sign in sign up
<<
Home , Glucosamine

Effects of glucosamine hydrochloride on the production of prostaglandin E2, nitric oxide and metalloproteases by and synoviocytes in H. Nakamura, A. Shibakawa, M. Tanaka, T. Kato, K. Nishioka

Department of Bioregulation, St. Marianna University School of Medicine, Kawasaki, Kanagawa, Japan

Abstract Objectives To determine the response of glucosamine hydrochloride on chondrocytes and synoviocytes in terms of

prostaglandin E2 (PGE2), nitric oxide (NO) and matrix metalloproteases (MMPs).

Methods Chondrocytes and synoviocytes were prepared from joint specimens of patients who underwent total knee arthroplasty for osteoarthritis (OA). Chondrocytes from patients with femoral neck fracture were served as a normal control. Culture cells were stimulated by 5 ng/ml of IL-1 and treated with various concentration of

glucosamine hydrochloride (from 1 g/ml to 500 g/ml). PGE2, NO, MMP-1, MMP-3, and MMP-13 levels were evaluated in the culture supernatant. Further, the expression of COX-2 mRNA was studied by semi- quantitative PCR.

Results With IL-1 stimulation, the levels of these mediators increased dramatically, except for NO from synoviocytes. After stimulation, levels of these mediators in OA chondrocytes were higher than synoviocytes and normal chondrocytes, and the level of MMP-3 was higher than those of MMP-1 and MMP-13. Glu- cosamine hydrochloride at a concentration of 100 g/ml suppressed PGE2 production, and partly suppressed NO production. It also suppressed the production of MMPs from normal chondrocytes and synoviocytes but not from OA chondrocytes.

Conclusion Glucosamine modulates the metabolism of chondrocytes and synoviocytes and its mode of action differs between cells and conditions.

Key words Glucosamine, osteoarthritis, chondrocytes, synovial cells, MMP, PGE2, NO, COX-2.

Clinical and Experimental Rheumatology 2004; 22: 293-299. Glucosamine forPGE2, NO, MMPproduction in OA/ H. Nakamura et al.

Hiroshi Nakamura, MD, PhD; Atsuyuki Introduction glucosamine on chondrocytes or articu- Shibakawa, MD; Michiaki Tanaka, DVM; Glucosamine is aminosugar that is lar (18-21). Here, we evaluat- Tomohiro Kato, MD, PhD; Kusuki formed from fructose in vivo by gluta- ed the effects of glucosamine hydro- Nishioka, MD, PhD. min-fructose-6-phosphate amidotrans- chloride on cultured cells derived from Supported in part by grants from the Jap- ferase and composes various glycos- normal cartilage, OA cartilage and OA anese Ministry of Education, Culture, aminoglycans, the major component of synovium in terms of production of Sports, Science and Technology. extracellular matrix such as articular P G E , NO, MMP-1, MMP-3 and Please address reprint requests and 2 cartilage (1). Hyaluronic acid and ker- MMP-13. correspondence to: Hiroshi Nakamura, MD, Institute of Medical Science, atan sulfate are gigantic polymers of St. Marianna University, 2-16-1, Sugao, the found in cartilage, Materials and methods Miyame-ku, Kawasaki, 216-8512, Japan. both of which consist of N-acetyl glu- Ch o n d r ocyte and synoviocyte culture E-mail [email protected] cosamine and galactose or glucuronic Chondrocytes and synovial cells were Received on August 25, 2003; accepted in acid. Metabolism of glycosaminogly- prepared from surgical specimens ob- revised form on February 6, 2004. can together with that of collagen fibers tained during arthroplasty for OA and © Copyright CLINICAL AND EXPERIMEN- is important in maintaining matrix of femoral neck fracture. As articular car- TAL RHEUMATOLOGY 2004. articular cartilage and the disruption of tilage from the latter samples was ma- balance between synthesis and break- croscopically and microscopically nor- down causes osteoarthritis (OA). mal, they were used as the normal con- Biological and chemical factors as well trols. OApatients (n =7) had an average as mechanical factors are involved in age of 70.1 (range 50 - 77 yrs) and were the pathogenesis of OA. Among vari- diagnosed as having primary OA. Arti- ous chemical mediators, interleukin-1 cular changes were advanced and the (IL-1) has been revealed to play a cen- patients were being treated with anal- tral role in degradation of articular car- gesics. Fracture patients (n = 6) had an tilage (2); it is produced by mononu- average age of 76.6 yrs (range 70-83). clear cells in arthritic synovium (3) and Cartilage was carefully cut from the chondrocytes (4). IL-1b stimulates the subchondral bone. proliferation of synoviocytes (5) and en- Fragments of cartilage and synovial tis- hances the production of various chem- sue were minced to fine pieces follow- ical mediators such as IL-6 (6), prosta- ed by digestion in Dulbecco’s modified

