The objectives of this study were to investigate the different effects on muscle mass and cancellous (proximal tibial metaphysis [PTM]) and cortical (tibial shaft [TX]) bone mass of sham-operated and orchidectomized (ORX) male rats by making rats rise to erect bipedal stance for feeding. Specially designed raised cages (RC) were used so that the rats had to rise to erect bipedal stance to eat and drink for 12 weeks. Dual-energy X-ray absorptiometry (DEXA) and peripheral quantitative computerized tomography (pQCT) were used to estimate the lean leg mass and bone mineral. Static and dynamic histomorphometry were performed on the triple-labeled undecalcified sections. We found that making the intact rats rise to erect bipedal stance for feeding increased muscle mass, cortical bone volume, and periosteal bone formation. Orchidectomy increased net losses of bone next to the marrow by increasing bone turnover. Making the ORX rats rise to erect bipedal stance increased muscle mass, partially prevented cancellous bone loss in the PTM, and prevented net cortical bone loss in TX induced by ORX by depressing cancellous and endocortical high bone turnover and stimulating periosteal bone formation. The bone-anabolic effects were achieved mainly in the first 4 weeks in the PTM and by 8 weeks in the TX. These findings suggested that making the rats rise to erect bipedal stance for feeding helped to increase muscle mass and cortical bone mass in the tibias of intact rats, increase muscle mass, and partially prevented cancellous and net cortical bone loss in ORX rats.
BETWEEN THE fourth and sixth decades, osteoporosis in men was not as common as in women, but starting from the sixth decade, the risks of hip fractures increased rapidly in men.(1–4) Among other risk factors, androgen deficiency, reduction of physical activities, and body weight help to increase hip fractures because of the decrease in bone mineral density (BMD).(5–11) The bone loss patterns in young or old orchidectomized (ORX) rats have been found to mimic the bone loss after androgen depletion in men and ORX rats have been used to study androgen effects on the development or maintenance of the male skeleton.(12–14) The bone loss induced by castration can be prevented by hormone-replacement therapy (androgen, testosterone, 5α-dihydrotestosterone, and estrogen),(15–19) calcitonin,(20) parathyroid hormone,(21) and prostaglandin E2 (PGE2).(22) Exercise has been reported to help prevent bone loss induced by ovariectomy.(23–29) Treadmill running might be able to prevent ORX-induced bone loss in 3-month-old young, growing rats,(30, 31) given that the training regiment was intense enough. So far, there is no report on erect bipedal exercise on ORX-induced bone loss.
The ways to exercise rats include treadmill running,(32–34) jumping,(35) swimming,(36) climbing ladder,(37) voluntary wheel running in cages,(38) and overloading one limb by immobilizing the other.(39, 40) However, rats do not cooperate well with such exercises, so it is hard to get optimal results. One other problem also exists. The exercise studies are labor intensive and time-consuming. For example, the treadmill running and swimming studies took from 30 minutes to 1 h/rat per day,(32–34, 36) jumping(35) took about 5 minutes/rat per day, and climbing the ladder took about 15 minutes/rat per day.(37) To minimize the time and labor problems and at the same time exercise the rats efficiently, we raised the cages so the rats had to stand on their hind limbs whenever they needed food or water. Minimal technician time and labor was involved in using this approach to exercise the rats. Furthermore, this exercise requires no surgical or other form of intervention and no regional accelerating phenomenon occurs. Also, there is no major change of the normal environment. It is less stressful than enforced treadmill or wheel running. The model also mimics the upright human posture.
Accordingly, this study investigated (1) the histomorphometric profiles of male rats between 6 and 9 months of age; (2) the effects of ORX on muscle mass, indices of bone strength, and bone histomorphometric profiles of rats of similar ages; (3) and the effects of making the rats rise to erect bipedal stance for feeding on muscle mass, indices of bone strength, and bone mass of selected bone sites (proximal tibial metaphysis [PTM] and tibial shaft [TX]) in both intact and ORX rats.
