Targeting osteoclasts with zoledronic acid prevents bone destruction in collagen-induced arthritis.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 50, No. 7, July 2004, pp 2338–2346 DOI 10.1002/art.20382 © 2004, American College of Rheumatology Targeting Osteoclasts With Zoledronic Acid Prevents Bone Destruction in Collagen-Induced Arthritis Natalie A. Sims,1 Jonathan R. Green,2 Markus Glatt,2 Stephen Schlict,3 T. John Martin,4 Matthew T. Gillespie,5 and Evan Romas1 Objective. To study the effect of zoledronic acid (ZA) on synovial inflammation, structural joint damage, and bone metabolism in rats during the effector phase of collagen-induced arthritis (CIA). Methods. CIA was induced in female dark agouti rats. At the clinical onset of CIA, rats were assigned to treatment with vehicle or single subcutaneous doses of ZA (1.0, 10, 50, or 100 g/kg). Clinical signs in all 4 paws were scored on a daily basis. After 2 weeks, the joints in the hind paws were assessed using plain radiographs, microfocal computed tomography (microCT), histologic scoring, and histomorphometry, and the serum levels of type I collagen crosslinks were measured by enzyme-linked immunosorbent assay. Results. Although ZA mildly exacerbated synovitis, it effectively suppressed structural joint damage. At doses of >10 g/kg, ZA significantly reduced radiographic bone erosions, Larsen scores, and juxtaarticular trabecular bone loss as quantified by micro-CT. ZA prevented increased type I collagen (bone) breakdown in CIA and diminished histologic scores of focal bone erosion by up to 80%. Increases in the percentage of eroded surface, osteoclast surface, and osteoclast num- bers associated with CIA were prevented by ZA, even though synovitis scores were unchanged. Conclusion. Single doses (>10 g/kg) of ZA strikingly reduced focal bone erosions and juxtaarticular trabecular bone loss, although synovitis was mildly exacerbated. Targeting osteoclasts with ZA may therefore be an effective strategy for preventing structural joint damage in rheumatoid arthritis. Rheumatoid arthritis (RA) is a chronic inflammatory disease characterized by progressive destruction of synovial joints. Osteoclasts, the cells responsible for bone degradation, play a pivotal role in the joint destruction seen in RA (1–3) as well as animal models of RA (4–7). Interactions between activated T cells and macrophages drive the process of bone destruction that occurs in RA by releasing proinflammatory cytokines, especially tumor necrosis factor ␣ (TNF␣), interleukin-1 (IL-1), and IL-17, which transform myeloid precursor cells and synovial fibroblasts into tissue-destructive effector cells (8,9). TNF␣ from activated macrophages, in the presence of permissive levels of RANKL (which is produced by osteoblasts, synovial fibroblasts, and activated T cells), stimulates production of osteoclasts (10). The importance of osteoclasts in arthritic bone destruction has been verified by administration of osteoprotegerin (OPG; a decoy receptor for RANKL) in spontaneously arthritic TNF␣-transgenic mice and models such as collagen-induced arthritis (CIA) and adjuvant-induced arthritis, as well as a serum transfer model of arthritis using RANKL-deficient mice, all of which demonstrated reduced bone damage and dissociated synovial inflammation from bone destruction (4,6,11,12). Osteoclasts were also directly implicated in TNF␣-mediated bone loss, because crossing spontaneously arthritic human TNF␣-transgenic mice with osteopetrotic c-fos–deficient mice lacking osteoclasts resulted Supported by the University of Melbourne Early Career Research Grant Scheme, the National Health and Medical Research Council (NHMRC; project grant 247909), and the Arthritis Foundation of Australia. Dr. Sims’ work was supported by an NHMRC Career Development Award. 1 Natalie A. Sims, PhD, Evan Romas, MBBS, PhD: St. Vincent’s Hospital, University of Melbourne, Melbourne, Victoria, Australia; 2Jonathan R. Green, PhD, Markus Glatt, PhD: Novartis Pharma, Basel, Switzerland; 3Stephen Schlict, MBBS: St. Vincent’s Hospital, Melbourne, Victoria, Australia; 4T. John Martin, MD, DSc: University of Melbourne, Melbourne, Victoria, Australia; 5Matthew T. Gillespie, PhD: St. Vincent’s Institute, Melbourne, Victoria, Australia. Address correspondence and reprint requests to Evan Romas, MBBS, PhD, Department of Medicine, University of Melbourne, St. Vincent’s Hospital, 41 Victoria Parade Fitzroy 3065, Melbourne, Victoria, Australia. E-mail: email@example.com. Submitted for publication October 7, 2003; accepted in revised form April 13, 2004. 