The age-related decline in fat-free mass is accelerated in women after menopause. The role of ovarian hormone deficiency in the regulation of fat-free mass, however, has not been clearly defined. To address this question, we examined the effect of ovarian hormone suppression on whole body protein metabolism. Whole body protein breakdown, oxidation, and synthesis were measured using [13C]leucine in young, healthy women with regular menstrual patterns before and after 2 mo of treatment with gonadotropin-releasing hormone agonist (GnRHa; n = 6) or placebo (n = 7). Protein metabolism was measured under postabsorptive and euglycemic-hyperinsulinemic-hyperaminoacidemic conditions. Ovarian suppression did not alter whole body or regional fat-free mass or adiposity. In the postabsorptive state, GnRHa administration was associated with reductions in protein breakdown and synthesis (P < 0.05), whereas no change in protein oxidation was noted. Under euglycemic-hyperinsulinemic-hyperaminoacidemic conditions, a similar reduction (P < 0.05) in protein synthesis and breakdown was noted, whereas, protein oxidation increased (P < 0.05) in the placebo group. Testosterone, steroid hormone precursors, insulin-like growth factor I, and their respective binding proteins were not altered by GnRHa administration, and changes in these hormones over time were not associated with GnRHa-induced alterations in protein metabolism, suggesting that changes in protein turnover are not due to an effect of ovarian suppression on other endocrine systems. Our findings provide evidence that endogenous ovarian hormones participate in the regulation of protein turnover in women.
- fat-free mass
aging is characterized by an erosion of fat-free tissue mass (6, 8). In women, this decline is accelerated during middle age, near the time of the menopause, and continues at an increased rate during the postmenopausal years (3, 8, 37, 44). The loss of fat-free mass contributes to reduced physical function (5, 13), which can increase the risk for falls and fractures and the subsequent development of disability (19). The mechanisms underlying this loss of fat-free mass, however, have not been clearly defined.
Cessation of ovarian function is the hallmark of menopause. Thus it is logical to hypothesize a role for ovarian hormones in the regulation of fat-free mass. To address this hypothesis, some investigators have measured changes in fat-free mass in postmenopausal women in response to pharmacological ovarian hormone replacement. Although these studies generally show little or no effect (4, 14, 15, 33, 34), this may be due to the fact that ovarian hormone replacement regimens do not recreate the endogenous hormonal milieu with respect to the type of hormone(s), the cyclicity of release, or the anatomic site of production. An additional important limitation to these studies is the difficulty in measuring changes in fat-free mass over time in humans. This methodological problem is particularly relevant to studies that manipulate ovarian hormone levels, given their role in regulating fluid homeostasis (36) and the fact that body water is included in estimates of fat-free mass. That is, changes in body water content associated with altered ovarian hormone levels can obscure changes in fat-free tissue mass. Because of these experimental and methodological limitations, the role of ovarian hormones in the regulation of fat-free mass in humans has not been resolved.
Protein is the primary structural and functional component of lean tissues. Changes in fat-free mass, therefore, are dictated by alterations in protein balance. In this context, a clearer understanding of the role that ovarian hormones play in the regulation of fat-free mass may be gained by examining their relationship to the metabolic processes controlling protein balance. To this end, the goal of the present study was to assess the effect of ovarian hormone deficiency on whole body protein metabolism. To accomplish this objective, we measured protein metabolism using [13C]leucine in young, healthy women during both the follicular and luteal phases of the menstrual cycle at baseline and directly after 2 mo of treatment with the gonadotropin-releasing hormone agonist (GnRHa) leuprolide acetate or placebo. GnRHa administration causes an abrupt reduction in circulating ovarian hormone concentrations to menopausal levels (17). Thus it provides an experimental model that permits evaluation of the physiological role of endogenous ovarian hormones in the regulation of protein metabolism. We chose to study healthy, normal cycling, young adult women (21–35 yr), rather than women near the age of menopause, to eliminate the possibility that natural fluctuations in ovarian hormone production that precede menopause (29) could confound our results. Measurements of protein metabolism were performed in the postabsorptive state and during euglycemia-hyperinsulinemia-hyperaminoacidemia. These conditions were chosen to simulate the catabolic and anabolic stimuli present during periods of fasting and feeding, respectively. On the basis of studies by our laboratory and others (22, 41) showing positive relationships between ovarian hormones and protein turnover, we hypothesized that ovarian hormone deficiency would reduce rates of protein turnover.
