COMPARATIVE CYTOKINE GENE EXPRESSION IN SYNOVIAL TISSUE OF EARLY RHEUMATOID ARTHRITIS AND SERONEGATIVE SPONDYLOARTHROPATHIES

British Journal of Rheumatology 1997;36:38 42 COMPARATIVE CYTOKINE GENE EXPRESSION IN SYNOVIAL TISSUE OF EARLY RHEUMATOID ARTHRITIS AND SERONEGATIVE SPONDYLOARTHROPATHIES J. D. CAN ETE, J. LLENA, A. COLLADO, R. SANMARTI, A. GAYA,* J. GRATACO S, M. BLAY and J. MUN OZ-GO MEZ Departments of Rheumatology and *Immunology, Hospital Clinic of Barcelona and University of Barcelona, Barcelona, Spain SUMMARY Interleukin 1-beta (IL-1 ), IL-2, IL-4, IL-5, IL-6, IL-8, tumour necrosis factor-alpha (TNF- ), transforming growth factor-beta (TGF- ) and granulocyte-macrophage colony-stimulating factor (GM-CSF) gene expression was determined in knee synovium of 16 patients with rheumatoid arthritis (RA) and 16 patients with seronegative spondyloarthropathies (SSP), by using polymerase chain reaction (PCR) amplification. The pattern of cytokines observed in RA synovium is of the macrophage-fibroblast type, with the highest expression of IL-1 and TGF-. GM-CSF and IL-2 bands were visualized in a minority of patients. Neither IL-4 nor IL-5 could be detected. No significant differences were observed in the cytokine profile between patients with early ( 12 months) and more advanced disease. No differences were observed according to gender, age, rheumatoid factor status and the duration of knee synovitis. The pattern of cytokines in the synovium of SSP patients is similar to that observed in RA patients and does not change in relation to disease duration. IL-2 was the only T-cell cytokine observed. These data provide evidence that the macrophage-fibroblast cells have an important role in early and more advanced rheumatoid synovitis, and show that this is also true for SSP peripheral synovitis. KEY WORDS: Gene, Cytokine, PCR, Synovium, Early disease, Rheumatoid arthritis, Seronegative spondyloarthropathies. CYTOKINES are polypeptide mediators of the immune and inflammatory response. Their action, generally at a paracrine or autocrine level, is pleiotropic and complex, with multiple relationships, and it is difficult to predict the exact role of any single cytokine in disease [1]. The role of cytokines in the pathogenesis of rheumatoid arthritis (RA) has been the subject of multiple studies that, generally, agree on the overproduction of proinflammatory cytokines in the rheumatoid joint [2]. There are several studies on cytokine expression in rheumatoid synovium, but most of them are focused on a limited spectrum of cytokines [3 9]. These studies, using immunohistochemistry [3, 4] or molecular biology techniques {Northern blot [5], in situ hybridization [6, 7] or polymerase chain reaction (PCR) [8, 9]}, detected expression of cytokines (protein or mrna) that originated predominantly in the macrophage-fibroblast cells [interleukin-1 alpha (IL-1 ) or IL-1, IL-6, tumour necrosis factor alpha (TNF- ), transforming growth factor beta (TGF- )], but it has been difficult to detect T-cell-derived cytokines (mainly IL-2 and IL-4). However, these studies analysed synovial samples of patients with a long-standing disease treated with anti-rheumatic drugs, which may alter the original profile of cytokine expression [10 12]. In this scenario, two questions merit being answered. Does the pattern of cytokines observed in these studies reflect the original cellular Submitted 6 March 1996; revised version accepted 20 June 1996. Correspondence to: J. D. Can ete, Department of Rheumatology, Hospital Clinic of Barcelona, Villarroel, 170, 08036 Barcelona, Spain. response, or does the pattern change with the duration of the disease? Is this pattern specific for RA? In order to answer these questions, we studied, by using PCR methodology, the mrna expression of a broad spectrum of cytokines [IL-1, IL-2, IL-4, IL-5, IL-6, IL-8, TNF-, TGF- and granulocytemacrophage colony-stimulating factor (GM-CSF)] in the synovial tissue of RA patients, with either early or more advanced disease. Moreover, a comparison with the profile of synovium cytokines in patients with peripheral seronegative spondyloarthropathies (SSP) was carried out. PATIENTS AND METHODS This study includes 32 patients: 16 suffered from RA (1987 ARA criteria). Rheumatoid factor was positive in 13 of them. Disease duration was 12 months in six patients (x S.D. = 4.5 3.7) and 12 months in 10 (x S.D.