Li P et al / Acta Pharmacol Sin 2004 Oct; 25 (10): 1262-1266

In vivo and in vitro chondrotoxicity of ciprofloxacin in juvenile rats¹

Ping LI, Neng-neng CHENG, Bin-yan CHEN, Yong-ming WANG²

Department of Pharmacology, School of Pharmacy, Fudan University, Shanghai 200032, China

¹ Project supported by the JRCC Fund.

² Correspondence to Prof Yong-ming WANG. Phn 86-21-5423-7465. Fax 86-21-6449-3495. E-mail Ymwang@shmu.edu.cn

Received 2003-12-31 Accepted 2004-04-22

KEY WORDS ciprofloxacin; knee joints; articular cartilage; glycosaminoglycans; cell survival; toxicokinetics; juvenile rats

ABSTRACT

AIM: To study the relationship between chondrotoxicity and toxicokinetics of ciprofloxacin (CPFX). METHODS: Rats, 4-week old, were treated with CPFX 0, 400, 800, and 1200 mg/kg ig once daily on seven consecutive days. The knee joint cartilage was examined histopathologically. The concentration of CPFX in venous blood and knee joint cartilage samples were determined by a microbioassay using Escherichia coli 44102. The effects of CPFX on proliferation of chondrocytes and secretion of soluble proteoglycans were determined with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) and 1,9-dimethylmethylene blue (DMB) assay, respectively. RESULTS: Cartilage was severely lesioned after treatment with CPFX 800 or 1200 mg/kg for 7 d, such as matrix swelling and loss of chondrocytes. The thickness of cartilage was significantly decreased compared with the control group. The maximum serum concentration (C_max), the area under the plasma concentration-time curve (AUC_0-∞), and concentration in cartilage was 16.3±2.1 mg/L, 97.2±12.3 mg·h·L^-1, and 13.4±2.8 µg·g^-1 and 21.8±2.5 mg/L,143.1±22.3 mg·h·L^-1, and 20.3±3.5 µg·g^-1 after oral administration of CPFX 800 or 1200 mg/kg on d 1, respectively. The data on d 6 were similar with that on d 1. CPFX inhibited proliferation of chondrocytes and the secretion of soluble proteoglycans. CONCLUSION: CPFX concentrations in serum and cartilage could provide a better basis for risk assessment.

INTRODUCTION

Quinolones (QNs) were widely used in clinical practice owing to their wide spectrum antibacterial activity and high degree of bioavailability. They were not approved for use in children and adolescents due their toxic effects on joint cartilage of immature animals^[1]. So QNs was a second-line drug following the failure of a well established antibiotic treatment^[2].

The extraordinary age-related drug toxicity of QNs was demonstrated in juvenile dogs, rats, rabbits, guinea pigs, and marmosets. However, the relationship between the adverse clinical effects of these drugs and lesions in target tissue of laboratory animals has not been clarified^[3,4].

A rational comparison or extrapolation of chondrotoxic potentials between different species should be based on kinetics. As very few toxicokinetic studies in immature animals have been published, reasonable extrapolations of the toxicological results are still limited. Because there exists significant differences in the kinetics of drugs in rodents and humans, it makes no sense to compare the doses applied in toxicological experiments in rats with those used therapeutically in humans. Instead, concentrations achieved in plasma or the target tissue represented a more reliable basis for such comparisons^[5].

Kato et al reported that levofloxacin inhibited glycosaminoglycan synthesis initially and DNA synthesis and mitochondrial function secondarily at actual arthro-pathic concentrations in cultured rabbit chondrocytes but that these changes were reversible and not enough to kill the cells^[6]. Egerbacher et al reported that enroflo-xacin and ciprofloxacin hydrochloride inhibited the proliferation of cultured canine and equine chondrocytes and decreased their ability to adhere to the culture dish^[7].

Ciprofloxacin (CPFX) is widely used in the clinic, while it is not clear at what concentrations in plasma and target tissue it will induce the lesions of articular joints. So in this paper we studied the kinetics of CPFX in plasma and determined its concentration in joint cartilage of juvenile rats and the related mechanisms.

MATERIALS AND METHODS

Materials Male 4-week old SD rats were provided by the Experimental Animal Center of Fudan University and housed under controlled conditions (tempera-ture 21±1 ºC and lighting 8:00-20:00) with food and water ad libitum. CPFX was provided by the He-nan Institute for Drug Control. HAM's F-12 and fetal calf serum were supplied by Gibco. Trypsin, colla-genase, 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetra-zolium bromide (MTT), and chondroitin sulfate C were from Sigma. 1,9-Dimethylmethylene blue (DMB) was obtained from Polysciences. Image Analysis System was purchased from Shanghai Shenteng Information Technology Co.

