Intra-articular Injection of Chloramphenicol Reduces Articular Cartilage Degeneration in a Rabbit Model of Osteoarthritis

Abstract

Background

Osteoarthritis (OA) is characterized by degeneration of articular cartilage. Studies have found that enhancement of autophagy, an intracellular catabolic process, may limit the pathologic progression of OA. Chloramphenicol is a potent activator of autophagy; however, the effects of chloramphenicol on articular cartilage are unknown.

Questions/purposes

Using human OA knee chondrocytes in vitro, we asked, does chloramphenicol (1) activate autophagy in chondrocytes; (2) protect chondrocytes from IL-1β-induced apoptosis; and (3) reduce the expression of matrix metallopeptidase (MMP)-13 and IL-6 (markers associated with articular cartilage degradation and joint inflammation). Using an in vivo rabbit model of OA, we asked, does an intra-articular injection of chloramphenicol in the knee (4) induce autophagy; (5) reduce OA severity; and (6) reduce MMP-13 expression?

Methods

Human chondrocytes were extracted from 10 men with OA undergoing TKA. After treatment with 25 μg/mL, 50 μg/mL, or 100μg/mL chloramphenicol, the autophagy of chondrocytes was detected with Western blotting, transmission electron microscopy, or an autophagy detection kit. There were four groups in our study: one group was untreated, one was treated with 100 μg/mL chloramphenicol, another was treated with 10 ng/mL of IL-1β, and the final group was treated with 10 ng/mL of IL-1β and 100 μg/mL of chloramphenicol. All groups were treated for 48 hours; cell apoptosis was detected with Western blotting and flow cytometry. Inflammation marker IL-6 in the cell culture supernatant was detected with an ELISA. Articular cartilage degradation-related enzyme MMP-13 was analyzed with Western blotting. A rabbit model of OA was induced by intra-articular injection of type II collagenase in 20 male 3-month-old New Zealand White rabbits' right hind leg knees; the left hind leg knees served as controls. Rabbits were treated by intra-articular injection of saline or chloramphenicol once a week for 8 weeks. Autophagy of the articular cartilage was detected with Western blotting and transmission electron microscopy. Degeneration of articular cartilage was analyzed with Safranin O-fast green staining and the semi-quantitative index Osteoarthritis Research Society International (OARSI) grading system. Degeneration of articular cartilage was evaluated using the OARSI grading system. The expression of MMP-13 in articular cartilage was detected with immunohistochemistry.

Results

Chloramphenicol activated autophagy in vitro in the chondrocytes of humans with OA and in an in vivo rabbit model of OA. Chloramphenicol inhibited IL-1-induced apoptosis (flow cytometry results with chloramphenicol, 25.33 ± 3.51%, and without chloramphenicol, 44.00 ± 3.61%, mean difference, 18.67% [95% CI 10.60 to 26.73]; p = 0.003) and the production of proinflammatory cytokine IL-6 (ELISA results, with chloramphenicol, 720.00 ± 96.44 pg/mL, without chloramphenicol, 966.67 ± 85.05 pg/mL; mean difference 74.24 pg/mL [95% CI 39.28 to 454.06]; p = 0.029) in chondrocytes. After chloramphenicol treatment, the severity of cartilage degradation was reduced in the treatment group (OARSI 6.80 ± 2.71) compared with the control group (12.30 ± 2.77), (mean difference 5.50 [95% CI 1.50 to 9.50]; p = 0.013). Furthermore, chloramphenicol treatment also decreased the production of MMP-13 in vitro and in vivo.

Conclusions

Chloramphenicol reduced the severity of cartilage degradation in a type II collagen-induced rabbit model of OA, which may be related to induction of autophagy and inhibition of MMP-13 and IL-6.

Clinical Relevance

Our study suggests that an intra-articular injection of chloramphenicol may reduce degeneration of articular cartilage and that induction of autophagy may be a method for treating OA. The animal model we used was type II collagen-induced OA, which was different from idiopathic OA and post-traumatic OA. Therefore, we need to use other types of OA models (idiopathic OA or a surgically induced OA model) to further verify its effect, and the side effects of chloramphenicol also need to be considered, such as myelosuppression.

