A Novel Synthesized Sulfonamido-Based Gallate-JEZTC Blocks Cartilage Degradation on Rabbit Model of Osteoarthritis: An in Vitro and in Vivo Study

Osteoarthritis (OA) is the most common joint disorder which would result in pain, disability and loss the quality of life [1]. It is reported that obesity, bone mass, joint injury and instability, developmental diseases, trauma, joint deformity and age are common factors in OA, especially in the hip and knee joints [2]. Follow by injury is the activation of catabolic factors such as pro-inflammatory cytokines including interleukin-1 (IL-1) [3] and tumor necrosis factor α (TNF-α) [4], which are contributed to the progression of OA. Chondrocytes exposed to IL-1β and TNF-α leaded to release of matrix metalloproteinases (MMPs), the main matrix-degrading enzymes, resulting in cartilage degeneration [5, 6]. A growing body of evidences showed that MMPs like MMP-1, MMP-3, and MMP-13 plays an important role in OA based on their ability to cleave components of the extracellular matrix (ECM) in cartilage [7, 8]. During the process of OA, various catabolic pathways have also been shown to mediate the expression of MMPs, including mitogen-activated protein kinases (MAPKs) [9], nuclear factor kappa B (NF-κB) [10] and phosphatidylinositol 3-kinase (PI3K) / Protein kinase B (AKT) signaling pathways [11].

Currently, the treatment of OA is still limited to steroidal and non-steroidal anti-inflammatory drugs (NSAIDS), which failed to block the progression of OA [12, 13]. These drugs merely provide symptomatic relief for pain and inflammation. Besides, they have severe side effects such as gastrointestinal bleeding and cardiovascular diseases [14, 15]. As an alternative, plant-derived agents that exhibit minimum side effects and are available in cost effective manner have received considerable attention in the treatment of OA [16, 17].

Gallic acid (GA) and its derivatives are a group of polyphenol compounds that have been known with strong anti-oxidant [18] and anti-inflammatory [19, 20] properties. GA can suppress the expressions of pro-inflammatory cytokines and chemokines [21-23]. The expression of matrix metalloproteinases (MMPs) was strongly suppressed by GA [24]. GA was also found to be involved in the modulation of several important pharmacological and biochemical pathways including MAPKs and NF-κB [25, 26], which were important in pathogenesis of OA. Yoon et al, reported that GA has pro-apoptotic and anti-inflammatory effects on fibroblast-like synoviocytes (FLS) from patients with rheumatoid arthritis [27]. These findings suggested GA was potential agent in the treatment of OA. However, GA was reported to suppress cell proliferation [28], which may attenuate its chondro-protective effect. Besides, GA showed much weaker antioxidant effects than its esters in cell systems due to its hydrophily [18].

Recently, we reported a new series of derivatives of GA that synthesized by coupling with sulfanmides, including sulfathiazole sodium, SMZ, sulfadimidine and sulfachloropyrazine sodium [29-33]. These compounds were found to be effective in promoting proliferation and maintaining the phenotype of chondrocytes in vitro. Among the novel analogs, a very promising compound namely3, 4, 5-trihydroxy-N-{4-[(5-methylisoxazol-3-yl) sulfamoyl] phenyl} benzamide (JEZTC), synthesized by GA and SMZ was demonstrated to be effective in chondroprotection in vitro [30]. Pilot study also showed that it has relatively stronger anti-inflammatory potential than others. It was hypothesized that JEZTC may hold promise in the treatment of OA by exerting both chondroprotive and anti-inflammatory effects. Thus, further investigation of its effect on OA and underlying mechanism is urgently needed.

In this study, we investigated the anti-inflammatory effect of JEZTC in vitro followed by the effects on the treatment of OA. For this purpose, we established an OA model in the knee joint of rabbit, and further exploration the underlying mechanism. Findings of this study are anticipated to provide new insights for the therapy of OA.

JEZTC was prepared from GA and SMZ according to previous study [30]. Briefly, appropriate distilled water was added to the mixture after reactions, and then the raw product precipitated was separated by vacuum filtration. The resultant product was recrystallized in a THF-methanol solvent system for purification.

JEZTC and GA were separately dissolved in sodium hydroxide solution (NaOH, Sigma, USA) at concentrations of 10 mg/ml as stock solutions and stored at - 4°C for further experiments.

