Effects of Transforming Growth Factor (beta) on the Production of ...

Download PDF
13 downloads 0 Views 360KB Size Report
Comment
EFFECTS OF TRANSFORMING GROWTH FACTOR b ON THE PRODUCTION. OF PROSTAGLANDIN E AND CASEINASE ACTIVITY OF UNSTIMULATED.
British Journal of Rheumatology 1997;36:729–734

EFFECTS OF TRANSFORMING GROWTH FACTOR b ON THE PRODUCTION OF PROSTAGLANDIN E AND CASEINASE ACTIVITY OF UNSTIMULATED AND INTERLEUKIN 1-STIMULATED HUMAN ARTICULAR CHONDROCYTES IN CULTURE F. W. FAWTHROP,*† A. FRAZER,* R. G. G. RUSSELL* and R. A. D. BUNNING‡ *Department of Human Metabolism and Clinical Biochemistry, Sheffield University Medical School, Sheffield S10 2RX and ‡Division of Biomedical Sciences and Health Research Institute, Sheffield Hallam University, Sheffield S1 1WB SUMMARY Transforming growth factor b (TGFb) has previously been shown to have actions on chondrocytes and cartilage both in vitro and in vivo which suggest a role in connective tissue repair. In particular, some of its actions have been shown to be antagonistic to those of interleukin 1 (IL-1). In this study, the effects of TGFb on prostaglandin E (PGE) production and caseinase activity, in the presence and absence of IL-1, in human articular chondrocytes were investigated. TGFb1 and TGFb2 were shown to modulate IL-1b-stimulated PGE production and caseinase activity. Both TGFb1 and b2 inhibited IL-1b-stimulated PGE production in the absence of serum and augmented it in the presence of serum. TGFb1 and TGFb2 inhibited IL-1-stimulated caseinase activity with and without serum. In general, the TGFbs had little or no effect on basal PGE or caseinase levels. TGFbs may be important modulators of chondrocyte metabolism, their effects on PGE production may depend on cytokine interactions; furthermore, their effects on caseinase activity may help prevent cartilage breakdown. K : Transforming growth factor b, Interleukin 1, Prostaglandin E, Caseinase activity, Chondrocytes.

T transforming growth factors b (TGFbs) are a family of homodimeric polypeptides with mol. wts of 025 kDa. There are at least five isoforms of TGFb, of which the best characterized are TGFb1 and TGFb2 [1, 2]. Although TGFbs 1 and 2 exhibit only 71% homology, they seem to share similar activities [2]. However, it has been shown that TGFb2 is a more active stimulator of chondrogenesis, proteoglycan synthesis and cell proliferation in vitro [3]. Our aim was to compare the actions of TGFbs 1 and 2 on chondrocyte metabolism in terms of prostaglandin E (PGE) production and caseinase activity. The TGFbs are multifunctional molecules and are thought to be involved in processes such as connective tissue repair [4]. In this context, TGFb has been shown to stimulate the synthesis of extracellular matrix components such as type I collagen and fibronectin [5], to inhibit the production by fibroblasts of proteinases capable of extracellular matrix breakdown, such as plasminogen activator and stromelysin, and to stimulate the production of inhibitors of plasminogen activator and metalloproteinases [1, 6–8]. TGFb has been shown to have a number of actions on chondrocytes and cartilage, some being antagonistic to those of interleukin 1 (IL-1). These include the inhibition of IL-1-stimulated protease activity, the reversal of IL-1-mediated inhibition of proteoglycan

