| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of 1Health and Kinesiology and 2Mechanical Engineering, Texas A&M University, College Station, Texas
Submitted 12 March 2004 ; accepted in final form 8 January 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
150% of peak isometric torque every third day for 8 wk. After the training period, groups were not significantly different with regard to peak torque or muscle size. However, the tibial midshaft of the trained leg in the OVX+Train mice exhibited greater stiffness (+15%) than that in the untrained OVX mice, which could not be explained by changes in cross-sectional geometry of the tibia. Scaling of bone mechanical properties to muscle strength were not altered by ovariectomy or training. These data indicate that eccentric exercise training in adult mice can significantly increase bone stiffness, despite the absence of ovarian hormones. ovariectomy; bone mechanical properties; bone density; bone geometry; muscle stimulation
Along with load-bearing exercise history, endocrine status is a primary determinant of bone metabolic characteristics (34). Endogenous estrogen plays a role in balancing the two bone metabolic processes of formation and resorption, acting to maintain bone integrity. The effects of estrogen deficiency on bone tissue have been investigated in both human and animal models. Losses in cancellous bone are extensive in both humans and rodents, although the rate of loss in rats exceeds that in both humans and mice (4, 5, 10, 35, 44, 45). This is accompanied by a significant weight gain in rats, whereas humans and mice show little or no gains in body weight (4, 5, 10, 32, 44, 45).
Although the effects of estrogen deficiency are well characterized in humans and rodents in cancellous bone, the effects on cortical bone are poorly understood. Understanding the effects of estrogen deficiency on cortical bone is important because some 80% of the human skeleton is cortical bone, the density and geometry of which are primary determinants of bone strength (13). Losses at cortical bone sites occur in estrogen-deficient rats, primarily on endocortical surfaces, but at a slow rate, whereas periosteal bone formation often increases (44, 45, 46). Indexes of bone formation and resorption change little after ovariectomy at cortical bone sites in skeletally mature mice (4, 15), but there are few data documenting whether bone mineral density or mechanical properties are altered. On the basis of similarities between humans and mice in cancellous bone and body weight changes after estrogen deficiency, it is logical to assume that cortical bone changes in the mouse would more closely model that in humans than changes in the rat.
Endogenous estrogen also plays a role in regulating skeletal muscle mass and strength in the adult female. Decreases in skeletal muscle mass and strength are observed in postmenopausal women (1, 14, 33). Furthermore, estrogen deficiency in mice has been shown to significantly reduce maximum isometric strength of hindlimb muscles (39). The effect of resistance exercise training on muscle characteristics in estrogen-deficient animals has not been established.
The individual and summative contributions of alterations in mechanical loading and endocrine status to bone mass and mechanical properties, as well as to skeletal muscle properties, in models of estrogen deficiency are not clear. The primary objective of this study was to investigate the effects of muscular contraction training on the tibiae in an estrogen-deficient mouse model. This was accomplished by simulating resistance training with high-force eccentric contractions of the anterior crural muscles [tibialis anterior (TA), extensor digitorum longus (EDL), and extensor hallucis longus] in the ovariectomized mouse to generate in vivo mechanical loading on the tibia. A secondary purpose was to investigate changes in skeletal muscle after this type of training in the ovariectomized mouse. We hypothesized that imposing high magnitudes of strain on bone with high-force contractions would cause simultaneous improvements in muscle strength and bone material and structural properties, while attenuating losses in bone mineral density (BMD) resulting from the removal of estrogen. Measurements of bone and skeletal muscle in vivo at 0, 4, and 8 wk, as well as ex vivo properties at the end of the 8-wk time course, were used to investigate concurrent changes in these tissues throughout the training period. Few studies published to date have utilized in vivo testing of both skeletal muscle function and bone density and geometry in an animal resistance training study. We demonstrate here that training of hindlimb muscle in ovariectomized mice with high-force contractions improves bone stiffness and muscle strength, even in the absence of circulating estrogen.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
A priori statistical power analyses were conducted using cortical BMD and maximal isometric torque as the criterion measures (paired t-test; 
= 0.2). Notomi et al. (30) detected a 7% greater middiaphyseal BMD in resistance-trained rat femur and, assuming that our high-force training program could produce half of that effect (3.5%), a power of 0.8 could be achieved with a sample size of five animals per group (
= 0.05, = 0.2, 
= 0.03955, 
= 0.02486). Previous work in our laboratory (39) has shown that maximal isometric torque differences of 18% exist between estrogen-replaced and estrogen-depleted mice. Therefore, a sample size of eight would be sufficient to detect significant differences between groups ( = 0.05, = 0.2, = 0.3, = 0.28). Given that we expected some failures of the nerve cuff over time, we compensated for an expected 25% attrition rate from each group, so 13–14 animals were used for each of the groups requiring nerve cuff implantation. All mice were allowed 1 wk of acclimation before any procedures were performed. This study was approved by the Texas A&M University Laboratory Animal Care Committee and followed the guidelines set by the American Physiological Society.
