Taipei Medical University Institutional Repository:Item 987654321/4246

全文

(1)台北醫學大學醫學科學研究所醫學檢驗暨生物技術組碩士論文. Taipei Medical University Graduate Institute of Medical Sciences Division of Medical Examination and Biotechnology Master Thesis. 指導教授: 陳建和. 博士 (Chien-Ho Chen Ph.D.). 指導教授: 許明照. 博士 (Ming-Thau Sheu Ph.D.). 乙型轉型生長因子藉由多重路徑抑制人類初代滑膜纖維母細胞上第一型間白素誘導之第二型蛋白酶活化受體表現 TGF- β inhibits PAR-2 expression induced by IL-1 β on human primary synovial fibroblasts through multiple pathways. 研究生: 鄭琇丹 (Hsiu-Tan Cheng). 中華民國九十七年七月 July 2008.

(2) 致謝兩年到底算長還是短呢？還記得剛進來研究所時的情景，感覺就像幾天前發生的事情，現在卻已經在打著論文的謝誌，論文寫完了，碩士生涯也即將結束，替自己高興的，又完成了一個階段的學業，但是卻也有些捨不得，兩年相處的同學們，常常陪伴我度過漫漫長夜的實驗室，有點嘮叨卻又跟爸爸一樣慈祥的陳老師，以及陪我走過六年的北醫校園。這些年來，感謝這一切，我從懵懂的大學生變成一個懂得獨立思考的研究生，首先，我要感謝三位老師：對學生很好的指導教授陳建和老師、點子很多的共同指導教授許明照老師，以及在實驗技術上提供許多支援的梁有志老師，有他們的指導，我的碩士生涯才能在兩年內順利落幕。此外，還要感謝陳家的實驗室夥伴：QOO、百鴻、藍絜、奕璇，有你們一起奮鬥，碩士兩年讓我隨時充滿衝勁；也感謝指點迷津的畢業學長姐：佳蓓、健翔、雅婷、育如，感謝你們在不同階段的指導以及鼓勵，我才能不斷進步；還有陳家的碩一學弟妹：詩芸、丞弘跟千翔，感謝你們一年來的陪伴以及最後幫我們打理口試時的一切，我才能順利通過口試；另外，特別感謝兩位大學部學弟：冠偉以及宥嘉，感謝你們整整一年的協助，常常提供大家細胞做實驗，也常常陪大家一起聊天說笑，讓實驗室感覺輕鬆許多。除此之外，我還要感謝梁老師實驗室的許多人：謝謝苓芳學姊、佩蓉學姊，常常幫我解惑；還有邵宇、乃琦跟阿信三個同學，讓實驗室變熱鬧許多；以及超級貼心的鈺蒨、活潑大方的紹玉、面惡心善的大皓三位學弟妹，也幫我很多忙；還有其他同屆的同學們：辜董、強哥、家慧、阿達、怡蓉跟宜珮，感謝大家兩年來互相切磋學習，讓我成長許多。然後，還要感謝許多夠義氣的幕後英雄：常常在百忙之中北上來探望我的大學同學們、在我心情低落時當我垃圾桶的高中死黨。最重要的，要感謝我的家人，尤其是老爸老媽，謝謝您們的經濟支援以及精神支持，讓我能專心在學業上。能順利畢業，要感謝的人也許兩頁都寫不完，文字能表達的實在有限，心裡面的感激卻是無限，在此僅能以這篇小小的謝誌，表達對所有師長，同學，學弟妹以及親人們大大的謝意，謝謝你們大家一路上給我支持以及鼓勵。最後，再次感謝老天厚愛，感謝神明保佑，感謝鄉親朋友老師同學們，託各位的福，我，熬過來啦!. 琇丹謹至於台北醫學大學醫學科學研究所民國九十七年七月 I.

(3) Abstract Proteinase-activated receptors (PARs) are a family of four G-protein-coupled receptors (GPCR) which included four members: PAR-1, PAR-2, PAR-3, and PAR-4. As reported, PAR-2 related mostly with inflammation. Osteoarthritis (OA) patients, chondrocytes have higher PAR-2 expression level than that of normal chondrocytes and this expression can be up-regulated by IL-1 β but repressed by TGF-β, an anabolic factor of OA pathogenesis. This suggests the relationship between PAR-2 and OA progression. Even though it is known that TGF- β can inhibit PAR-2 expression in OA chondrocytes, the phenomenon and the mechanisms have never been discussed in human primary synovial cells. The aim of this study is to investigate the mechanism how TGF- β represses IL-1 β induced PAR-2 expression in human primary synovial cells. In this study, we have demonstrated that IL-1 β induces PAR-2 expression via p38 pathway and this induction can be repressed by TGF- β in human primary synovial cells. Also, we observed that this inhibition persists for at least 48 hours, and this suggests TGF- β inhibits PAR-2 expression through multiple pathways. First, TGF- β inhibits PAR-2 by inhibiting IL-1 β induced p38 signal transduction. Then, we demonstrated that the inhibition also indirectly due to MMP-13 inactivation. Finally, TGF- β induces CTGF, and CTGF also represses PAR-2 expression by inhibit IL-1 β induced phospho-p38 level. Because PAR-2 gene expression is up-regulated by IL-1 β in short time in human primary synovial cells, this suggests that PAR-2 expression may play an important role in early phase of OA. In light of this, PAR-2 might be a novel ideal therapeutic target to prevent OA from progressing.. Key words: OA, PAR-2, IL-1 β, TGF- β.. II.

(4) 摘要蛋白酶活化受體家族屬於 GPCR，由四個成員組成，分別是第一型到第四型蛋白酶活化受體，四個成員之中，以第二型和發炎反應最為相關。目前已知第二型蛋白酶活化受體在關節炎病人以及正常人的軟骨細胞均有，其中在關節炎病人的軟骨細胞上表現較正常人多，並且其表現會受第一型間白素影響而上升，但是此上升現象在乙型轉型生長因子存在下則會被抑制，而乙型轉型生長因子在關節炎中是重要的同化作用因子。這些發現表示第二型蛋白酶活化受體和骨關節炎病程機制有所關聯。雖然已經確定乙型轉型生長因子會抑制關節炎病人軟骨細胞上的第二型蛋白酶活化受體表現，但其詳細分子機制仍然未知，並且，這些現象在人類滑膜纖維母細胞模式中，並未被研究過，因此，本篇研究即是探討人類滑膜纖維母細胞模式中，乙型轉型生長因子如何抑制第二型蛋白酶活化受體表現。我們在本研究中已經證實了，在人類滑膜纖維母細胞中，第一型間白素會經由 p38 路徑引發第二型蛋白酶活化受體表現，並且此種誘導也可被乙型轉型生長因子抑制。同時，因為此種抑制作用可持續達四十八小時，因此，我們認為乙型轉型生長因子對第二型蛋白酶活化受體的抑制行為可能是受到多條路徑所調控。首先，我們發現乙型轉型生長因子會在短時間內直接抑制 p38 路徑；其次，我們也證實了此種抑制現象間接和 MMP-13 的去活化有關；最後，乙型轉型生長因子誘導結締組織生長因子產生後，p38 路徑也會被結締組織生長因子所抑制。在人類滑膜纖維母細胞中，當第一型間白素存在時，第二型蛋白酶活化受體基因表現在很短時間內即被活化，此表示第二型蛋白酶活化受體在骨關節炎病程早期佔有重要角色，有鑑於此，抑制第二型蛋白酶活化受體在關節炎初期的表現，是避免關節炎病程進展的理想方式。. 關鍵字：骨關節炎、第二型蛋白酶活化受體、第一型間白素、乙型轉型生長因子. III.

(5) Table of Contents. 致謝.................................................................................................................... I. Abstract.............................................................................................................. II. 摘要..................................................................................................................... III. Table of contents………………………………………………………………. IV. Introduction........................................................................................................ 1. Materials and Methods..................................................................................... 6. Materials.......................................................................................................... 6. Primary culture of human synovial cells......................................................... 7. RNA extraction and RT-PCR.......................................................................... 8. Polymerase Chain Reaction............................................................................. 9. Preparation of cell lysates................................................................................ 10. SDS PAGE………………............................................................................... 10. Western Blotting.............................................................................................. 10. Preparation of medium protein……................................................................ 11. Collagen zymography……….......................................................................... 11. Statistical analysis........................................................................................... 11. Results................................................................................................................. 13. Induction of PAR-2 mRNA and protein level by IL-1 β in human primary synovial cells................................................................................................... 13. Inhibition of IL-1 β -induced PAR-2 expression by p38 MAPK inhibitor but not by MEK or JNK MAPK inhititor in human primary synovial cells.... 13. TGF- β 1 suppresses excess expression PAR-2 level induced by IL-1 β in human primary synovial cells.......................................................................... IV. 14.