Introduction

Osteoarthritis (OA) is characterized by articular cartilage degradation and joint inflammation [34]. The cartilage extracellular matrix has abundant fibrillar collagens, especially large, Type II aggregating proteoglycans and smaller hydrophilic macromolecules that give lubricative and swelling properties to the tissue [24]. Chondrocytes are the only cells in cartilage in adults and regulate the secretion of the extracellular matrix. A normal composition of the extracellular matrix is closely related to the articular cartilage homeostasis and normal biomechanical function of cartilage. Chondrocytes depend on a normal cartilage homeostasis to maintain cell survival and normal biosynthetic function [18]. As OA progresses, the metabolic homeostasis of the extracellular matrix is disrupted; the component synthesis of the extracellular matrix is reduced and activity of catabolic enzymes in the cartilage is enhanced, resulting in degradation of the extracellular matrix [29]. Consequently, targeting chondrocyte injury is important for developing effective therapies to treat OA.

Autophagy is a catabolic self-digestion process that, in most cases, maintains cellular integrity via the removal of damaged macromolecules and dysfunctional organelles [23]. Studies have found that autophagy protects cells from apoptosis induced by drugs to treat many diseases (for example, sulfaphenazole, an antimicrobial agent, can induce autophagy to protect the heart against ischemia reperfusion injury) and is associated with drug resistance [4, 39]. Therefore, autophagy is considered a protective process for cell survival under stressful conditions. Studies have revealed that age-related OA in humans and mice is associated with reduced expression of the autophagy-related proteins ULK1, Beclin1, and light chain 3 (LC3) and increased apoptosis in chondrocytes [6, 9]. Excessive mechanical load on cartilage also suppresses autophagy regulators [7]. Although the role of autophagy in chondrocytes is complex, emerging data suggest that up-regulation of autophagy can reduce cartilage degradation in vitro and in experimental OA in mice [5, 7, 14, 32]. These results indicate that autophagy may preserve joint health and prevent OA progression [21].

Chloramphenicol is a bacteriostatic antimicrobial, an inhibitor of cytochrome P450 monooxygenase, and an important regulator of autophagy [28]. Chloramphenicol can attenuate ischemia-reperfusion injury and plays a cardioprotective role by promoting the formation of autophagosomes [28]. Moreover, chloramphenicol can induce an autophagic response in osteosarcoma cells [26]. Chloramphenicol has also induced autophagy by inhibiting the synthesis of mitochondrial proteins and may up-regulate autophagosome formation in TP53^+/+ cells [38]. However, to our knowledge, the effect of chloramphenicol on OA has not been studied.

These studies indicate that inducing autophagy may delay the progression of OA. Using human OA knee chondrocytes in vitro, we asked, does chloramphenicol (1) activate autophagy in chondrocytes; (2) protect chondrocytes from IL-1β-induced apoptosis; and (3) reduce the expression of matrix metallopeptidase (MMP)-13 and IL-6 (markers associated with articular cartilage degradation and joint inflammation, which greatly influence OA pathogenesis [10, 11])? Using an in vivo rabbit model of OA, we asked, does an intra-articular injection of chloramphenicol in the knee (4) induce autophagy; (5) reduce the severity of OA; and (6) reduce the expression of MMP-13?

Materials and Methods

Overview of Experiments

This study was comprised of two parts: an in vitro assessment of articular cartilage tissue from the knee of humans with OA and an in vivo assessment of experimentally induced OA in the articular knee cartilage of young adult rabbits. For the in vitro assessment, the outcome measures were chondrocytes autophagy, cell apoptosis, articular cartilage degradation marker MMP-13 and joint inflammation marker IL-6. For the in vivo assessment in a rabbit model, the outcome measures were articular cartilage tissue autophagy, OARSI score (a semi-quantitative scoring system of joint degeneration) and MMP-13 (Fig. 1).

Fig. 1.

The current experimental overview diagram is shown here; CAP = chloramphenicol.

Human Cartilage Sampling

Samples of articular cartilage tissue were obtained from the knees of patients with OA who had undergone TKA (Table 1). This research was approved by the human research ethics committee of Xi’an Hong Hui Hospital; all patients’ OA fulfilled the American College of Rheumatology’s criteria for OA [1], and all patients provided written informed consent. A non-weightbearing area of cartilage without any macroscopically visible abnormalities was harvested and washed in sterilized saline. Next, the tissue was sliced into smaller sections of 1 mm³ and incubated in trypsin (2.5 mg/mL) (Sigma Co, St. Louis, Missouri, USA) at 37 °C for 40 minutes. After the trypsin solution was removed, the tissue slices were treated for 8 hours with Type II collagenase (2 mg/mL) (Sigma Co) in a Dulbecco's Modified Essential Medium/F12 medium (Thermo Scientific, Waltham, MA, USA) at 37 °C. The isolated chondrocytes were placed in Dulbecco’s Modified Essential Medium/F12 supplemented with 10% fetal bovine serum (Gibco Life Technologies, Grand Island, NY, USA) and 100 units/mL of penicillin (BOSTER, Wuhan, China) and 100 μg/mL of streptomycin (BOSTER), and were transferred to a culture flask. The cells were incubated at 37 °C in a humidified gas mixture containing 5% CO₂ balanced with air. Following the above procedure, we used second-passage chondrocytes for further experiments.

