We employed skeletally matured rats to study changes in biochemical markers of bone turnover, bone mineral density (BMD), and bone biomechanics produced by continuous elevation of parathyroid hormone (PTH) in estrogen-deplete and -replete rodents. Ninety-six 7-month-old virgin female rats were divided randomly into 12 groups (n = 8) and treated as follows. One group was killed on the day of surgery. The remaining groups were either bilaterally ovariectomized (Ovx) or sham-operated and left untreated for 8 weeks, at which point, two groups, one sham and one Ovx, were killed. The remaining nine groups were treated for 2 weeks or 4 weeks. One sham and two Ovx groups received subcutaneous implants of Alzet miniosmotic pumps with vehicle for PTH. Two Ovx groups were given pumps with vehicle as well as a subcutaneous implant of 17β-estradiol, which delivered 10 μg/kg per day. Two Ovx groups were implanted with rat PTH(1–34) in Alzet miniosmotic pumps, which delivered 30 μg PTH/kg per day. Two Ovx groups were implanted with both estradiol pellets and PTH-loaded pumps. One group of Ovx animals from each treatment was killed after 2 weeks and the other after 4 weeks. Biochemical markers of bone turnover, serum osteocalcin and urinary free pyridinoline, BMD, and mechanical strength of excised bones were measured. As expected, there was a significant increase in N-terminal PTH and serum calcium levels in all PTH infusion groups. Both serum osteocalcin and urinary pyridinoline showed a rapid increase within the first 2 weeks of the PTH infusion and remained elevated at week 4. In estrogen-replete groups, osteocalcin increased by week 2 of PTH infusion but pyridinoline did not increase until week 4. BMD of the distal and proximal femur showed the expected decrease 8 weeks after ovariectomy but did not exhibit any further changes during the 4 weeks of treatment with vehicle. Four weeks of PTH infusion in Ovx animals resulted in BMD loss at the midshaft, distal, and proximal regions of the femur. Estrogen repletion by itself, beginning 8 weeks after ovariectomy, produced no change in BMD at any site when compared with from Ovx vehicle-treated rats. Estrogen repletion in PTH-infused Ovx animals resulted in significant improvements of BMD comparable with sham-operated animals at all three femoral regions. The indentation test at the cancellous bone of the distal femur, three-point bending test at the midshaft femur, and cantilever bending test at the femoral neck showed that the changes in mechanical strength in these sites were consistent to the changes found in BMD. Our results showed that (1) continuously elevated levels of PTH induced additional loss of BMD in estrogen-deficient animals beyond the rapid bone loss phase associated with ovariectomy, (2) estrogen repletion, given by implant, to PTH-infused Ovx animals, reversed these BMD changes increasing BMD to levels comparable with estrogen-sufficient rats, and (3) these changes were reflected in the mechanical strength determined at these sites. These results lend experimental support that hormone replacement therapy may benefit bone health in postmenopausal women with primary hyperparathyroidism (PHPT). In addition, it raises the possibility that a continuous elevation of PTH could exert anabolic effects on skeletal tissue if its catabolic component can be minimized.
Several studies have taken advantage of a parathyroid hormone (PTH) infusion device to produce continuously elevated PTH conditions either in normal or thyroparathy-roidectomized (TPTX) animals.(1–7) Five days of PTH infusion studies in TPTX animals produced many of the characteristics of human primary hyperparathyroidism (PHPT): modest hypercalcemia and hypophosphatemia, renal loss of phosphate, elevated 1,25-dihydroxyvitamin D [1,25(OH2)D] and increased bone turnover.(6,7) Longer periods, 2–4 weeks, of PTH infusion in rats induce significant bone loss mimicking lower bone mass observed in patients with PHPT.(1–5) PTH given as pulsatile delivery for longer than 2 h/day or in continuous fashion for a total of 6 days gave rise to abnormalities observed in patients with chronic PHPT, including peritrabecular marrow fibrosis and focal bone resorption.(5) Thus, PTH infusion in rats has been proposed as a surrogate to study PHPT in human.(6) Because a majority of PHPT patients are postmenopausal women, the possible interaction between estrogen deficiency and the elevation of PTH could be of clinical significance.(8) Furthermore, short-term hormone replacement therapy has been shown to increase bone mineral density (BMD) in postmenopausal women with PHPT.(9,10) Rats have been a successful animal model to study estrogen deficiency–induced bone loss and the efficacy of estrogen replacement therapy.(11) Although rats have been used extensively to study the anabolic action of intermittent PTH in estrogen-deficient and -replete animals, there have not been any studies on the effects of continuously elevated PTH in estrogen-deficient rats.(12) Therefore, we have undertaken this study to examine the skeletal effects of a prolonged elevation of PTH in estrogen-deficient and estrogen-replete animals to study the changes in biochemical markers of turnover and BMD.
