An Uncoupling Agent Containing Strontium Prevents Bone Loss by Depressing Bone Resorption and Maintaining Bone Formation in Estrogen-Deficient Rats

Authors


Abstract

Trabecular bone loss in estrogen deficiency is associated with enhanced bone resorption with a smaller increase in bone formation. We previously reported that low doses of strontium can increase trabecular bone volume in rodents by affecting bone resorption and formation. In this study we determined the effect of a new divalent strontium salt (S12911) on bone loss induced by E2 deficiency. Sprague-Dawley female rats (230 g, n = 15–25 per group) were sham operated or ovariectomized (OVX) and treated with 17β-estradiol (E2, 10 μg/kg/day, sc) or S12911 by gavage at the dose of 77, 154, or 308 mg/kg/day or the vehicle. Treatment for 60 days with S12911 resulted in a dose-dependent increase in plasma, urine, and bone strontium concentrations without any deleterious effect on total or skeletal growth. OVX rats were osteopenic compared to sham rats as shown by decreased femoral dry bone weight and mineral content measured on bone ash and by DXA. Treatment of OVX rats with S12911 prevented bone loss as bone ash and bone mineral content were restored to the values in sham rats. Trabecular bone volume measured by histomorphometry on the tibial metaphysis was decreased by 46% in OVX rats and was corrected by E2. Treatment of OVX rats with S12911 increased the trabecular bone volume by 30–36%. Histomorphometric indices of bone resorption (osteoclast surface and number) were increased in OVX rats and were reduced by S12911 to the levels in sham rats. In contrast to this inhibitory effect on bone resorption, the osteoid surface, osteoblast surface, mineral apposition rate, and bone formation rate were as high in OVX rats treated with S12911 as in untreated OVX rats. In addition, plasma osteocalcin (OC) and alkaline phosphatase (ALP) levels remained elevated or were further increased in OVX rats treated with S12911. In contrast, treatment with E2 reduced both bone resorption and formation and plasma ALP and OC to the levels in sham rats. The data indicate that the divalent strontium salt S12911 is acting as an uncoupling agent that can prevent the femoral osteopenia and partially prevent the trabecular bone loss in E2-deficient rats by inhibiting bone resorption without reducing bone formation.

INTRODUCTION

Estrogen deficiency in postmenopausal women is associated with an increased rate of bone turnover and accelerated rate of trabecular bone loss, which leads to osteopenia and to an increased incidence of vertebral fractures.(1) The trabecular bone loss induced by estrogen deficiency appears to result from an imbalance between bone resorption and formation.(2) Estrogen therapy reduces the accelerated bone turnover and prevents trabecular bone loss.(3,4) This depressing effect of estrogens on bone turnover appears to be mediated by inhibition of bone resorption(5–7) with subsequent diminution of bone formation.

Since bone formation in estrogen deficiency is not as increased as bone resorption, accurate prevention of bone loss may require stimulation of bone formation together with inhibition of bone resorption. Several antiresorbing agents(8,9) were found to partially or completely prevent the oestrogen deficiency-induced bone loss in young ovariectomized (OVX) rats, an animal model with characteristics that are comparable with those of early postmenopausal trabecular bone loss.(10,11) By contrast, few agents capable of stimulating bone formation have been used in OVX rats. Parathyroid hormone (PTH)(12,13) and fluoride,(14) which have anabolic effects on bone, were shown to increase bone mass in ovariectomized rats. Until now, uncoupling agents having a depressing efect on bone resorption and a stimulatory effect on bone formation have not been tested to prevent osteopenia in estrogen deficiency. We have reported that strontium, a natural trace element,(15) is able to increase the trabecular bone volume when administered at low concentrations to normal animals.(16) Such a beneficial effect on bone mass appears to result from an inhibitory effect on bone resorption(17) and a stimulatory effect on bone formation as evaluated by bone histomorphometry.(16–18) This raises the possibility that strontium salts may have beneficial effects on the increased osteoclastic resorption and the relatively low bone formation associated with osteopenia in estrogen deficiency.

To determine whether a strontium salt may prevent the trabecular osteopenia induced by estrogen deficiency, we studied the effects of S12911, a new divalent strontium salt, on bone mineral content and on biochemical and histomorphometric indices of bone resorption and formation in ovariectomized rats. Our results show that this new strontium derivative is an uncoupling agent that can partially prevent the bone loss induced by estrogen deficiency by inhibiting bone resorption whereas the high bone formation is maintained.

