Bisphosphonate modulates morphological and mechanical properties in distraction osteogenesis through inhibition of bone resorption

Bone 39 (2006) 573 – 581 www.elsevier.com/locate/bone

Bisphosphonate modulates morphological and mechanical properties in distraction osteogenesis through inhibition of bone resorption Mitsuhiko Takahashi, Kiminori Yukata, Yoshito Matsui, Aziz Abbaspour, Shinjiro Takata, Natsuo Yasui ⁎ Department of Orthopedics, Institute of Health Biosciences, the University of Tokushima Graduate School, 3-18-15, Kuramoto, Tokushima 770-8503, Japan Received 27 December 2005; revised 15 March 2006; accepted 17 March 2006 Available online 18 May 2006

Abstract Despite the general clinical acceptance of distraction osteogenesis and much attention to bone formation in this method, little is recognized about activated bone resorption in the regenerated bone. The purpose of this study was to demonstrate the simultaneously activated bone resorption with activated bone formation and to investigate the role and efficacy of bisphosphonate in distraction osteogenesis. Left tibiae of 54 immature rabbits were lengthened for 3 weeks at a rate of 0.7 mm/day after a 1-week lag. Regenerated bone was quantitatively investigated by radiographic bone density, bone histomorphometry, and three-point bending testing. Animals received either vehicle or nitrogen-containing bisphosphonate (N-BP), YM529/ONO5920 at doses of 0.4 mg/kg/w or 0.004 mg/kg/w for 6 weeks. Regenerated bone of the vehicle group showed a radiologically characteristic zone structure containing the osteopenic zones adjacent to the sclerotic zones. The regenerated bone of the 0.4-mg/kg/w group showed no osteopenic zones during the course and eventually became homogeneously radiodense. The bone volume corresponding to the osteopenic zone of this group was 5.6-fold greater compared with that of the vehicle group. The lengthened bone strength of this group was 3.3-fold greater in ultimate force than that of the vehicle group and equivalent to the contralateral tibia. The 0.004-mg/kg/w group had no substantial differences compared with the vehicle group, despite radiological enhancement of the mineralized front as well as somewhat delayed bone resorption. These results demonstrate that not only bone formation but also bone resorption is highly activated in the regenerated bone, implying high bone turnover. Sufficient N-BP caused a notable modulation in morphological properties of the regenerated bone through inhibition of highly activated bone resorption and eventually increased mechanical properties. © 2006 Elsevier Inc. All rights reserved. Keywords: Distraction osteogenesis; Bone resorption; Bisphosphonate; Morphological modulation; Bone strength

Introduction Distraction osteogenesis has been accepted as an important and effective technique in musculoskeletal problems. Many successful cases have been reported even in the treatment of large bone defects and nonunions in various congenital and posttraumatic conditions [1–5]. In this method, a long bone is separated by corticotomy and is subjected to slow distraction using an external fixator. The distraction gap is gradually filled with newly formed bone during distraction and even after completion of distraction, followed by remodeling into cortical

⁎ Corresponding author. E-mail address: [email protected] (N. Yasui). 8756-3282/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2006.03.010

bone. It is believed that mechanical tension stimuli promote exuberant bone formation subsequent to recruitment of mesenchymal cells in distraction osteogenesis through production of various endogenous growth factors [6–9]. Every effort has been made to enhance an anabolic process in distraction osteogenesis. Injection of exogenous growth factors [10–15] into the distraction gap has been applied. Supplementation of growth factors after completion of distraction seems to be useful, as most growth factors are overexpressed during distraction but significantly reduce after completion of distraction [9,16,17]. Injections of fibroblast growth factor (FGF)-2 and bone morphogenetic protein (BMP)-2 into the distraction gap after completion of distraction have demonstrated considerable effects on volume of the regenerated bone [10,13]. However, anabolic agents have been shown to provide

574

M. Takahashi et al. / Bone 39 (2006) 573–581

negligible effects, if any, as in the case of BMP-7, now clinically available as OP-1, which had small effects notwithstanding an application after completion of distraction [15]. Previous studies on the basic aspects of distraction osteogenesis have focused on the mechanism of bone formation in response to mechanical tension stimuli as well as the enhancement of the anabolic process. Bone resorption in the regenerated bone has been largely neglected, though we have demonstrated in a rabbit model that the regenerated bone had radiologically three characteristic zones, consisting of not only a central radiolucent zone and adjacent sclerotic zones but also subsequent osteopenic zones [18]. The sclerotic zones that represent predominant bone formation migrate toward the central radiolucent zone in accordance with distraction. Subsequently, these sclerotic zones are substituted by the osteopenic zones, undoubtedly caused by predominant bone resorption. It is for this reason; we feel that bone resorption should be of interest. In the current study, we encouraged highly activated bone resorption in distraction osteogenesis. Bisphosphonates (BPs) are useful therapeutic agents in high turnover bone diseases. Nitrogen-containing BPs (N-BPs) are currently available as one of the most effective anti-catabolic agents [19]. Several studies have applied N-BPs to distraction osteogenesis, increasing bone volume around the regenerated bone [20–23]. However, these studies have not evaluated discrete aspects of the regenerated bone, which consists of fundamentally dissimilar characteristics. Thus, the argument is still unsettled as to whether the increase in bone volume results from only anti-catabolic effects or feasible anabolic effects by N-BP. In the current study, we also investigated an effect of a third-generation N-BP, YM529/ONO5920, on bone resorption and bone formation in a rabbit tibial lengthening model, where the anti-catabolic effect was indicated by preservation of bone volume and reduction of osteoclastic parameters in the osteopenic zones. Inversely, the anabolic effect was evaluated in the mineralized front of the sclerotic zone contiguous to the central radiolucent zone. YM529/ONO5920 caused noticeable morphological modulation of the regenerated bone that has not been reported previously.

the osteotomy was followed by 3 weeks of distraction at a rate of 0.35 mm every 12 h (0.7 mm/day). After achieving 15 mm of lengthening, a 4-week consolidation phase was instituted with the external fixator in situ for bone consolidation. The rabbits were allowed to run freely in the cage throughout the experiment and were randomly divided into three groups of 18 animals each. The vehicle group (Group V) intravenously received 5 ml/week of normal saline for 6 weeks from the osteotomy. Group L and Group H were intravenously administered YM529/ONO5920 for the same period at weekly doses of 0.004 (low dose) and 0.4 (high dose) mg/kg of body weight (BW), respectively.