2338 ZOLEDRONATE PREVENTION OF BONE EROSIONS IN CIA in inflammatory arthritis without bone destruction (7). Together, these lines of evidence indicate that suppressing osteoclast numbers or function can be a strategy for preserving skeletal integrity in inflammatory arthritis. The bisphosphonates inhibit osteoclast formation, function, and survival at least in part by inhibiting the mevalonate pathway enzyme, farnesyl diphosphate synthase (13–15). Zoledronic acid (ZA; Zoledronate) is a third-generation bisphosphonate (16,17) that is reported to have the greatest suppressive effect on bone resorption. Although it was shown to be beneficial in clinical trials of osteoporosis (18), Paget’s disease of bone (19), renal transplantation–related bone loss (20), and metastatic and osteolytic bone disease (21,22), ZA has not been rigorously studied for its potential to prevent bone destruction in models of RA. The present study was designed to investigate the effects of ZA on synovial inflammation, joint structure, and bone resorption in rats during the effector phase of CIA. The results show that single doses of ZA (ⱖ10 g/kg) administered at the onset of arthritis are sufficient to prevent focal and juxtaarticular bone loss caused by arthritis, in the absence of therapeutic effects on synovitis. MATERIALS AND METHODS Reagents. Lyophilized native bovine type II collagen (Sigma, St. Louis, MO) was dissolved at a concentration of 2 mg/ml in 0.01 moles/liter acetic acid. ZA was supplied by Novartis Pharma AG (Basel, Switzerland). Experimental animals. Female 9-week-old dark agouti rats (n ⫽ 46) were used as the model for arthritis. Six rats served as nonarthritic controls, and CIA was induced in the remaining 40 (8 rats per treatment group). All animals were fed standard rodent food and given water and 0.9% NaCl for drinking ad libitum. The animals were weighed and monitored daily for signs of arthritis. All experimental procedures conformed to the National Health and Medical Research Council guidelines and were approved by an institutional animal ethics committee. The rats were maintained at 22°C in an animal room that was illuminated for 12 hours daily. After a 1-week acclimatization period, the animals were used for experiments. Induction of CIA. CIA was induced as previously described (5,6). CIA developed in all of the immunized rats after a mean (⫾SEM) interval of 19.5 ⫾ 0.5 days; the day of arthritis onset was similar in all treatment groups. Clinical assessment of CIA. Arthritis was monitored daily by a macroscopic scoring system (range 0–4 for each paw), as previously described (5,6). The scoring system is as follows: 0 ⫽ no arthritis, 1 ⫽ swelling and/or redness of 1–2 interphalangeal (IP) joints, 2 ⫽ involvement of 3–4 IP joints or 1 larger joint, 3 ⫽ more than 4 joints red/swollen, and 4 ⫽ severe arthritis of an entire paw. Use of this system yielded a score between 0 and 16 (maximum possible score) per animal. In addition, the width of each paw was measured, using digital calipers, at baseline and 3 times weekly beginning at the onset 2339 of arthritis. The percentage of change in paw thickness for each paw was calculated from baseline measurements and was expressed as an average for each rat, for statistical analysis. Administration of ZA. At the clinical onset of arthritis (the first day on which the clinical score was ⱖ2), rats were randomly assigned to receive a single subcutaneous injection of phosphate buffered saline (PBS) (vehicle) or ZA (1.0, 10, 50, or 100 g/kg). Tissue collection and specimen preparation. Two weeks after the onset of arthritis, blood was collected by cardiac puncture exsanguination. The hind paws of each rat were dissected and fixed in 4% paraformaldehyde/PBS overnight at 4°C, then stored in 70% ethanol for radiographic and microfocal computed tomographic (micro-CT) analyses. Following these procedures, individual toes of the hind paws were dissected and decalcified with 15% EDTA in 0.5% paraformaldehyde/PBS at 4°C for 1 week before standard processing for paraffin wax embedding. Serial 5-m sagittal sections were obtained through the digits and stained with hematoxylin and eosin (H&E) for general histologic evaluation or toluidine blue for the specific assessment of cartilage. To identify osteoclasts, sections were stained for tartrate-resistant acid phosphatase (TRAP) using an acid phosphatase kit (Sigma). Radiographic examination. Contact radiographs of both hind feet were obtained using a Faxitron x-ray cabinet (Faxitron, Wheeling, IL) and high-resolution mammography film (Eastman Kodak, Rochester, NY). Destruction of bone and cartilage was classified and scored according to the Larsen method (23), without knowledge of the assigned treatment groups. Micro-CT. Left hind paws were analyzed with a CT 20 scanner (Scanco Medical, Bassersdorf, Switzerland). The paws were scanned perpendicular to their longitudinal axis by a fixed x-ray fan beam (10 m spot-size tube, 0.1 mA, 50 kVp) while the holder rotated along its axis. For 3-dimensional (3-D) data accumulation, the sample was moved incrementally along its axis after each turn. A total of 300 slices (1,024 ⫻ 1,024–pixel matrix per slice) of 13 m thickness were determined, yielding a voxel size of 13 ⫻ 13 ⫻ 13 m3. The resulting