Skeletal muscle wasting: the estrogen side of sexual dimorphism

Abstract

The importance of defining sex differences across various biological and physiological mechanisms is more pervasive now than it has been over the past 15–20 years. As the muscle biology field pushes to identify small molecules and interventions to prevent, attenuate, or even reverse muscle wasting, we must consider the effect of sex as a biological variable. It should not be assumed that a therapeutic will affect males and females with equal efficacy or equivalent target affinities under conditions where muscle wasting is observed. With that said, it is not surprising to find that we have an unclear or even a poor understanding of the effects of sex or sex hormones on muscle wasting conditions. Although recent investigations are beginning to establish experimental approaches that will allow investigators to assess the impact of sex-specific hormones on muscle wasting, the field still needs rigorous scientific tools that will allow the community to address critical hypotheses centered around sex hormones. The focus of this review is on female sex hormones, specifically estrogens, and the roles that these hormones and their receptors play in skeletal muscle wasting conditions. With the overall review goal of assembling the current knowledge in the area of sexual dimorphism driven by estrogens with an effort to provide insights to interested physiologists on necessary considerations when trying to assess models for potential sex differences in cellular and molecular mechanisms of muscle wasting.

INTRODUCTION

Skeletal muscle wasting, also referred to as muscle atrophy, occurs in response to a variety of conditions. The onset, duration, and magnitude of muscle wasting vary depending on the condition (i.e., muscle disuse, cachexia, or aging) and may by influenced by sex. One cause of sexual dimorphism is sex hormones: estrogens, progestins, and androgens. The focus of this review is on female sex hormones, specifically estrogens, and the roles that these hormones and their receptors play in skeletal muscle wasting conditions. The overall goal of this review is to advance readership knowledge in the area of sexual dimorphism driven by estrogens to enable physiologists to better elucidate cellular and molecular mechanisms of muscle wasting. First, we provide an overview of estrogen biology focusing on mechanisms of action in skeletal muscle (Fig. 1). Second, we present data from conditions that induce skeletal muscle wasting including disuse, sarcopenia, cachexia, and injury demonstrating how estrogens may or may not be involved. Finally, we end with a summary of estrogenic therapeutic interventions that may become beneficial in muscle wasting conditions as the field continues to unravel sex hormone mechanisms contributing to the maintenance of skeletal muscle mass.

Figure 1.

Effects of 17β-estradiol deficiency on conditions of skeletal muscle wasting caused by disuse, injury, cachexia, and sarcopenia. [Images were generated at Biorender.com and published with permission.]

OVERVIEW OF ESTROGEN BIOLOGY

Estrogens are primarily produced in and secreted from the granulose cells of the ovaries in females and the Sertoli cells of the testes in males in response to follicle-stimulating hormone released from the anterior pituitary (1–4). In both sexes, adipose, fibroblast, mesenchymal stromal, and epithelial cells also produce estrogens locally (5). The production of estrogens in all cells is dependent on the expression and abundance of the enzyme aromatase (5–7). Aromatase regulates the conversion of androgens into estradiol [17-β estradiol (E₂)] and estrone (E₁), which are converted to estriol (E₃) during pregnancy (1, 8). Aromatase expression and activity have been detected in human (9) and rat (10) skeletal muscle homogenates, suggesting that muscle may have the ability to convert testosterone to estrogen locally; in contrast, aromatase mRNA was not detected in mouse skeletal muscle (11). Immunofluorescence staining for aromatase indicated that aromatase was localized to either the sarcolemma or subsarcolemmal zone of skeletal muscle fibers from pre- and postmenopausal women (12). Important to note is that aromatase activity in nonfiber cell types within a skeletal muscle could contribute to skeletal muscle androgen-to-E₂ action (13). For example, aromatase expression has been detected in isolated vascular smooth muscle cells (14, 15), endothelial cells (16), and macrophages (17), cell types that are known to be within skeletal muscle tissue. Thus, whether or not muscle cells (fibers) per se contain aromatase, particularly in rodent muscle, is not clear but likely as a tissue, skeletal muscle, has the ability to convert androgens to estrogens because of nonfiber cells in the environment containing aromatase.

In reproductive aged females, circulating concentrations of estrogens fluctuate during the menstrual [human (18, 19)] or estrous cycle [rodent (20–22)]. The menstrual cycle consists of the follicular and luteal phases (19). The estrous cycle is divided into four stages: proestrus and estrus, which make up the follicular phase, and the metestrus and diestrus, which consist of the luteal phase (22). E₂ peaks at the end of the follicular phase (28–475 pg/mL) in women (19) and during proestrus in rodents (rats: ∼35 pg/mL, mice: ∼8 pg/mL) in rodents (21). Accurate measurement of female hormones is notoriously difficult due to a high degree of variability across the women’s lifespan [e.g., E₂ can range from <20 pg/mL in postmenopausal women to 20,000 pg/mL during pregnancy (23)] (Table 1). In rodent models, challenges include the compressed 5-day cycle of fluctuating E₂, small volumes of blood (serum) sample available to assay, and commercial assays that are developed primarily for human samples having potential less specificity for rodent blood. To address these issues, commonly employed methods such as ELISA, enzyme immunoassay, or radioimmunoassay that have limited accuracy, specificity, reproducibility, and standardization (29) are now being replaced by chromatographic methods combined with mass spectrometry in both human and rodent research (30).