synthesis in rabbit chondrocytes and the partial inhibition of IL-1-stimulated resorption of porcine cartilage in vitro [9, 10]. In organ cultures of bovine cartilage, TGFb stimulates proteoglycan synthesis and decreases its catabolism [11]. Human osteoarthritic cartilage has been shown to be more sensitive to TGFb than normal cartilage in terms of its ability to stimulate proteoglycan synthesis [12]. Depending on the culture conditions, TGFb may inhibit or stimulate the proliferation of rabbit chondrocytes and may stimulate proliferation in human chondrocytes [13, 14]. TGFb has also been shown to have a marked protective effect on cartilage destruction in vivo. Injection of TGFb together with IL-1 into murine knee joints prevented the cartilage destruction observed in injecting IL-1 alone [15]. Both IL-1 and TGFb are present in synovial fluids of patients with osteoarthritis and rheumatoid arthritis [16, 17], and may interact to modulate chondrocyte metabolism. There are a number of potential sources of synovial fluid TGFb. Rheumatoid synovium produces TGFb [18] and there is some evidence for the production of TGFb by human chondrocytes. mRNAs for TGFbs 1, 2 and 3 have been demonstrated in cultured human chondrocytes in addition to the production of TGFb by these cells in culture medium [19, 20]. Explants of human cartilage in culture have also been shown to produce TGFb [12]. In this study, we have investigated the effect of TGFb1 and 2, alone and in combination with IL-1, on prostaglandin production and caseinase activity in human chondrocytes. Caseinase activity is used as a measure of stromelysin activity [21]. PGE is a potent mediator of inflammation [22] which may also have actions on connective tissue cell

Submitted 10 May 1996; revised version accepted 13 December 1996. Correspondence to: R. A. D. Bunning, Division of Biomedical Sciences and Health Research Institute, Sheffield Hallam University, Pond Street, Sheffield S1 1WB. Present address: †Level D, Rotherham District General Hospital, Moorgate, Rotherham S60 3UD.

= 1997 British Society for Rheumatology 729

730

BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 7

metabolism, PGE2 stimulating both bone resorption and formation under certain conditions [23]. IL-1 stimulates the production of PGE2 by human chondrocytes [24]. In addition, IL-1 and TGFb in combination produce a synergistic stimulation of PGE2 production in human rheumatoid synovial cells [25]. However, little is known of the effects of TGFb alone or in combination with IL-1 on chondrocyte PGE production. Stromelysin is probably involved in both normal and pathological connective tissue turnover as it can degrade collagen types III, IV, V and IX, and proteoglycans [26], and can activate procollagenase [27]. IL-1 has been shown to stimulate the stromelysin activity of cultured human cartilage explants and chondrocytes [28]. Furthermore, stromelysin activity is increased in osteoarthritic cartilage [29]. Production of TGFb by chondrocytes and its actions on chondrocytes may be important in cartilage repair mechanisms both by limiting cartilage breakdown, and promoting new matrix synthesis and cellular proliferation. Also of importance are its interactions with other cytokines, such as IL-1, which are known to mediate cartilage destruction. Much of the previous work has been performed using animal cells and tissues, and to date there are few published studies of effects of TGFb on human articular chondrocytes. MATERIALS AND METHODS Cytokines Recombinant human (rh) TGFb1 was a gift from Dr E. Amento, Genentech Inc., San Francisco, USA, supplied at a concentration of 28.6 m and diluted in medium containing 1% bovine serum albumin (BSA). rhTGFb2, a gift from Sandoz AG, Basel, Switzerland, was obtained at a concentration of 4 × 10−5  and diluted as required in medium containing 1% BSA. rhIL-1b (108 U/mg) was generously donated by Dr A. Shaw, Glaxo Wellcome Research, Stevenage, Herts. Preparation and culture of adult human articular chondrocytes Macroscopically normal human articular cartilage was obtained from femoral heads, removed following trauma, and from the femoral and tibial condyles of above-knee amputation specimens. Samples were from a predominantly elderly age group, mean age 76 yr (range 54–93 yr), and are therefore relevant to mature adult cartilage. Chondrocytes were dispersed by sequential enzymatic digestion as previously described [30] and were cultured in Eagle’s minimum essential medium containing 10% fetal calf serum, 100 U/ml penicillin, 100 mg/ml streptomycin, 2 m glutamine (MEM) and kept in a humidified incubator at 37°C (5% CO2/95% air). Cells were used at first passage after 3–6 weeks in culture. Chondrocytes have been shown to maintain their phenotype, i.e. produce type II collagen, under these conditions [31]. For experimental purposes, chondrocytes were passaged in 24 well plates at a density of 3–4 × 104 cells/well and grown to confluence. After washing the cells three times with