MATERIALS AND METHODS

Experimental protocol

A total of 112 3-month-old Sprague-Dawley female rats weighing 230 g (Charles River Laboratories, France) were separated at random into six groups. All rats were given a random number to obscure the identification of morphologic, biochemical, and histomorphometric samples by the observers. A group of 26 rats were sham operated (controls). A group of 26 rats had a bilateral ovariectomy and were given no treatment. A group of 15 rats were ovariectomized and were injected subcutaneously every day starting 1 day after surgery with 10 μg/kg/day of 17β-estradiol (E2, Sigma) which was dissolved in absolute alcohol with the final concentration adjusted in a large volume of distilled water. Three other groups of rats (n = 15 per group) were ovariectomized and received every day starting 1 day after surgery a dose of 77, 154, or 308 mg/kg/day of a strontium salt (S12911; Institut de Recherches Internationales Servier, Courbevoie, France). This acid compound [3-(3-cyano-4-carboxymethyl-5-carboxy)2-thienyl)-3-azapentanedioic, distrontium salt] contains two strontium atoms per molecule (Fig. 1). It was given by gavage as a suspension after dilution in carboxymethylcellulose. Both OVX rats and sham-operated rats were given the same amount of solvent by gavage every day. The rats were fed a standard rat chow containing 0.6% calcium. The amount of diet eaten by the sham-operated rats was measured throughout the experiment, and the food consumption of all OVX rats was restricted to that of sham rats (pair feeding) to prevent the hyperphagy resulting from estrogen deficiency in rats. Each rat was injected intraperitoneally 8 and 3 days before sacrifice, with 20 mg/kg body weight of tetracycline (Terramycin) to obtain a fluorescent double label at the sites of active bone-forming surfaces. All animals were killed 60 days after the onset of the experiment, and blood was obtained by aortic puncture. All rats were placed individually in metabolic cages 1 day before death; the urine samples were collected for 24 h and were frozen. The success of ovariectomy in OVX rats was confirmed at necropsy by failure to detect ovarian tissue and by examination of uterine horn atrophy.

Figure FIG. 1..

Structural formula of the strontium salt S12911.

Biochemistry

Alkaline phosphatase (ALP) levels were measured by a colorimetric method. Plasma magnesium and plasma and urinary calcium and strontium were determined by atomic absorption spectrometry. Plasma osteocalcin (OC) concentrations were determined by radioimmunoassay using a specific rabbit antibody raised in our laboratory against rat bone osteocalcin. The sensitivity of the assay is 0.39 ng/ml, and the inter- and intravariabilities of the assay are 7 and 9%, respectively.(19)

Bone mineral content

The right tibia was dehydrated at 100°C for 14 h, weighed, and calcinated at 600°C for 16 h. Dry bone ashes were weighed and dissolved in 6 N HCl, and the bone ash contents of calcium, magnesium, and strontium were determined by atomic absorptiometry after dilution in 0.5% lanthanum chloride.

The right and left femora were removed, freed from soft tissues, and the length measured. They were then used to measure the bone mineral content (BMC) by dual-energy x-ray absorptiometry (DXA) using a previously validated Sophos apparatus (Sopha Medical, France).(20) Since the attenuation coefficient of strontium is higher than that of calcium, the BMC values were corrected for the strontium content in bone after calibration using tablets containing strontium (0–2%) in hydroxyapatite. The ash weights of the femora were also measured to further validate the BMC values obtained by DXA.