Nerve cuff implantation.
Stimulating electrode nerve cuffs were implanted on the common peroneal nerve within the left leg of animals in the Sham, OVX, and OVX+Train groups as described previously (40, 41). The nerve cuff is made up of two lengths of Teflon-coated 90% platinum/10% iridium wire (MedWire; Mt. Vernon, NY), distally anchored in the limb at the lateral gastrocnemius and proximally externalized in the dorsal cervical region of the animal. When not in use for muscle function testing or training, the proximal end of the wire was recessed under the skin and the opening was closed with a wound clip. Success of nerve cuff implantation was checked 21 days after implantation by in vivo activation (see Anterior crural muscle testing below) of the anterior crural muscles, during which isometric torques about the left ankle of the mouse were measured. Implantation was deemed successful if the peak isometric tetanic torque of the anterior crural muscles was 
2.54 N·mm (39).
Ovariectomy. Mice with nerve cuffs judged successful at 21 days then underwent either ovariectomy (OVX and OVX+Train) or sham ovariectomy (Sham); at this time point, mice were 13 wk old. After anesthetization with pentobarbital sodium (80 mg/kg ip), a 2–3 mm incision of the posterior abdominal wall was made, vascular supply to each ovary was tied off, and both ovaries were removed. Animals were allowed to recover for 7 days before the onset of training, during which time their health and recovery were monitored.
Anterior crural muscle testing. Anterior crural muscle function assessments were performed at 0, 4, and 8 wk on animals in the Sham, OVX, and OVX+Train groups as described previously (39, 41). Briefly, animals were anesthetized with fentanyl (0.33 mg/kg), droperidol (16.7 mg/kg), and diazepam (5.0 mg/kg). After aseptic preparation of the dorsal cervical region, each mouse was placed in right lateral recumbency on a temperature (39°C) controlled platform, the left foot was secured in a machined aluminum shoe attached to a servomotor (Cambridge Technology 300B dual-mode servomotor system), and the left knee was clamped so that the lower leg was perpendicular to the foot. A stimulator (Grass Instruments S48) and a stimulus isolation unit (Grass Instruments SIU5) were used to stimulate the common peroneal nerve via the implanted electrode wire externalized at the animal's dorsal neck. Voltage at which peak isometric torque was elicited was optimized during a trial of 8–16 isometric tetanic contractions (200-ms train duration with 0.1-ms pulses at 300 Hz). Peak torque produced about the ankle, as well as torque-frequency curves, were then recorded by using the Cambridge Technology servomotor system, Keithley-Metabyte DAS-4062 and 1602 interface boards, data-acquisition/process control software (Viewdac Version 2.1), and a 80486 66-MHz computer. During the torque-frequency measurements, isometric contractions were performed every 45 s at the following stimulation frequencies: 20, 40, 60, 80, 100, 125, 150, 200, 250, 300, 350, and 400 Hz.
In vivo anterior crural muscle training.
During skeletal muscle training, both anesthesia and optimization of voltage procedures were identical to those used in muscle function testing described above. Training consisted of three sets of 10 eccentric contractions with 10 s between contractions and 2 min between sets. The three sets of 10 contractions were used to approximate a human resistance training program, but rest between contractions was shortened to optimize force production and thus cumulative bone strain. Contractions were performed over a 40° angle using stimulation at 180 Hz and 300 ms, parameters designed to elicit 150% of maximal isometric tetanic torque as measured at the initiation of the study. Two minutes after the final eccentric contraction, a final isometric contraction was performed (posttraining bout torque). The OVX+Train group underwent skeletal muscle exercise training every third day for 8 wk. Those animals not in the training group were given an intraperitoneal injection of fentanyl, droperidol, and diazepam every third day to control for the effects of the anesthetic over a total of 18 injections.