Table 1.

Reported serum estradiol, estrone, and progesterone levels measured by GC- or LC-MS/MS from human and rodent blood

	Estradiol, pg/mL	Estrone, pg/mL	Progesterone, pg/mL	Method	References
Women
Premenopausal	15–350	17–200	1,000–45,000	LC-MS/MS	(24, 25)
Postmenopausal	1–9	5–33		LC-MS/MS	(26)
<5 yr	5	40		LC-MS/MS	(27)
>5 yr	1.3	35			(27)
Men	10–40	15–60		LC-MS/MS	(24, 28)
Rats
Female (3–4 mo)	<1–45	1–15	5–20	GC-MS/MS	(21)
Mice
Female (3–4 mo)	1–12	1–2	10–60	GC-MS/MS	(21)

<5 yr and >5 yr indicate data from women who were postmenopausal for a time less than or greater than 5 yr since their last menstrual cycle. GC, gas chromatography; LC, liquid chromatography, MS/MS, tandem mass spectrometry.

The impact of female sex hormones, primarily E₂, on skeletal muscle has been studied to a limited extent. There is some evidence in muscle physiology, as in many other systems (31, 32), that cycling levels of estrogens do not have a significant influence on biological variability. For example, in women, the fluctuation of sex hormones during the menstrual cycle did not elicit variation in muscle contractile properties including strength, power, and fatigability (33, 34). Similarly, female mice did not demonstrate alterations in running power, running economy (at >50% max oxygen consumption), or grip strength across the various phases of the estrous cycle (35). The phase of the estrous cycle did not impact cardiac function (e.g., heart rate, cardiac output, or stroke volume) or system mean arterial blood pressure in female rats (36). However, when challenged by hemorrhage shock, cardiac functions were depressed in all phases except during the proestrus phase, the phase of the cycle when E₂ levels are the highest (36). This latter observation suggests that fluctuation of E₂ during the normal menstrual or estrous cycle under homeostatic conditions may have little to no impact on physiological muscle function, but rather changes in E₂ concentrations may play a more critical role in the ability of tissue to respond to an acute stress, perhaps such as those that elicit skeletal muscle wasting. A complexity to consider is that some acute stressors associated with muscle atrophy can induce alterations in the estrous cycle. For example, hindlimb unloading in rats causes altered duration and frequency of the estrous cycle (37, 38). In contrast, animals enduring the stress of space flight had very little change in the estrous cycle (39) suggesting that the investigator may need to determine how the acute stress model affects the estrous cycle to accurately assess the role of E₂ in female muscle during conditions of muscle wasting.

Estrogen Transport, Receptors, and Mechanism of Action

In humans, E₂ is transported in the bloodstream to target tissues primarily by being reversibly bound to sex-hormone-binding globulin (SHBG; ∼55%) or carrier proteins such as albumin (∼42%) with only ∼3% being free and nonprotein bound (40, 41). The “free hormone hypothesis” proposes that the biological activity of hormones is determined by the free concentration of E₂ in the bloodstream (42, 43). However, whether E₂ bound to SHBG or carrier proteins alters the biological activity of E₂ in humans is currently unknown (42). Although rats and mice lack SHGB, experiments using transgenic mice expressing SHBG (SHBG-tg) provide support to the free hormone hypothesis (43). E₂ treatment of ovariectomized (OVX) SHGB-tg mice increased circulating E₂ concentrations; yet the elevated E₂-SHBG levels failed to prevent the loss of uterine mass like E₂ treatment to nontransgenic OVX mice did (43). These results suggest that the concentration of free, nonprotein-bound E₂ levels is important for biological activity.

Estrogens are cholesterol derivatives that when free unbound can diffuse across membranes of target cells and bind intracellular or membrane-bound receptors to mediate physiological effects (40). There are three known estrogen receptors (ER): ERα, ERβ, and G protein-coupled receptor (GPER, formally designated as GPR30). Each of the estrogen receptors have been identified in numerous tissues and cell types throughout the body including skeletal muscle (44–48).

ERα and ERβ are both members of the nuclear receptor protein family (49) yet differ in the COOH-terminal ligand binding domain and in the NH₂-terminal transactivation domain, binding affinity for E₂, and tissue distribution and expression levels (40, 47). The ESR1 gene encodes for ERα protein and produces at least two variants (66 and 46 kDa; for review, see Refs. 50–53). The ERα (66 kDa) protein contains 595 amino acids and is highly expressed in the uterus and ovaries but is also found in the prostate stroma, testes, epididymis, breast, skeletal muscle, hypothalamus, pituitary, liver, pancreas, adipose, heart, bone, lungs, and blood vessels. The smaller ERα (46 kDa) variant has a truncated NH₂-terminal resulting in a missing activation function (AF)-1 domain. This 46-kDa variant has not been described in skeletal muscle but has been detected in a membrane fraction of cardiac muscle (54), with other studies in nonmuscle tissue, suggesting it can be found in the plasma membrane (55, 56). The ESR2 gene encodes for ERβ, with the full-length protein containing 530 amino acids (60 kDa; for review, see Ref. 57). Multiple variants of ERβ have been identified with various masses because of alternative splicing, which results in the lack of the ligand-dependent AF-2 domain (53, 58). ERβ is highly expressed in ovaries and has also been identified in prostate epithelium, testes, epididymis, bone marrow, hypothalamus, adipose, lungs, bone, and blood vessels (5, 45, 47).

