Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Normal bone formation and maintenance of bone mass in weight-bearing limbs are dependent of gravity. Skeletal unloading induced by space flight or immobilization causes a decrease in bone mass in both human and animal models (Li et al., 1990; Collet et al., 1997) and alteration of bone architecture in rats (Morey-Holton and Arnaud, 1991). In the model of tail-suspended growing rats, we (Vico et al., 1991) and others (Wronski and Morey-Holton, 1987) demonstrated that changes in the tibial metaphysis [i.e., decrease in trabecular number (Tb.N) and trabecular thickness (Tb.Th)] resulted in alterations in longitudinal bone growth and bone cellular activities. Immobilization-induced bone loss is related to a rapid and transient increase in bone resorption and a sustained reduction in bone formation (Globus et al., 1984, 1986; Weinreb et al., 1989; Vico and Alexandre, 1994). Treatments increasing osteoblast recruitment, such as insulin-like growth factor-I (Machwate et al., 1993) or transforming growth factor-2 (Machwate et al., 1994), or those inhibiting bone resorption, such as bisphosphonates (Apseloff et al., 1993; Bikle et al., 1994; Kodama et al., 1997), have shown their ability to prevent bone loss in rapidly growing rats. Although beneficial effects of bisphosphonates are well documented in osteoporosis with high bone turnover (Fleisch, 1993), their mechanisms of action on the immobilization-related bone loss are not clearly understood (Bikle et al., 1994; Murakami et al., 1994; Grynpas et al., 1995). Whether bisphosphonates can prevent immobilization-induced bone loss either by directly decreasing osteoclast activity and number during the early phase or by indirectly acting later on osteoblasts (Sato et al., 1991; Sahni et al., 1993) remains unclear. In addition, their potential effects in the two metabolically distinct areas of the metaphysis have never been assessed in tail-suspended rat model. These two areas are the primary spongiosa (ISP) located under the growth plate metaphyseal junction, where modeling activity occurred (tissue balance is positive or null), and the secondary spongiosa (IISP) located between ISP and diaphysis, where tissue balance could be either positive or negative depending on circumstances (Baron et al., 1984).

The aim of our study was to analyze the sequential effects of the bisphosphonate tiludronate (chloro-4-phenylthiomethylene bisphosphonate; Sanofi Recherche, Montpellier, France) during each of the characteristic phases previously described in tibiae of the tail-suspended rat (Globus et al., 1984, 1986; Vico and Alexandre, 1994). These phases are characterized by a transient increase in bone resorption and the beginning of bone formation inhibition during the first week, a sustained reduction in bone formation during the second week, and a progressive normalization between the second and third weeks. We evaluated the effects of tiludronate in rats at the end of the rapid growing phase to minimize the bias induced by growth alteration due to bisphosphonate treatment (Schenk et al., 1986; Bikle et al., 1994; Kodama et al., 1997). We assessed the effects of tiludronate treatment administered during the last week of short-term [7 days (D7)], median-term [13 days (D13)], and long-term [23 days (D23)] suspensions.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals

After 1 week of acclimatization under standard conditions of a temperature-controlled (23 ± 1°C) and light-controlled environment (12-h light/dark cycle), 78 male Wistar rats, aged 12 weeks (300 g; IFFA Credo, Les Onçins, France) were randomly assigned to 13 groups of 6 animals each. They were fed a standard rodent chow and water ad libitum. Rats were tail-suspended (Susp) or not suspended as control animals (Ctr) for 7, 13, or 23 days. The baseline group was sacrificed the first day of experiment. For each time point, we determined four groups: one group of suspended rats treated with 0.5 mg/kg b.wt. tiludronate s.c. (Susp Til 0.5), one group of suspended rats treated with 5 mg/kg b.wt. tiludronate s.c. (Susp Til 5), or one Susp group treated with s.c. vehicle (i.e., saline solution) and one Ctr group treated with vehicle. Treatment was administered every other day for the 8 days before sacrifice.

Rats were suspended without anesthesia by wrapping two-thirds of the tail with orthopedic tape and allowing the animals to move freely in specifically designed cages as described previously by Morey et al. (1979). Rats were suspended at an angle of 30 degrees with approximately 50% of the body weight loaded onto the forelimbs. Calcein (15 mg/kg) and demeclocycline (20 mg/kg) were injected i.p. 1 and 7 days before sacrifice. All rats were sacrificed with an anesthetic overdose (0.1 mg/kg Nesdonal).

