Inhibition of Leukotriene A 4 Hydrolase Suppressed Cartilage Degradation and Synovial Inflammation in a Mouse Model of Experimental Osteoarthritis

Study Approval

Animal studies were approved by the Institutional Animal Care and Use Committees and Institutional Biosafety Committees of Central South University, China (CSU; 2013-05). Six-week-old male C57BL/6 mice were purchased from the Animal Centre of Central South University (Changsha, Hunan, China). Collection of human OA tissues from discarded knee replacement surgeries was approved by the Queensland University of Technology (QUT) and the St Vincent Private Hospital Ethics Committees (Human ethics number: #1400001024). Informed consent was obtained from all human subjects.

Human Cartilage Specimen Preparation

Human cartilage was obtained from patients with OA (Kellgren-Lawrence grade 4) undergoing total knee replacement. Following our previous publication, the cartilage tissues collected from the relatively smooth region were classified into the grade 1 (G1, non-OA) group, and the cartilage collected from the severely damaged region were classified into the grade 4 cartilage (G4, OA) group. ²⁴ Osteochondral plugs (including cartilage and subchondral bone) were taken from the tibial plateau region. All samples were prepared for histology and immunohistochemical analysis by washing in phosphate-buffered saline (PBS) and immediately fixed in 4% paraformaldehyde. The samples were then decalcified in 10% ethylenediaminetetraacetic acid (EDTA, pH = 7.2–7.4). The specimens were embedded in paraffin, and 5-µm-thick sections were obtained as described in our previous study. ²⁵

Cell Culture and Bestatin Treatment In Vitro

Human chondrocyte cell line C28I2 was seeded into 6-well plates at a density of 1 × 10⁶ cells/well and cultured in DMEM containing 10% fetal bovine serum (FBS) and 1% of penicillin and streptomycin (PS). At 90% confluence, cells were serum-starved for 24 hours and then were incubated in the presence or absence of 10 ng/ml interleukin-1β (IL-1β) or 50 μM bestatin with culture medium for 24 hours. Bestatin is a dipeptide immunomodulator that was discovered in the culture supernatant fraction from Streptomyces olivoreticuli, and it is a well-characterized inhibitor of LTA₄H to attenuate LTB₄ synthesis in many studies and particularly in animal models.^{26 -28}

Cell Viability Test

Human chondrocyte cell line C28I2 was seeded into 96-well plates at a density of 5000 cells/well with culture medium. After 24 hours, cells were serum-starved for 24 hours and then were incubated in 0, 10, 50, 100, and 200 μM of bestatin separately. Cell viability test and cell count were performed according to our previous publication using CyQUANT proliferation assay (Life Technologies, Thermo Fisher). ²⁹ Briefly, 100 μl of lysis buffer containing the CyQUANT GR dye was added to each well and incubated at 37 °C for 30 minutes. Fluorescence was measured using a fluorescence microplate reader with an excitation of 485 nm and emission of 530 nm.

Preparation of Bestatin

Bestatin was purchased from Cayman Chemical (Michigan, USA), and 2-hydroxypropyl-β-cyclodextrin (HPCD) and polyethylene glycol 400 (PEG-400) were obtained from Sigma Aldrich (Missouri, USA). The oral formulation for bestatin was prepared as previously described. ¹³ Briefly, 100 µl PEG-400 and 100 mg of bestatin were gently mixed before dissolving in a 20% (w/v) aqueous solution of HPCD. Then, the solution was mixed at room temperature to obtain a homogeneous mixture at 2 mg/ml concentration.

