Reduced testosterone as a result of catabolic illness or aging is associated with loss of muscle and increased adiposity. We hypothesized that these changes in body composition occur because of altered rates of protein synthesis under basal and nutrient-stimulated conditions that are tissue specific. The present study investigated such mechanisms in castrated male rats (75% reduction in testosterone) with demonstrated glucose intolerance. Over 9 wk, castration impaired body weight gain, which resulted from a reduced lean body mass and preferential sparing of adipose tissue. Castration decreased gastrocnemius weight, but this atrophy was not associated with reduced basal muscle protein synthesis or differences in plasma IGF-I, insulin, or individual amino acids. However, oral leucine failed to normally stimulate muscle protein synthesis in castrated rats. In addition, castration-induced atrophy was associated with increased 3-methylhistidine excretion and in vitro-determined ubiquitin proteasome activity in skeletal muscle, changes that were associated with decreased atrogin-1 or MuRF1 mRNA expression. Castration decreased heart and kidney weight without reducing protein synthesis and did not alter either cardiac output or glomerular filtration. In contradistinction, the weight of the retroperitoneal fat depot was increased in castrated rats. This increase was associated with an elevated rate of basal protein synthesis, which was unresponsive to leucine stimulation. Castration also decreased whole body fat oxidation. Castration increased TNFα, IL-1α, IL-6, and NOS2 mRNA in fat but not muscle. In summary, the castration-induced muscle wasting results from an increased muscle protein breakdown and the inability of leucine to stimulate protein synthesis, whereas the expansion of the retroperitoneal fat depot appears mediated in part by an increased basal rate of protein synthesis-associated increased inflammatory cytokine expression.
a decreased circulating concentration of testosterone in males is a consequence of normal aging but also occurs in response to various catabolic and stress conditions (5). This hypogonadal state in otherwise healthy males alters body composition, including decreases in bone and skeletal muscle mass (9, 25, 40) as well as an increase in fat mass (11, 25, 26). As might be expected, treatment of hypogonadal males with replacement doses of testosterone largely reverses these metabolic changes (10, 11, 44). For example, testosterone treatment of hypogonadism increases lean body mass (LBM) in part by stimulating muscle protein synthesis, thereby increasing muscle cross-sectional area and strength (11). Furthermore, the metabolic actions of testosterone on LBM and fat mass appear to dose dependent (4, 52). The effect of testosterone on muscle protein breakdown is more variable and may be dependent upon the prevailing basal testosterone concentrations (e.g., hypo- vs. eugonadal), the dose of testosterone administered (e.g., replacement vs. pharmacological), and whether whole body or muscle-specific protein breakdown is assessed (15, 16, 40, 61). In contrast to skeletal muscle, there are few data pertaining to the effect of testosterone on the rate of protein synthesis in other tissues, such as heart and fat.
The effect of testosterone on insulin action in muscle can be described as an “inverted U-shaped” dose-response curve (24), with impaired insulin action and glucose intolerance noted in patients with endogenous hyperandrogenicity or in response to pharmacological doses of testosterone (14, 48). Female rats treated with testosterone are also insulin resistant (23). However, hypogonadism per se also produces insulin resistance that can be reversed by replacement doses of testosterone (22, 45, 60). In part, replacement testosterone improves insulin action via its ability to prevent the gain in visceral adipose tissue (1).
In contrast to the well-documented muscle insulin resistance, little attention has been paid to the responsiveness of muscle to amino acids in hypogonadal males. The availability of amino acids, especially leucine, represents an important nutritional signal and stimulates muscle protein synthesis (27). However, other catabolic conditions also impair the ability of leucine to stimulate muscle protein synthesis (33, 36, 47), which thereby contributes to the atrophic response. Whether hypogonadism produces such a muscle “leucine resistance” has not been investigated. Therefore, the present study sought to determine 1) whether basal protein synthesis was altered in a coordinated manner in selected tissues (e.g., skeletal muscle, heart, liver, kidney, and adipose tissue) and 2) whether the ability of leucine to stimulate protein synthesis in these tissues was impaired. To address these aims, the current study used hypogonadal male rats, which manifested muscle atrophy, preferential sparing of whole body fat, and glucose intolerance.