Dual X-Ray Bone Densitometry

Bone mineral density (BMD) was measured in the entire left tibia (without the fibula), in a 2.5-cm-deep water bath, according to the recommendations of Ammann et al. (1992). A dual-energy X-ray densitometer (QDR 1000; Hologic) was used with a 0.9-mm-diameter collimator and ultra-high-resolution mode adapted to the measurements of small animals. The scan was divided into proximal, distal, and diaphyseal portions (approximately 1.2, 1.4, and 1.4 cm in height, respectively) for analysis to take into account the changes in cortical and cancellous bone densities. The coefficient of variation of five measurements of BMD was 0.33, 1.7, and 0.40% for proximal, diaphyseal, and distal tibia, respectively.

Three-Dimensional Microtomography (3D-µCT)

To compensate for any difference in longitudinal growth between groups, the region of interest representing IISP was delimited anatomically from the bottom of ISP to a zone located proximally to the diaphysis-metaphysis border (Vico et al., 1991).

The proximal IISP of the excised tibia was analyzed. The microcomputed tomography (Scanco Medical, Zurich, Switzerland) was performed using a micro-X-ray source directed toward the sample (maximum dimensions, 36 mm length and 14 mm diameter; Rüegsegger et al., 1996). The modifications induced by bone crystals in the X-ray beam were analyzed with a plane detector (coupled derived detector array, 1024 elements). After a set data acquisition time (200 ms), the sample was rotated <1 degree, and a new acquisition was performed. This was repeated until the entire plan was covered (approximately 600 measurements). The sample was then moved upward by a fixed height (10 µm), and a new series of acquisitions were performed. The entire procedure was repeated on 200 successive slices (total length of sample analyzed, approximately 2 mm; total time for data acquisition, 400 min).

After the acquisition of the two-dimensional slices and the selection of the volume of interest, three-dimensional reconstruction of bone was performed using triangulation algorithms. Pixel size was estimated at approximately 10 µm, the volume of interest was approximately 150 × 150 × 150 voxels, and the resolution was of 16 µm. The morphometric parameters directly determined from the binarized volume of interest were trabecular bone volume/tissue volume (BV/TV) and Tb.N. Tb.Th and trabecular separation (Tb.Sp) were derived from these parameters.

Bone Histomorphometry (HM)

The right tibia and humerus were immediately excised after sacrifice, fixed, dehydrated in absolute acetone, and embedded in methylmethacrylate at a low temperature according to the method developed in our laboratory (Chappard et al., 1987). Samples were frontally sliced 7 µm thick for subsequent measurements of quantitative bone HM (Parfitt et al., 1987) in the proximal metaphyses in the region of interest representing ISP and IISP (Vico et al., 1991). The operator was not informed of the origin of the sample, and the following parameters were either measured or calculated.

BV/TV. BV/TV (percentage of cancellous bone area) and structural indices (Tb.Th and Tb.N) in ISP and IISP measurements were carried out on six modified Goldner's sections with an automatic image analyzer (TAS+).

Osteoid Surfaces. Osteoid surfaces OS/BS (percentages) were measured on four Goldner's sections, and osteoclast surfaces (Oc.S/BS) and osteoclast number (N.Oc/B.Pm; cell number per millimeter of trabecular bone perimeter) were determined on four sections stained for tartrate-resistant acid phosphatase activity (Chappard et al., 1987).

Statistical Analysis

Data are reported as mean ± S.D. A two-way ANOVA was performed to compare the effects of growth (i.e., experiment duration) and the effects of loading in untreated rats. Similar analysis was used to compare dose- and time-dependent effects of tiludronate in treated rats. ANOVA test was used to compare the different groups. When significant difference was observed, a post hoc test (Scheffé) was performed. Differences were considered significant at the .05 level.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Body Weight

No death or health deterioration occurred during this study. Body weight increased at each time point in all groups (Ctr, 12.3, 17.9, and 26.5%; Susp, 4.4, 5.8, and 17.9%; D7, D13, and D23, respectively, p  .05; Susp Til 0.5, 10.8 and 14.8%; Susp Til 5, 11.4 and 13.8%; D13 and D23, respectively, p  .05) except Susp Til 0.5 and Susp Til 5 at D7 (3.2 and 1.2%, respectively; N.S.).

Growth curves of the animals suspended for 13 days and their controls are shown in Fig. 1. Slopes of growth curves of the other groups suspended for 7 or 23 days were similar.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Growth body weight at D13 for Ctr and suspended and suspended/treated animals. Data are presented as mean ± S.D. a, statistically different (p < .05) versus control. , Ctr13; , Susp S13; , Susp S13 0.5; , Susp S13 5.

