Animal preparation and surgical procedures
This study was approved by the animal research committee of Kyoto University (approval number: Med Kyo 19016) and conducted in accordance with the ARRIVE guidelines [47].
Sixty-four male Wistar rats (12 weeks old), purchased from SHIMIZU Laboratory Supplies Co. Ltd. (Kyoto, Japan), were placed in a plastic cage with paper bedding on a 12-h light/dark cycle at a constant temperature. The rats could move freely in the cages and had free access to food and water. All rats underwent DMM surgery for the OA model [46]. Briefly, DMM surgery was performed with anteromedial capsule incision and transection of the medial meniscotibial ligament in the right knee under anesthesia with 1.0 mL/kg pentobarbital sodium (Somnopentyl; Kyoritsu Seiyaku Corp., Tokyo, Japan). The rats were randomly separated into TBP1901 and saline solution (control) injection groups (n = 32 for each group). They were sacrificed at four periods of 1, 2, 6, or 10 weeks postoperatively (n = 8 for each). The synthesis of TBP1901 was performed as described previously [42]. TBP1901 and saline injections of 50 µL were administered to the right knee joints through the patellar tendon. The concentration of TBP1901 was chosen to be 30 mg/mL based on a preliminary study. The rats sacrificed at 1 week postoperatively were injected at day-3 and day-7 postoperatively, and at 2 weeks were injected at day-3, day-7, day-10, and day-14, and at 6 and 10 weeks were injected once a week from 1 week postoperatively. An hour after the final injection in each observation period, the rats were sacrificed by a lethal dose of pentobarbital sodium, and their right knee joints were harvested for micro-computed tomography (micro-CT), histochemical, and immunohistochemical analysis. Bodyweight was measured every week for all rats.
Micro-CT analysis
After fixation of the knees with 4% paraformaldehyde overnight, the tibiofemoral (TF) joints were scanned using a micro-CT system (SMX-100CT, Shimadzu, Kyoto, Japan) at 43 kV and 43 µA with a scan time of approximately 10 minutes. After scanning, the three-dimensional (3D) reconstruction and assessment of the TF joint were performed with ImageJ (National Institutes of Health, Bethesda, MD, USA) and Amira® (version 5.5, EFI Visualization Science Group, Burlington, MA, USA) software. The subchondral bone (SB) plate thickness, number of SB plate perforations, diameter of SB plate perforations, bone volume (BV), total volume (TV), and osteophyte volume in the medial tibia SB (above epiphyseal plate) were calculated according to protocols described in previous studies [48, 49]. SB plate thickness was measured at the thickest region around the center of the medial articular surface. The diameter of the SB plate perforation was defined as the largest diameter of the largest perforation in each sample. To assess the bone density of the SB, bone volume was divided by total volume (BV/TV).
Histological analysis
After micro-CT, the knee joint samples were decalcified in 10% ethylene-diamine-tetra-acetic acid (EDTA) for three to four weeks and cut along the mid-sagittal plane at the halfway point. After decalcification, the samples were paraffin-embedded and cut into 6 µm sections at 50 µm intervals. First, the paraffin sections were observed with an epifluorescence microscope (MVX10, Olympus Corporation, Tokyo, Japan) to confirm the presence of curcumin in the articular cartilage and the synovium. Curcumin is a fluorescent substance with an excitation wavelength of 300–550 nm (maximum excitation wavelength: 467 nm) and an emission wavelength of 548–600 nm (maximum emission wavelength: 571 nm) [50]. Fluorescence detection was used for curcumin quantification [51]. Based on this report, curcumin fluorescence was observed with the mirror unit of U-MYFPHQ/XL (Olympus Corporation, Tokyo, Japan). Second, the paraffin sections were stained with hematoxylin and eosin (H/E) to evaluate the inflammation of the synovium in knee joints and Safranin O/Fast green to evaluate the severity of the cartilage degeneration and the SB changes. Inflammation was assessed using an inflammation scoring system described in a previous study [52]. Three membrane features (synovial lining cell layer, stroma cell density, and inflammatory infiltrate) were assessed in the whole knee joint as 0 (none), 1 (slight), 2 (moderate), or 3 (strong). The inflammation score was determined as the summed score for all parameters. The highest inflammation score was recorded for each sample. Cartilage degeneration was assessed using the OARSI score [53] and the modified Mankin (MM) score [54], according to a previous study. The OARSI score consists of six grades and four stages on a scale from 0 (intact) to 24 (severe damage). The MM score consists of three features (pericellular matrix staining, spatial arrangement of chondrocytes, and inter-territorial matrix staining) on a scale from 0 (intact) to 8 (severe). Cartilage degeneration was evaluated on the medial TF joint and patellofemoral (PF) joint. The maximum score was used for all scoring systems and samples.
Immunohistochemical analysis
Immunohistochemistry of type I collagen, type II collagen, Interleukin-1β (IL1β), Interleukin-6 (IL6), and tumor necrosis factor-α (TNFα) was performed to determine the principal collagen expression and inflammation in the cartilage. Antigen retrieval was performed for 20 minutes by heating with HistoVT One (Nacalai Tesque, Inc., Kyoto, Japan). Blocking was performed using 0.3% H2O2 for 15 minutes and 5% goat serum for 20 minutes at room temperature. The sections were then incubated at 4 ºC overnight with primary antibodies against type I collagen (AB755P, diluted 1:1000; Merck KGaA, Darmstadt, Germany), type II collagen (diluted 1:100; KYOWA PHARMA CHEMICAL CO., LTD., Toyama, Japan), IL1β (ab9722, diluted 1:100; Abcam Co., Cambridge, UK)), IL6 (ab9324, diluted 1:250; Abcam Co., Cambridge, UK)), and TNFα (ab6671, diluted 1:100; Abcam Co., Cambridge, UK). Detection was performed using the streptavidin-biotin-peroxidase complex technique with an Elite ABC kit (Vector Laboratories, CA, USA), and immunoreactivity was visualized by incubation with a diaminobenzidine solution (Vector Laboratories, CA, USA) followed by counterstaining with hematoxylin. The primary antibody was omitted from the negative control slides. The expression of type I and type II collagen in the medial tibia and patella cartilage was analyzed by measuring the minimum and mean pixel intensity values using TIFF images (magnification × 40) taken by microscopy and ImageJ, respectively, on a scale of 0 (maximum staining) to 255 (no staining), according to a previous study [55]. The expression of IL1β, IL6, and TNFα in the medial tibia and patella cartilage was analyzed by measuring the percentage of positive chondrocytes to detect the severity of cartilage inflammation within the middle region of the medial tibia with an anteroposterior width of 0.5 mm or the central region of the patella with a superoinferior width of 0.5 mm.
Statistical analyses
The median and interquartile range (IQR) of the inflammation score, OARSI score, and MM score (non-parametric data) were calculated for each group, and the Wilcoxon test was used to compare differences between the two groups. In addition, the mean and 95% confidence intervals (CIs) of body weight, immunohistochemical, and micro-CT analysis data (parametric data) were calculated, and Welch’s T-test was used to compare differences between the groups. The normality of the data was assessed using normal quantile–quantile plots and the Shapiro-Wilk test. P-values < 0.05 were considered statistically significant in all tests. Statistical analyses were performed with JMP® Pro 14 software (SAS Institute Inc., Cary, NC, USA).