Intracrine and Myotrophic Roles of 5α-Reductase and Androgens: A Review

Abstract

Historically, the circulation was thought to be the primary source of androgens influencing skeletal muscle. However, a growing body of research indicates that skeletal muscle expresses several androgen-synthesizing enzymes, including 5α-reductase. The intramuscular expression of these enzymes suggests that skeletal muscle is capable of synthesizing bioactive androgens, which could induce myotrophic effects via intracrine action.

Purpose:

The aim of this brief review is to discuss recent research related to the intracrine and myotrophic roles of androgens, with particular focus on 5α-reductase as a myotrophic mediator.

Methods:

Included in the review are 17 reviews and 58 original studies that were identified by a systematic review from MEDLINE and deemed particularly relevant to our purpose. Results are summarized to provide an overview of 5α-reductase as a mediator of the myotrophic effects of androgens. In particular, discussions are included regarding androgen biosynthesis and androgen signaling within skeletal muscle, the effects of exercise on intramuscular androgen biosynthesis, and clinical applications of androgens and of a new class of myotrophic agonists termed selective androgen receptor modulator.

Results:

The ability of several peripheral tissues to synthesize bioactive androgens is well documented in the literature. Herein, we summarize newer studies that demonstrate that 1) skeletal muscle has the capability to synthesize both testosterone and dihydrotestosterone from dehydroepiandrosterone, which is present in abundance within the circulation, and 2) that exercise increases the expression of certain androgen-biosynthesizing enzymes within muscle.

Conclusions:

Intramuscularly synthesized androgens have the potential to influence skeletal muscle via intracrine action; however, their exact role in skeletal muscle development and maintenance requires further elucidation.

Both resistance training and endurance training improve skeletal muscle health and performance (31); however, the pathways underlying these training-induced adaptations require further elucidation. One such pathway involves androgen signaling, which occurs via classic genomic-mediated protein translation after cytosolic androgen receptor (AR) activation (19) or via newly recognized rapid nongenomic receptor-mediated pathways (11,25,48). 17β-Hydroxyandrost-4-en-3-one (testosterone) is the most abundant bioactive androgen in circulation and, historically, has been considered to be the primary androgen regulating skeletal muscle. However, testosterone undergoes localized biotransformation to the more potent androgen 17β-hydroxy-5α-androstan-3-one (dihydrotestosterone; DHT) in tissues that highly express any of the 3-oxo-steroid-4-ene dehydrogenase (5α-reductase) isozymes, and as such, androgen signaling is amplified within these tissues. Recently, it has been discovered that skeletal muscle expresses several 5α-reductase isozymes (4,9,32) (Table 1) and several other enzymes involved in localized androgen biosynthesis (3,66) (Fig. 1). The expression of these enzymes opens the possibility that locally produced androgens may exhibit intracrine and/or autocrine actions within muscle, similar to effects observed in the prostate, hair follicles, and other tissues (42). Our primary objective is to review recent research related to the role of the 5α-reductase isozymes in mediating the actions of androgens. We will present brief overviews on androgen biosynthesis, androgen signaling, and the effects of exercise on the aforementioned factors, with a particular focus on skeletal muscle. In addition, we will discuss the clinical applications of androgen administration and the recent development of selective AR modulators (SARM) as potent myotrophic agonists that may not be reliant on 5α-reductase activity.

TABLE 1.

Expression of 5α-reductase isozymes within select androgen-responsive tissues.

Tissue	Species	Type 1 Isozyme	Type 2 Isozyme	Type 3 Isozyme
Skeletal muscle	Human	+	−	+
	Rat	+	+	Unknown
Cardiac muscle	Human	Unknown	−	Unknown
Prostate	Human	−	+	+

+, Expressed; −, not expressed (9,21,32,64).

FIGURE 1—

Androgen metabolism. Leydig cells of the testis express the full enzymatic machinery necessary for testosterone synthesis, whereas the adrenals synthesize and secrete DHEA into the circulation. Peripheral androgen biosynthesis also occurs locally in extragonadal tissues expressing 3β-HSD, 17β-HSD, 5α-reductase, and steroid sulfatase. Skeletal muscle expresses 3β-HSD, 17β-HSD, several 5α-reductase isozymes, and steroid sulfatase. As such, skeletal muscle can synthesize testosterone and DHT starting from circulating DHEA and/or DHEAS.