phosphate-buffered saline, test substances were added in either serum-free medium containing 0.1% BSA or medium containing 1% fetal calf serum. For each experiment, chondrocytes from a single patient were used, control and test conditions being set up in quadruplicate wells. Each set of experiments was carried out on cells from at least three different donors. PGE assay PGE in cell supernatants was measured by radioimmunoassay as described previously [32] using an antiserum raised in rabbits against PGE2 (Sigma, Poole). This antibody does not discriminate between PGE1 and PGE2, hence results are expressed as PGE. Caseinase assay The non-specific caseinase assay using 14C-acetylated casein as substrate was used as previously described [21]. Latent enzyme was activated by the addition of 4-aminophenylmercuric acetate (0.8 m). One unit of stromelysin activity degrades 1 mg of casein per minute at 37°C to peptides soluble in 3% trichloroacetic acid. It has previously been shown that the casein-degrading activity produced by human chondrocytes is mainly due to a latent metalloproteinase, and therefore probably represents stromelysin [33]. Statistical methods Results are expressed as the mean 2 ... for each group. Statistical analysis was by analysis of variance using Scheffe’s method for multiple comparisons [34]. RESULTS TGFb1 or TGFb2, 10−13–10−7  in the presence or absence of 1% serum, had no significant effect on the basal PGE production of adult human chondrocytes as compared to controls. In all experiments, IL-1 markedly stimulated PGE levels. In the absence of serum, both TGFb1 and TGFb2, at 10−9–10−7 , significantly inhibited IL-1-stimulated PGE production. Since we found that the effects of TGFb1 and TGFb2 on PGE production were similar, results for TGFb2 only are shown (Fig. 1a). The inhibitory activity of TGFb1 or TGFb2 on IL-1stimulated PGE production was generally reduced or lost at low concentrations: 10−13–10−10 . Indeed, in some experiments, TGFb1 or TGFb2 at 10−13  enhanced IL-1-stimulated PGE production (Fig. 1b). In the presence of 1% serum, 10−9–10−7  TGFb1 or TGFb2 significantly enhanced IL-1-stimulated PGE production (Fig. 2a). This effect was lost at lower concentrations of TGFb1 (10−13–10−11 ) or TGFb2 (10−13–10−10 ) (Fig. 2b). In both the presence and absence of 1% serum, IL-1 significantly stimulated the caseinase activity of human chondrocytes. TGFb1 alone had no significant effect on basal caseinase levels, whereas TGFb2, in the absence of serum, produced a slight increase in basal levels and in the presence of serum a slight decrease which reached significance in some experiments. IL-1-stimulated caseinase activity was decreased by TGFb1 or

FAWTHROP ET AL.: TGFb EFFECTS ON PGE AND CASEINASE

731

F. 1.—The effect of rhTGFb2, 10−9–10−7  (a) and 10−13––10−10  (b), on the PGE production of human chondrocytes and its interaction with rhIL-1 (10 U/ml). Chondrocytes were incubated with MEM, containing 0.1% bovine serum albumin instead of fetal calf serum and test substances for 24 h. PGE was measured in cell supernatants by radioimmunoassay. Results are representative of at least three experiments and replicates, from the same donor, are expressed as mean 2 ... (n = 4), *P E 0.05, **P E 0.01 in comparison to the rhIL-1-treated group, Scheffe’s multiple range analysis.