Bone histomorphometry

The left tibia was removed and fixed in cold neutral phosphate-buffered formaldehyde, dehydrated in ethanol in graded concentrations, and embedded undecalcified in methyl methacrylate.(16,21) Longitudinal sections (5 μm) of the proximal tibia were cut with a Jung K microtome and stained with toluidine blue, whereas 15 μm thick sections were mounted unstained for visualization under fluorescent microscopy. Histomorphometric indices of formation and resorption were determined on three different sections per rat using a Zeiss II eyepiece integrator and a semiautomatic image analyzer (Leitz AMS).(16,17) All histomorphometric parameters were measured in the metaphysis in a standardized region (length 3 mm) below the primary spongiosa at distance (50 μm) from the growth plate. Conventional histomorphometric indices were measured(22): the bone volume (percentage of bone tissue consisting of unmineralized and mineralized bone), the percentage of trabecular bone surface with osteoid (osteoid surface), with osteoblasts (osteoblast surface), or with osteoclasts (osteoclast surface), the number of osteoclasts per mm2 of bone surface, and the osteoid thickness (μm). The dynamic indices of bone formation were measured on all double tetracycline-labeled surfaces in the same area as for static histologic indices. They include the percentage of trabecular surface with two labels (double-labeled surface) and the mineral apposition rate (mean distance between double labels divided by the interval of time between the two tetracycline injections). The mineralization lag time was calculated as the ratio of osteoid thickness to the mineral apposition rate. The bone formation rate at the tissue level was calculated as the product of the double-labeled surface times the mineral apposition rate.

Statistics

All results are expressed as the mean ± standard error of the mean (SEM). Differences between mean values in untreated and treated groups were evaluated using analysis of variance followed by a multiple-comparison procedure (Dunnett's test). P values of 0.05 or less were considered statistically significant.

RESULTS

Growth

Despite pair feeding the final body weight was slightly higher in OVX rats (330.8 ± 3.2 g) compared to sham-operated rats (294.6 ± 3.9 g, p < 0.05). The body weight was similar in OVX rats treated with S12911 and in untreated OVX rats (data not shown). The skeletal growth evaluated by the femur length was not significantly different in sham rats (3.58 ± 0.01 cm), in OVX rats (3.61 ± 0.01 cm), in OVX rats treated with estrogen (3.56 ± 0.02 cm) or treated with the strontium salt (3.65 ± 0.02, 3.63 ± 0.02, and 3.65 ± 0.02 cm at the three doses used, respectively).

Biochemistry

As shown in Table 1, ovariectomy induced a rise in plasma ALP and OC, two markers of bone formation, and treatment with E2 lowered ALP and OC levels, which were not different from the values in sham rats. In contrast, treatment with S12911 increased plasma ALP by 14–32% and OC by 43–71% compared to values in untreated OVX rats (Table 1).

Table Table 1.. Plasma and Urine Biochemistry in Sham-Operated and OVX Rats Treated with the Divalent Salt S12911a
original image

As expected, strontium concentration in plasma and urine increased dose dependently in OVX rats treated with the strontium derivative, whereas it was low or undetectable in OVX and sham rats (Table 1). Ovariectomy and treatment with the strontium salt had no effect on plasma calcium or magnesium concentrations (Table 1). Urinary calcium was twofold decreased in OVX rats and was corrected by E2 treatment. Treatment with increasing doses of S12911 progressively raised urinary calcium to normal values.

Bone mineral content

As expected, ovariectomy induced a significant decrease in bone dry weight and ash weight that was associated with a reduction in bone calcium and magnesium content (Table 2 and Fig. 2A). Treatment with 17β-estradiol partially prevented the decrease in bone mineral content (Table 2 and Fig. 2A). However, the dry bone weight and bone calcium content in E2-treated rats were not completely restored to the values in sham rats (Table 2). Treatment of OVX rats with S12911 maintained the bone dry weight and ash weight as well as calcium and magnesium contents to the values in sham rats (Table 2 and Fig. 2A). This beneficial effect of S12911 on the total bone mineral content was associated with a dose-dependent increase in bone strontium content (Table 2). The bone ash weight in OVX rats was corrected at the higher dose of S12911, which achieved a Sr/Ca ratio of 0.76 ± 0.05% in the tibia.

Table Table 2.. Tibia Weight and Bone Mineral Content in Sham-Operated and OVX Rats Given E2, S12911 or Vehiclea
original image

Ovariectomy induced a significant 6% decrease in bone mineral content as measured by DXA on femora of OVX rats. E2 therapy in OVX rats slightly increased the BMC, which remained lower than in sham rats (Fig. 2B). By contrast, treatment of OVX rats with the strontium salt S12911 completely normalized the BMC, which was increased to values similar to those in sham rats (Fig. 2B). Normal BMC values in treated OVX rats were obtained at the higher dose of S12911, which gave a Sr/Ca ratio of 0.98 ± 0.08% in femora. No clear dose-dependent effect of S12911 on BMC or on bone ash weight could be seen, presumably because of the narrow range of the dosage levels used in this study.

Figure FIG. 2..

