This study was designed to investigate the effect of high-impact and low-repetition jump training on bones in ovariectomized (OVX) rats. Forty female Wistar rats were sham-operated (sham) or OVX at the age of 11 weeks. The rats were divided randomly into the following four groups: sham-sedentary (SS; n = 10), sham-exercised (SE; n = 10), OVX-sedentary (OS; n = 10), and OVX-exercised (OE; n = 10). The rats started the jump training at the age of 12 weeks. The jump-training protocol was 10 times/day, 5 days/week and the jumping-height was 40 cm. After 8 weeks of training, the mass and breaking force in the tibia and ulna, cross-sectional areas of diaphysis in the tibia, and serum bone turnover markers were measured. The jump training significantly increased the fat-free dry weight, ash weight, and ultimate breaking force in the tibia. The rate of increase in these parameters was similar in both the sham and the OVX groups. On the other hand, in the ulna, there were no significant changes in the ultimate breaking force. The jump training significantly increased the periosteal perimeter and cortical area, although the increase in these parameters in OE compared with OS was lower than that in SE compared with SS. The jump training significantly increased serum osteocalcin in the OVX groups, as well as in the sham groups. These results suggest that high-impact and low-repetition training had beneficial effects on bone formation and bone biomechanical properties in OVX rats, as well as in sham rats.
NUMEROUS STUDIES have reported that physical exercise increases bone mass or density in humans and animals. One of the most important reasons why exercise increases bone mass is the mechanical stress caused by physical exercise.(1) Therefore, high-impact exercise has been studied closely to increase and maintain bone density,2-5) or prevent bone loss after menopause.(6) However, there are few studies on impact training that have evaluated the effects on postmenopausal women,(3, 6) although several studies reported the positive effects of other types of exercise in postmenopausal women.7-8) Bassey et al.(3) reported that high-impact training (jump training) had a beneficial effect for premenopausal women, although it did not affect postmenopausal women. Thus, the adaptation to high-impact training in the postmenopausal state is unclear.
Ovariectomized (OVX) animals are presented as a model of postmenopausal women. In many animal studies using OVX rats to study exercise effects on bone, the type of exercises were almost all treadmill running and were mainly aerobic and endurance exercises.9-14) However, it is doubtful whether running creates sufficient mechanical stress to increase bone mass. Some studies reported the strains at the limbs measured during running in vivo.15-17) According to their reports, the strains created by running might be below or a little over the minimum effective strain (modeling threshold) suggested by Frost,(1) although they varied in locomotion, speed, age, or spices. In addition, running training gave different results according to the training intensity, duration, period, etc.(10, 11, 13, 14)
Turner(18) pointed out the following three rules for adaptation to mechanical stress: (1) bone adaptation was driven by dynamic loading, (2) only a short duration of mechanical loading was necessary to initiate an adaptive response, and (3) bone cells accommodate to a customary mechanical loading environment. The loading imposed by jump training was dynamic, of short duration, and uncustomary. Moreover, compared with running training, jump training gave a greater mechanical stress and higher strain rate; therefore, greater increases in bone mass and strength were expected.19-20) On the other hand, the aerobic component seemed to have little connection with jump training, because only five jumps per day increased bone mass.(20) The purpose of this study was to investigate the effect of high-impact and low-repetition training on bones in OVX rats.
