Tsai Et Al-2011-Journal of Orthopaedic Research

Ciprooxacin Up-Regulates Tendon Cells to Express MatrixMetalloproteinase-2 with Degradation of Type I Collagen

Wen-Chung Tsai,1 Chih-Chin Hsu,2 Carl P.C. Chen,1 Hsiang-Ning Chang,1 Alice M.K. Wong,1

Miao-Sui Lin,1 Jong-Hwei S. Pang3

1Department of Physical Medicine and Rehabilitation, Chang Gung Memorial Hospital at Linkou, College of Medicine, Chang Gung University,Linkou, Taiwan, 2Department of Physical Medicine and Rehabilitation, Chang Gung Memorial Hospital at Keelung, College of Medicine, ChangGung University, Taiwan, 3Graduate Institute of Clinical Medical Sciences, Chang Gung University, Taiwan

Received 22 January 2010; accepted 19 May 2010Published online 2 July 2010 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jor.21196

ABSTRACT: Ciprooxacin-induced tendinopathy and tendon rupture have been previously described, principally affecting the Achillestendon. This study was designed to investigate the effect of ciprooxacin on expressions of matrix metalloproteinases (MMP)-2 and -9,tissue inhibitors of metalloproteinase (TIMP)-1 and -2 as well as type I collagen in tendon cells. Tendon cells intrinsic to rat Achillestendon were treated with ciprooxacin and then underwent MTT (tetrazolium) assay. Real-time reverse-transcription polymerase chainreaction (RT-PCR) and Western blot analysis were used, respectively, to evaluate the gene and protein expressions of type I collagen, andMMP-2. Gelatin zymography was used to evaluate the enzymatic activities of MMP-2 and -9. Reverse zymography was used to evaluateTIMP-1 and -2. Immunohistochemical staining for MMP-2 in ciprooxacin-treated tendon explants was performed. Collagen degradationwas evaluated by incubation of conditioned medium with collagen. The results revealed that ciprooxacin up-regulated the expressionof MMP-2 in tendon cells at the mRNA and protein levels. Immunohistochemistry also conrmed the increased expressions of MMP-2 inciprooxacin-treated tendon explants. The enzymatic activity of MMP-2 was up-regulated whereas that of MMP-9, TIMP-1 or TIMP-2was unchanged. The amount of secreted type I collagen in the conditioned medium decreased and type I collagen was degraded afterciprooxacin treatment. In conclusion, ciprooxacin up-regulates the expressions of MMP-2 in tendon cells and thus degraded type Icollagen. These ndings suggest a possible mechanism of ciprooxacin-associated tendinopathy. 2010 Orthopaedic Research Society.Published by Wiley Periodicals, Inc. J Orthop Res 29: 6773, 2011

Keywords: tendon; ciprooxacin; matrix metalloproteinase; collagen

The uoroquinolone class of antibiotics (e.g.,ciprooxacin, levooxacin, moxioxacin) has beenused to treat a wide range of infections. It was reportedin the literature that quinolone-induced tendinopa-thy or even tendon rupture principally affected theAchilles tendon.1,2 However, the mechanisms by whichciprooxacin predisposes tendinopathy or even tendonrupture remain to be investigated.

In animal studies with uoroquinolone-treated rats,disorganization of the extracellular matrix (ECM),inammation of the paratenon and degenerativechanges in tendon cells have been demonstrated.3,4

Besides, uoroquinolone class of antibiotics has beendocumented to exert a number of effects on vari-ous cell types in vitro, including reduced expressionof some ECM proteins,5,6 decreased mitochondrialactivity,6 enhanced matrix metalloproteinase (MMP)expression5,7 noncytotoxic inhibition of tendon cellproliferation5,8 and inhibition of tendon cell migration.9

Further matrix degradation or repeated micro-traumato a tendinopathic tendon might inevitably lead to atendon rupture.

The basic constituent of a tendon is collagen, whichaccounts for 70% of the dryweight of a tendon.10 Approx-imately 90% of collagen in normal tendons is type Iand less than 10% is type III collagen.11 Type I colla-gen is organized into brils grouped in parallel to form

Correspondence to: Jong-Hwei S. Pang (T: 886-3-2118800,ext. 3482; F: 886-3-3280170; E-mail: [email protected]) 2010 Orthopaedic Research Society. Published by Wiley Periodicals, Inc.

organized bundles while type III collagen is almost com-pletely conned to the endotendineum which surroundsthe bundles.12 Tendon cells (broblasts), which are itsbasic cellular component of a tendon, are the sourceof collagen production, protein mediators of repair, andmatrix proteoglycans.10 It appears that the physiologicresponse of tendon cells to trauma induces productionof both types I and III collagen.13 The biomechanicalproperties of a tendon are primarily a feature of theECM (mainly collagen), which is in a state of dynamicequilibrium between synthesis and degradation.14

