| Sign In to gain access to subscriptions and/or personal tools. |
Effects of TGF-β on Hyaluronan Anabolism in Fibroblasts Derived from the Synovial Membrane of the Rabbit Temporomandibular JointDepartment of Orthodontics and Craniofacial Developmental Biology, Hiroshima University Graduate School of Biomedical Sciences, 1-2-3 Kasumi, Minami-ku, Hiroshima 734-8553, Japan; Correspondence: * corresponding author, tkotaro{at}hiroshima-u.ac.jp
Hyaluronan (HA) synthesis in the synovial membrane is affected by various chemical mediators. It is hypothesized that transforming growth factor-beta1 (TGF-β1) would be a mediator to modulate HA synthesis in cultured synovial membrane fibroblasts of the temporomandibular joint (TMJ). Fibroblasts were extracted from the TMJ synovial membrane of four-week-old Japanese white rabbits. The amount of HA and expression levels of HA synthase (HAS) mRNAs induced by TGF-β1 treatment were analyzed by means of high-performance liquid chromatography and real-time polymerase chain-reaction, respectively. Both medium and large amounts of HA were enhanced by the stimulation of TGF-β1. HAS2 mRNA expression was enhanced 13-fold after six-hour stimulation with TGF-β1 (10 ng/mL), whereas HAS3 mRNA expression was not changed significantly. These results suggest that TGF-β1 enhances the expression of HAS2 mRNA in the TMJ synovial membrane fibroblasts and may contribute to the production of high-molecular-weight HA in the joint fluid.
Key Words: hyaluronan HAS synovial membrane TMJ transforming growth factor
Hyaluronan (HA), a glycosaminoglycan with repeated disaccharide units of D-glucuronic acid and N-acetyl-D-glucosamine, is a major component of the extracellular matrix. In general, HA has a molecular weight of 800–1900 kDa in its native state and has various physiological and biological functions (Laurent and Fraser, 1992). The viscoelasticity and lubrication of synovial fluid produced by high-molecular-weight HA are crucial to normal jaw function or condylar movement in the temporomandibular joint (TMJ). In normal joints, high-molecular-weight HA is dominant in the synovial fluid, whereas low-molecular-weight HA increases under inflammatory or pathologic conditions (Bjelle et al., 1983). Previous studies have demonstrated the effects of various cytokines (Hamerman and Wood, 1984; Meyer et al., 1990) and growth factors (Heldin et al., 1989; Westergren-Thorsson et al., 1990) on HA synthesis in cultured fibroblasts. Transforming growth factor beta (TGF-β), a 25-kDa disulfide-linked dimeric protein, is a general stimulator of connective tissue formation and increases the amounts of several connective tissue components, such as collagens (Sporn et al., 1987), fibronectin (Varga et al., 1987), and dermatan sulfate proteoglycan (Bassols and Massague, 1988). Among the isoforms of TGF-β, TGF-β1 is known to be a potent mediator of HA synthesis in fibroblasts (Heldin et al., 1989; Westergren-Thorsson et al., 1990). Previous studies showed that the level of TGF-β1 in synovial fluid was increased in the knee joint with trauma (Wei and Messner, 1998), rheumatoid arthritis (Olsson et al., 2001), and osteoarthrosis (Fahlgren et al., 2001). An increase in the level of TGF-β1 was also demonstrated in the synovial fluid of patients with temporomandibular disorders (Fang et al., 1999). TGF-β1 stimulates cartilage matrix synthesis in injured cartilage and osteoarthritis (Trippel, 1995; van den Berg, 1995; Wei and Messner, 1998). It is suggested from these findings that TGF-β1 may be induced during the repair process of damaged joints and may play an important role in the improvement of intra-articular pathologic status. It is essential, in the repair of joints, that the biomechanical properties of synovial fluid, such as viscoelasticity and lubrication, controlled by high-molecular-weight HA, be improved. Injection of high-molecular-weight HA into the joint was found to be effective for the treatment of arthritis (Balazs and Denlinger, 1989). At present, however, we do not fully understand whether synovial membrane cells have the potential to control HA synthesis mediated by TGF-β1. Recently, three putative HA synthase (HAS1, HAS2, and HAS3) genes have been cloned in mice and humans, and their characteristics have been partially clarified (Itano et al., 1999). HAS1 and HAS2 polymerize HA chains of similar large size (up to 2000 kDa), whereas HAS3 produces shorter chains (from 200 to 300 kDa) (Itano et al., 1999). However, the effects of TGF-β1 on the expression of HAS mRNAs and the relevant HA synthesis have not been clarified in synovial membrane fibroblasts treated with TGF-β1. We conducted the present study to investigate the effects of TGF-β1 on the mRNA expression of HAS isoforms and the production of various molecular weights of HA in cultured fibroblasts derived from rabbit TMJ synovial membrane.