Induction of OA in Mice and Bestatin Treatment

Mice were group-housed (5 animals/cage) in ventilated cages in a temperature-controlled room on 12-hour light/dark cycles with ad libitum access to standard mouse food and water and routine veterinary assessment. Six-week-old male C57BL/6 mice were purchased following our previous publications.^4,25 After that, all the animals were fed with the standard diet from the Animal Center (control diet) for at least 2 weeks as an observation period. Eight-week-old C57BL/6 mice (similar to other studies,^{30 -32} while 8-week-old of mice were considered as around 18-year-old in human age) ³³ were randomly assigned and underwent destabilization of the medial meniscus (DMM) surgery to induce OA as previously described. ²⁵ Briefly, OA was induced by dissecting the medial meniscus ligament to destabilize the medial meniscus in the knee joint of the hind limb on the right side. Mice in the sham group were surgically treated in which the ligament was visualized but not dissected. One week after experimental OA induction, mice were treated orally with bestatin ³⁴ or vehicle (PEG-400 and HPCD). All experimental mice were then received twice-weekly oral bestatin after DMM surgery at a dose of 5 mg/kg for a total of 8 weeks. Bestatin (5 mg/kg) significantly decreased serum levels of LTB₄ and BLT1 production in a mouse model and showed no difference to high dose of bestatin (15 mg/kg). ²⁷ Mice were sacrificed 8 weeks after surgery for histological analysis. Bestatin is well absorbed in the gastrointestinal tract and is widely distributed in the body, including the kidney, liver, lung, spleen, and heart. It is metabolized in the liver, and its active metabolites are further metabolized and excreted in the urine.^35,36 Overall, bestatin is rapidly absorbed and metabolized and is safe for in vivo experiments.

Histologic Evaluation of Knee Joint in Mice

After euthanasia, the knee joints of hind limbs were fixed in 4% paraformaldehyde, decalcified in 10% EDTA, embedded in paraffin, and sectioned for histological and immunohistochemical analysis. Sections were stained with Safranin-O (Saf-O)/Fast green to evaluate cartilage degradation and synovitis. The severity of OA was assessed in the medial compartment of the knee using a modified Mankin scoring system (0–14) as previously described. ²⁵ The degree of synovitis was scored using a 0 to 6 scoring system that measured the thickness of the synovial lining cell layer on a scale of 0 to 3 (0 = 1–2 cells, 1 = 2–4 cells, 2 = 4–9 cells, and 3 = 10 or more cells) and cellular density in the synovial stroma on a scale of 0 to 3 (0 = normal cellularity, 1 = slightly increased cellularity, 2 = moderately increased cellularity, and 3 = greatly increased cellularity) as previously described. ²⁵ Both the Mankin scoring and synovial inflammation scoring were carried out by 2 independent blinded readers (Indira and Antonia). The sum of both parameters was used for analysis. Synovitis scores obtained from all 4 quadrants (lateral femur, lateral tibia, medial femur, and medial tibia).

Immunohistochemistry

Immunohistochemistry was performed according to our previously published methods.^37,38 Briefly, after dewaxing, sections were incubated with proteinase K for 20 minutes for antigen retrieval. The endogenous peroxidase activity was eliminated with 3% H₂O₂ in PBS for 20 minutes. Sections were then eliminated non-specific protein binding by incubation with 1% Bovine serum albumin for 1 hour at room temperature. Sections were incubated overnight at 4 °C with antibodies against LTA₄H (Abcam, Melbourne, VIC, Australia; dilution 1:250) and LTB₄ receptor (BLT1, LSBio, Sapphire Bioscience, Redfern, NSW, Australia; dilution 1:50) followed by incubation with anti-rabbit or anti-rat secondary antibody (1:5000 dilution; LI-COR Biosciences, Nebraska, USA) and visualization with 3,3-diaminobenzidine chromogen. To exclude antigen-independent staining, negative controls for which the incubation step with the primary antibody was omitted were examined. The percentage of positive cells was determined by counting the number of immunostained cells and dividing by the total number of cells (100 cells per sample) visualized by a hematoxylin counterstain using ImageJ (National Institute of Health, Bethesda, BA, USA).

Quantitative Real-Time PCR

Total RNA was extracted with TRIzol, and cDNA was synthesized using the SensiFAST cDNA Synthesis Kit according to the manufacturer’s protocol. SYBR Green detection chemistry was used on the QuantStudio Real-Time PCR system (Applied Biosystems, Thermo Scientific, Scoresby, VIC, Australia). Quantitative measurements of all primers used in this study were determined using (2^{−ΔΔ Ct}) method, and GAPDH expression was used as the internal control,^27,39 as described previously by our group.^{37,40 -42}

Statistical Analysis and Sample Size Justification

All experiments were performed in triplicate using different samples: 7 donors of human knee cartilage for histologic evaluation; 5 donors for immunohistochemistry; and 6 mice per group were used for animal study. Statistical difference of the results was tested using the unpaired Student’s t-test comparison of 2 variables. One-way analysis of variance followed by post hoc test was used for comparison of more than 2 variables. All analyses were performed using GraphPad Prism 7, and P < 0.05 is significant. All data are presented as mean ± SD. Sample size for each experiment was determined using G power program (https://clincalc.com/stats/samplesize.aspx). Eighty percent power and an alpha level of 0.05 were used for the calculation. Given concerns about the ability to complete the experiments, we increased the sample sizes in the experiments above.