MATERIALS AND METHODS
Virgin female Sprague-Dawley rats, 4 months of age, were purchased from Harlan Sprague-Dawley, Inc. (Indianapolis, IN, U.S.A.) and maintained at the Animal Research Facility of the Helen Hayes Hospital for an additional 3 months before the start of the experiment. They were housed in a room maintained at 20°C on 12-h light/12-h dark cycles and maintained on Laboratory Rodent Chow (no. 5001, Purina Mills, St. Louis, MO, U.S.A.) and tap water ad libitum until the onset of the experiment. The experimental protocol was approved by the Institutional Animal Care and Use Committee at Helen Hayes Hospital.
Table Table 1.. Experimental Protocol
At 7 months of age, the rats were divided into 12 groups of 8 animals each. One group was killed at the onset of the experiment. The remaining groups were then subjected to either bilateral ovariectomy (Ovx) or sham operation under 12.5:2.5 mg/kg ketamine/xylazine anesthesia, given intraperitoneally. Ketamine was supplied as Ketalar (Parke-Davis, Morris Plains, NJ, U.S.A.) and xylazine hydrochloride was purchased from Sigma Chemical Co., St. Louis, MO, U.S.A. The animals were housed two per cage and remained untreated for an additional 8 weeks to allow post-Ovx bone loss to occur. All animals were fed 15g/day per rat of Laboratory Rodent Chow, the mean food intake of the sham-operated animals during the experimental phase, to prevent the excessive hyperphagia associated with Ovx in rats. The schematic experimental protocol is as shown in Table 1. At the initiation of the treatment phase, one sham and one Ovx group were killed. One sham and two Ovx groups were subcutaneously implanted with Alzet miniosmotic pumps (model 2002, Alza Corp., Palo Alto, CA, U.S.A.) with vehicle (1 mM acetic acid). Two Ovx groups were given pumps with vehicle and an additional subcutaneous 17β-estradiol pellet implant (Innovative Research, Inc., Sarasota, FL, U.S.A.), which delivered estrogen at a rate of 10 μg/kg per day. Two Ovx groups were implanted with rat PTH(1–34) (Bachem, Inc., Torrance, CA, U.S.A.) in Alzet miniosmotic pumps, which delivered 30 μg PTH/kg per day. Two Ovx groups were implanted with both estradiol pellets and PTH-loaded pumps. Mniosmotic pumps were replaced every 2 weeks, corresponding to the functional life span of the pump. All implants, accomplished within 2 minutes in each rat, were performed under light ether anesthesia. One group of Ovx animals from each type of treatment was killed 2 weeks into the treatment and the second set was killed after 2 additional weeks. The death was performed via exsanguination of the abdominal aorta after the animals were anesthetized with ketamine/xylazine. Blood and urine samples were collected at the time of death. The right femurs were collected and stored at 4°C in saline, with 0.1% sodium azide as preservative, for later BMD and mechanical strength measurements. The success of Ovx operation was confirmed by a lack of ovarian tissue, by visual examination, after the animals were killed.