To estimate the loading imposed on the bone by this training protocol, strain on the tibial surface was calculated assuming the tibia to behave as a beam column under combined bending and axial compression. The moment arm distance for the anterior crural muscles has been estimated to be 1.0 mm for this system (3), and the line of action of the force was assumed to be offset from the center of the tibia cross section by one-half of the anterior-to-posterior external diameter at middiaphysis. The maximum strain in this case is compressive in nature and occurs on the surface of the tibia at middiaphysis and on the posterior aspect (where both the axial and bending strains are compressive).
In vivo bone testing. Tibial bone density and polar cross-sectional moment of inertia (CSMI) were measured using peripheral quantitative computed tomography (pQCT; Stratec XCT Research M; Norland Medical Systems, Fort Atkinson, WI). This pQCT device has a scanning fan beam thickness of 0.5 mm and maximal voxel resolution of 70 µm. Daily machine calibration was performed by using a hydroxyapatite standard cone phantom to ensure measurement precision throughout the study. Before training and at 4 and 8 wk, one scan slice was taken at 50% (tibial midshaft) of the total bone length, a site that is virtually 100% cortical bone and subjected to bending forces during active muscle contraction. Scans were simultaneously taken of right and left tibiae, allowing for the unexercised right tibia from a given animal to serve as a matched within-animal control. Analyses were performed with Stratec software (version 5.40B), using a standardized analysis for diaphyseal bone (separation = 1, attenuation threshold = 0.605 g/cm3) applied to each scan. Machine precision (based on manufacturer data) is 9 mg/cm3 for cortical bone. The moment of inertia around the neutral bending axis (CSMIy), pertinent to the three-point bending procedure later applied, was estimated by dividing the polar CSMI in half. This estimation procedure was judged reasonable, given that the mouse tibia at midshaft is nearly cylindrical.
Reproducibility of the pQCT results were determined by scanning two adult mice five times each, with repositioning of the hindlimbs with each scan. Coefficients of variation are ±5.5% for cortical bone area, ±7.3% for cortical BMD, and ±13.1% for polar CSMI.
Tissue harvest. Animals were killed at the age of 22 wk after final in vivo muscle and pQCT bone testing. After the completion of in vivo testing, animals were anesthetized with pentobarbital sodium (80 mg/kg), and the TA and EDL muscles were dissected free, weighed, and frozen for future testing. Tibiae from both legs were harvested, packed in PBS-soaked gauze and PBS saline, and frozen at –80°C until mechanical testing could be performed. Mice were euthanized with an overdose of pentobarbital sodium (200 mg/kg). Abdominal examination during necropsy confirmed absence of ovarian tissue in all ovariectomized mice.
Mechanical testing.
Tibiae from animals that completed the study were measured for anterior-posterior and medial-lateral periosteal diameters by precision hand calipers and mechanically tested 2 wk after death. The unexercised right tibia from a given animal served as a within-animal control for each (left) exercised bone. Tibial bone structural and material properties were determined by a three-point bending test on an Instron 1125 machine at the middiaphyseal pQCT sampling site (50% total bone length). Tibiae (thawed to room temperature) were placed lateral side down on custom-built metal pin supports, and quasi-static loading was applied at 2.54 mm/min to the upper bone surface until fracture occurred. Bones were placed lateral side down to create the most stable position possible during testing. The small displacements of the servo-controlled Instron were monitored by a linear variable differential transformer, and the applied load measured with a 453.6-kg load cell at 22.7-kg maximum load. Load and displacement outputs were digitized to a personal computer at 10 Hz by using LabTech Notebook Pro software (version 8.01; Laboratory Technologies, Wilmington, MA). Load-displacement curves were later analyzed with TableCurve 2.0 (Jandel; San Rafael, CA). Ultimate load (N) was defined as the highest load obtained before fracture, and stiffness was defined as the slope of the linear portion of the load vs. displacement curve in the elastic region, utilizing the greatest number of data points possible. The same investigator visually inspected all load-displacement curves to assure that consistent criteria were used in designating ultimate load and in the selection of data points for the calculation of stiffness.