TGFb2 at 10−9–10−7  in the presence or absence of serum (Fig. 3a and b). In the presence of serum, the lowering of IL-1-stimulated activity with TGFbs 1 and 2 did not always reach significance, as seen in Fig. 3b. In the absence of serum, 10−13  TGFb1 or TGFb2 enhanced the action of IL-1 on caseinase activity (Fig. 4). Since observations for TGFb1 or TGFb2 were similar, only the results for TGFb1 are shown. DISCUSSION TGFb1 has previously been identified in fetal and adult non-human cartilage [11, 35, 36], and cultured explants of human cartilage and human chondrocytes in culture have been shown to secrete TGFb into the medium [20], suggesting that TGFb may have an important role in the metabolism of normal human articular cartilage. TGFb1 and b2 had no significant effect on basal PGE

levels in the presence or absence of serum. However, in the absence of serum, both TGFb1 and b2 inhibited IL-1-stimulated PGE production, possibly the result of the downregulation of IL-1 receptors on chondrocytes by TGFb [37, 38]. In contrast, in the presence of 1% serum, both growth factors augmented IL-1-stimulated PGE production. The reason for this is uncertain, but is probably the result of a complex interaction(s) between IL-1 and/or TGFb with serum components. Platelet-derived growth factor (PDGF) and fibroblast growth factor, which are present in serum, have been shown to augment IL-1-stimulated PGE production in rabbit articular chondrocytes [39, 40]. The effect of PDGF is thought to be mediated by the induction of IL-1 receptors by PDGF. However, we have observed that PDGF inhibits IL-1-stimulated PGE production in human articular chondrocytes [41]. Another possibility may be that the active concentration of TGFb is reduced in the presence of serum by binding

F. 2.—The effect of rhTGFb2, 10−9–10−7  (a) and 10−13–10−10  (b), on the PGE production of human chondrocytes and its interaction with rhIL-1 (10 U/ml). Chondrocytes were incubated with MEM containing 1% fetal calf serum and test substances for 24 h. PGE was measured in cell supernatants by radioimmunoassay. Results are representative of at least three experiments and replicates, from the same donor, are expressed as mean 2 ... (n = 4), **P E 0.01 in comparison to the rhIL-1-treated group, Scheffe’s multiple range analysis.

732

BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 7

F. 3.—The effect of rhTGFb1 (10−9–10−7 ) on the caseinase activity of human chondrocytes and its interaction with rhIL-1 (10 U/ml). Chondrocytes were incubated with MEM, containing (a) 0.1% bovine serum albumin or (b) 1% fetal calf serum and test substances for 24 h. Caseinase activity was measured in cell supernatants. Results are representative of at least three experiments and replicates, from the same donor, are expressed as mean 2 ... (n = 4), **P E 0.01, compared with the rhIL-1-treated group, Scheffe’s multiple range analysis.

to a2-macroglobulin [42] to levels where TGFb augments IL-1-stimulated PGE production, as observed by us at 10−13  TGFb1 and b2 in the absence of serum (see Fig. 1b). Additionally, transcription factors present in serum may effect the synthesis of mediators involved in the responsiveness of cells to TGFb or IL-1 [43]. A systematic study of the effect of serum components on the action of IL-1 and TGFb would be required to explain the results fully. In general, TGFb reduced IL-1-stimulated caseinase activity either in the presence or absence of 1% serum. This observation is in accord with those of Chandrasekhar et al. [9] and Bandara et al. [44] who noted a TGFb1-mediated suppression of neutral metalloproteinase activity in rabbit chondrocytes, activated by synoviocyte-conditioned medium. Interestingly, they also observed an increase in PGE synthesis in these activated cells. Little or no significant effect of TGFb was seen on basal caseinase levels, possibly because activity levels were so low. Augmentation of IL-1-xstimulated caseinase activity at low concentrations of TGFb (10−13 ) in the absence of serum mirrors its effects on PGE production. TGFb appears to have biphasic effects, depending on concentration. This may result from the presence of TGFb receptors on chondrocytes with differing affinities and opposing effects [45]. Stromelysin is thought to be an important mediator of extracellular matrix degradation. Its key role may be in the activation of other latent metalloproteinases, since although it is capable of breaking down the cartilage proteoglycan, aggrecan, the major site at which aggrecan is cleaved in vivo is not a stromelysin cleavage site [46]. In osteoarthritic cartilage, an increase in stromelysin has been observed, suggesting a role in cartilage destruction [29]. The lowering of IL-1-stimulated stromelysin activity by TGFb may be important in the prevention of cartilage breakdown. The action of rhTGFb2 was markedly similar to that