glandin E2 (PGE2) (4) and nitric oxide Eagle’s medium (DMEM; Gibco BRL, (NO) (7) from synoviocytes. IL-1b also NY) with 1 mg/ml of bacterial collage-

stimulates the production of PGE2 (8), nase (Sigma, MO). The cells were fil- NO (9) and metalloproteases (10) from tered, washed, and seeded into 75 cm2 chondrocytes. flasks (Becton Dickinson Labware, Matrix metalloproteases (MMPs) are Bedford, UK). The cells were cultured important molecules that induce carti- in DMEM containing 10% fetal calf lage depletion in OA. MMP-1 and serum (FCS; Gibco BRL,NY), 100 MMP-13 cleave predominantly type II units/ml penicillin (Gibco BRL), and collagen (11), and MMP-3, also known 100 mg/ml streptomycin (Gibco BRL) as stromelysin-1, degrades proteogly- and cells from the primary cultures cans (12). Activated chondrocytes in were used for the experiments. OA produce these MMPs and degrade After the cells became confluent in 24- extracellular matrix surrounding them. well plates, medium was changed to

P G E2, an inflammatory mediator, is 2.5% FCS/DMEM and the cells were involved in the homeostasis of chondro- stimulated with 5 ng/ml of IL-1b (R o s c h , cytes (13) and enhances (14). Mannheim, Germany) in the presence -derived NO directly breaks of 1, 10 or 100 mg/ml of glucosamine down cartilage matrix and induces hydrochloride (Yaizu Marine Products, chondrocyte apoptosis (15,16 ) . Japan) for 24 hours. For synoviocytes, glucosamine has been 5, 50 and 500 mg/ml of glucosamine used as a supplement for the treatment hydrochloride were used as they were of osteoarthritis for decades, possibly considered to be less responsive to glu- because the extracellular matrix was be- cosamine. Supernatants were then col-

lieved to wear out in OA c a r t i l a g e . lected to measure the levels of PGE2, Many clinical results were reported (re- NO, MMP-1, -3, and -13. The cells were viewed in 17) and some in vitro studies harvested to detect COX-2 by semi- have been published on the effects of quantitative RT-PCR and Western blot.

294 Glucosamine for PGE2, NO, MMPproduction in OA/ H. Nakamura et al.

Measurement of PGE2, NO, MMP-1, phosphate) and sonicated for 3 seconds Wilcoxon signed rank test for compar- MMP-3 and MMP-13 on ice following centrifugation at 3000 isons of paired values, and the Man-

PGE2 and NO were assessed by a PGE2 rpm for 5 minutes. Supernatant was dis- Whitney U test for inter-group compar- EI A kit (Cayman Chemical Co., MI) and solved in the same amount of sample isons. a Nitrite colorimetric assay kit (COXIS buffer (0.1M Tris, 4% SDS, 12 mM International, Inc., OR), respectively. DTT, 20% glycerol, 1% BPB) and boil- Results