=80 80). The remaining 16 patients suffered from SSP [six with 12 months (x S.D. = 2.6 2.7) and 10 with 12 months (x S.D. = 90.7 71.5)]. Twelve of these patients had reactive arthritis (Calin s criteria; eight HLA-B27 positive) and four suffered from ankylosing spondylitis with peripheral arthritis (New York criteria; all HLA-B27 positive). Both RA and SSP patients were out-patients with active disease and knee synovitis, and none of them had been treated with slow-acting anti-rheumatic drugs (SAARDs) or prednisone in a dose higher than 5 mg/day 3 months before their inclusion in this study. The study was approved by the Ethical Committee of the Hospital Clinic of Barcelona. Demographic, clinical and biological characteristics of RA patients and SSP patients are shown in Table I. 38 1997 British Society for Rheumatology

CAN ETE ET AL.: CYTOKINE GENE EXPRESSION IN EARLY ARTHRITIS 39 Blood, for the determination of erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP), and synovial samples were obtained at the same time as clinical assessment. Clinical activity of synovitis was measured according to Thompson s articular index [13] and disability according to the Health Assessment Questionnaire (HAQ) previously validated for Spanish people [14]. Multiple samples of synovial tissue of the knee were obtained using a needle biopsy (Parker & Pearson). The inflammatory synovitis was confirmed in each case by haematoxylin eosin staining. The other samples were dipped in guanidinium-thiocyanate (GTC) buffer [15] and frozen in liquid nitrogen immediately. Synovial tissue RNA isolation and reverse transcription Total cellular RNA was isolated from snapfrozen synovial membranes by using GTC buffer for disruption of the tissue, extracted with phenol/ chloroform and precipitated with isopropanol, washed in 70% ethanol and dissolved in diethylpyrocarbonatetreated double-distilled (DEPCdd) water [15]. The RNA was spectrophotometrically quantitated, and a 2.5 g aliquot was reverse transcribed with 200 U of Moloney murine leukaemia virus reverse transcriptase according to standard procedures (Boehringer Ingelheim). The cdna was stored at 20 C and an aliquot was used in a PCR. PCR amplification Each PCR was carried out in a total volume of 50 l, containing 26.6 l of DEPCdd water, 5 l of cdna, 5 l of 10 PCR buffer (Promega, Madison, WI, USA), 8 l of dntp mix (1.25 mm each), 2.5 l of each primer (Clontech, CA, USA) and 0.4 l (2U)ofTaq DNA polymerase (Promega). The reaction was performed in a Thermocycler 480S (Perkin-Elmer, Norwalk, CT, USA) using a 35 cycle programme (95 C for 1 min, 60 C for 2 min and 70 C for 3 min). The reverse transcription and the DNA amplification were performed according to the manufacturer s protocol. TABLE I Demographic, clinical and biological data of patients RA SSP Number of patients 16 16 Sex (M/F) 11/5 7/9 Age (x S.D. yr) 53 ( 15) 37 ( 13) Disease duration (x S.D. months) 52 ( 73) 49 ( 68) HAQ score 1.7 ( 1.6) 0.48 ( 0.62) Thompson s index 193 ( 90) 130 ( 59) Knee synovitis duration (x S.D. months) 11.5 ( 14) 6.6 ( 6.4) Rheumatoid factor positive (%) 13 (81) 0 HLA-B27 positive (%) 0 12 (75) CRP (x S.D. mg/dl) 3.9 ( 4.5) 3.3 ( 4.3) ESR (x S.D. mm/h) 42.8 ( 28) 32.9 ( 33.8) RA, rheumatoid arthritis; SSP, seronegative spondyloarthropathies; HAQ, Health Assessment Questionnaire. FIG. 1. PCR amplification with specific primers for IL-4 and IL-5 from