Material properties were calculated from the relevant structural properties [ultimate load (UL) and stiffness (k)] by using the pQCT-derived CSMIy, bone anterior-posterior outside diameter at midshaft (d), and the bottom support span (L; 10 mm) as input data. Classical beam theory analysis was used to generate expressions estimating intrinsic material properties. Ultimate stress (US) was calculated by using the equation US = (UL·d·L)/(8·CSMIy). Modulus of elasticity (E) was calculated by using the equation E = (k·L3)/(48·CSMIy).
Statistical analyses. Changes in peak torques of the anterior crural muscles were assessed by use of a group-by-time ANOVA with repeated measures over time. Change scores for midtibia BMD at 4 and 8 wk were used for two-way ANOVA because baseline values for this variable were significantly different among the three groups at 0 wk. Bone mechanical properties were compared by a two-way ANOVA (treatment group x leg). When two-way interactions and main effects were found, the Student-Newman-Keuls post hoc test was applied. Simple linear regression was used to determine correlation coefficients between bone mechanical properties and peak muscle torque values. Statistical tests were performed with SigmaStat Version 1.0. All statistical tests resulting in P < 0.05 were judged significant. Results reported are means ± SE.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
There were no significant differences in baseline anterior crural muscle torques among groups as measured during isometric and eccentric contractions 7 days after ovariectomy (day 1). However, peak muscle torque in the ovariectomized groups tended to be lower (4–9%) than that of the Sham group (Fig. 1A). Increases (vs. baseline) in peak anterior crural muscle isometric torque over the 8 wk were 17.2 ± 5.7% in the OVX+Train group, 9.9 ± 4.5% in the OVX group and 8.8 ± 6.0% in the Sham group. These values were not significantly different among groups (P = 0.3). Similarly, although peak eccentric torque rose 15% in the trained group, there were no statistically significant differences in peak eccentric torque of the anterior crural muscles among groups or over time (Fig. 1B). Anterior crural muscle strength at all frequencies was not different between the OVX and OVX+Train groups at any time (Fig. 2). Slight rightward shifts in torque at frequencies between 60 and 125 Hz were seen in the Sham group only.
|
|
for Sham, OVX, and OVX+Train mice, respectively, at this training intensity. Finally, when peak isometric torque was remeasured at the end of the first training bout for OVX+Train mice, a 19% decrease was observed compared with the prebout peak torque (Fig. 3). After 4 and 8 wk of training, peak isometric torque declined by only 5 and 9%, respectively, by the end of a training bout. This training-induced reduction in strength deficit is similar to the persistent but reduced strength deficits associated with multiple bouts of eccentric contraction-induced injury (22). The ability of the "trained" muscle to maintain a higher force output during the bout of eccentric contractions would result in greater total strain on the bone during these later training sessions.
|
Tibial midshaft BMD and geometry.
Cortical BMD for the left tibial midshaft changed similarly among Con, Sham, and OVX+Train groups across time, with an average increase of 4.6% vs. baseline values (Fig. 4). Concurrently, left tibial BMD for OVX mice decreased 1.7% after 8 wk. However, these changes in BMD over time and among groups at 8 wk showed only a statistical trend by ANOVA (P = 0.13). When examining the right (untrained) tibiae in OVX mice, a significant loss of bone density was observed (–20.24 ± 26.1 mg/cm3 or 2% of baseline) after 8 wk compared with the increase observed in Con mice (+30.8 ± 21.9 mg/cm3), verifying a significant decrement in cortical BMD with ovariectomy. Neither cortical bone area nor the CSMIy for the left tibia was different among groups (Table 2).
|
|
19%) in OVX+Train animals than in OVX mice.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
-estrogen receptor in bone cells is downregulated during estrogen deficiency (21, 29). Given that the cortical bone response to mechanical loading is significantly attenuated in mice lacking -estrogen receptor (28), it is reasonable to ask whether any benefits of resistance training might accrue in the absence of normal circulating estrogen. Indeed, there is evidence that bone in estrogen-deficient postmenopausal women is less sensitive to mechanical loading than in estrogen-sufficient women (7, 26). The primary finding of this study was that training with high-force eccentric muscle contractions (at 150% of peak isometric torque) effectively stimulated an increase in tibial stiffness. The scaling of bone strength parameters to peak muscle torque in the present study (Table 3) was not affected, however, by either estrogen deficiency or training.