Table 1.

Clinical characteristics of OA patients

Experimental OA in Rabbits

All animal experiments were approved by the institutional animal care and use committee of Xi’an Jiaotong University. Twenty male 3-month-old New Zealand White rabbits (weight 2.5 kg to 2.8 kg) were purchased from the Animal Center of Xi’an Jiaotong University for use in this study. All rabbits were kept in individual cages in a room at 22 ± 3 °C with 55 ± 20% humidity and a 12-hour light-dark cycle. The knees of all rabbits’ right hind legs were injected with 1 mg of Type II collagenase (Sigma Co) to induce experimental OA [16]; the knees of the left hind legs served as controls. Twenty rabbits were marked with corresponding numbers (from 1 to 20), and randomly assigned to two groups according to the random number table by two observers (ZH, LY). In the first group, the knees of the rabbits’ hind legs were injected with 0.5 mL of saline (Control/Saline (-/-), Collagenase/Saline (+/-)), and in the second group, the knees of the rabbits’ hind legs were injected with 0.5 mL of saline containing 100 μg/mL of chloramphenicol (Rongsheng Company, Jiaozuo, China) (Control/Chloramphenicol (-/+), Collagenase/Chloramphenicol (+/+)). Type II collagenase was administered twice (once on the first day and once on the fourth day), and the other drugs (saline and chloramphenicol) were injected once per week until the rabbits were euthanized 8 weeks after the first injection. The tibial plateaus of the rabbits’ hind legs were harvested.

Western Blotting Analysis

To determine whether chloramphenicol activates the formation of autophagosome in chondrocytes of humans with OA, cells were treated with chloramphenicol at concentrations of 25 μg/mL, 50 μg/mL, and 100 μg/mL for 48 hours, and the expression of autophagy-related proteins LC3 and P62 were detected. To test whether chloramphenicol protects chondrocytes from IL-1β-induced apoptosis, the first group of chondrocytes was not given any treatment as a negative control, the second group was treated with 10 ng/mL of IL-1β, the third group was treated with 100 μg/mL of chloramphenicol and the last group was treated with 10 ng/mL of IL-1β and 100 μg/mL of chloramphenicol. All the cells were treated for 48 hours, and the expression of Bcl-2 and Bax were detected. To assess whether chloramphenicol reduces the expression of MMP-13 and IL-6, the first group of chondrocytes did not undergo any treatment as a negative control, the second group was treated with 10 ng/mL of IL-1β, the third group was treated with 100 μg/mL of chloramphenicol and the last group was treated with 10 ng/mL of IL-1β and 100 μg/mL of chloramphenicol. All cells were treated for 48 hours, and the expression of MMP-13 was detected. To determine if intra-articular injection of chloramphenicol induces autophagy in an in vivo rabbit model of OA, five rabbits were treated with saline (Control/Saline (-/-), Collagenase/Saline (+/-)) and five were injected with Chloramphenicol (Control/Chloramphenicol (-/+), Collagenase/Chloramphenicol (+/+)) for 8 weeks; the tibial plateaus were harvested and the expression of Beclin1 was measured. The treated cells or tissues were collected and lysed in a radioimmunoprecipitation assay buffer (Beyotime, Shanghai, China) containing a protease inhibitor cocktail (Beyotime, Shanghai, China) on ice. Bicinchoninic acid (Thermo Scientific, Waltham, MA, USA) was used to quantify protein. We loaded 30 μg of lane protein into sodium dodecyl sulfate-polyacrylamide gel electrophoresis and electrophoretically transferred the protein to nitrocellulose membranes (Millipore, Billerica, MA, USA). The membranes were blocked with 10% skimmed milk at room temperature for 2 hours and incubated in antibodies directed against microtubule associated protein 1 LC3 (1:1000, Sigma, St. Louis, Missouri, USA), Beclin1 (1:500, Bioss, Beijing, China), p62 (1:8000, Abcam, Cambridge, UK), B cell lymphoma 2 (Bcl-2) (1:1000, Abcam, Cambridge, UK), Bcl-2-associated X (BAX) (1:1000, Cell Signaling Technology, San Francisco, X, USA), MMP-13 (1:3000, Abcam, Cambridge, UK), and β-actin (1:1000, Bioss, Beijing, China) overnight at 4 °C. The membranes were washed and incubated in appropriate horseradish peroxidase-conjugated secondary antibodies (1:10000, Bioss, Beijing, China) for 2 hours at room temperature. Afterwards, the membranes were washed and examined with an electrochemiluminescence system (GeneGnome 5, Synoptics Ltd., Cambridge, UK). Relative densitometric analysis (a semi-quantitative analysis) was done after densitometric scanning of western blots by electrochemiluminescence.