GPER is a 375-amino-acid (42 kDa) protein encoded by the GPER1 gene (59, 60). Although single nucleotide polymorphism variations of GPER have been detected and are associated with cancer predisposition and progression (61), little is known if the variants play a role in normal physiological functions. GPER is a seven transmembrane-domain G protein-coupled receptor with a subcellular localization on the plasma membrane; however, it has also been suggested to be localized on the endoplasmic reticulum (60, 62). GPER is expressed in the central nervous system, reproductive tissues, pancreatic islets, adipose, liver, skeletal muscle, heart, intestine, and inflammatory cells (48).

ER and GPER knockout mice have shown that these receptors have differing roles in estrogen physiology and distinct intracellular signaling mechanisms (63, 64). The “classic” or genomic mechanism of estrogen-mediated biological effects involves E₂ diffusing directly across the cell membrane into the cytosol or nucleus and binding to ERα or ERβ (50, 53). The E₂-ER then forms a dimer and binds to estrogen response elements (ERE) in the promoter and enhancer regions of genes, thereby regulating the transcription of estrogen responsive genes (40, 65, 66). In addition to the classic genomic mechanisms, E₂ induces rapid activation of numerous protein-kinase cascades that can lead to indirect changes in gene expression because of protein phosphorylation. These nongenomic actions involve extracellular signal-regulated pathways such as phospholipase C (PLC)/protein kinase C (PKC) and p38/mitogen-activated protein kinase (MAPK) (60, 67). These pathways are thought to be primarily linked with membrane-associated ERs as well as the more recently identified GPER (60, 67). For more details on estrogen genomic and nongenomic mechanisms, see thorough reviews by Arnal et al. (50), Fuentes and Silveyra (67), and Olde and Leeb-Lundberg (48).

Estrogen Receptors in Skeletal Muscle

Identifying estrogen receptors in skeletal muscle tissue and especially cells (fibers) has been challenging because of 1) unreliable antibodies for ERα and ERβ producing false-positive results (54), 2) ∼60% of total protein in skeletal muscle being actin and myosin, making detection of low abundance proteins such as ERs difficult (68, 69), and 3) whole skeletal muscle tissue homogenates being “contaminated” by multiple cell types in the tissue that are known to express estrogen receptors [e.g., neurons (70), endothelial cells (71), and hematopoietic cells (72, 73)]. This latter challenge makes interpretation of the detection and quantification of ER proteins or mRNA distinctly in myofibers difficult.

To attribute ERs to specific cell types in skeletal muscle, strategies of fractionation, immunohistochemistry, or in situ hybridization can be used (with a caveat of understanding the challenges of working with ER antibodies or that mRNA localization should not be used as a substitute for protein localization). For example, the use of cellular protein fractionation has identified ERα and ERβ in both the crude nuclear and cytosolic/membrane fractions in skeletal muscle homogenate from postmenopausal women (74). Yet, immunostaining of the vastus lateralis muscle of children, adult men and women, and postmenopausal women has identified the primary localization of both ERα and ERβ to be in the myonuclei of muscle fibers (75). Both ERα and ERβ have been identified via immunofluorescence in mouse (76) and rat (77) skeletal muscles with localization in the nuclei. We and others have delivered fluorescently labeled ERα as means to determine ERα localization and circumvent ER antibodies challenges (54, 78). Following the electroporation of cDNA that encodes for GFP-tagged ERα (46 kDa) or ERα (66 kDa) into mouse skeletal muscle fibers, we largely found that both variants localize to the nucleus almost exclusively (78) in agreement with data from others (54). Although the degree to which ERα and ERβ are expressed on the sarcolemma has yet to be determined, membrane-bound GPER expression has been identified in human and rodent skeletal muscle and myotubes (12, 44, 79–81). Additionally, ERα and ERβ have been identified in satellite cells and play a key role in satellite cell dynamics during muscle regeneration (82, 83).

In addition to expression in the nucleus, competitive binding assays using the C2C12 skeletal muscle cell line suggest that there are nonclassic estrogen binding sites in mitochondria and microsomes (84). Immunolocalization of ERα and ERβ in C2C12 myoblast suggest that ERα is localized in the nucleus, microsomes, cytosol, and mitochondria whereas ERβ is primarily expressed in the mitochondria and to a lesser extent in the cytosol, nucleus, and microsomes (85). These nonclassic receptor localizations of ERα and ERβ have been proposed to contribute to E₂ antiapoptotic effects in skeletal muscle (86), and thus, could play a role in maintaining skeletal muscle mass during muscle wasting conditions.