MATERIALS AND METHODS
Animals and experimental protocol
Thirty male 1-month-old Sprague-Dawley rats were acclimated to local vivarium conditions (Simonsen Laboratories, Gilroy, GA, U.S.A.). They were housed six per cage (58 × 36 × 20 cm) with the room temperature maintained at 72°F and 12:12 light/dark cycles. The rats were allowed free access to water and pelleted commercial natural diet (Teklad Rodent Laboratory Chow 8604; Harlan Teklad, Madison, WI, U.S.A.), which contains 1.46% calcium, 0.99% phosphorus, and 4.96 IU/g of vitamin D3. At 6 months of age, the rats were divided randomly into the following five body-weight-matched groups with six rats per group; (1) Baseline control (Baseline); (2) Sham-ORX + housed in the normal-height cage (Sham + NC); (3) bilaterally ORX + housed in the NC (ORX + NC); (4) Sham-ORX + housed in the raised cage (Sham + RC); (5) and ORX + housed in the RC (ORX + RC). The study lasted for 12 weeks.
The baseline rats received 20 mg/kg of demeclocycline (Sigma Chemical Co., St. Louis, MO, U.S.A.) on 14 days (−14 days) and 10 mg/kg of Calcein (Sigma Chemical Co.) on 4 days (−4 days) before being killed. All other rats received 90 mg/kg xylenol orange (Sigma Chemical Co.) at the beginning of the study (day 0) and 20 mg/kg of demeclocycline on 14 days (day 70) and 10 mg/kg Calcein on 4 days (day 80) before being killed.
At first, the rats in the sham + RC and ORX + RC groups were housed in the specially designed cages with the initial height of 28 cm (58 × 36 × 28 cm) for 1 week. Then the heights of the cages were raised 2.5 cm every third day until they reached the final heights of 35.5 cm. At this height, the rats needed to stand upright on their hind limbs and reach upward for food and water that were at the heights of 22–29.5 cm (Fig. 1). This height was maintained for 10 weeks. During the first 2 weeks, the body weights were recorded daily and then after that they were recorded weekly. The rats were observed carefully to be sure that they were able to stand upright and reach the food and water.
Dual-energy X-ray absorptiometry and peripheral quantitative computerized tomography measurements
The bone mineral content (BMC) and BMD and lean leg mass of the right tibias (from the proximal end to the distal end of the tibia) were measured by dual-energy X-ray absorptiometry (DEXA) using a bone densitometer adapted for small bone animal research (pDEXA; Norland Medical Systems, Fort Atkinson, WI, U.S.A.). The measurements were carried out at baseline and at 4, 8, and 12 weeks in rats anesthetized by an intraperitoneal injection of ketamine (50 mg/kg) and xylazine (10 mg/kg). Throughout the study, the scans were performed with a scan speed of 20 mm/s and a resolution of 0.5 mm × 0.5 mm.
The trabecular and cortical BMD and BMC measurements were carried out using an XCT-960A peripheral quantitative computerized tomography (pQCT; Norland Medical Systems) on the PTM at 5 mm distal to the knee joint and on the TX 1 mm distal to the tibiofibular junction at 0, 4, 8, and 12 weeks. A voxel size of 0.295 mm was chosen to scan the PTMs and a voxel size of 0.148 mm was chosen to scan the TXs. A threshold of 0.930 cm−1 for cortical bone and a threshold of 0.630 cm−1 for cancellous bone were used throughout the experiment. With repositioning, the in vivo precision of the BMC and BMD of total bone, trabecular, and cortical regions ranged from 1.9% to 6.1%. Because the cross-sectional moment of inertia (CSMI) of long bone is related to the architectural fitness of the modeling-dependent diaphyseal design,(41–44) we used it as an index to estimate bone strength.
During the study period, we lost five rats from anesthesia with pDEXA and pQCT scanning.
At autopsy the rats were anesthetized by an intraperitoneal injection of ketamine (50 mg/kg) and xylazine (10 mg/kg) and killed by cardiac puncture. The lung, heart, thymus, liver, spleen, kidneys, adrenal glands, testicles, and quadriceps, gastrocnemius, soleus muscles were removed and weighed separately.