Autophagosome Formation

We detected whether autophagosome formation in chondrocytes was elicited by chloramphenicol using a Cyto-ID™ autophagy detection kit (Enzo Life Sciences, Farmingdale, NY, USA), which detected autophagic vacuoles in live cells, according to the manufacturer’s recommendations. Briefly, articular cartilage cells of humans with OA were plated and treated with or without 100 μg/mL of chloramphenicol for 48 hours and stained with a dual detection reagent for 30 minutes at 37 °C in the dark. A fluorescence microscope (Olympus, Tokyo, Japan) was used to observe the expression of green dot fluorescence (autophagosome) in cells.

Transmission Electron Microscopy

To confirm the presence of autophagosomes, we used transmission electron microscopy. In the experiment that assessed whether chloramphenicol activates the formation of autophagosome in chondrocytes of humans with OA, some chondrocytes were treated with 100 μg/mL of chloramphenicol for 48 hours (other chondrocytes were left untreated as negative controls), and then autophagosomes were detected. To determine if intra-articular injection of chloramphenicol induces autophagy in an in vivo rabbit model of OA, five rabbits were treated with saline (Control/Saline (-/-), Collagenase/Saline (+/-)) and five were injected with chloramphenicol (Control/Chloramphenicol (-/+), Collagenase/Chloramphenicol (+/+)) for 8 weeks; the tibial plateaus were harvested and the autophagosomes were detected. Cartilage tissue or chondrocytes were fixed in ice-cold 2% glutaraldehyde/0.1 M phosphate-buffered saline (pH 7.2) and post-fixed in 1% osmium tetroxide. After being washed and dehydrated with a series of graded ethanol (30% to 100%), tissue or cells were embedded in propylene oxide or embedding resin (1:1). Resin blocks were cut into thin sections (60 nm) with an LKB V ultramicrotome (LKB, Bromma, Stockholm, Sweden), and the sections were placed on 200-mesh copper grids and stained with uranyl acetate and lead citrate. An H-7650 transmission electron microscope (Hitachi, Ibaraki, Japan) was used for examining the autophagic vesicles (double membrane-enclosed vesicles containing engulfed organelles or other cell components).

Cell Apoptosis Assay

To investigate the impact of chloramphenicol on IL-1β-stimulated chondrocytes, the first group of chondrocytes did not undergo any treatment as a negative control, the second group was treated with 100 μg/mL of chloramphenicol, the third group was treated with 10 ng/mL of IL-1β and the last group was treated with 10 ng/mL of IL-1β and 100 μg/mL of chloramphenicol. All cells were treated for 48 hours, and cell apoptosis was assayed with flow cytometry (guava easyCyte HT; EMD Millipore, Billerica) using the Annexin V-FITC/PI apoptosis detection kit (7SeaPharmTech, Shanghai, China). Counting 10,000 cells (including normal cells, apoptotic cells and dead cells), the percentage of apoptotic cells was then calculated.

ELISA

The first group of chondrocytes did not undergo any treatment as a negative control, the second group was treated with 100 μg/mL of chloramphenicol, the third group was treated with 10 ng/mL of IL-1β and the last group was treated with 10 ng/mL of IL-1β and 100 μg/mL of chloramphenicol, all the cells were treated for 48 hours, and culture supernatants were collected and stored at -80 °C until they were assayed. The expression of IL-6 was measured with an ELISA (BOSTER), according to the manufacturer’s directions.

Histologic Analysis

Degeneration of articular cartilage was assessed using the semi-quantitative scoring system Osteoarthritis Research Society International’s histopathology grading system of cartilage OA [27]. In preparation for the histologic analysis, we fixed the tibial plateaus of the rabbits’ hind legs in 4% paraformaldehyde for 24 hours and decalcified them for 2 months with 10% ethylenediaminetetraacetic acid (EDTA). The tissues were dehydrated, infiltrated with paraffin, and embedded in paraffin wax. The paraffin blocks were sectioned into 5-μm slices along the sagittal plane using a microtome, and Safranin O-fast green staining was performed. Three slices were selected from each medial tibial plateau, and two observers (XS, KX) who were blinded to the animal study independently used a semi-quantitative scoring system (OARSI’s histopathology grading system of cartilage OA) to evaluate articular cartilage degeneration.