scan length of 3.9 mm was sufficiently long to cover the distal metatarsal joint area of the third metatarsal bone. The distal growth plate (“visible” as empty clefts) was used to position a core volume of interest for 3-D reconstruction. A cancellous core of 100 slices of 13 m thickness (31.3-mm length) was finally evaluated with MicroCT software (Scanco Medical) for bone morphometry. The primary spongiosa was excluded from measurement. The spatial resolution with the given settings of the instrument was 24 m; the precision of repeated measurements was ⬍1%. Threshold settings were optimized with histomorphometric methods. Total volume, bone volume, bone surface, trabecular number, trabecular separation, and trabecular thickness were measured in the volume of interest after removal of cortical bone using a semiautomatic morphing procedure. Trabecular number, trabecular separation, and trabecular thickness denote model-independent 3-D calculations (24). Enzyme-linked immunosorbent assay (ELISA) for type I collagen crosslinks. Serum was isolated by centrifugation and stored at ⫺80°C before analysis. The serum concentration of type I collagen crosslinks, a measure of bone 2340 resorption breakdown products, was measured by ELISA (RatLaps ELISA; Nordic Bioscience, Herlev, Denmark). Histopathologic assessment. The histopathologic assessment focused on the hind paw digits that show the most severe effects of joint destruction. Because the distal IP (DIP) joints were the first to become inflamed, followed by the proximal IP (PIP) and metatarsophalangeal (MTP) joints, each joint was scored separately to allow analysis of the effects of ZA at different stages of inflammation and arthritic destruction. For standardization, sections bisecting the digits were used to ensure that the proximal, middle, and distal phalanges with their opposing cortices and cartilage ends were present. At least 2 digits were randomly selected from each rat for histologic assessment, without knowledge of specific interventions. H&E-stained sections were scored by 2 independent observers, at low power for inflammation and pannus, and at low (⫻10) and high (⫻100 or ⫻200) power for bone erosion, with consultation to TRAP-stained sections when required. Sequential toluidine blue–stained sections were then scored for cartilage damage and proteoglycan loss. Inflammation was scored 0–4 according to the following criteria: 0 ⫽ normal, 1 ⫽ minimal inflammatory infiltration, 2 ⫽ mild infiltration, 3 ⫽ moderate infiltration with lymphoid aggregates, 4 ⫽ marked infiltration with lymphoid aggregates and edema. Pannus formation was defined as synovial proliferation adjacent to cartilage and filling the joint space and was scored 0–3 as follows: 0 ⫽ none, 1 ⫽ minimal, 2 ⫽ moderate (invasion of ⬍50% of the cartilage surface), 3 ⫽ severe (invasion of ⱖ50% of the cartilage surface). Articular cartilage damage was scored 0–4 according to the following criteria: 0 ⫽ normal, 1 ⫽ minimal (loss of toluidine blue staining only, indicating proteoglycan loss but no tissue damage), 2 ⫽ mild (loss of toluidine blue staining, and cartilage thinning of the superficial zone only), 3 ⫽ moderate (loss of cartilage to the second articular cartilage zone or beyond), 4 ⫽ marked (regions of exposed bone below the articular cartilage). Bone resorption was scored 0–4 according to the following criteria: 0 ⫽ none, 1 ⫽ minimal (1–2 sites of resorption, visible only at high magnification), 2 ⫽ mild (at least 3 sites of resorption, visible only at high magnification), 3 ⫽ moderate (obvious foci of resorption, visible at low power), 4 ⫽ marked (large erosions extending through to the marrow space). To enumerate osteoclasts, histomorphometric analysis was performed on masked H&E-stained sections by adapting standard procedures using the OsteoMeasure analysis system (OsteoMetrics, Decatur, GA). Osteoclasts were defined using the standard criterion of multinucleated cells residing in resorptive lacunae on the bone surface. Both IP and MTP joints were included in the analysis, which was carried out on the periosteal surface (the cortical bone surface facing the external environment) over a region 300 m from the chondroosseus junction toward the midshaft. At each joint, the eroded surface (ES), osteoclast surface (OcS), osteoclast number (NOc), and osteoclast area were expressed as a ratio in relation to the bone surface (BS), bone perimeter (BPm), or ES. Statistical analysis. Differences between groups were analyzed by one- or two-way analysis of variance followed by Fisher’s post hoc test. P values less than 0.05 were considered significant. SIMS ET AL Figure 1. A, Inflammation scores in rats treated with or without zoledronic acid (ZA). Arthritis was evaluated daily by a clinical score (maximum score per animal 16; see Materials and Methods). B, Effect of ZA on paw swelling. For each hind paw, the footpad thickness was measured and