dopamine neurons of the ventral tegmental area and in prefrontal cortical neurons projecting to the ventral tegmentum in adult rats. European Journal of Neuroscience 45 106–120. (https://doi.org/10.1111/ejn.13380)
Loos RJF, Rankinen T, Tremblay A, Pérusse L, Chagnon Y & Bouchard C 2005 Melanocortin-4 receptor gene and physical activity in the Québec Family Study. International Journal of Obesity 29 420–428. (https://doi.org/10.1038/sj.ijo.0802869)
Lorentzon M, Lorentzon R, Bäckström T & Nordström P 1999 Estrogen receptor gene polymorphism, but not estradiol levels, is related to bone density in healthy adolescent boys: a cross-sectional and longitudinal study. Journal of Clinical Endocrinology and Metabolism 84 4597–4601.
MacLean HE, Chiu WSM, Notini AJ, Axell A-M, Davey RA, McManus JF, Ma C, Plant DR, Lynch GS & Zajac JD 2008 Impaired skeletal muscle development and function in male, but not female, genomic androgen receptor knockout mice. FASEB Journal 22 2676–2689. (https://doi.org/10.1096/fj.08-105726)
Magnaghi V, Cavarretta I, Zucchi I, Susani L, Rupprecht R, Hermann B, Martini L & Melcangi RC 1999 Po gene expression is modulated by androgens in the sciatic nerve of adult male rats. Molecular Brain Research 70 36–44. (https://doi.org/10.1016/S0169-328X(99)00124-2)
Magnaghi V, Ballabio M, Gonzalez LC, Leonelli E, Motta M & Melcangi RC 2004 The synthesis of glycoprotein Po and peripheral myelin protein 22 in sciatic nerve of male rats is modulated by testosterone metabolites. Molecular Brain Research 126 67–73. (https://doi.org/10.1016/j.molbrainres.2004.03.009)
Mahfouz A, Lelieveldt BPF, Grefhorst A, van Weert LTCM, Mol IM, Sips HCM, van den Heuvel JK, Datson NA, Visser JA, Reinders MJT, et al.2016 Genome-wide coexpression of steroid receptors in the mouse brain: identifying signaling pathways and functionally coordinated regions. PNAS 113 2738–2743. (https://doi.org/10.1073/pnas.1520376113)
Mao J-H, Langley SA, Huang Y, Hang M, Bouchard KE, Celniker SE, Brown JB, Jansson JK, Karpen GH & Snijders AM 2015 Identification of genetic factors that modify motor performance and body weight using Collaborative Cross mice. Scientific Reports 5 16247. (https://doi.org/10.1038/srep16247)
Marie-Luce C, Raskin K, Bolborea M, Monin M, Picot M & Mhaouty-Kodja S 2013 Effects of neural androgen receptor disruption on aggressive behavior, arginine vasopressin and galanin systems in the bed nucleus of stria terminalis and lateral septum. General and Comparative Endocrinology 188 218–225. (https://doi.org/10.1016/j.ygcen.2013.03.031)
McCarthy MM, Schlenker EH & Pfaff DW 1993 Enduring consequences of neonatal treatment with antisense oligodeoxynucleotides to estrogen receptor messenger ribonucleic acid on sexual differentiation of rat brain. Endocrinology 133 433–439. (https://doi.org/10.1210/endo.133.2.8344188)
McDermott JL, Kreutzberg JD, Liu B & Dluzen DE 1994 Effects of estrogen treatment on sensorimotor task performance and brain dopamine concentrations in gonadectomized male and female CD-1 mice. Hormones and Behavior 28 16–28. (https://doi.org/10.1006/hbeh.1994.1002)
McGinnis MY, Lumia AR, Tetel MJ, Molenda-Figueira HA & Possidente B 2007 Effects of anabolic androgenic steroids on the development and expression of running wheel activity and circadian rhythms in male rats. Physiology and Behavior 92 1010–1018. (https://doi.org/10.1016/j.physbeh.2007.07.010)
McInnes KJ, Smith LB, Hunger NI, Saunders PTK, Andrew R & Walker BR 2012 Deletion of the androgen receptor in adipose tissue in male mice elevates retinol binding protein 4 and reveals independent effects on visceral fat mass and on glucose homeostasis. Diabetes 61 1072–1081. (https://doi.org/10.2337/db11-1136)