mrna derived from either non-stimulated PBMC cells or PHA-stimulated PBMC cells. Lane 1, 2-microglobulin (non-stimulated); lane 2, IL-4 (non-stimulated); lane 3, IL-4 (stimulated); lane 4, IL-5 (non-stimulated); lane 5, IL-5 (stimulated). Numbers indicate the size of expected PCR products and are expressed in base pairs. The primers utilized for amplification were IL-1, IL-2, IL-4, IL-5, IL-6, TGF-, TNF-, GM-CSF and IL-8 (Clontech, CA, USA). Previously, the primers were tested with phytohaemagglutinin (PHA)- stimulated peripheral blood mononuclear cells (PBMC) (Fig. 1). Following specific cdna amplification by PCR, ethidium bromide-stained 2% agarose gels were used to separate and visualize the amplified DNA. A HaeIII-digested X174 marker (Promega) was used in each run [16]. The identity of the amplified products was confirmed by their specific respective sizes. Amplifications with specific primers for 2-microglobulin were used in duplicate reactions as controls. To increase the sensitivity of the method, samples not yielding a PCR product were subjected to a second round of amplification. PCR results on the same samples were reproducible. Statistics Data were analysed by 2 test and Yate s correction when necessary. The differences of distribution in different continuous variables such as age, rheumatoid factor and synovitis duration were analysed by the Mann Whitney U-test in relation to the presence of each cytokine. RESULTS Profile of cytokine mrna expression in RA The pattern of cytokines observed in RA (Table II) is of the macrophage-fibroblast type, with the highest expression of IL-1 (94%) and TGF- (81%), and not as high expression of IL-6, TNF- and IL-8. GM-CSF was only expressed in one patient (6%). IL-2 is the only T-cell cytokine found (12.5%). Neither IL-4 nor IL-5 could be detected. No significant differences were

40 BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 1 TABLE II Cytokine mrna expression in synovial tissue of RA and SSP patients (% in parentheses) Disease duration (months) IL-1 IL-2 IL-4 IL-5 IL-6 IL-8 TNF- TGF- GM-CSF Rheumatoid arthritis 12 (n = 6) 6 (100) 0 0 0 3 (50) 3 (50) 2 (33) 5 (82) 0 12 (n = 10) 9 (90) 2 (20) 0 0 5 (50) 2 (20) 4 (40) 8 (80) 1 (10) Seronegative spondyloarthropathies 12 (n = 6) 3 (50) 1 (16) 0 0 2 (33) 1 (16) 2 (33) 4 (66) 2 (33) 12 (n = 10) 8 (80) 0 0 0 5 (50) 1 (10) 3 (33) 7 (70) 0 observed in the profile of cytokines between patients with early ( 12 months) and more advanced disease. There are fewer patients positive for IL-8 in the more advanced group. The only two patients with IL-2 expression had advanced disease (36 and 48 months). Moreover, no differences were observed when considering gender, age, rheumatoid factor (RF) status, HAQ score, Thompson s articular index and the duration of knee synovitis. Profile of cytokine mrna expression in SSP The pattern of cytokines in SSP patients is similar to that observed in RA patients (Table II), IL-1 and TGF- being the most common cytokines found. However, there is a smaller number of positive patients for IL-1, IL-6, IL-8 and TGF- in the group of SSP of 12 months with respect to RA patients. As in RA, GM-CSF and IL-2 were observed in very few patients. Neither IL-4 nor IL-5 was found. No significant FIG. 2. PCR amplification of cytokine mrnas with specific primers in one patient with reactive arthritis (A) and another patient with rheumatoid arthritis (B), both expressing IL-2. Lane 1, 2-microglobulin; lane 2, IL-1; lane 3, IL-2; lane 4, IL-4; lane 5, IL-5; lane 6, IL-6; lane 7, IL-8; lane 8, TNF- ; lane 9, TGF- ; lane 10, GM-CSF. Numbers indicate the size of expected PCR products and are expressed in base pairs. differences were observed in the cytokine expression when patients were classified according to the disease duration ( 12 or 12 months), gender, age, presence of HLA-B27, HAQ score, Thompson s articular index, the duration of knee