Effects of estrogen deficiency and exercise training on bone.
A significant decrease in cortical BMD was detected in the right tibiae of the OVX mice at the study's end (9 wk after ovariectomy) compared with controls, verifying a significant impact of estrogen deficiency on cortical bone in this model. Tibial BMD levels in the OVX+Train mice were not significantly different from that of the control group after 8 wk of training, suggesting that this training protocol successfully attenuated the decline in BMD observed with estrogen deficiency. We detected a statistical trend for a significant difference between the OVX and OVX+Train group (P = 0.13) for cortical BMD change over time in the left leg. Although we were not able to measure tibial strain directly during the eccentric contraction training bouts, our estimate of a loading magnitude of 1,700 µ on the midshaft tibia with this eccentric contraction training is roughly double that observed (850 µ) to produce significant increases in cortical bone mass in young mice with in vivo cantilever bending of the tibia (18).
Bone mechanical properties are dependent on bone material properties (such as mineral density) but also geometric properties that can independently affect bone strength (13). Our direct testing of mechanical properties revealed a 15% increase in stiffness in the tibia of the trained leg in OVX mice; whereas ultimate load (the maximum force absorbed by bone before fracture) was not affected by training.
Improvements in stiffness are especially important in the small terrestrial animal model, in which bone stiffness is fundamental in optimizing skeletal muscle mechanics during normal locomotion (9). High-impact forces approaching the bone's ultimate load are likely to be rare events in the average small rodent's lifetime. Our data, however, do not provide a clear mechanism for this increase in stiffness in the OVX+Train mice. It is possible that the statistical trends toward increases in CMSIy and BMD over the 8 wk of training, although not statistically significant individually, combined to yield a significant effect on stiffness. Cortical BMD, in particular, showed a statistical trend in the trained vs. the OVX mice (P = 0.13). One source of variation in our study was the scanning technique used for pQCT scanning. Our coefficients of variation for cortical BMD are higher than those shown for ex vivo scans [±7.3% in our study vs. 1.1% in Jamsa et al. (23)]. With this in mind, it will be important for future studies to use single leg scanning protocols to try to increase reliability for in vivo pQCT-derived BMD measures.
The supposition that our stiffness increase was due to small increases in several contributing factors is further supported by the suggested improvement in elastic modulus (material stiffness independent of bone geometry) in the trained mice. Kinetic histomorphometric measurements of bone formation rate would be required to verify whether periosteal or endocortical modeling was stimulated with this training protocol.
Endurance training protocols, which incur concentric and eccentric muscle actions but at a lower intensity than with resistance-type training, produce somewhat variable effects on long bone mechanical properties in ovariectomized rodents. Treadmill running initiated immediately after ovariectomy in 3-mo-old rats does not produce any significant changes in tibial mechanical properties (24). However, a slightly lower intensity treadmill running protocol initiated in 9-mo-old ovariectomized rats successfully enhanced femoral ultimate load, with no effect on elastic modulus (6). Jump training, which involves both high-force concentric and eccentric muscle contractions, successfully induces increases in ultimate load of the tibia in 3-mo-old ovariectomized rats, achieved with only 10 jumps per day (20). There are fewer data available in exercise-trained ovariectomized mouse models, so it remains unclear whether a species difference may account for the lack of effect observed on ultimate load in the present study. Warren et al. (39) observed increases in material properties (i.e., ultimate strength and modulus of elasticity) in ovariectomized mice but no significant changes in stiffness after a single bout of 150 eccentric contractions.
Effects of estrogen deficiency and exercise training on skeletal muscle.