Radiological examination and radiographic densitometry Each lengthened tibia was evaluated in vivo every other week by anteroposterior radiographs taken with a Softex model CMB-2T (Softex Co., Kanagawa, Japan). Under definite conditions (50 kVp, 20 mA, 8.0-s exposure time, and 60-cm distance), exposures were carried out with an aluminum wedge that allowed computing radiographic bone densities later. These radiographs were scanned in 300 dpi (ES-8000, Seiko Epson Co., Nagano, Japan). Respective pixel values of the aluminum wedge where the thickness was already known were measured with NIH Image (version 1.63, NIH, USA). Calibration functions were generated on each radiograph by regressions measured pixel intensities against known aluminum thicknesses. An intramedullary rectangle area of the regenerated bone with adjacent osteotomized bones was established (Fig. 1). Respective pixel values in this area were measured by NIH Image and were graphed along the longitudinal axis of tibia as a bone density curve expressed by aluminum thicknesses (mm Al) from the regression equation. Mean bone density curves were calculated from twelve of each group.

Histomorphometry and tissue harvest For bone histomorphometric analyses, rabbits were sacrificed at two different time points: 4 weeks (at the end of distraction) and 8 weeks (at the end of consolidation) postoperation. Fluorochrome bone labels (calcein green 20 mg/kg) were administered intravenously at 7 days and 2 days before euthanasia. The rabbits were euthanized with an overdose of pentobarbiturate (60 mg/kg), and lengthened tibiae were harvested by disarticulation at the knee and ankle. The specimens were fixed in 70% ethanol, and the undecalcified bones were embedded in glycolmethacrylate. Undecalcified 3-μm-thick frontal sections were stained (1) doubly with tartrate-resistant acid phosphatase (TRAP) and alkaline phosphatase (ALP) and (2) with toluidine blue O. Histomorphometric analyses were performed at a minimum of eight optical fields with 100fold magnification using a semi-automated image analysis system (Osteoplan II; Carl Zeiss, Thornwood, NY) and a digitizing tablet with a light/fluorescence microscope. At first, we measured the spongiosa in the osteopenic zone at 4 weeks and 8 weeks postoperation to investigate mainly an anti-catabolic effect. Secondly, to investigate mainly an anabolic effect we measured the interval of

Materials and methods Rabbit lengthening model and protocol All experimental procedures were approved by the local animal protection and ethics committee. Mid-diaphyseal tibial lengthening using Orthofix M-100 external fixators was performed on 54 male immature Japanese white rabbits ranging from 2.2 to 2.7 kg of body weight. The animals were housed individually in environmentally controlled room and were fed standard rabbit chow (RC4, Oriental yeast Co., Tokyo, Japan). Anesthesia was administered with ketamine 20 mg/kg and xylazine 5 mg/kg by intravenous injection with lidocaine 5 mg/kg locally. Under sterile conditions, four stainless steel half pins were inserted into the medial aspect of the left tibia after percutaneous predrilling. Transverse osteotomy was performed below the tibio-fibular junction between the second and third screw, using a threadwire saw. The periosteum was carefully opened before osteotomy, protected throughout the procedure, and then closed using 4-0 nylon threads. A 1-week lag phase after

Fig. 1. Region of interest for radiographic density in the lengthened tibia of Group V at 4 weeks postoperation (at completion of distraction). This intramedullary rectangular area from 5.0 mm proximal to the proximal osteotomy line is 25 mm in length and 2.5 mm in width. Note a characteristic zone structure of the regenerated bone, which contains not only two sclerotic zones adjacent to the central radiolucent zone but also two osteopenic zones at the proximal and distal ends. The zone structure is proximally and distally symmetrized from the central radiolucent zone.

M. Takahashi et al. / Bone 39 (2006) 573–581 4.8 mm to 6.0 mm from the original osteotomized bone at 4 weeks postoperation, which corresponded to the mineralized front of the sclerotic zone. To avoid dispersion due to the fibula, both measurements were made at the distal half. Nomenclature, symbols, and units used are those recommended by the Histomorphometry Nomenclature Committee of the American Society of Bone and Mineral Research [24].

575

The sample size in each group was six for all analyses. The data of histomorphometric bone volume at both time points and all of the mechanical analyses showed that powers of ≥0.9 were achieved to determine differences among three groups.

Results Mechanical analysis The mechanical properties of tibiae at 8 weeks postoperation were examined. The rabbits were euthanized in an identical method as mentioned above. The lengthened and contralateral untreated tibial specimens were cleaned of soft tissue and preserved in saline-soaked gauze under refrigeration at −80°C. They were thawed at room temperature 12 h before testing and kept moist during preparation and testing. Three-point bending strength was measured until failure with a support span of 40 mm between the second and third pin holes, using a servohydraulic materials testing system (S-100; Shimazu, Kyoto, Japan) with a 1 kN load cell under displacement control (5 mm/min). The specimens were positioned horizontally with anterior cortex compression and pressed vertically at the center of the supports, i.e., the center of the regenerated bone. Load and displacement data were recorded on an accompanying computer, and then ultimate force (the maximum force that the specimen sustained), stiffness (slope of the linear portion of the load-displacement curve), and work to failure (integrating the load-displacement curve to the point of failure) were determined.