Immunohistochemistry

Five rabbits were treated with saline (Control/Saline (-/-), Collagenase/Saline (+/-)) and five were injected with Chloramphenicol (Control/Chloramphenicol (-/+), Collagenase/Chloramphenicol (+/+)) for 8 weeks, the tibial plateaus were harvested and the expression of MMP-13 in rabbit articular tissue was evaluated by an immunohistochemistry test. After the tissues were dewaxed, 3% H₂O₂ was used to inhibit the levels of endogenous peroxide on the histologic slices, after which microwave heating was used to retrieve antigen. The slices were blocked for 30 minutes with 5% goat serum (Beyotime Institute of Biotechnology, Shanghai, China). The slices were incubated in MMP-13 primary antibodies (1:100, Abcam); 5% bovine serum albumin was used for the negative controls. After the slices were incubated for 12 hours at 4 °C, a horseradish peroxidase-conjugated secondary antibody (Abcam) was used for 30 minutes, and cells were stained with 3,3'-diaminobenzidine (Beyotime Institute of Biotechnology) and mounted. A microscope (Olympus, Tokyo, Japan) was used for examining the percentage of MMP-13 positive cells (brown cells).

Statistical Analysis

All the data are presented as the mean ± SD. Differences between two groups were analyzed by unpaired t- test or the Mann-Whitney U test for experiments in which the datasets were not normally distributed. Differences in multiple groups were analyzed by ANOVA. GraphPad Prism 5 software (GraphPad Software, La Jolla, CA, USA) was used for all statistical analyses. P values less than 0.05 were considered statistically significant.

Results

Chloramphenicol Activated the Formation of Autophagosomes in Chondrocytes of Humans with OA

We performed a Western blotting analysis to detect LC3 and p62, two selective markers of autophagy (Fig. 2A). Chloramphenicol treatment led to a dose-dependent increase in the LC3-II to LC3-I ratio (Fig. 2B) and a decrease in the amount of p62 (Fig. 2C). The expression of autophagosome (green dots) in chondrocytes was induced after treatment with 100 μg/mL of chloramphenicol for 48 hours (Fig. 2D). After treatment with 100 μg/mL of chloramphenicol for 48 hours, many autophagic vesicles—double-membrane-enclosed vesicles containing engulfed organelles (arrows)—were observed in the cytoplasm of chondrocytes (Fig. 2E).

Fig. 2 A-E.

Chloramphenicol activated the formation of autophagosome in chondrocytes of humans with OA. The expression of LC3, p62 and β-actin (as a loading control) were detected with Western blotting (A), and the levels of LC3 (B), p62 (C), were analyzed with relative densitometric analyses. Chondrocytes in humans with OA were treated or not treated with 100 μg/mL of chloramphenicol for 48 hours, and autophagosome formation (green dots and arrows) was detected using an autophagy detection kit (D) and transmission electron microscopy (E). The LC3-II to LC3-I and p62 to β-actin ratios represent three experiments using cells from different patients and are expressed as the mean ± SD. The data were analyzed using ANOVA.

Chloramphenicol Protected Chondrocytes from IL-1β-induced Apoptosis

The flow cytometry analysis results showed that chloramphenicol decreased apoptosis induced by IL-1β (with chloramphenicol, 25.33 ± 3.51%, and without chloramphenicol, 44.00 ± 3.61%; mean difference, 18.67% [95% CI 10.60 to 26.73]; p = 0.003) (Fig. 3A). The expression of Bcl-2 and BAX were detected with Western blotting (Fig. 3B). After treatment with 10 ng/mL of IL-1β for 48 hours, the expression of Bcl-2 (Fig. 3C) was increased and the expression of BAX (Fig. 3D) was reduced with 100 μg/mL of chloramphenicol. These data demonstrated that chloramphenicol inhibited IL-1β-induced apoptosis in chondrocytes.

Fig. 3 A-D.

Chloramphenicol protected chondrocytes from IL-1β-induced apoptosis. (A) Cell apoptosis was measured via an Annexin-V-FITC/PI double-staining assay. (B) The expression of Bcl-2 (C) and BAX (D) was measured with Western blotting and relative densitometric analyses. The results represent three experiments using cells from different patients and are expressed as the mean ± SD. The data were analyzed using an unpaired t-test.