MMP-1, MMP-3, and MMP-13 were ed for 5 minutes. Electrophoretically IL- 1 stimulated production of PGE2, determined by ELISA or EIA k i t s analyzed proteins (15 µg/lane) on NO, MMP-1, MMP-3, and MMP-13 (MMP-1 and MMP-13; Amersham Phar- 12.5% SDS-PAGE gel were transferred With IL-1b stimulation, the level of all macia Bioteck, UK, MMP-3, Daiichi onto Nitrocellulose Membrane (Bio- mediators from chondrocytes and syn- Kagaku Yakuhin, Japan). Precursor and RAD, CA). The membrane was block- oviocytes increased dramatically, except active MMP as well as active complex ed in 5% skim milk for 1 hour at room for NO from synoviocytes. These levels with tissue inhibitor of metallopr o t e - temperature and subsequently immun- and the response to IL-1b were higher in ase-1 (TIMP-1) are measured with these oblotted with anti-COX-2 antibody OA chondrocytes than in normal chon- systems. (Cayman Chemical Co.) at a 1000 dilu- drocytes and synoviocytes. It is worth tion and then incubated with anti-rab- noting that the net level of MMP-3 in Preparation of mRNAand semi- bit-IgG antibody conjugated with HRP stimulated OA chondrocytes was ex- quantitative PCR (Nichirei, Japan) at a 10,000 dilution tremely high compared to those of COX-2 mRNA was quantified using a for 1 hour. The membrane was wa s h e d MMP-1 and MMP-13 (Fig. 1). Gene Specific Relative RT-PCR kit 3 times between each step. Th e n t h e

(Ambion,Inc, UK) following the manu- membrane was developed using an Effects of glucosamine on PGE2 f a c t u r e ’s instruction. Briefly, mRNA ECLsystem (Amersham, IL). production was extracted from 106 chondrocytes Prior to assay of the effects of glucos- using Isogen (Nippon Gene, Japan). Statistical analysis amine hydrochloride on the production cDNA was synthesized in a 20 ml reac- Each value was from individual samples. of various mediators, its effect on cell tion mixture containing 5 mg of total Values were expressed as the mean ± viability was evaluated by MTT assay RNA, 2.5 mM of each dNTP, 1 mM of SE. Statistical analysis included the (22). Cell viability was not affected at random hexamer primers, 40 units of ribonuclease inhibitor (RNasin; Toyo- bo, Japan), and 200 units of Superscript II RT (Gibco BRL) at 42°C for 2 hours. The resulting cDNAs were amplified in 50 ml reaction mixture containing COX- 2 and 18S primer pairs (included in the kit), 2.5 mM of each dNTPand 0.3ul of Taq polymerase (Takara, Japan) at 94° C for 30 seconds, 59°C for 30 seconds, and 72°C for 30 seconds for 24 cycles in a PCR thermal cycler (Perkin-Elmer, CT). PCR products were electrophor- esed on 2% agarose gels, stained with ethidium bromide (10 mg/ml), and the intensity of the bands were quantified by a fluoro-image analyzer, FLA-2000 (Fuji Film, Japan). The result was ex- pressed as the intensity of COX-2 PCR products against those of 18S.

Western blot for COX-2 Normal chondrocytes in 24-well dish were stimulated by 5 ng/ml of IL-1b, in the presence of 100 mg/ml glucosamine hydrochloride or 1 mM . Af- ter 24 hours, cells were dissolved in Fig. 1. PGE2, NO, MMP-1, MMP-3 and MMP-13 values in the supernatant of normal chondrocytes (n lysis buffer (20 mM Tris, 250 mM NaCl, = 6), OAchondrocytes (n = 7) and OAsynoviocytes (n = 5) at baseline and after stimulation with IL- 10 1%NP-40, 1mM DTT, 10 mM NaF, 1b. Except for NO in normal chondrocytes and synoviocytes, IL-1b dramatically stimulated the pro- duction of these mediators. Blank bars indicate the baseline and dotted bars indicate the stimulation. 2mM Na3VO4, 10 mM pyro-

295 Glucosamine forPGE2, NO, MMPproduction in OA/ H. Nakamura et al. concentrations ranging from 1 to 500 mg/ml (data not shown).

