Twenty-four skeletally-mature, healthy, mixed small breed dogs of both genders, age 2 to 6 years and weighing 2.8 to 9 kg, were used for this experiment. The experimental dogs were purchased from the Preclinical Research Center (Chemon, Korea). The animals were kept in quarantine for 1 month and acclimated in cage confinement. Before participating in the experiment, physical and radiographic examinations were performed to ensure the absence of neurologic disease, patellar luxation or other orthopedic diseases, or to determine the animals had previously undergone surgical procedures involving either hind limb. The dogs were housed in individual cages in the departmental animal shed at the Department of Surgery, College of Veterinary Medicine, Chonbuk National University, Korea. They were fed a standard commercial diet (Precept, USA) and had access to water ad libitum.

After the animals were premedicated with atropine sulfate (0.05 mg/kg, SC; Dai Han Pharm, Korea), anesthesia was induced with thiopentone sodium (25 mg/kg, IV; Dai Han Pharmaceutical, Korea) and maintained with enflurane and oxygen delivered through a cuffed endotracheal tube.

In all of the experimental animals (n = 24), left stifle arthrotomy was performed using a standard parapatellar approach. An incision was made through the fascia lata just lateral to the patellar ligament. The joint was thoroughly examined to evaluate the condition of the cruciate ligaments and menisci. MPL was induced by placing purse string sutures around the parapatellar fibrocartilage and anchoring the patella to the fabellar ligament using monofilament sterile nylon sutures. Medial retinacular reinforcement (imbrication) and lateral release were also performed to interfere with the neutral tracking of the patella. Dogs in the sham group (n = 12) underwent sham operation on their right stifles, whereas animals in the non-surgical control group (n = 12) remained intact. For the sham operation, a scalpel blade was inserted into the right stifle through a lateral parapatellar stab incision to inflict minor trauma to the synovium while the other joint components remained intact.

Post-surgically, the animals were kept in close observation in a postoperative room and were administered cephalexin (25 mg/kg, IV; Union Korea Pharmaceutical, Korea) every 8 h, for 5 days and dexamethasone (0.2 mg/kg, IV; Daewon Pharmaceutical, Korea) every 6 h for 3 days. The external stitches were removed after the surgical wounds healed, and the animals were moved to the experimental animal shed after 10 days.

SF specimens were collected preoperatively and postoperatively at 1.5 month intervals from both stifles (index and contralateral). As much SF was collected as possible (0.2~0.45 mL) by arthrocentesis from the stifle joint and special care was taken to avoid contamination with blood. After noting the volume and appearance of the SF, the specimens were centrifuged at 12,000 × g for 10 min at 4℃ to remove the cells, and the supernatant was stored at -80℃ until biomarker assays were performed.

Blood samples were collected prior to and 3, 6, 9, and 12 months after experimental induction of OA, and subjected to TRAP and TIMP-2 assays. Three mL of blood samples from each animal were collected using a 5 mL disposable syringe (Hwajin Medical, Korea) and transferred to sterile screw-capped tubes without anticoagulant (Soyagreentec, Korea), left undisturbed at room temperature for 1 h to permit coagulation, and centrifuged at 1,500 × g for 15 min. The serum was collected and transferred to Eppendorf tubes and stored at -80℃ until it was used for the assays.

TRAP activity in the SF and serum specimens was measured using a biochemical assay in 96-well plates with para-nitrophenylphosphate (pNPP) (Sigma-Aldrich, USA) as a substrate. All reagents for this assay were purchased from a commercial supplier. To reduce sample viscosity and improving pipetting accuracy, the SF specimens were digested with 50 IU/mL of streptomyces hyaluronidase (Sigma-Aldrich, USA) at at 37℃ for 30 min prior to the assay. The serum specimens were also incubated at 37℃ without streptomyces hyaluronidase for 30 min to inactivate the acid phosphatase activity of the erythrocytes. All of the SF and serum samples were diluted 1:4 with 0.9% NaCl before the assay. The 0.9% NaCl solution was used as the negative control. SF and serum samples (100 µL) were added to 100 µL of the reaction mixture so that the final incubation medium contained 2.5 mM pNPP, (ditris salt), 0.1 M sodium acetate buffer, pH 5.8; 0.2 M KCl, 0.1% Triton X-100, 10 mM sodium tartrate, 1 mM ascorbic acid, and 100 µM FeCl₃. The concentration of TRAP was determined using 200 µL of the final incubation medium per well in the 96-well plate (Immulon 2 HB Flat Bottom Microtiter Plate; Dynex Technologies, USA). After incubating the 96-well plate for 1 h at 37℃, the liberated p-nitrophenol was converted to p-nitrophenolate by adding 50 µL of 0.9 M NaOH, and the absorbance was measured at 405 nm using a spectrophotometer (Cess UV 90c; Bioteck, USA). One unit (U) of TRAP activity corresponded to 1 mol of p-nitrophenol liberated per 1 min at 37℃. If the TRAP activity exceeded the range of the standard curve, the samples were further diluted and assayed.