METHODS

A systematic literature search for all relevant articles was performed through August 2011 using MEDLINE (beginning January 1950). The search strategy was developed using the following key words related to androgen biosynthesis (3β-hydroxysteroid dehydrogenase, 3β HSD, 5α-reductase, 5AR, 17β hydroxysteroid dehydrogenase, 17β HSD, androgen, dehydroepiandrosterone (DHEA), DHT, dutasteride, finasteride, steroid sulfatase, testosterone) either alone or in combination with the following: anabolic, biosynthesis, exercise, intracrine, metabolism, muscle, myotrophic, performance, SARM, steroid, strength, synthesis. Citations were limited to those in the English language. A manual search for relevant references from the articles we identified was also performed. Authors independently screened titles and abstracts for potential inclusion and reviewed full-text versions for relevance. Owing to the nature of this brief review, full-text citations were only included if they were deemed of particular relevance to the review by the authors who are experts in this subject matter. In total, 58 original and 17 review articles were used.

Androgen biosynthesis.

The endogenous production of sex steroid hormones begins with cholesterol synthesis and involves a series of enzymatic reactions, which produce several androgens that vary in their degree of biological activity (Fig. 1). Of these, testosterone is the most abundant bioactive androgen within the circulation and is the principal androgen acting in tissues that do not express the 5α-reductase isozymes; conversely, androgen action is amplified in tissues that express 5α-reductase (42). Testosterone is primarily produced in the Leydig cells of the testes in males, underlying the approximate 10-fold higher circulating testosterone concentrations in postpubertal males, compared with females (67). In addition, DHEA is produced in the adrenals and circulates as both DHEA and its more abundant sulfated form (DHEAS) whose concentration exceeds that of testosterone by greater than 100-fold (26,42). The biological activity of DHEA is lower than that of testosterone; however, its presence in the circulation underlies the possibility that peripheral tissues that express the proper enzymatic machinery may locally synthesize testosterone and DHT from their precursor DHEA (42). Both testosterone and DHT bind to the same AR, although DHT has approximately three times the affinity of testosterone for AR (10). As such, the ability of tissues to locally synthesize testosterone and DHT from DHEA provides a pathway for enhanced tissue-specific androgen action, while limiting the systemic androgen response. Locally produced androgens may subsequently induce intracrine and/or autocrine actions within the tissues that synthesize them, as has been demonstrated in the prostate (42). In this regard, extragonadal steroidogenesis is estimated to account for 40%–50% of the total androgen concentrations within men and is considered the primary source of androgens and estrogens within postmenopausal females (42). Thus, circulating androgens represent only a fraction of the actual androgenic influence within tissues that are capable of localized androgen biosynthesis. The ability of androgens to induce intracrine effects within skeletal muscle has received only recent attention (3,4,6,56,58,66) but requires careful consideration because intratissue androgen concentrations may not mimic circulating androgen concentrations (42). As evidence, bone (another androgen-sensitive tissue that responds to mechanical overload) of male and female rodents contains approximately 30% greater intraskeletal testosterone compared with that found in the circulation (72,74).

The density of AR also provides means of influencing androgen action within peripheral tissues. For example, a high proportion of myonuclei in the rat levator ani/bulbocavernosus (LABC) muscle complex express AR (approximately 74%), and as a result, this muscle experiences robust hypertrophy in response to testosterone administration and severe atrophy after androgen ablation (50). Conversely, the AR density in certain rodent hindlimb muscles of rats is low (e.g., extensor digitorum longus with approximately 7% AR positive nuclei), and only a small degree of growth or atrophy occurs after testosterone administration or androgen ablation, respectively (50). As such, peripheral androgen action is collectively influenced by the concentrations of circulating androgens, the local expression of androgen-synthesizing enzymes, and AR density.