Kinetics Rats were randomly assigned to 4 groups (each including 5 animals). CPFX at 400, 800, and 1200 mg/kg body weight was administered ig once daily on seven consecutive days. CPFX was suspended in 0.5 % sodium carboxymethylcellulose (CMC). Rats in control group were ig 0.5 % CMC. Venous blood was collected from an incised tail vein just before and at 15, 30, 45 min and 1, 1.5, 2, 4, 6, 8, 12 h after the dosing on d 1 and d 6. Serum was separated and stored at -20 ºC until assay.

Histopathology Four hours after final dosing the rats were killed. The left knee joint was removed, fixed in 10 % formalin, and stored. Cartilage was obtained from the femoral condyle of the right knee joint, weighed and homogenized in saline. After centrifugation, the supernatant was stored at -20 ºC until assay.

The joint samples were decalcified and embedded in paraffin, serially sectioned in a sagittal plane at 8 µm and routinely-stained with hematoxylin and eosin. Four tissue sections were used randomly for the histopathological examination for one animal. The thickness of the cartilage of the femoral condyle was measured with Image Analysis System. The thickness of one sample was obtained from the average of those from four sections.

Microbioassay CPFX concentration in serum and cartilage was determined by a validated agar diffusion microbioassay using Escherichia coli 44102^[8]. The standard curve was prepared with rat normal serum and saline, and the linear relationship was determined between the concentrations of 0.125 and 16 mg/L. The intra- and inter-day coefficients of variation were less than 5 % in standard curve and the limit of sensitivity of the assay was approximately 0.125 mg/L.

Isolation and culture of articular chondrocytes Cartilage fragments were excised from the femuro-tibial joints of the rats and cut into small pieces (1 mm×1 mm×1 mm). Cartilage pieces were trypsinized for 30 min and treated with 0.1 % collagenase for 3 h at 37 ºC. Cells were filtered through nylon mesh (50 µm pore size) to remove matrix residues, centrifuged, and seeded 1×10⁵ cells per 50 mm³ flask in HAM's F-12 culture medium supplemented with 10 % fetal calf serum. After 5-d culture with growth medium in flasks the cells were trypsinized and counted. For the MTT assay, cells were dispensed into 96-well microtiter plates (4×10³ cells per well) and for the DMB assay cells were dispensed into 24-well plates (3×10⁴ cells per well). After 24 h of pre-incubation with growth medium, CPFX was added at concentrations of 0, 5, 10, 20, 40, 80, and 160 mg/L.

Determination of proliferation Cell proliferation was determined by the MTT assay^[9]. Chondrocytes were incubated with CPFX for 1, 3, and 5 d, respectively. Then MTT (5 g/L, in PBS, pH 7.4) was added and incubated at 37 ºC for 4 h. The medium was removed and Me₂SO was added to dissolve the crystals. The absorbance was measured at 490 nm in a microtiter plate photometer. Data were presented as inhibitory ratio.

Determination of proteoglycans After 5-d incubation proteoglycans were determined in culture media supernatants by the DMB assay^[10]. Samples in volume of 0.5 mL were mixed with 0.5 mL DMB solution (32 mg/L in 0.05 mol/L formate buffer, pH 3.5) and pipetted into cuvettes. Absorbance was measured using a spectrophotometer at 530 nm. A standard curve was constructed at chondroitin sulfate C 5-100 mg/L . The cells were trypsinized and counted. The content of proteoglycans was presented as mg per 1×10⁵ cells.

Data analysis Pharmacokinetic analysis was performed using the 3p97 program. Results were expressed as mean±SD. Statistical comparisons of the means were performed using multivariate analysis of variance with SPSS software.

RESULTS

Histopathology No obvious cartilage alterations were observed after treatement with CPFX 400 mg/kg for 7 d. However cartilage was severely lesioned after treatment with CPFX 800 or 1200 mg/kg for 7 d, such as matrix swelling and loss of chondrocytes. The thickness of cartilage in CPFX 800 and 1200 mg/kg groups was significantly decreased compared with the control group, respectively (Fig 1).

Fig 1. Articular cartilage of femoral condyle from 4-week old rats. (A) Control; (B) CPFX 1200 mg/kg. ×200. (C) Effect of CPFX 400-1200 mg/kg on thickness of the cartilage. n=5. Mean±SD. ^bP<0.05 vs Control.

Toxicokinetics CPFX was rapidly absorbed after dosing. The concentration of CPFX reached the peak at 1.5 h and then declined gradually (Fig 2). The C_max, AUC_0-∞, and CPFX concentration in cartilage was increased in a dose-dependent manner. The C_max and AUC_0-∞ are similar on d 1 and d 6 (Tab 1).