Gelatinase such as MMP-2 and MMP-9 are MMPswith collagenolytic activity.15,16 The activity of MMPis inhibited by tissue inhibitors of MMPs (TIMPs)17,18

and the balance in MMPs and TIMPs regulated tendonremodeling. The expression of MMP-2 was up-regulatedin Achilles tendinopathy and MMP-9 expression wasalso up-regulated in the ruptured area of Achillestendon.19,20 It is concerned that the potential combi-nation of increased local matrix-degrading activity byenhanced MMP expression and/or decreased ECM pro-duction in tendon cells after ciprooxacin treatmentmight induce the occurrence of tendinopathy or tendonrupture. The aim of this study is to investigate the effectof ciprooxacin on expressions of type I collagen, MMP-2, MMP-9, TIMP-1, and TIMP-2 of tendon cells.

METHODSAll procedureswere approved by the Institutional Animal Careand Use Committee.

JOURNAL OF ORTHOPAEDIC RESEARCH JANUARY 2011 67

68 TSAI ET AL.

Primary Culture of Rat Achilles Tendon CellsTendon cells were obtained from SpragueDawley rats(200250 g) as previously described.21 Samples from passages2 to 4, with proper growth rate and normal broblast shape,were used. All the following experiments were performed atleast in triplicate.

MTT (3-[4,5-Dimethylthiazol-2-yl]-2,5-DiphenyltetrazoliumBromide) AssayTendon cells were left untreated or treated with 5, 10, 20,and 50g/mL ciprooxacin. Cell survival was determinedby MTT assay 24h after treatment. MTT (50g/mL) wasadded and incubated at 37C for 1h. Then, the MTT solu-tion was discarded and 1mL DMSO was added to dissolveformazan crystals. Optical density at 570nm (OD 570nm) inaliquots was read using a spectrophotometer (NanoDrop 1000,Thermo Scientic, Wilmington, DE). Percentages of OD valuefor ciprooxacin treated cells relative to the control were cal-culated.

Quantitative Real-Time RT-PCR AnalysisTotal RNA was extracted from tendon cells using solu-tion D (1mL solution D/107 cells). Subsequently, total RNAwas extracted with phenol and chloroform/isoamyl alcohol(49:1) to remove proteins and genomic DNAs. Comple-mentary (c) DNA was synthesized using 1mg total RNAin a 20mL volume RT reaction mix containing 0.5mgof random primers, 0.8mM dNTP, 0.1M DTT, and 1Lrst strand buffer. Quantitative real-time RT-PCR was per-formed using an SYBR Green and MxPro-Mx3000P QPCRmachine (Stratagene, NeoMarkers, Fremont, CA). Aliquots(20ng) of cDNA were used for each quantitative PCR, andeach reaction was run in triplicate. The following primerswere used: GAPDH: 5-TTCATTGACCTCAACTACAT-3 (for-ward) and 5-GAGGGGCCATCCACAGTCTT-3 (backward),type I collagen: 5-TGGAGACAGGTCAGACCTG-3 (forward)and 5-TATTCGATGACTGTCTTGCC-3 (reverse), as wellas MMP-2: 5-GGAAGCATCAAATCGGACTG-3 (forward)and 5-GGGCGGGAGAAAGTAGCA-3(reverse). Relative geneexpressions between experimental groups were determinedusing MxPro software (Stratagene) and GAPDH was used asan internal control. All real-time RT-PCRs were performed intriplicate, and changes in gene expressions were reported asmultiples of increases relative to the untreated controls.

Western Blot AnalysisThe levels of MMP-2 in the conditioned medium were ana-lyzed by Western blot analysis. The methods were describedas previous study.9 Primary antibodies such as mouse mono-clonal antibody against tubulin, type I collagen, and MMP-2were used. The semi-quantitative measurement of the banddensity was calculated by using 1D Digital Analysis Software(Kodak Digital ScienceTM, Eastman Kodak, Rochester, NY).The band densities were normalized to relative cell numberwith the results being band density divided by the percent-age of cell number from the results of corresponding MTTassay. Normalized data were expressed as 100% in controlgroup.