Articular chondrocytes were harvested from knee joint cartilage of 1-week-old New Zealand rabbits by enzymatic digestion. In brief, cartilage slices from two rabbits were digested with 0.25% trypsin (Solarbio, China) for 30 min and then with 2mg/ml collagenase type II (Gibco, USA) in alpha-modified Eagle’s medium (α-MEM, Gibco, USA) for 3 h. After centrifugation, the chondrocytes were re-suspended and cultured with α-MEM containing 20% (v/v) fetal bovine serum (FBS) (Gibco, USA) and 1% (v/v) penicillin/streptomycin (Solarbio, China) in a 5% CO₂ humidified incubator at 37°C with the culture medium replaced every other day. Articular chondrocytes at passage two were used for further studies.

To investigate the effects of JEZTC and GA on IL-1β-induced chondrocytes, the cells were grouped as follow: (1) normal control group, chondrocytes treated with culture medium only; (2) IL-1β (OA model) group, chondrocytes treated with culture medium contained 10 ng/ml IL-1β (Gibco, USA); (3) JEZTC-1 group, chondrocytes pre-incubated with 2.344 µg/ml JEZTC for 1 h followed by stimulation with IL-1β for 24 h; (4) JEZTC-2 group, chondrocytes pre-incubated with 4.688 µg/ml JEZTC for 1 h followed by stimulation with IL-1β for 24 h; (5) JEZTC-3 group, chondrocytes pre-incubated with 9.376 µg/ml JEZTC for 1 h followed by stimulation with IL-1β for 24 h; (6) GA group, chondrocytes pre-incubated with 4.688 µg/ml GA for 1 h followed by stimulation with IL-1β for 24 h. The concentrations of JEZTC were determined according to previous study [26]. The dose of GA was derived from the preliminary cytotoxicity assay, which was optimal for cell growth.

A total of twenty-four 12 weeks old male New Zealand white rabbits (Animal Centre of Guangxi Medical University, Guangxi, China) weighing 2.0-2.5 kg were used in this study. All experiments were conducted with the approval of the Guangxi Medical University Animal Care and Use Committee (2013-12-3). Sixteen rabbits underwent bilateral anterior cruciate ligament transection (ACLT) on the knee joints to induce OA were randomly assigned to two groups. The other eight rabbits underwent sham operations (normal group), of which the articular cavity was opened and sutured without cutting the short anterior cruciate ligament. At 8 weeks after surgery, the JEZTC group received intra-articular injections of 0.5 ml JEZTC (4.688µg/ml, the dose exhibiting the best performance in the in vitro study) once per week for 8 weeks. The PBS (OA model) group was injected with 0.5 ml of PBS alone in both knees under the same conditions. In the sham-operation group, no other procedures were conducted.

The mRNA expression of OA and apoptosis-specific gene was detected by qRT-PCR. For the in vitro study, cells were washed once with PBS and the total RNA was extracted using an RNA isolation kit (Tiangen Biotechnology; Beijing, China) according to the manufacturer’s instructions. For the in vivo investigation, cartilage obtained at 4 and 8 weeks post-treatment was pulverized in liquid nitrogen before RNA extraction. Approximately 300 ng of total RNA was used as a template and reverse transcribed into cDNA using a reverse transcription kit (Fermentas Company, USA). The qRT-PCR reactions were performed using a Quantitative PCR Detection System (Realplex 4, Eppendorf Corporation, USA) with FastStart Universal SYBR Green Master (Mix, Roche company, Germany) and at the conditions of 10 min at 95°C, 15 s at 95°C and 1 min at 60°C. The primers used for PCR are shown in Table 1. Each gene was analyzed in triplicate to diminish operation errors. The relative gene expression levels were calculated by using the 2^-ΔΔCt method and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) served as the internal reference.

After removal the soft tissue, macroscopic evaluation was performed by two observers (ZL, CL) at 4 and 8 weeks after treatment. The depth of erosion in articular cartilage were performed according to the criteria described by Pelletier et al. [34] using a four-grade scale (0 = a normal-appearing surface, 1= minimal fibrillation or a slight yellowish discoloration of the surface, 2 = erosion extending into the superficial or middle layers only, 3 = erosion extending into the deep layers, and 4= erosion extending to the subchondral bone). A higher score indicates greater cartilage damage.