(6) TGF- β 1 inhibits IL-1 β mediated p38 MAPK activation, but not JNK or ERK level of human primary synovial cells.................................................... 14. TGF- β 1 induces CTGF mRNA and protein expression level of human primary synovial cells...................................................................................... 15. CTGF suppresses PAR-2 protein level induced by IL-1 β in human primary synovial cells...................................................................................... 15. CTGF inhibits IL-1 β -mediated p38 MAPK activation, but activated ERK pathway of human primary synovial cells............................................... 16. TGF- β 1 down-regulates TIMP-3 mRNA and protein level of human primary synovial cells...................................................................................... 17. TGF- β 1 induces TIMP-3 expression via Akt pathway in human primary synovial cells................................................................................................... 17. TIMP-3 blocks MMP-13 activity in culture medium of human primary synovial cells.................................................................................................... 18. Blocking of MMP-13 function resulted in down-regulation of PAR-2 protein level of human primary synovial cells................................................ 18. Discussion........................................................................................................... 19. References.......................................................................................................... 22. Figures................................................................................................................ 28. Figure 1............................................................................................................ 28. Figure 2............................................................................................................ 31. Figure 3............................................................................................................ 33. Figure 4............................................................................................................ 36. Figure 5............................................................................................................ 39. Figure 6............................................................................................................ 43. V.

(7) Figure 7............................................................................................................ 45. Figure 8............................................................................................................ 53. Figure 9............................................................................................................ 56. Figure 10.......................................................................................................... 60. Figure 11.......................................................................................................... 62. Appendix Figures............................................................................................... 64. AF. 1................................................................................................................ 65. AF. 2................................................................................................................ 66. AF. 3................................................................................................................ 69. AF. 4................................................................................................................ 70. AF. 5................................................................................................................ 71. AF. 6................................................................................................................ 72. AF. 7................................................................................................................ 73. VI.

(8) Introduction OA is a degenerative disease characterized by depletion of articular cartilage and formation of osteophytes, it’s often seen in elders and those who have bad posture and use their joint inappropriately 3. There are various factors can be involved in the development of OA: mostly traumatic events are causative but there are other factors like genetic predisposition, defective position of joints, ageing and malnutrition 12. On the view of histology, articular cartilage is composed of 5~10% chondrocytes and more than 90% of matrix component, including type II collagen, proteoglycan, fibronectin and glycosaminoglycan (GAG) 20. OA formed under the condition of imbalance between anabolic and catabolic mediators, when catabolism is grater than anabolism, the risk of OA raises. The catabolic mediators include a serious of matrix degrading enzymes named “matrix metalloproteinase” such as MMPs, ADAMTS, and ADAM, and various proinflammatory cytokines such as IL-1 β, IL-17, IL-18 and TNF-α, which increase degradation of cartilage and inhibit synthesis of metalloproteinase inhibitors such as TIMPs, tenascin and YKL-40. The anabolic mediators include TGF- β, IGF-1, FGFs and BMPs, which stimulate synthesis and repairing of cartilage. OA primarily affects articular cartilage, this contrasts with rheumatoid arthritis, which is primarily a disease of the synovial membrane 3. When an inappropriate mechanical stress is present, articular cartilage (AC) damaged and release small pieces of cartilage 19. At this time, increased secreting of proinflammation cytokines such as IL-1 β and TNF-α, and catabolic enzymes like MMP-1, MMP-3, MMP-9, MMP-13, aggrecanase-4, aggrecanase-5, and TACE from articular chondrocytes and synoviocytes. are. activated.. The. secreted. proinflammatory. cytokines. and. metalloproteinases up-regulate expression of chondrocyte PAR-2, stimulating more proinflammatory. cytokines. and. metalloproteinases 1. secrete. and. enhancing.

(9) inflammatory response. 5. and degradation of ECM components of cartilage tissue,. causing progressive loss of cartilage, furthermore, fragments of protein degradation, like fibronectin fragments and collagen type II fragments, seem to play a role in inducing degradation of cartilage. 21. as well as stimulating chondrocytes to repair the. matrix. Connective tissue growth factor (CTGF), a member of CCN family, is a cysteine-rich matricellular protein. Expression of this protein is potently induced by TGF- β via Smad pathway. It is strongly expressed in mesenchyme, and it performs a critical role in formation of cartilage, bone and tooth. CTGF knockout animals displayed an inability of expression of matrix specific protein and failure of proliferation of chondrocytes. It promotes chondrocytes proliferation through p38 MAPK and differentiation via p42/p44 MAPK. Thus, CTGF is important for cell proliferation and matrix remodeling during chondrogeneis and is a key regulator coupling ECM remodeling. 14. . Several studies have proved that CTGF can stimulate. the proliferation and expression of the cartilage phenotype by promoting type II collagen and aggrecan production, but did not stimulate the terminal hypertrophy or calcification of articular cartilage cells, suggesting that CTGF might be useful in the repair of damaged articular cartilage. 18. . Moreover, it was reported that a strong. immuno-positive reaction for CTGF was seen in the clustering chondrocytes next to the damaged cartilage surfaces, and excessive TGF- β production in osteoarthritic cartilages 2. Besides, there has been reported that TGF- β stimulate type II collagen and GAG producing and growth promotion in rabbit articular chondrocytes 4. Other report suggested that TGF- β antagonizes IL-1 β -mediated inflammation via decreasing its receptor expression on chondrocytes. 22-24. . In summary, TGF- β and. CTGF play a critical role in cartilage matrix repairing; in addition, TGF- β is an anabolic and anti-catabolic factor of articular cartilage. 2.

(10) PARs are a family of four G-protein-coupled receptors which included four members: PAR-1, PAR-2, PAR-3, and PAR-4. PAR-1 and PAR-2 distribute in many tissues, including airway, blood, bone, joint system, cardiovascular system, epidermis, immune system, kidney, nervous system, skeletal muscle, and digestive system. Activation of PAR-1 is related in proliferation, inhibited differentiation, smooth muscle contraction/relaxation chemokine production and activation of PAR-2 is related in proliferation, inflammation and immune regulation. PAR-3 does not signal when it is activated by proteinase, so it is thought that PAR-3 acts as a thrombin tethering protein rather than a fully active receptor. PAR-4 expresses highly in lung, pancreas, thyroid, testis and small intestine but not found in other tissues. Similar to PAR-3, PAR-4 also does as an adjunct to PAR-115. Among PARs, PAR-2 is unique in that it is activated by trypsin and mast cell tryptase, but not by thrombin which activates the other three members of the PAR family. This study focuses on PAR-2, which plays an important role in inflammation and pain. Trypsin cleaves PAR-2 at R34↓S35LIGKV of the extracellular N-terminus to expose the hexameric tethered peptide that binds to conserved regions in extracellular second loop of the receptor to initiate signaling. During activation, PAR-2 couples to Gαq/11, resulting in activation of phospholipase C- β, production of inositol 1,4,5 -trisphosphate and diacylglycerol, and then activation of protein kinase C. In addition, PAR-2 can activate ERK1/2 MAPK, mediating cell proliferation 14. Recently, it was reported that expression of PAR-2 on chondrocytes and synovial cells and it was overexpressed on osteoarthritic chondrocytes. The expression level of PAR-2 on chondrocytes is up-regulated by IL-1β, TNF-α and down-regulated by TGF- β 5. Tissue inhibitor of metalloproteinase (TIMP), comprise a family of four glycoproteins (TIMP-1, 2, 3, and 4)37. TIMP-1, -3, -4 expression is inducible and often exhibits tissue specificity: TIMP-1 is enriched in reproductive organ systems; 3.

(11) TIMP-3 is uniquely located in ECM; and TIMP-4 in neural tissue, fetal testes, and Sertoli cells or ovaries, and cardiac, breast, and skeletal muscle tissues, whereas TIMP-2 expression is largely constitutive34. TIMPs inhibit metalloproteinases activity in a similar way, by binding to their catalytic sites. TIMPs differ in their affinity for specific metalloproteinases, for example, TIMP-3 inhibits MMP-13, aggrecanases (ADAMTS-4, 5) and TACE (ADAM-17). Transforming growth factor- β (TGF- β), a prominent member of the TGF- β superfamily of ligands which include TGF- β s and bone morphogenetic proteins (BMPs), is vital for the homeostasis of numerous cellular functions, including cell growth, differentiation, and apoptosis in a broad spectrum of tissues38. TGF- β signals are propagated through direct physical interactions with the extracellular domain of essentially two transmembrane serine/threonine kinase receptors (T β RI and T β RII), which transduce a number of secondary signals, most notably Smads 2 and 3 as well as PI3-kinase and various members of the mitogen activated protein kinase (MAPK) family39-42. In human chondrocytes, a recent study has reported that TGF- β induced TIMP-3 via PI3K/Akt signaling pathway34. Several models have been proposed to explain how TGF- β may activate the PI3K/Akt pathway. A recent study suggests that T β RI associates with the p85 regulatory subunit of PI3K, thus enabling activation of the p110α catalytic subunit of PI3K38. In summary, it is possible that TGF- β repress human synovial PAR-2 via Akt-TIMP3-MMP13 pathway, as TGF- β induces TIMP-3 expression by Akt activation, MMP13 activity can be blocked by raised TIMP. Once MMP13 is blocked, inducing factors of PAR-2 decrease, so repressive effect of TGF- β to PAR-2 can be performed. In light of this, the aim of this thesis is to study the roles and molecular 4.