synovitis and disease group (AS or reactive arthritis). Cytokine patterns in the synovium of one patient with reactive arthritis and another with RA are shown in Fig. 2. DISCUSSION In this study, cytokine patterns were assessed by PCR in the synovium of RA and SSP patients, with either early or more advanced disease. Our results show that RA patients have a profile of mrna cytokine expression in synovial tissue that is characteristic of the macrophage-fibroblast lineage, with the highest frequency of expression of IL-1 and TGF-, and not as high of IL-6, TNF- and IL-8. GM-CSF was found in only one patient. IL-2 was the only T-cell cytokine detected (12% of positivity). This cytokine pattern is not significantly altered by disease duration ( 12 or 12 months), gender, age, HAQ score, Thompson s articular index or RF status. This work also shows that the synovium of SSP patients has a pattern of cytokine expression similar to that of rheumatoid synovium. The results in RA patients are in good agreement with previous studies, in which the macrophagefibroblast cytokines are commonly detected in the rheumatoid joint, but T-cell cytokine expression is difficult to demonstrate [3, 6, 8, 9, 17, 18]. Firestein et al. [6], studying by in situ hybridization seven synovial samples of RA patients with long-standing disease, showed that the macrophage and the fibroblast-like type B synoviocyte were the origin of most of the IL-1, TNF-, IL-6, GM-CSF and TGF- transcription in the synovium. However, TGF- and GM-CSF were found in only 1 2% of the cells, and little or no interferon gamma (IFN ) mrna was detected. In our study, GM-CSF was rarely expressed, but TGF- was expressed in 80% of the patients. It is possible that these differences are due to disease duration and/or the sensitivity of the different methodologies used. The reason for the low T-cell cytokine expression in the synovium of RA patients is unclear, although recently it has been demonstrated that T cells are essential in the pathogenesis of the flare of RA synovitis [19]. It has been proposed that these cytokines could be expressed at levels too low to be detected by

CAN ETE ET AL.: CYTOKINE GENE EXPRESSION IN EARLY ARTHRITIS 41 methods previously used [20]. In this study, we determined cytokine patterns by PCR, a very sensitive method, and we confirmed a low expression of IL-2 and a null expression of IL-4 and IL-5 in the synovium of RA patients. Recently, Simon et al. [21], using PCR analysis of synovium mrna, detected expression of IL-1 in 60% and IFN in 80% of RA patients, but did not find IL-2 or IL-4 expression. The same investigators [21], by using the most sensitive method of hybridizing a radioactive probe to the PCR products, detected expression of IL-2 and IL-4 genes in 20% of the same samples. On the other hand, Dolhain et al. [12] also found mrna of IFN in RA synovium. However, Miossec [8] could not detect IL-4 by PCR, and Chen et al. [9] reported a very low expression of IL-2 and IL-4 mrna using PCR with radioactive nucleotides. Another proposed reason for the low T-cell cytokine expression is that it could be inhibited by TGF- [18, 22] which is highly expressed in our patients. Furthermore, the pattern of cytokine expression could also be affected by disease duration or previous anti-rheumatic therapy [23]. By contrast, our results clearly show that the cytokine pattern does not change significantly with the duration of the disease. A recent study [4] using a new immunohistochemical technique reported the predominance of mainly monocytemacrophage-derived cytokines (protein) in the RA synovium, although it also detected T-cell cytokines (including IL-5) in some samples. Globally, these findings support the paradigm that the inflammatory synovitis in RA is predominantly directed and perpetuated by the macrophage-fibroblast cells [24, 25]. Results of therapies directed against