Cell Culture Permission for a series of experiments in this study was granted by the Ethics Committee of Hiroshima University. Fibroblasts were isolated from the TMJ synovial membrane of four-week-old Japanese white rabbits as described previously (Werb and Burleigh, 1974). Fibroblasts were maintained in 100-mm culture dishes (Corning, New York, NY, USA) with 10 mL alpha-minimum essential medium (Sanko Junyaku, Tokyo, Japan) containing 10% fetal bovine serum (FBS; Mitsubishikasei, Tokyo, Japan), 32 U/mL penicillin, 60 µg/mL kanamycin, and 250 ng/mL amphotericin B under an atmosphere of 5% CO2 in a humidified incubator. The conditioned medium was changed every other day until the cells became confluent on the culture plate. Finally, the conditioned medium was changed to medium containing 0.5% FBS 12 hrs prior to the following experiments, and then incubated in fresh conditioned medium with or without 1–100 ng/mL TGF-β1 (Sigma, St. Louis, MO, USA). The 5th- to 8th-passaged cells from 3 different preparations isolated from 3 different rabbits were used for the following experiments.
Quantification of HA The pre-treated cell layer and conditioned medium were moved into disposable extraction columns (Bond Elut SCX and BondElut SAX; GL Sciences, Tokyo, Japan). After being mixed with 1 mL of 50% MeOH and 5 mL H2O, the solvents were eluted by 3 mL of 50 mM MeOH/HCl. After drying, the solvents in 500 µL of 0.1 M NaCl were filtrated with a 0.45-µm millipore filter (Gelman Sciences, Ann Arbor, MI, USA). The solvents were analyzed by means of high-performance liquid chromatography (HPLC: Waters 600E; Waters, Milford, MA, USA) with gel filtration columns (Ohpak KB-804 for the fraction under 1000 kDa, Ohpak KB-806 for the fraction 1000~20,000 kDa; Shodex, Tokyo, Japan). Elution was carried out with 0.1 mM NaCl at a flow rate of 1.0 mL/min. The column effluent was monitored by a differential refractometer (RI DETECTOR 504R; GL Sciences). The accumulation levels of various molecular weights of HA were quantified based on the calibration curves for pure HAs of 300 and 1900 kDa supplied by Denki Kagaku Kogyo (Tokyo, Japan).
Real-time Polymerase Chain-reaction (PCR) Analysis
PCR amplification was carried out by means of an ABI Prism 7700 Sequence Detection System (PE Applied Biosystems) as described previously (Leutenegger et al., 1999). The HAS signals were normalized with reference to the G3PDH signals.
Statistical Analysis
Effects of TGF-β1 on the Accumulation of HA in Cultured Synovial Membrane Fibroblasts The level of low-molecular-weight HA (< 300 kDa) was not changed markedly by the treatment of TGF-β1 in both the cell layer and conditioned medium of cultured fibroblasts derived from the synovial membrane of TMJ (Figs. 1A
Effects of TGF-β1 on the Expression of HAS mRNAs in Cultured Synovial Membrane Fibroblasts HAS2 and HAS3 mRNAs were expressed in unstimulated cultured fibroblasts from the TMJ synovial membrane, whereas HAS1 mRNA could not be detected even at high cycle numbers in reverse-transcriptional PCR analysis (data not shown). Dose-dependent and time-dependent effects of TGF-β1 were examined for the expression of HAS mRNAs in cultured fibroblasts from the synovial membrane of TMJ by use of a real-time PCR analysis.