MATERIALS AND METHODS
l-[1-13C]leucine (99% 13C) and sodium [13C]bicarbonate (99% 13C) were obtained from Cambridge Isotope Laboratories (Andover, MA), and l-[2H7]leucine (99%) was obtained from MSD Isotopes (St. Louis, MO; no longer in business). Leuprolide acetate (Lupron Depot; 3.75 mg) was obtained from Tap Pharmaceuticals (Lake Forest, IL).
Thirteen healthy young women ranging in age from 21 to 35 yr (mean ± SE: 27 ± 1 yr) were recruited. Women were nonobese (body mass index <28 kg/m2; 22 ± 1 kg/m2), had stable body weights (±2 kg) for 6 mo before study, were healthy based on medical history, physical exam, and routine blood tests, were glucose tolerant (glucose <7.77 mM 2 h after 75 g of oral glucose load), had no history of tobacco use, and were not on any medication that could affect protein metabolism or ovarian/reproductive function. All women were nulliparous, had not been exposed to any form of hormone-based contraceptive therapy for at least 6 mo before study, and reported having at least two spontaneous cycles in the 3 mo before recruitment. The nature, purpose, and possible risks of the study were explained to each subject before written consent to participate was given. The experimental protocol was approved by the Committee on Human Research at the University of Vermont.
Each volunteer underwent an outpatient screening visit at which time medical history, physical examination, biochemical laboratory tests, an exercise stress test, and an oral glucose tolerance test were performed. Volunteers who met the eligibility criteria were randomized by the study coordinator with the use of a stratified (age and body mass index) block approach to receive the GnRHa leuprolide acetate (n = 6; Lupron Depot, 3.75 mg im) or placebo (n = 7; 0.9% saline). All personnel performing metabolic testing on volunteers were blinded to treatment status. Before study, each volunteer's menstrual cycle was monitored for at least two cycles with the use of menstrual diaries, ovulation prediction kits (Ovu-Quick One-Step; Quidel, San Diego, CA), and mid-luteal phase blood draws. Menstrual cycle length ranged from 25 to 32 days (29 ± 1 days).
Each woman was tested on two occasions before treatment: during the early to mid follicular phase (cycle days 3 to 8) and mid-luteal phase (cycle days 19 to 25). The order of pretreatment metabolic testing with respect to cycle phase (follicular-luteal or luteal-follicular) was randomized. After pretreatment testing, GnRHa or placebo was administered by intramuscular injection during the mid-luteal phase. The second injection was given 30 days after the first injection in the GnRHa group and 25–32 days after the first injection in the placebo group, depending on the volunteer's menstrual cycle length. Posttreatment metabolic testing was performed 60 days after the first injection in the GnRHa group. Posttreatment testing in the placebo group was performed in the phase of menstrual cycle that was tested second during pretreatment testing ∼50–64 days after the first injection. Directly preceding each bout of metabolic testing, volunteers were provided 3 days of a weight-maintenance, standardized diet (20% protein, 25% fat, and 55% carbohydrate). The diet was designed to provide at least 1 g of protein per kilogram of body weight and 200 g of carbohydrate per day.
Protein metabolism measurements were performed under basal, postabsorptive conditions and during a euglycemic-hyperinsulinemic-hyperaminoacidemic clamp the morning after an overnight visit to the General Clinical Research Center. Volunteers were fasted after 1900 the evening of admission. At ∼0600, catheters were placed in an antecubital vein for infusion and retrograde in a dorsal hand vein of the contralateral arm (1). Baseline blood and breath samples were taken, and a primed (4.5 μmol/kg), continuous (4.5 μmol·kg−1·h−1) infusion of [1-13C]leucine was started and maintained for 330 min. The bicarbonate pool was primed (1.6 μmol/kg) with sodium [13C]bicarbonate. At 210 min, insulin was infused at 40 mU·m−2·min−1 to approximate postprandial insulin levels with euglycemia maintained by a variable rate infusion of 20% dextrose. In addition, a constant rate (1.5 ml·kg−1·h−1) infusion of 10% Aminosyn (Abbott Laboratories, Chicago, IL) was started at 210 min. [1-13C]Leucine was added to the Aminosyn infusate to preserve steady-state plasma leucine enrichment.