MATERIALS AND METHODS
Postpubertal male Sprague-Dawley rats (8 wk of age; Charles River Breeding Laboratories, Cambridge, MA) were castrated or underwent sham surgery. Animals received bilateral orchiectomy via scrotal incision or sham surgery under sterile conditions. Rats were housed individually in a light-controlled room (12:12-h light-dark cycle) under constant environmental conditions. Standard rat chow (Harlan Teklad no. 2018), containing ∼18% protein, 6% fat, and 3.8% fiber, and water were provided ad libitum. All experiments were approved by the Institutional Animal Care and Use Committee of The Pennsylvania State University College of Medicine and adhere to the National Institutes of Health (NIH) guidelines for the use of experimental animals.
After a 2-wk surgical recovery period, food consumption, body weight (BW), and body composition were determined weekly in all rats for the ∼9-wk experimental protocol. This time course is comparable with other rodent studies where an atrophic response to castration has been demonstrated (10). Body composition, i.e., body fat and LBM, was measured using 1H nuclear magnetic resonance (NMR) (LF90 Minispec; Bruker, The Woodlands, TX) (46). An oral glucose tolerance test (OGTT) was performed at the end of the 7th week on overnight-fasted rats. Rats were gavaged with 30% glucose at 1.5 g/kg BW, and the blood glucose concentration was determined at times 0 (basal), 30, 60, 90, and 120 min after glucose. These samples were also analyzed for insulin. The area under the curve (AUC) for the dynamic glucose and insulin response produced by the OGTT was calculated using the trapezoidal rule (SigmaPlot; Systat Software, San Jose, CA). Two days thereafter, an insulin tolerance test (ITT) was performed between 0900 and 1200 on fasted animals. Human recombinant insulin (Humulin R; Eli Lilly, Indianapolis, IN) was injected intraperitoneally into conscious rats (0.75 U/kg BW), and blood samples were collected from the tail vein at 0, 15, 30, 60, 90, and 120 min.
At the start of the final week of the study, rats were housed individually in metabolic cages and acclimated for 24 h. During the subsequent 24-h period, urine was collected and an aliquot used for HPLC determination of 3-methylhistidine (3-MH) and data normalized to the urinary creatinine concentration (46). 3-methylhistidine is formed by posttranslation modification of histidine released from actin and myosin and provides an in vivo estimate of myofibrillar protein breakdown (42).
Thereafter, whole body carbon dioxide (CO2) production and oxygen (O22 consumed. Rats were acclimated to the cages used for indirect calorimetry for 24 h prior to the start of data collection, which was collected continuously for the subsequent 24-h period. Food consumption was monitored during the 24-h test period but not during the prior acclimation period.
On the morning of the final day, heart function and in vivo tissue protein synthesis were determined, as described below. Both the control and castrated groups were randomly divided, and half of the rats in each group orally gavaged with saline (0.155 mol/l) or leucine (1.35 g/kg BW, prepared as 54.0 g/l of l-amino acid in distilled water) (2, 3, 33, 34, 36). This leucine dose was selected because it is equivalent to that consumed daily by rats (2). Saline was administered to time-matched control rats, because previous studies indicated that isonitrogenous administration of nonbranched chain amino acids (e.g., alanine) failed to stimulate either protein synthesis or peptide-chain initiation (2). Thereafter, rats were lightly anesthetized by the intraperitoneal injection of the combination of ketamine (40 mg/kg) and acepromazine (1 mg/kg) and M-mode echocardiography performed (Sequoia C256; Siemens Medical Solutions, Mountain View, CA), equipped with a 7.5-MHz transducer (46). All data were collected and analyzed by a single investigator in a blinded manner. During the procedure, which lasted <5 min, body temperature was maintained by a warming pad.
Tissue protein synthesis.