of rhTGFb1, which is not surprising as TGFb1 and b2 share 70% homology and in other systems have a common action [2]. It is not possible to comment on their relative potencies as the growth factors were obtained from different sources and their specific activities were not available. The production of TGFb by chondrocytes and its actions on them suggests that it may be an important autocrine and/or paracrine modulator of chondrocyte metabolism. In the arthritic joint, any effects on PGE production may be dependent on cytokine interactions and levels of TGFb. Effects on IL-1-stimulated stromelysin activity may help tip the balance of cartilage metabolism from breakdown towards attempts at repair.

F. 4.—The effect of rhTGFb1 (10−13–10−10 ) on the caseinase activity of human chondrocytes and its interaction with rhIL-1 (10 U/ml). Chondrocytes were incubated with MEM, containing 0.1% bovine serum albumin and test substances for 24 h. Caseinase activity was measured in cell supernatants. Results are representative of at least three experiments and replicates, from the same donor, are expressed as mean 2 ... (n = 4), *P E 0.05, **P E 0.01, compared with the rhIL-1-treated group, Scheffe’s multiple range analysis.

FAWTHROP ET AL.: TGFb EFFECTS ON PGE AND CASEINASE

A This project was supported by the Arthritis and Rheumatism Council (UK) and the Biotechnology and Biological Sciences Research Council.

R 1. Sporn MB, Roberts AB, Wakefield LM, de Crombrugghe B. Some recent advances in the chemistry and biology of transforming growth factor beta. J Cell Biol 1987;105:1039–45. 2. Roberts AB, Sporn MB. The transforming growth factor bs. In: Sporn MB, Roberts AB, eds. Handbook of experimental pharmacology, peptide growth factors and their receptors I, Vol. 95. Heidelberg: Springer Verlag, 1990:419–72. 3. Okazaki R, Sakai A, Nakamura T, Kunugita N, Norimura T, Suzuki K. Effects of transforming growth factor bs and basic fibroblast growth factor on articular chondrocytes obtained from immobilised rabbit knees. Ann Rheum Dis 1996;55:181–6. 4. Sporn MB, Roberts AB. Transforming growth factor-b: multiple actions and potential clinical applications. J Am Med Assoc 1989;262:938–41. 5. Ignotz RA, Massague J. Transforming growth factorbeta stimulates the expression of fibronectin and collagen and their incorporation into the extracellular matrix. J Biol Chem 1986;264:4337–45. 6. Matrisian LM, Leroy P, Ruhlmann C, Gesnel M-C, Breathnach R. Isolation of the oncogene and epidermal growth factor-induced transin gene: complex control in rat fibroblasts. Mol Cell Biol 1986;6:1679–88. 7. Keski-Oja J, Blasi F, Leof EB, Moses HL. Regulation of the synthesis and activity in A549 human lung carcinoma cells by transforming growth factor-b. J Cell Biol 1988;106:451–9. 8. Laiho M, Saksela O, Andreasen PA, Keski-Oja J. Enhanced production and extracellular deposition of the endothelial type plasminogen activator inhibitor in cultured human lung fibroblasts by transforming growth factor-beta. J Cell Biol 1986;103:2403–10. 9. Chandrasekhar S, Harvey AK. Transforming growth factor-b is a potent inhibitor of interleukin-1 induced protease activity and cartilage proteoglycan degradation. Biochem Biophys Res Commun 1988;157:1352–9. 10. Andrews HJ, Edwards TA, Cawston TE, Hazelman BL. Transforming growth factor-beta causes partial inhibition of interleukin-1-stimulated cartilage degradation in vitro. Biochem Biophys Res Commun 1989;162:144– 50. 11. Morales TI, Joyce ME, Sobel ME, Danielpour D, Roberts AB. Transforming growth factor-beta in calf articular cartilage organ cultures: synthesis and distribution. Arch Biochem Biophys 1991;288:397–405. 12. Lafeber FPJG, Van der Kraan PM, Huber-Bruning O, Van den Berg WB, Bijlsma JWJ. Osteoarthritic human cartilage is more sensitive to transforming growth factor b than is normal cartilage. Br J Rheumatol 1993;32:281– 6. 13. Vivien D, Gale´ra P, Lebrun E, Loyau G, Pujol J-P. Differential effects of transforming growth factor-b and epidermal growth factor on the cell cycle of cultured rabbit articular chondrocytes. J Cell Physiol 1990; 143:534–45. 14. Frazer A, Bunning RAD, Russell RGG. Effects of transforming growth factor b and interleukin-1b on [3H]