Blood and breath samples were drawn at 165, 180, 195, and 210 min for measurement of postabsorptive leucine kinetics and at 285, 300, 315, and 330 min for measurement of leucine kinetics under euglycemic-hyperinsulinemic-hyperaminoacidemic conditions. Oxygen consumption and carbon dioxide production rates were determined at 60, 165, and 300 min using the ventilated hood technique (DeltaTrac, Yorba Linda, CA). Indirect calorimetry measurements were not performed on one volunteer at the 300-min time point because of technical problems.
Plasma α-ketoisocaproate enrichment was measured by electron impact ionization gas chromatography-mass spectrometry (GC-MS), as described previously (26). Leucine enrichment and concentration of the Aminosyn infusate were measured by negative chemical ionization GC-MS, as described previously (26). Enrichment and concentration of analyates were calculated from ion current ratios (11). The enrichment of expired CO2 was measured by isotope ratio mass spectrometry (VG Sira II). Plasma glucose concentrations were measured with a glucose analyzer (Yellow Springs Instruments, Yellow Springs, OH). Plasma amino acid concentrations were measured by cation exchange high-performance liquid chromatography with fluorescent detection (Dionex, Sunnyvale, CA) by the Yale General Clinical Research core laboratory, with an interassay coefficient of variation (CV) of between 5 and 10%.
The rate of leucine turnover (Q) is described by the equation (1) where S is the rate of leucine incorporation into protein (i.e., nonoxidative leucine disposal), C is the rate of leucine oxidation, B is the rate of appearance into the plasma from protein breakdown (i.e., endogenous leucine appearance), and I is the rate of exogenous leucine input from dietary intake or infusion. In the postabsorptive state, I is equal to zero. Thus leucine turnover is equal to the rate of appearance of leucine into plasma (B), which was calculated as described previously (41). The rate of 13CO2 excretion into expired air (F13C) was calculated as (2) where FCO2 is the CO2 production rate (μmol·kg−1·h−1), ECO2 is the enrichment of expired 13CO2 (atom percent excess), the constant 10 accounts for unit changes, and the factor 0.81 accounts for the recovery in expired air of 13CO2 released into the body bicarbonate pool (2). Dividing F13C by the rate of 13C infusion gives the fraction of the tracer that is oxidized to CO2 (fox); fox was then multiplied by B to derive leucine oxidation (C). Finally, leucine incorporation into protein (S), a proxy of protein synthesis, can be calculated as the difference between B and C (i.e., S = B − C).
During the euglycemic-hyperinsulinemic-hyperaminoacidemic clamp, leucine enters the plasma from endogenous protein breakdown and the exogenous amino acid infusion. Therefore, to calculate leucine turnover (Q), we must consider [13C]leucine input from the tracer infusion and the exogenous amino acid infusion. This was accomplished using an equation that we have described previously in detail (40). The rate of exogenous leucine infusion (I) was then subtracted from total leucine turnover (Q) to derive endogenous leucine appearance rate (B) during the clamp. For the calculation of leucine oxidation, we had to consider that the use of endogenous energy substrates will be decreased in favor of the use of exogenous glucose and amino acids. To account for changes in 13CO2 excretion due to changes in substrate oxidation (31), we measured 13CO2 excretion in a separate group of six young women (30 ± 1 yr; 22 ± 1 kg/m2) undergoing an identical clamp protocol but without administration of 13C isotope tracers. Average 13CO2 enrichments were calculated at each time point and were used to correct 13CO2 enrichments for contribution from exogenously administered glucose and amino acids. The correction factors for 13CO2 excretion at 285, 300, 315, and 330 min were 1.55, 2.19, 2.56, and 2.70 mole percent excess × 1,000 above the 13CO2 obtained for each subject at baseline, respectively. Leucine incorporation into protein (S) was calculated as described above.