PGE2 production was enhanced by IL-1 stimulation in normal chondrocytes and OA synoviocytes as well as OA chon- drocytes. The PGE2 level in IL-1 stimu- lated OA chondrocyte cultures was 10 times higher than that in unstimulated ones. PGE2 production was suppressed by 100 mg/ml glucosamine in normal and OAchondrocytes, whereas dexam- ethasone and diclofenac, which were tested as positive controls, almost com- pletely inhibited stimulated PGE2 pro- duction. IL-1b enhanced the expression of COX-2 mRNA, but glucosamine had no effect on this response. Dexam- ethasone suppressed the expression of COX-2 mRNA, whereas diclofenac en- hanced its expression in OA chondro- cytes, which may reflect a feedback re- gulation caused by diclofenac-mediat- ed COX-2 inhibition (Fig.2). This re- Fig. 2. Production of PGE2 and expression of COX-2 mRNA in normal chondrocytes (n = 6), OA sult was also supported by Western blot chondrocytes (n = 7) and OA synoviocytes (n = 5). Cells were stimulated with IL-1b (5 ng/ml) and analysis which showed that COX-2 treated with glucosamine hydrochloride (1-100 mg/ml), dexamethasone (0.1 mM) or diclofenac (1 expression was enhanced by IL-1b, and mM). Glucosamine (100 mg/ml) suppressed the production of PGE2 in normal and OAchondrocytes. In that addition of 100 mg/ml glucosamine synoviocytes, 500 mg/ml of glucosamine showed suppressed PGE2 production but this change was not had no effect whereas the addition of significant. The expression of COX2 mRNAwas not affected by glucosamine, whereas dexamethasone suppressed it and diclofenac enhanced its expression in normal chondrocytes. Statistical analysis was diclofenac slightly increased the ex- performed using the Wilcoxon signed rank test. *p < 0.05. N: normal chondrocytes, OA: OAchondro- pression (Fig. 3). cytes, S: OAsynoviocytes. Gluco: Glucosamine, Dex: Dexamethasone, Diclo: Diclofenac

Effects of glucosamine on NO production Glucosamine slightly suppressed the production of NO from IL-1 treated normal chondrocytes, whereas it did not affect OA chondrocytes and OA synoviocytes (Fig. 4).

Effects of glucosamine on MMP production The production of MMP-1, MMP-3 and MMP-13 was measured in the supernatants of chondrocytes and syn- oviocytes. The net level of MMP-3 was Fig. 3. The effect of glucosamine hydrochloride and diclofenac on IL-1b stimulated COX-2 protein the highest among the MMPs tested expression was analyzed by Western immunoblot. COX-2 expression was enhanced by IL-1b stimula- tion; 100 mg/ml of glucosamine had no effect on the expression and 1 mM dicloefenac slightly and the level of MMP13 was the low- enhanced the expression. Repeated experiments showed the same results. est. All MMP levels were enhanced by IL-1b. After stimulation with IL-b, the production of MMP-1, -3 and -13 was suppressed by glucosamine at the con- include proinflammatory cytokines such partly suppressed by 100 mg/ml glu- centrations to 100 mg/ml in OA chon- as IL-1 and T N F, proteases such as cosamine in normal chondrocytes. In drocytes (Fig. 5). MMPs and cathepsins, and NO that in- OA synoviocytes, 500 mg/ml glucosa- duces chondrocyte apoptosis as well as mine suppressed the production of Discussion matrix damage. IL-1 also enhanced the MMP-1, -3 and -13. Constant suppres- Degradation of articular cartilage is dr i - production of PGE2. In this experiment, sion by dexamethasone was shown in ven by an imbalance of catabolic and the level of MMP-3 in the supernatant all cell types. MMP production was not anabolic factors in OA(23). The forme r of stimulated OA chondrocytes was 4