proinflammatory cytokines in RA patients are also in good agreement with this pathogenetic hypothesis [26]. However, it is possible that the original cellular response in the synovium has already been expressed in the pre-clinical stage of RA, when an aberrant immune response is detected [27]. Thus, in view of the results of this and the different studies mentioned, the role of T-cell cytokines in the pathogenesis of RA has not been fully demonstrated and remains controversial [19, 21]. A systematic study on cytokines in SSP patients had not been carried out before. Recently, a similar pattern of cytokines in the serum of patients with RA and SSP, with higher levels in the RA patients, was found by our group [28]. Moreover, the present study shows that the pattern of cytokine gene expression in the synovium of SSP patients is similar to that of the RA patients. Simon et al. [21] also studied the expression of T-cell cytokine mrna in knee synovium of nine patients with reactive arthritis, following infection by specific intracellular pathogens, by using qualitative PCR and hybridizing with radioactive probes, and detected high expression of IL-1 and IL-4 (60%) and low expression of IL-2 (16%). They did not assess IL-5 (Th2-derived cytokine probably involved in the stimulation of B-cell-secreting IgA). In our study, we also detected a comparable expression of IL-1 and IL-2, but no expression of IL-4 or IL-5. We do not have a satisfactory explanation for the discrepancies in the IL-4 expression between the two studies, although differences due to the type of patients studied (our patients did not have a demonstrable infection by serological or microbiological criteria) or in the methodologies could not be ruled out. These authors proposed that their reactive arthritis patients had a Th2 cytokine pattern, while in RA patients a Th1 cytokine pattern predominates. However, Ulfgren et al. [4] reported some samples of RA synovium with a Th1/Th2 cytokine pattern. Thus, new studies on the expression of Th1 (IFN, IL-2) and Th2 cytokines (IL-4 and IL-5) in RA and SSP patients are warranted. In conclusion, this study shows an overexpression of proinflammatory cytokines in the synovium of RA patients. The expression of T-cell cytokines was low (IL-2) or even absent (IL-4, IL-5). This profile of cytokines was not influenced by disease duration and was similar to that observed in patients with peripheral SSP. From this and other studies, it can be concluded that the macrophage-fibroblast cells have an important role in early and more advanced rheumatoid synovitis. This study shows that this is also true for SSP synovitis. Although there is experimental evidence implicating T-cell cytokines in RA pathogenesis [19], it is difficult to detect their mrna in synovial tissue. To clarify this important issue, further studies on these cytokines in early RA and SSP patients will be needed. ACKNOWLEDGEMENTS This work was supported by Grants from DGICYT (Ministerio de Sanidad), Project PM91-0042, and from Hospital Clinic of Barcelona. The authors are grateful to Jaume Farre s and Salvador Bartolome, from the Department of Biochemistry and Molecular Biology, Universidad Autónoma de Barcelona, for their valuable discussion on this work. REFERENCES 1. Feldmann M, Brennan F, Maini RN. Role of cytokines in autoimmune disease. In: Oppeheim JJ, Rossio JL, Gearing AJH, eds. Clinical applications of cytokines. Role in pathogenesis, diagnosis and therapy. Oxford: Oxford University Press, 1993:109 13. 2. Arend WP, Dayer JM. Cytokines and growth factors. In: Kelley WN, Harris ED Jr, Ruddy S, Sledge CB, eds. Textbook of rheumatology, 4th edn. Philadelphia: WB Saunders, 1993:227 47. 3. Nabil M, Yanni G, Poston R, Panayi GS. Cytokine expression in synovial membranes of patients with rheumatoid arthritis. Ann Rheum Dis 1994;52:870 5. 4. Ulfgren AK, Lindblad S, Klareskog L, Anderson J, Anderson U. Detection of cytokine producing cells in the synovial membrane from patients with rheumatoid arthritis. Ann Rheum Dis 1995;54:654 61. 5. Buchan G, Barret K, Turner M, Chantry D, Maini RN, Feldmann M. Interleukin-1 and tumor necrosis factor- mrna expression in rheumatoid arthritis: prolonged production of IL-1. Clin Exp Immunol 1988;3:449 55. 6. Firestein GS, Alvaro-Gracia JM, Maki R. Quantitative analysis of cytokine gene expression in rheumatoid arthritis. J Immunol 1990;144:3347 53. 7. Wood NC, Simons JA, Dickens E, Duff GW. In situ