Body mass was measured on a metabolic scale (Scale-Tronix, Wheaton, IL). Fat mass, fat-free mass, and bone mineral mass were each measured by dual-energy X-ray absorptiometry using a GE Lunar Prodigy densitometer (Madison, WI). Appendicular skeletal muscle mass was measured using discrete skeletal landmarks, as described previously (18).
Peak oxygen consumption (V̇o2) was measured during a graded treadmill test to volitional fatigue. Briefly, a comfortable initial walking speed was found for each volunteer and was maintained throughout the test. The grade was increased 2.5% every 2 min until volitional fatigue. Peak V̇o2 was defined as the highest 30-s average V̇o2 value measured during the last 2 min of the test.
Serum levels of estrone, estradiol, testosterone, androstenedione, and dehydroepiandrosterone were measured by radioimmunoassay. Before measurement, steroids were extracted from serum with hexane-ethyl acetate (3:2). Androstenedione, dehydroepiandrosterone, and testosterone were then separated by Celite column partition chromatography by using increasing concentrations of toluene in trimethylpentane. Estrone and estradiol were separated in a similar fashion using ethyl acetate in trimethylpentane. Dehydroepiandrosterone sulfate and sex hormone-binding globulin were measured by direct chemiluminescent immunoassays using the Immulite analyzer (Diagnostic Products, Inglewood, CA). Free estradiol and testosterone were calculated using their respective total serum concentration, sex hormone-binding globulin levels, and an assumed constant for albumin in a validated algorithm (35, 43). Intra- and interassay CV for steroid hormones and their binding proteins varied from 4 to 8% and 8 to 13%, respectively. Insulin-like growth factor I (IGF-I) was measured by radioimmunoassay (ALPCO, Windham, NH) with an intra-assay CV of 5%. IGF-I-binding protein-3 (IGFBP-3) was determined by enzyme-linked immunosorbent assay (Diagnostic Systems Laboratory, Webster, TX) with an intra-assay CV of 3%. Insulin levels were determined by radioimmunoassay (Linco Research, St. Charles, MO) with an interassay CV of 7%.
Paired t-tests were used to compare data between pretreatment follicular and luteal phase measurements in the entire cohort (n = 13). Pretreatment follicular and luteal phase measurements were averaged and are referred to as pretreatment values. A 2 × 2 repeated-measures analysis of variance model was used to detect group, time, and group × time interaction effects with treatment group (GnRHa vs. placebo) as the between-subjects factor and time (pretreatment vs. posttreatment) as the within-subjects factor. If a significant group × time interaction effect was found, a post hoc analysis was performed to assess the unique effect of time within each group through an analysis of the simple effects. Pearson product and Spearman rank correlation coefficients were calculated for the relationship between changes in protein metabolism and changes in circulating hormone levels. All analyses were conducted with SPSS software (SPSS version 9.0; Chicago, IL).
Pre- and posttreatment body composition data are shown in Table 1. No differences were found in average pretreatment body size or composition between women randomized to GnRHa and placebo. Moreover, no difference in baseline peak V̇o2 was found between women randomized to the two treatment groups (GnRHa: 41.6 ± 2.5 vs. placebo: 39.9 ± 2.9 ml·kg fat-free mass−1·min−1). No group, time, or group × time interaction effects were found for any of the body size/composition measures.
The effects of ovarian suppression on serum steroid hormones, IGF-I, and binding proteins are shown in Table 2. No group or group × time interaction effects were found for any hormone or binding protein. There were time effects (P < 0.05) observed for estrone, estradiol, and free estradiol levels. No other time effects were noted. In the GnRHa group, blood was drawn 10 days after the first injection to confirm ovarian suppression (plasma estradiol level <50 pg/ml). Ovarian suppression was confirmed in five of six volunteers 10 days after injection. In one volunteer, ovarian suppression was confirmed at 16 days after injection. Moreover, ovarian suppression was confirmed at posttreatment testing in all women randomized to the GnRHa group as a circulating estradiol level >50 pg/ml (range of values: 14–38 pg/ml).