Gelatin ZymographyMMP-2 and -9 in conditioned mediumwere detected by gelatinzymography. It was performed under nonreducing conditionsusing a 7.5 % SDSpolyacrylamide gel containing 2mg/mLgelatin (Mini-PROTEAN II system, Bio-Rad Laboratories Ltd,Hempstead, UK). Gels were washed in 2.5% Triton X-100 to

remove SDSand allow renaturation ofMMPs, then transferredto 50mM Tris (pH 7.5), 5mM CaCl2, 1mM ZnCl2 and incu-bated at 37C for 18h. After staining with Coomassie brilliantblue R250 (Bio-Rad Laboratories, Hercules, CA), pro-MMPsand active MMPs result in white lysis bands, due to gelatindegradation.

Reverse ZymographyReverse zymography was performed similarly as zymographywith the exception that conditioned medium which containedactivities of MMP-2 and -9, was included in the gel mix exceptgelatin. All washes and incubations procedures were the sameas for zymography. TIMP-1 and -2, which inhibited gelatindigestion byMMP-2 and -9, appeared as dark bands on a whitebackground.

Ex Vivo Experimental Design and ImmunohistochemistryAchilles tendon (six tendons from three rats) was choppedinto two pieces and put separately into two dishes. Culturemediummade of Dulbeccos modied Eagles medium (DMEM;HyClone, Logan, UT), with 10% FBS (Cansera, Rexdale, ON,Canada), 100U/mL penicillin, and 100mg/mL streptomycin inthe presence or absence of ciprooxacin was added for 24h.The parafn-embedded tissue sections were stained by thestandard immunoperoxidase staining procedure. In brief, theparafn embedded tendon blocks were cut into 5m sections,de-parafnized, washed and then sequentially blocked forendogenous peroxidase activity with 3%H2O2, and nonspecicantibody binding sites with 1% BSA and 1% goat serum. Afterthree washings in PBS, the tissue sections were incubated for1h with mouse monoclonal antibodies against MMP-2 (Neo-Marks, Fremont, CA) diluted in blocking solution. Negativecontrol was performed following the same procedures exceptincubation with primary antibody. The signal was detectedwith DAKO labeled streptavidinbiotin system and colordevelopment was performed by incubation with diaminobenzi-dine substrate-chromogen (DAKO, Via Real, Carpinteria, CA)for 5min.

Collagen DegradationThe condition media of tenocytes treated with or without vari-ous concentrations of ciprooxacin (5, 10, 20, and 50g/mL)were collected and then mixed with collagen (0.5mg/mL)extracted from tendon (1:1). After incubation in 37C for 24h,the reaction products were revealed by nondenaturing gel elec-trophoresis. The collagen degradation, as demonstrated by thedecreased amount (band density) of collagen, was observedafter the gel was stained with Coomassie blue.

Statistical AnalysisAll nonparametric data obtained by MTT assay and den-sitometric analysis were expressed as the meanSEM.Ciprooxacin-treated and control cells were compared usingthe KruskalWallis test. The MannWhitney test was used todetermine the signicance of differences. p values less than0.05 were considered signicant.

RESULTSThe Effect of Ciprooxacin on Tendon Cell ViabilityMTT data revealed that ciprooxacin reduced relativeOD 570nm in a dose-dependent manner. The OD valuerelative to control group was 100.11.6%, 90.80.2%,80.84.4%, and 74.52.0% for cultures treated with 5,10, 20, and 50g/mL ciprooxacin, respectively. These

JOURNAL OF ORTHOPAEDIC RESEARCH JANUARY 2011

CIPROFLOXACIN MODULATES MMPS OF TENDON CELLS 69

Figure 1. The results of real-time RT-PCR (A) and densitometricanalysis of the results of Western blot analysis normalized to cellnumber (MTT result) (B) (*p

70 TSAI ET AL.

Figure 2. (A) Zymography of conditioned medium (CM) revealed that the enzymatic activities of MMP-2 (lower band: 72kDa) andMMP-9 (upper band: 95kDa). (B) Densitometric analysis of MMP-2 (*p


Figure 4. Immunohistochemical staining for MMP-2. Tendon cells in tendon explants treated with 50 g/mL ciprooxacin (D) revealingmore brown staining within the cytoplasm as compared to tendon cells in control explants (C) (A and B, negative: negative control).

Figure 5. (A) Western blot analysis of tubulin (as an internal control) and type I collagen in cytosol and conditioned medium whichwere identied at 57 and 95kDa, respectively. (B) Real-time RT-PCR revealed the mRNA expression of type I collagen. (C) Densitometricanalysis of secreted type I collagen in conditioned medium normalization to cell number was decreased by ciprooxacin (*p

72 TSAI ET AL.

Figure 6. Type I collagen was degraded by conditioned mediumfrom cells treated with ciprooxacin. Two major bands revealed byCoomassie blue staining after nondenaturing gel electrophoresiswere indicated by arrow.