BMD of excised femurs was measured by dual-energy X-ray absorptiometry (DPX-L, Lunar Corp., Madison, WI, U.S.A.) using small animal software. Three areas of interest, the distal, midshaft, and proximal femur, were selected for analysis. Twenty-five percent of the length of the femur, as measured from the distal and proximal ends, areas rich in cancellous bone, and the remaining 50% of the femur (the midshaft region, primarily cortical bone) were analyzed. Triplicate determinations of five different femurs, with a new placement of the bone after each determination, showed a maximum coefficient of variation of 2.5%.
Rat serum calcium and urinary creatinine were measured using a routine chemistry autoanalyzer. The serum concentration of N-terminal PTH was measured using a chick anti-PTH antibody (CK67.57) in a radioimmunoassay.(13) Serum osteocalcin concentration was measured with a commercial radioimmunoassay kit (Biomedical Technologies, Inc., Stoughton, MA, U.S.A.) using rat osteocalcin as standard and a region-specific domain of goat anti-rat osteocalcin antibody. Urinary free pyridinoline was measured by an ELISA kit (Pyrilinks) from Metra Biosystems (Mountain View, CA, U.S.A.). Intra-assay coefficients of variation for rat osteocalcin and pyridinoline were 6% and 12%, respectively.
Three sites of the femurs used in BMD measurements were subjected to mechanical testing procedures using a Materials Testing System (Model 810, MTS Systems Corp., Minneapolis, MN, U.S.A.): (1) Indentation test of the distal femur. An indenter was used to test the mechanical properties of the marrow area of the distal femur, using a modification of the methods used to determine mechanical strength of knee replacements in humans.(14,15) A 4-mm section of the distal femur was cut just proximal to the femoral condyle with low-speed saw (Isomet, Beuhler LTD, Lake Bluff, IL, U.S.A.) under constant saline irrigation. A cylindrical indenter (with flat testing face) of 1.6-mm diameter, attached to a 125 Newton load cell (MTS model 3397–101) was applied to the center of the marrow cavity on the distal face of the section at a constant displacement velocity of 0.1 mm/s. The indenter was allowed to penetrate the cavity to a depth of 2 mm before load reversal. The load results were calculated from the load-displacement curve recorded. (2) Three-point bending test of the midshft femur. The midshaft of the femur was subjected to three-point bending to failure at a displacement rate of 0.1 mm/s, using the method of Turner and Burr, monitored using a 2500 Newton load cell (MTS model 661.14A-03).(16) The load results were calculated from the load-displacement curve. The moment of inertia at the midpoint of the midshaft femur was calculated using the formula MI = (3.14/64)[ab3 – (a–2t)(b–2t)3] where a is the width of the cross-section in the medial-lateral direction, b is the width of the femur in the anterior-posterior direction, and t is the average thickness of the cortical bone obtained with a peripheral quantitative computed tomography (pQCT) scan. The ultimate stress at the midshaft femur was calculated using the formula, US = (FuLb)/8(MI) where Fu is the breaking force and L is the loading distance (15 mm). (3) Cantilever compression test of the femoral neck. After the three-point bending test, a 1.2-cm segment of the proximal femur was obtained and subjected to the cantilever compression test similar to that described by Sogaard et al.(17) The proximal femurs were placed in an aluminum block with a notch to anchor the greater trochanter. The load was applied with a 2500-N load cell at a displacement rate of 0.1 mm/s until failure. The load results were obtained from the load-displacement curve.
The effects of treatments at each treatment period were compared by the analysis of variance (ANOVA) method using Duncan's multiple group comparison procedure with SAS software (SAS Institute, Inc., Cary, NC, U.S.A.). A P-value of 0.05 was used to determine statistical significance. Analyses were performed using any groups killed at the same time point. Power calculations indicated 80% power to see diffeences between groups at P < 0.05 level.
After 4 weeks of treatment, Ovx rats treated with vehicle gained more body weight than sham-operated animals despite food restriction (Ovx, 320 ± 4g; Sham, 290 ± 5g). Estrogen replacement of Ovx animals resulted in a body weight slightly less than sham-operated animals (Ovx + estrogen, 281 ± 3g). PTH-infused Ovx animals showed a reduction of body weight to that of about 90% of the sham-operated animals (Ovx + PTH, 261 ± 6g). The estrogen replacement in PTH-infused animals showed a further small decrease (Ovx + PTH + estrogen, 247 ± 4g). All animals survived the experiment and no noticeable discomfort was evident.