expressed as a percentage of baseline thickness; an average value was determined for each rat. Values are the mean ⫾ SEM. ⴙ ⫽ first day at which the P value was ⬍0.05 versus nonarthritic rats; ⴱ ⫽ P ⬍ 0.05 versus collagen-induced arthritis (CIA)–untreated rats at the same time point. Z ⫽ zoledronic acid. RESULTS Changes in clinical markers of incidence and severity of arthritis. The rate of progression of CIA after onset was not significantly altered by ZA treatment (Figure 1A). In rats treated with 100 g/kg of ZA, the arthritis score was higher on days 10–13 compared with untreated rats with CIA (Figure 1A). Paw thickness peaked 1 week after the onset of CIA. All doses of ZA enhanced paw swelling (Figure 1B); the most severe exacerbation was detected in rats treated with 100 g/kg of ZA 4–7 days after disease onset. These differences in paw thickness between groups were no longer present at 2 weeks. Findings on plain radiographic examination. Although ZA mildly exacerbated arthritis, the treated rats exhibited clear protection against structural joint damage. In fact, doses of ⱖ10 g/kg were sufficient to block the significant increase in Larsen scores observed in ZOLEDRONATE PREVENTION OF BONE EROSIONS IN CIA 2341 Figure 2. Zoledronic acid (ZA; Zol) prevents structural damage in collageninduced arthritis (CIA) model. A, Plain radiographs of the hind paw obtained 2 weeks after arthritis onset in a normal rat (left), a rat with CIA that did not receive ZA (middle), and a rat with CIA that received 100 g/kg of ZA (right). Note the damage highlighted in the boxed area (middle) and prevention of damage by ZA treatment (right). B, Scoring of the tarsometatarsal and metatarsophalangeal joints using the Larsen method showed prevention of arthritic joint destruction by ZA doses of ⱖ10 g/kg. max ⫽ maximum. C, Microfocal computed tomographic images showing dorsal surface reconstruction of the hind paws from a normal rat (left), an untreated arthritic rat (middle), and an arthritic rat treated with 100 g/kg of ZA (right). Note the joint destruction at the distal third metatarsal joint (open arrow) and periosteal new bone formation due to periostitis at the tarsus and proximal metatarsal joints. D, Percentage of trabecular bone volume (Vol) ascertained at the distal third metatarsal bone, as described in Materials and Methods. Values are the mean and SEM. ⴱ ⫽ P ⬍ 0.05 versus no CIA; ⴱⴱ ⫽ P ⬍ 0.01 versus no CIA; ⴙⴙⴙ ⫽ P ⬍ 0.001 versus untreated CIA. untreated CIA rats at the hind paw, because this parameter of disease severity was reduced by ⬃50% compared with untreated CIA rats (Figures 2A and B). Findings on micro-CT analysis. To further elucidate the effects of ZA on periarticular bone integrity, micro-CT analysis was performed on arthritic rats treated with or without ZA. Micro-CT imaging demonstrated that ZA prevented focal bone erosions at the tarsal, MTP, and IP joints (Figure 2C). The new periosteal bone formation observed in the tarsus has been described previously in CIA (25,26), and, to avoid confounding effects, the tarsal region was not included in the micro-CT analysis. To reveal the effect of ZA on juxtaarticular bone, the trabecular bone volume at the distal end of the middle metatarsal bone was evaluated by quantitative imaging. Trabecular bone volume at this site was reduced by ⬎60% in the untreated CIA group; ZA not only abrogated this loss but led to trabecular bone 2342 SIMS ET AL Table 1. ZA-induced increase in serum type I collagen crosslinks (RatLaps) associated with CIA* Treatment group RatLaps, ng/ml No CIA Untreated CIA CIA ⫹ 1 g/kg ZA CIA ⫹ 10 g/kg ZA CIA ⫹ 50 g/kg ZA CIA ⫹ 100 g/kg ZA 19.1 ⫾ 2.8 31.2 ⫾ 4.0† 16.9 ⫾ 5.2 18.9 ⫾ 4.3 22.1 ⫾ 3.6 20.5 ⫾ 5.2 * Values are the mean ⫾ SEM for 6–8 rats per group. Serum samples were collected 2 weeks after the onset of clinical arthritis symptoms. ZA ⫽ zoledronic acid; CIA ⫽ collagen-induced arthritis. † P ⬍ 0.05 by one-way analysis of variance followed by Fisher’s post hoc test, versus controls (no CIA). accrual despite continued inflammation (Figure 2D). ZA maintained trabecular bone volume by preventing CIA-induced reductions in both trabecular number and thickness (data not shown). Changes in systemic bone resorption. To ascertain the degree of bone resorption occurring in the whole animal, the levels of circulating type I collagen crosslinks were quantitated by ELISA. At 2 weeks, the mean level of type I collagen crosslinks was almost 2-fold higher in the arthritic rats compared with the nonarthritic rats (Table 1). ZA abrogated the increased concentration of type I collagen crosslinks in CIA, at all doses tested. Histopathology of arthritis. To elucidate the effects of ZA treatment on joint structure at the histologic level, inflamed joints were scored with semiquantitative grading scales. In the toes, inflammatory synovitis was most consistent and severe in the IP and MTP joints, which therefore were