As compared with this previous study, the presentstudy documented that ciprooxacin could directly up-regulate expressions of MMP-2 without the presenceof IL-1. Furthermore, to our knowledge, the resultof immunohistochemical staining for MMP-2 in thisstudy is the rst one to document the ex vivo effectof ciprooxacin on up-regulating MMP-2. Besides, thisstudy specically documents that type I collagen mightbe degraded by MMPs which are up-regulated byciprooxacin. Because uoroquinolones may also stimu-late inammatory pathways in or around the tendon,30

the combined effect on tendon matrix turnover mayaccount for the mechanisms of tendinopathy in somepatients treated with ciprooxacin.

An analysis of the results obtained from this studysuggests that enhanced MMP expression might signi-cantly compromise the integrity of the tendon ECM andthus induce the occurrence of tendinopathy or tendonrupture. The peak serum concentrations of ciprooxacingiven orally or intravenously were reported to rangefrom 0.5 to 10g/mL3134 and 5g/mL ciprooxacin wasthe initial concentration to induce the expressions ofMMP-2. Although, the concentration of ciprooxacin intendon tissue after standard dosing regimens remainsunknown, the result of this study suggest a potentiallink between ciprooxacin-associated tendinopathy andincreased dosage of ciprooxacin.

Because of the small size of the explants (about 0.5 cmin length, 0.2 cm in diameter), it is difcult to perform abiomechanical test to evaluate the changes in mechani-cal properties of a tendon after ciprooxacin treatment.Further animal studies to investigate if there is a dete-rioration of the mechanical properties of a tendon afterciprooxacin treatment are needed to validate the nd-ings of this study.

In conclusion, ciprooxacin up-regulates the expres-sion of MMP-2 in tendon cells with concomitantdegradation of type I collagen. These ndings providenovel molecular mechanisms of ciprooxacin-inducedtendinopathy or tendon rupture.

ACKNOWLEDGMENTSThe author thank the National Science Council, Taiwan fornancially support this research.

REFERENCES1. Piertte C, Royer RJ. 1996. Tendon disorders with uoro-

quinolones. Therapie 51:419420.2. Zabraniecki L, Negrier I, Vergne P, et al. 1996. Fluo-

roquinolone induced tendinopathy: Reports of 6 cases. JRheumatol 23:516520.

3. Kato M, Takada S, Kashida Y, et al. 1995. Histologi-cal examination on Achilles tendon lesions induced byquinolone antibacterial agents in juvenile rats. ToxicolPathol 23:385392.

4. Shakibaei M, Stahlmann R. 2001. Ultrastructure of Achillestendon from rats after treatment with eroxacin. Arch Toxi-col 75:97102.

5. WilliamsRJ, Attia E,Wickiewicz TL, et al. 2000. The effect ofciprooxacin on tendon, paratenon, and capsular broblastmetabolism. Am J Sports Med 28:364369.

6. Bernard-Beaubois K, Hecquet C, Hayem G, et al. 1998. Invitro study of cytotoxicity of quinolones on rabbit tenocytes.Cell Biol Toxicol 14:283292.

7. Corps AN, Harrall RL, Curry VA, et al. 2002. Ciprooxacinenhances the stimulation of matrix metalloproteinase 3expression by interleukin-1 in human tendon-derivedcells: A potential mechanism of uoroquinolone-inducedtendinopathy. Arthritis Rheum 46:30343040.

8. Tsai WC, Hsu CC, Tang FT, et al. 2008. Ciprooxacin medi-ated cell proliferation inhibition and G2/M cell cycle arrestin tenocytes. Arthritis Rheum 58:16571663.

9. Tsai WC, Hsu CC, Chen HC, et al. 2009. Ciprooxacin-mediated inhibition of tenocyte migration and down-regulation of focal adhesion kinase phosphorylation. Eur JPharmacol 607:2326.

10. OBrien M. 1992. Functional anatomy and physiology of ten-dons. Clin Sports Med 11:505520.

11. Amiel D, Frank C, Harwood F, et al. 1984. Tendons andligaments: A morphological and biochemical comparison. JOrthop Res 1:257265.

12. Duance VC, Restall DJ, Beard H, et al. 1977. The loca-tion of three collagen types in skeletal muscle. FEBS Lett79:248252.

13. Maffulli N, Ewen SW, Waterston SW, et al. 2000. Tenocytesfrom ruptured and tendinopathic Achilles tendons producegreater quantities of type III collagen than tenocytes fromnormal Achilles tendons. An in vitro model of human tendonhealing. Am J Sports Med 28:499505.