Recently, human and rodent skeletal muscle have been shown to express several enzymes involved in androgen biosynthesis, including 3β-hydroxysteroid dehydrogenase (3β-HSD), 17β-HSD (3,4,6,56,58,66), and several 5α-reductase isozymes (9,32). In addition, skeletal muscle expresses steroid sulfatase (51), which is not required for androgen biosynthesis, but which transforms DHEAS to DHEA, thus increasing the availability of substrate for subsequent peripheral androgen biosynthesis. The presence of these enzymes underlies the ability of skeletal muscle to synthesize both testosterone and DHT from DHEA (3,58), similar to other peripheral tissues (42). Importantly, the expression of the above-mentioned enzymes within muscle provides two potential pathways for localized DHT synthesis, one that is dependent on the 5α reduction of testosterone and one that begins with DHEA but is independent of testosterone (42) (Fig. 1). In this regard, testosterone-independent DHT synthesis provides the potential for locally enhanced androgen action in a tissue-specific manner, while limiting systemic androgenic effects. Biosynthesis of DHT from both testosterone and DHEA has been demonstrated in isolated rat skeletal muscle (58). In addition, a recent report indicates that postmenopausal females have higher intramuscular testosterone than premenopausal females, despite having lower circulating testosterone (56), suggesting that intramuscular androgen biosynthesis occurs in humans. However, no study to date has directly evaluated the ability of human skeletal muscle to synthesize bioactive androgens via DHEA-dependent pathways.

5α-reductase.

The three known 5α-reductase isozymes (types 1, 2, and 3) are membrane-bound proteins that are primarily responsible for the localized tissue-specific steroidogenic conversion of testosterone to its more potent metabolite DHT (42). In addition, peripheral tissues that express both 5α-reductase and 17β-HSD are capable of DHT biosynthesis through an alternative pathway that begins with androstenedione rather than testosterone (42) (Fig. 1). Although testosterone and DHT bind to the same AR, localized production of DHT amplifies androgen signaling because 1) DHT has an approximate threefold higher affinity for AR than does testosterone (10) and 2) testosterone may be rapidly converted to estradiol (via aromatase) or to the weaker androgen androstenedione (via 17β-HSD), whereas DHT cannot (42) and maintains a longer presence within the tissue.

The localized synthesis of DHT is essential to normal in utero and adolescent development of the male pheno-type as demonstrated by the observations that mutations in the 5α-reductase type 2 gene cause pseudohermaphroditism in men (71) and that type 3 mutations cause multiorgan developmental disorders, including muscle hypotonia (18). We are unaware of reports detailing deficiencies of the type 1 5α-reductase isozyme in humans; however, male 5α-reductase type 1 knockout mice appear normal (45), whereas female knockouts experience reduced litter sizes and parturition defects as a result of elevated estrogen (44). These findings provide evidence that 5α-reductase influences development in an isozyme- and tissue-specific manner.

The expression of 5α isozymes varies among muscle (Table 1) and other highly androgen-sensitive tissues, which illustrates the ability of certain tissues to amplify androgen action in the presence of either circulating testosterone or DHEA. Importantly, human skeletal muscle highly expresses the type 3 isozyme (32) and moderately expresses the type 1 isozyme within the subsarcolemmal zone and within close proximity to the sarcolemma (9,56), whereas the type 2 isozyme does not seem to be expressed within human striated muscle (21,64). Some species differences also seem to exist because both the type 1 and 2 isozymes have been isolated from cultured rat skeletal muscle, albeit in relatively low concentrations (58), whereas it remains unknown whether rodent skeletal muscle expresses the type 3 isozyme. Interestingly, Thigpen et al. (64) reported that neither type 1 nor type 2 5α-reductase activity was detected in human skeletal muscle extracts. In addition, human skeletal muscle extracts do not convert radiolabeled testosterone to DHT (63), and relatively little (0.28%) radiolabeled testosterone is converted to DHT in vivo within human skeletal muscle (40). However, these experiments only examined muscle in the resting state and did not address the role of the recently identified type 3 5α-reductase isozyme, which is highly expressed in human skeletal muscle (32) or the possibility that DHT can be synthesized independently of testosterone via a DHEA-dependent pathway. Alternatively, rodent skeletal muscle actively synthesizes both testosterone and DHT from DHEA (58), and both acute and chronic (endurance) exercise seem to increase intramuscular biosynthesis of these androgens (4,5,59). To date, we are unaware of any studies that have examined the influence of exercise on intramuscular DHT biosynthesis in humans. Regardless, the expression of 3β-HSD, 17β-HSD, and the types 1 and 3 5α-reductase isozymes within human skeletal muscle indicates that local biosynthesis of DHT remains possible (32,56).