No group × time interaction effects were found for insulin levels under postabsorptive (GnRHa: 12.5 ± 1.0 to 12.2 ± 1.1 vs. placebo: 11.3 ± 0.5 to 13.5 ± 1.1 μIU/ml) or hyperinsulinemic conditions (GnRHa: 134 ± 6 to 122 ± 7 vs. placebo: 128 ± 5 to 124 ± 6 μIU/ml). Similarly, no group × time interaction effects were found for total amino acid levels under postabsorptive (GnRHa: 2,134 ± 56 to 2,380 ± 61 vs. placebo: 2,179 ± 71 to 2,213 ± 100 μM) or hyperinsulinemic conditions (GnRHa: 3,456 ± 140 to 3,769 ± 108 vs. placebo: 3,429 ± 82 to 3,518 ± 126 μM). Similarly, no group or time effects were found for insulin or total amino acid concentrations for either condition. However, the clamp procedure increased circulating concentrations of insulin and total amino acid levels in both treatment groups during pretreatment (follicular and luteal) and posttreatment testing (P < 0.01 for all variables). Similar increases (P < 0.01 for all variables) in subgroups of amino acids, such as the essential, nonessential, and branched-chain amino acids, were found with glucose-insulin-amino acid infusion (data not shown). Moreover, no group × time interaction effects were found when comparing average pretreatment and posttreatment values for subgroups of amino acids (data not shown).
The effect of menstrual cycle phase on leucine metabolism under both postabsorptive and euglycemic-hyperinsulinemic-hyperaminoacidemic clamp conditions was examined in the total cohort (n = 13). In the postabsorptive state, leucine rate of appearance did not differ by menstrual cycle phase (follicular: 110.4 ± 4.7 vs. luteal: 108.1 ± 4.1 μmol·kg−1·h−1), whereas leucine oxidation (follicular: 18.7 ± 0.9 vs. luteal: 22.1 ± 0.7 μmol·kg−1·h−1) was higher (P < 0.01) and nonoxidative leucine disposal (follicular: 91.7 ± 4.2 vs. luteal: 86.0 ± 4.0 μmol·kg−1·h−1) was lower (P < 0.01) during the luteal phase. Under hyperinsulinemic-hyperaminoacidemic clamp conditions, leucine rate of appearance (follicular: 114.1 ± 12.1 vs. luteal: 103.1 ± 10.3 μmol·kg−1·h−1) was greater during the follicular phase (P < 0.05), whereas no differences in leucine oxidation (follicular: 36.4 ± 3.4 vs. luteal: 34.2 ± 3.5 μmol·kg−1·h−1) or nonoxidative leucine disposal (follicular: 75.3 ± 9.6 vs. luteal: 68.5 ± 8.4 μmol·kg−1·h−1) were found.
Changes in postabsorptive leucine kinetics with GnRHa (A) and placebo (B) administration are shown in Fig. 1. GnRHa administration reduced leucine rate of appearance (108.1 ± 8.3 to 96.0 ± 7.8 μmol·kg−1·h−1) compared with placebo (110.3 ± 4.3 to 108.1 ± 5.5 μmol·kg−1·h−1; group × time interaction effect: P < 0.05; time effect: P < 0.01). Post hoc analysis showed that the unique effect of time was only significant in the GnRHa group (P < 0.01). Similarly, nonoxidative leucine disposal was reduced in women receiving GnRHa (87.1 ± 7.2 to 75.9 ± 7.3 μmol·kg−1·h−1) compared with women receiving placebo (90.3 ± 4.7 to 85.5 ± 5.1 μmol·kg−1·h−1; group × time interaction effect: P < 0.05; time effect: P < 0.01). Post hoc analysis showed that the unique effect of time was only significant in the GnRHa group (P < 0.01). No group, time, or group × time interaction effects were found for leucine oxidation (GnRHa: 21.0 ± 1.2 to 20.0 ± 1.5 vs. placebo: 20.0 ± 0.91 to 22.6 ± 1.2 μmol·kg−1·h−1). The directionality of group × time interaction effects was similar when individual pretreatment measurements (i.e., follicular and luteal phase measurements) were compared with posttreatment data. When follicular phase measurements were compared with posttreatment data, leucine rate of appearance (P < 0.05) was reduced by GnRHa administration (GnRHa: 110.5 ± 9.0 to 96.0 ± 7.8 vs. placebo: 110.3 ± 4.9 to 108.1 ± 5.5 μmol·kg−1·h−1), whereas nonoxidative leucine disposal showed a trend (P = 0.09) toward being reduced in the GnRHa group (GnRHa: 91.3 ± 7.4 to 75.9 ± 7.3 vs. placebo: 91.9 ± 5.2 to 85.5 ± 5.1 μmol·kg−1·h−1). When luteal phase measurements were compared with posttreatment data, leucine rate of appearance showed a strong trend toward being reduced (P = 0.06) with GnRHa administration (GnRHa: 105.6 ± 7.7 to 96.0 ± 7.8 vs. placebo: 110.2 ± 4.1 to 108.1 ± 5.5 μmol·kg−1·h−1), whereas no group × time interaction effect was found for nonoxidative leucine disposal (GnRHa: 82.9 ± 7.1 to 75.9 ± 7.3 vs. placebo: 88.6 ± 4.5 to 85.5 ± 5.1 μmol·kg−1·h−1; P = 0.22). Finally, no correlations were found between calculated changes in protein metabolism (average pretreatment to posttreatment) and changes in any circulating hormone measured.