PTH infusion in Ovx rats, regardless of estrogen status, induced a similar significant elevation of serum calcium (Fig. 1A). N-terminal PTH levels were higher in all PTH infusion groups, as expected (Fig. 1B). The bone formation marker osteocalcin was elevated in the Ovx group treated with vehicle when compared with the sham group and this elevation was blunted by estrogen repletion (Fig. 1C). PTH infusion induced significant elevation of the serum osteo-calcin level at each of time points. The estrogen-replete groups showed a less robust response at 2 weeks when compared with the estrogen-deficient group. Free pyridino-line, a bone resorption marker, showed a nonstatistically significant increase in the Ovx groups when compared with the sham group and this increase was again reduced by estrogen repletion (Fig. 1D). PTH infusion resulted in a significant increase of urinary free pyridinoline in the PTH infusion group but not in the PTH infusion group repleted with estrogen at week 2. By week 4 of the treatment, excretion of free pyridinoline was significantly elevated in both groups.
BMD measurements at the distal and proximal femur, both rich in cancellous bone, exhibited similar responses to the treatments (Figs. 2A and 2B). As expected, 8 weeks of estrogen deficiency resulted in a significant loss of BMD in the Ovx group. There was little further bone loss during weeks 8–12 post-Ovx in the vehicle-treated groups. Consequently, estrogen repletion during these 4 weeks did not produce any significant effects on BMD. PTH-infused animals continued to lose bone in an almost linear fashion, which gained statistical significance after 4 weeks, except in the midshaft region, where there did not appear to be bone loss in the initial 2 weeks. Estrogen repletion coupled with PTH infusion protected the skeleton from the detrimental effects of PTH and by 4 weeks of treatment BMD was restored to the Sham control levels at all three sites (Figs. 2A–2C).
Three different sites of the femur were subjected to mechanical testing after BMD measurements. Ovx-induced significant loss of the cancellous bone strength at the distal femur before the onset of the treatment. Among all treatments, only estrogen replacement in the PTH-infused Ovx animals restored the strength to the level of sham-operated animals at week 4 (Fig. 3A). Cantilever compression testing performed on femoral neck showed no significant changes in loading strength after Ovx with or without estrogen replacement. PTH-infused animals experienced a significant decrease in femoral neck loading strength and estrogen replacement in these animals showed an initial decrease of the strength in week 2 but it was restored to the levels of sham-operated animals at week 4 (Fig. 3b). The loading strength at the midshaft of the femur showed somewhat similar changes when the results were examined either as breaking load or ultimate stress (Figs. 3C and 3D). Ovx induced time-dependent loss of midshaft breaking strength and ultimate stress when compared with sham-operated animals, the changes being significant by week 4. Estrogen replacement of Ovx animals prevented further loss of loading strength from the point of estrogen administration. PTH-infused Ovx animals exhibited similar breaking strength and ultimate stress as the Ovx animals and estrogen replacement in these animals restored both to a level similar to that of the sham-operated animals.
PHPT in postmenopausal women has been associated with lower bone mass and short-term estrogen replacement has been suggested to be a potential treatment for bone loss in these women. Continuously elevated PTH, by infusion, in rats with intact ovarian function, has shown repeatedly to result in a catabolic response in the skeleton.(1–5) How this catabolic action might be modified in an estrogen-deficient state in rats has not been examined. In the present study, we have examined this question by infusing PTH in rats with established osteopenia caused by estrogen deficiency. Our findings suggest that a continuous elevation of PTH results in further loss of BMD and mechanical strength in estrogen-deficient animals and that estrogen repletion protects the skeleton from these detrimental effects of PTH.