(12) mechanisms of TGF- β against OA, including the anabolic ability to induce CTGF production in order to maintain ECM integrity, and the anti-catabolic ability to induce TIMP-3 production to inhibit MMPs and avoid PAR-2 over-expression.. 5.

(13) Materials and Methods Materials Human primary chondrocytes and synovial cells. Dulbecco’s modified eagle medium (DMEM) was purchased from Gibco (Grand Island, NY). Fetal bovine serum from Gibco (FBS, Grand Island, NY). Penicillin-streptomycin solution, amphotericin B solution, and L-glutamine from Sigma chemical (St. Louis, Mo, USA). Trizol (REzol) from C&T (UK). Chloroform from Sigma chemical (St. Louis, Mo, USA). Isopropanol from J.T. Baker (Ph, USA). DEPC water (Ultraspec) from Biotecx (TX, USA). Oligo dT18 primer from BioTeks (RU Saint Petersburg). 40mM dNTP from VIOGENE (CA, USA). MMLV- derived reverse transcriptase and RT buffer from TOYOBO (Osaka, Japan). PCR buffer and Taq-polymerase from Yeastern Biotech (Taipei, Taiwan). Ethidium bromide from AMRESCO (Solon, Ohio, USA). Bio-Rad protein assay dye from BIO-RAD (Hercules, CA, USA). PVDF membrane from PALL-Corporation (Mi, USA). Hyaluronidase, pronase, collagenase and BSA from Sigma chemical (St. Louis, Mo, USA). Rabbit anti-CTGF, Rat anti-tubulin, mouse anti-TIMP3, and mouse anti-type II collagen from NOVUS (Littleton, CO). Mouse anti-PAR2 from Santa Cruz (CA, USA). Mouse anti-pSMAD and mouse anti-pp38 from BD (CA, USA). 6.

(14) Rabbit anti-pAkt from cell signaling (MA, USA). Mouse anti-GAPDH from abcam (Cambridge, UK). Anti-rat HRP-conjugated antibody, anti-rabbit HRP-conjugated antibody, and anti-mouse HRP-conjugated antibody from Jackson ImmunoResearch (West Grove, PA). Enhanced Chemiluminescent (ECL) Substrate from Visual protein Biotechnology Corp (MA, USA). Microcon centrifugal filter devices from Millipore (MA, USA). Recombinant human IL-1β and recombinant human TGF- β 1 from R&D system (Minneapolis, USA). Recombinant human CTGF from MBL (Nagoya, Japan). Recombinant human MMP-13 and bovine collagen type III from Sigma chemical (St. Louis, Mo, USA). Signal inhibitors from Merck (Darmstadt, Germany).. Primary culture of human chondrocytes and synovial cells Human chondrocytes and synovial cells were isolated from cartilages and synovium of OA patients undergoing total joint replacement surgery. Tissues were cut into small pieces by using surgical scissors and then treated with 0.1% hyaluronidase for 15 minutes, 0.5% pronase for 30 minutes, and 0.2% collagenase for 12 hours to digest ECM of synovial tisssues to release chondrocytes and synovial cells from tissues. Released cells were inoculated in 60 mm diameter dish and cultured in Dulbecco’s modified eagle medium (DMEM) containing 10% FBS, 1% penicillin-streptomycin solution, 1% amphotericin B solution, and 1% L-glutamine in an atmosphere of 37°C with 5% CO2 and medium is replaced every two days. When cells reach 80% confluence, treat them with FBS-free medium with various 7.

(15) concentration of stimulant for dose-course analysis, or with certain concentration of stimulant for various time for time-course analysis.. RNA extraction and RT-PCR For evaluating the expression of mRNA level of CTGF, PAR2 and TIMP-3, total RNA was extracted from primary synovial cells using Trizol reagent. The Trizol solutions were then extracted with chloroform (Trizol/chloroform: 5/1) and centrifuged at 13000 rpm for 15 minutes. The upper phase of samples were collected with another tubes and precipitated with the same volume of isopropanol and spinning at 13000 rpm for 30 minutes to obtain RNA pelletes. Then the pelletes were washed with 75% ethanol in diethylpyrocarbonate (DEPC) - treated water. After drying the pelletes at room temperature, the samples were resuspended in 10 μl of DEPC water. And 2 μg of total RNA was reverse transcribed into cDNA by using oligo-dT18 primer at concentration of 0.2 μg/μl, 0.5 mM of dNTP, 2 U/μl of MMLV-derived reverse transcriptase and 5 μl of 5X RT buffer for 60 min at 42℃.. 8.

(16) Polymerase Chain Reaction 0.08 μg cDNA was amplified with 1 μl primers described in following table: Primers CTGF. PAR-1. PAR-2. PAR-3. PAR-4. Sequence. bps. Sense. 5’ CCG TAC TCC CAA AAT CTC CA. 210. 60. 35. Anti-sense. 5’ GTA ATG GCA GGC ACA GGT CT. Sense. 5’ACC ACA TTT GCT CCA TCC TC. 248. 60. 35. Anti-sense. 5’ CCT ATT GGA GTG CCC ACA GT. Sense. 5’ CTG CCT ATG TGC TGA T. 317. 60. 35. Anti-sense. 5’ CGG ACA CTT CGG CAA A. Sense. 5’ACC CTC CAC CAC TTC ACA AG. 198. 60. 35. Anti-sense. 5’AGC AAG AGG TTT GGT TGG TG 5’GTG GGC CTT ACA TCC AGT GT 241. 60. 35. 60. 35. 57. 35. Sense Anti-sense. TIMP-3. Sense Anti-sense. GAPDH. Sense Anti-sense. AT cycles (℃). 5’CCT TCT GCC TCA GTC TCC TG 5’ TCT GGA CTA CCC CAA GAT GG 233 5’ AAG AAG CCT CTA CCC CCA AA 5’CAA GGC TGA GAA CGG GAA GC 195 5’ AGG GGG CAG AGA TGA TGA CC. by using 0.2 mM dNTP, 2.5 μl of 10X PCR buffer which contains 20 mM Mg2+ per ml and 0.05 U/μl of Taq DNA polymerase in a 25 μl-PCR reaction. The RNA level of GAPDH was determined in every sample as an internal control. The PCR is carried out for 35 cycles of 30 seconds at 94℃, 30 seconds at primer-specific annealing temperature, and 30 seconds at 72℃. After amplification, the products were visualized by electrophoresis through a 2% agarose gel, and stained with ethidium bromide, and illumination with a UV lamp.. 9.

(17) Preparation of cell lysates Whole cell lysates were obtained from primary synovial cells and performed after washing with PBS using 50 μl golden lysis buffer: 20 mM Tris/HCl, pH 7.9; 137 mM NaCl; 5 mM EDTA; 1 mM EGTA, pH 8.0; 10 mM NaF; 1 mM sodium orthovanadate; 1. mM. sodium. pyrophosphate;. 0.1. mM. β. -Glycerophosphate;. 2. mM. phenylmethylsulfo-nylfluoride (PMSF), as well as 0.8 nM aprotinin, 10 nM leupeptin, and 5 mM DTT. The protein concentration was determined using the Bio-rad assay.. SDS PAGE 30 μg of cellular proteins was separated by SDS-PAGE. 12% resolving gel: 3M Tris pH8.9, 12~15%; Acrylamide/Bis (1:37.5), 12%; SDS, 0.1%; APS, 0.1%, and TEMED, 0.1%. 5% stacking gel: 1.5M pH6.8, 13%; Acrylamide/Bis (1:37.5), 5%; SDS, 1%; APS, 1%; TEMED, 0.2%.. Western Blotting After electrophoresis, the proteins were transferred onto PVDF membranes. Then the membranes are blocked with TBST containing 1% BSA and 0.2% sodium azide at room temperature for 1 hour and subsequently detected by immunoblotting using the following primary antibodies: anti-CTGF, 1:500; anti-PAR2, 1:2000; anti-pAkt, 1:1000; anti-pp38, 1:1000; anti-TIMP3, 1:500; anti-GAPDH, 1:10000; and anti-α-tubulin, 1:2000 at room temperature for 1~2 hours or 4℃ overnight. Then membranes were washed by using TBST for 10 min for three times. After washing, the membranes were incubated with the following secondary antibodies: anti-mouse HRP, anti-rabbit HRP, and anti-rat HRP in titer of 1:10000 in TBST at room temperature for 1 hour, then washing by using TBST for 10 min for three times. After washing,. membranes. were. visualized 10. with. ECL. detection. reagents. and.