assessed in detail. Bone erosions.The most striking effects of ZA were reflected in the bone erosion scores. Arthritic rats treated with ZA had significantly lower bone damage scores compared with vehicle-treated rats (Figure 3B). At a dose of 100 g/kg, ZA reduced bone erosion scores by at least 80% compared with vehicle. A dose-dependent relationship was Figure 3. A, Joint regions used for histologic evaluation of B, bone erosion, C, cartilage damage, D, pannus formation, and E, synovial inflammation in rats with collageninduced arthritis (CIA). The distal interphalangeal (DIP), proximal IP (PIP), and metatarsophalangeal (MTP) joints from normal (nonarthritic) rats and arthritic rats treated with or without zoledronic acid (ZA; Z) were scored using a semiquantitative system (described in Materials and Methods). Results are the mean and SEM score. max ⫽ maximum; ⴙ ⫽ P ⬍ 0.05 versus control (nonarthritic) rats; ⴙⴙ ⫽ P ⬍ 0.01 versus control (nonarthritic) rats; ⴙⴙⴙ ⫽ P ⬍ 0.001 versus control (nonarthritic) rats; ⴱ ⫽ P ⬍ 0.05 versus untreated rats; ⴱⴱ ⫽ P ⬍ 0.01 versus untreated rats; ⴱⴱⴱ ⫽ P ⬍ 0.001 versus untreated rats. ZOLEDRONATE PREVENTION OF BONE EROSIONS IN CIA 2343 Figure 4. Effects of zoledronic acid (ZA) on osteoclasts and bone erosion sites in collagen-induced arthritis (CIA). Low-power images of an untreated joint (A) and a joint treated with 10 g/kg of ZA (B) clearly show sites of reduced erosion in ZA-treated joints (white arrows). Area boxed in black (A) indicates region of new periosteal bone formation. Higher-power images (C and D) of regions boxed in green in A and B show that although numerous osteoclasts (black arrows) can be seen in erosion sites in untreated CIA (C), these were rarely detected in joints from rats treated with 10 g of ZA (D). Histomorphometric analysis (E) (see Materials and Methods) demonstrated significantly reduced erosion sites (eroded surface/bone surface [ES/BS]), osteoclast surface (OcS/BS), and osteoclast number (NOc/bone perimeter [BPm]) per unit of joint surface in CIA rats treated with 10 g, 50 g, and 100 g/kg of ZA. In contrast, the number of osteoclasts per unit of eroded surface (NOc/eroded perimeter [EPm]) was significantly increased in rats with CIA that were treated with 50 g and 100 g of ZA, indicating that an increased number of osteoclasts is required for the same unit of eroded surface (reduced osteoclast activity). Values are the mean and SEM. ⴙⴙⴙ ⫽ P ⬍ 0.001 versus control (nonarthritic) rats; ⴱ ⫽ P ⬍ 0.05 versus untreated rats with CIA; ⴱⴱⴱ ⫽ P ⬍ 0.001 versus untreated rats with CIA. Bar ⫽ 500 m in A; 100 m in C. present at the DIP joints, with 100 g/kg of ZA providing the greatest bone protection, and a dose of 1 g/kg providing no significant protection. Cartilage damage. Arthritic rats exhibited similar degrees of cartilage damage regardless of the assigned treatment group (Figure 3C). 2344 SIMS ET AL Synovial inflammation and pannus. Consistent with the lack of clinical antiinflammatory effect, ZA in doses ranging from 1–100 g/kg did not influence histologic scores for synovial inflammation or pannus invasion at either the IP or MTP joints (Figures 3D and E). Bone histomorphometry. To further quantify the effects of ZA, histomorphometric analysis of the bone surface at the joint was performed (Figure 4). This analysis showed that CIA was associated with increased eroded surface (ES/BS), osteoclast surface (OcS/BS), and osteoclast numbers (NOc/BPm) in the vicinity of inflamed joints, and all of these parameters were inhibited by ZA at doses of ⱖ10 g/kg (Figure 4). Furthermore, in arthritic rats treated with ⱖ50 g/kg of ZA, the number of osteoclasts per unit of eroded surface (NOc/ ES) was significantly increased (Figure 4). This increase in the NOc/ES quotient suggests reduced osteoclast activity in ZA-treated rats, because a greater number of osteoclasts are required to produce the same unit of eroded surface. Finally, there was no significant change in osteoclast size or the number of nuclei per osteoclast with either CIA or with ZA treatment at any dose (data not shown). DISCUSSION The ability of bisphosphonates to regulate bone turnover by suppressing osteoclast activity, together with their selective localization to bone, have inspired their widespread use for osteoporosis, Paget’s disease of bone, humoral hypercalcemia of malignancy, and metastatic cancer in bone (18–22). RA is characterized by multifaceted bone pathology, specifically generalized and juxtaarticular osteoporosis, as well as focal bone erosions at the joint (2,9,27). Experimental models of RA provide compelling verification that these focal bone erosions are dependent on osteoclastic activity that in turn is regulated by proinflammatory cytokines such as IL-1, TNF␣, and IL-17, as well as RANKL (4,6,7,11,13). We therefore hypothesized that ZA, as an osteoclast inhibitor, might be effective in preventing structural damage in