Biochemical markers of bone turnover have been used to examine changes in bone formation and resorption. Serum osteocalcin has been shown to be associated with the bone formation rate and osteocalcin has been used as a formation marker to monitor drug actions.(18,19) Urinary free pyridino-line has served the same role to describe the bone resorption process.(20) At 2 weeks of PTH infusion, we saw a rapid increase of osteocalcin in PTH-infused Ovx animals, which was accompanied by a similar rapid rise in free pyridinoline. This combination of marker changes was reflected in decreased BMD in PTH-infused animals, when compared with Ovx animals treated with vehicle. In estrogen-replete animals infused with PTH the increase of osteocalcin at 2 weeks seemed to be less vigorous although still significantly elevated. In contrast, free pyridinoline did not increase at 2 weeks when PTH and estrogen were combined. In combination these biochemical changes may be the underlying mechanism for the observation of later increase of BMD. The delayed but marked increase in free pyridinoline after 4 weeks of PTH infusion in estrogen-replete Ovx animals may signal an increase in bone resorption and/or remodeling activity. This could either be the result of an attempt to modify the increase in bone mass or a delayed onset of the catabolic action of PTH in the presence of estrogen. Four weeks of treatment in rats is equivalent to approximately 2–3 years of drug intervention in humans, based on their relative life expectancy. Experiments using longer-term PTH infusion, with or without estrogen repletion, will be needed to ascertain whether the combination of PTH infusion and estrogen can provide long-lasting beneficial effects.
PHPT is a disease that occurs primarily in postmeno-pausal women.(8,21) It is associated with a lower bone mass at all skeletal sites though loss is generally more prominent at cortical bone–rich sites.(22) Although estrogen treatment has been shown to be effective in the prevention and treatment of postmenopausal osteoporosis, its role in the treatment of parathyroid-related bone disease is less established.(23,24) It has been reported, in a cross-sectional study that estrogen-replete postmenopausal women with PHPT have higher BMD than their nonestrogen-replete counter-parts.(25) Hormone replacement therapy in postmenopausal women with PHPT has resulted in a small improvement in BMD.(9,10) It has been suggested that the beneficial effects of estrogen may be the result of a reduction of plasma calcium without an accompanied increase in PTH or the result of reduced bone turnover as reflected by a reduction in radiocalcium bone turnover and decrease in bone turnover markers.(10,26–28) Longer-term studies with physiological levels of hormone replacement therapy in patients with PHPT did not alter serum calcium levels.(9,10) Thus the latter explanation, decreased bone turnover, probably better reflects the physiological reaction to estrogen replacement. Our results showed that the use of estrogen in PTH-infused Ovx animals indeed prevented an increase in bone resorption markers during the first 2 weeks of infusion. However, it was accompanied by some blunting of the PTH-stimulated increase in bone formation, a result that might be expected because estrogen is known to reduce bone formation in rats. These results with osteocalcin perhaps also suggest some de novo stimulation of bone formation, an anabolic effect known to occur with intermittent PTH administration in both rodents and humans.(12) Our findings of significant bone gain in estrogen-replete, PTH-infused Ovx animals after 4 weeks of treatment may support that conclusion, but clearly further evaluation will be required.
Bone loss differs among skeletal sites in patients with PHPT. The extent of bone loss appears to be correlated positively to the proportion of cortical bone content at each site, that is, bone loss is the highest in the forearm and the lowest in the spine.(29) Histomorphometric measurements of the iliac crest bone biopsies in postmenopausal women with PHPT shows a remarkable preservation of cancellous bone while the cortical thickness was severely reduced.(30) Ovx in rats induces a rapid cancellous bone loss at different skeletal sites with marginal effects on cortical bone.(31–33) When we supplied exogenous PTH to Ovx animals, beginning 8 weeks post-Ovx, at a time in which additional cancellous bone loss is minimal, further bone loss in the femur was observed at the two cancellous-rich regions, the distal and proximal regions, and the cortical region, the midshaft of the femur. Our results suggest that PTH infusion can induce cortical bone loss even when the effects of estrogen-deficiency are significantly diminished in these regions. Rats serve well as a model of estrogen deficiency–induced osteoporosis because their cancellous bone responds somewhat similarly to estrogen deficiency as observed in humans. Rats are not an ideal model to study changes in cortical bone, in part, because of their lack of a Haversian canal system as well as the fact that cortical bone loss has been difficult to induce in rats. Previously, Uzawa et al convincingly showed that PTH infusion produced cortical tunneling in intact rats.(2) Our results suggest that PTH infusion, in estrogen-deficient rats, could serve as an animal model to study cortical bone loss in postmenopausal women.