The effects of GnRHa (A) and placebo (B) administration on leucine kinetics under hyperinsulinemic-hyperaminoacidemic conditions are shown in Fig. 2. As with the postabsorptive condition, GnRHa reduced leucine rate of appearance (112.5 ± 15.7 to 88.4 ± 22.6 μmol·kg−1·h−1; group × time interaction effect: P < 0.05; post hoc analysis showed that the unique effect of time was only significant in the GnRHa group: P < 0.02) and nonoxidative leucine disposal (81.6 ± 12.3 to 57.9 ± 16.7 μmol·kg−1·h−1; group × time interaction effect: P < 0.05; time effect: P < 0.05; post hoc analysis showed that the unique effect of time was only significant in the GnRHa group: P < 0.01) compared with placebo (109.9 ± 7.7 to 115.4 ± 10.1 and 81.3 ± 6.4 to 77.3 ± 9.5 μmol·kg−1·h−1, respectively). Leucine oxidation, in contrast, increased in the placebo group (28.2 ± 1.5 to 39.1 ± 3.1 μmol·kg−1·h−1) compared with the GnRHa group (30.9 ± 3.4 to 30.5 ± 5.0 μmol·kg−1·h−1; group × time interaction effect: P < 0.05; time effect: P < 0.05; post hoc analysis showed that the unique effect of time was only significant in the placebo group: P < 0.01). The directionality of group × time interaction effects were similar when individual pretreatment measurements (i.e., follicular and luteal phase measurements) were compared with posttreatment data. When follicular phase measurements were compared with posttreatment data, leucine rate of appearance was reduced (P < 0.03) by GnRHa administration (GnRHa: 119.4 ± 23.6 to 88.4 ± 22.6 vs. placebo: 109.6 ± 11.9 to 115.4 ± 10.1 μmol·kg−1·h−1) and leucine oxidation was increased (P < 0.02) in the placebo group (GnRHa: 39.0 ± 6.1 to 30.5 ± 5.0 vs. placebo: 33.9 ± 3.3 to 39.1 ± 3.1 μmol·kg−1·h−1). When luteal phase measurements were compared with posttreatment data, leucine rate of appearance was reduced (P < 0.04) with GnRHa administration (GnRHa: 103.8 ± 19.7 to 88.4 ± 22.6 vs. placebo: 102.4 ± 10.6 to 115.4 ± 10.1 μmol·kg−1·h−1) and leucine oxidation increased (P < 0.01) in the placebo group (GnRHa: 37.0 ± 6.3 to 30.5 ± 5.0 vs. placebo: 31.3 ± 3.5 to 39.1 ± 3.1 μmol·kg−1·h−1). Finally, no correlations were found between calculated changes in protein metabolism (average pretreatment to posttreatment) and changes in any circulating hormone measured.
Several studies have shown that menopause accelerates the normal age-related loss of lean tissue (3, 8, 37, 44), suggesting that ovarian hormones may participate in the regulation of fat-free mass. In an effort to explore the role of ovarian hormones in the regulation of fat-free mass, the present study examined the hypothesis that ovarian suppression with GnRHa administration would reduce whole body protein turnover. Our results show that ovarian hormone deficiency coincident with GnRHa administration was associated with decreased rates of protein metabolism. To our knowledge, this is the first study to demonstrate a role for endogenous ovarian hormones in the regulation of protein metabolism in adult women.