Bone Mass and Structural Parameters

Tibia. Bone densitometry. Proximal tibia BMD did not change with age, confirming that the animals achieved their rapid growth phase. No change was observed after only 7 days of suspension (6.4%, N.S.). Thirteen and 23 days of tail suspension significantly decreased proximal tibia BMD compared with Ctr (12.3 and 17.1%, respectively; Fig. 2A). This decrease was prevented by tiludronate treatment in a dose-dependent manner (Fig. 2A). Diaphyseal and distal tibia BMD levels were not significantly different between Ctr and vehicle- or tiludronate treated-suspended groups at any time of suspension (data not shown).

View larger version (46K): [in this window] [in a new window]   Fig. 2.   A, BMD in proximal tibia metaphysis (expressed in g/cm2). Data are presented as mean ± S.D. B, secondary spongiosa bone volume in tibial metaphysis measured by 3D-µCT. Data are presented as mean ± S.D. C, primary spongiosa bone volume in tibial metaphysis measured by conventional HM. Data are presented as mean ± S.D. Statistically different (p < .05): a, versus control; b, versus suspended; c, versus tiludronate 0.5 mg/kg b.wt. Columns under 0 days indicate basal control group. The next four columns (from left to right) indicate Ctr, Susp, Susp Til 0.5, and Susp Til 5 groups.

Macroscopic analysis of tibial metaphysis and growth plate. Analysis of Goldner-stained sections (Fig. 3A) revealed that tail suspension reduced the amount of primary cancellous bone compared with Ctr. No significant change was observed in growth plate height between groups (data not shown). As shown in Fig. 3A and Table 1, tiludronate treatment markedly increased ISP height in Susp, whereas no significant difference in IISP height was observed (data not shown). Tiludronate treatment not only compensated the suspension-related decrease in ISP height but also further increased the ISP height compared with Ctr (Table 1). We observed a trend to an increase in the whole metaphysis height after 23 days of suspension in the treated groups.

View larger version (133K): [in this window] [in a new window]   Fig. 3.   A, histological Goldner's sections of tibial ISP control (a), suspended (b), and suspended/treated (c) rats after 7 days (original magnification, 40×). Note the increased amount of cartilage (dark) in central cores of trabeculae in tiludronate-treated rats. B, 3D-µCT illustrations of control (a), suspended (b), and suspended/treated (c) rats after 7 days.

Bone mass and structural parameters of ISP. As shown by conventional HM, in Fig. 2C and Table 1, ISP BV/TV was significantly reduced in Susp (16.2% at D7, 22% at D13, and 17.1% at D23) compared with Ctr. This decrease in bone mass was related to a decrease in Tb.Th with no change in Tb.N (Table 1). Tiludronate markedly increased BV/TV in Susp in a dose-dependent manner but with no influence of time (Fig. 2C). The decrease in Tb.Th was prevented by tiludronate in a dose-dependent manner. Moreover, Tb.N increased under tiludronate treatment (Table 1) as early as 7 days of suspension.

Bone mass and structural parameters of IISP. These parameters were assessed by both 3D-µCT (Figs. 2B and 3B, Table 2) and HM (data not shown). Three-dimensional BV/TV decreased in Susp by 28.8% (N.S.) after 13 days and by 79.5% after 23 days (p < .05) compared with their respective Ctr (Fig. 2B). This bone loss was related to a decrease in Tb.Th and Tb.N (Table 2). Tiludronate treatment was effective in preventing bone loss at D13 at high dose only, whereas it was active at both doses during the third week (Fig. 2B). These preventive effects were mainly due to an increase in Tb.N during the second and third weeks (Fig. 3B).

Humerus.

Quantitative HM 1.  ISP: BV/TV was not affected by suspension. Tiludronate influenced BV/TV in a dose-dependent manner in Susp groups (18.1% at D7, 8.9% at D13, and 10.8% at D23 for the Til 0.5 dose; 27.2% at D7, 7.4% at D13, and 36.1% at D23 for the Til 5 dose) by increasing Tb.N (Table 1).

Bone Cellular Activities

Tibial IISP. Bone resorption. Oc.S/BS (Table 4) and N.Oc/B.Pm (data not shown) were similar in control and tail-suspended rats at every time points. Tiludronate treatment reduced Oc.S/BS in a dose-dependent manner, with significant changes for the 5 mg/kg dosage at 7D and 13D (Table 4).