Recent evidence suggests that E₂ can affect skeletal muscle independent of the ERs by localizing directly in the membranes affecting the fluidity of the membrane structure (87). In this case, E₂ appears to have the ability to integrate with the mitochondrial membrane and influence mitochondrial energetics through effects on membrane fluidity (87). It is unclear at this point, if E₂ can integrate with any other membranes such as the sarcolemma. However, since the structure of E₂ is similar to cholesterol and the sarcolemma is enriched in cholesterol, it is conceivable that E₂ may locate there as well and affect mechanisms that regulate skeletal muscle properties such as muscle mass.

THE INFLUENCE OF ESTROGEN ON MUSCLE WASTING

The hormonal profile is cyclic in women for the majority of their lifespan with the cyclic nature becoming less pronounced as ovarian estrogen and progesterone production decrease with age. This can result in women spending approximately one-third of their lives in a postmenopausal, E₂-deficient state (88), and studies suggest that there is an accelerated loss of muscle mass associated with the onset of menopause (89–91). E₂ deficiency is relevant for younger women as well as it can be a consequence of eating disorders, energy imbalances due to high-load exercise training, chemotherapy, or hysterectomy, for example. Thus, the mechanism by which E₂ deficiency in females of any age couples with events that induces muscle wasting such as disuse or injury is critical to understand. Studies examining sex differences may hint at possible hormonal influences on muscle mass; however, additional research is needed to determine the cellular and molecular mechanisms underlying E₂-mediated effects on muscle wasting as there are numerous other differences between the sexes than simply sex hormones. Below we present four conditions that cause skeletal muscle wasting and experimental results indicating sex differences and/or an influence by E₂, as well as some mechanisms by which they occur.

Muscle Disuse Atrophy

Muscle mass is largely determined by the coordinated balance between muscle protein synthesis and muscle protein degradation. During periods of prolonged disuse (e.g., bedrest, limb immobilization, spaceflight), rates of muscle protein synthesis decrease significantly with concurrent elevations in degradation resulting in muscle atrophy. Because of the historical preference of using males in nonclinical investigations of disuse pathology, it was previously accepted that males and females exhibit similar cell signaling responses to inductions of atrophic pathways (92). However, recent works have highlighted sex-based disparities in the mechanisms that regulate disuse atrophy, suggesting an increased susceptibility to muscle wasting in females (93). Yoshihara et al. (94) found greater reductions in soleus muscle mass and cross-sectional area (CSA) in female rats than in males following 7 days of hindlimb unloading (HLU). Similarly, HLU led to significantly greater reductions in Type IIb myofiber CSA in the tibialis anterior muscles of female mice compared with males (95). Future research should investigate potential sexual dimorphisms in muscle wasting during long periods of physical inactivity considering the shorter duration of unloading (1–7 days) within these studies.

Mechanisms that may underlie greater loss of muscle mass in female rodents include lower rates of fractional protein synthesis and mRNA profile changes indicative of muscle catabolism during the first 24 h of disuse (95). Specifically, unloaded muscles of female rodents exhibited increased levels of ubiquitinated proteins because of alterations in the atrophic Foxhead box O3 (FoxO3a)/ubiquitin-proteasome pathway (94, 96). It appears that the enhanced severity of muscle wasting in females following disuse is attributed partly to elevated protein ubiquitination and degradation, especially during the initial phase of inactivity. Additional future research is necessary to determine whether these mechanisms are mediated by E₂ as ubiquitination of the ERs has been demonstrated to disrupt receptor stability and function, ultimately impairing E₂-ER signaling (97).

The seemingly greater predisposition of females than males to disuse-induced muscle wasting is further evident in conditions of E₂ deficiency, such as that which occurs following the onset of menopause. Menopausal-induced E₂ deficiency is modeled most often in preclinical rodent studies via bilateral surgical removal of the ovaries [i.e., ovariectomy (OVX)], resulting in substantial reductions of circulating ovarian hormones (98). Studies have shown that reloading of hindlimb skeletal muscles of OVX rats following HLU resulted in failure of muscle to recover atrophied muscle mass compared with ovary-intact rats (99, 100). Ovarian E₂ deficiency was associated with impaired myofiber growth, myofiber regeneration, and extracellular matrix remodeling during muscle recovery from HLU (101). Importantly, impairments in the recovery of skeletal muscle mass from mechanical unloading can be rescued by E₂ treatment pinpointing E₂ as the major ovarian hormone impacting muscle mass (100, 101). Furthermore, E₂ treatment facilitated the muscle mass recovery response of OVX rats during a rehabilitation exercise program (102), an effect also seen in postmenopausal women who were subjected to dual interventions of E₂-based hormone therapy (HRT) and exercise [E₂-related therapies; Sipilä et al. (103)]. A summary of the current knowledge surrounding the effects of HRT on muscle mass is discussed below.

In contrast to rodents, men but not women exhibit a loss in fiber size under conditions of reduced physical activity; however, women demonstrate lower muscle power output after physical inactivity compared with men (104). The authors suggested that in response to this type of muscle disuse (i.e., physical inactivity), actomyosin cross-bridge kinetics are slowed leading to reductions in myofiber force production and shortening velocity in fibers from women (104). These effects were further attributed partly to greater oxidative stress on contractile proteins in females than in males. Interestingly, estrogens appear to have tissue-specific effects on reactive oxygen species (ROS) emission potential, with E₂ decreasing ROS emission in skeletal muscle (105). Thus, it can be speculated that any modulation in E₂ levels or E₂-ER signaling could lead to alterations in redox balance resulting in decreased muscle mass and/or strength. Overall, such studies indicate that across nonclinical models and in humans, the regulation of skeletal muscle mass in females following periods of disuse appears to be affected by ovarian hormone status and that ovarian function is important to recover from disuse that induces muscle atrophy.