Few studies have examined the relationship between ovarian hormones and protein metabolism. Prior work from our laboratory in a cohort of middle-aged women showed a positive association between postabsorptive protein turnover and both circulating estradiol and progesterone concentrations (41). Our current results agree with the directionality of these relationships and extend these preliminary observations to show that pharmacologically induced ovarian suppression reduces protein turnover in the postabsorptive state and under euglycemic-hyperinsulinemic-hyperaminoacidemic conditions. Collectively, these data support a role for ovarian hormones in the maintenance of normal rates of protein turnover in healthy women.
Our findings differ from those of Mauras et al. (27), who showed that ethinyl estradiol or depot estradiol administration to prepubertal girls with Turner's syndrome and hypogonadotropic hypogonadism had no effect on whole body protein metabolism. Although seemingly contradictory, these findings should be interpreted within the context of the populations examined and the study designs employed. For example, prepubertal girls, in particular those with preexisting pathologies and/or abnormalities in ovarian hormone production, may not be as responsive to estrogen compared with healthy, adult women with normal ovarian function. Perhaps more importantly, pharmacological replacement of estradiol, which achieves a constant, supraphysiological plasma concentration of the hormone, is not directly comparable to the cessation of cyclical patterns of endogenous hormone release caused by GnRHa administration. Finally, replacement of only estradiol may have limited effects if other ovarian hormones, such as progesterone, affect protein metabolism individually or in combination with estradiol (22, 41). Thus divergent results between studies may relate to the different study populations and/or experimental approaches.
How might ovarian hormones regulate protein metabolism? Receptors for ovarian hormones are found in a number of nonreproductive tissues, including liver (7) and skeletal muscle (20, 24), which together account for approximately two-thirds of whole body protein turnover (38). Thus ovarian hormones may regulate protein metabolism in these tissues directly. In support of this notion, ovarian hormones have been shown to regulate skeletal muscle protein metabolism (20, 30, 39), although little is known about their effects on hepatocytes. In addition to direct effects, ovarian hormones may affect protein metabolism by modulating other hormonal systems, such as the growth hormone-IGF-I axis (25). In agreement with some (16, 45) but not all studies (9), we found no effect of GnRHa administration on circulating levels of IGF-I or IGFBP-3. Moreover, no effect of GnRHa administration was found on circulating total or free testosterone levels, androstenedione, dehydroepiandrosterone, dehydroepiandrosterone sulfate, or sex hormone-binding globulin, and changes in circulating hormones were not correlated to GnRHa-induced changes in protein metabolism. Thus our results argue against an indirect effect of ovarian suppression on protein metabolism via other endocrine systems. We should mention that ovarian hormones may modify the actions of other hormones at target tissues in a manner that is independent of their circulating concentrations (21, 25). In addition, we cannot discount the possibility that ovarian hormones influence protein metabolism indirectly through their effect to modify activity or dietary habits.
What are the consequences of the observed alterations in protein metabolic rates? Although the clinical implications of changes in protein turnover are difficult to surmise, the physiological relevance of our findings can be put into context by comparing our data to similar studies in men. Mauras et al. (28) reported a 13% reduction in protein turnover in men made hypogondal by GnRHa administration for 10 wk, a decrease that is very similar to the 11% reduction in protein turnover found in the present study. The comparable effects of hypogonadism on protein metabolism in men and women are surprising. Although the anabolic effects of testosterone have been well described (12, 28, 42), to our knowledge there has been no clear evidence of a protein anabolic effect of endogenous ovarian hormones in humans. Our findings would suggest that, like their male counterparts, ovarian hormones play an important role in the maintenance of normal rates of protein turnover in women.