(18) autoradiographic film.. Preparation of medium protein Conditioned media were collected, and concentrated 50-fold by using Microcon centrifugal filter devices. The concentrated media were denatured with 5X SDS loading dye for Western blotting analysis directly.. Collagen zymography To analyze MMP-13 activities, concentrated media were mixed with sample buffer without reducing agent or boiling. The sample was loaded into 0.5 mg/ml collagen-containing SDS-polyacrylamide gel, and then undergoing electrophoresis. After electrophoresis, remove SDS from the gel by washing in 2.5% Triton X-100 solution for 30 minutes at room temperature that allows enzymes to renature and degrade the protein substrate. Then, the gel was rinsed with developing buffer containing 25 mM Tris-HCl , 0.2 M NaCl, 6mM CaCl2, and 0.02% Brji for 30 minutes at room temperature and then incubation with developing buffer at 37℃ overnight. Afterward, zymographic activities were revealed by staining with 0.5% CBB R-250 for 30 minutes at room temperature and destaining with CBB destainer.. Statistical analysis Results were normalized to the copy numbers of GAPDH or α-tubuin. The mean and standard deviation were used to evaluate the CTGF, PAR-2, and TIMP-3 mRNA level. Student’s t test was used for comparison of the difference in expression level of target proteins and target gene between samples within the same set. The effects of the first stimulation were done as positive control and were analyzed as changes relative to un-stimulated baseline, using repeated measure analysis of variance, and 11.

(19) the inhibitory effects of the second drug co-treated with the first stimulant were analyzed as relative to positive control by using repeated measurement of variance. These analyses were performed individually and for the group as a whole. Statistical significance was set at P < 0.05 and very significant was set at P < 0.01.. 12.

(20) Results Induction of PAR-2 mRNA and protein level by IL-1 β in human primary synovial cells It has been reported that PAR-2 express in human chondrocytes and synovial cells and its expression level in human chondrocytes was up-regulated by IL-1 β in both mRNA and protein level5. In this study, we demonstrated IL-1 β can also up-regulate PAR-2 protein and mRNA level of human primary synovial cells. To induce PAR-2 mRNA expression, we treated human primary synovial cells with 10 ng/ml of IL-1 β for various time period then PAR-2 mRNA level was detected by RT-PCR. In order to induce PAR-2 protein level, we treated 10 ng/ml of IL-1 β for various time period during 12 to 48 hours then analyzed PAR-2 protein level by western blotting. As shown in Fig. 1, IL-1 β induced human primary synovial cells mRNA expression started at 30 minute and reached maximal effect at the time point of 1 hour (Fig. 1a). And IL-1 β induced PAR-2 protein expression at 12 hour that elevated up to 48 hour (Fig.1b). Inhibition of IL-1 β -induced PAR-2 expression by p38 MAPK inhibitor but not by MEK or JNK MAPK inhititor in human primary synovial cells Several recent studies suggest that trypsin and tryptase can activate PAR-2 via MAPK pathway, including ERK1/2, JNK, and p38 according to cell type15. In order to investigate the pathways which regulate human primary synovial cells PAR-2 level by IL-1 β, we pre-treated MAPK inhibitors for 1 hour followed by stimulating with IL-1 β for 12 hours, and PAR-2 protein level was determined by western blotting. As shown in Fig.2, SB202190, a p38 inhibitor, significantly inhibited IL-1β induced PAR-2 expression which was not inhibited by U0126, a MEK inhibitor or SP600125,. 13.

(21) a JNK inhibitor.. TGF- β 1 suppresses excess expression PAR-2 level induced by IL-1 β in human primary synovial cells According to a recent report, in human chondrocytes, TGF- β can decrease over-expressed PAR-2 protein level of cells from OA patients, but this phenomenon has not been observed in cell from non-OA patients that expressed normal PAR-2 level5. These results suggest that TGF- β is a regulator of PAR-2 expression in human chondrocytes. In light of this, we will demonstrate TGF-β can suppress excess expression of PAR-2 induced by IL-1 β in human primary synovial cells. According to our data, PAR-2 mRNA level was induced by IL-1 β after 30 minutes but has not reached maximal level (Fig. 1a). We pre-treated the cells with IL-1 β for 30 minutes to mimic early OA condition in order to increase PAR-2 protein level of human primary synovial cells, and then co-treated with various concentrations of TGF- β 1 for 12 hours or co-treated with 10 ng/ml of TGF- β 1 for 12, 24, and 48 hours. Then cellular PAR-2 protein level was determined by western blotting. As shown in Fig. 3, increasing PAR-2 level induced by IL-1 β could be down-regulated by TGF- β 1 dose-dependently (Fig. 3a) and time-dependently from 12 to 48 hours (Fig 3b).. TGF- β 1 inhibits IL-1 β mediated p38 MAPK activation, but not JNK or ERK level of human primary synovial cells Since we have demonstrated that IL-1 β induced PAR-2 expression via p38 MAPK pathway and TGF- β 1 could reduce this induction, we next investigated the mechanisms by which TGF- β 1 inhibits IL-1 β -mediated PAR-2 expression in human primary synovial cells. We pre-treated IL-1 β for 30 minutes to mimic OA condition followed by co-treating with TGF- β 1 for various time periods from 15 to 14.

(22) 60 minutes and then p-p38, p-ERK, and p-JNK level were analyzed using western blotting. As shown in Fig. 4, TGF- β 1 reduced IL-1 β mediated p-p38 level, at the time points of 15, 30, and 60 minutes. But TGF- β 1 had no inhibition effect to p-ERK or p-JNK of human primary synovial cells.. TGF- β 1 induces CTGF mRNA and protein expression level of human primary synovial cells It was reported that TGF- β 1 is a strong inducer of CTGF in many cell types, such as fibroblast, hepotocytes, renal cells, and keratinocytes12. First, we demonstrated that TGF- β 1 can induce CTGF mRNA and protein expression in human primary synovial cells. We treated human primary synovial cells with different dosage of TGF- β 1 for 6 hours and CTGF mRNA level was demonstrated by RT-PCR. As shown in Fig. 5, TGF- β 1 induced CTGF mRNA expression in human primary synovial cells dose-dependently (Fig. 5a). Then we treated 10 ng/ml of TGF- β 1 for 12, 24, and 48 hours or various concentration of TGF- β 1 for 24 hours and both cellular and medium CTGF protein levels were determined by western blotting. As shown in Fig. 5, TGF- β 1 induced CTGF production in a time-dependent manner after 12 hours treatment and persisted to at least 48 hours (Fig. 5b). TGF- β 1 induced CTGF production also in a dose-dependent manner after 24 hours of treatment (Fig. 5c).. CTGF suppresses PAR-2 protein level induced by IL-1 β in human primary synovial cells Because CTGF is the major down-stream molecular of TGF- β 1 in human primary synovial cells, we wondered whether inhibition of PAR-2 by TGF- β 1 not only via direct inhibition of p38 pathway, but also through the induction of CTGF 15.

(23) which further reduces PAR-2 protein level. Next, we have to demonstrate that PAR-2 induced by IL-1 β can be repressed by CTGF. We pre-treated human primary synovial cells with 10 ng/ml of IL-1 β for 30 minutes and then treated 50 ng/ml of CTGF for 12 and 24 hours16, 17, then cellular PAR-2 protein level was detected by western blotting. As shown in Fig. 6, CTGF repressed IL-1β induced PAR-2 protein level after 24 hours of treatment.. CTGF inhibits IL-1 β -mediated p38 MAPK activation, but activated ERK pathway of human primary synovial cells According to available references, CTGF is an anabolic factor in OA process because this factor stimulates cell proliferation and ECM marker production, such as type II collagen and aggrecan in human chondrocytes16, 17. And CTGF was known to interact with TGF- β and present TGF- β to TGFRII12, so it can enhance TGF- β activity. In light of this, we thought that it was possible CTGF enhances TGF- β -mediated p38 inhibition. To prove our hypothesis, we treated 50 ng/ml of CTGF to human primary synovial cells for 15, 30, and 60 minutes following exposed to 10 ng/ml of IL-1 β and p-p38 level was determined by western blotting. As shown in Fig. 7, CTGF inhibited IL-1 β induced p-p38 level but not p-ERK or p-JNK level after 30 minutes of treatment (Fig. 7a) and when treated with CTGF singly, it does not affect p-p38 level of human primary synovial cells (Fig. 7b). In order to validate that TGF- β inhibits p-p38 is mediated by CTGF- TGF- β interaction, we compared p-p38 inhibition effect between treating with TGF- β 1 singly, CTGF singly, and combination treated with TGF- β 1 and CTGF. As shown in Fig. 7, when combination treated with TGF- β 1 and CTGF, it had better p-p38 inhibition effect than treated with TGF- β 1 or CTGF singly (Fig 7c).. 16.