296 Glucosamine forPGE2, NO, MMPproduction in OA/ H. Nakamura et al.

mg/ml whereas its concentration in OA has been reported to be a little higher (from 13 mg/ml to 15 mg/ ml) (24, 25). On the contrary, MMP-1 and MMP-13 levels in the supernatant of OA chondrocytes (551.6 ng/ml and 12.8 ng/ml, respectively) were higher than those in OA synovial fluid (356.1 ng/ml and 5.27 ng/ml, respectively) ( 2 5 , 26). Overall, OA c h o n d r o c y t e s showed a higher sensitivity to the stim-

ulation of MMP, PGE2 and NO by IL-1 than normal chondrocytes did. T h i s Fig. 4. Production of NO in normal chondrocytes (n = 6), OAchondrocytes (n = 7) and OAsynovio- could be attributed to the fact that OA cytes (n = 5). Cells were stimulated with L-1b (5 ng/ml) and treated with glucosamine hydrochloride chondrocytes had a higher (about 2- (1-100 mg/ml) or dexamethasone (0.1 mM). Glucosamine (100 mg/ml) suppressed the production of fold) of IL-1 receptor com- NO only in normal chondrocytes. Statistical analysis was performed using the Wilcoxon signed-rank test. *p < 0.05. N: normal chondrocytes, OA: OA chondrocytes, S: OA synoviocytes. Gluco: Glu- pared with normal chondrocytes (27). cosamine, Dex: Dexamethasone. As the cartilage samples of OA were derived from an advanced stage of the disease in which arthroplasty is needed, the chondrocytes are supposed to have been exposed to the specific condition in the OA joint for a long time, which may have altered their characteristics.

PGE2 modulates cartilage degradation in the presence of IL-1 (28) and seems to have a bivalent effect on chondro-

cytes (29). PGE2 also induces inflam- mation and augments pain (14). The latter function is important with regard to the clinical effects of glucosamine on OA, as one of the major short-term e ffects of glucosamine is pain relief

(30). PGE2 is spontaneously released from OA chondrocytes and its produc- tion is enhanced by IL-1 (27). In this

study, the production of PGE2 from IL- 1-stimulated chondrocytes, especially chondrocytes derived from OA p a- tients, was higher than that from syno- viocytes. Thus, it is speculated that the

source of PGE2 in the OA joint was chondrocytes as well as synoviocytes. The suppressive effect of glucosamine

on PGE2 production might reflect its beneficial role in the treatment of OA.

PGE2 synthesis is regulated by the co- ordination of phospholipase A2 (PLA2), COX and PGE synthase (PGES) activi- ties (31,32). Nonsteroidal anti-inflam- matory drugs inhibit COX activity, whereas glucocorticoids suppress its Fig. 5. MMP-1, -3 and -13 levels in the normal chondrocytes (n = 6), OAchondrocytes (n = 7) and expression at the transcriptional level. OAsynoviocytes (n = 5). After stimulation with IL-1b (5 ng/ml), cells were treated with various con- centration (1-100 mg/ml) of glucosamine hydrochloride or dexamethasone (0.1 mM). Glucosamine Shikhman et al. reported that 10 mM (100 mg/ml) showed a suppressive effect on MMPproduction from normal chondrocytes and OAsyn- N-acetyl glucosamine suppressed COX- oviocytes. The Wilcoxon signed-rank test was used for the statistical analysis. *p < 0.05. N: normal 1 and COX-2, but did not observe sup- chondrocytes, OA: OAchondrocytes, S: OAsynoviocytes. Gluco: Glucosamine, Dex: Dexamethasone. pressive effects on the expression of