Sarcopenia

Sarcopenia is defined as the age-associated reduction in skeletal muscle mass and strength that can lead to a decline in functional independence and an increased risk of disability, hospitalization, and mortality rates in aged adults (106). Factors contributing to sarcopenia include increased inflammation and oxidative stress, impaired mitochondrial function, satellite cell senescence, elevated apoptosis and proteasome activity, and suppressed protein synthesis and myocyte regeneration (107). Sarcopenic mechanisms appear to be sex dependent as elderly men exhibit more severe muscle wasting than elderly women because of male’s relatively larger muscle mass (108). However, the extent to which alterations in sex hormones contribute to the progression of sarcopenia remains to be fully elucidated. Despite the wide degree of variation among techniques and populations involved in longitudinal studies of sarcopenia, median rates of muscle mass loss in men and women aged ≥75 yr have been reported as 0.80–0.98% and 0.64–0.70% per year, respectively (109). It is important to note these data do not account for lifestyle factors (e.g., diet, degree of physical activity, socioeconomic status) among included population, which can impact severity of sarcopenia observed between sexes.

A chronic state of low-grade inflammation is a hallmark of aging (often termed “inflammaging”) and is marked by increased levels of proinflammatory cytokines such as tumor necrosis factor α (TNFα) and interleukin 6 (IL-6) (110). TNFα and IL-6 are well-characterized inflammatory biomarkers of aged skeletal muscle that stimulate muscle catabolism directly via activation of the UPS or indirectly by inducing muscle insulin resistance and inhibiting protein synthesis (111, 112). Although TNFα has been demonstrated to activate the UPS through NF-κB signaling in skeletal muscle (110), its contribution to muscle wasting during aging does not appear to occur through this pathway (113). Importantly, elevations in proinflammatory cytokines coupled with extended periods of physical inactivity (113) are strong predictors of morbidity and mortality in the aged population (114, 115).

Currently, there is no means to prevent sarcopenia. Thus, many groups are pursuing potential pharmacological agents and therapeutic interventions to minimize age-related muscle wasting. It is widely accepted that androgens exhibit anabolic effects on skeletal muscle in sarcopenic men (116), but unfortunately the ability of estrogens to prevent muscle wasting in sarcopenic women is not readily known. Previous research has highlighted potential mechanisms that may occur with the loss of ovarian-derived E₂ and senescent female skeletal muscle. For example, E₂ deficiency is associated with apoptosis-linked microRNAs (117) and increased levels of aforementioned cytokines (109, 111). Accordingly, E₂ treatment in OVX rats decreased levels of proinflammatory cytokines and reduced muscle damage as evidenced by lower serum creatine kinase expression compared with that in untreated OVX rats (118). These data may position early anti-inflammatory treatment as an especially useful therapeutic to minimize the development of sarcopenia in the aged female population because of the apparent exacerbation of sarcopenic muscle wasting by ovarian loss of E₂.

In addition to inflammation, alterations in mitochondrial energetic regulation influence the onset of sarcopenia (119–121). A significant body of literature across a variety of tissues has demonstrated myriad effects of estrogens and E₂-ER signaling on mitochondrial function (e.g., ATP production, ROS emission, membrane potential regulation, calcium handling) as well as mitochondrial biogenesis (122). Thus, it is logical to consider that alterations in skeletal muscle mitochondria during aging may be mediated by decreases in E₂ and ultimately contribute to the onset of sarcopenia in women. A number of groups have found that skeletal muscle from OVX mice demonstrate reduced ADP-stimulated mitochondrial respiration and an oxidative shift in redox homeostasis compared with control mice (105, 123). Loss of ovarian-derived E₂ results in reduced complex I activity in skeletal muscle alongside an oxidative shift in the glutathione ratio and an increase in ROS emission, all of which were reversed with E₂ treatment in OVX mice (87). As OVX mice are not commonly aged, it may be difficult to distinguish between the independent effects of decreased E₂ and aging on mitochondrial function. However, elevated ROS emission potential is often found in various models of muscle wasting and in sarcopenia, suggesting that increased ROS emission due to E₂ deficiency may contribute to or exacerbate sarcopenic muscle wasting (reviewed in Ref. 124).

Determining the manner in which E₂ influences mitochondria in skeletal muscle has led to conflicting results concerning the role of ERα. The conditional deletion of ERα in cultured myotubes resulted in lower respiratory values (125), whereas in an inducible skeletal muscle-specific ERα knockout mouse model, no differences were found across multiple different mitochondrial respiratory measures (78). Similar results were confirmed by Torres et al. (87) who found that E₂ associated with the mitochondrial membrane affected the fluidity of the membrane, thus influencing activity of the respiratory complexes. Furthermore, the association of E₂ with the mitochondrial membrane occurred independent of ERα (87). It remains to be empirically determined if loss of E₂ is a major leader to the onset of sarcopenia due to excessive ROS emission from muscle mitochondria; however, circumstantial evidence suggests it is possible and future experiments are needed.