Bone formation. All parameters of bone formation: OS/BS, MAR, MS/BS, and BFR/BS were markedly decreased by suspension after the first and second weeks. After 23 days of suspension, these parameters were all normalized. Bone formation was further decreased under tiludronate with a marked dose-dependent effect at days 23 probably related to formation normalization in corresponding suspended rats.

Humeral IISP. Globally, at the humerus level suspension did not modify the bone cellular parameters compared with their age-matched controls (Table 5).

Bone resorption. Bone resorption activity (Oc.S/BS) was similar in Ctr and in Susp at every time points. Tiludronate decreased Oc.S/BS with a dose-effect (45.1, 27.1, and 56.1% for dose 0.5 mg/b.wt.; 60.4, 58.7, and 76.3% for dose 5 mg/b.wt.; D7, D13, and D23, respectively; p < .05, versus Ctr).

Bone formation. All parameters of bone formation activity (OS/BS, MAR, and BFR/BS) were not affected by suspension. However, these parameters were reduced in rats treated with tiludronate, especially the high dose. This decrease was more marked for BFR/BS at D7 and D23 for dose 5 mg/b.wt. (52.7, 10.3, and 52.9% for dose 0.5 mg/b.wt.; 49.2, 39.9, and 85.4% for dose 5 mg/b.wt.; D7, D13, and D23, respectively; p < .05, versus Ctr).

Loading Status versus Drug Dose-Dependent Effects

Table 6 summarizes the results of two-way ANOVAs. The first two-way ANOVA showed that suspension mainly had major effects on IISP bone formation and that growth effects had major effects on IISP bone mass and structural parameters mainly. In the second two-way ANOVA, duration of suspension influenced both ISP and IISP bone mass and structural parameters as well as bone formation parameters. Tiludronate dose had preponderant effects on most of the measured parameters: BV/TV and Tb.N in both ISP and IISP, resorption activity as expected, and MAR.

View this table: [in this window] [in a new window]   TABLE 6 Two-way ANOVA in tibia For each parameter in each two-way ANOVA, p values are classified into one of three categories: 0, p not significantno effect; +, p statistically significantsignificant effect; or ++, p is more significant than in +predominant effect.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

A number of studies in animals have demonstrated the efficacy of bisphosphonates in the prevention of bone loss in models characterized by high bone turnover, such as ovariectomy (Amman et al., 1993; Balena et al., 1993a; Lauritzen et al., 1993; Mosekilde et al., 1994; Lepola et al., 1996) and thyroid hormone-induced osteopenia in rats (Balena et al., 1993b; Yamamoto et al., 1993). In contrast, little is known about their effectiveness in a model associated with low bone turnover, including rodent single hindlimb immobilization (e.g., plaster cast, immobilization against the abdomen, or unilateral sciatectomy) and tail suspension. In a recent review, Jee et al. (1997) stressed that the immobilization model in mature rats should be more appropriate than the rat ovariectomy-induced bone loss for human senile osteoporosis, at least at the cancellous envelope. Thus, we investigated the effect of tiludronate treatment on bone loss and microarchitectural deterioration in the hindlimbs of the tail-suspended rat. Furthermore, tiludronate was administered during the last week of each suspension period to evaluate its effects on the different phases of the bone formation changes that are commonly observed in this model.

The kinetics of bone alterations reported in the present study was in accordance with previous experiments (Globus et al., 1984; Halloran et al., 1988; Vico et al., 1991; Machwate et al., 1993). Moreover, 3D-µCT revealed early microarchitecture modifications that were not detected with HM (O.B., L.V., T.T., C.M., F.T., A.B., C.A., M.-H.L.-P., unpublished data). We found progressive BMD decrease in proximal tibia according to the time spent suspended, whereas no change occurred at the diaphysis level, confirming that trabecular bone was more affected by unloading (Li and Jee, 1991). In the ISP, we observed bone loss at a rather constant level (around 20%) according to the time of suspension as assessed by HM. Endochondral processes were impaired after 13 days of suspension (as assessed by either longitudinal growth rate or ISP height), whereas a relative recuperation was seen at D23.