In general, the mechanical testing results agree with the findings with BMD with few differences. A lack of change in mechanical load in the cancellous bone at the distal femur in PTH-infused animals at week 4 of the treatment agrees with our previous suggestion that further bone loss may be occurring in cortical bone. The majority of the bone mass in the femoral neck in rats is cortical bone. The lack of change in the response to compressive load at the femoral neck in Ovx animals and Ovx animals treated with estrogen suggests a minimal effect of estrogen at this region. The differences in BMD observed in the proximal femur are most likely to be the results of cancellous bone present distal to the femoral neck. The PTH-infused animals continued to lose loading strength at the femoral neck and this result lends support to the concept that continuously infused PTH can induce cortical bone loss at the femoral neck. Qualitatively, the breaking load and calculated ultimate stress at the midshaft of the femur agree with BMD measurements at the same site, that is, estrogen replacement in PTH-infused Ovx animals protects bone strength. However, the correlation was not perfect because of the changes in the endosteal and periosteal perimeters and average cortical thickness. We noted a trend of increased endosteal and periosteal perimeters in Ovx animals and Ovx animals infused with PTH. Estrogen replacement prevented Ovx-induced endocortical and periosteal cortical expansion. The combination of estrogen replacement and PTH infusion resulted in no increase in periosteal surface and a reduction in endosteal surface. This suggests an expansion of endosteal surface by erosion or periosteal formation in estrogen-deficient animals treated with vehicle or PTH infusion. Estrogen repletion prevented these changes and combined estrogen and PTH infusion resulted in infilling of the marrow cavity without an accompanied expansion in periosteal surface. We speculate that these changes in physical dimensions resulted in the minor variations in the relationship between BMD and mechanical measurements. The precise contribution of each factor cannot be ascertained at this time because of the small sample size used in the experiment and the multiple variables involved in the expression of mechanical strength. Further experiments with larger sample sizes need to be performed in order to delineate this problem.
It is well recognized that intermittent PTH administration exerts anabolic action on bone in rats whereas continuous use of PTH results in a catabolic response.(1–5) This is in agreement with observations made from human studies in which the subjects underwent either intermittent PTH administration or had PHPT.(12) The preliminary success in using hormone replacement therapy in postmenopausal women with HPT and the present study of estrogen replacement in PTH-infused Ovx rat, raises the question of whether the previous axiom should be modified.(9,10) The use of hormone replacement therapy or a form of antiactivation agent treatment may serve to reduce the deleterious catabolic effects of PTH administration while maintaining the anabolic effects of PTH. The protective effects of hormone replacement therapy are only seen with a sustained elevation of estrogen. The question of why intact ovarian function fails to protect bone from elevated PTH, as in the case of premenopausal women with PHPT or in intact animals that have undergone PTH infusion or from bone gained as a result of the anabolic action of intermittent PTH administration, remains a question that has yet to be answered.(34)
Although our results provide interesting insights into the use of a combination of continuously elevated PTH and estrogen, there are several caveats that need to be pointed out. The dose of PTH used in our infusion study resulted in a significant increase of serum calcium well beyond the normal physiological range. Future studies would have to be designed with PTH levels at one-third to one-half of the dose described here. Although PTH infusion in this animal model showed some similarity to PHPT in humans, the adenoma in the parathyroid glands cannot be mimicked accurately. If the model is to represent HPT in postmenopausal state, further characterization of the model, including a second confirming experiment, histomorphometry at multiple sites, using animals with no longitudinal growth or open growth plates, multiple doses of estrogen, possible confounding effects of progesterone surges in rats at 7–9 months of age, and better examination of changes in mechanical strength will need to be studied.
The authors thank Dr. XG Liang for his technical assistance. This study was supported by the National Institutes of Health grant AR 39191.