(24) TGF- β 1 down-regulates TIMP-3 mRNA and protein level of human primary synovial cells It was reported that TGF- β 1 can reduce MMPs level in human chondrocytes due to its ability to induce TIMP-3, a nature inhibitor of MMPs33 in vivo. So, we also investigated whether TGF- β can induce TIMP-3 production in human primary synovial cells. We treated 10 ng/ml of TGF- β 1 in a time course then TIMP-3 mRNA level was determined by RT-PCR and protein level was detected by western blotting. As shown in Fig. 8, TIMP-3 mRNA level was induced by TGF- β 1 after 1 hour treatment and had best effect at the time point of 4th hour (Fig. 8a). TGF- β 1 also induced TIMP-3 protein expression time-dependently during 6 to 24 hours after treatment (Fig. 8b).. TGF- β 1 induces TIMP-3 expression via Akt pathway in human primary synovial cells It has been reported that TGF- β 1 induced TIMP-3 expression through Akt pathway in human chondrocytes33. We then determined whether this induction through the same pathway in human primary synovial cells. We treated 10 ng/ml of TGF- β 1 in various time periods from 15 to 240 minutes and then analyzed p-Akt level by western blotting. As shown in Fig. 9, TGF- β 1 induced p-Akt level after 120 minutes of treatment and this induction persisted at least 240 minutes (Fig. 9a). To confirm our results, we pre-treated SH-6, an inhibitor of Akt, for 30 minutes to block Akt pathway, and then co-treated with 10 ng/ml of TGF- β 1, then analyzed TIMP-3 mRNA and protein level at the time point of 4 hour and 24 hour by RT-PCR and western blotting. As shown in Fig. 8, SH-6 almost blocked total TGF- β 1-induced TIMP-3 mRNA (Fig. 9b) and protein (Fig. 9c) expression.. 17.

(25) TIMP-3 blocks MMP-13 activity in culture medium of human primary synovial cells It is known that TIMP-3 inhibits MMP-13 by competing the catalytic pocket of MMP-13 with its substrate, collagen. So we analyzed MMP-13 activity in culture medium of human primary synovial cells to make sure that TIMP-3 functioned. We first pre-treated 10 ng/ml of IL-1 β for 12 hours to induce MMP-13 secretion from human primary synovial cells, and then co-treated with 10 ng/ml of TGF- β 1 for 12 hours to induce TIMP-3 production and then we investigated MMP-13 activity in conditioned medium by collagen zymograpgy. As shown in Fig. 10, MMP-13 activity was significantly induced by IL-1 β, but when we treated the cells with TGF- β 1, TIMP-3 was produced and MMP-13 activity was blocked.. Blocking of MMP-13 function resulted in down-regulation of PAR-2 protein level of human primary synovial cells According to available reports, it was observed that reducing of MMP-13 activity also reduced PAR-2 protein level in human chondrocytes, because inhibition of MMP-13 could slow down cartilage break and then reducing stimulating factor for proinflammatory cytokines releasing, so inhibiting of MMP-13 function leads to reduce PAR-2 expression indirectly. Based on available study, concentration of MMP-13 in synovial fluid of OA patients lies in 0.1~1.5 ng/ml45, so we treated various dosages of MMP-13 for 24 hours and then cellular PAR-2 protein level was determined by western blotting. As shown in Fig. 11, when we treated the cells with higher concentration of MMP-13, PAR-2 protein had higher expression level, but when MMP-13 level reduced, PAR-2 protein level was lowered.. 18.

(26) Discussion This study is the first report to demonstrate the pathway how PAR-2 is repressed by TGF- β in human primary synovial cells. According to available reports, we know that PAR-2 is unique in its distribution property because PAR-2 is the only member of this family expressed in human chondrocytes5. In contrast, human primary synovial cells express PAR-1, PAR-2, and PAR-35. But in our study, we observed that PAR-1, PAR-2, and PAR-3 express in both human chondrocytes and human primary synovial cells (AF. 2), this may be due to the polymorphism of human. Nevertheless, we found that the mRNA levels of PAR-1, PAR-2, and PAR-3 were significantly induced by IL-1 β in human chondrocytes, but only PAR-2 mRNA was up-regulated by IL-1 β in human primary synovial cell. In addition, PAR-2 had higher expression level in human primary synovial cells than that in human chondrocytes (AF. 2), which suggests that PAR-2 plays a critical role of OA in human primary synovial cells. PAR-1 is related with chemokine release, and PAR-2 is directly related with inflammation15. It is possible that during early phase of OA, IL-1β stimulates PAR-2 expressing by human primary synovial cells via autocrine pathway and results more IL-1β secreting to up-regulate PAR-1 of human chondrocytes and stimulating chemokine such as IL-6, IL-8, eotaxin and MCP-1 releasing by human chondrocytes that induce OA severity. So PAR-2 might be a novel ideal therapeutic target to prevent OA from progressing. In human primary synovial cells, IL-1β up-regulated PAR-2 gene expression in short time (Fig. 1a) suggesting that PAR-2 expresses mainly occur in early phase of acute inflammation and induction of PAR-2 protein level by IL-1 β from 12 to 48 hour in human primary synovial cells (Fig. 1b) suggesting that PAR2-mediated inflammatory response in OA synovial cells through an amplification effect. In our studies, we found that TGF-β inhibited PAR-2 protein level induced by 19.

(27) IL-1 β time dependently (Fig. 3b), and the inhibition effect persisted for at least 48 hours, suggesting that TGF- β might inhibit IL-1 β induced PAR-2 through multiple pathways and we have proved that TGF- β inhibits PAR-2 expression in human primary. synovial. cells. through. three. pathways:. directly. inhibiting. p38. phosphorelation or by inducing TIMP-3 which in turn results MMP-13 dysfunction to ease inflammation cascade, and by induction CTGF to block p38 so that inhibit inflammatory cells proliferating. It was seen that TGF- β inhibited p-p38 pathway, the major signal pathway which mediated PAR-2 expression by human primary synovial cells (Fig. 2), soon after 15 minutes of treatment (Fig. 4). This is the first mechanism of TGF- β suppressed PAR-2 which occurred at the time point of 12 to 24 hour we observed in Fig. 3b. It should be noticed that PAR-2 was up-regulated under higher MMP-13 exposure for 24 hours in human primary synovial cells (Fig. 11). This might be due to higher MMP-13 enhanced further inflammatory response of synovial cells and stimulating IL-1β secretion, so we know that PAR-2 might be up-regulated by MMPs indirectly. Hence, blocking MMP-13 can also prevent over-expression of PAR-2. In light of this, besides of direct inhibited p38 pathway, TGF- β might inhibit PAR-2 through TGF β -TIMP-MMP-PAR2 cascade. According to our studies, TIMP-3 protein was significantly induced by TGF- β after 6 to 24 hours of treatment (Fig. 8b) to block MMP-13. Once MMP-13 was blocked, PAR-2 level decreased within 24 hours (Fig. 11). In brief, inhibiting PAR-2 by TGF β -TIMP-MMP-PAR2 cascade lies during 24th to 36th hour. It was observed that CTGF activated ERK but inhibited p38 pathway induced by IL-1β (Fig 7a and 7b) in human primary synovial cells, this is different with CTGF in human chondrocytes which activated both ERK and p38 pathway12. CTGF stimulates proliferation via p38 MAPK results in chondrocytes growth and differentiation 20.

(28) through p42/p44 MAPK pathway results in ECM specific marker expression. In contrast, CTGF activated p42/p44 rather than p38 in human primary synovial cells, suggests that CTGF mediates differentiation and ECM specific marker expression (data not shown) via p42/p44 pathway but not proliferation via p38 pathway (Fig. 7a, and 7b). This is the reason that over-expression of CTGF in human chondrocytes resulting in osteophytes formation but over-expression of CTGF in human primary synovial cells does not lead to osteophytes formation. It is noticeable that CTGF enhanced TGF- β -mediated p38 inhibition in human primary synovial cells (Fig. 7c). This suggests that TGF- β also inhibits p38 through CTGF and CTGF enhances TGF- β activity by presenting TGF- β to TGFR12. So, another pathway that TGF- β inhibits p38 is by inducing CTGF to reduce phospho-p38 level. We observed that TGF- β significantly induced CTGF protein after 12 to 48 hours (Fig. 5b) and then PAR-2 was inhibited by CTGF to repress PAR-2 after 24 hours (Fig.6). So, inhibition of p38 induced by IL-1β via TGF β -CTGF-p38 cascade occurs during 36th to 48th hour. In summary, our studies suggest that TGF- β inhibits IL-1β induced PAR-2 expression level through three pathway at various time points (AF. 4). First, TGF- β directly inhibits IL-1 β induced p-p38 level within 24 hours (AF. 5). Secondary, TGF- β induces TIMP-3 expression and results in suppression of MMP-13 function, so that PAR-2 over-expression can be eased off. In a word, inhibiting of PAR-2 expression by TGF- β -TIMP3-MMP13 cascade occurs during 24th to 36th hour (AF. 6). And finally, TGF- β induces CTGF after 12 hours, and CTGF reduces PAR-2 of human primary synovial cells after 24 hours (Fig. 6), so TGF β -CTGF-PAR2 cascade occurs during 36th to 48th hour (AF. 7). Because PAR-2 plays an important role in early phase of OA, it might be a novel ideal therapeutic target to prevent OA from progressing. 21.