Statistical analysis and power All results and error bars on graphs are presented as the means ± standard deviations. Statistical analyses to detect differences among respective groups were performed by one-way analysis of variance (ANOVA) using statistical package StatView (version 5.0, SAS Institute, Inc., Cary, NC). If significances were indicated by ANOVA, comparisons between means of respective groups were tested by Fisher's protected least significant difference (PLSD) tests for post hoc analyses. Comparisons between individual time points were made with unpaired t test or Mann-Whitney U test. The level of statistical significance was set at P < 0.05.

Radiological findings Bony continuities of the regenerated bone were achieved in all animals at 8 weeks postoperation, notwithstanding the presence of central radiolucent zone at 4 weeks postoperation. Representative radiographs of each group are shown in Fig. 2. Group V showed a typical zone structure on each radiograph through the consolidation phase. There were two sclerotic zones proximally and distally adjacent to the central radiolucent zone at 4 weeks postoperation. Both sclerotic zones migrated toward the center of regenerated bone and eventually merged with each other. The osteopenic zones expanded from the proximal and distal original osteotomized bones during the course, compressing the sclerotic zones toward the center of regenerated bone. Group L also showed same zone structure as Group V, but less resorption of the osteopenic zones. Group H showed no osteopenic zones at any time points during the consolidation phase. In this group, neither radiographic tubular structure nor cortical configuration in the regenerated bone was detected at 8 weeks when they were detectable in Group V. In Group L, the central radiolucent zone was already undetectable at 6 weeks, whereas it was not yet in the other groups.

Fig. 2. Consecutive radiographs in the regenerated bones of each group. Distraction is underway at 2 W (2 weeks postoperation), and the consolidation phase starts at 4 W and ends at 8 W. The proximal sides are on the left of each radiograph. In Group V, the osteopenic zones are already noticeable between each sclerotic zone and original osteotomized bone at 4 W. In Group L, the osteopenic zones are also noticeable, but their borders are vague. No osteopenic zones are perceptible in Group H.

576

M. Takahashi et al. / Bone 39 (2006) 573–581

Fig. 3. Mean bone density curves in the regenerated bones of each group. Bone densities are expressed by aluminum thickness (mm Al). The proximal sides are on the left of each graph. Gray, dashed, and solid lines represent bone density curves at 4, 6, and 8 weeks postoperation, respectively. Dotted lines express proximal cortical density immediately after the operation as reference. Lower black bars represent approximately 15 mm of the regenerated bones. Note that the curves of Group L are similar to those of Group V, whereas the curves of Group H show continuous increase in bone density entirely.

Radiographic densitometry The mean bone density curves for each group are shown in Fig. 3. The central radiolucent zone was identified at the center of regenerated bone as a valley at 4 weeks postoperation in all groups. Group V The peak of two sclerotic zones exceeded 3 mm Al of bone density at 4 weeks. The peaks of the sclerotic zones migrated toward the center of regenerated bone and eventually fused into a single peak with 4 mm Al of bone density. The two depressions were osteopenic zones between each sclerotic zone and osteotomized bone through the consolidation phase and showed approximately 3 mm Al of bone density (arrow). Group L The prominences of each sclerotic zone were wider, and their descending curves to each osteopenic zone were less steep than those in Group V, indicating slower bone resorption. However, two osteopenic zones were still identified at 6 to 8 weeks, and they showed equivalent bone density to the osteopenic zones of Group V (arrow). The valley of central radiolucent zone already

disappeared and the peaks of sclerotic zones merged at 6 weeks consistent with the radiological finding, suggesting an accelerated bone formation from mesenchymal cells. Group H The valley of central radiolucent zone was seen at 4 weeks but no depressions of osteopenic zones. Proximally and distally ascending curves of the sclerotic zone were steeper than those of the other groups. The whole bone densities of regenerated bone increased during the treatment period with over 5 mm Al uniformly at 8 weeks. The bone density of proximal osteotomized bone approximately doubled at 8 weeks compared with immediately postoperation. Histology and histomorphometry At 4 weeks postoperation, the central radiolucent zone and the sclerotic zone predominantly consisted of longitudinal fibrous tissue and abundant trabecular bone, respectively. Both zones were connected by various rates of enchondral, membranous and transchondroid ossifications in all groups. At 8 weeks postoperation, the fibrous tissue and cartilage tissue have been substituted by mature trabecular bone.

Fig. 4. Representative photomicrographs (Toluidine blue O stain) in the regenerated bones of each group at 8 weeks postoperation. Each photomicrograph corresponds to the lower (fibular) side of distal osteopenic zone in Fig. 2. Arrowheads indicate that newly formed trabecular bone absorbed in various manners. Newly formed cortex (arrows) and original cortex (stars) seem to be similar in all groups. Scale bar = 1 mm.