Increased LTA₄H and BLT1 in OA Cartilage

To study the leukotriene A₄ pathway in the articular cartilage, we first divided the cohort of OA patients into 2 groups based on a histomorphometry analysis for the semi-quantitative estimation of OA extent on Safranin-O-stained cartilage sections. As exemplarily shown in Fig. 1A for intact smooth cartilage surface (grade 1, G1) and severely degenerated and fibrillated OA regions (grade 4, G4), significant cartilage damage associated with elevations of Mankin score (12.14 ± 1.574, n = 7) were observed in G4 OA sites; compared with G1 group (Mankin score = 3.286 ± 0.951; n = 7) ( Fig. 1B ).

OPEN IN VIEWER

Figure 1.

Increased leukotriene A₄ hydrolase (LTA₄H) and leukotriene B4 receptor 1 (BLT1) in osteoarthritis cartilage. (A) Representative Safranin-O (Saf-O) and fast green stained sections of G1 or G4 knee regions. Similar sections were stained with LTA₄H and BLT1. Scale bars, 200 μm. (B) Severity of articular cartilage degradation was graded using Mankin scoring system. Graphs represent mean ± SD (n = 7). The percentage of LTA₄H (C) and BLT1 (D) positive cells per knee section were counted. Graphs represent mean ± SD (n = 5). (E) Sham and DMM mice knees were stained with LTA₄H and BLT1 (n = 6). Positive cells per knee section were counted (F). Scale bars, 20 μm. Graphs represent as mean ± SD. DMM = the surgical destabilization of the medial meniscus. P < 0.05 is considered significant.

We then examined LTA₄H and BLT1 expression in OA cartilage of humans and mice DMM OA model. The percentage of LTA₄H immunostained chondrocytes was markedly elevated in OA-affected G4 cartilage compared with undamaged G1 regions of patient-matched OA cartilages ( Fig. 1A and C ). BLT1 levels were also significantly increased in severe G4 versus mild G1 cartilage ( Fig. 1A and D ). To further ascertain whether leukotriene plays a role in OA pathogenesis, we examined LTA₄H and BLT1 expression in OA cartilage of mice. Similar results were found in experimental OA mice, in which LTA₄H and BLT1 were significantly increased in knee cartilage from DMM mice compared to Sham surgery mice ( Fig. 1E and F ). These results indicated that the expression of LTA₄H and BLT1 is in accordance with the disease severity.

Inhibition of LTA₄H Level Suppressed Articular Cartilage Degradation In Vivo

The articular cartilage of mice with DMM-induced OA exhibited pathological OA changes characterized by Safranin-O loss and detectable cartilage roughness ( Fig. 2A ), with a Modified Mankin score of 11.833 ± 1.169 ( Fig. 2B ). In contrast, the knee joint of sham mice exhibited no visible changes on the articular cartilages with a significant lower Mankin score (0.857 ± 0.690) than that in mice with DMM-induced OA ( Fig. 2A and B ). Bestatin is a well-characterized inhibitor of LTA₄H and was shown to attenuate LTB₄ synthesis in several previous studies.^27,43 Inhibition of LTA₄H level in mice with DMM-induced OA suppressed OA progression with a lower Mankin score (8.000 ± 1.095) compared with DMM-induced OA mice with vehicle ( Fig. 2A and B ). These results suggested that increased leukotriene may be one of the reasons causing OA as inhibition of which could suppress OA progress in DMM mice.

OPEN IN VIEWER

Figure 2.

Inhibition of LTA₄H level suppressed articular cartilage degradation in a mouse model of experimental osteoarthritis. (A) Representative Saf-O and fast green stained sagittal sections of sham and DMM mice knee with/without LTA₄H inhibitor. Scale bars in the top panel, 100 μm; bottom panel, 20 μm. (B) Severity of articular cartilage degradation was graded using Mankin scoring system. Graphs represent mean ± SD (n = 6). LTA₄H = leukotriene A₄ hydrolase; DMM = the surgical destabilization of the medial meniscus. P < 0.05 is considered significant.