297 Glucosamine forPGE2, NO, MMPproduction in OA/ H. Nakamura et al.

MAP kinases or translocation of NFkB ment of OA, excess administration and its regulation by rabbit synoviocytes. J 21). On the other hand, Largo et al. could be harmful and concentrations of Rheumatol 1994; 21: 1892-8. 8. CAMPBELL IK, PICCOLI DS, HAMILTON JA: demonstrated that glucosamine sulfate up of 100 mg/ml are supposed to be Stimulation of human chondrocyte prosta- inhibited IL-1 induced NFkB activity rational to investigate clinical effects of glandin E2 production by recombinant hu- in a dose-dependent manner and sup- glucosamine. Moreover, as our data man interleukin-1 and tumour necrosis fac- pressed COX-2 expression at the con- originated from human samples, the tor. Biochim Biophys Acta 1990; 1051: 310-8 9. S TA D L E R J, STEFANOVIC-RACIC M, BIL- centration of 1 mg/ml (33). In these re- results are considered to reflect the LIAR TR et al.: Articular chondrocytes syn- ports, authors used different glucosa- practical clinical response of glu- thesize nitric oxide in response to cytokines mine derivatives with a higher concen- cosamine. and . J Immunol 1 9 9 1 ; tration than those used in this study. Our data show that glucosamine hydro- 147: 3915-20. 10. TE T L O W LC, ADLAM DJ, W O O L L E Y D E: Glucosamine hydrochloride did not af- chloride had pharmacological eff e c t s Matrix metalloproteinase and proinflamma- fect the expression of COX-2 mRNA on chondrocytes and synoviocytes by tory cytokine production by chondrocytes of and protein at 100 mg/ml, thus the sup- inhibition of IL-1 stimulated produc- human osteoarthritic cartilage: associations with degenerative changes. Rheum pressive effect of glucosamine hydro- tion of PGE2, NO, MMP-1, MMP-3 2001; 44: 585-94. chloride on PGE2 in this study should and MMP-13. Glucosamine hydrochlo- 11. BI L L I N G H U R S T RC, DAHLBERG L, IONES- be investigated further. Phospholipase ride (100 mg/ml) suppressed the MMP CU M et al.: Enhanced cleavage of type II activity might be involved in this pro- production of normal chondrocytes and collagen by collagenases in osteoarthritic ar- ticular cartilage. J Clin Invest 1997; 99: cess as Piperno et al. clearly showed synoviocytes but not of OA chondro- 1534-45. that glucosamine suppressed PLA2 cytes. It was revealed that glucosamine 12. O K A D A Y, SHINMEI M, TA N A K A O et al.: activity (18). modulates the metabolism of chondro- Localization of matrix metalloproteinase 3 Recently, Dodge et al. reported that 25 cytes and synoviocytes, and that its (stromelysin) in osteoarthritic cartilage and synovium. Lab Invest 1992; 66: 680-90. mM glucosamine sulfate suppressed mode of action differs between cells 13. FULKERSON JP, DAMIANO P: Effect of pro- MMP-3 from IL-1 stimulated OA ch o n - and experimental conditions. These re- staglandin E2 on adult pig articular cartilage drocytes (34) and Byron et al. demon- sults may reflect the mechanism of ac- slices in culture. Clin Orthop 1983; 179: 266- strated that glucosamine suppressed tion of glucosamine on pain relief and 9. 14. F E R R E I R A SH, NAKAMURA M, CASTRO MMP-1, -3 and -13 from LPS stimulat- the prevention of articular destruction. MSA: The hyperalgesic effects of prostacy- ed horse chondrocytes (35). In this clin and prostaglandin E2. P ro s t a g l a n d i n s study, glucosamine was shown to sup- References 1978; 16: 31-7. press MMP-1, -3 and -13 in normal 1. HUGENBERG ST, BRANDT KD, COLE C A: 15. HA S H I M O TO S, TAKAHASHI K, A M I E L D , E ffect of , , and COUTTS RD, LOTZ M: Chondrocyte apopto- chondrocytes and OA s y n o v i o c y t e s . on enzymes required by the chon- sis and nitric oxide production during experi- However, OA