M. Takahashi et al. / Bone 39 (2006) 573–581

Effect on the osteopenic zone Most distinct feature among three groups was bone volume of newly formed trabeculae in the osteopenic zone at 8 weeks (Fig. 4). Newly formed trabeculae were not only preserved in this area but also intimately connected with newly formed cortex in Group H. Hence, cortical configuration was undetectable in the radiological examination. This connection ought to reinforce the mechanical properties. Results from bone histomorphometric analyses in the osteopenic zone at 4 weeks and 8 weeks are presented in Tables 1 and 2, respectively. New bone volume (BV/TV) of Group H was approximately 2-fold greater than that of Group V or L at 4 weeks. BV/TVs decreased in Group V (38%) and Group L (54%) from 4 weeks to 8 weeks, whereas it increased in Group H (14%). BV/TV of Group H was preserved 5.6-fold greater than that of Group V and 3.8-fold than that of Group L at 8 weeks. BV/TVs correlated with trabecular numbers (Tb.N), which significantly decreased in Group V (19%) and Group L (32%) from 4 weeks to 8 weeks. Tb.N of Group H was preserved 7.6-fold greater than that of Group V and 3.9-fold than that of Group L at 8 weeks. Trabecular thicknesses (Tb.Th) increased in all groups from 4 weeks to 8 weeks, and there were no significant differences at 8 weeks. At 4 weeks, osteoclast number (N.Oc/B.Pm) of Group H was approximately 2.5-fold lower than that of Group V or L, and osteoclast surface (Oc.S/BS) of Group H was 3.7-fold lower than that of Group V and 3.1-fold than that of Group L. The differences of N.Oc/B.Pm and of Oc.S/BS among the groups reduced at 8 weeks because these osteoclastic parameters more decreased in Group V and L than in Group H. Osteoid parameters (OV/BV, OS/BS, and O.Th) revealed no significant differences in all groups at 4 weeks, whereas osteoblast surface (Ob.S/BS) of Group L as well as Group H Table 1 Histomorphometric parameters of the osteopenic zone at the completion of distraction (4 weeks postoperation) Parameters

Group V

Group L

Group H

BV/TV (%) Tb.Th (μm) Tb.N (/mm) N.Oc/B.Pm (/100 mm) Oc.S/BS (%) OV/BV (%) OS/BS (%) O.Th (μm) Ob.S/BS (%) MAR (/day) MS/BS (%) BFR/BS (mm3/cm2/Y)

26.04 ± 11.00 63.94 ± 5.17 4.00 ± 1.51 196.13 ± 66.01 6.01 ± 1.64 5.15 ± 3.24 21.89 ± 9.24 7.04 ± 2.42 31.47 ± 9.34 3.31 ± 0.96 1.10 ± 0.46 1.40 ± 0.96

27.48 ± 10.44 58.25 ± 10.96 4.57 ± 1.12 187.30 ± 75.62 5.04 ± 2.49 1.45 ± 1.55 6.17 ± 4.46 5.10 ± 1.81 9.45 ± 6.32 2.55 ± 1.21 1.22 ± 1.32 1.09 ± 1.18

49.11 ± 13.38a,b 76.84 ± 13.18c,b 6.33 ± 1.11a,d 78.07 ± 35.39a,b 1.63 ± 0.44b,e 4.06 ± 5.75 13.88 ± 10.46 8.97 ± 3.14 9.46 ± 3.90e,f 2.02 ± 0.40 c 1.52 ± 2.03 1.33 ± 2.03

Data expressed as means ± SD. Data expressed as means ± SD. P < 0.001 for high dose group vs. low dose group. a P < 0.01 for high dose group vs. low dose group. b P < 0.01 for high dose group vs. control group. c P < 0.05 for high dose group vs. control group. d P < 0.05 for high dose group vs. low dose group. e P < 0.001 for high dose group vs. control group. f P < 0.05 for low dose group vs. control group.

577

Table 2 Histomorphometric parameters of the osteopenic zone at the end of consolidation (8 weeks postoperation) Parameters BV/TV (%) Tb.Th (μm) Tb.N (/mm) N.Oc/B.Pm (/100 mm) Oc.S/BS (%) OV/BV (%) OS/BS (%) O.Th (μm) Ob.S/BS (%) MAR (/day) MS/BS (%) BFR/BS (mm3/cm2/Y)

Group V †

9.97 ± 7.32 119.31 ± 56.65† 0.75 ± 0.37‡ 152.49 ± 136.85 3.47 ± 2.62 3.09 ± 4.36 18.57 ± 18.87 6.25 ± 4.40 17.60 ± 16.34 2.14 ± 0.86 11.36 ± 7.47‡ 9.55 ± 8.74†

Group L

Group H

14.73 ± 11.04 87.58 ± 35.13 1.45 ± 0.89§ 68.28 ± 36.77‡ 1.38 ± 0.76‡ 2.80 ± 2.29 19.88 ± 10.40† 5.85 ± 3.97 17.99 ± 7.75 2.40 ± 0.79 8.58 ± 7.78† 6.56 ± 4.41†

56.15 ± 10.29a,c 99.66 ± 22.68 5.69 ± 0.48a,c 75.67 ± 18.88 0.97 ± 0.41†,a,f 0.23 ± 0.26 2.02 ± 2.23†,a,d 3.51 ± 2.89† 1.07 ± 1.04§,a,e 1.15 ± 0.63†,d 0.34 ± 0.13 0.16 ± 0.14a

Expression of data, abbreviations, and units are identical to Table 1. †P < 0.05; ‡ P < 0.01; §P < 0.001 for each group at 8 weeks vs. at 4 weeks.

significantly diminished compared with that of Group V. The trends of these osteoid parameters and ObS/BS from 4 weeks toward 8 weeks were varied in each group (i.e., slightly decrease in Group V, increase in Group L and distinctly decrease in Group H). Dynamic bone formation parameters (MAR, MS/BS, and BFR/BS) of Group V and Group L rather increased from 4 weeks to 8 weeks, but in contrast, those of Group H decreased. These results suggest trabecular bone in the osteopenic zone of Groups V and L are still under high turnover at 8 weeks. Effect on the sclerotic zone As mentioned above, various ossifications were shown from the central radiolucent zone to the sclerotic zones at 4 weeks postoperation. Despite 2-time injection of fluorochrome bone label, we were not able to identify two consecutive linear labels but nonuniform staining at this area. Therefore, dynamic bone formation data are not reported. However, osteoid volume and osteoblast surface were relatively higher in Group L, indicating enhancement of bone formation (Table 3). There were no significant differences of bone volume and osteoclast number, as well as bone formation parameters in the mineralized front of the sclerotic zone in all groups. Recruitment of osteoclast has been observed in all groups even in the mineralized front of the sclerotic zone, though