Androgen signaling pathways.

Androgen signaling occurs through three main pathways: classic genomic signaling, genomic Wnt/β-catenin signaling, and nongenomic signaling (Fig. 2) (11,25,48). Classic genomic signaling mediates most of the familiar effects of androgens on skeletal muscle and bone that require protein synthesis and take weeks or months to develop (11). In classic genomic signaling, testosterone or DHT binds to cytosolic ARs and the hormone-receptor complex translocates to the nucleus where it binds to regions of DNA called hormone response elements (19), resulting in increased protein synthesis. Alternatively, genomic Wnt/β-catenin signaling begins with Wnt-induced inhibition of glycogen synthase kinase 3, an enzyme that phosphorylates β-catenin and marks it for degradation. Thus, Wnt increases β-catenin, which translocates to the nucleus and binds to DNA, influencing transcription and ultimately activation of muscle satellite cells (61). In contrast, newly identified nongenomic AR signaling mediates rapid responses to androgens and occurs through cell surface and cytosolic AR (25,48) as well as nontraditional cell surface receptors such as the epidermal growth factor (EGF) receptor (33) or other G protein–coupled receptors (27). Of particular interest to this review is evidence indicating that testosterone promotes cell proliferation and differentiation through activation of G protein–coupled receptors in a rat skeletal muscle myoblast cell line that lacks AR (27) and that DHT enhances skeletal muscle force generation (33) within isolated Type II muscle fibers after cell surface–mediated EGF receptor activation, an effect that does not occur with testosterone.

FIGURE 2—

Genomic and nongenomic androgen signaling pathways. Genomic signaling mediates the cumulative effects of androgens on protein accretion in two ways. First, androgens (A) pass through the plasma membrane and bind to cytosolic AR. The hormone–receptor complexes enter the nucleus and bind to regions of DNA termed hormone response elements to promote protein accumulation. A second genomic pathway occurs via interactions with Wnt signaling. Wnt binds to cell surface Frizzled (FZ) receptors that activate Disheveled and, in turn, inhibit glycogen synthase kinase 3—an enzyme that phosphorylates β-catenin and marks it for degradation. Subsequently, β-catenin accumulates, enters the nucleus, and binds to T-cell factor/lymphoid enhancer factor 1 (TCF/LEF) response elements that cause activation and proliferation of muscle satellite cells. Nongenomic androgen signaling mediates rapid actions of androgens and may occur through binding of androgens to the cell surface AR or other receptors. The response is mediated through a guanine nucleotide binding protein (GP) and can result in the activation of protein kinases or in the release of calcium ions from the sarcoplasmic reticulum (SR).

Androgens and muscle.

Androgens exert a profound influence on skeletal muscle development and performance, as evidenced by the myocyte-specific AR knockout (mARKO) (52) and myofiber-specific AR knockout mice (20), which exhibit lower skeletal muscle mass compared with wild-types. Orchiectomized mice also experience, reductions in voluntary running distance, maximal running speed, and skeletal muscle mass—results that are completely reversed with physiologic testosterone replacement (37). Furthermore, in the absence of endogenous androgens, both men (39) and male rodents (unpublished laboratory data) experience little to no improvement in lean mass or strength after progressive resistance training. In addition, testosterone administration can 1) dose-dependently increase skeletal muscle mass, muscle fiber cross-sectional area, satellite cell number (35), and mitochondrial area (62); 2) augment the benefits of resistance training on muscle mass and strength (13); and 3) improve aerobic exercise performance by 40%–50% in men with low-normal testosterone (60).