The production of reactive oxygen species (ROS) was assayed with an ROS detection reagent kit (Nanjing Jiancheng Bioengineering Research Institute, China) according to the manufacturer’s instructions. Chondrocytes were treated with JEZTC or GA for 1h followed by IL-1β-stimulation for 24h, then washed twice with PBS and incubated with dichlorodihydrofluorescein diacetate (DCFDA, 10 μM) in PBS in the dark for 60 min. After washing with PBS, the fluorescence was measured immediately using a fluorescence microplate reader (FLx800, Biotek, USA) with excitation and emission at 485 and 525 nm, respectively. At 4 and 8 weeks post-treatment, cartilage homogenate was used to detect ROS generation under frozen conditions. After centrifugation, the supernatant was incubated with DCFDA, and the experiments were performed as previously described.

After macroscopic evaluation, the cartilage tissues were fixed in 4% paraformaldehyde and decalcified by buffered ethylenediaminetetraacetate (EDTA). Samples were embedded in paraffin and then cut into 3-μm sections. Paraffin sections were de-waxed and stained with hematoxylin-eosin (HE, JianCheng Biotech, China) and Masson (Fuzhou, Fujian, China) for histological examination and safranin O (Sigma, USA) for glycosaminoglycan (GAG) secretion orientation. Sections from each animal were randomized for histological evaluation to eliminate bias. Each section was scored separately by 3 independent researchers (ZL, HW and XL). The severity of the OA was graded using the histologic criteria (Table 2) according to Mankin et al, [35] based on structural changes (score range 0-6), cellular changes (score range 0-3), loss of safranin O staining (score range 0-4) and tidemark integrity (scored rang 0-1).

The evaluation of MMP-1 secretion was performed by an immunohistochemical staining kit (Bioss, China). Cells and de-waxed sections were washed in PBS and exposed to 3% H₂O₂ for 15 min at room temperature to quench endogenous peroxidase activity. Then the cells were blocked with normal goat serum for 20 min at room temperature. After incubation overnight at 4°C with the primary antibody (1: 200), the cells were incubated with the second antibody and biotin-labeled horseradish peroxidase. Subsequently, the antibody binding was visualized with a 3, 3’-diaminobenzidine tetrahydrochloride (DAB) kit (Boster, China) before brief counterstaining with hematoxylin. Eventually, cells were gradually dehydrated, sealed with neutral gum, observed and photographed with an inverted phase contrast microscope (Zeiss Corporation, Germany).

To prepare total protein from tissues and cells, cartilage obtained at 4 or 8 weeks after treatment in vivo was pulverized in liquid nitrogen and chondrocytes treated with JEZTC or GA followed by induction with IL-1β were washed with cool PBS, then lysis buffer (Sigma-Aldrich, USA) containing a protease inhibitor mixture was used for protein extraction. Equal amounts of protein (60 µg) were separated by SDS-polyacrylamide gel (Sigma, USA), and then transferred onto polyvinylidene fluoride (PVDF) membranes (Millipore, Billerica, MA, USA). PVDF membranes were blocked with 5% nonfat milk for 1h, and then incubated with primary antibodies (Abcam, UK) of P38(1: 1000), p-P38 (1: 800), Akt (1: 500), p-Akt (1: 1000), NF-κB (1: 500) overnight at 4°C. The membranes were then incubated with the secondary antibody (Invitrogen, USA) and visualized using the Odyssey Infrared Imaging System (LI-COR, USA) according to the manufacturer’s instructions. Data were normalized by housekeeping protein (GAPDH).

Data were presented as the means ± SD. Significant differences were determined by one-way analysis of variance (ANOVA) followed by LSD post hoc test. The level of significance was set at P< 0.05.

Effects of JEZTC and GA on the expression of MMPs and TIMP-1 in IL-1β-induced chondrocytes. To investigate the effects of JEZTC and GA on OA, qRT-PCR was performed to determine the gene expression levels of MMP-1, MMP-3, MMP-13 and TIMP-1 in chondrocytes incubated with GA or JEZTC prior to stimulate with IL-1β. As shown in Fig. 1A, Chondrocytes stimulated by IL-1β exhibited up-regulation of MMP-1, MMP-3, MMP-13 genes expression and down-regulation of TIMP-1 expression. By contrast, JEZTC inhibited the IL-1β-mediated induction of MMP-1, MMP-3 and MMP-13 genes expression and induced the expression of TIMP-1 in a dose-dependent manner. JEZTC-2 exhibited superior effect compared with JEZTC-1 and JEZTC-3. Comparatively, GA prevented the induction of MMPs gene expression or up-regulate TIMP-1 expression, its effect is not as marked as JEZTC did.