Table 3 Histomorphometric parameters in the mineralized front of the sclerotic zone at the completion of distraction (4 weeks postoperation) Parameters

Group V

Group L

Group H

BV/TV (%) Md.V/TV (%) N.Oc/B.Pm (/100 mm) OV/BV (%) Cho/TV (5) Ob.S/BS (%)

13.14 ± 9.52 12.27 ± 8.92 124.22 ± 82.72 5.28 ± 4.67 8.71 ± 4.58 30.11 ± 9.82

18.47 ± 13.74 17.56 ± 13.75 92.39 ± 34.86 7.53 ± 8.10 8.14 ± 5.42 33.55 ± 14.92 a

19.46 ± 11.23 18.76 ± 10.95 70.07 ± 51.94 3.79 ± 2.19 4.77 ± 5.27 21.14 ± 4.72 b

Expression of data, abbreviations, and units are identical to Table 1. a P < 0.01. b P < 0.001 for each group in the sclerotic zone vs. in the osteopenic zone.

578

M. Takahashi et al. / Bone 39 (2006) 573–581

each value of osteoclast number was lower than that in the osteopenic zone of the corresponding group. YM529/ONO5920 probably prevented the further recruitment of osteoclast after the primary bone formation as an anti-catabolic driver. Mechanical analysis The lengthened tibiae of Group H had significantly higher mechanical properties compared with Group V or L. Mean ultimate force of Group H was 3.3-fold greater than that of Group V and equivalent to the untreated contralateral tibia (Fig. 5A). Mean stiffness (Fig. 5B) and work to failure

Fig. 5. Ultimate force (A), stiffness (B), and work to failure (C) of the lengthened tibiae (left columns) and the untreated contralateral tibiae (right columns) in three-point bending strength. Data are shown for Group V (white), Group L (hatched), and Group H (black). On the lengthened tibiae, Group H was significantly stronger in ultimate force (P < 0.0001), in stiffness (P = 0.0007), and in work to failure (P < 0.0001) as compared with Group V, and also in ultimate force (P = 0.0002), in stiffness (P = 0.0221), and in work to failure (P = 0.0015) as compared with Group L.

(Fig. 5C) of Group H were approximately 3-fold greater than those of Group V. There were no significant differences in any mechanical properties of the lengthened tibiae between Group V and L, whereas there were still significant differences in all mechanical properties between Groups L and H. On the untreated contralateral tibiae, there were no significant differences among three groups in all mechanical properties. Discussion The current study is based on the consecutive process of distraction osteogenesis, where mechanical tension stimuli promote recruitment of mesenchymal cells and bone formation in the center of regenerated bone and followed by consolidation proximally or distally [6,7,18]. The spatial distribution of central radiolucent, sclerotic, and osteopenic zones in the regenerated bone correspond to the inflammatory, repair, and remodeling phases in fracture healing, respectively. Although various animal models of distraction osteogenesis have been developed in the last two decades, most of them practiced rapid distraction rates or short distraction distance, both of which are inadequate for detecting the typical osteopenic zones. Accordingly, bone formation has been paid more attention than bone resorption in distraction osteogenesis. Insufficiency of bone volume of the regenerated bone has been thought to be due to the attenuation of bone formation. However, this attenuation is mainly due to highly activated bone resorption, as we presented in the current study and as presented by Smith et al. [20]. The newly formed trabecular bone is rapidly absorbed, resulting in osteopenic zones. In Group V, the osteopenic zones were already detectable at 4 weeks postoperation (at the completion of 3-week distraction), which implies that the perceptible amount of newly formed bone has a very short life span, and that the regenerated bone is under high bone turnover. Thus, NBPs that increase bone strength by decreasing bone remodeling (anti-catabolic effect) [19] is a reasonable application to distraction osteogenesis. The regenerated bone would have more strength if it were not absorbed, in terms of radiodense character of the sclerotic zone. High doses of the potent N-BP, YM529/ONO5920, significantly preserved radiological and histomorphological bone volume in the osteopenic zone and eventually increased mechanical properties of the regenerated bone. In the osteopenic zone, where bone resorption overwhelmingly predominates over bone formation, there was an increase in bone volume that demonstrates decreased bone catabolism. Anti-catabolic effects were also confirmed by decreases in osteoclastic parameters in the osteopenic zone. Our results are consistent with previous literature such as Smith et al. [20] which discussed results of bone histomorphometry and Little et al. [21] which conducted mechanical testing. However, these studies that applied N-BPs to distraction osteogenesis have not discriminated between the sclerotic zone and the osteopenic zone. Despite positive effects around the regenerated bone and anti-catabolic action of N-BPs as a possible mechanism, further explication has been required. Why do these powerful and typical anti-catabolic agents