The right proximal tibiae (PTs) and the middle third of the right tibiae were stained with Villanueva bone stain, dehydrated in graded concentrations of ethanol, defatted in acetone, and embedded in methyl methacrylate (Fisher Scientific, Fairlawn, NJ, U.S.A.). Longitudinal sections of PTs and cross-sections at the tibio-fibular junction of the TXs were cut to 230 μm thickness using a low-speed metallurgical saw and then ground to 20 μm (PT) and 30 μm (TX) for histomorphometric measurement.(45)
Histomorphometry was done with a semiautomatic image analysis system (OsteoMeasure; OsteoMetrics, Inc. Decatur, GA, U.S.A.) linked to a microscope equipped with transmitted and fluorescence light.
Cancellous bone measurements
The region of the PTM that was studied was from 1 to 4 mm distal to the growth plate-metaphyseal junction. Static measurements included total tissue area (T.Ar), bone area (B.Ar), and bone perimeter (B.Pm). Dynamic measurements included single-labeled perimeter (sL.Pm) and double-labeled perimeter (dL.Pm), eroded perimeter (E.Pm), and interlabel width (L.Wi 70- to 80-day interval). These indices were used to calculate percent trabecular bone area (%B.Ar/T.Ar), trabecular number (Tb.N), trabecular width (Tb.Wi), and trabecular separation (Tb.Sp); percent mineralizing perimeters (%sLPm, %dL.Pm, and %L.Pm/B.Pm); percent E.Pm (E.Pm/B.Pm), mineral apposition rate (MAR; 70- to 80-day interval); bone formation rate (BFR) per unit of bone area (BFR/B.Ar); BFR per unit of T.Ar (BFR/T.Ar); BFR per unit of bone surface (BFR/B.Pm); and activation frequency (Ac.f) according to Parfitt et al.(46, 47) Longitudinal growth rate (LGR) was derived by measuring the L.Wi at the growth plate-metaphyseal junction.(48)
Cortical bone measurements
Cortical bone measurements included total cross-sectional area (T.Ar), cortical width (Ct.Wi), E.Pm, sL.Pm and dL.Pm, and L.Wi (70- to 80-day interval). These parameters were used to calculate the cortical B.Ar (Ct.Ar), marrow area (Ma.Ar), percent Ct.Ar (%Ct.Ar), percent Ma.Ar (%Ma.Ar), percent mineralizing perimeter (%L.Pm/B.Pm), MAR (70- to 80-day interval), and BFR per unit of bone surface (BFR/B.Pm) according to Jee et al.(49) In addition, because newly apposed bone was seen at the periosteal surfaces of TXs in the RC groups, we also measured the modeling-dependent new periosteal BAr (Ps-NB.Ar) and width (Ps-NB.Wi), and IrLWi (0- to 80-day interval). These parameters were defined as the BAr and distance between the pretreatment label (xylenol orange) and the last label (calcein). Because there were no obvious changes of intracortical region within groups, we did not perform the quantitative measurement.
Results are presented as mean ± SD. The statistical differences among different groups were evaluated using analysis of variance (ANOVA) with the Fisher projected least significance difference (PLSD) test. The value of p < 0.05 was considered significant.
Baseline and Sham + NC controls
The control animals were growing slowly with LGR of the PTM of 5.7 μm/day, bone resorption surface (%E.Pm) of 3.8%, BFR (BFR/B.Pm) of 14 μm/day × 100, and bone turnover rate (BFR/B.Ar) of 212.8% per year. Endocortical bone resorption surface of the TX was 5.5%, and BFR (BFR/B.Pm) was 7 μm/day × 100 at the periosteal surface and 3 μm/day × 100 at the endocortical surface. Except for the 31% decrease in LGR between 6 and 9 months of age, no other age-related changes of body weight, select muscle weights, bone mineral, and bone “mass” were found in sham-operated control rats (Figs. 2, 3, and 4; Tables 1, 2, 3, 4, 5, 6, 7, 8, and 9).