Fig 2. Serum concentration-time profile after a single oral dose of CPFX 400, 800, and 1200 mg/kg in 4-week old rats. n=5. Mean±SD.

Tab 1. Main pharmacokinetic parameters on d 1 and d 6 and the concentration of CPFX in cartilage 4 h after dosing on d 7 in 4-week old rats after treatment with CPFX 400, 800, and 1200 mg/kg. n=5. Mean±SD.

CPFX/mg·kg-1	Cmax/mg·L-1	AUC/mg·h·L-1	Cmax/mg·L-1	AUC/mg·h·L-1	Concentration in
	d 1	d 1	d 6	d 6	cartilage/µg·g-1
400	9.6±1.3	48.1±7.2	9.4±1.6	51.1±6.0	7.4±2.1
800	16.3±2.1	97.2±12.3	15.2±2.2	95.4±10.6	13.4±2.8
1200	21.8±2.5	143.1±22.3	23.7±3.0	148.8±19.8	20.3±3.5

Proliferation of cultured chondrocytes CPFX inhibited chondrocytes proliferation in a dose-dependent and time-dependent manner. The threshold concentration of CPFX to inhibit chondrocytes proliferation was 40 mg/L after 1-d culture, 20 mg/L after 3-d culture, and 10 mg/L after 5-d culture (Fig 3).

Fig 3. Inhibitory effects of CPFX on proliferation of cultured chondrocytes from 4-week old rats. n=4. Mean±SD. ^bP<0.05, ^cP<0.01 vs control.

Soluble proteoglycans CPFX 10-80 mg/L decreased the secretion of soluble proteoglycans after incubation with chondrocytes for 5 d (Fig 4).

Fig 4. Effect of CPFX on secretion of soluble proteoglycans from cultured chondrocytes of 4-week old rat. n=5. Mean±SD. ^bP<0.05, ^cP<0.01 vs control.

DISCUSSION

For most xenobiotic compounds, kinetics in animals and humans differ considerably. Therefore, data from toxicological studies can be interpreted only on the basis of the kinetics of the compound in the species studied. In the present study, the toxicity of CPFX on knee joint cartilage in juvenile rats was investigated after oral administration once daily for seven consecutive days, resembling the length of a clinical treatment period. CPFX decreased thickness of the articular cartilage of the femoral condyle at 800 and 1200 mg/kg but not at 400 mg/kg. Although the doses we used was 50, 100, and 150 fold than the therapeutic doses in humans, the mean peak concentrations in the serum of juvenile rats were only approximately 4, 6, and 8 times than those in human volunteers after administration of a single oral dose of CPFX 500 mg^[11]. Thus, it is very important to include toxicokinetic investigations to provide a reasonable basis for risk assessment. Our study showed that concentration of CPFX in serum reached the peak at 1.5 h after administration of an oral dose of 400, 800, and 1200 mg/kg, respectively. The C_max and AUC_0-∞ were similar on d 1 and d 6 indicating no accumulation after multiple doses. The cartilag lesions become severe with an increase of C_max, AUC_0-∞, and concen-traion in cartilage.

Stahlmann et al reported that lesions in knee joint cartilage from juvenile rats were detectable only after oral treatment with sparfloxacin 1800 mg/kg but not 600 mg/kg. The corresponding C_max was 12.4 mg/L and the AUC_0-∞ was 228 mg·h·L^-1 in serum. At all time points the concentrations of sparfloxacin in joint cartilage were significantly higher than those in serum^[5]. Kato et al discovered that repeated oral administration of levofloxacin 100 mg/kg in rats for 7 d induced lesions in the articular cartilage of the femoral condyle, while at dose of 30 mg/kg did not. The corresponding data for levofloxacin concentration in serum and articular cartilage were 6.24 g/L and 4.93 µg/g in 30 mg/kg group, and 21.98 g/L and 12.20 µg/g in 100 mg/kg group, respectively ^[6]. The concentration of levofloxacin in plasma was higher than that in cartilage. In our study no significant differences between concentration of CPFX in serum and cartilage were observed. Nagai et al demonstrated that the although garenoxacin concentrations in plasma and joint tissue in beagel dogs were higher than those for ciprofloxacin and norfloxacin, the articular toxicity of garenoxacin was much less than that of the other two antimicrobials^[12].

To confirm our results obtained in vivo, we investigated the effects of CPFX on proliferation of cultured chondrocytes and secretion of soluble proteo-glycans. CPFX inhibited chondrocytes proliferation and secretion of soluble proteoglycans in a concentration- and time-dependent manner. The minimal toxic concentration of CPFX was 10 mg/L similar with that in in vivo experiments. So the cultured cells would be appropriate to evaluate the toxicity of QNs on chondro-cytes in vitro. CPFX caused cartilage lesion by inhibiting proliferation of chondrocytes and secretion of proteoglycans.