chondrocytes showed no drocyte for synthesis of . J mentally induced osteoarthritis. A rt h r i t i s response to glucosamine at 100 mg/ml. Rheumatol 1993; 20: 2128-33. Rheum 1998 ; 41: 1266-74. As OA chondrocytes intensely express 2. VAN DE LOO FA, JOOSTEN LA, VAN LENT PL, 16. PELLETIER JP, JOVANOVIC DV, LASCAU- ARNTZ OJ, VA N D E N B E R G W B: IL-1/NO/ COMAN V et al.: Selective inhibition of in- IL-1 receptor and are sensitive to IL-1 chondrocyte Role of interleukin-1, tumor ducible nitric oxide synthase reduces pro- stimulation (27), they are supposed to necrosis factor alpha, and interleukin-6 in gression of experimental osteoarthritis i n be less responsive to counteracting glu- cartilage metabolism and des- vi v o : possible link with the reduction in chon- truction. Effect of in situ blocking in murine drocyte apoptosis and caspase 3 level. Arthri - cosamine. antigen- and zymosan-induced arthritis. tis Rheum 2000; 43: 1290-9. Nitric oxide is also a catabolic factor Arthritis Rheum 1995; 38: 164-72. 17. MCALINDON TE, LAVALLEY MP, GULIN JP, for extracellular matrix in articular car- 3. FA R A H AT MN, YA N N I G, POSTON R, PA N- F E L S O N D T: Glucosamine and chondroitin tilage and glucosamine hydrochloride AY I G S: Cytokine expression in synovial for treatment of osteoarthritis: a systematic membranes of patients with rheumatoid arth- quality assessment and meta-analysis. JAMA slightly suppressed IL-1 induced NO, ritis and osteoarthritis. Ann Rheum Dis 1993; 2000; 283: 1469-75. while N-acetyl glucosamine dramati- 52: 870-5. 18. PIPERNO M, REBOUL P, HELLIO LE G R A- cally suppressed it (21). In other posi- 4. MO O S V, FICKERT S, MULLER B , W E B E R VERAND MP et al.: Glucosamine sulfate mo- tive reports, the concentrations of glu- U, SIEPER J: Immunohistological analysis of dulates dysregulated activities of human os- cytokine expression in human osteoarthritic teoarthritic chondrocytes in vitro. Osteoar - cosamine derivatives were considera- and healthy cartilage. J Rheumatol 1999; 26: thritis Cartilage 2000; 8: 207-12. bly higher than those used in this study; 870-9. 19. BASSLEER C, ROVATI L, FRANCHIMONT P: thus it is possible that a higher concen- 5. KUMKUMIAN GK, LAFYATIS R, REMMERS Stimulation of proteoglycan production by tration of glucosamine hydrochloride EF, CASE JP, KIM SJ, WILDER RL: Platelet- glucosamine sulfate in chondrocytes isolated derived growth factor and IL-1 interactions from human osteoarthritic articular cartilage might suppress the mediators that we in rheumatoid arthritis. Regulation of syn- in vitro. Osteoarthritis Cart i l a g e 1998; 6: investigated. However it should be not- oviocyte proliferation, prostaglandin produc- 427-34. ed that high-dose glucosamine (0.25 - tion, and collagenase transcription. J Im - 20. GOUZE JN, BORDJI K, GULBERTI S et al.: munol 1989; 143: 833-7. Interleukin-1beta down-regulates the expres- 25 mg/ml) impairs the metabolic activ- 6. GU E R N E PA, ZURAW BL, VAUGHAN JH, sion of glucuronosyltransferase I, a key en- ity of chondrocytes in terms of proteo- CA R S O N DA, LOTZ M: Synovium as a source zyme priming biosynthe- glycan synthesis, cell viability and cell of interleukin 6 in vitro. Contribution to local sis: influence of glucosamine on interleukin- death (36). Though there have been no and systemic manifestations of arthritis. J 1beta-mediated effects in rat chondrocytes. Clin Invest 1989; 83: 585-92. Arthritis Rheum 2001; 44: 351-60. reports describing the optimal clinical 7. STEFANOVIC-RACIC M, STADLER J, GEOR- 21. SHIKHMAN AR, KUHN K, ALAAEDDINE N, dosage of glucosamine for the treat- GESCU HI, EVANS CH: Nitric oxide synthesis LOTZ M: N-acetylglucosamine prevents IL-1