Only limited published data exist on the influence of DHT on skeletal muscle mass or strength. Ly et al. (43) reported that 3 months of transdermal DHT administration reduced body fat mass and improved isokinetic knee flexion strength of the dominant leg but failed to improve lean body mass, knee extension strength, or shoulder flexion/extension strength in hypogonadal elderly men. Similarly, Marin et al. (46) reported that 3 months of transdermal DHT administration increased muscle strength and diameter of Type II muscle fibers, albeit to a lesser extent than testosterone administration. In rodents, physiologic DHT replacement also prevents orchiectomy-induced reductions in the mass of the highly androgen-sensitive LABC (36), again to an apparently lesser degree than testosterone replacement (65). It is not completely understood why the myotrophic efficacy of administered DHT is less than that of administered testosterone, especially considering that DHT binds to AR with approximately three times the affinity of testosterone (10) and maintains a longer presence in the tissues which synthesize it. Potential explanations for this discrepancy are that the doses of DHT administered only replaced DHT to within normal circulating physiologic concentrations (36,43,46), a range that remains only mildly anabolic (65), or that a portion of the myotrophic effects of testosterone may be mediated by tissue-specific conversion of testosterone to estradiol (41), whereas DHT cannot form estradiol. Alternatively, it is also possible that intramuscular DHT remains low after systemic DHT administration because of a lack of substrate (i.e., testosterone or 5α androstenedione) after feedback inhibition, although this remains to be determined.

When administered at physiologic doses, androgens increase skeletal muscle mass primarily via inhibition of protein degradation. As evidence, Ferrando et al. (23) examined arteriovenous phenylalanine balance across the leg and reported that older men undergoing testosterone replacement with the long-acting ester testosterone enanthate (titrated to maintain serum T within physiologic ranges) experienced reduced protein loss during the fasted state but not increased protein synthesis during the fed state. In addition, supraphysiologic testosterone may also increase protein synthesis, as demonstrated by the ability of supraphysiologic testosterone enanthate (at dosages of 300 and 600 mg·wk⁻¹) to increase muscle satellite cell number—a result that was not observed with a replacement dosage of 125 mg·wk⁻¹ (62). Recently, DHT has also been shown to directly enhance Type II skeletal muscle fiber force generation (33) through a rapid nongenomic/nonclassic phosphorylation of the MAPK/ERK1/2 pathway after cell surface–mediated EGF receptor activation. This newly identified pathway, which seems independent of testosterone (33), provides further rationale to examine the role of DHT and 5α-reductase activity within muscle.

5α-reductase and muscle.

Several studies have examined the role of 5α-reductase in androgen-induced myotrophic action. In intact rodents, administration of finasteride (a type 2 5α-reductase inhibitor) (1) reduces the mass of androgen-sensitive organs (e.g., prostate and kidney), which express the type 2 5α-reductase isozyme but does not alter LABC muscle mass (29,65). Inhibition of the type 2 isozyme also produces no decrement in androgen-induced development of lean mass in hypogonadal elderly men, as evidenced by reports that testosterone plus finasteride increases lean body mass and strength to a similar magnitude as testosterone alone (54), despite an approximate 50% reduction in circulating DHT induced by finasteride (8). Conversely, dutasteride (an inhibitor of all known 5α-reductase isozymes [32]) seems to prevent the accrual of lean mass induced by testosteroneundecanoate, in female-to-male transsexuals (47). Dutasteride produces a 990% reduction in circulating DHT by inhibiting all known 5α-reductase isozymes, as opposed to finasteride that produces an approximate 50%–75% reduction in DHT by selectively inhibiting the type 2 isozyme (7). As such, it seems that a critical level of DHT may be necessary for androgen-induced myotrophic action, at least within this unique population. Our laboratory has also performed studies evaluating the systemic inhibition of 5α-reductase using MK-434, a dual type 1 and 2 5α-reductase inhibitor. The results of these studies indicate that administration of MK-434 in combination with supraphysiologic testosterone ablates DHT concentrations and reduces prostate mass to below that found in sham animals but does not impede the myotrophic or bone-protective effects of testosterone (15,16).

Together, these studies suggest that neither type 1 nor 2 5α-reductase activity is required for postnatal androgen-induced myotrophic action, at least in the presence of sufficient testosterone. However, some discrepancy exists between the aforementioned results regarding the lack of lean body mass gain in humans receiving testosterone plus dutasteride and between our rodent testosterone plus MK-434 model. The recent identification of a type 3 5α-reductase isozyme within skeletal muscle (32) may explain this apparent discrepancy, although it remains unknown whether this isozyme is expressed in rodent muscle.

Exercise and androgens.