15.

16. 17. 18.

19.

20.

21. 22. 23. 24.

25.

26. 27. 28.

29. 30.

31.

733

thymidine incorporation by human articular chondrocytes in vitro. Biochim Biophys Acta 1994;1226:193–200. Van Beuningen HM, Van der Kraan PM, Arntz OJ, Van den Berg WB. Protection from interleukin 1 induced destruction of articular cartilage by transforming growth factor b: studies in anatomically intact cartilage in vitro and in vivo. Ann Rheum Dis 1993;52:185–91. Wood DD, Ihrie EJ, Dinarello CA, Cohen PL. Isolation of an interleukin-1-like factor from human joint effusions. Arthritis Rheum 1983;26:975–83. Fava R, Olsen N, Keski-Oja J, Moses HL, Pincus T. Active and latent forms of transforming growth factor b activity in synovial effusions. J Exp Med 1989;169:291–6. Lafyatis R, Thompson NL, Remmers EF et al. Transforming growth factor-b production by synovial tissues from rheumatoid patients and streptococcal cell wall arthritic rats. J Immunol 1989;143:1142–8. Frazer A, Seid JM, Hart KA, Bentley H, Bunning RAD, Russell RGG. Detection of mRNA for the transforming growth factor b family in human articular chondrocytes by the polymerase chain reaction. Biochem Biophys Res Commun 1991;180:602–8. Villiger PM, Kusari AB, ten Dijke P, Lotz M. IL-1b and IL-6 selectively induce transforming growth factor b isoforms in human articular chondrocytes. J Immunol 1993;151:44–50. Cawston TE, Galloway WA, Mercer E, Murphy G, Reynolds JJ. Purification of rabbit bone inhibitor of collagenase. Biochem J 1981;195:159–65. Vane JR. The mode of action of aspirin and similar compounds. J Allergy Clin Immunol 1976;58:690–712. Raisz LG. The role of prostaglandins in the local regulation of bone metabolism. Prog Clin Biol Res 1990;332:195–203. Campbell IK, Piccoli DS, Hamilton JA. Stimulation of human chondrocyte prostaglandin E2 production by recombinant human interleukin-1 and tumour necrosis factor. Biochim Biophys Acta 1990;1051:310–8. Taylor DJ, Feldmann M, Evanson JM, Woolley DE. Comparative and combined effects of transforming growth factors a and b, interleukin 1 and interferongamma on rheumatoid synovial cell proliferation, glycolysis and prostaglandin E production. Rheumatol Int 1989;9:65–70. Woessner JF. Matrix metalloproteinases and their inhibitors in connective tissue remodeling. FASEB J 1991;5:2145–54. Murphy G, Cockett MI, Stephens PE, Smith BJ, Docherty AJP. Stromelysin is an activator of procollagenase. Biochem J 1987;248:265–8. Martel-Pelletier J, Zafarullah M, Kodama S, Pelletier J-P. In vitro effects of interleukin 1 on the synthesis of metalloproteinases, TIMP, plasminogen activators and inhibitors in human articular cartilage. J Rheumatol 1991;18(suppl. 27):80–4. Woessner J, Gunja-Smith Z. Role of metalloproteinases in human osteoarthritis. J Rheumatol 1991;18(suppl. 27):99–101. Meats JE, Elford PR, Bunning RAD, Russell RGG. Retinoids and synovial factor(s) stimulate the production of plasminogen activator by cultured human chondrocytes: a possible role for plasminogen activator in the resorption of cartilage in vitro. Biochim Biophys Acta 1985;838:161–9. Frazer A, Bunning RAD, Russell RGG. Modulation of human articular chondrocyte phenotype by transforming

734

32.