(29) References [1] Shunsuke Omoto, Keiichiro Nishida, Yuichiro Yamaai, Motoi Shibahara, Takashi Nishida, Takeshi Doi, Hiroshi Asahara, Tohru Nakanishi, Hajime Inoue. and Masaharu Takigawa. Expression and localization of connective tissue growth factor (CTGF/Hcs24/CCN2) in osteoarthritic cartilage. Osteoarthritis and Cartilage (2004) 12, 771-778. [2] H. M. van Beuningen, H. L. Glansbeek, P. M. van der Kraan and W. B. van den Berg. Osteoarthritis-like changes in the murine knee joint resulting from intra-articular transforming growth factor- β injections. Osteoarthritis and Cartilage (2000) 8, 25-33. [3] Peter M. van der Kraan and Wim B. van den Berg. Anabolic and destructive mediators in osteoarthritis. Current Opinion in Clinical Nutrition and Metabolic Care (2000) 3: 205-211. [4] F. Redini, P. Galera, A. Mauviel, G. Loyau and J.-P. Pujol. Transforming growth factor- β stimulates collagen and glycosaminoglycan biosynthesis in cultured rabbit articular chondrocytes. FEBS Lett (1988) 234: 172-6. [5] Y. Xiang, K. Masuko-Hongo, T. Sekine, H. Nakamura, K. Yudoh, K. Nishioka. and T. Kato. Expression of proteinase-activated receptors (PAR)-2 in articular chondrocytes is modulated by IL-1 β, TNF-α and TGF- β. OsteoArthritis and Cartilage (2006) 14, 1163-1173. [6] JP Pelletier, C Boileau, F Mineau, C Geng, M Boily, J MarteI-Pelletier. Upregulation of proteinase-activated receptor (PAR)-2 in human osteoarthritic tissues: A newly pathway for the mediation of joint destruction. Osteoarthritis and Cartilage (2003) 13, Supplement A. [7] Elizabeth B. Kelso, William R. Ferrell, John C. Lockhart, Iona Elias-Jones, Todd Hembrough, Lynette Dunning, J. Alastair Gracie, and Iain B. McInnes. 22.

(30) Expression and Proinflammatory Role of Proteinase-Activated Receptor 2 in Rheumatoid Synovium. Arthritis & Rheumatism (2007) 56, 3, 765-771. [8] S. Varghese, P. Theprungsirikul B.S., S. Sahani B.S., N. Hwang B.S., K. J. Yarema, and J. H. Elisseeff. Glucosamine modulates chondrocyte proliferation, matrix synthesis, and gene expression. OsteoArthritis and Cartilage (2007) 15, 59-68. [9] K. Tateishi, C. Higuchi, W. Ando, K. Nakata, J. Hashimoto, D. A. Hart, H. Yoshikawa. and N. Nakamura. The immunosuppressant FK506 promotes development of the chondrogenic phenotype in human synovial stromal cells via modulation of the Smad signaling pathway. OsteoArthritis and Cartilage (2007) 15, 709-718. [10] J. A. Roman-Blas, D. G. Stokes. and S. A. Jimenez. Modulation of TGF- β signaling by proinflammatory cytokines in articular chondrocytes. OsteoArthritis and Cartilage (2007). Article in press. [11] Helga Lorenz, Wiltrud Richter. Osteoarthritis: Cellular and molecular changes in degenerating cartilage. Progress in Histochemistry and Cytochemistry (2006) 40, 135-163. [12] Andrew Leask and David J. Abraham. The role of connective tissue growth factor, a multifunctional matricellular protein, in fibroblast biology. Biochem. Cell Biol. (2003) 81, 355-363. [13] G.S. Cottrell, S. Amadesi, F. Schmidlin and N. Bunnett. Protease-activated receptor 2: activation, signalling and function. Biochemical Society Transactions (2003) 31, part 6. [14] Marta Ruiz-Ortega, Juan Rodríguez-Vita, Elsa Sanchez-Lopez, Gisselle Carvajal, Jesús Egido. TGF- β signaling in vascular fibrosis. Cardiovascular Research (2007). Article in press. 23.

(31) [15] Scotter. Macfarlane, Michael J. Seatter, Toru Kanke, Gary D. Hunter, and Robin Plevin. Proteinase-Activated Receptors. Pharmacol Rev (2001) 53, 245-282. [16] Takashi Nishida, Tohru Nakanishi, Masahiro Asano, Tsuyoshi Shimo, and Masaharu. Takigawa1.. Effects. of. CTGF/Hcs24,. a. Hypertrophic. Chondrocyte-Specific Gene Product, on the Proliferation and Differentiation of Osteoblastic Cells in vitro. Journal of Cellular Physiology (2000) 184, 197-206. [17] Takashi Nishida, Satoshi Kubota, Tohru Nakanishi, Takuo Kuboki, Gen Yosimichi, Seiji Kondo, and Masaharu Takigawa. CTGF/Hcs24, a Hypertrophic Chondrocyte-Specific Gene Product, Stimulates Proliferation and Differentiation, but Not Hypertrophy of Cultured Articular Chondrocytes. Journal of Cellular Physiology (2002), 192, 55-63. [18] Sven Knecht a, Benedicte Vanwanseele, Edgar Stussi. A review on the mechanical quality of articular cartilage – Implications for the diagnosis of osteoarthritis. Clinical Biomechanics (2006) 21, 999-1012. [19] M. F. L’Hermette, C. Tourny-Chollet, G. Polle, F. H. Dujardin. Articular Cartilage, Degenerative Process, and Repair: Current Progress. Orthopedics and Biomechanics (2006) 27, 738-744. [20] Jennings L, Wu L, King KB, Hammerle H, Cs-Szabo G, Mollenhauer J. The effects of collagen fragments on the extracellular matrix metabolism of bovine and human chondrocytes. Connect Tissue Res (2001) 42, 71-86. [21] Transforming growth factor- β is a potent inhibitor of IL-1 induced protease activity and cartilage proteoglycan degradation. Biochemical and Biophysical ResearchCommunications. , (1988) 30, 157: 3, 1352-1359. [22] Transforming growth factor-beta causes partial inhibition of interleukin 1 Stimulated cartilage degradation in vitro. Biochemical and Biophysical ResearchCommunications. (1989) 14, 144-150. 24.

(32) [23] TGF- β exerts opposite effects from IL-1 β on cultured rabbit articular chondrocytes through reduction of IL-1 β receptor expression. Arthritis Rheum (1993) 36, 44-50. [24] Verrecchia F, Pessah M, Atfi A, Mauviel A. Tumor necrosis factor-alpha inhibits transforming growth factor- β /Smad signaling in human dermal fibroblasts via AP-1 activation. J Biol Chem (2000) 275, 30226-31. [25] Verrecchia F, Tacheau C, Wagner EF, Mauviel A. A central role for the JNK pathway in mediating the antagonistic activity of pro-inflammatory cytokines against transforming growth factor- β -driven SMAD3/4- specific gene expression. J Biol Chem (2003) 278, 1585-93. [26] Bitzer M, von Gersdorff G, Liang D, Dominguez-Rosales A, Beg AA, Rojkind M, et al. A mechanism of suppression of TGF- β /SMAD signaling by NFkappa B/RelA. Genes Dev (2000) 14, 187-97. [27] Germaine F.J.D. Benus, Albertus T.J. Wierenga, David J.J. de Gorter, Jan Jacob Schuringa, Arie¨tte M. van Bennekum, Loes Drenth-Diephuis, Edo Vellenga, and Bart J.L. Eggen. Inhibition of the Transforming Growth Factor β (TGF β) Pathway by Interleukin-1β Is Mediated through TGF β -activated Kinase 1 Phosphorylation of SMAD3. Molecular Biology of the Cell (2005) 16, 3501-3510. [28] 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 β: studies in anatomically intact cartilage in vitro and in vivo. Ann Rheum Dis (1993) 52, 185-91. [29] Van Beuningen HM, van der Kraan PM, Arntz OJ, van den Berg WB. In vivo protection against interleukin- 1-induced articular cartilage damage by transforming growth factor-b 1: age-related differences. Ann Rheum Dis (1994) 25.