After completion of echocardiography, rats were provided supplemental anesthesia (pentobarbital sodium, 50 mg/kg ip), and a catheter was inserted into the carotid artery. At time 0 an arterial blood sample was collected (1 ml), and exactly 20 min after the oral gavage of saline or leucine described above, rats were injected with l-[2,3,4,5,6-3H]phenylalanine (Phe; 150 mM, 30 Ci/ml, 1 ml/100 g BW). Radiolabeled isotope was delivered via percutaneous puncture of the jugular vein, and serial arterial blood samples were collected into heparinized syringes at 2, 6, and 10 min for measurement of plasma Phe-specific radioactivity. After the final blood sample (i.e., 30 min after oral leucine), the gastrocnemius, heart, liver, kidney, and fat (retroperitoneal depot) were excised and frozen with liquid nitrogen-cooled clamps. This time point was selected because muscle protein synthesis is maximally stimulated by leucine between 30 and 45 min (3). All tissue and plasma samples were stored at −80°C until they were analyzed. Tissues were powdered under liquid nitrogen, and a portion was used to estimate the incorporation of [3H]Phe into protein, exactly as described (57). The total RNA was measured from homogenates of tissue samples (56). Translational efficiency was calculated as the rate of protein synthesis (nmol Phe incorporated·mg protein−1·h−1) divided by the total RNA content (μg RNA/mg protein).
Time 0 (e.g., prior to injection of radiolabeled Phe) blood samples were used to determine insulin (Linco Research, St. Charles, MO) and total IGF-I (Immunodiagnostic Systems, Fountain Hills, AZ), as well as testosterone (R & D Systems, Minneapolis, MN), by enzyme-linked immunosorbent assay. Plasma amino acid concentrations were determined using reverse-phase HPLC after precolumn derivatization of amino acids with phenylisothiocyanate. Glucose was determined using an Analox analyzer (Lunenburg, MA). The plasma and urinary concentrations of blood urea nitrogen (BUN) were determined using the Vitros Chemistry System (DT60 II and DTSCII; Ortho-Clinical Diagnostics, Rochester, NY).
Western blot analysis.
Fresh tissue was homogenized in ice-cold homogenization buffer [20 mM HEPES (pH 7.4), 2 mM EGTA, 50 mM NaF, 100 mM KCl, 0.2 mM EDTA, 50 mM β-glycerophosphate, 1 mM DTT, 0.1 mM PMSF, 1 mM benzamidine, 0.5 mM sodium vanadate] exactly as described (33, 35, 36, 47). Equal amounts of protein per sample were subjected to standard SDS-PAGE for total and phosphorylated ribosomal protein S6 kinase 1 (S6K1) (Thr389; Cell Signaling Technology, Beverly, MA), ribosomal protein S6 (Ser235/Ser236; Cell Signaling Technololgy), and eukaryotic initiation factor (eIF) 4E-binding protein-1 (4E-BP1) (Thr37/46; Bethyl Laboratories, Montgomery, TX) (33, 36, 46, 47). The blots were washed with TBS-T (1× TBS, including 0.1% Tween-20) and incubated with secondary antibody (horseradish peroxidase-conjugated goat anti-mouse or goat anti-rabbit) at room temperature. Blots were developed with enhanced chemiluminescence Western blotting reagents as per the manufacturer's (Amersham) instructions and then exposed to X-ray film in a cassette equipped with a DuPont Lightning Plus intensifying screen. After development, the film was scanned (Microtek ScanMaker IV) and analyzed using NIH Image 1.6 software.
From an aliquot of supernatant, eIF4E was immunoprecipitated using an anti-eIF4E monoclonal antibody (kindly provided by Drs. L. Jefferson and S. Kimball, Hershey, PA), and the eIF4E·4E-BP1 and eIF4E·eIF4G complexes were quantified as described (33, 36, 46, 47).
Nuclease protection assay (RPA).
Primer selection for rat genes of interest [i.e., atrogin-1, MuRF1 (muscle RING finger 1), IGF-I (insulin-like growth factor), and various IGF-binding proteins as well as tumor necrosis factor (TNF)α, interleukin (IL)-1α, IL-6, and nitric oxide synthase-2 (NOS2)] has previously been published by our laboratory (39, 43, 46). An aliquot of template was labeled using T7 Polymerase, RNaisin, and DNase (Promega, Madison, WI), NTPs and tRNA (Sigma), and [32P]UTP (Amersham, Piscataway, NJ). The entire RPA procedure, including labeling conditions, component concentrations, sample preparation, and gel electrophoresis, was performed according to previously published protocols (BD Biosciences Pharmingen, San Diego, CA). Polyacrylamide gels were run, transferred to chromatography paper, and dried (FB GD 45 Gel Dryer; Fisher Scientific). Gels were exposed to a PhosphorImager screen (Molecular Dynamics, Sunnyvale, CA), and the resultant data were quantified using Molecular Dynamics' ImageQuant software (version 5.2). Data were normalized to L32 or GAPDH, and the data were qualitatively comparable (data not shown).