298 Glucosamine forPGE2, NO, MMPproduction in OA/ H. Nakamura et al.

beta-mediated activation of human chondro- BATTISTAJ et al.: The interleukin-1 receptor mal glutathione-dependent, inducible en- cytes. J Immunol 2001; 166: 5155-60. in normal and osteoarthritic human articular zyme, constituting a potential novel drug tar- 22. D E N I Z O T F, LANG R: Rapid colorimetric ch o n drocytes. Identification as the type I get. Pr oc Natl Acad Sci USA 1999; 96: 7220-5. assay for cell growth and survival. J Immunol receptor and analysis of binding kinetics and 33. LARGO R, A LVA R E Z - S O R I A MA, DIEZ-OR- Methods 1986; 89: 271–7. biologic function. Arthritis Rheum 1992; 35: TEGO I et al.: Glucosamine inhibits IL-1beta- 23. G O L D R I N G M B: The role of the chondro- 530-40. induced NFkappaB activation in human oste- cytes in osteoarthritis. Arthritis Rheum 2000; 28. HARDYMM, SEIBERT K, MANNING PT et al.: oarthritic chondrocytes. Osteoarthritis Carti - 43: 1916-26. Cyclooxygenase 2-dependent prostaglandin lage 2003; 11: 290-8. 24. RIBBENS C, ANDRE B, KAYE O et al.: Syn- E2 modulates cartilage proteoglycan degra- 34 DODGE GR, JIMENEZ SA: Glucosamine sul- ovial fluid matrix metalloproteinase-3 levels dation in human osteoarthritis explants. Arth - fate modulates the levels of aggrecan and are increased in inflammatory arthritides whe- ritis Rheum 2002; 46: 1789-803. matrix metalloproteinase-3 synthesized by ther erosive or not. Rheumatology (Oxford) 29. A B R A M S O N S B: The role of COX-2 pro- cultured human osteoarthritis articular chon- 2000; 39: 1357-65. duced by cartilage in arthritis. Osteoarthritis drocytes. Osteoarthritis Cartilage 2003; 11: 25. YOSHIHARA Y, NAKAMURA H, OBATA K et Cartilage 1999; 7: 380-1. 424-32. al.: Matrix metalloproteinases and tissue in- 30. NAKAMURAH, NISHIOKA K: Effects of glu- 35 BYRON CR, ORTH MW, V E N TA P J et al.: hibitors of metalloproteinases in synovial cosamine/chondroitin supplement on osteo- Influence of glucosamine on matrix metallo- fluids from patients with rheumatoid arthritis arthritis: Involvement of PGE2 and YKL-40. proteinase expression and activity in lipopo- or osteoarthritis. Ann Rheum Dis 2000; 59: Jpn J Rheum Surg 2002, 21: 175-84. lysaccharide-stimulated equine chondro- 455-61. 31. LESLIE CC: Properties and regulation of cy- cytes. Am J Vet Res 2003; 64: 666-71. 26. ITO T, UZUKI M, SHIMAMURA T, SAWAI T: tosolic phopholipase A2. J Biol Chem 1997; 36. D E MAT T E I M, PELLAT I A, PASELLO M e t Dynamics of matrix metalloproteinase 27: 16709-12. a l .: High doses of glucosamine-HCl have (MMP)-13 in the patients with rheumatoid 32. J A K O B S S O N P-J, THOREN S, MORGEN- detrimental effects on bovine articular carti- arthritis. Ryumachi 2002; 42: 60-9. STERN R, SAMUELSSON B: Identification of lage explants cultured in vitro. Osteoarthritis 27. MA RT E L - P E L L E T I E R J, MCCO L L U M R, DI- human prostaglandin E synthase: a microso- Cartilage 2002; 10: 816-25.

299