M. Takahashi et al. / Bone 39 (2006) 573–581

intensify the anabolic process of osteogenesis? Again, the regenerated bone includes the antithesis of osteogenesis. The current study, by the distinctive evaluation of each zone, demonstrated that N-BP strikingly modulated distraction osteogenesis through inhibition of critical bone resorption. YM529/ONO5920 is experimentally 10,000 times more potent than etidronate and has the same anti-catabolic effect as zoledronic acid [25,26]. Three doses of oral administration of YM529/ONO5920, which is approximately equivalent to an intravenous doses of 0.0014 to 0.035 mg/kg/w, showed significant effects in an animal model of osteoporosis on preservation of bone mass and strength [27]. However, in Group L whose dose corresponded to that for osteoporosis, there were little effect on density and strength of the regenerated bone. The dose of Group H had a sufficient effect on inhibition of bone resorption and corresponded with the other experimental models [28,29], where injected YM529/ONO5920 markedly suppressed metastasis formation and tumor growth in bone. In bone histomorphometry, the osteoclast surface of Group V at 4 weeks was larger than that of the metaphysis of growing rabbits in another study [30], indicating bone resorption is highly activated in the osteopenic zones. Bone resorption in the regenerated bone is more aggressive than that in osteoporosis and may be rather close to that in bone metastasis. Mechanical tension stimuli promote plentiful bone formation in the center of distraction gap, suggesting prominent bone formation. However, an external fixator, which provides mechanical tension stimuli, concomitantly causes the regenerated bone stressshielding effect unfavorable for maintaining bone volume, as previously surmised [21,22,23]. Promoted bone formation would consequently lead to bone resorption in accordance with the coupling mechanism between osteoblasts and osteoclasts [31,32] and with direct or indirect stimulation of osteoclast by BMPs or FGFs [33,34]. Recent research has generated new insights that N-BPs can enhance the bone forming activities of osteoblasts in addition to anti-catabolic effect [35–37]. This enhancement of anabolic activity is generally observed in weekly exposure of N-BPs at lower concentration [38–40]. Radiographic migration of the sclerotic zones toward the center of distraction gap was rather accelerated in Group L. This accelerated migration indicates enhancement of primary bone formation from mesenchymal cells and absolutely excludes the contribution of anti-catabolic effect of N-BPs because osteoclasts cannot work until bone matrix is formed. However, histomorphometry in the mineralized front of the sclerotic zone showed positive but nonsignificant differences in terms of bone formation parameters. Furthermore, low doses of YM529/ONO5920 were unable to preserve the newly formed trabecular bone from consequently activated bone resorption. Therefore, the anabolic effect of NBPs, albeit radiologically detectable in the regenerated bone, must be unimportant for distraction osteogenesis compared with the anti-catabolic effect. Anabolic agents usually seem to have little effect on increasing bone volume of the regenerated bone. In fact, injection of several growth factors into the regenerated bone resulted in no effects [11,12,14]. Based on the fact that most of

579

the growth factors significantly decreased after completion of distraction [9,16,17], injection of BMP-7 at this time point also ended in no success [15]. These results suggest that there is little indication of anabolic agents to distraction osteogenesis. However, it may be advantageous to apply the growth factors to distraction osteogenesis with aged individuals that are anticipated to have low generating potential of such growth factors [41]. Even in this condition, anti-catabolic activity is thought to be necessary for preservation of the newly formed trabecular bone and for suppression of high bone turnover. This is upheld by the noteworthy outcome of bone volume in the bone defect model where combination of BMP-7 and zoledronic acid synergistically modulate both the anabolic and catabolic responses [42]. Primary bone formation from mesenchymal cells lasted approximately 6 weeks in the current model. As a consequence of strong affinity to bone mineral, N-BP injected intravenously is eliminated from serum in 2 h [43]. Once taken up by the bone, it is conceivable that a single injection of N-BP no longer functions on the bone formed thereafter. Therefore, we have determined it would be beneficial to use continuously repeated doses of N-BP. Little et al. reported that two doses of 0.1 mg/kg zoledronic acid 2 weeks apart yielded an 89% increase in ultimate force compared with the control group [21]. Our results showed a 230% increase in Group H compared with Group V. We assume that this increase results from the repeated and increased dosage of N-BP, regardless of the differences in N-BP architecture, postoperative protocol, or testing method. However, this increase is less impressive than our expectation, considering that a more than 10-fold dose had been applied all together in the current study. The dose of Group H might have gone beyond the threshold. Moreover, side effects of N-BPs such as renal toxicity and osteonecrosis of the jaw [44,45] should be considered in intravenously high dose application. The dose-dependent but nonsignificant decrease in longitudinal bone growth was also observed (data not shown). Delayed cortical development is another concern, though histologically new cortical formation could be seen in Group H. Smith et al. [20] have reported that the regenerated bones treated with two doses of 0.1 mg/kg zoledronic acid were remodeled into radiologically normal cortical bones by 44 weeks. Completion of the remodeling may be prolonged in a dose-dependent manner. On fracture healing models, single systemic dose of NBP rather accelerated bone strength compared with local and probably continuous doses [46]. Distribution of N-BP in the newly formed callus, albeit less than that in the animal injected continuously, was also detected in the animal in which injection had ceased at fracture [47]. These results indicate that a bolus dose of N-BP may be pertinent even to distraction osteogenesis. Advanced studies optimizing dose and consequent remodeling would be needed in distraction osteogenesis models. In conclusion, despite induction of bone formation by the distraction procedure, our data have shown that bone resorption is also occurring with the result of the definite osteopenic zones. These zones of the regenerated bone were noteworthily modified by sufficient N-BP, eventually resulting in increased mechanical properties of the regenerated bone. Although the