As previously discussed, endogenous androgens produce myotrophic effects (20,52) and enhance exercise-induced improvements in lean mass and muscle performance (37,39). In addition, exogenous testosterone dose-dependently augments skeletal muscle mass and strength (35). As such, exercise-induced elevations in endogenous androgens (67) or AR (2) may assist in optimizing the muscular adaptations to training. Briefly, high-volume, moderate- to high-intensity resistance training, composed of large muscle group exercises performed with short-rest intervals is the exercise stimulus that most robustly elevates circulating testosterone concentrations (67). In addition, endurance exercise acutely elevates circulating DHEA-sulfate, testosterone, and DHT concentrations within previously sedentary men (34,57). Regardless, it remains somewhat controversial as to whether the transient exercise-induced elevations in circulating androgens are of sufficient magnitude to influence training adaptations in muscle (69,70).

Recently, endurance exercise has been shown to increase intramuscular androgen biosynthesis in rodents, providing an alternative source of androgens that may produce intracrine and/or autocrine actions within skeletal muscle. Specifically, a single bout of endurance treadmill running (30 m·min⁻¹ for 30 min) upregulates mRNA expression and protein content of 3β-HSD, 17β-HSD, and the 5α-reductase type 1 isozyme within rodent skeletal muscle and increases the intramuscular concentrations of testosterone, free testosterone, and DHT (4,6). Similarly, endurance training (30–60 min of treadmill running at 25–30 m·min⁻¹, 5 d·wk⁻¹ for 6–12 wk) increases circulating and intramuscular concentrations of DHEA (59), skeletal muscle mRNA expression of 3β-HSD, and skeletal muscle mRNA and protein expression of 5α-reductase (5). As a result, the intramuscular DHT concentrations of endurance-trained animals are higher than those of age-matched sedentary animals (5). Only a single study has evaluated the intramuscular steroidogenic potential of human skeletal muscle after exercise in healthy, young eugonadal men and women (66) and reported that the intramuscular protein content of 3β-HSD type 1 and 2 and 17β-HSD type 3 did not change in the quadriceps musculature at 10 and 70 min after a single bout of heavy resistance exercise (6 sets × 10 repetitions of squats) and that no alterations in intramuscular concentrations of total testosterone were present after completion of the exercise bout. However, this study did not evaluate alterations in the 5α-reductase isozymes or intramuscular DHT. The possibility exists that differences in training mode (i.e., endurance vs resistance) or exercise intensity differentially effect intramuscular androgen biosynthesis. Regardless, these studies provide evidence that a single bout of endurance exercise increases the synthesis of bioactive androgens and that endurance training enhances this response, ultimately increasing intramuscular androgen concentrations that have the potential to produce intracrine actions within skeletal muscle.

Clinical applications of androgen administration.

Greater than 20% of men older than 60 yr experience symptomatic clinical androgen deficiency (38), which results in limited work capacity (60), reduced skeletal muscle mass and strength (12), and diminished benefits from resistance training (39). Within this population, androgen replacement produces a moderate myotrophic effect (53) and may augment the known benefits of exercise on skeletal muscle mass (35) and aerobic performance (60). However, the improvements in muscle are not robust enough to recommend broad application of androgen replacement (12). For example, Wang et al. (68) compared the effects of testosterone patch versus gel in hypogonadal men and reported only a modest ~1.5-kg increase in lean mass and an ~10% increase in lower body maximal strength after 6 months of administration. Similarly, several studies have demonstrated that transdermal DHT administration over 3 months only modestly increased skeletal muscle fiber area and muscle strength (43,46). In contrast, Bhasin et al. (14) treated older eugonadal men with graded doses of intramuscular testosterone enanthate for 20 wk and reported a robust dose-dependent anabolic effects, including a 7.3-kg increase in lean mass and a 15% increase in lower body maximal strength with the highest dosage administered (600 mg·wk⁻¹). Ferrando et al. (24) also treated older men with testosterone enanthate (titrated to increase serum testosterone approximately twofold) and reported a 4.2-kg increase in lean mass and a 15.3-kg increase in leg extension strength after 6 months. Unfortunately, higher doses of testosterone produce several adverse events, the most common being polycythemia, increased incidence of prostate related events (17), and slight reductions in high-density lipoprotein cholesterol (22). In addition, understanding the influence of 5α-reductase on skeletal muscle also remains important within this population because finasteride and dutasteride are commonly prescribed for the treatment of several conditions associated with aging, including benign prostate hyperplasia and male pattern baldness (1). Considering that dutasteride is a potent inhibitor of all known 5α-reductase isozymes, its use certainly limits synthesis of DHT within tissues expressing 5α-reductase (e.g., prostate and skeletal muscle). The effects of complete inhibition of intramuscular 5α-reductase remain largely unknown but may be detrimental to the accretion skeletal muscle as demonstrated by the inability of testosterone-undecanoate to induce lean mass gains when administered in combination with dutasteride (47). As such, future preclinical and clinical trials focusing on the development of strategies to produce strong myotrophic effects, along with a low incidence of adverse androgenic events, seem warranted for the treatment androgen deficiency and associated muscle loss.