Muscle Cachexia

Cachexia is a complex syndrome that is characterized by the unintentional loss of body weight of >5% or >2% for people with a body mass index <20 kg/m², or >2% with individuals already presenting with established sarcopenia (appendicular muscle index: men <7.26 kg/m²; women <5.45 kg/m²) (126, 127). Muscle cachexia is associated with several diseases including autoimmune diseases (e.g., rheumatoid arthritis, multiple sclerosis), cancer (type and stage), chronic obstructive pulmonary disease, congestive heart failure, diabetes, renal failure, and sepsis (128, 129). Cachexia is considered to be caused by the dysregulation of metabolic, immunological, and neurological factors and involves multiple pathways (for review, see Refs. 129–134).

Elucidating the potential of E₂ to alter cachexia by comparing men and women is difficult to determine because of diagnostic age (e.g., pre- vs. postmenopausal women), type of primary disease, body composition sex differences, level of physical activity, and pharmaceutical interventions that may be sex specific. Yet, given these limitations, the susceptibility and progression of cachexia seem to differ between the sexes (107, 135). For example, men with rheumatoid arthritis have a greater deficit in muscle mass than women (136); yet women with multiple sclerosis have less loss in muscle fiber cross-sectional area compared with men (137). In a study comparing muscle wasting in men and women patients with lung cancer, results indicated that although both sexes had a similar ∼6% total percent body weight loss over 6 mo, significantly more men (61%) than women (38%) were classified as sarcopenic (138). Similarly, in patients with progressive cachexia defined as >10% weight loss, men had a greater reduction in muscle mass loss compared with women (139). Taken together, these studies hint at the possibility that sex-specific hormones may be relevant to the prevention and/or treatment of muscle cachexia (130, 140).

The immune system is known to play a central role in the development of cachexia (128, 132, 133), with proinflammatory cytokines such as IL-1, IL-6, and TNF-α, all being associated with the progression of cachexia (141) from multiple origins [autoimmune (142), cancer (132), sepsis (131, 132, 140–143)]. E₂ is a known regulator of the immune system (144, 145), and treatment with E₂ has been demonstrated to alter expression of both IL-6 and TNF-α in humans (146) and mice (147) during conditions associated with cachexia. Thus, it has been purported that E₂ could be protective against the cachexia (135); yet studies to directly examine muscle mass in response to estrogen treatment in diseases associated with cachexia are lacking. Future nonclinical and clinical studies are needed to determine if an estrogenic treatment can mitigate inflammation-induced progression of cachexia and further to substantiate molecular mechanisms involved.

Some mechanistic insight on how estrogens influence cachexia can be gleaned from studies using rodent models. Similar to studies in humans, a study examining cachexia in a nonclinical tumor-bearing mouse model (Apc^Min/+) indicates that male and female mice differ with male Apc^Min/+ mice having greater reductions in body mass than females (148). Additionally, it was hypothesized that the sex difference in muscle loss was IL-6 dependent as male Apc^Min/+ mice expressed higher (∼58%) circulating IL-6 levels than female Apc^Min/+ mice (148). Hetzler et al. (149) further demonstrated that tumor-bearing acyclic mice had elevated muscle cytokine gene expression (TNFα and IL-6) compared with female mice with normal estrous cycles. Ovariectomized colorectal cancer (azoxymethane/dextran sulfate sodium-treated) mice that were treated with E₂ had decreased expression of both TNFα and IL-6 expression (147), additionally suggesting that estrogen alters cytokine expression and thus could be beneficial to reducing the progression of cachexia.

Direct impact of E₂ on muscle wasting in a cachexic model is shown in a study by Counts and coworkers (150). They examined muscle cachexia in a nonclinical tumor-bearing mouse model (Apc^Min/+) and showed that E₂ treatment of OVX cachexic mice leads to greater muscle mass compared with those mice not treated with E₂ (151). Additionally, the loss of body and muscle masses was less in female cachexic mice with normal estrous cycles compared with cachexic mice with abnormal cycles and presumably low E₂ (149, 151). Collectively, these data demonstrate that estrogen is beneficial for maintaining muscle mass during Apc^Min/+-induced cachexia. Counts et al. (151) proposed that the beneficial estrogenic effects on muscle mass may occur through the mechanistic target of rapamycin (mTOR) pathway. Skeletal muscle of E₂-treated Apc^Min/+ mice had an increased activation of mTORC targets including eukaryotic initiation factor 4E binding protein-1 (4E-BP1:T37/44) and ribosomal S6 kinase 1 (rpS6:S240/244), suggesting an increase in protein synthesis rates (151). Interestingly, sepsis, which induces elevated cytokine expression and results in cachexia, blunts muscle protein synthesis by impairing mTOR phosphorylation of 4E-BP1 (150, 152). In contrast, when cytokine production is inhibited 4E-BP1 phosphorylation and protein synthesis are increased (152, 153). Taken together, such results indicate that E₂ alterations of mTOR signaling may be a secondary effect of E₂ modulating the immune system. More research is needed to further determine the extent that E₂ directly regulates mTOR signaling in muscle as well as the extent to which the beneficial effects of E₂ treatment on muscle cachexia are involved in other cachexia-inducing diseases (autoimmune, renal failure, etc.).