Proteasome activity was assessed by quantifying the chymotryptic-like peptidase activity in gastrocnemius. Muscle was homogenized in buffer containing 50 μM Tris·HCl (pH 7.4), 5 mM MgCl2, 250 mM sucrose, 2 mM ATP, and 1 mM DTT. The homogenate was clarified by sequential centrifugation steps to isolate the 20S and 26S proteasomes. Proteasome chymotryptic-like activity was determined as the release of 7-amino-4-methylcoumarin (AMC) from the fluorogenic peptide substrate LLVY-AMC (58).
Data are summarized as means ± SE, with the sample size indicated in the legends to each figure and table. Data were analyzed by a two-tailed, unpaired Student's t-test (2-group comparison) or two-way ANOVA with post hoc Student-Neuman-Keuls test (4-group comparison). Differences were considered significant when P < 0.05.
Body weight and composition, food consumption, and energy expenditure.
Measured within 7–10 days of castration, there was no significant difference in the body weight, LBM, or fat mass between control and castrated male rats (Fig. 1). However, by the 5-wk time point, body weight was decreased in castrated rats compared with time-matched control values, and this reduction persisted throughout the remainder of the protocol (Fig. 1A). The decrease in total body weight in castrated rats results largely from a reduction in LBM and not fat mass, as determined by 1H-NMR (Fig. 1, B and C, respectively). Over the 9-wk duration of the study, the increment in body weight was reduced 26% by castration (controls = 241.7 ± 5.4 vs. castrated = 178.3 ± 7.5 g/rat, P < 0.05), as was the increment in LBM (144 ± 4 vs. 104 ± 5 g/rat, P < 0.05). However, the increment in body fat mass was not different between the groups [66 ± 3 vs. 63 ± 3 g/rat, P = not significant (NS)].
Control rats consumed an average of 29.2 ± 1.3 g of chow/rat during the final 5 wk of the experiment. This time range was selected because it corresponded to when differences in body weight and LBM were most pronounced. In contrast, castrated rats consumed less food than time-matched control rats (26.1 ± 1.2 g, P < 0.05). However, both groups consumed essentially the same amount of food per day when normalized to body weight (controls = 5.9 ± 0.2 vs. castrated = 5.9 ± 0.3 g food·day−1·100 g BW−1).
Energy expenditure in the light and the dark phase was determined by indirect calorimetry. Although both groups had an ∼25% greater energy expenditure in the dark vs. light cycle, castration did not significantly alter energy expenditure during either the light (controls = 118 ± 14 vs. castrated = 119 ± 12 J·min−1·100 g LBM−1, P = NS) or dark (147 ± 19 vs. 154 ± 19 J·min−1·100 g LBM−1, P = NS) phase. Similar conclusions were reached whether energy expenditure was normalized to LBM or total body weight (data not shown). The RER was determined every 15 min from ∼1100 to noon the following day. RER during the light period (e.g., fasting) was increased in castrated compared with control rats (control = 0.78 ± 0.01 vs. castrated = 0.81 ± 0.01, P < 0.05). In contrast, RER during the dark period (e.g., food freely available) was not different between the two groups (0.89 ± 0.01 and 0.89 ± 0.01). Food consumption by the two groups of rats during the 24-h period for which energy expenditure was determined was not different from that determined in their standard wire-bottom cages (data not shown). Body temperature did not differ between the control and castrated rats (controls = 37.9 ± 0.1 vs. castrated 38.0 ± 0.2°C) and therefore cannot explain group differences in RER.
The wet weights of the gastrocnemius and soleus were decreased by 10% in castrated rats compared with time-matched control values (Table 1). The dry-to-wet weight ratio for the gastrocnemius was not different between the two groups (data not shown). Cardiac atrophy was also detected, with the heart weight being reduced 14% in castrated rats. These decreases in striated muscle mass in castrated rats were proportional to the above-mentioned decreases in body weight and LBM, and thus no differences between groups were detected when muscle weights were normalized to body weight (Table 1). Castration decreased kidney weight by 25%, and the reduction was still evident when normalized to the lower body weight in castrated rats (Table 1). It is noteworthy that the castration-induced reduction in kidney mass was not associated with a detectable change in either the BUN (Table 2) or the creatinine clearance between groups (controls = 447 ± 38 vs. castrated = 426 ± 41 μl/min, P = NS). The weight of the liver did not differ between the control and castrated groups. Finally, the weight of the retroperitoneal adipose depot was increased 45% in castrated rats compared with control values. Because of the increased weight of this tissue in rats with lower body weights, the relative percentage of fat was further increased when data were normalized to body weight (Table 1).