SF and serum samples were assayed for TIM-2 activity using a commercially available TIMP-2 Biotrak ELISA system (Amersham, UK). The samples were assayed according to the protocol provided by the kit manufacturer. Briefly, 100 µL of the prepared standard and 100 µL of each sample were incubated at 25℃ for 1 h with 100 µL of the anti-TIMP-2 peroxidase conjugate. The wells were then aspirated, washed, and incubated with 100 µL of tetramethylbezidine hydrogen peroxide at 25℃ for 30 min. This reaction was stopped by the addition of 1.0 M sulphuric acid and the optical density was measured at 450 nm using a spectrophotometer (Cess UV 90c; Bioteck, USA). The concentrations of TIMP-2 were determined according to a standard curve and reported in ng/mL.

SF samples (40 µL) were incubated in Laemmli sample buffer (LSB) with β-mercaptoethanol (40 µL), heated to 90℃ for 5 min, and separated by SDS polyacrylamide gel electrophoresis on 4~20% Tris-glycine gradient gels (Bio-Rad Laboratories, USA). A sample of human MMP-2 protein (3 µL; Sigma-Aldrich, USA) was mixed in LSB (3 µL), heated to 90℃ for 5 min, and run on each gel as a control. The separated proteins were then electrophoretically transferred to a polyvinylidene difluoride (PVDF) membrane (Sigma-Aldrich, USA) in a tank system with plate electrodes. The membranes were blocked for 1 h at 25℃ in 5% non-fat dried milk in TBS-Tween, and then incubated in room temperature for 1 h with mouse monoclonal anti-MMP-2 antibody (1:1,000 dilution in TBS-Tween; Sigma-Aldrich, USA). After incubating, the membranes were washed four times with 0.1% TBS-Tween, and then incubated for 1 h at room temperature with a horseradish peroxide-conjugated secondary antibody (Amersham, UK) diluted 1 : 10,000 in TBS-Tween with 0.1% Tween. After four washes with TBS-Tween, the immunoreactive bands were visualized with Pico chemiluminescent substrate (Pierce Biotechnology, USA) and the blots were exposed to radiographic film (Pierce Biotechnology, USA).

Histochemical staining to visualize TRAP expression was performed on 15 µm frozen sections of the CCL and synovium specimens as described by Muir et al. [17]. All reagents for histochemical staining were obtained from Sigma-Aldrich (USA). A solution of naphthol AS-BI phosphate was prepared by dissolving 25 mg of naphthol AS-BI phosphate in 2.5 mL of n,n-dimethylformamide to which 45 mL of 0.05 M Tris-maleate buffer (pH 5) was added. A solution of hexazotized pararosanaline was prepared by dissolving 0.25 g of pararosaniline hydrochloride in 5 mL of distilled water and then adding 1.25 mL of hydrochloric acid. This solution was mixed with an equal volume of 4% sodium nitrite immediately before use. The final reaction mixture for histochemical staining was prepared by adding 4 mL of the hexazotized pararosanaline solution to the naphthol AS-BI phosphate solution along with 50 mM sodium-potassium tartrate. The final reaction mixture was filtered using syringe filters (Nalge, USA) before use. The 15 µm thick sections were incubated in the reaction mixture at 37℃ for 1~2 h, rinsed in distilled water, counterstained with Mayer's hematoxylin, and mounted. For each batch of slides, a negative control was prepared by omitting the naphthol AS-BI phosphate. The CCL and synovium specimens were examined by light microscopy for presence of the cells that contained TRAP.

The Western blots were scanned using a digital imaging system (Fusion FX7; Vilber Lourmat, Germany) and the relative intensity of the bands was determined by densitometry (Bio-Rad Laboratories, USA). The data obtained in the present study were analyzed using an ANOVA and Student's t-test. A p value < 0.05 was considered to be statistically significant. The data are presented as the mean ± SD.

The level of TRAP in the SF in OA-induced stifles increased significantly (p < 0.05) 3 months postoperatively in comparison to that of the pre-surgical control, post-surgical contralateral control and sham-operated stifles. This elevated level was sustained throughout the experimental period (Fig. 1). The serum level of TRAP was found to be increased 3 months postoperatively and was significantly increased (p < 0.05) 6 months after the induction of OA (Fig. 2).

The level of TIMP-2 in the SF collected from OA-induced stifles was significantly decreased (p < 0.05) 3 months postoperatively compared to the pre-surgical control, post-surgical contralateral control, and sham-operated stifles (Fig. 3). The serum level of TIMP-2 was decreases 3 months postoperatively and significantly decreased (p < 0.05) 6 months after the induction of OA (Fig. 4).

The level of MMP-2 in the SF was found to increase significantly (p < 0.05) on 3 months after induction of OA and the levels observed on the subsequent experimental period were significantly (p > 0.05) above the control value (Fig. 5).

Histochemical staining specific for TRAP revealed the presence of TRAP-positive cells in the synovium, CCL, and cartilage 3 months postoperatively in the OA-induced stifles but not in the contralateral control or sham-operated stifles (Figs. 6, 7, 8). The synovium, CCL, and cartilage collected from the index stifles after 6 and 12 months of OA induction showed markedly increased numbers of TRAP-positive cells. Moderate synovial inflammation is also seen as mononuclear cell infiltration into the synovium. However, the contralateral control and shamoperated stifles also showed a mild increase in the number of TRAP-positive cells in the synovium and CCL tissues during the later stage of OA.