Table Table 1.. Effects of ORX and RC on Body Weights and Muscle Weights in 6-Month-Old SD Rats
Table Table 2.. Static Histomorphometric Changes of PTMs
Table Table 3.. Statistical Comparisons of Static Histomorphometric Indices of PTMs
Table Table 4.. Dynamic Histomorphometric Changes of PTMs
Table Table 5.. Statistical Comparisons of Dynamic Histomorphometric Indices of PTMs
Table Table 6.. Static Histomorphometric Changes of TXs
Table Table 7.. Statistical Comparisons of Histomorphometric Indices of TXs
Table Table 8.. Dynamic Histomorphometric Changes of TXs
Table Table 9.. Statistical Comparisons of Histomorphometric Indices of TXs
ORX + NC compared with the Sham + NC
From week 4 and on, lean mass and the BMC and BMD of the right tibia were lower by 10, 11, and 12%, respectively (Fig. 2). The BMC and BMD of the PTM were lower by 48% and 31% from week 4 thereafter (Fig. 3). The BMC, BMD, and CSMI of the TX changed nonsignificantly (Figs. 3-4).
At week 12, the LGR, %B.Ar/T.Ar, and Tb.N of the PTM was lower by 31, 64, and 60%, respectively; whereas TbSp was higher by 226% (Tables 2 and 3). The %E.Pm, %dL.Pm, BFR/B.Ar, and Acf were higher by 84, 276, 41, and 614%, respectively (Tables 4 and 5).
In the TX, Ma.Ar and %Ma.Ar were higher by 23% and 30%, whereas Ct.Ar, Ct.Wi, and %Ct.Ar were lower by 11, 9, and 6%, respectively (Tables 6 and 7). The endocortical %E.Pm and %L.Pm were higher by 346% and 59%, and the MAR (70- to 80-day interval) and BFR/B.Pm were higher from being undetectable in controls to 80 μm/day × 100 and 27 μm/day × 100, respectively (Tables 8 and 9).
Sham+RC compared with the Sham + NC
Body, quadriceps, and gastrocnemius weights were higher by 21, 12, and 27%, respectively (Table 1). Lean leg mass was higher by 10% at week 12 (Fig. 2). The BMC of the TX was higher by 3% at week 4, 21% at week 8, and 23% at week 12 (Fig. 3). The CSMI was higher by 20% at week 4, 74% at week 8, and 50% at week 12 (Fig. 4).
At week 12, the static and dynamic histomorphometry of PTM only differed in the %E.Pm, which was lower by 35% (Tables 2, 3, 4 and 5)
The Ps-NBAr was 0.16 mm2 (3.5%), which added 53 μm to the cortical thickness (Tables 6 and 7). The only dynamic histomorphometric change of the TX was that the MAR between the xylenol orange label and calcein label (0–80 days interval) was higher from being undetectable double labels in controls to 0.6 μm/day (Tables 8 and 9).
ORX + RC compared with the Sham + NC
Body, quadriceps, and gastrocnemius weights were higher by 13, 11, and 24%, respectively (Table 1). BMC and BMD of the PTM were lower by 24% and 13% at week 4 and then stabilized.
At week 12 the LGR, %B.Ar/T.Ar, and TbN of the PTM were lower by 33, 12, and 16%, respectively, whereas the TbSp was higher by 23% (Tables 2 and 3). Dynamic histomorphometry only differed in the %E.Pm, which was lower by 24% (Tables 4 and 5).
In the TX, Ma.Ar, endocortical %E.Pm, and %L.Pm were higher by 16, 138, and 50%, respectively (Tables 6, 7, 8, and 9). MAR (70- to 80-day interval) and BFR/B.Pm were higher from being undetectable in controls to 0.3 μm/day and 8.8 μm/day × 100, respectively (Tables 8 and 9).
ORX + RC compared with the ORX + NC
At week 4, the lean body mass, BMC, and BMD of the right tibias were higher by 10, 12, and 8%, respectively, and remained at these levels for the next 8 weeks (Fig. 2). From week 4 and on, the BMC and BMD of the PTM were higher by 44% and 26%, respectively (Fig. 3). The CSMI was higher by 62% at week 12 (Fig. 3).
At week 12, the %B.Ar/T.Ar and Tb.N of the PTM were higher by 135% and 94%, whereas Tb.Sp was lower by 60% (Tables 2 and 3). Bone resorption and formation indices and Acf were all significantly lower, varying between 25% and 83% respectively, (Tables 4 and 5).