580

M. Takahashi et al. / Bone 39 (2006) 573–581

discrete evaluations in the current study also indicated the anabolic effects of N-BP in distraction osteogenesis, the effects were inconsequential under the highly activated bone resorption. In contrast, the anti-catabolic effects were indicated even in the mineralized front of the sclerotic zones as well as in the osteopenic zones, which were important effects of N-BP on distraction osteogenesis and designated the superiority of application of N-BP to distraction osteogenesis. As further lengthening and time course are required for clinical cases, the application of N-BP to human lengthening should be verified in quest of both efficacy and convenience within an absolute safety. Acknowledgments This work was supported, in part, by grants from the Japan Society for the Promotion of Science (grant no. 16390441 and no. 14370465). We thank Toshiharu Yokoyama for facilitating mechanical testing and Astellas Pharmaceutical Company for donation of YM529/ONO5920. References [1] Green SA, Jackson JM, Wall DM, Marinow H, Ishkanian J. Management of segmental defects by the Ilizarov intercalary bone transport method. Clin Orthop Relat Res 1992;280:136–42. [2] Cattaneo R, Catagni M, Johnson EE. The treatment of infected nonunions and segmental defects of the tibia by the methods of Ilizarov. Clin Orthop Relat Res 1992;280:143–52. [3] Paley D, Catagni M, Argnani F, Prevot J, Bell D, Armstrong P. Treatment of congenital pseudoarthrosis of the tibia using the Ilizarov technique. Clin Orthop Relat Res 1992;280:81–93. [4] Dendrinos GK, Kontos S, Lyritsis E. Use of the Ilizarov technique for treatment of non-union of the tibia associated with infection. J Bone Jt Surg, Am 1995;77(6):835–46. [5] Tsuchiya H, Abdel-Wanis ME, Sakurakichi K, Yamashiro T, Tomita K. Osteosarcoma around the knee. Intraepiphyseal excision and biological reconstruction with distraction osteogenesis. J Bone Jt Surg, Br 2002; 84(8):1162–6. [6] Ilizarov GA. The tension-stress effect on the genesis and growth of tissues: Part I. The influence of stability of fixation and soft-tissue preservation. Clin Orthop Relat Res 1989;238:249–81. [7] Ilizarov GA. The tension-stress effect on the genesis and growth of tissues: Part II. The influence of the rate and frequency of distraction. Clin Orthop Relat Res 1989;239:263–85. [8] Sato M, Yasui N, Nakase T, Kawahata H, Sugimoto M, Hirota S, et al. Expression of bone matrix proteins mRNA during distraction osteogenesis. J Bone Miner Res 1998;13(8):1221–31. [9] Sato M, Ochi T, Nakase T, Hirota S, Kitamura Y, Nomura S, et al. Mechanical tension-stress induces expression of bone morphogenetic protein (BMP)-2 and BMP-4, but not BMP-6, BMP-7, and GDF-5 mRNA, during distraction osteogenesis. J Bone Miner Res 1999;14(7):1084–95. [10] Okazaki H, Kurokawa T, Nakamura K, Matsushita T, Mamada K, Kawaguchi H. Stimulation of bone formation by recombinant fibroblast growth factor-2 in callotasis bone lengthening of rabbits. Calcif Tissue Int 1999;64(6):542–6. [11] Stewart KJ, Weyand B, van't Hof RJ, White SA, Lvoff GO, Maffulli N, et al. A quantitative analysis of the effect of insulin-like growth factor-1 infusion during mandibular distraction osteogenesis in rabbits. Br J Plast Surg 1999;52(5):343–50. [12] Rauch F, Lauzier D, Travers R, Glorieux F, Hamdy R. Effects of locally applied transforming growth factor-beta1 on distraction osteogenesis in a rabbit limb-lengthening model. Bone 2000;26(6):619–24. [13] Li G, Bouxsein ML, Luppen C, Li XJ, Wood M, Seeherman HJ, et al. Bone

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25] [26]

[27]

[28]

[29]

[30]

[31]

[32]

consolidation is enhanced by rhBMP-2 in a rabbit model of distraction osteogenesis. J Orthop Res 2002;20(4):779–88. Eckardt H, Bundgaard KG, Christensen KS, Lind M, Hansen ES, Hvid I. Effects of locally applied vascular endothelial growth factor (VEGF) and VEGF-inhibitor to the rabbit tibia during distraction osteogenesis. J Orthop Res 2003;21(2):335–40. Hamdy RC, Amako M, Beckman L, Kawaguchi M, Rauch F, Lauzier D, et al. Effects of osteogenic protein-1 on distraction osteogenesis in rabbits. Bone 2003;33(2):248–55. Rauch F, Lauzier D, Croteau S, Travers R, Glorieux FH, Hamdy R. Temporal and spatial expression of bone morphogenetic protein-2, -4, and -7 during distraction osteogenesis in rabbits. Bone 2000;27(3):453–9. Pacicca DM, Patel N, Lee C, Salisbury K, Lehmann W, Carvalho R, et al. Expression of angiogenic factors during distraction osteogenesis. Bone 2003;33(6):889–98. Kojimoto H, Yasui N, Goto T, Matsuda S, Shimomura Y. Bone lengthening in rabbits by callus distraction. The role of periosteum and endosteum. J Bone Jt Surg, Br 1988;70(4):543–9. Riggs BL, Parfitt AM. Drugs used to treat osteoporosis. the critical need for a uniform nomenclature based on their action on bone remodeling. J Bone Miner Res 2005;20(2):177–84. Smith EJ, McEvoy A, Little DG, Baldock PA, Eisman JA, Gardiner EM. Transient retention of endochondral cartilaginous matrix with bisphosphonate treatment in a long-term rabbit model of distraction osteogenesis. J Bone Miner Res 2004;19(10):1698–705. Little DG, Smith NC, Williams PR, Briody JN, Bilston LE, Smith EJ, et al. Zoledronic acid prevents osteopenia and increases bone strength in a rabbit model of distraction osteogenesis. J Bone Miner Res 2003;18(7): 1300–7. Little DG, Cornell MS, Hile MS, Briody J, Cowell CT, Bilston L. Effect of pamidronate on distraction osteogenesis and fixator-related osteoporosis. Injury 2001;32(Suppl 4):SD14–20. Little DG, Cornell MS, Briody J, Cowell CT, Arbuckle S, CookeYarborough CM. Intravenous pamidronate reduces osteoporosis and improves formation of the regenerate during distraction osteogenesis. A study in immature rabbits. J Bone Jt Surg, Br 2001;83(7):1069–74. Parfitt AM, Drezner MK, Glorieux FH, Kanis JA, Malluche H, Meunier PJ, et al. Bone histomorphometry. standardization of nomenclature, symbols, and units. Report of the ASBMR Histomorphometry Nomenclature Committee. J Bone Miner Res 1987;2(6):595–610. Fleisch H. Bisphosphonates in Bone Disease. From the Laboratory to the Patient. 4th ed. San Diego Academic; 2000. Dunford JE, Thompson K, Coxon FP, Luckman SP, Hahn FM, Poulter CD, et al. Structure-activity relationships for inhibition of farnesyl diphosphate synthase in vitro and inhibition of bone resorption in vivo by nitrogen-containing bisphosphonates. J Pharmacol Exp Ther 2001;296(2): 235–42. Yoshida Y, Moriya A, Kitamura K, Inazu M, Okimoto N, Okazaki Y, et al. Responses of trabecular and cortical bone turnover and bone mass and strength to bisphosphonate YH529 in ovariohysterectomized beagles with calcium restriction. J Bone Miner Res 1998;13(6):1011–22. Sasaki A, Kitamura K, Alcalde RE, Tanaka T, Suzuki A, Etoh Y, et al. Effect of a newly developed bisphosphonate, YH529, on osteolytic bone metastases in nude mice. Int J Cancer 1998;77(2):279–85. Miwa S, Mizokami A, Keller ET, Taichman R, Zhang J, Namiki M. The bisphosphonate YM529 inhibits osteolytic and osteoblastic changes and CXCR-4-induced invasion in prostate cancer. Cancer Res 2005;65(19): 8818–25. Cadet ER, Gafni RI, McCarthy EF, McCray DR, Bacher JD, Barnes KM, et al. Mechanisms responsible for longitudinal growth of the cortex. coalescence of trabecular bone into cortical bone. J Bone Jt Surg, Am 2003;85-A(9):1739–48. Martin TJ, Ng KW. Mechanisms by which cells of the osteoblast lineage control osteoclast formation and activity. J Cell Biochem 1994;56 (3):357–66. Jimi E, Nakamura I, Amano H, Taguchi Y, Tsurukai T, Tamura M, et al. Osteoclast function is activated by osteoblastic cells through a mechanism involving cell-to-cell contact. Endocrinology 1996;137(8):2187–90.