(33) 53, 593-600. [30] Lum ZP, Hakala BE, Mort JS, Recklies AD. Modulation of the catabolic effects of interleukin-1 b on human chondrocytes by transforming growth factor-beta. J Cell Physiol (1996) 166, 351-9. [31] Verrecchia F, Mauviel A. TGF-beta and TNF-alpha: antagonistic cytokines controlling type I collagen gene expression. Cell Signal (2004), 16, 873-80. [32] Christelle Boileau1, Nathalie Amiable, Johanne Martel-Pelletier, Hassan Fahmi, Nicolas Duval and Jean-Pierre Pelletier. Activation of proteinase-activated receptor 2 in human osteoarthritic cartilage upregulates catabolic and proinflammatory pathways capable of inducing cartilage degradation: a basic science study. Arthritis Research & Therapy (2007), 9, R121. [33] Hamid Yaqoob Qureshi, Rasheed Ahmad, Judith Sylvester, Muhammad Zafarullah. Requirement of phosphatidylinositol 3-kinase/Akt signaling pathway for regulation of tissue inhibitor of metalloproteinases-3 gene expression by TGF- β in human chondrocytes. Cellular Signalling (2007) 19, 1643-1651. [34] Hideaki Nagase, Robert Visse , Gillian Murphy. Structure and function of matrix metalloproteinases and TIMPs. Cardiovascular Research (2006) 69, 562-573. [35] Robert Visse and Hideaki Nagase. Matrix Metalloproteinases and Tissue Inhibitors of Metalloproteinases: Structure, Function, and Biochemistry. Circ Res. (2003) 92, 827-839. [36] Xiaoyun Fu, William C. Parks, Jay W. Heinecke. Activation and silencing of matrix metalloproteinases. Seminars in Cell & Developmental Biology (2008) 19, 2-13. [37] David Danielpour, Kyung Song. Cross-talk between IGF-I and TGF-b signaling pathways. Cytokine & Growth Factor Reviews (2006), 17, 59–74. [38] Wrana JL, Attisano L. The Smad pathway. Cytokine Growth Factor Rev (2000) 26.

(34) 11(1-2): 5-13. [39] Wrana JL. TGF-beta receptors and signalling mechanisms. Miner Electrolyte Metab (1998) 24,120-30. [40] Wrana JL. Regulation of Smad activity. Cell (2000) 100(2), 189-92. [41] Wrana JL, Attisano L. MAD-related proteins in TGF-beta signalling. Trends Genet (1996) 12, 493-6. [42] S. Andereya, N. Streich, B. Schmidt-Rohlfing, T. Mumme, R. Muller-Rath, U. Schneider. Comparison of modern marker proteins in serum and synovial fluid in patients with advanced osteoarthrosis and rheumatoid arthritis. Rheumatol Int (2006) 26: 432-438. [43] Ludwig Institute for Cancer Research, Uppsala University, P.O. Box 595, Biomedical Center, SE-751 24 Uppsala, Sweden. Dynamic control of TGF-b signaling and its links to the cytoskeleton. FEBS Letters (2008). Article in press. [44] Horiuchi K, Saito S, Sasaki R, Tomatsu T, Toyama Y. Expression of granzyme B in human articular chondrocytes. J Rheumatol (2003) 30, 1799-810. [45] S. Andereya A N. Streich A B. Schmidt-Rohlfing T. Mumme A R. Muller-Rath A U. Schneider. Comparison of modern marker proteins in serum and synovial fluid in patients with advanced osteoarthrosis and rheumatoid arthritis. Rheumatol Int (2006) 26, 432-438.. 27.

(35) 28.

(36) 29.

(37) Fig.1 Induction of PAR-2 mRNA and protein level by IL-1 β in human primary synovial cells. (a) Human primary synovial cells were incubated in 10 ng/ml of IL-1 β in serum-free DMEM for 30, 60, and 90 minutes following 24 hours if starvation. IL-1 β significantly induced human synovial primary cells PAR-2 mRNA expression started at 30 minute and reach maximal effect at the time point of 1 hour, and this effect persisted for at least 90 minutes after treatment. Equal loading of DNA samples in each lane was indicated by equal intensity of internal control, GAPDH. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2 to lane 4 were compared with lane 1. Star-labled lane represents P < 0.01 and was thought as very significant difference. (b) Human primary synovial cells were incubated in 10 ng/ml of IL-1 β for 12, 24, and 48 hours. IL-1 β significantly induced PAR-2 protein expression at 12, 24, and 48 hours. Equal loading of protein samples in each lane was indicated by equal intensity of loading control protein, GAPDH. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2 to lane 4 were compared with lane 1. Star-labled lane represents P < 0.01 and was thought as very significant difference.. 30.

(38) 31.

(39) Fig. 2 Inhibition of IL-1 β -induced PAR-2 expression by p38 MAPK inhibitor but not by MEK or JNK MAPK inhibitor. Human primary synovial cells were pre-treated with 5 μM of U0126, SP600125, and SB202190 in serum-free DMEM for 1 hour and then incubated with 10 ng/ml of IL-1 β for 12 hours in serum-free DMEM. Only SB202190, a p38 inhibitor, significantly inhibited IL-1 β induced PAR-2 expression, but U0126, an ERK inhibitor and SP600125, a JNK inhibitor can’t significant inhibit IL-1 β induced PAR-2 expression. Equal loading of protein samples in each lane was indicated by equal intensity of loading control protein, GAPDH. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2 as positive control, lane 3 to lane 5 were compared with lane 2. Star-labled lane represents P < 0.01 and was thought as very significant difference.. 32.

(40) 33.

(41) 34.

(42) Fig. 3 TGF- β 1 suppresses PAR-2 protein level induced by IL-1 β in human primary synovial cells. (a) Human primary synovial cells were pre-treated with 10 ng/ml of IL-1 β in serum-free DMEM for 30 minutes and then incubated in 2, 5, 10, and 30 ng/ml of TGF- β 1 in serum-free DMEM for 12 hours. IL-1 β induced PAR-2 expression was significantly inhibited after treated with 2, 5, 10, and 30 ng/ml of TGF- β 1. Equal loading of protein samples in each lane was indicated by equal intensity of loading control protein, GAPDH. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2 as positive control, lane 3 to lane 6 were compared with lane 2. Single star-labeled lane represents P < 0.05 and was thought as significant difference. Double star-labled lane represents P < 0.01 and was thought as very significant difference. (b) Human primary synovial cells were pre-treated with 10 ng/ml of IL-1 β in serum-free DMEM for 30 minutes and then incubated in 10 ng/ml of TGF- β 1 for 12, 24, and 48 hours in serum-free DMEM. TGF- β 1 significantly inhibited IL-1 β induced PAR-2 expression after treated with 10 ng/ml of TGF- β 1 for 12, 24, and 48 hours. Equal loading of protein samples in each lane was indicated by equal intensity of loading control protein, GAPDH. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2, 4, and 6 as positive controls of lane 3, 5, and 7. Lane 2, 4, and 6, were compared with lane 1; lane 3 was compared with lane 2; Lane 5 was compared with lane 4; and lane 7 was compared with lane 6. Star-labeled lane represents P < 0.05 and was thought as significant difference.. 35.

(43) 36.

(44) 37.

(45) Fig. 4 TGF- β 1 inhibits IL-1 β -mediated p38 MAPK activation, but not JNK or ERK level. Human primary synovial cells were first incubated in serum-free DMEM for 24 hours and then pre-treated with 10 ng/ml of IL-1 β in serum-free DMEM for 30 minutes and then incubated in 10 ng/ml of TGF- β 1 for 15, 30, and 60 minutes in serum-free DMEM. TGF- β 1 significantly inhibited IL-1 β induced p-p38 level after treated with 10 ng/ml of TGF- β 1 for 15, 30, and 60 minutes. But TGF- β 1 had no inhibition effect to p-ERK or p-JNK level. Equal loading of protein samples in each lane was indicated by equal intensity of loading control protein, GAPDH. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2, 4, and 6 as positive controls of lane 3, 5, and 7. Lane 2, 4, and 6, were compared with lane 1; lane 3 was compared with lane 2; Lane 5 was compared with lane 4; and lane 7 was compared with lane 6. Single star-labeled lane represents P < 0.05 and was thought as significant difference. Double star-labled lane represents P < 0.01 and was thought as very significant difference.. 38.

(46) 39.

(47) 40.

(48) 41.

(49) Fig. 5 TGF- β 1 induces CTGF mRNA and protein expression level of human primary synovial cells. (a) Human primary synovial cells were incubated with 1, 2, 5, 10, and 30 ng/ml of TGF- β 1 for 6 hours in serum-free DMEM following 24 hours of starvation. TGF- β 1 significantly induced CTGF mRNA expression after treatment the cells with 1, 2, 5, 10, and 30 ng/ml of TGF- β 1. Equal loading of DNA samples in each lane was indicated by equal intensity of internal control, GAPDH. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2 to lane 6 were compared with lane 1. Star-labled lane represents P < 0.01 and was thought as very significant difference. (b) Human primary synovial cells were incubated in 10 ng/ml of TGF- β 1 in serum-free DMEM for 12, 24, and 48 hours. TGF- β 1 significantly induced cellular and medium CTGF protein level after 12, 24, and 48 hours. Equal loading of protein samples in each lane was indicated by equal intensity of loading control protein, α-tubulin. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2 to lane 4 were compared with lane 1. Star-labled lane represents P < 0.01 and was thought as very significant difference. (c) Human primary synovial cells were incubated with 1, 2, 5, 10, and 30 ng/ml of TGF- β 1 for 24 hours in serum-free DMEM. TGF- β 1 significantly induced cellular and medium CTGF protein level when treated with 1, 2, 5, 10, and 30 ng/ml of TGF- β 1. Equal loading of protein samples in each lane was indicated by equal intensity of loading control protein, α-tubulin. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2 to lane 6 were compared with lane 1. Star-labled lane represents P < 0.01 and was thought as very significant difference. 42.