The T.Ar, Ma.Ar, Ct.Ar, Ct.Wi, and %Ma.Ar of the TX were all higher, varying between 4% and 16%, respectively (Tables 6 and 7). The endocortical %E.Pm, MAR (70–80 days interval), and BFR/B.Pm were lower by 47, 64, and 67%, respectively (Tables 6 and 7). The MAR between xylenol orange label and calcein label (0- to 80-day interval) increased from being undetectable in controls to 0.5 μm/day (Tables 8 and 9).
ORX + RC compared with the Sham + RC
There were no significant changes between these two groups in the body and muscle weights and bone mineral (Figs. 2, 3, 4 and Table 1). Histomorphometric changes of PTM only differed in Acf, which was higher by 44%, and C.Wi, Ct.Ar, and NB.Wi of the TX were lower by 9, 8, and 24%, respectively; whereas the endocortical %E.Pm was higher by 136% (Tables 2, 3, 4, 5, 6, 7, 8, and 9).
This study showed that making the rats rise to erect bipedal stance for feeding improved muscle mass, estimated indices of bone strength, and increased cortical bone mass in theTXs because of new periosteal bone formation. ORX of rats at 6 months of age tended to decrease indices of muscle mass (lean leg mass) and estimated indices of bone strength (CSMI) and induced high-turnover bone loss in cancellous bone of the PTM and cortical bone of the TX. Making the ORX rats rise to bipedal stance increased muscle mass and reduced cancellous bone loss in the PTM and net cortical bone loss in TXs. These effects were accomplished by suppressing bone turnover (i.e., BFR/B.Ar and Acf) induced by ORX on trabecular and endocortical surfaces (i.e., bone next to the marrow) and stimulating modeling-dependent bone formation at the periosteal surface. The most rapid stage of bone gain occurred at the first 4 weeks in the PTM and by 8 weeks in the TX. These findings suggest that making the rats rise to erect bipedal stance for feeding helped to increase indices of bone strength and mass in intact rats and to prevent decreases of the estimated indices of bone strength and mass in orchidectomy rats.
Our previous study showed that male Sprague-Dawley rats gain their peak bone mass by 7 months of age.(50) In this study, the body weights of the rats, the lean mass, the bone density, and the bone leg mass of the PTM and TX did not change between 6 and 9 months of age. Hadaman et al. reported that the growth plate of the male Wistar rats did not close until they were 24 months old,(51) which suggested that some male rats might keep growing throughout their entire life span. Our results indicated that there was a tendency toward decreased LGR with age, but with no age-related bone loss or gain occurring between 6 and 9 months, it suggested that these male rats had gained their peak bone mass by 6 months of age. Although some slow growth did occur, bone mass did not change significantly during this period.
It has been shown that ORX can induce cancellous bone loss in young, growing, or old rats by increasing bone resorption with or without increasing bone formation.(52–54) In this study, we found that after 12 weeks of ORX, there was a 64% cancellous bone loss in the PTM (Tables 2 and 3). Dynamic histomorphometry indicated that the bone loss was accomplished by high bone turnover, as suggested by increased eroded surface, dL.Pm, BFR (BFR/B.Pm), bone turnover (BFR/B.Ar), and Ac.f (Tables 4 and 5). In theTX, ORX induced 11% of the cortical bone loss (Tables 6 and 7). This was accomplished with increases of bone resorption and bone turnover (Tables 8 and 9). Similar to age-related bone loss seen in humans,(55) ORX-induced bone loss occurred next to the marrow and therefore decreased the cortical width. In this respect, ORX may be a better model to study bone loss in cortical bone, because it does not affect the periosteal bone formation. This is unlike OVX, in which net bone loss was not so apparent because increased periosteal bone formation offset increased endocortical bone loss.(56, 57)
This seems to be the first report of muscle and histomorphometric data on the tibias of rats by making them rise in an erect bipedal stance instead of by running all four limbs on treadmill, wheel, or by swimming and jumping. We found that in sham-operated rats, the indices of muscle strength (lean leg mass) improved after just 4 weeks of exercise; the estimated indices of bone mass and bone strength (BMC, BMD, and CSMI) of the tibia improved from 8 weeks (Figs. 4 and 5). At week 12, all animals had new periosteal bone on theTXs, thickening and strengthening the cortices. Compared with the ORX rats housed in NCs, muscle mass of the rats housed in RCs increased at week 4, as did bone mass (BMC and BMD) and estimated bone strength (Figs. 2, 3, and 4). At week 12, bone losses were only 12% in the PTM whereas no net bone loss was seen in TX. In contrast, bone losses in the rats housed in NCs were 64% in PTM and 11% in the TX (Tables 2, 3, 4, 5, 6, 7, 8, and 9). Because of the labeling schedule, most of our dynamic data were derived between 70 and 80 days of the study. Because the dynamic histomorphometric changes had already reached their steady stage, we could not detect the early dynamic changes of RCs on the animals. However, the periosteal MAR of the TX and the cortical area increased between week 0 and week 12 (0- to 80-day interval) in the RC group compared with undetectable in the NC groups (Tables 8 and 9); we postulated increased bone formation may occur in the early stage of the study. In ORX rats, exercise by making the rats rise to erect bipedal stance depressed high bone turnover induced by ORX, that is, decreased the eroded surface, MAR, BFR, and Acf. Moreover, like its effects on the sham animals, it also stimulated periosteal bone formation. Our results differ somewhat from the previous reports: Tuukanen et al.(30) reported that low-intensity treadmill running (10 m/minute, 60 minutes/day for 56 days) was not significant enough to prevent bone loss after ORX in 3-month-old rats, whereas Horcajada et al.(31) reported that by increasing the intensity of the running (30 m/minute, 60 minutes/day for 105 days), ORX-induced trabecular and cortical bone loss could be prevented. This prevention effect was achieved mainly by the inhibition of bone resorption. No increased bone formation was observed in their study.
Frost proposed that bone structure be maintained by a negative feedback system that keeps mechanical strains from exceeding minimum effective strains.(58–61) Bone modeling drifts and remodeling each have strain thresholds (set points). Where bone strains exceed a modeling threshold range, modeling usually begins to increase bone strength and bone mass via lamellar bone formation or woven bone formation drifts. The strain level for modeling lies in the 1500- to 3000-microstrain region. In contrast, the strain threshold for bone remodeling is in the 100- to 300-microstrain region. Larger strain would depress bone remodeling and preserve bone tissue. Although we could not measure the magnitude of loads at different bone sites in this study, we found that formation drifts increased at the periosteal surfaces of the TX, while bone resorption was depressed on the trabecular and endocortical surfaces. These observations suggest that the muscle force on bone generated by rising to erect bipedal stance in rats caused strains in the lower limbs that exceeded the bone modeling thresholds and therefore bone mass increased. Another possible explanation of this bone-anabolic effect involved interstitial fluid pressure. Iwamoto et al. reported that the bone gained in the distal tibial metaphysis (a point far below the heart, higher fluid pressure) exceeded that in the PTM in young rats subjected to treadmill running.(62) By ligating the femoral vein, Bergula et al. found that the femoral bone density correlated strongly with femoral intramedullary pressure.(63) By making the rats rise to erect bipedal stance, gravity could further increase interstitial fluid pressure in lower limbs and therefore might help to increase bone mass.
It has been proposed and seems to be true that muscle force is the main determinant of the postnatal and whole-bone strength and bone mass.(64–67) In our previous immobilization studies, we found that the loss of muscle weight preceded the bone loss and the recovery of muscle preceded the recovery of bone mass.(68) In this study, we found that lean leg mass and BMC measured by pDEXA correlated strongly, suggesting that the higher bone mass could relate to the increase of lean leg mass. Similar results occur in humans.(69–72) Moreover, the wet weight of the quadriceps increased significantly in the RC groups. The quadriceps seem very important in body balance and in preventing falls and fall-related fractures;(73) so increasing the quadriceps strength should be taken into account when considering treatments for bone loss.
In summary, ORX tended to decrease the estimated indices of muscle strength and bone strength and induced high bone turnover cancellous and cortical bone loss in 6-month-old rats. Making the rats rise to erect bipedal stance for feeding helped to increase muscle mass and bone mass in tibias of the intact rats, increased muscle mass, and partially prevented cancellous and net cortical bone loss in ORX rats.