Bone ash weight (A) and bone mineral content (BMC) measured by DXA (B) on femora from sham-operated and OVX rats treated with vehicle (OVX), 17β-estradiol (E2), or S12911. Data are expressed as mean ± SEM. aSignificantly different compared to sham (p < 0.05); bsignificantly different compared to OVX (p < 0.05).

Bone histomorphometry

Ovariectomy induced a 46% decrease in the trabecular bone volume measured at the tibial metaphysis. This effect was almost completely prevented by E2 treatment because the bone volume was 71% higher in E2-treated rats compared to OVX rats (Fig. 3). Administration of the strontium salt partially prevented the trabecular bone loss induced by ovariectomy as the bone volume was 30–36% higher (P < 0.05) in rats treated with S12911 than in untreated OVX rats (Fig. 3).

Figure FIG. 3..

Trabecular bone volume measured in the metaphyseal region of the proximal tibia in sham rats and OVX rats treated with vehicle (OVX), E2, or S12911. Data are the mean ± SEM. aSignificantly different compared to sham rats (p < 0.05); bsignificantly different compared to OVX rats (p < 0.05).

As expected ovariectomy induced an increase in the indices of bone resorption that was prevented by E2 administration (Fig. 4). In OVX rats treated with the strontium salt S12911, the histologic indices of bone resorption were reduced by 13–24% compared to untreated OVX rats, and the values were not different from those in sham rats or in OVX rats treated with E2 (Fig. 4).

Figure FIG. 4..

Histomorphometric indices of bone resorption and osteoclast surface (A) and number (B) in sham and OVX rats treated with vehicle (OVX), E2, or S12911. Data are the mean ± SEM. aSignificantly different from sham rats (p < 0.05); bsignificantly different from OVX rats (p < 0.05).

In parallel with the increased bone resorption, static and dynamic indices of bone formation were increased in OVX rats. Treatment with E2 abolished the increased bone formation induced by ovariectomy (Table 3 and Fig. 5). In contrast to the inhibitory effect of E2 on bone turnover, S12911 had no depressing effect on bone formation. The osteoid surface and osteoblast surface, the mineral apposition rate, and the bone formation rate in OVX rats treated with S12911 were as high as in untreated ovariectomized rats (Table 3 and Fig. 5). Treatment with S12911 had no depressing effect on bone mineralization as evaluated by the mineral apposition rate, osteoid thickness, and mineralization lag time (Table 3).

Table Table 3.. Histomorphometric Indices of Bone Formation in Sham-Operated and OVX Rats Given E2, S12911, or Vehiclea
original image
Figure FIG. 5..

Histomorphometric indices of bone formation, osteoblast surface (A), and bone formation rate (B) in sham and OVX rats treated with vehicle (OVX), E2, or S12911. Data are the mean ± SEM. aSignificantly different from sham rats (p < 0.05).

DISCUSSION

The initial rapid phase of trabecular bone loss in estrogen deficiency has been attributed to stimulation of bone resorption that is more increased than bone formation.(5,6,23) The main effect of estrogen replacement is to suppress bone turnover, which prevents trabecular bone loss. The decreased bone formation observed in OVX rats treated with E2 appears secondary to the reduced bone resorption since E2 administration was shown to stimulate bone formation in vivo.(24,25) Inhibitors of bone resorption, such as bisphosphonates(8,26) or calcitonin,(9) were also shown to prevent osteopenia in OVX rats by inhibiting bone resorption and formation. In the present study we show that the trabecular bone loss induced by ovariectomy in rats can be partially prevented by a nonhormonal uncoupling agent that suppresses the high bone resorption but not bone formation induced by estrogen deficiency.

Osteopenia in OVX rats was evidenced at 60 days post-ovariectomy by decreased dry bone weight and bone mineral content measured either on bone ashes or by DXA and by the marked decrease in metaphyseal bone volume. Although treatment with 17β-estradiol prevented the decreased metaphyseal bone loss in OVX rats, the total bone mineral content measured by ash weight or BMC was not completely corrected by E2. The most likely explanation for the discordance between the effects of E2 on trabecular bone volume and BMC is that DXA measures combined cortical and cancellous bone, whereas the histomorphometric method measures only cancellous bone, which is mainly affected in OVX rats.(27) We found that the administration of S12911 had beneficial effects on bone mass in OVX rats. As evaluated on bone ashes and by DXA, 2 months of treatment with the strontium derivative prevented the decrease in bone mineral content in OVX rats. Although bone strontium content was increased after treatment, the rise in bone ash and bone mineral content induced by S12911 resulted mainly from an increase in bone calcium content. We also found that the loss in trabecular bone volume measured at the metaphyseal level was partially prevented by S12911. This effect was not due to stimulation of the growth rate or skeletal growth since the body weight and femur length in OVX rats treated with S12911 were not different from the values in untreated OVX rats.