We next examined the effects of JEZTC and GA on the secretion of MMP-1 protein in IL-1β-induced chondrocytes using immunohistochemical analysis (Fig. 1C). Treating with IL-1β resulted in the up-regulation of MMP-1 protein expression, as indicated by more intensely positive staining (dark brown staining) compared with normal control. JEZTC could down-regulate the secretion of MMP-1 protein, as demonstrated by the obviously lower positive staining. This was consistent with the results of qRT-PCR. Among the JEZTC groups, JEZTC-2 exhibited the lightest staining of MMP-1. In addition, the effect of GA is relatively weaker than JEZTC.

Effects of JEZTC and GA on IL-1β-induced ROS production. The influence of JEZTC and GA on intracellular ROS production in IL-1β-stimulated chondrocytes was also detected. Treating with IL-1β resulted in an increase of intracellular ROS generation to 59.0 ± 5.0 (Fig. 1B). JEZTC significantly reduced intracellular ROS production by 9.5%, 26.2% and 19.1% in the JEZTC-1, JEZTC-2 and JEZTC-3 groups, respectively. However, this effect was much weaker in the GA group compared with JEZTC-treated groups.

Macroscopic observation. An OA experimental animal model of ACLT was used to evaluate the effects of JEZTC (JEZTC-2) on the development of tOA. After 8weeks of ACLT in rabbits, no meniscal tears were observed (data not showed). In the sham-operation (Normal) group, the cartilage on the femoral condyles was macroscopically normal, with a smooth, glistening surface, and no cartilage defect or osteophyte formation (Fig. 2A). In the PBS group, general characteristics of OA, including erosion and osteophyte formation were seen on the side of the femoral condyles after 4 and 8 weeks post-treatment. The degree of cartilage destruction was increased over time. The JEZTC group showed less bone wear than the PBS group as evidenced by gross appearance in the same period. Consistent with these findings, the macroscopic scores were reduced in the JEZTC group compared with the OA model group (Fig. 2B) by 66% and 55.3% at 4 and 8 weeks post-treatment, respectively.

ROS generation in articular cartilage. The levels of intracellular ROS generation in cartilage at 4 and 8 weeks post-treatment were detected by DCFDA probe (Beyotime, China). As shown in Fig. 2C, in the PBS group, ROS production increased in a time-dependent manner, by 21.4% and 64.6% at 4 and 8 weeks, respectively. However, JEZTC effectively reduced the intracellular ROS production by 8.9% and 46.2% at 4 and 8 weeks, respectively.

Histological changes in articular cartilage. Histological changes in the cartilage after surgery mainly involved the thinner cartilage layer, abraded surface and reduced safranin O depositing in the cartilage. In the PBS group, extensive degeneration of cartilage was observed, as illustrated by a mass of superficial layer loss, the general disappearance of chondrocytes (Fig. 3A) and the loss of GAG (Fig. 3B) from the femoral condyle. The cartilage in PBS group exhibited much greater destruction at week 8 compared to week 4. Nevertheless, the cartilage destruction was significantly diminished in the treatment of JEZTC which prevented damage to the cartilage structure, reduced cellular abnormalities, and preserved matrix architecture. However, osteophyte proliferation could not be reversed (Fig. 3A-C). The Mankin score was used to further evaluate the cartilage damage. Overall, the Mankin scores were lower in the JEZTC group at 4 and 8 weeks post-treatment compared with the PBS group (Fig. 3D and E), these was consistent with the macroscopic findings (Fig. 2B).

Effects of JEZTC on the expression of MMPs and TIMP-1 in articular cartilage. The expression levels of MMPs in articular cartilage at 4 and 8 weeks after treatment was determined by qRT-PCR. In the PBS group, an increase in MMP-1, MMP-3, and MMP-13 genes expression but decrease in TIMP-1 gene expression was showed. There was a statistically significant decrease in mRNA expression of MMP-1, MMP-3, and MMP-13 and up-regulation of the mRNA expression of TIMP-1 in the JEZTC group compared to the PBS group (Fig. 4A and B). These results suggested that JEZTC could inhibit the development of OA by down-regulating MMPs and up-regulating TIMP-1 genes expression.