Hildebrand et al reported that QNs decreased soluble proteoglycan levels in the culture medium and the incorporation of ³⁵S-sulfate into high-molecular-weight molecules in cultures of dog chondrocytes^[13]. Egerbacher et al also observed that in cultivated horse and dog chondrocytes, cell proliferation decreased in CPFX- and enrofloxacin-treated groups^[14]. Our results were consistant with them.

The specific mechanism responsible for the initiation of quinolone-induced arthropathy has not been determined. The most plausible explanation for the mechanism involved the chelation of magnesium ions by QNs, resulting in changed function of chondrocyte surface integrin receptors^[15,16]. QNs treatment plus Mg²⁺-free medium increased the negative effects on cell morphology and adhesion seen in the QNs or Mg²⁺-free groups. Mg²⁺ supplement reduced the effects of QNs in vitro^[14].

In conclusion, concentrations of CPFX in serum and cartilage could provide a better basis for a reasonable risk assessment.

REFERENCES

1 Cuzzolin L, Fanos V. Safety of fluoroquinolones in paediatrics. Expert Opin Drug Saf 2002; 1: 319-24.

2 Dominique G, Florence M. Fluoroquinolones in paediatrics. Paediatr drugs 2001; 3: 365-77.

3 Stahlmann R, Lode H. Toxicity of quinolones. Drugs 1999; 58 Suppl 2: 37-42.

4 Stahlmann R. Children as a special population at risk —quinolones as an example for xenobiotics exhibiting skeletal toxicity. Arch Toxicol 2003; 77: 7-11.

5 Stahlmann R, Zippel U, Forster C, Schwabe R, Shakibaei M, Merker HJ, et al. Chondrotoxicity and toxicokinetics of sparfloxacin in juvenile rats. Antimicrob Agents Chemother 1998; 42: 1470-5.

6 Kato M, Takada S, Ogawara S, Takayama S. Effect of levo-floxacin on glycosaminoglycan and DNA synthesis of cultured rabbit chondrocytes at concentrations inducing cartilage lesions in vivo. Antimicrob Agents Chemother 1995; 39: 1979-83.

7 Egerbacher M, Edinger J, Tschulenk W. Effects of enrofloxacin and ciprofloxacin hydrochloride on canine and equine chondrocytes in culture. Am J Vet Res 2001; 62: 704-8.

8 Zhang LR, Wang YM, Chen BY, Cheng NN. Neurotoxicity and toxicokinetics of norfloxacin in conscious rats. Acta Pharmacol Sin 2003; 24: 605-9.

9 Cheung WH, Lee KM, Fung KP, Leung KS. Growth plate chondrocytes inhibit neo-angiogenesis — a possible mechanism for tumor control. Cancer Lett 2001; 163: 25-32.

10 Enobakhare BO, Bader DL, Lee DA. Quantification of sulfated glycosaminoglycans in chondrocyte/alginate cultures, by use of 1,9-dimethylmethylene blue. Anal Biochem 1996; 243: 189-91.

11 Li GC, Chen CY, Wang YH. Comparative pharmacokinetic study on ciprofloxacin in healthy volunteers of the han, uigur and kazak nationalities. Chin J Clin Pharmacol 1997; 13: 145-8.

12 Nagai A, Miyazaki M, Morita T, Furubo S, Kizawa K, Fukumoto H, et al. Comparative articular toxicity of gareno-xacin, a novel quinolone antimicrobial agent, in juvenile beagle dogs. J Toxicol Sci 2002; 27: 219-28.

13 Hildebrand H, Kempka G, Schluter G, Schmidt M. Chondro-toxicity of quinolones in vivo and in vitro. Arch Toxicol 1993; 67: 411-5.

14 Egerbacher M, Wolfesberger B, Gabler C. In vitro evidence for effects of magnesium supplementation on quinolone-treated horse and dog chondrocytes. Vet Pathol 2001; 38: 143-8

15 Stahlmann R, Forster C, Shakibaei M, Vormann J, Gunther T, Merker HJ. Magnesium deficiency induces joint cartilage lesions in juvenile rats which are identical to quinolone-induced arthropathy. Antimicrob Agents Chemother 1995; 39: 2013-8.

16 Lozo E, Riecke K, Schwabe R, Vormann J, Stahlmann R. Synergistic effect of ofloxacin and magnesium deficiency on joint cartilage in immature rats. Antimicrob Agents Chemother 2002; 46: 1755-9.