The expression of HAS2 mRNA was significantly enhanced after six-hour stimulation with TGF-β1 (0.1 ng/mL, p < 0.05; 1–100 ng/mL, p < 0.01), and reached a maximum 13-fold increase at the dose of 10 ng/mL relative to the controls (Fig. 2A
In our previous study, expression of HAS2 and HAS3 mRNAs, but not HAS1, was detected in the synovial membrane of rabbit knee joints in both in vivo and in vitro experiments (Tanimoto et al., 2001). In this study, only HAS2 and HAS3 mRNAs were detected in the cultured synovial membrane fibroblasts of rabbit TMJ, and HAS1 mRNA could not be found. The accumulation of HA, high-molecular-weight HA in particular, was enhanced by the stimulation of TGF-β1 in fibroblasts derived from the rabbit TMJ synovial membrane, as is the case in fibroblasts derived from human knee joints (Haubeck et al., 1995). Furthermore, HAS2 mRNA was remarkably up-regulated by the TGF-β1 treatment. These findings suggest that the accumulation of high-molecular-weight HA induced by TGF-β1 treatment is caused by the up-regulation of HAS2 in the TMJ. Since high-molecular-weight HA is essential for maintaining the viscoelasticity of synovial fluid (Kobayashi et al., 1994), the enhancement of HA synthesis by HAS2 may improve the biomechanical function of synovial joints. In contrast, low-molecular-weight HA causes a decrease in the viscoelasticity of synovial fluid and modulates the immune or inflammatory processes (McKee et al., 1996; Hodge-Dufour et al., 1997). An imbalance of synthesis and degradation of HA caused by various pathological conditions leads to the accumulation of low-molecular-weight HA in joint fluid (Bjelle et al., 1983). The pathological accumulation of low-molecular-weight HA is suggested to be due to various degrading mechanisms, such as depolymerization with reactive oxygen species (McNeil et al., 1985) and enzymatic cleavage (Orkin and Toole, 1980), and acceleration of low-molecular-weight HA synthesis by HAS3. Such pro-inflammatory cytokines as interleukin-1 beta and tumor necrosis factor-alpha markedly enhance the gene expression of HAS3 as well as HAS2 in cultured human orbital fibroblasts (Kaback and Smith, 1999), gingival fibroblasts (Ijuin et al., 2001), and rabbit knee synovial membrane fibroblasts (Tanimoto et al., 2001). However, the gene expression of HAS3 was not affected by TGF-β1 in the cultured synovial membrane fibroblasts in the present study. Although a constant level of low-molecular-weight HA was detected in both the cell layer and conditioned medium, no significant differences in the HA levels were found between the TGF-β1-treated and control groups during the experimental period. It was demonstrated that some hyaluronidases are expressed and exert activity in normal fibroblasts (Huey et al., 1990). The catabolic action for HA may thus be exerted in the normal state in cultured fibroblasts derived from the synovial membrane. Since there has been no evidence to demonstrate that TGF-β1 induces the degradation of HA, it would be reasonable to assume that TGF-β1 may not cause the accumulation of low-molecular-weight HA different from the effects of pro-inflammatory cytokines on HA accumulation. These findings suggest that TGF-β1 is capable of enhancing the synthesis of only high- but not low-molecular-weight HA, by selective up-regulation of HAS2 mRNA. The availability of TGF-β1 for the treatment of TMJ arthritis has not been confirmed, but it can be speculated that the injection of TGF-β1 in combination with pumping manipulation would be effective for the treatment of TMJ arthritis. The amount of TGF-β1 for the treatment may be substantially less than that of sodium hyaluronate currently used for joint diseases. Further investigation would be expected to clarify the effectiveness of TGF-β1 for the treatment of patients with TMJ arthritis. In conclusion, it is shown that TGF-β1 may contribute to the accumulation of high-molecular-weight HA needed for the maintenance of the viscoelasticity and lubrication of synovial fluid by enhancing HAS2 expression in the synovial membrane fibroblasts in TMJ.
This research was supported by a Grant-in-Aid (No. 1347045100) for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology, Japan. Received for publication March 9, 2003. Revision received October 1, 2003. Accepted for publication October 1, 2003.
Journal of Dental Research, Vol. 83, No. 1,
40-44 (2004) This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||