In specimens of cartilage from the femoral condyle of normal rabbits, chondrocytes within the superficial layers were stained positive for MMP-1, which plays a key role in the progression of OA. In the PBS group, greater numbers of MMP-1-positive cells were detected in the middle and deep cartilage layers compared with normal cartilage, indicating extensive destruction of cartilage. However, in the JEZTC group, only a small number of positive cells were observed in the shallow and middle cartilage layers and most of the cartilage tissue remained intact (Fig. 3C).

To explore the molecular mechanisms of JEZTC inhibited IL-1β-induced destruction in chondrocytes and OA, we investigated the involvement of the MAPK, NF-κB and PI3K/Akt signaling pathways using western blot analysis. In the in vitro study, IL-1β-induced increased of phospho-P38/P38 (p-P38/P38), phospho-Akt/Akt (p-Akt/Akt) and NF-κB expression were diminished by JEZTC in a dose-dependent manner (Fig. 5B). Moreover, among the JEZTC groups, the greatest reduction of related proteins expression in JEZTC-2 group was superior to the JEZTC-1 and JEZTC-3 groups.

The in vivo investigation was consistent with the results in vitro. JEZTC significantly down-regulated p-P38/P38, p-Akt/Akt and NF-κB expression when compared to the PBS group (Fig. 6B). A scheme of signaling pathways involved in the anti-osteoarthritic effect of JEZTC is presented in Fig. 5C.

Further analysis revealed that as pro-inflammatory factors (tnf-α) decreased, the expression levels of proapoptotic genes bax and caspase 3 were up-regulated by JEZTC in a dose-dependent manner (Fig. 5A and 6A). The in vitro findings showed the expression of pro-apoptotic genes in JEZTC-2 group was higher than the other two groups. What is more, JEZTC also distinctly down-regulated tnf-α expression and increased the expression of pro-apoptotic genes in the in vivo study.

Drug therapy is among the most typical choices for the treatment of OA. GA, which can suppress the expressions of pro-inflammatory cytokines and chemokines have been widely studied [21-23]. IL-1 is a pivotal cytokine in many inflammatory disorders including OA, which induces MMPs and TIMP-1 involved in cartilage degradation in vitro. In our in vitro study, we found that JEZTC ameliorated IL-1β-induced chondrocytes destruction by inhibiting the expression of MMP-1, MMP-3 and MMP-13 as well as up-regulating the expression of TIMP-1 (Fig. 1A). MMP-3 protein was reported to be expressed in the synovium and the surface of cartilage in the knee joints and pannus-like tissue of patients with OA [36]. Among the MMPs, MMP-1 and MMP-13 are interstitial collagenases mainly responsible for type II collagen degradation, which is a committed step in the progression of OA [37]. It has been suggested that inhibition of IL-1 induced MMPs expression may benefit for chondro-protection on the pathological conditions [38], which was further demonstrated in in vivo investigation.

The effect of JEZTC on cartilage degradation in vivo was investigated in an ACLT-based experimental model of OA．This model is a well-established animal model of OA and has been widely used to evaluate the efficacy of agents in the treatment of OA. ACLT in the rabbit resulted in cartilage degradation due to the mechanical instability [39]. As evidenced by the improvement in macroscopic and histopathologic scores, injection of JEZTC for 4 and 8 weeks obviously improved the structure of the injured joints thereby alleviating the development of OA (Fig. 2A and B, Fig. 3). More specifically, the impact of JEZTC on the development of OA was manifested by a decrease in the OA-specific gene expression of MMP-1, MMP-3 and MMP-13 and an increase in TIMP-1 expression in animals treated with JEZTC (Fig. 4A and B). The results were in consistence with the in vitro findings (Fig. 1A) which indicated the beneficial effects of JEZTC on protection of cartilage degradation both in vitro and in vivo.