Inhibition of LTA₄H Level Reduced Synovitis in OA Mice

Mice with DMM-induced OA also revealed an increased influx of synoviocytes and lining layer thickening in synovium compared with the respective control mice ( Fig. 3A ). The synovitis scores reiterated these observations with a significantly higher score in DMM-induced OA mice (Synovitis score: 4.417 ± 0.540 compared with 1.208 ± 0.485) ( Fig. 3B ). The role of leukotriene in synovitis in mice was examined by oral administration of LTA₄H inhibitor-bestatin. Bestatin treatment reduced the thickness of the synovial lining cell layer and cellular density in the synovial stroma in mice with DMM (Synovitis score: 2.917 ± 0.585) compared with vehicle-treated mice with DMM (Synovitis score: 4.417 ± 0.540) ( Fig. 3A and B ). However, oral administration of bestatin had no significant effect on synovitis in mice without DMM ( Fig. 3A and B ). These results indicated that bestatin treatment also reduced synovitis.

OPEN IN VIEWER

Figure 3.

Inhibition of LTA₄H level reduced synovitis in osteoarthritis mice. (A) Representative Saf-O and fast green stained sagittal sections of sham and DMM mice knee with/without LTA₄H inhibitor. Scale bars, 100 μm. (B) Severity of articular synovitis inflammation was graded using the Synovitis scoring system. Graphs represent mean ± SD (n = 6). LTA₄H = leukotriene A₄ hydrolase; DMM = the surgical destabilization of the medial meniscus. P < 0.05 is considered significant.

In Vitro Inhibition of LTA₄H Significantly Affect Gene Expression Profile in Chondrocytes

We further tested the effect of LTA₄H inhibitor on C28I2 chondrocytes cell line in vitro. To test whether bestatin plays a direct role in chondrocytes, we validated the potential effect of bestatin by gene expression profile analysis. We first tested cell viability under different concentrations of bestatin ( Fig. 4A and B ). Our results indicated that 0 to 200 μM of bestatin treatment showed no toxic effect on chondrocytes for 24 hours. Next, we tested the effects of 50 μM bestatin treatment on chondrocytes phenotype. Bestatin treatment significantly upregulated COL2A1 and inhibited SOX9 expression in chondrocytes in the absence of IL-1β ( Fig. 4D and F ). No significant effects of bestatin on the expression of aggrecan (ACAN), Runt-related transcription factor 2 (RUNX2), a disintegrin and metalloproteinase with thrombospondin motifs 5 (ADAMTS5) and MMP13 were observed (Fig. 4C, E, G, and H). To establish an in vitro OA model, the cells were stimulated with IL-1β (10 ng/ml). The expression of RUNX2 gene caused by IL-1β in chondrocytes was significantly inhibited by bestatin ( Fig. 4E ). The gene expression of ACAN, COL2A1, and SOX9 was robustly upregulated by bestatin, but there were no significant changes in the cartilage degradation marker, ADAMTS5 ( Fig. 4G ). In response to bestatin treatment, MMP13 showed a decreased gene expression; however, this effect was not significant ( Fig. 4H ).

OPEN IN VIEWER

Figure 4.

In vitro inhibition of LTA₄H significantly affect gene expression profile in chondrocytes. Cell viability (A), cell counts (B), and qualitative polymerase chain reaction analysis of ACAN (C), COL2A1 (D), RUNX2 (E), SOX9 (F), ADAMTS5 (G), and MMP13 (H) messenger RNA in C28I2 chondrocytes cell line with/without LTA₄H inhibitor under normal and inflammation condition (interleukin-1β treatment). Data represented as mean ± SD (n = 6). NS = not significant; LTA₄H = leukotriene A₄ hydrolase; ACAN = aggrecan; COL2A1 = collagen 2A1; RUNX2 = Runt-related transcription factor 2; SOX9 = SRY-Box transcription factor 9; ADAMTS5 = a disintegrin and metalloproteinase with thrombospondin motifs 5. P < 0.05 is considered significant.