Glucose and insulin tolerance.
To confirm that castrated rats had metabolic changes comparable with those reported previously, OGTTs were performed on a representative cohort of both control and castrated rats at the end of the 7-wk time point (Fig. 2). There was no difference in the basal glucose concentration between groups. In both groups, the blood glucose peaked at the 30-min point and thereafter returned toward basal levels (Fig. 2A). However, the blood glucose was greater in castrated rats at each time point. The glucose AUC after the OGTT was increased 30% in castrated rats (332 ± 24 vs. 257 ± 21 mg·h−1·dl−1, P < 0.05). The glucose-stimulated insulin release in castrated rats tended to be elevated throughout the 120-min sampling period, but only two time points achieved statistical significance (Fig. 2B). Despite the impaired glucose tolerance in castrated rats, the ITT did not reveal any difference in the ability of insulin to decrease plasma glucose (Fig. 2C).
Plasma concentrations of substrates and hormones.
The plasma testosterone concentration in castrated rats was reduced by 75% under basal fasted conditions compared with control values (Table 2). Oral administration of leucine to control rats unexpectedly increased the testosterone by 60%; such an elevation was not seen in castrated animals. There was no difference in the plasma concentration of individual amino acids between control and castrated rats under basal fasted conditions (Table 3). The oral gavage of leucine increased the plasma leucine concentration in both groups to a comparable extent (6- to 8-fold; Table 2). Basal plasma insulin concentrations were not different between groups, and the leucine gavage increased plasma insulin ∼3.5-fold in both groups (Table 2). The free fatty acid concentration in the basal fasted condition was not different between groups and was not altered by leucine administration (Table 2).
Tissue protein synthesis.
Under basal (e.g., no exogenous leucine) fasted conditions, there was no significance difference in the protein synthesis in gastrocnemius, heart, or liver obtained from control and castrated rats (Fig. 3). In contrast, castration increased basal protein synthesis in adipose tissue 70%. Although a change in the number of ribosomes may be responsible for differences in rates of protein synthesis under basal condition, there was no difference in the RNA content for gastrocnemius, heart, liver, kidney, or fat between control and castrated animals (data not shown). Oral leucine increased protein synthesis in both gastrocnemius and adipose from control rats but not in either tissue from castrated animals (Fig. 3). Leucine tended to increase myocardial protein synthesis, but the difference did not achieve statistical significance. Total rates of protein synthesis in liver and kidney were not altered by leucine in either control or castrated rats (Fig. 6).
In the current study, the phosphorylation of 4E-BP1 and ribosomal protein S6 was used as functional readouts of mammalian target of rapamycin (mTOR) kinase activity. The phosphorylation of S6 was used instead of S6K1 because phosphorylated S6 is detectable under fasted conditions in muscle, where little or no phosphorylated S6K1 is detected (30). For gastrocnemius, there was no difference in basal level of phosphorylated 4E-BP1 between control and castrated rats (Fig. 4A); however, there was a significant 40% reduction in basal S6 phosphorylation in muscle from castrated rats (Fig. 4B). Basal S6K1 phosphorylation (Thr389) in muscle was essentially undetectable in both groups (data not shown). Upon leucine stimulation, the phosphorylation of 4E-BP1 and S6 increased approximately twofold in gastrocnemius from control rats but not in muscle from castrated rats. Leucine stimulation in control rats increased the amount of the active eIF4E·eIF4G complex and decreased the amount of the inactive eIF4E·4EBP1 complex (Fig. 4, C and D). Such a redistribution of eIF4E between protein complexes was not seen in castrated rats after leucine administration.