Analysis of the molecular mechanisms revealed that several signaling pathways are involved in the anti-arthritis effects of JEZTC, including MAPK, NF-κB and PI3K/Akt signaling pathways, which are known to play major roles in the biomechanical and metabolic pathways involved with the etiology and pathogenesis of OA [40, 41]. NF-κB is abundant in inflamed tissue and contributes to the progression of extracellular matrix damage and cartilage destruction [42]. During the process of OA, the cytokines, such as IL-1, activate the NF-κB dimers by triggering a signaling pathway that leads to the phosphorylation of IkB. Once phosphorylated, IκBs undergo poly-ubiquitination and ultimately proteosomic degradation, allowing NF-κB to enter the nucleus and promote the transcription of inflammatory genes like TNF-α and IL-1β [43]. The results showed that JEZTC markedly inhibited the activation of NF-κB, as evidenced by the down-regulation of the expression of NF-κB both in vitro (Fig. 5B) and in vivo (Fig. 6B). These indicate that JEZTC may down-regulate the expression of MMPs via the inactivation of NF-κB signaling pathway.

P38/ MAPK is part of an intracellular signaling pathway activated by proinflammatory cytokines and environmental stressors including IL-1 and TNF-α which have been linked to pathogenesis od OA [44-47]. In OA model, the level of phosphorylated MAPKs, including extracellular signal regulated kinase (ERK) and p38 appears to be higher than that in normal cartilage [48], these was in agreement with our study (Fig. 5B and 6B). However, the activity of P38 was decreased in cells treated with JEZTC in a dose-dependent manner. Further study showed that P38 was significantly down-regulated by JEZTC in OA model, suggesting the role of JEZTC in the regulation of p38 MAPK signaling pathway (Fig. 5A and 6B) to block OA.

The PI3K/AKT signaling pathway also plays an important role in OA [49]. PI3K and phosphorylated Akt act as survival signals in preventing TNF-α-induced apoptosis [50, 51]. In this study, the results showed that the p-AKT increased significantly in IL-1β induced in vitro model and in vivo OA model (Fig. 5A and 6B). In the treatment of JEZTC, the expression of p-AKT was significantly down-regulated in a dose-dependent. It indicates that JEZTC inhibits the production of inflammatory mediators in OA, which may be mediated partly through suppression of the PI3K/AKT pathway or MAPK cascades, leading to inactivation of NF-κB.

ROS are toxic and play an important role in the initiation and pathophysiology of OA [52]. Intracellular ROS can also mediate destroy and degradation of the extracellular matrix by causing of chondrocytes apoptosis in cartilage [53, 54]. As a strong antioxidant and anti-inflammatory agent, GA was found to attenuate ROS release to prevent OA progression [55], which was in agreement with our study. However, GA has much weaker effect than JEZTC (Fig. 1B and Fig. 2C). Our previous study showed that GA derivatives had stronger anti-oxidant effects than GA [29-33, 55], which may be due to that GA is much more hydrophilic and cytotoxic than its esters [14].

In summary, we investigated the protective effects of JEZTC on IL-1β-stimulated chondrocytes in vitro and OA in vivo. The inhibition of MMPs by JEZTC was associated in part with inhibition the PI3K/AKT, p38/MAPK and NF-κB signaling pathways. Our results indicate that JEZTC may be a promise therapeutic agent for the treatment of OA.

However, these preliminary findings in anti-osteoarthritic effect of JEZTC were only affirmed on the cell and animal level, further clinical studies are needed to confirm and extend.

This study was financially supported by National key research and development program of China (2016YFB0700804), National Natural Science Fund of China (Grant Nos. 81760326, 81472054 and 81860390), the Guangxi Science and Technology Major Project (Guike AA17204085), the Guangxi Scientific Research and Technological Development Foundation (Grant No. GuikeAB16450003), High level innovation teams and outstanding scholars in Guangxi Universities (The third batch), the Distinguished Young Scholars Program of Guangxi Medical University, the Key Program of Guangxi Collaborative Innovation Center for Biomedicine (GCICB-SR-2017002), the 2018 Basic Ability Improvement Project for Middle-aged and Young Teachers in Colleges and Universities in Guangxi and the Science Foundation for The Excellent Young Scholars of Guangxi Collaborative Innovation Center for Biomedicine (GCICB-TC-2017005).

No conflict of interests exists.

Z. Lu, Q. Liu and L. Liu contributed equally to this work.