Tiludronate prevented tail-suspension-induced tibial bone loss, based on metaphyseal BMD, in a dose-dependent manner. Tiludronate effects on bone mass were related to an increase in Tb.Th in both ISP and IISP. In addition, ISP height in tibia was 2- and 1.6-fold increased under tiludronate treatment compared with Susp and Ctr rats, respectively. Furthermore, a two-way ANOVA revealed that Tb.N was more strongly related to tiludronate dose than to the time spent in suspension. At the humeral ISP, high-dose tiludronate also induced a 1.2-fold significant increase in BV/TV, height, and Tb.N, without any change in Tb.Th. Conversely, in humeral IISP, neither suspension nor tiludronate affected bone architecture parameters. Our results are in accordance with two previous studies evaluating the effects of other bisphosphonates in the rat tail-suspension model (Bikle et al., 1994; Kodama et al., 1997). Pamidronate was unable to prevent Tb.Th decrease, whereas it increased 21-fold Tb.N in treated suspended group (Kodama et al., 1997). Similarly, Bikle et al. (1994) showed that suspension induced a bone loss through a decrease in Tb.Th, which was not corrected under alendronate treatment. However, alendronate increased markedly Tb.N in both control and suspended rats. Although these studies did not analyzed ISP separately, histologic sections clearly showed an increased ISP height and ISP calcified volume under bisphosphonate treatment. Thus, bisphosphonates may prevent immobilization-induced bone loss in rats by altering bone modeling in ISP. In a different bone loss model, Murakami et al. (1994) also showed that tiludronate acts in the ISP early after administration. Because bisphosphonates prevent resorption through inhibition of osteoclast function, the ISP may be not remodeled in treated rats, leading to hyperostosis of the metaphysis in both tibia and humerus regardless of the loading status. In contrast to the lack of resorption in the ISP, the resorption of cartilage below the hypertrophic zone of the growth plate appeared normal under treatment, as assessed on the basis of the growth plate thickness. These observations supported the concept that cartilage resorption was not mediated by osteoclasts/chondroclasts (Gilson et al., 1995). At the growth plate metaphyseal junction, the number of calcified cartilage septa was normally reduced so that only half of the septa persisted and remained accessible as templates for further bone deposition (Schenk et al., 1986). Tiludronate-induced inhibition of calcified tissue resorption within the ISP might have resulted in an increase in the number of persisting calcified cartilage septa and therefore in a higher density of the tibial ISP and IISP trabeculae. Overall, our data suggested that the inhibition of resorption in ISP could increase Tb.N and, subsequently, bone volume of the whole metaphysis, whereas the decrease in trabecular thickness persisted under treatment. The trabecular thinning seemed more related to the depressed bone formation because anabolic factors such as PTH (Turner et al., 1998) did prevent bone loss through prevention of trabeculae thinning.

Bone cellular effects of tiludronate were assessed histomorphometrically in IISP. Resorption was not affected by hindlimb unloading. The well known dose-dependent reduction in bone resorption (Oc.S/BS) was found in both loaded and unloaded bones of tiludronate-treated animals. Bikle et al. (1994) found similar results in tail-suspended rats treated with alendronate before the suspension, and at the ultrastructural level, osteoclasts had poorly developed brush borders and appeared not to adhere to the bone surface when examined.

As expected, bone formation was greatly depressed in tibiae after 7 and 13 days of suspension. Both static and dynamic parameters indicated a decrease in osteoblast recruitment (more than 40% reduction in Susp rats compared with Ctr for OS/BS and MS/BS), whereas the activity of mature osteoblasts (i.e., MAR) was less altered (20% at D7). The rebound in bone formation activity was observed at D13. This confirmed our previous results (Vico et al., 1991). In addition, we showed here that a rebound in osteoblast recruitment (i.e., MS/BS) occurred later (23 days). Tiludronate further decreased the unloading-induced reduction in bone formation, but the associated effects of suspension and tiludronate were not fully additive. Although the decreased bone formation observed in treated animals may be secondary to the reduction in bone resorption, a direct effect of bisphosphonates on osteoblast activity might occur (Khokher and Dandona, 1989).

In summary, after the first week of tail suspension, tibial bone loss occurred only in ISP and was prevented by the use of tiludronate. During the second and third weeks, tiludronate, despite a dramatic decrease in bone formation, prevented the tibial bone loss induced by immobilization in ISP and IISP. Tiludronate-induced decrease in bone resorption altered bone modeling with an increase in Tb.Th of the ISP and, subsequently, the IISP. In contrast, tiludronate, like other bisphosphonates, was unable to prevent or restore the decrease in Tb.Th secondary to bone formation impairment in this model.

Thus, the effects of bisphosphonate treatment in the widely used rat animal model, even at the end of the rapid growing phase, should be interpreted with caution, and other immobilization-induced osteoporosis models should be used to evaluate bisphosphonates in this specific pathology.