Chloramphenicol Reduced the Expression of MMP-13 and IL-6

The expression of MMP-13 was detected with Western blotting (Fig. 4A), and the results showed that IL-1β promoted MMP-13 expression, and chloramphenicol suppressed IL-1β-induced MMP-13 expression (Fig. 4B). Moreover, chloramphenicol alleviated IL-1β-induced inflammation and reduced the expression of IL-6 (ELISA results, with chloramphenicol, 720.00 ± 96.44 pg/mL, without chloramphenicol, 966.67 ± 85.05 pg/mL; mean difference 74.24 pg/mL [95% CI 39.28 to 454.06]; p = 0.029) (Fig. 4C). The results indicate that chloramphenicol protected chondrocytes from IL-1β-induced injury.

Fig. 4 A-C.

Chloramphenicol inhibited the expression of MMP-13 and IL-6. The expression MMP-13 was measured with Western blotting (A) and relative densitometric analyses (B). The expression of IL-6 in chondrocyte supernatants was analyzed with an ELISA (C). The results represent three experiments using cells from different patients and are expressed as the mean ±SD. The data were analyzed using an unpaired t-test with Welch’s correction.

Intra-articular Injection of Chloramphenicol-Induced Autophagy in an In Vivo Rabbit Model of OA

After the intra-articular injection, side effects such as allergy, diarrhea, and wound infection were not observed in any rabbits. Intra-articular injection of chloramphenicol increased the expression of Beclin1 in the non-collagenase and collagenase groups (Control/Chloramphenicol (-/+) and Collagenase/Chloramphenicol (+/+), Fig. 5A). The expression of Beclin1 in the collagenase group (Collagenase/Saline (+/-) and Collagenase/Chloramphenicol (+/+)) was decreased compared with that in the non-collagenase group (Control/Saline (-/-) and Control/Chloramphenicol (-/+), Fig. 5B). Chondrocytes from chloramphenicol-treated rabbits (Control/Chloramphenicol (-/+) and Collagenase/Chloramphenicol (+/+)) exhibited more double-membrane-enclosed vesicles containing engulfed organelles than those from saline-treated rabbits in the control (Control/Saline (-/-)) and collagenase groups (Collagenase/Saline (+/-)) with transmission electron microscopy (Fig. 5C). All results indicated that an intra-articular injection of chloramphenicol increased chondrocyte autophagy in rabbits with collagenase-induced OA.

Fig. 5-C.

An intra-articular injection of chloramphenicol induced autophagy in experimental OA. The expression of Beclin1 was measured with Western blotting (A) and relative densitometric analyses (B) (n = 5 each group). (C) Autophagosomes were detected with transmission electron microscopy. -/-: Control/Saline, -/+: Control/ Chloramphenicol, +/-: Collagenase/Saline, +/+: Collagenase/ Chloramphenicol. The results of Beclin1 expression represent five experiments using cartilage proteins from different rabbits and are expressed as the median and interquartile range. The data were analyzed using the Mann-Whitney test.

Intra-articular Injection of Chloramphenicol Reduced the Severity of OA in the Rabbit Model

Cartilage in rabbits in the collagenase group (+/- and +/+) showed OA-like changes, with loss of the cartilage surface and reduced expression of proteoglycan (Fig. 6A). In the collagenase group, the severity of cartilage degeneration was decreased in chloramphenicol-treated rabbits (+/+) compared with saline-treated rabbits (+/-) (Fig. 6A). The histologic evaluation did not reveal any degenerative changes in saline- (-/-) or chloramphenicol-treated (-/+) rabbits in the control group (Fig. 6A). Evaluation of the extent of OA using the OARSI system revealed that chloramphenicol treatment (+/+) (6.80 ± 2.71) decreased the severity of cartilage degeneration compared with treatment of the saline-treated (+/-) (12.30 ± 2.77), (mean difference 5.50 [95% CI 1.50 to 9.50]; p = 0.013) rabbits with collagenase-induced OA (Fig. 6B). The results indicate that an intra-articular injection of chloramphenicol reduced OA severity in the in vivo rabbit model.

Fig. 6 A-B.

An intra-articular injection of Chloramphenicol reduced the severity of experimental OA. Degeneration of articular cartilage degeneration was evaluated with (A) Safranin O-fast green staining and (B) the OARSI grading system (n = 5 each group). -/- = Control/Saline, -/+ = Control/Chloramphenicol, +/- = Collagenase/Saline, +/+ = Collagenase/Chloramphenicol. Values are expressed as the mean ± SD. The data were analyzed using an unpaired t-test.