Skeletal Muscle Injury

Skeletal muscle incurs injuries throughout life and requires repeated regeneration to regain mass and contractile function. With aging, the capacity to efficiently regenerate and recover from injury is impaired and contributes to a decrease in muscle mass (154–156). A type of stem cell specific to skeletal muscle (MuSC), also known as the satellite cell, mediates skeletal muscle regeneration. MuSCs are quiescent until activated in response to a growth or injury signal and then transition from quiescence into the myogenic program to proliferate and repair the muscle (for review, see Ref. 157). Decreases in MuSC number, proliferation, or activation hamper the muscle regeneration process leading to a loss in muscle mass (156, 158).

In humans and mice, there is recent evidence that a deficiency of ovarian hormones is related to a decline in the number of MuSCs (82). Both ERα and ERβ have been detected in MuSC (82, 83). The loss of ERα in MuSC results in upregulation of apoptotic genes (e.g., p53, p38, and DIABLO) and a reduction in MuSC population of up to 56% (82). The deletion of ERβ in MuSC does not alter the number of MuSC per muscle fiber but rather results in a suppression of MuSC proliferation and impaired expression of cyclinA2 (Ccna2), a cell-cycle regulator (83). Taken together, estrogens can impact the muscle regeneration process via receptor-mediated mechanisms within MuSCs.

In female mice lacking estrogens, the loss of MuSC is detrimental to muscle mass adaptation following repeat injuries (158), suggesting that effect of E₂ on MuSC is beneficial for muscle mass adaptation. In addition to estrogen cell-autonomous mechanisms, estrogen molecules are a part of and have been implicated in modulating the MuSC microenvironment known as the niche (82, 159–161). The importance of estrogens in the MuSC niche is demonstrated via in vivo transplantation studies whereby injured muscle is regenerated from donor MuSCs. Specifically, MuSC harvested from control female mice transplanted into an E₂-depleted environment (injured muscle in an OVX mouse) had cell engraftment reduced 75% compared with those transplanted into an E₂-repleted environment indicating incomplete regeneration (82). Moreover, the transfer of MuSC harvested from an E₂-depleted environment (OVX mouse) into an E₂-present environment rescued MuSC engraftment (82). The MuSC niche consists of extracellular matrix, vascular and neural networks, immune cells, and circulating hormones (155, 161). In response to injury, E₂ has been implicated in altering extracellular matrix (162), vascular (163) and neural (164) networks, and immune cells (165). However, how E₂ modulates the MuSC niche and how alterations in the niche contribute to the maintenance or adaptation of muscle mass need further investigation. Furthermore, to what extent estrogenic mechanisms in MuSC are the same or differ from that of muscle fibers is not clear, though both likely impact the mass of muscle.

E₂-RELATED THERAPEUTIC INTERVENTIONS, CONSIDERATIONS, AND ALTERNATIVES

Evidence is beginning to accumulate that E₂ may be an important factor to consider in the treatment of muscle wasting. Although interrogation of mechanisms regulating muscle mass by E₂ continues, there are several interventions to consider including elevating E₂ signaling using hormone replacement therapy (HRT) and selective estrogen receptor modulators (SERMs) or through the reduction in E₂ levels with aromatase inhibition. Here we will give an overview of these estrogenic interventions.

Estrogen-Based Hormone Replacement Therapy/Treatments

Estradiol is the most common form of HRT for the treatment of menopause and is prescribed via several different modes of administration (e.g., tablets, injections, and topical creams) and dosages (88, 166). The use of HRT has been reported to increase the levels of circulating estrogens. However, the physiological implications for the mode of administration and dosage for clinical benefits still need clarification (88, 167, 168). Healthy pre- and postmenopausal women treated with a transdermal E₂ patch (0.1 mg E₂ per day) for 14 days had an ∼1.3-fold increase in the ratio of E₂-to-estrone in their serum yet had no effect on protein fractional synthesis rate or expression of genes involved in the regulation of muscle mass (e.g., MYOD1, FST, MSTN, and FOXO3) (169). The use of a conjugated equine estrogen of 0.625 mg daily for 1 yr has been shown to elevate E₂ serum levels ∼2.7-fold in healthy postmenopausal women compared with women who received placebo (170). A recent meta-analysis investigating the effect of HRT on muscle mass demonstrated that a HRT of 0.625 mg of E₂ in women >50 yr was associated with ∼0.06–0.2 kg greater muscle mass retention compared with the postmenopausal control group not receiving HRT (171). Although the analysis did not prove statistically significant, it does point to the potential benefits of E₂-based HRT on maintaining muscle mass in healthy postmenopausal women. In another study, healthy pre- and postmenopausal women treated with a transdermal E₂ patch (0.1 mg E₂ per day) for 14 days had an ∼1.3-fold increase in the ratio of E₂-to-estrone in their serum; yet there was no effect on protein fractional synthesis rate or expression of genes involved in the regulation of muscle mass (e.g., MYOD1, FST, MSTN, and FOXO3) (169). Thus, more clinical studies are needed to determine if HRT effects on muscle mass would be beneficial during conditions of muscle wasting including but not restricted to menopause.