SARM.

SARMs are a growing class of steroidal and nonsteroidal AR ligands that are designed to produce anabolic effects in specific androgen responsive tissues (e.g., skeletal muscle and bone), while minimizing the common adverse androgenic responses associated with testosterone administration in other nontarget tissues (e.g., prostate) (49). Classes of SARMs that are currently under investigation include antiandrogens (e.g., monoarylpropionamide flutamide) and nonsteroidal AR agonists (e.g., proprionamides, bicyclic hydantoins, and quinolinones) that possess both myotrophic and bone-protective effects (49). Briefly, the orally active proprionamides S-4 and S-1 possess similar myotrophic activity as DHT, as evidenced by the ability of S-4 to dose-dependently increase muscle strength and restore levator ani muscle mass to the level of intact rats after orchiectomy, while limiting prostate growth (30), and by the ability of S-1 to augment muscle mass with minimal response in the prostate (49). Similarly, several bicyclic hydantoins and quinolinone derivatives also restore lean body mass in orchiectomized rats (49). Further work is required to elucidate the mechanism(s) that allow their tissue-specific AR activation, although it has been proposed that SARMs lack the ability to undergo 5α reduction (28), which would reduce their potency in tissues highly expressing 5α-reductase (e.g., prostate), while maintaining anabolic potency in tissues with lower 5α-reductase activity (e.g., muscle) (Table 1). In addition, several synthetic androgens possess SARM-like activity because they do not undergo 5α reduction (75). As evidence, both 17β-hydroxyestra-4,9,11-trien-3-one (trenbolone) (73) and 19-nor-4-androstenediol-3β,17β-diol (3β,19-NA) (55) increase LABC muscle mass and protect against orchiectomy-induced bone loss in rats, without inducing prostate enlargement.

CONCLUSIONS

The maintenance of skeletal muscle mass positively influences health throughout aging. Exercise represents the most efficacious means of safely improving skeletal muscle mass and physical function (31); however, androgen-deficient populations (and others unable or unwilling to exercise) may not reap the benefits of exercise training (39). As such, understanding the mechanism(s) through which exercise influences skeletal muscle may lead to alternative means of improving muscle mass and physical function within these populations. Androgen signaling is one such mechanism that provides a profound positive influence on the development and maintenance of muscle (20,52). Historically, the circulation was considered to be the primary source of androgens capable of influencing skeletal muscle. However, recent evidence indicates that skeletal muscle expresses the full machinery of enzymes (i.e., 3β-HSD, 17β-HSD, and 5α-reductase) necessary to synthesize both testosterone and DHT from DHEA (3,58). In addition, both acute and chronic (endurance) exercise increase the expression of several of these enzymes, ultimately elevating intramuscular testosterone and DHT concentrations (4,5). Thus, in addition to what is found in the circulation, locally synthesized androgens may influence skeletal muscle via intracrine and/or autocrine actions. Currently, preclinical and clinical research is ongoing in an effort to develop alternative therapies to testosterone replacement (e.g., DHT, testosterone plus finasteride/dutasteride, or various SARMs) that use androgen-mediated pathways to enhance skeletal muscle mass and performance, while limiting adverse events. Future research focused on determining the role(s) that the 5α-reductase isozymes play in the myotrophic effects of androgens is warranted and would assist in elucidating the influence of intramuscular androgen biosynthesis on skeletal muscle.