Intra-articular Injection of Chloramphenicol Reduced the Expression of MMP-13 in OA in the Rabbit Model

The MMP-13 expression was detected with immunohistochemical test (Fig. 7A), and the statistical results revealed that chloramphenicol treatment (+/+) decreased the expression of MMP-13 compared with saline-treated (+/-) rabbits in collagenase group (median [range] 41% [32 to 47] versus 57% [51 to 64]; difference of medians [16%]; p = 0.008) (Fig. 7B). These results suggest that an intra-articular injection of chloramphenicol may protect cartilage from damage by reducing MMP-13 production.

Fig. 7 A-B.

An intra-articular injection of chloramphenicol reduced the expression of MMP-13 in experimental OA. The knees of rabbits’ hind legs were treated with saline or chloramphenicol once per week, and 8 weeks after the first intra-articular injection, the tibial plateaus were harvested (n = 5 each group). (A) MMP-13 was detected with immunohistochemistry tests. (B) The percentages of MMP-13-positive cells (arrows) were randomly counted using three fields. -/- = Control/Saline; -/+ = Control/Chloramphenicol; +/- = Collagenase/Saline; +/+ = Collagenase/Chloramphenicol. The results are expressed as the median and interquartile range. Data were analyzed using the Mann-Whitney test.

Discussion

OA is the most common clinical chronic joint disorder [3]; therefore, the development of an effective therapy to inhibit its progression remains a research topic of considerable clinical interest. Chondrocytes are a kind of non-renewable cells, which are closely related to the pathological mechanism and progression of OA. In our study, we reported that chloramphenicol, administered to treat OA, inhibited IL-1β-induced apoptosis and IL-6 production in chondrocytes and activated autophagy in vitro in human chondrocytes and in vivo in a rabbit model of OA. Moreover, our results revealed that chloramphenicol reduced the severity of collagenase-induced OA, which may be achieved by reducing MMP-13 expression.

There were limitations to our study. First, all the human cartilage chondrocytes were derived from males, and all the OA animal models used were male rabbits, which was a potential selection bias and may not be representative of female patients with OA. Second, all chondrocytes were derived from patients undergoing knee replacement surgery, which means that OA had progressed to a late stage. In general, drug therapy is often used in the early stages of OA, and we had no data on chondrocyte experiments in the early stages of OA in this study; readers should pay attention to this point. Third, the animal model we used in this paper was type II collagen-induced OA, which has a different mechanism from idiopathic OA and post-traumatic OA, thus, it may not be broadly representative of all OA conditions. Fourth, the mechanism of chloramphenicol treatment of OA was not fully elucidated. We found that chloramphenicol induced autophagy in chondrocytes in this study, but no evidence showed that chloramphenicol inhibited IL-1β-induced apoptosis and IL-6 production by inducing autophagy in chondrocytes; further research should examine the role of chloramphenicol after autophagy inhibition.

Autophagy is a highly conserved cellular homeostasis mechanism in which subcellular components (including damaged organelles and protein aggregates) are delivered to lysosomes by autophagosomes, followed by degradation by lysosomal enzymes [8, 17, 19, 36]. Autophagosome formation is controlled by Atg proteins, including Atg12, Atg5, and LC3 [6]. LC3 is present in two forms: LC3-I in cytoplasm and LC3-II bound to the autophagosome membrane. LC3-I is converted to LC3-II through lipidation, resulting in the association of LC3-II with autophagic vesicles. In this process, the LC3-II to LC3-I ratio increases [15]. Age-related OA in humans and mice is associated with reduced autophagy and increased apoptosis in chondrocytes, and autophagy may help maintain joint health and prevent OA progression [2, 6-7, 9, 21, 30]. Carames et al. [5] found that intraperitoneal injection of rapamycin, an autophagy inducer, which works at least in part by activating chondrocyte autophagy, reduced the severity of experimental osteoarthritis. One study found that chloramphenicol regulated autophagy by inhibiting the synthesis of mitochondrial proteins and could up-regulate the formation of autophagosomes in TP53^+/+ cells [38]. Although chloramphenicol-induced autophagy has been detected in some diseases [26, 28], to our knowledge, the role of chloramphenicol in OA has not been reported. In the present study, we found that the LC3-II to LC3-I ratio was increased and the expression of the autophagic flux-related protein p62 was decreased in a dose-dependent manner in chloramphenicol-treated chondrocytes. The transmission electron microscopy study found that autophagy became more pronounced and many more autophagosomes were observed in chloramphenicol-treated chondrocytes. The results indicated that autophagy was induced by chloramphenicol in human chondrocytes.