M. Takahashi et al. / Bone 39 (2006) 573–581 [33] Kaneko H, Arakawa T, Mano H, Kaneda T, Ogasawara A, Nakagawa M, et al. Direct stimulation of osteoclastic bone resorption by bone morphogenetic protein (BMP)-2 and expression of BMP receptors in mature osteoclasts. Bone 2000;27(4):479–86. [34] Kawaguchi H, Chikazu D, Nakamura K, Kumegawa M, Hakeda Y. Direct and indirect actions of fibroblast growth factor 2 on osteoclastic bone resorption in cultures. J Bone Miner Res 2000;15(3):466–73. [35] Tenenbaum HC, Torontali M, Sukhu B. Effects of bisphosphonates and inorganic pyrophosphate on osteogenesis in vitro. Bone 1992;13(3): 249–255. [36] Giuliani N, Pedrazzoni M, Negri G, Passeri G, Impicciatore M, Girasole G. Bisphosphonates stimulate formation of osteoblast precursors and mineralized nodules in murine and human bone marrow cultures in vitro and promote early osteoblastogenesis in young and aged mice in vivo. Bone 1998;22(5):455–61. [37] Giuliani N, Pedrazzoni M, Passeri G, Negri G, Impicciatore M, Girasole G. Bisphosphonates stimulate the production of basic fibroblast growth factor and the formation of bone marrow precursors of osteoblasts. New findings about their mechanism of action. Minerva Med 1998;89(7–8):249–58. [38] Reinholz GG, Getz B, Pederson L, Sanders ES, Subramaniam M, Ingle JN, et al. Bisphosphonates directly regulate cell proliferation, differentiation, and gene expression in human osteoblasts. Cancer Res 2000;60(21): 6001–7. [39] Pan B, To LB, Farrugia AN, Findlay DM, Green J, Gronthos S, et al. The nitrogen-containing bisphosphonate, zoledronic acid, increases

[40]

[41] [42]

[43]

[44] [45]

[46]

[47]

581

mineralisation of human bone-derived cells in vitro. Bone 2004;34 (1):112–23. Schindeler A, Little DG. Osteoclasts but not osteoblasts are affected by a calcified surface treated with zoledronic acid in vitro. Biochem Biophys Res Commun 2005;338(2):710–6. Aronson J. Modulation of distraction osteogenesis in the aged rat by fibroblast growth factor. Clin Orthop Relat Res 2004;425:264–83. Little DG, McDonald M, Bransford R, Godfrey CB, Amanat N. Manipulation of the anabolic and catabolic responses with OP-1 and zoledronic acid in a rat critical defect model. J Bone Miner Res 2005; 20(11):2044–52. Hoffman A, Stepensky D, Ezra A, Van Gelder JM, Golomb G. Mode of administration-dependent pharmacokinetics of bisphosphonates and bioavailability determination. Int J Pharm 2001;220(1–2):1–11. Body JJ, Diel I, Bell R. Profiling the safety and tolerability of bisphosphonates. Semin Oncol 2004;31(5 Suppl 10):73–8. Marx RE. Pamidronate (Aredia) and zoledronate (Zometa) induced avascular necrosis of the jaws. A growing epidemic. J Oral Maxillofac Surg 2003;61(9):1115–7. Amanat N, Brown R, Bilston LE, Little DG. A single systemic dose of pamidronate improves bone mineral content and accelerates restoration of strength in a rat model of fracture repair. J Orthop Res 2005;23(5): 1029–1034. Li J, Mori S, Kaji Y, Kawanishi J, Akiyama T, Norimatsu H. Concentration of bisphosphonate (incadronate) in callus area and its effects on fracture healing in rats. J Bone Miner Res 2000;15(10):2042–51.