Few studies have examined the effect of E₂ treatment on skeletal muscle mass in ovarian hormone deficient rodents. Experimental variables such as age of rodents and E₂ method of administration, dosage, and duration of treatment differ greatly among studies and complicate interpretation of muscle mass data. For example, reported E₂ dosages in mice range from 0.02 µg/day to 20 mg/day (172–174). Dosing is critical because resulting serum levels can be nonphysiological and not comparable with ovary-intact, control animals. As an example, an E₂ dose of ∼120 µg/day raised serum E₂ levels ∼88-fold in OVX mice over ovary-intact mice (174). In another study, a dose of ∼4.2 µg/day raised serum E₂ levels ∼4-fold in OVX mice and resulted in an increase in uterine mass beyond physiological levels as well as adverse effects on kidney physiology (172). Taken together, these studies suggest that an excessive dose of E₂ can limit the risk-benefit ratio and complicates the interpretation of E₂ effects in preclinical models.

Following ovariectomy in adult female mice, there is an increase skeletal muscle mass due to an increase in collagen and/or nonprotein content (99, 175–177). This increase in muscle weight gain by nonprotein content is attenuated with E₂ treatment of ∼3 µg/day that yielded E₂ serum levels ∼2- to 4-fold greater in OVX than ovary-intact mice (175, 178). The duration of ovarian hormone deficiency and that of E₂ treatment in these mouse studies are relatively short, on the order of 2–12 wk, and may demonstrate transient effects on muscle mass. Little is known on how long-term OVX with or without E₂ treatment affects muscle mass in preclinical animal models. In summary, although estrogen-based HRT is clinically relevant for women, the effects and mechanisms of any effects on skeletal muscle mass are not clear and further research is warranted.

Selective Estrogen Receptor Modulators

Selective estrogen receptor modulators (SERMs) have emerged as alternative therapeutic agents that show promise in treating muscle wasting associated with estrogen deficiency because of their more favorable side effect profile than estrogens (179). SERMs are a class of ER ligands that act as tissue- and cell-specific estrogen agonists/antagonists via conformational changes of ERs that are induced on ligand binding (180). Mechanisms of action of SERMs are reliant on different cell signaling properties and ER cofactors, which result ultimately in several physiological effects that are tissue dependent. Examples of SERMs commonly used in the clinic include tamoxifen (TAM; first-generation SERM), raloxifene (RAL; second-generation SERM), and bazedoxifene (BZA; third-generation SERM). Specifically, TAM and RAL have been shown to improve numerous pathogenic factors involved in mouse models of myopathy (181–183), which may hold implications in ameliorating declines in skeletal muscle mass. However, evidence surrounding the direct effects of SERMs on skeletal muscle in estrogen-deficient rodents and humans remains scarce. Thus, further research is necessary to determine the efficacy of this drug class in the treatment of muscle wasting.

Aromatase Inhibition

Aromatase inhibitor therapy is a common adjuvant therapy for patients with breast cancer and postmenopausal breast cancer survivors to reduced E₂ levels (184, 185). Breast cancer patients who underwent aromatase inhibition demonstrated maintenance of total body fat and an increase in lean body mass and free testosterone levels compared with no-aromatase inhibition controls (185). Similarly, an increase in body mass and an increase in testosterone are observed in rats treated with an aromatase inhibitor (186) and in aromatase knockout mice (187, 188). Collectively, muscle mass may be maintained while taking aromatase inhibitors because of the rise in free testosterone (185) and therefore should be a consideration if using aromatase inhibitor in the study of estrogenic effects.

SUMMARY

The purpose of the review is to summarize the current understanding of E₂’s impact on skeletal muscle wasting and highlight aspects of sex dimorphism on the topic. Our review of the literature has elucidated that the field needs additional studies to directly assess effects of E₂ on the susceptibility of skeletal muscle atrophy or wasting. Several of our interpretations of the literature require some form of extrapolation or assumption regarding the muscle wasting conditions that we report on particularly when considering the estrogenic mechanisms that may be involved across the several models used to study wasting. That is, although sex differences may suggest that estrogen is involved, there are clearly many other aspects that differ between females and males than just sex hormones. Ultimately, determining if sex-based differences exist and the mechanisms that drive them will be necessary as the field pushes toward developing approaches to prevent, attenuate, and ideally reverse muscle wasting.

GRANTS

This study was funded by National Institutes of Health Grants R61AR078100 and RO1AR066660 (to E.E.S.), RO1AG031743 and RO1AG062899 (to D.A.L.), and T32AR007612 (to S.L.M.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

S.L.M. and E.C.M. prepared figures; S.L.M., E.C.M., D.A.L., and E.E.S. drafted manuscript; S.L.M., E.C.M., D.A.L., and E.E.S. edited and revised manuscript; S.L.M., E.C.M., D.A.L., and E.E.S. approved final version of manuscript.