McNamara KM, Harwood DT, Simanainen U, Walters KA, Jimenez M & Handelsman DJ 2010 Measurement of sex steroids in murine blood and reproductive tissues by liquid chromatography-tandem mass spectrometry. Journal of Steroid Biochemistry and Molecular Biology 121 611–618. (https://doi.org/10.1016/j.jsbmb.2010.02.001)
Mhillaj E, Morgese MG, Tucci P, Bove M, Schiavone S & Trabace L 2015 Effects of anabolic-androgens on brain reward function. Frontiers in Neuroscience 9 295. (https://doi.org/10.3389/fnins.2015.00295)
Model Z, Butler MP, LeSauter J & Silver R 2015 Suprachiasmatic nucleus as the site of androgen action on circadian rhythms. Hormones and Behavior 73 1–7. (https://doi.org/10.1016/j.yhbeh.2015.05.007)
Mohler ML, Bohl CE, Jones A, Coss CC, Narayanan R, He Y, Dong JH, Dalton JT & Miller DD 2009 Nonsteroidal Selective Androgen Receptor Modulators (SARMs): dissociating the anabolic and androgenic activities of the androgen receptor for therapeutic benefit. Journal of Medicinal Chemistry 52 3597–3617. (https://doi.org/10.1021/jm900280m)
Morimoto M, Amano Y, Oka M, Harada A, Fujita H, Hikichi Y, Tozawa R, Yamaoka M & Hara T 2018 Amelioration of sexual behavior and motor activity deficits in a castrated rodent model with a selective androgen receptor modulator SARM-2f. PLoS ONE 12 e0189480. (https://doi.org/10.1371/journal.pone.0189480)
Morin LP & Cummings LA 1981 Effect of surgical or photoperiodic castration, testosterone replacement or pinealectomy on male hamster running rhythmicity. Physiology and Behavior 26 825–838. (https://doi.org/10.1016/0031-9384(81)90106-2)
Musatov S, Chen W, Pfaff DW, Mobbs CV, Yang X-J, Clegg DJ, Kaplitt MG & Ogawa S 2007 Silencing of estrogen receptor alpha in the ventromedial nucleus of hypothalamus leads to metabolic syndrome. PNAS 104 2501–2506. (https://doi.org/10.1073/pnas.0610787104)
Nilsson ME, Vandenput L, Tivesten Å, Norlén AK, Lagerquist MK, Windahl SH, Börjesson AE, Farman HH, Poutanen M, Benrick A, et al.2015 Measurement of a comprehensive sex steroid profile in rodent serum by high-sensitive gas chromatography-tandem mass spectrometry. Endocrinology 156 2492–2502. (https://doi.org/10.1210/en.2014-1890)
Novak CM, Burghardt PR & Levine JA 2012 The use of a running wheel to measure activity in rodents: relationship to energy balance, general activity, and reward. Neuroscience and Biobehavioral Reviews 36 1001–1014. (https://doi.org/10.1016/j.neubiorev.2011.12.012)
Nugent BM, Wright CL, Shetty AC, Hodes GE, Lenz KM, Mahurkar A, Russo SJ, Devine SE & McCarthy MM 2015 Brain feminization requires active repression of masculinization via DNA methylation. Nature Neuroscience 18 690–697. (https://doi.org/10.1038/nn.3988)
Ogawa S, Chester AE, Hewitt SC, Walker VR, Gustafsson J-Å, Smithies O, Korach KS & Pfaff DW 2000 Abolition of male sexual behaviors in mice lacking estrogen receptors α and β (αβERKO). PNAS 97 14737–14741. (https://doi.org/10.1073/pnas.250473597)
Ogawa S, Chan J, Gustafsson JÅ, Korach KS & Pfaff DW 2003 Estrogen increases locomotor activity in mice through estrogen receptor α: specificity for the type of activity. Endocrinology 144 230–239. (https://doi.org/10.1210/en.2002-220519)
Okura T, Koda M, Ando F, Niino N, Ohta S & Shimokata H 2003 Association of polymorphisms in the estrogen receptor alpha gene with body fat distribution. International Journal of Obesity and Related Metabolic Disorders 27 1020–1027. (https://doi.org/10.1038/sj.ijo.0802378)
Onakomaiya MM, Porter DM, Oberlander JG & Henderson LP 2014 Sex and exercise interact to alter the expression of anabolic androgenic steroid-induced anxiety-like behaviors in the mouse. Hormones and Behavior 66 283–297. (https://doi.org/10.1016/j.yhbeh.2014.04.008)