Previously, we documented a 15% reduction in peak isometric torque of the anterior crural muscles in estrogen-deficient mice compared with ovariectomized animals supplemented with estrogen (39). Furthermore, Fisher et al. (16) showed decreases in peak force per kilogram body mass and in other contractile properties of the leg muscles in ovariectomized rats. Conversely, estrogen deficiency had minor effects on skeletal muscle strength in this study. Peak isometric torque in the ovariectomized animals in our study was 5–10% lower than those in the Sham group, but this was not statistically significant. These results are similar to those found in postmenopausal women, in whom muscle specific force decreases. Although one study (33) demonstrated an attenuation of this decline in specific force with hormone replacement therapy, most studies have shown no association between serum estrogen levels and muscle strength (14, 27) and no effect of hormone replacement therapy on strength levels in postmenopausal women (2, 38).
Although there were trends for greater peak isometric strength and muscle size in the trained group, we believe that our data set was simply underpowered to detect significance in 7–10% advantages in peak isometric strength and 12–20% advantages in wet weight for the TA and EDL in the trained animals. Muscle function as tested via the torque-frequency relationship over a series of frequencies was not different between OVX and OVX+Train groups. Slight shifts in the torque-frequency relationship in the Sham group at certain lower frequencies is unexplained and not due to significant changes in muscle size.
The interactive effects of estrogen deficiency and resistance exercise training on bone and muscle are not firmly established. Few studies have investigated the effects of endurance training on skeletal muscle characteristics in ovariectomized rodents. Treadmill running stimulates increases in muscle mass in ovariectomized rats (31), and endurance exercise has been shown to affect fast-to-slow myosin heavy chain isoform shifts in rats (25). Although we are not aware of any studies in the mouse model, there are many human studies that examine changes in muscle strength and morphology in postmenopausal women with strength training (8, 11, 37). These studies have agreed that muscle function can be improved in postmenopausal women after resistance training, although data concerning the independent contributions of hormone replacement therapy on this improvement are equivocal.
In a previous study, we showed that functional bone-muscle relationships in ovariectomized mice were not altered by estradiol treatment in ovariectomized mice after a single bout of eccentric exercise (39). The results of the present study further these findings, demonstrating that functional bone-muscle relationships are stable over an 8-wk training program.
Further studies are needed to address issues of stimulus intensity (exercise training at levels other than 150% of peak isometric tetanic torque). Skeletal muscle testing in this study revealed a low level of persistent eccentric contraction-induced loss in skeletal muscle strength throughout the 8 wk of training, although the postexercise isometric force loss was decreased by half over the 8-wk training period. A lower intensity stimulus (e.g., 125% of peak isometric tetanic torque) could possibly optimize skeletal muscle strain on bone while avoiding these small decrements in muscle strength over single training bouts. On the other hand, a higher intensity stimulus could produce increases in muscle size and strength, if it could be implemented with minimal eccentric contraction-induced injury.
In conclusion, the data presented above lend support to our preliminary hypothesis that high-intensity eccentric exercise training can create significant bone adaptation over an 8-wk time course in the estrogen-deficient mouse. We have demonstrated that the eccentric exercise training regimen used in this study elicited variable gains in bone mechanical properties. However, the improved stiffness observed in trained OVX mice is unlikely to be due solely to geometric changes, because the relevant cross-sectional moment of inertia at midtibia was minimally affected, suggesting that some change in material properties occurred with training. Although our study was underpowered to detect such changes, we detected statistical trends in several variables that would likely be significant given a greater number of animals per group or decreased variability in the pQCT measures. Overall, these results lend support to the concept that cortical bone adaptations can occur with resistance training in an estrogen-deficient state.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Present address for M. J. Hubal: Department of Exercise Science, 110 Totman, 30 Eastman Lane, Amherst, MA 01003.
|
FOOTNOTES |
|---|
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.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. M. Greising, K. A. Baltgalvis, D. A. Lowe, and G. L. Warren Hormone Therapy and Skeletal Muscle Strength: A Meta-Analysis J Gerontol A Biol Sci Med Sci, June 26, 2009; (2009) glp082v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. L. Warren, A. L. Moran, H. A. Hogan, A. S. Lin, R. E. Guldberg, and D. A. Lowe Voluntary run training but not estradiol deficiency alters the tibial bone-soleus muscle functional relationship in mice Am J Physiol Regulatory Integrative Comp Physiol, November 1, 2007; 293(5): R2015 - R2026. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. P. Rader and J. A. Faulkner Effect of aging on the recovery following contraction-induced injury in muscles of female mice J Appl Physiol, September 1, 2006; 101(3): 887 - 892. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