MATERIALS AND METHODS
Forty female Wistar rats, aged 10 weeks, were obtained from Japan SLC, Inc. (Hamamatsu, Japan). The rats were housed individually in standard cages under constant temperature (23 ± 1°C). The light/dark cycle was 12:12 h with darkness from 6:30 a.m. to 6:30 p.m. Food (CE-2; CLEA, Inc., Hamamatsu, Japan) and water were provided ad libitum. After 1 week of stabilization, they were divided randomly into the following four groups (10 rats/group): (1) sham-operated (sham) sedentary (SS), (2) sham-exercised (SE), (3) OVX-sedentary (OS), and (4) OVX-exercised (OE). The rats were OVX or sham under anesthesia with pentobarbital sodium via an abdominal approach when aged 11 weeks. One week after the operation, exercise training with 10 jumps daily was started. The exercise-training period was 8 weeks when the rats were from 12 to 20 weeks old. At the end of the experiment, the rats were anesthetized with diethyl ether and killed by exsanguination. Both tibias, the right ulna, and the uterus were dissected from each rat. Immediately after careful removal of soft tissue, the ultimate breaking force of the right tibia was measured. The left tibia and right ulna were stored at −80°C until the other measurements were performed. Each uterus was weighed immediately after dissection. The experimental protocol was approved by The Animal Subjects Committee at Chukyo University Graduate School of Health and Sport Sciences.
The rats in the exercise groups were jumped 10 times/day, 5 days/week for 8 weeks. The jump-training protocol was reported in detail by Umemura et al.(19) The height of the jumping board was 40 cm throughout the experiment (maximum height was about 45-50 cm). Initially, the rats jumped with electrical stimulation, but a few days later, they jumped without stimulation. The sedentary groups were handled only with the same frequency as the exercise groups. The training and handling were started at the age of 12 weeks.
Mechanical testing procedures
After the length of the right tibia and ulna were measured with sliding calipers, the bones were loaded for three-point bending until fracture using a servocontrolled electromechanical testing system (RX1600 I; Techno Corp., Tokyo, Japan). Tests were conducted on both bones at the midlength. The distance between the bottom supports was 16 mm for the tibia or 8 mm for the ulna, and the crosshead speed for all tests was 10 mm/minute. The maximum force at the ultimate breaking point was recorded.
Bone mass and bone morphometry
After the fracture test, the right tibia and ulna were immersed in a solvent (2 vol of chloroform combined with 1 vol of methanol) for 1 week and then dried at 80°C for 24 h. Then, the fat-free dry weights (FFDW) were measured. For cross-sectional analysis, the right tibias were embedded in polyester resin (Rigolac 2004; Okenshouji, Inc., Tokyo, Japan) by submersion at room temperature for 3 days after restoration with a bonding agent. The midshaft cross-section of each bone was cut within 1 mm distal to the site of the fracture, which was premarked before the fracture test. The cross-section was photographed and enlarged. A digitizing pad was used to determine the medullary and cortical area, the endosteal and periosteal perimeters, and moment of inertia.(21) The bending stress was then calculated.(20) The left tibia and right ulna were burned at 800°C for 24 h in a muffle furnace (MPN-2N; Shimazu Co., Ltd., Tokyo, Japan) and ash weights were obtained.
Blood was allowed to clot for 20 minutes and spun at 3000 rpm for 15 minutes at room temperature, and serum was obtained. All samples were stored at −80°C until analysis. Serum osteocalcin was measured with a double-antibody radioimmunoassay (RIA) using the rat standard (Biomedical Technologies, Inc., Stoughton, MA, USA) and pyridinoline cross-linked carboxy-terminal telopeptides of type I collagen (ICTP) was measured with an RIA using the human standard (Orion Diagnostica, Oulunsalo, Finland).
Data are presented as means ± SD. A two-way (OVX × jump training) analysis of variance (ANOVA) was used to examine the individual main effect and interaction between these factors by SPSS 9.0 J. for Windows. If a significant interaction was found, the effects of training or OVX were assessed by post hoc analysis. A significance level of p < 0.05 was used for all statistical tests.
Body weight and uterine weight
All groups had a similar initial body weight (means, 209-212 g). The final body weight in the OVX groups (OS = 299.2 ± 19.6 g and OE = 303.1 ± 19.4 g) was significantly higher (p < 0.01) than the sham groups (SS = 260.9 ± 11.5 g and SE = 264.4 ± 12.8 g). The uterine weight significantly decreased (p < 0.01) in the OVX groups (SS, SE, OS, and OE were 577 ± 282 mg, 503 ± 105 mg, 83 ± 1 mg, and 90 ± 20 mg, respectively). Jump training did not affect these parameters.