The prevention of bone loss induced by S12911, an acid compound containing two atoms of strontium per molecule, was associated with a dose-dependent increase in plasma, urine, and bone strontium concentrations. Comparable dose-dependent changes were previously reported in rats, mice, and humans treated with natural monovalent strontium salts.(16–18,28) The marked increase in plasma strontium induced by S12911 treatment had no effect on plasma calcium and magnesium. By contrast, treatment with this strontium salt completely corrected the decreased urinary calcium excretion observed in OVX rats. Although low doses of strontium in normal rats does not affect urinary calcium,(16) our results show that the divalent strontium salt S12911 was able to restore normal urinary calcium in OVX rats, as did estrogen treatment.

The examination of histomorphometric indices of bone formation and resorption allowed us to evaluate the effects of the strontium derivative S12911 on bone cell activity in E2-deficient rats. As expected, ovariectomy resulted in stimulation of both bone resorption and formation. We found that treatment with S12911 prevented the increased bone resorption induced by estrogen deficiency. This indicates that the prevention of trabecular bone loss in OVX rats resulted in part from inhibition of bone resorption by the strontium derivative. This is in accordance with the findings that strontium salts may inhibit bone resorption in vitro(29) and in vivo.(16) Despite the reduction in bone resorption induced by S12911 in OVX rats, both static and dynamic indices of bone formation remained as high as in untreated OVX rats. In addition, plasma ALP and OC, which reflect bone formation in rats,(30,31) remained elevated or were further increased after treatment with S12911. This maintenance of high bone formation in OVX rats treated with the divalent strontium salt is in accordance with previous observations in normal rats(16,32) and in iliac crest bone biopsies in osteoporotic patients,(18) suggesting that strontium salts may activate bone formation in vivo. In contrast to the treatment with S12911, E2 treatment in OVX rats reduced bone formation, presumably as a result of the decreased bone resorption.(33) Inhibitors of bone resorption, such as bisphosphonates and calcitonin, were also reported to reduce both bone resorption and formation in OVX rats.(8,9) The present results indicate that the strontium derivative S12911 is an uncoupling agent that can inhibit bone resorption without reducing bone formation in E2-deficient rats.

Several similarities underlie the effects of early estrogen deficiency on bone turnover in rats and humans.(11) Since early estrogen deficiency in women is characterized by increased bone resorption with a relatively smaller increase in bone formation, prevention of postmenopausal bone loss may be more effective by using treatment with an uncoupling factor capable of inhibiting bone resorption while maintaining bone formation. Our finding that S12911 suppressed bone resorption but not bone formation and partially prevented the trabecular bone loss in OVX rats without causing deleterious effects indicates that this uncoupling agent may be of potential interest in the prevention and treatment of postmenopausal osteoporosis. This therapeutic agent may be applicable to humans since preliminary data indicate that S12911, even at high dosage levels, does not produce significant extraskeletal side effects that could contraindicate its use in humans. The potential beneficial effects of this compound on bone in osteoporosis therefore warrants further experimental and clinical investigations.

Acknowledgements

Part of this work was presented at the 13th Annual Meeting of the American Society for Bone and Mineral Research, San Diego, California, August 1991.

JBMR Anniversary Classic

An Uncoupling Agent Containing Strontium Prevents Bone Loss by Depressing Bone Resorption and Maintaining Bone Formation in Estrogen-Deficient Rats

PJ Marie, M Hott, D Modrowski, C DePollak, J Guillemain, P Deloffre, Y Tsouderos

Originally published in Volume 8, Number 5 pp 607–615 (1993)

How many readers recognized that this study showing that a strontium salt could prevent bone loss in ovariectomized rats might lead to a new therapeutic approach? Strontium ranelate is the first agent that has been shown to inhibit bone resorption, stimulate bone formation and reduce fracture incidence in osteoporosis.

Ancillary