Animal experiments were performed at the Department of Orthopedics at the University of Muenster. All procedures were performed with the permission of the local government's animal rights protection authorities and in accordance with the National Institute of Health guidelines for the use of laboratory animals (G49/2000). Twenty-eight female NZW rabbits were used in the study. All animals were fully grown, clinically healthy, and had a mean ± SD weight of 4,193.75 ± 298.84 gm at the time of surgery and a mean ± SD weight of 4,274.64 ± 462.49 gm at the time of death.
The animals were divided into 3 groups. In 12 rabbits, the anterior cruciate ligament (ACL) was resected to create anterior instability of the knee (group 1). In 12 control rabbits, only a sham operation was performed, without resection of the ACL (group 2). The 4 animals of group 3 underwent no surgical treatment.
All operations were performed under general anesthesia (intramuscular administration of ketamine and xylazine). All animals received an intramuscular injection of broad-spectrum antibiotics (Tardomyocel; Bayer, Leverkusen, Germany) as prophylaxis against infection. For postoperative pain medication, intramuscular metamizole (Novalgin; Aventis Pharmaceuticals, Bad Soden, Germany) was administered. Operations on both knee joints were performed under a single dose of anesthesia. The knee joints were opened using a medial paramedian skin incision of 3 cm in length and a medial parapatellar incision of the capsule. In animals of group 1, the ACL was resected using a tapering scalpel. The patella was dislocated laterally. In animals of group 2 (control group), the joint and skin were closed again with no further surgical treatment (sham operation). An intraoperative Lachman test was carried out to evaluate anterior instability. Before closure, the joints were rinsed thoroughly. Postoperatively, all animals were allowed to move freely in their cages (which measured 100 cm × 70 cm × 40 cm).
Four animals from each group were killed 3, 6, and 12 weeks after the operations. The 4 animals from group 3 (no surgical treatment) were killed after they reached maturity. The joints were evaluated macroscopically by using a self-developed grading system consisting of 4 different criteria: fibrillations and ulcerations of the hyaline cartilage, osteophyte formation, and joint effusion. Scores for fibrillations ranged from 0 for intact hyaline cartilage to 3 for marked fibrillations, scores for ulcerations ranged from 0 for normal to 2 for a large area of ulceration, scores for osteophyte formation ranged from 0 for no osteophytes to 3 for marked osteophyte formation, and scores for joint effusion ranged from 0 for no effusion to 3 for marked effusion. The total score ranged from 0 to 11, with 0 being a macroscopically intact knee joint and 11 being late-stage OA. After macroscopic grading, both distal femurs were immediately stored at ~80°C until further examination.
Tissue samples (3-5-mm thick) were fixed in 4% buffered paraformaldehyde for 2 days. After decalcification with buffered EDTA (20% EDTA, pH 7.4), the samples were dehydrated and embedded in paraffin. Sections were cut at a thickness of 5 μm, mounted on poly-L-lysine-coated glass slides, deparaffinized in xylene, and washed 3 times with distilled water and then with Tris buffered saline (TBS; pH 7.5) for 2 minutes each (washing procedure). Some of the sections were stained with Safranin O or with hematoxylin and eosin (H&E) to evaluate histologic changes of the cartilage and bone tissue according to the Mankin scale . The Mankin score included assessments of the structure, cellularity, Safranin stainability, and integrity of the tidemark. The scores for structure ranged from 0 for normal to 6 for complete disorganization of the cartilage, scores for cellularity ranged from 0 for normal to 3 for hypocellularity, scores for Safranin O stainability ranged from 0 for normal to 4 for no stainability, and scores for tidemark integrity ranged from 0 for an intact tidemark to 1 for blood vessels crossing the tidemark. The total score ranged from 0 to 14, where 0 = intact hyaline cartilage and 14 = late-stage OA.
Immunohistochemical method for labeling PTH1R
Deparaffined sections were hydrated and incubated for 5 min in proteinase K (ready-to-use; Dako, Hamburg, Germany). After washing in TBS, the slides were treated for 10 min with 3% H2O2 and blocked for 30 min in 3% bovine serum. A monoclonal mouse anti-rabbit PTH1R was used as primary antibody (IgG1, clone VFF-18, Bender MedSystems, Vienna, Austria). The sections were incubated with the primary antibody in PBS + 1% bovine serum albumin (diluted 1:50) over night at 4°C. After washing once again sections were incubated for 30 min in the secondary antibody anti-mouse IgG (K4001; Dako) at room temperature. After washing in TBS, the reaction was developed with aminoethylcarbazole chromogen substrate (ready-to-use; Dako) for 30 minutes at room temperature. To demonstrate specificity of binding of mouse anti-rabbit PTH1R in control experiments, sections were incubated (a) without primary antibody or, instead of primary antibody, (b) mouse IgG1 with irrelevant specificity (Aspergillus niger glucose oxidase; Dako) was used at the same concentration as the primary antibody. In all controls tested the specificity no positive staining of the tissue slices was visible.
Quantitative counting of cells
Pictures of histologic sections were taken with a CoolSnap-Pro Color camera (model A00J82025; Media Cybernetics, Silver Spring, MD). The cells were counted using Image-Pro Plus software for Windows, version 4.1 (Media Cybernetics). First, the total number of chondrocytes in an immunohistochemically stained section was determined. Afterward, the number of immunohistochemically stained cells was counted in the same section and measured in relation to the total number of chondrocytes. Zonal attribution of PTH1R positive chondrocytes was done with respect to the typical shape of chondrocytes at the different zones of articular cartilage according to Buckwalter et al. .
Statistical analysis was performed using the Statistical Package for Social Sciences, release 11.0 (SPSS, Munich, Germany). Student's t-test was performed for comparison of scores in the control and experimental groups. Spearman's coefficient was calculated to determine correlations. P values less than 0.05 were considered significant.