The FFDW, ash weights, ash weight/FFDW, and bone lengths are shown in Table 1. In the tibia, jump training influenced the FFDW and ash weight but the OVX did not. These training group parameters were significantly higher than those of the sedentary group. The percent increases of the FFDW in the exercise group compared with each sedentary group were 18% for the sham and 19% for the OVX groups, respectively. The ash weight/FFDW did not change. Significant interactions between the OVX effect and the jump-training effect were not detected in bone mass parameters.
Table Table 1.. Bone Mass and Length
In the ulna, the FFDW and ash weight were significantly higher in the training groups than in the sedentary groups. The ash weight/FFDWs in the OVX groups were significantly higher than that in the sham groups. The length of the tibia was similar among all groups. However, the ulna was slightly longer in the OVX groups than in the sham groups.
Ultimate breaking forces in the fracture test are presented in Fig. 1. OVX had no effect on this parameter. However, in the tibia, we found that jump training significantly increased the ultimate breaking force in the OVX group as well as in the sham group. The percent increases of maximum force for the sham and OVX groups, in comparison with the sedentary values, were approximately 36% and 34%, respectively. On the other hand, there were no significant differences in the ulna. There were no significant interactions between the OVX effect and the jump-training effect on the ultimate breaking force.
Cross-sectional analyses in the tibia are presented in Table 2. Jump training significantly increased the periosteal perimeter, medullary area, and cortical area so that the bone became larger to the outside. There were significant interactions in the periosteal perimeter and cortical area, as the increase in these parameters was marked in the sham groups. The moment of inertia in the training groups was significantly higher than in the sedentary groups. With respect to the bending stress, there was a significant main effect of OVX and interaction. OE was significantly higher than SE, but there were no significant differences between SS and OS.
Table Table 2.. Bone Morphometric and Mechanical Properties in the Tibia
Serum osteocalcin levels in SS, SE, OS, and OE were 20.5 ± 1.8 ng/ml, 23.6 ± 1.6 ng/ml, 25.9 ± 4.2 ng/ml, and 28.2 ± 1.9 ng/ml, respectively, and ICTP levels were 3.9 ± 0.5 ng/ml, 3.9 ± 0.4 ng/ml, 4.2 ± 0.4 ng/ml, and 4.1 ± 0.2 ng/ml, respectively. Jump training significantly increased serum osteocalcin (p < 0.01). Serum osteocalcin (p < 0.01) and ICTP levels (p < 0.05) in the OVX groups were significantly higher than those in the sham groups. There were no significant interactions between the jump-training effect and the OVX effect.
The findings in this study were that high-impact and low-repetition training had a beneficial effect on bones in estrogen-deficient rats. The increases in bone mass and mechanical force caused by jump training were similar in the sham and OVX groups, although the increase in the cortical area in the OVX groups was less than that in the sham groups. The increase in ultimate breaking force in two jump-training groups was achieved via increased geometries, not via material properties. Moreover, serum osteocalcin, an index of bone formation, was significantly higher in both training groups. These data indicated that jump training stimulated bone formation and increased bone mass and strength in the OVX rats, as well as in the sham rats.