In adipose tissue, under basal conditions, there was a discordant pattern of phosphorylation for 4E-BP1 compared with either S6K1 or S6 in castrated rats (Fig. 5). That is, under basal conditions where in vivo protein synthesis was elevated, adipose tissue from castrated rats demonstrated a twofold increase in 4E-BP1 phosphorylation but a reduction in the phosphorylation of both S6K1 and S6 (70 and 35%, respectively). S6K1 phosphorylation was assessed in fat because this tissue, as opposed to muscle, shows constitutive phosphorylation of this protein. In response to oral leucine, 4E-BP1, S6K1, and S6 phosphorylation were coordinately increased in adipose tissue from control rats. Conversely, only S6K1 and S6 phosphorylation were increased in adipose tissue of castrated rats after leucine, but this increase was less than that seen in adipose tissue from control rats.
The urinary excretion of 3-MH was increased in castrated rats by 40% compared with values in control rats (Fig. 6A). Furthermore, the 3-MH-to-creatinine ratio also tended (P = 0.07) to be elevated by a comparable amount (35%), but the difference failed to achieve statistical significance because of the larger variability (data not shown). Other end points were determined to confirm an increase in muscle proteolysis per se because a fraction of excreted 3-MH is derived from nonmuscle tissues (42). In this regard, we detected an increase in the ubiquitin proteasome pathway based on the nearly twofold increase in proteasome chymotryptic-like peptidase activity in gastrocnemius from castrated rats compared with control values (Fig. 6B). We also determined the mRNA content of the muscle-specific ubiquitin E3 ligases atrogin-1 and MuRF1, and contrary to expectation, the mRNA content for atrogin-1 (control = 1.00 ± 0.07 vs. castrated = 0.54 ± 0.06 AU/L32, P < 0.05) and MuRF1 (1.00 ± 0.06 vs. 0.58 ± 0.08 AU/L32, P < 0.05) was 40–45% lower in muscle from castrated rats (Fig. 6B, bottom).
Tissue cytokine expression.
Cytokine mRNA expression was determined because the overexpression of proinflammatory mediators is recognized as mediating at least part of the metabolic phenotype in diet-induced obesity (21). The mRNA content for the inflammatory mediators TNFα, IL-6, IL-1, and NOS2 was differentially regulated in muscle and fat in response to castration (Fig. 7). This was evidenced by the two- to fourfold increase in mediator mRNA content in adipose tissue, but not muscle, of castrated rats. Leucine did not alter the mRNA content for any of these inflammatory mediators in either tissue (data not shown).
Next, the plasma IGF-I concentration as well as the mRNA content for IGF-I and its various binding proteins were assessed within liver [e.g., primary source for circulating IGF-binding proteins (IGFBPs)] and muscle. The plasma IGF-I concentration was not different between castrated and control rats either under basal conditions or after oral leucine (Table 2). In contrast, castration selectively decreased IGF-I mRNA in muscle (−26%) but not liver compared with time-matched control values (Table 4). The content of IGFBP-3 mRNA was increased in both liver (42%) and skeletal muscle (87%) of castrated rats. No other changes were noted in IGFBP mRNA content in liver or muscle, except for the 32% reduction in IGFBP-5 mRNA observed in muscle of castrated rats. Oral leucine did not significantly alter the mRNA content of IGF-I or any binding protein in either liver or gastrocnemius (data not shown).
Cardiac structure and function.
Because castration decreased heart weight, selected aspects of myocardial function were examined in vivo by echocardiography (Table 5). Despite the 14% reduction in cardiac mass in castrated rats, there was no significant change in any cardiac end point assessed by echocardiography compared with control values. Moreover, leucine did not significantly alter any of the end points assessed (data not shown). It was particularly relevant that cardiac output was not different between the two groups under basal conditions, nor was it altered by leucine. Hence, it seems unlikely that the observed alterations in tissue metabolism were caused by overt changes in heart function in castrated rats.