Autophagy is likely a self-protective process in OA in response to stimulation with chloramphenicol. A previous study indicated that IL-1β treatment promoted the expression of pro-apoptosis regulators and the apoptosis of primary chondrocytes from cartilage with OA [13]. In the current study, we found that IL-1β promoted the expression of the pro-apoptosis regulator BAX and attenuated the expression of the anti-apoptosis modulator Bcl-2, consistent with the results of previous studies [25, 35]. Furthermore, we observed that chloramphenicol abrogated the effects of IL-1β-induced apoptosis in chondrocytes. The Annexin-V-FITC/PI double-staining assay and Western blotting analysis showed that chloramphenicol decreased chondrocyte apoptosis induced by IL-1β. These findings are consistent with the results of Sala-Mercado et al. [28], who demonstrated that chloramphenicol can attenuate ischemia-reperfusion injury and play a cardioprotective role by promoting autophagosome formation. To our knowledge, the current report is the first demonstrating that chloramphenicol modulates the IL-1β induction of chondrocyte apoptosis in articular cartilage with OA.

Two studies have confirmed that IL-1β greatly impacts chondrocytes to produce extracellular matrix components, interfering with the synthesis of pivotal structural proteins such as Type II collagen [31, 33]. In addition, IL-1β influences the synthesis of MMPs such as MMP-13 in chondrocytes, which have a destructive effect on cartilage components [37]. Therefore, we detected the expression of MMP-13 in chondrocytes. Consistent with prior studies [37], we found that IL-1β increased MMP-13 expression in chondrocytes. We also found that chloramphenicol efficiently decreased MMP-13 production in chondrocytes. However, this finding was different from the results of Li et al. [20], who found that chloramphenicol increased MMP-13 expression in cancer cells by inducing PI-3K/Akt and c-Jun N-terminal kinase phosphorylation and activating activator protein 1 [20]. We do not know why our results were opposite from these results; further research is needed. Studies have found that IL-1β stimulates the secretion of the pro-inflammatory cytokine IL-6 from chondrocytes [121, 22]. Therefore, we detected the expression of IL-6 in chondrocytes using an ELISA. Our results showed that chloramphenicol down-regulated IL-1β-induced IL-6 secretion. The above evidence suggests that chloramphenicol treatment had anti-catabolic and anti-inflammatory effects on OA.

We also investigated the role of an intra-articular injection of chloramphenicol on autophagy in rabbits with collagenase-induced OA. Consistent with cell experiments, we found that intra-articular chloramphenicol injection increased the autophagy-related protein Beclin1 expression, and more autophagosomes were observed by transmission electron microscopy. Bcl-2 can inhibit autophagy by binding with Beclin 1 [39]. The present results indicated that chloramphenicol may inhibit the binding of Beclin1 to Bcl-2. These results indicated that intra-articular injection of chloramphenicol can effectively induce autophagy.

We studied the role of chloramphenicol in the severity of collagenase-induced OA with Safranin O-fast green staining and the OARSI histopathology grading system of cartilage OA [27]. The histologic evaluation showed that osteoarthritis-like changes, with loss of the cartilage surface and reduction of proteoglycan, were observed in rabbits at 8 weeks after the injection of collagenase, but the intra-articular injection of chloramphenicol decreased the severity of cartilage degeneration. The OARSI system yielded a similar result; chloramphenicol-treated rabbits with collagenase-induced OA had a lower OARSI score than saline-treated rabbits. These findings indicated that chloramphenicol treatment reduced the severity of collagenase-induced OA.

Using immunohistochemistry, we investigated the role of intra-articular injection of chloramphenicol on MMP-13 expression in collagenase-induced OA. The results showed that chloramphenicol inhibited the expression of MMP-13 in cartilage, which was consistent with cell experiments. These results suggested that an intra-articular injection of chloramphenicol may protect cartilage from damage by reducing the production of MMP-13.

In summary, our results suggest that chloramphenicol promoted in vitro and in vivo autophagy. Chloramphenicol administration inhibited IL-1-induced apoptosis and production of the pro-inflammatory cytokine IL-6 in chondrocytes. Chloramphenicol treatment in a rabbit model reduced the severity of collagenase-induced OA with no observable side effects, which may be the result of autophagy induction and MMP-13 and IL-6 inhibition. Our results suggest that inducing autophagy may be a novel therapeutic strategy to confer cytoprotection against OA, and an intra-articular injection of chloramphenicol might be a therapeutic approach to OA treatment. Future experiments should be performed in other types of OA models (such as, idiopathic OA or surgically induced OA models). In addition, we should also pay attention to the side effects of chloramphenicol, such as myelosuppression.