arthritis. ZA is a third-generation bisphosphonate with inhibitory effects on bone resorptive activity that are 100-fold to 10,000-fold stronger than those of the second- and first-generation bisphosphonates pamidronate and etidronate, respectively (16,17). In the ovariectomized rat, ZA administered once weekly at a dose of 0.3 g/kg slowed this bone loss, and at a dose of 1.5 g/kg, bone loss was completely prevented (16). Similarly, 4 mg of ZA administered intravenously once per year to postmenopausal women with osteoporosis reduced biochemical markers of bone metabolism and increased bone mineral density (18). In the present study, ZA dose-dependently suppressed bone erosions in CIA even though inflammation was not reduced. In fact, at early time points, arthritis scores and paw swelling were exacerbated, but this did not result in worsening of the joint damage. This dissociation of synovial inflammation from bone erosion was also shown with the osteoclastogenesis inhibitor, OPG (4,6,12). Both the histopathologic and radiographic analyses revealed a significant bone protective effect of ZA in CIA. This was particularly clear in the histomorphometric and micro-CT analyses. This protection was less obvious in the Larsen damage scores, probably because of the lower resolution of plain radiographs, the later onset of inflammation in the ankle region, and the confounding factor of new periosteal bone formation that exists in this model (25,26). The effectiveness of ZA in suppressing bone erosions was comparable with the effects of OPG injections, highlighting the preeminence of osteoclastic activity in mediating these bone lesions (4,6,9,12). The effects of ZA in CIA are consistent with the inhibition of osteoclastic bone resorption by this agent in other models of clinical bone loss, such as osteoporosis in the ovariectomized rat and rhesus monkey (16). The main perturbation of bone metabolism in RA is increased osteoclastic bone resorption (28), which is correlated to high inflammatory disease activity (2). Focal bone destruction is strongly associated with generalized osteoporosis, indicating a common mechanism between these types of bone loss (27). A high bone turnover rate and decreased bone mineral density were also demonstrated in the periarticular bone of rats with experimental arthritis (4,28–32). In our study, micro-CT analysis clearly demonstrated that ZA prevented juxtaarticular trabecular bone loss. At doses of ⬎1.0 g/kg, ZA inhibited the increase in eroded surface (ES/BS), osteoclast surface (OcS/BS), and osteoclast number (NOc/BPm) that is associated with CIA. Furthermore, in arthritic rats treated with 50 g/kg or 100 g/kg of ZA, the number of osteoclasts per millimeter of eroded surface (NOc/ES) was significantly elevated, indicating a feedback mechanism responding to reduced osteoclast activity. Thus, ZA may exert its beneficial effect by down-regulating both osteoclast numbers and osteoclast activity. Like other bisphosphonates, ZA may induce osteoclast apoptosis (13–15) and suppress osteoclast function via accessory cells, such as osteoblasts (33). Aminobisphos- ZOLEDRONATE PREVENTION OF BONE EROSIONS IN CIA phonates also interact with the ␥/␦ T cell receptor, and their effects on osteoclastic resorption may be secondary to action on T cells (34), which are critical for osteoclastogenesis in RA (4,9). ZA also stimulates the release of OPG by osteoblasts (35), and OPG is expressed by synoviocytes in CIA and RA (6), suggesting that several distinct mechanisms of osteoclast inhibition might contribute to the actions of ZA in arthritis. It is important to note that in our study, ZA at doses as low as 1.0 g/kg suppressed increased levels of bone type I collagen breakdown products, but higher doses (ⱖ10 g/kg) were necessary for protection against focal bone erosions. Other third-generation bisphosphonates, such as minodronic acid, have also been reported to suppress bone turnover markers and bone mineral density loss caused by CIA (28). The basis for the mild exacerbation of CIA observed with high-dose ZA may be related to proinflammatory cytokine release, as described previously for both ZA and ibandronate (36). In the context of bisphosphonate studies in experimental arthritis, the suppressive effect of single doses of ZA on focal bone destruction is noteworthy. The earlier-generation bisphosphonates usually reduced markers of increased bone resorption and physiologic bone resorption at the growth plate, although evidence of inhibition of focal bone erosions was not compelling (37–42). A notable exception was pamidronate, which reduced the size of radiographic bone erosions in spontaneously arthritic TNF␣-transgenic mice to an extent similar to that observed with the TNF inhibitor infliximab (12). Clinical trials of bisphosphonates in RA generally demonstrated suppressive effects on generalized bone loss but not on focal bone erosions (43–46). In