The loss of muscle mass in hypogonadal males represents a mismatch between rates of muscle protein synthesis and degradation (4, 9, 10, 26, 40, 48). The ability of basal levels of testosterone (as opposed to pharmacological concentrations) to influence muscle protein synthesis has been evidenced in healthy hypogonadal males where replacement testosterone improved whole body nitrogen retention and increased the protein synthetic rate in skeletal muscle (11, 28). In contrast, although in men the acute suppression of testosterone produced by administration of a gonadotropin-releasing hormone agonist decreased whole body protein synthesis (40), we found no comparable studies where muscle protein synthesis had been directly assessed in control and otherwise healthy hypogonadal male subjects. Our current data fail to equivocally demonstrate that castration significantly reduces basal muscle protein synthesis, although a slight (P = 0.1), 9% decrease was observed. We estimate that the sample size for each group would need to be more than doubled to demonstrate that a difference of this magnitude and variance is statistically significant. Importantly, our results are consistent with one of the major mechanisms regulating protein synthesis (i.e., 4E-BP1 phosphorylation), which was also not different between the two groups under basal conditions. One caveat of the current study is that protein synthesis was determined only on the predominantly fast-twitch gastrocnemius muscle and not on slow-twitch muscles such as the soleus. However, previous reports have demonstrated that fast- and slow-twitch muscles undergo a similar degree of atrophy in castrated rats (10).
In contrast, our data unequivocally show that the ability of leucine to increase muscle protein synthesis is markedly impaired in castrated rats. The inability of leucine to increase muscle protein synthesis is mediated in part by reduced mTOR activity, as evidenced by the diminished leucine-induced phosphorylation of 4E-BP1 and S6. The leucine-induced increase in 4E-BP1 phosphorylation in muscle of control rats was consistent with the increased abundance of the active eIF4E·eIF4G complex. Conversely, muscle from castrated rats showed no leucine-induced increase in eIF4E·eIF4G, which suggests inhibition of cap-dependent mRNA translation (27). This differential effect of leucine on muscle protein synthesis, 4E-BP1, and eIF4E·eIF4G could not be explained by differences in the prevailing plasma concentrations of either leucine or insulin. Furthermore, based on the relatively normal ITT, it seems unlikely that the muscle leucine resistance is a secondary consequence of peripheral insulin resistance. Hence, we posit that this muscle leucine resistance limits the normal anabolic response to dietary amino acid stimulation and thereby blunts developmental accretion of muscle mass and body weight in castrated rats. Although excess glucocorticoids can produce muscle leucine resistance (50), this mechanism does not seem applicable in castrated rats because studies report little alteration in the plasma glucocorticoid concentration in hypogonadal males or in response to exogenous testosterone (60). Finally, although the muscle leucine resistance observed after bacterial infection is associated with an increase in muscle inflammatory cytokines (e.g., TNFα, IL-1α, and IL-6) (34), such a mechanism is also not operational in muscle from castrated animals.
The increased urinary excretion of 3-MH, which implies accelerated myofibrillar degradation, suggests that castration-induced atrophy is also caused in part by an enhanced rate of skeletal muscle protein degradation. Such a contention is supported by the increased ubiquitin proteasome activity observed in gastrocnemius. However, the decreased expression for both atrogin-1 and MuRF1 mRNA in muscle from castrated rats in which we detected an elevated proteasome activity was unexpected. These results contrast with those showing that testosterone decreases atrogin-1 mRNA and promoter activity in C2C12 myotubes (61). Furthermore, the enhanced expression of these two “atrogenes” has been reported in other catabolic conditions with muscle atrophy, and mice deficient in these proteins are resistant to muscle atrophy (8, 19, 29, 37). However, a dissociation of proteolysis and the mRNA expression of these E3 ligases in muscle has been reported in other catabolic conditions (55). Collectively, these data indicate that castration-induced muscle atrophy is produced in part by enhanced protein degradation via a stimulation of the ubiquitin proteasome pathway.
IGF-I is an important hormone and autocrine mediator for the accretion of muscle mass during development and its maintenance in adults (18). Although muscle mass is directly proportional to the circulating IGF-I concentration, the concentration of the hormone in muscle per se also regulates protein balance (13, 43). IGF-I is capable of affecting both sides of the protein balance equation since a reduction in IGF-I is correlated with inhibition of protein synthesis and stimulation of protein breakdown in skeletal muscle. Others have reported that the plasma concentration for both total and free IGF-I is not significantly altered in hypogonadal men or after testosterone administration (11, 54). However, muscle IGF-I mRNA and protein are reduced in hypogonadal men and castrated rats, and that replacement testosterone increases muscle IGF-I (11, 40, 44, 54). Our results, showing a decrease in muscle IGF-I mRNA in the presence of unaltered circulating IGF-I, are consistent with these previous observations. In addition, because liver is the primary site of blood-borne IGF-I, the similar hepatic IGF-I mRNA content of castrated and control rats is consistent with the unaltered plasma IGF-I in our study. The bioavailability and bioactivity of IGF-I are modulated by changes in the any of the six structurally related high-affinity binding proteins (17). In this regard, castration decreased IGFBP-5 and increased IGFBP-3 mRNA in gastrocnemius. The reduction in muscle IGFBP-5 mRNA is most likely due to the drop in muscle IGF-I (38). In contrast, elevations in IGFBP-3 decrease insulin-stimulated glucose uptake (12), and therefore, we cannot exclude the possibility that an elevation in this particular IGFBP protein may impair insulin action in skeletal muscle that was not detected by the ITT.