33.

34. 35.

36.

37.

38.

39.

BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 7 growth factor-b and interleukin-1b in vitro. Osteoarthritis Cartilage 1994;2:235–45. Richardson JH, Elford PR, Sharrard RM, Meats JE, Russell RGG. Modulation of connective tissue metabolism by partially purified human interleukin 1. Cell Immunol 1985;90:41–51. Bunning RAD, Russell RGG. The effect of tumour necrosis factor a on the resorption of human articular cartilage and on the production of prostaglandin E and of caseinase activity by human articular chondrocytes. Arthritis Rheum 1989;32:780–4. Godfrey K. Statistics in practice: comparing the means of several groups. N Engl J Med 1988;313:1450–6. Flanders KC, Thomson NL, Cissel DS et al. Transforming growth factor-b1: histochemical localization with antibodies to different epitopes. J Cell Biol 1989;108:653– 60. Ellingsworth LR, Brennan JE, Fok K et al. Antibodies to the N-terminal portion of cartilage-inducing factor A and transforming growth factor b: immunohistochemical localization and association with differentiating cells. J Biol Chem 1986;261:12362–7. Harvey AK, Hrubey PS, Chandrasekhar S. Transforming growth factor-beta inhibition of interleukin-1 activity involves down-regulation of interleukin-1 receptors on chondrocytes. Exp Cell Res 1991;195:376–85. Re´dini F, Mauviel A, Pronost S, Loyau G, Pujol J-P. Transforming growth factor b exerts opposite effects from interleukin-1b on cultured rabbit articular chondrocytes through reduction of interleukin-1 receptor expression. Arthritis Rheum 1993;36:44–50. Smith RJ, Justen JM, Sam LM, Rohloff NA, Ruppel PL,

40.

41.

42.

43.

44.

45.

46.

Brunden MN et al. Platelet derived growth factor potentiates cellular responses of articular chondrocytes to interleukin-1. Arthritis Rheum 1991;34:697–706. Bandara G, Lin CW, Georgescu HI, Evans CH. Chondrocyte activation by interleukin-1: synergism with fibroblast growth factor and phorbol myristate acetate. Agents Actions 1989;27:439–41. Fawthrop FW, Russell RGG, Bunning RAD. The modulation of interleukin-1b-stimulated prostaglandin E production and caseinase activity by platelet derived growth factors AA, AB and BB in human articular chondrocytes. Calcif Tissue Int 1992;50(suppl.):A24. Danielpour D, Sporn MB. Differential inhibition of transforming growth factor beta 1 and transforming growth factor beta 2 activity by alpha 2-macroglobulin. J Biochem 1990;265:6973–7. Goldring MB. Degradation of articular cartilage in culture: regulatory factors. In: Woessner JF, Howell DS, eds. Joint cartilage degradation, basic and clinical aspects. New York: Marcel Dekker Inc., 1993:281–345. Bandara G, Lin CW, Georgescu HI, Evans CH. The synovial activation of chondrocytes: evidence for complex cytokine interactions involving a possible novel factor. Biochim Biophys Acta 1992;1134:309–18. Glansbeek HL, van der Kraan PM, Vitters EL, van den Berg WB. Correlation of the type II transforming growth factor b (TGFb) receptor with TGF-b responses of isolated bovine articular chondrocytes. Ann Rheum Dis 1993;52:812–6. Lohmander LS, Neame PJ, Sandy JD. The structure of aggrecan fragments in human synovial fluid. Arthritis Rheum 1993;36:1214–22.