fact, protection against structural joint damage was reported only with high-dose pamidronate (44). Factors such as the absence of an extensive bone surface at the joint, lack of drug potency, or underdosing may explain why clinical trials of bisphosphonates have thus far failed to demonstrate modification of structural joint damage (47). The most potent bisphosphonates (e.g., ZA), which are associated with the most profound reductions in bone turnover markers, have not yet been examined in sufficiently powered controlled clinical trials. In conclusion, single doses of ZA effectively suppressed focal bone erosions and juxtaarticular bone loss in CIA, despite continuing synovitis. Targeting osteoclasts with ZA may be an effective adjunctive strategy for preventing structural joint damage and generalized osteoporosis in RA. 2345 ACKNOWLEDGMENTS We sincerely thank Ingrid Kriechbaum, Jeanette Dickens, and William Paspaliaris for expert technical assistance. REFERENCES 1. Gravallese EM, Harada Y, Wang JT, Gorn AH, Thornhill TS, Goldring SR. Identification of cell types responsible for bone resorption in rheumatoid arthritis and juvenile rheumatoid arthritis. Am J Pathol 1998;152:943–51. 2. Gough A, Sambrook P, Devlin J, Huissoon A, Njeh C, Robbins S, et al. Osteoclastic activation is the principal mechanism leading to secondary osteoporosis in rheumatoid arthritis. J Rheumatol 1998;25:1282–9. 3. Goldring SR. The final pathogenetic steps in focal bone erosions in rheumatoid arthritis. Ann Rheum Dis 2000;59 Suppl 1:i72–4. 4. Kong YY, Feige U, Sarosi I, Bolon B, Tafuri A, Morony S, et al. Activated T cells regulate bone loss and joint destruction in adjuvant arthritis through osteoprotegerin ligand. Nature 1999; 402:304–9. 5. Romas E, Bakharevski O, Hards DK, Kartsogiannis V, Quinn JM, Ryan PF, et al. Expression of osteoclast differentiation factor at sites of bone erosion in collagen-induced arthritis. Arthritis Rheum 2000;43:821–6. 6. Romas E, Sims NA, Hards DK, Lindsay M, Quinn JW, Ryan PF, et al. Osteoprotegerin reduces osteoclast numbers and prevents bone erosion in collagen-induced arthritis. Am J Pathol 2002;161: 1419–27. 7. Redlich K, Hayer S, Ricci R, David JP, Tohidast-Akrad M, Kollias G, et al. Osteoclasts are essential for TNF-␣-mediated joint destruction. J Clin Invest 2002;110:1419–27. 8. Rodan GA, Martin TJ. Therapeutic approaches to bone diseases. Science 2000;289:1508–14. 9. Romas E, Gillespie MT, Martin TJ. Involvement of receptor activator of NFB ligand and tumor necrosis factor-␣ in bone destruction in rheumatoid arthritis. Bone 2002;30:340–6. 10. Lam J, Takeshita S, Barker JE, Kanagawa O, Ross FP, Teitelbaum SL. TNF-␣ induces osteoclastogenesis by direct stimulation of macrophages exposed to permissive levels of RANK ligand. J Clin Invest 2000;106:1481–8. 11. Pettit AR, Ji H, von Stechow D, Muller R, Goldring SR, Choi Y, et al. TRANCE/RANKL knockout mice are protected from bone erosion in a serum transfer model of arthritis. Am J Pathol 2001;159:1689–99. 12. Redlich K, Hayer S, Maier A, Dunstan CR, Tohidast-Akrad M, Lang S, et al. Tumor necrosis factor ␣–mediated joint destruction is inhibited by targeting osteoclasts with osteoprotegerin. Arthritis Rheum 2002;46:785–92. 13. Hughes DE, Wright KR, Uy HL, Sasaki A, Yoneda T, Roodman GD, et al. Bisphosphonates promote apoptosis in murine osteoclasts in vitro and in vivo. J Bone Miner Res 1995;10:1478–87. 14. Luckman SP, Hughes DE, Coxon FP, Graham R, Russell G, Rogers MJ. Nitrogen-containing bisphosphonates inhibit the mevalonate pathway and prevent post-translational prenylation of GTP-binding proteins, including Ras. J Bone Miner Res 1998;13: 581–9. 15. Russell RG, Croucher PI, Rogers MJ. Bisphosphonates: pharmacology, mechanisms of action and clinical uses. Osteoporos Int 1999;9 Suppl 2:S66–80. 16. Green J. Zoledronate: the preclinical pharmacology. Br J Clin Pract Suppl 1996;87:16–18. 17. Widler L, Jaeggi KA, Glatt M, Muller K, Bachmann R, Bisping M, et al. Highly potent geminal bisphosphonates: from pamidronate disodium (Aredia) to zoledronic acid (Zometa). J Med Chem 2002;45:3721–38. 18. Reid IR, Brown JP, Burckhardt P, Horowitz Z, Richardson P, Trechsel U, et al. Intravenous zoledronic acid in postmenopausal 2346 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. women with low bone mineral density. N Engl J Med 2002;346: 653–61. Chung G, Keen RW. Zoledronate treatment in active Paget’s disease. Ann Rheum Dis 2003;62:275–6. Haas M, Leko-Mohr Z, Roschger P, Kletzmayr J, Schwarz C, Mitterbauer C, et al. Zoledronic acid to prevent bone loss in the first 6 months after renal transplantation. Kidney Int 2003;63: 1130–6. Rosen LS, Gordon D, Kaminski M, Howell A, Belch A, Mackey J, et al. Zoledronic acid versus pamidronate in the treatment of skeletal metastases in patients with breast cancer or osteolytic lesions of multiple myeloma: a phase III, double-blind, comparative trial. Cancer J 2001;7:377–87. Saad