14. Jones GC, Corps AN, Pennington CJ, et al. 2006. Expressionproling of metalloproteinases and tissue inhibitors of met-alloproteinases in normal and degenerate human Achillestendon. Arthritis Rheum 54:832842.

15. Nagase H, Woessner JF, Jr. 1999. Matrix metallopro-teinases. J Biol Chem 274:2149121494.

16. Aimes RT, Quigley JP. 1995. Matrix metalloproteinase-2 isan interstitial collagenase. Inhibitor-free enzyme catalyzesthe cleavage of collagen brils and soluble native type I col-lagen generating the specic 3/4- and 1/4-length fragments.J Biol Chem 270:58725876.

17. Gomez DE, Alonso DF, Yoshiji H, et al. 1997. Tissueinhibitors of metalloproteinases: Structure, regulation andbiological functions. Eur J Cell Biol 74:111122.

18. Goupile P, Jayson MIV, Valat J, et al. 1998. Matrix metal-loproteinases: The clue to intervertebral disc degeneration?Spine 23:16121626.

19. Alfredson H, Lorentzon M, Backman S, et al. 2003. cDNA-arrays and real-time quantitative PCR techniques in theinvestigation of chronic Achilles tendinosis. J Orthop Res21:970975.

20. Karousou E, Ronga M, Vigetti D, et al. 2008. Colla-gens, proteoglycans, MMP-2, MMP-9 and TIMPs in humanAchilles tendon rupture. Clin Orthop Relat Res 466:15771582.

21. Tsai WC, Pang JH, Hsu CC, et al. 2006. Ultrasound stimula-tion of types I and III collagen expression of tendon cell andupregulation of transforming growth factor beta. J OrthopRes 24:13101316.



22. Gotoh M, Hamada K, Yamakawa H, et al. 1997. Signi-cance of granulation tissue in torn supraspinatus insertions:An immunohistochemical study with antibodies againstinterleukin-1 beta, cathepsinD, andmatrixmetalloprotease-1. J Orthop Res 15:3339.

23. Fu SC, Chan BP, Wang W, et al. 2002. Increased expressionof matrix metalloproteinase 1 (MMP1) in 11 patients withpatellar tendinosis. Acta Orthop Scand 73:658662.

24. Riley GP, Curry V, DeGroot J, et al. 2002. Matrix metal-loproteinase activities and their relationship with collagenremodeling in tendon pathology. Matrix Biol 21:185195.

25. Lo IK,MarchukLL,HollinsheadR, et al. 2004.Matrixmetal-loproteinase and tissue inhibitor ofmatrixmetalloproteinasemRNA levels are specically altered in torn rotator cuff ten-dons. Am J Sports Med 32:12231229.

26. Choi HR, Seiji K, Kazuyoshi H, et al. 2002. Expressionand enzymatic activity of MMP-2 during healing process ofacute supraspinatus tendon tear in rabbits. J Orthop Res20:927933.

27. Bigg HF, Rowan AD, Barker MD, et al. 2007. Activity ofmatrix metalloproteinase-9 against native collagen types Iand III. FEBS J 274:12461255.

28. Yoshihara Y, Hamada K, Nakajima T, et al. 2001. Biochem-ical markers in the synovial uid of glenohumeral joints

from patients with rotator cuff tear. J Orthop Res 19:573579.

29. Demirag B, Sarisozen B, Ozer O, et al. 2005. Enhancementof tendon-bone healing of anterior cruciate ligament graftsby blockage of matrix metalloproteinases. J Bone Joint SurgAm 87:24012410.

30. Kashida Y, Kato M. 1997. Characterization ofuoroquinolone-induced Achilles tendon toxicity inrats: Comparison of toxicities of 10 uoroquinolones andeffects of anti-inammatory compounds. Antimicrob AgentsChemother 41:23892393.

31. Bergeron MG. 1989. The Pharmacokinetics and tissue pene-tration of the uoroquinolones. Clin Invest Med 12:2022.

32. Dan M, Golomb J, Gorea A, et al. 1986. Concentration ofciprooxacin in human prostatic tissue after oral adminis-tration. Antimicrob Agents Chemother 30:8889.

33. MacGown AP, White LO, Brown NM, et al. 1994. Serumciprooxacin concentrations in patients with severe sepsisbeing treated with ciprooxacin 200mg i.v. bd irrespective ofrenal function. J Antimicrob Chemother 33:10511054.

34. Shah A, Lettieri J, Kaiser L, et al. 1994. Comparative phar-macokinetics and safety of ciprooxacin 400mg i.v. thricedaily versus 750mg po twice daily. J Antimicrob Chemother33:795801.


Documents

Tsai Et Al-2011-Journal of Orthopaedic Research