(50) 43.

(51) Fig. 6 CTGF suppresses PAR-2 protein level induced by IL-1 β in human primary synovial cells. Human primary synovial cells were pre-treated with 10 ng/ml of IL-1 β in serum-free DMEM for 30 minutes and then incubated in 50 ng/ml of CTGF for 12 and 24 hours in serum-free DMEM. CTGF significantly inhibited IL-1 β induced PAR-2 expression after 24 hours. Equal loading of protein samples in each lane was indicated by equal intensity of loading control protein, GAPDH. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2 and lane 4 as positive controls of lane 3 and 5. Lane 2 and 4 were compared with lane 1; lane 3 was compared with lane 2 and Lane 5 was compared with lane 4. Star-labeled lane represents P < 0.01 and was thought as significant difference.. 44.

(52) 45.

(53) 46.

(54) 47.

(55) 48.

(56) 49.

(57) 50.

(58) Fig. 7 CTGF inhibits IL-1 β -mediated p38 MAPK activation, but activate ERK pathway of human primary synovial cells. (a) Human primary synovial cells were first incubated in serum-free DMEM for 24 hours and then pre-treated with 10 ng/ml of IL-1 β in serum-free DMEM for 30 minutes and then incubated in 50 ng/ml of CTGF for 15, 30, and 60 minutes in serum-free DMEM. CTGF significantly inhibited IL-1 β induced p-p38 level after 30 minutes of treatment. But CTGF have no inhibition effect to p-JNK level, and on the contrary, it induced p-ERK level after 15 and 30 minutes. Equal loading of protein samples in each lane was indicated by equal intensity of loading control protein, GAPDH. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2, 4, and 6 as positive controls of lane 3, 5, and 7. Lane 2, 4, and 6, were compared with lane 1; lane 3 was compared with lane 2; Lane 5 was compared with lane 4; and lane 7 was compared with lane 6. Star-labeled lane represents P < 0.01 and was thought as very significant difference. (b) Human primary synovial cells were first incubated in serum-free DMEM for 24 hours and then pre-treated with 10 ng/ml of IL-1β in serum-free DMEM for 30 minutes and then incubated in 50 ng/ml of CTGF for 30 minutes in serum-free DMEM. CTGF significantly inhibited IL-1 β induced p-p38 level after 30 minutes of treatment and treated CTGF singly did not increase p-p38 level. But CTGF have no inhibition effect to p-JNK level, and on the contrary, it induced p-ERK level after 30 minutes singly. Equal loading of protein samples in each lane was indicated by equal intensity of loading control protein, GAPDH. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2 as positive control, lane 3 was compared with lane 2 and 51.

(59) lane 4 was compared with lane 1. Star-labeled lane was thought as significant difference. (c) Human primary synovial cells were first incubated in serum-free DMEM for 24 hours and then pre-treated with 10 ng/ml of IL-1 β in serum-free DMEM for 30 minutes and then treated with 10 ng/ml of TGF- β 1 or 50 ng/ml of CTGF singly or combined for 30 minutes in serum-free DMEM. Singly treated of TGF- β 1 and CTGF each significantly reduced IL-1 β -induced p-p38 level but combination of the two drugs had more inhibiting effect than using them singly but there was no significant difference to p-ERK and p-JNK level. Equal loading of protein samples in each lane was indicated by equal intensity of loading control protein, GAPDH. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2 as positive control, lane 3 to lane 5 was compared with lane 2. Star-labeled lane represents P < 0.01 and was thought as very significant difference.. 52.

(60) 53.

(61) 54.

(62) Fig. 8 TGF- β 1 up-regulates TIMP-3 mRNA and protein level of human primary synovial cells. (a) Human primary synovial cells were incubated with 10 ng/ml of TGF- β 1 in serum-free DMEM for 1, 2, 4, and 6 hours following starvation for 24 hours. TIMP-3 mRNA was significantly up-regulated after 1, 2, 4, and 6 hours and had the best effect in the time of 4th hour. Equal loading of DNA samples in each lane was demonstrated by equal intensity of internal control, GAPDH. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2 to lane 5 were compared with lane 1. Star-labeled lane represents P < 0.05 and was thought as significant difference. (b) Human primary synovial cells were incubated with 10 ng/ml of TGF- β 1 in serum-free DMEM for 6, 12, and 24 hours. TIMP-3 protein was significantly induced after 6, 12, and 24 hours. Equal loading of protein samples in each lane was indicated by equal intensity of loading control protein, GAPDH. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2 to lane 4 were compared with lane 1. Star-labeled lane represents P < 0.01 and was thought as very significant difference.. 55.

(63) 56.

(64) 57.

(65) 58.

(66) Fig. 9 TGF-β1 induces TIMP-3 expression via Akt pathway in human primary synovial cells. (a) Human primary synovial cells were incubated in starvation medium for 24 hours and then treated with 10 ng/ml of TGF- β 1 in serum-free DMEM for 15, 30, 60, 120, and 240 minutes. p-Akt level was significantly induced after 120 minutes and persisted at least 240 minutes. Equal loading of protein samples in each lane was indicated by equal intensity of loading control protein, GAPDH. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2 to lane 6 were compared with lane 1. Star-labeled lane represents P < 0.01 and was thought as very significant difference. (b) Human primary synovial cells were incubated in starvation medium for 24 hours and then treated with 10 μM of SH-6 for 30 minutes followed by incubated with 10 ng/ml of TGF- β 1 in serum-free DMEM for 4 hours. TGF- β 1 induced TIMP-3 mRNA was almost inhibited by SH-6. Equal loading of DNA samples in each lane was indicated by equal intensity of loading internal control, GAPDH. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2 as positive control, lane 3 was compared with lane 2. Star-labeled lane represents P < 0.01 and was thought as very significant difference. (c) Human primary synovial cells were treated with 10 μM of SH-6 for 30 minutes followed by incubated with 10 ng/ml of TGF- β 1 in serum-free DMEM for 24 hours. TGF- β 1 induced TIMP-3 protein expression was significantly inhibited by SH-6. Equal loading of protein samples in each lane was indicated by equal intensity of loading control protein, GAPDH. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2 as positive control, lane 3 was compared with lane 2. Star-labeled lane represents P < 0.01 and was thought as very significant difference. 59.

(67) 60.

(68) Fig. 10 TIMP-3 blocks MMP-13 activity. Human primary synovial cells were incubated with 10 ng/ml of IL-1 β in serum-free DMEM for 12 hours and then co-incubated with 10 ng/ml of TGF- β 1 in serum-free DMEM for 12 hours. MMP-13 activity was significantly inhibited by TGF- β 1. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2 as positive control, lane 3 was compared with lane 2. Star-labeled lane represents P < 0.01 and was thought as very significant difference.. 61.

(69) 62.

(70) Fig. 11 Blocking of MMP-13 function results in down-regulation of PAR-2 protein level of human primary synovial cells. Human primary synovial cells were incubated with 2, 1, 0.5, and 0.1 ng/ml of MMP-13 in serum-free DMEM for 24 hours. PAR-2 protein level was down-regulated when treat with lower concentration of MMP-13. Equal loading of protein samples in each lane was indicated by equal intensity of loading control protein, GAPDH. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2 to lane 5 were compared with lane 1. Single star-labeled lane represents P < 0.05 and was thought as significant difference. Double star-labled lane represents P < 0.01 and was thought as very significant difference.. 63.

(71) 64.

(72) 65.

(73) 66.

(74) 67.

(75) AF. 2 Induction of PAR-1, PAR-2, PAR-3, and PAR-4 mRNA level by IL-1 β in human primary synovial cells and chondrocytes. Human primary synovial cells and chondrocytes were incubated in 10 ng/ml of IL-1 β in serum-free DMEM for 30, 60, and 90 minutes following 24 hours of starvation and PAR-1, PAR-2, PAR-3, and PAR-4 mRNA level of human primary synovial cells and chondrocytes were determined by RT-PCR. Equal loading of DNA samples in each lane was indicated by equal intensity of internal control, GAPDH. The data shown here are representative of three independent experiments. Lane 1 as basal control and lane 2 to lane 4 were compared with lane 1. Star-labled lane represents P < 0.05 and was thought as significant difference.. 68.

(76) AF. 3. Human primary synovial fibroblast (FLS) P2. 69.

(77) 70.

(78) 71.

(79) 72.

(80) 73.

(81)