F, Gleason DM, Murray R, Tchekmedyian S, Venner P, Lacombe L, et al. A randomized, placebo-controlled trial of zoledronic acid in patients with hormone-refractory metastatic prostate carcinoma. J Natl Cancer Inst 2002;94:1458–68. Larsen A. How to apply Larsen score in evaluating radiographs of rheumatoid arthritis in long-term studies. J Rheumatol 1995;22: 1974–5. Ruegsegger P, Koller B, Muller R. A microtomographic system for the nondestructive evaluation of bone architecture. Calcif Tissue Int 1996;58:24–9. Tsi D, Khow A, Iino T, Kiso Y, Ono H. Effect of Brand’s glucosamine with essence of chicken on collagen-induced arthritis in rats. Life Sci 2003;73:2953–62. Jamieson TW, De Smet AA, Cremer MA, Kage KL, Lindsley HB. Collagen-induced arthritis in rats: assessment by serial magnification radiography. Invest Radiol 1985;20:324–30. Forsblad D’Elia H, Larsen A, Waltbrand E, Kvist G, Mellstrom D, Saxne T, et al. Radiographic joint destruction in postmenopausal rheumatoid arthritis is strongly associated with generalised osteoporosis. Ann Rheum Dis 2003;62:617–23. Yamane I, Hagino H, Okano T, Enokida M, Yamasaki D, Teshima R. Effect of minodronic acid (ONO-5920) on bone mineral density and arthritis in adult rats with collagen-induced arthritis. Arthritis Rheum 2003;48:1732–41. Osterman T, Virtamo T, Lauren L, Kippo K, Pasanen I, Hannuniemi R, et al. Slow-release clodronate in prevention of inflammation and bone loss associated with adjuvant arthritis. J Pharmacol Exp Ther 1997;280:1001–7. Muller K, Wiesenberg I, Jaeggi K, Green JR. Effects of the bisphosphonate zoledronate on bone loss in the ovariectomized and in the adjuvant arthritic rat. Arzneimittelforschung 1998;48: 81–6. Tanahashi M, Funaba Y, Itoh M, Kawabe N, Nakadate-Matsushita T. Inhibitory effects of TRK-530 on rat adjuvant arthritis. Pharmacology 1998;56:242–51. Francis MD, Hovancik K, Boyce RW. NE-58095: a diphosphonate which prevents bone erosion and preserves joint architecture in experimental arthritis. Int J Tissue React 1989;11:239–52. Sahni M, Guenther HL, Fleisch H, Collin P, Martin TJ. Bisphos- SIMS ET AL 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. phonates act on rat bone resorption through the mediation of osteoblasts. J Clin Invest 1993;91:2004–11. Das H, Wang L, Kamath A, Bukowski JF. V␥2V␦2 T-cell receptormediated recognition of aminobisphosphonates. Blood 2001;98: 1616–8. Viereck V, Emons G, Lauck V, Frosch KH, Blaschke S, Grundker C, et al. Bisphosphonates pamidronate and zoledronic acid stimulate osteoprotegerin production by primary human osteoblasts. Biochem Biophys Res Commun 2002;291:680–6. Richards PJ, Amos N, Williams AS, Williams BD. Pro-inflammatory effects of the aminobisphosphonate ibandronate in vitro and in vivo. Rheumatology (Oxford) 1999;38:984–91. Markusse HM, Lafeber GJ, Breedveld FC. Bisphosphonates in collagen arthritis. Rheumatol Int 1990;9:281–3. Osterman T, Kippo K, Lauren L, Hannuniemi R, Sellman R. Effect of clodronate on established collagen-induced arthritis in rats. Inflamm Res 1995;44:258–63. Nakamura M, Ando T, Abe M, Kumagai K, Endo Y. Contrast between effects of aminobisphosphonates and non-aminobisphosphonates on collagen-induced arthritis in mice. Br J Pharmacol 1996;119:205–12. Glatt M, Bray M, Bisping M, Blattler A. Effects of diphosphonate (APD, EHDP) on inflammation and bone turnover in adjuvant arthritis [abstract]. Experientia 1983;39:681. Kinne RW, Schmidt-Weber CB, Hoppe R, Buchner E, PalomboKinne E, Nürnberg E, et al. Long-term amelioration of rat adjuvant arthritis following systemic elimination of macrophages by clodronate-containing liposomes. Arthritis Rheum 1995;38: 1777–90. Matsuo A, Shuto T, Hirata G, Satoh H, Matsumoto Y, Zhao H, et al. Antiinflammatory and chondroprotective effects of the aminobisphosphonate incadronate (YM175) in adjuvant induced arthritis. J Rheumatol 2003;30:1280–90. Eggelmeijer F, Papapoulos SE, van Paassen HC, Dijkmans BA, Valkema R, Westedt ML, et al. Increased bone mass with pamidronate treatment in rheumatoid arthritis: results of a three-year randomized, double-blind trial. Arthritis Rheum 1996;39:396–402. Maccagno A, Di Giorgio E, Roldan EJ, Caballero LE, Perez Lloret A. Double blind radiological assessment of continuous oral pamidronic acid in patients with rheumatoid arthritis. Scand J Rheumatol 1994;23:211–4. Valleala H, Laasonen L, Koivula MK, Mandelin J, Friman C, Risteli J, et al. Two year randomized controlled trial of etidronate in rheumatoid arthritis: changes in serum aminoterminal telopeptides correlate with radiographic progression of disease. J Rheumatol 2003;30:468–73. England AJ, Reid DM, Capell H, Ralston SH. Effects of oral clodronate on disease activity and bone mineral density in rheumatoid arthritis [abstract]. Br J Rheumatol 1996;35:S21. Rehman Q, Lane NE. Bone loss: therapeutic approaches for preventing bone loss in inflammatory arthritis. Arthritis Res 2001;3:221–7.