Many studies reported the effects of running training as an exercise model on bones in OVX rats. However, the results were confusing. Generally, treadmill running was more effective in OVX rats than in sham rats. Barengolts et al.(9) reported that treadmill running at 21 m/minute, 7% grade, 40 minutes/day, 4 days/week, for 3 months had a beneficial effect on OVX rats but no effect on sham rats, except for ultimate breaking force. Similarly, Peng et al.(13) reported that running at 10 m/minute, 30 minutes/day, twice a day, 6 days/week, for 9 weeks had a positive effect on OVX rats but not on sham rats. Contrary to these reports, another study reported a similar effect in sham and OVX rats. Omi et al.(22) suggested that voluntary running exercise is effective for bone mineral density (BMD) in sham and OVX rats aged 8 months, although the running distance was one-fourth for the OVX rats compared with the sham rats. Iwamoto et al.(11) indicated that the beneficial effects of treadmill running were recognized only when the optimal level of exercise was applied in OVX rats. They studied three different running protocols—12 m/minute, 1 h/day; 18 m/minute, 1 h/day; and 12 m/minute, 2 h/day—and found only the first protocol was effective in OVX rats. Therefore, treadmill running, mainly aerobic training, has a different effect according to training intensity, duration, period, frequency, etc.
However, the principal reason for bone increase due to exercise is mechanical stress.(1) In this study, we imposed a large mechanical stress via jump-training. Moreover, the number of daily repetitions was small (10 times/day) so that the training style was nonaerobic. As a result, high-impact and low-repetition training had a beneficial effect on the OVX rats as well as the sham rats. Jump training was shown previously by Umemura et al.(19) to be more effective than running training to develop lower bone mass. Additionally, only 10 repetitions/day had a large positive effect on bones.(20) This probably was because jump training caused a great mechanical stress because the ground reaction force on the lower leg with a 40-cm jump was 499 ± 11% body weight (our laboratory data). The increase in ultimate breaking force caused by jump training was interpreted mainly by the geometric change but was not interpreted by the change in material properties because we observed a significant increase in the moment of inertia and no significant change in the bending stress. This interpretation was supported by the increased periosteal perimeter and would be strengthened by the normalized ash weight/FFDW. This geometric change accompanied by no material properties was observed not only in the sham or intact rats, but also in OVX rats.
In a human study, Bassey et al.(3) reported no jump-training effect on bone densities for postmenopausal women with and without hormone-replacement therapy (HRT) but some effects for premenopausal women. Their exercise consisted of 50 vertical jumps of 8.5-cm height, and the mean ground reaction force was three to four times the body weight. Considering their results on postmenopausal subjects with HRT, estrogen deficiency was not the main reason they could not observe positive effects in postmenopausal subjects. Yong et al.(23) reported that young ballet dancers had normal BMD in weight-bearing sites despite oligomenorrhea or amenorrhea, although amenorrheic girls with anorexia nervosa had lower BMD compared with control girls or ballet dancers. They concluded that hypogonadism might not result in generated osteoporosis, and that vigorous exercise might offset the effects of estrogen deficiency in weight-bearing sites. Our data supported the idea that high-impact, low-repetition exercise training could improve bones, even if the subjects were estrogen deficient, when the stress was strong enough and the other conditions (calcium intake and so on) were sufficient.
In this study, exercise increased bone mass and strength in the OVX and sham rats, but the increases in cortical area and periosteal perimeter in the OVX group were less than in the sham groups. Barengolts et al.(10) reported that exercise training could reduce the medullary area increase in OVX rats without changing the cortical area. However, running training at different intensities or durations did not affect either the cortical or medullary areas.(9, 10) In our jump study, both the cortical and the medullary areas increased, although the OVX had somewhat lower periosteal perimeter growth compared with the sham group.
The bone turnover markers were significantly higher in the OVX groups than in the sham groups. This indicated that OVX caused a high bone turnover. Jump training stimulated only bone formation marker in both groups. Because the training increased the serum osteocalcin level, bone formation was stimulated and a higher bone mass and strength in OVX were observed.
In conclusion, high-impact and low-repetition training increased bone mass and strength in estrogen-deficient rats and intact rats. Although the beneficial effects of exercise on biomechanical properties were observed only in weight-bearing bones, bone formation was stimulated in estrogen-deficient rats.
This study was supported by a grant-in-aid for Scientific Research (C) of the Japan Society for the Promotion of Science (11680062), Japan, 1999.