How does a reduction in protein turnover impact the quantity and/or functionality of body protein? Our study was not specifically powered to detect changes in fat-free mass. However, we predict that a reduction in protein turnover, which also occurs in parallel with the age-related loss of lean tissue (32), would promote a decline in fat-free mass. In support of this notion, reports have shown that prolonged (4 mo) treatment with GnRHa reduces fat-free mass in women (46). In addition to effects on protein mass (46), a decline in the turnover rate of body proteins could impact the structural and/or functional integrity of proteins. The constant breakdown and resynthesis of protein is essential for ridding the body of damaged and/or dysfunctional proteins. A decline in this cyclical repair process could lead to the accumulation of damaged proteins, which in turn could impair cellular function. Thus ovarian hormone deficiency may have implications for changes in the quantity and functional integrity of body protein.
We are careful not to extrapolate our findings in young, healthy, normal cycling women to those undergoing natural menopause. The experimental model used in this study causes an abrupt cessation of ovarian hormone production, whereas the loss of ovarian hormones associated with menopause is more episodic and protracted (29). In addition, there may be age-related differences in physiological function that could affect the response of protein metabolism to ovarian hormone deficiency in younger vs. older adult women. For example, aging could alter steroid hormone receptor density or responsiveness in target tissues. Moreover, age-related changes in other hormonal/physiological systems or the presence of various disease states could enhance or diminish the effect of ovarian hormone deficiency on protein metabolism. Thus our results are only directly applicable to ovarian deficiency in young adult women. Nevertheless, inasmuch as our experimental model allows us to evaluate the effect of endogenous ovarian hormone deficiency on protein turnover, our findings support a physiological role for ovarian hormones in the maintenance of normal rates of protein metabolism.
Our finding that menstrual cycle phase did not affect postabsorptive protein breakdown is at odds with the data of Lariviere et al. (23), which showed increased protein breakdown during the luteal phase. The reason for this disparity is not clear, because both studies used similar methodology to assess protein metabolism and studied women at about the same time of the menstrual cycle. An earlier study from this group using a different isotope tracer approach to assess protein metabolism, however, failed to find an effect of menstrual cycle phase on protein turnover (10). In contrast, our finding of increased (19%) postabsorptive leucine oxidation during the luteal phase agrees well with the 18% increased leucine oxidation reported by the same group (23). These results suggest that the luteal phase of the menstrual cycle increases the oxidative disposal of amino acids. Under euglycemic-hyperinsulinemic-hyperaminoacidemic clamp conditions, we found greater rates of protein breakdown during the follicular phase but no differences in protein oxidation or synthesis. No other study has examined protein metabolism under similar experimental conditions, although Kriengsinyos et al. (22) reported no effect of menstrual cycle phase on protein breakdown during 4 h of oral feeding. Although differences in study designs and measurement techniques prevent definitive conclusions from being drawn, these studies nonetheless highlight a potential effect of menstrual cycle phase on protein metabolism.
Because of these cycle-dependent differences in protein metabolism, in our analysis we used an unweighted average of follicular and luteal phase measurements for the pretreatment value to estimate average protein metabolic rates during the menstrual cycle. That our measurements were taken at points of the cycle characterized by relative hormone deficiency (early to mid-follicular phase) and excess (mid-luteal phase) means that an average value should be reflective of how fluctuations in hormone levels during the cycle impact protein metabolism. Our results derived from this average value are reinforced by the fact that qualitatively similar findings were obtained when data from individual cycle phases were used as pretreatment values (see results). For this reason, we believe that the effect of GnRHa administration on protein metabolism observed in our study is not an artifact of our analytical approach.
In conclusion, our study shows that ovarian suppression with GnRHa reduces protein metabolic rates under both postabsorptive and euglycemic-hyperinsulinemic-hyperaminoacidemic conditions. To our knowledge, this is the first study to report of a role for endogenous ovarian hormones in the regulation of protein metabolism in adult women. Together with studies showing that longer periods of ovarian suppression with GnRHa reduce fat-free mass (46), our findings suggest that ovarian hormone deficiency may alter body protein content through effects on protein turnover. Identification of the tissues responsible for these changes with ovarian suppression and the specific hormone(s) that regulate protein metabolism will be of importance in discerning the potential relevance of ovarian hormone deficiency to the erosion of fat-free mass and physical function observed after menopause.
This work was supported by National Institutes of Health Grants AG-021602 and RR-00109.
We thank all the participants who volunteered time for this study.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 by American Physiological Society