Castration also decreased the weight of the heart and kidney, and such changes have been reported previously (9, 10, 51). The castration-induced atrophy in these two tissues is not mediated by an altered rate of protein synthesis either under basal conditions or in response to leucine stimulation. However, it is noteworthy that for both tissues the loss of mass did not overtly impair organ function, since cardiac output and glomerular filtration rate were unaltered by castration. In contrast to the normal cardiac function determined under in vivo conditions in our castrated rats, impaired function has been unmasked using the isolated, perfused working heart preparation (49). The reason for this apparent difference is unclear but may be related to cardiac function being assessed under basal nonstressed conditions in our study, whereas under in vitro conditions the most pronounced myocardial defects were observed in response to conditions of elevated preload and afterload. Differences between in vivo and in vitro estimates of cardiac performance have previously been reported in other catabolic conditions (31, 41). Regardless, the preservation of apparently normal cardiac output suggests that the observed metabolic derangements are unlikely to be attributable to differences in blood flow and nutrient delivery to individual organs between castrated and control rats.
Previous studies have reported an increased weight of various adipose tissue depots and whole body fat mass in castrated male rats and hypogonadal men (10, 20, 25) and that testosterone replacement ameliorates the gain in visceral adipose tissue (1). In the current study, castration led to a relative increase in whole body adiposity and an absolute increase in the mass of the retroperitoneal fat depot. These changes were independent of an increased food intake. In contradistinction to other tissues, the in vivo-determined rate of protein synthesis in the retroperitoneal fat depot was increased in castrated rats under basal fasted conditions. This increased protein synthesis was associated with a selective increase in 4E-BP1 phosphorylation but not the phosphorylation of either S6K1 or S6. Because the phosphorylation of both 4E-BP1 and S6K1 is usually attributed to mTOR kinase activity, the mechanism for this differential response in adipose tissue is unclear. Additionally, although oral leucine increased protein synthesis in adipose tissue from control rats, it was unable to further augment protein synthesis in fat from castrated animals. These changes in protein synthesis were associated with an increased inflammatory tone within the fat depot, as evidenced by increased expression of TNFα, IL-1α, IL-6, and NOS2. Such an enhanced inflammatory state in adipose tissue has been observed in diet-induced obesity (53). Moreover, indirect calorimetry indicated that whole body lipid oxidation was decreased in castrated animals, suggesting an increased storage of lipids in these rats. However, these changes were independent of a change in the prevailing free fatty acid concentration. These latter data are consistent with the work of Birzniece et al. (6), which demonstrated that testosterone increased whole body fat oxidation in hypogonadal males. Also, our findings are consistent with the normal inhibitory actions of testosterone on adipogenesis and lipolysis (7, 59). Hence, depletion of testosterone would be expected to increase adipogenesis and lipid storage, with a compensatory increase in protein synthesis within this tissue under basal but not nutrient-stimulated conditions.
In summary, the castration of male rats is associated with changes in body composition that are exemplified by the loss of muscle mass and the relative or absolute gain in adipose tissue. The erosion of skeletal muscle results from both an increase in protein degradation and the inability of leucine to stimulate protein synthesis. In contradistinction, the increased adiposity in castrated rats results in part from an increase in the basal rate of protein synthesis that is associated with the upregulated expression of multiple inflammatory cytokines.
This work was supported in part by NIH Grants GM-38032, GM-39277, AA-11290, AA-18290, and DK-072909.
We thank Gina Deiter and Rachel Lantry for their technical assistance.
- Copyright © 2009 the American Physiological Society