42 BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 1 hybridization of IL-6 in rheumatoid arthritis. Clin Exp Immunol 1992;87:183 9. 8. Miossec P. Cytokine abnormalities in inflammatory arthritis. Baillie` re s Clin Rheumatol 1992;6:373 92. 9. Chen E, Keystone EC, Fish EN. Restricted cytokine expression in rheumatoid arthritis. Arthritis Rheum 1993;36:901 10. 10. Barrera P, Boerbooms AMTH, Jannssen EM et al. Circulating soluble TNF receptors, interleukin-2 receptor, tumor necrosis factor- and interleukin-6 levels in rheumatoid arthritis: longitudinal evaluation during methotrexate and azathioprine therapy. Arthritis Rheum 1993;36:1070 9. 11. Crilly A, Madhok R, Watson J, Capell HA, Sturrock RD. Production of interleukin-6 by monocytes isolated from rheumatoid arthritis patients receiving second-line drug therapy. Br J Rheumatol 1994;33:821 5. 12. Dolhain RJ, de Kuiper P, Verweij CL et al. Tenidap, but not nonsteroidal anti-inflammatory drugs, inhibits T-cell proliferation and cytokine induction. Scand J Immunol 1995;42:686 93. 13. Thompson PW, Silman AJ, Kirwan SR, Currey HLF. Articular index of joint inflammation in rheumatoid arthritis. Correlation with the acute-phase response. Arthritis Rheum 1987;6:618 23. 14. Esteve-Vives J, Batlle E, el Grupo para la Adaptación del HAQ a la Poblacio n Espan ola. Adaptación del Health Assessment Questionnaire (HAQ) a la población espan ola. Rev Esp Reumatol 1991;18:258 62. 15. Chomczynski P, Sacchi N. Single step method of RNA isolation by acid guanidinium thiocyanate-phenolchloroform extraction. Anal Biochem 1987;162:156 9. 16. Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual. Plainview, NY: Cold Spring Harbor Laboratory Press, 1989. 17. Chu CQ, Field M, Feldmann M, Maini RN. Localization of tumor necrosis factor- in synovial tissues and at the cartilage-pannus junction in patients with rheumatoid arthritis. Arthritis Rheum 1991;34:1125 32. 18. Miossec P, Naviliat M, Dupuy A, Sany J, Banchereau J. Low levels of interleukin-4 and high levels of transforming growth factor- in rheumatoid synovitis. Arthritis Rheum 1990;33:1180 7. 19. Jendro M, Ganten T, Matteson EL, Weyand CM, Goronzy JJ. Emergence of oligoclonal T cell populations following therapeutic T cell depletion in rheumatoid arthritis. Arthritis Rheum 1995;9:1242 51. 20. Panayi GS, Lanchbury JS, Kingsley GH. The importance of T cell in initiating and maintaining the chronic synovitis of rheumatoid arthritis. Arthritis Rheum 1992;36:729 35. 21. Simon AK, Seipelt E, Sieper J. Divergent T-cell cytokine patterns in inflammatory arthritis. Proc Natl Acad Sci USA 1994;91:8562 6. 22. Ahuja SS, Paliogianni F, Yamada K, Balow JB, Bonumpas DT. Effect of transforming growth factor- on early and late activation events in human T cell. J Immunol 1993;150:109 18. 23. Zvaifler NJ, Boyle D, Firestein GS. Early synovitis synoviocytes and mononuclear cells. Semin Arthritis Rheum 1994;23:11 6. 24. Firestein GS, Zvaifler NJ. How important are T cells in chronic rheumatic synovitis? Arthritis Rheum 1990;33:768 73. 25. Firestein GS, Zvaifler NJ. The role of T cells in rheumatoid arthritis. Ann Rheum Dis 1993;52:765. 26. Maini RN, Elliot MJ, Brennan FM et al. Monoclonal anti-tnf antibody as a probe of pathogenesis and therapy of rheumatoid disease. Immunol Rev 1995;144:195 223. 27. Aho K, Paulusuo T, Kurki P. Marker antibodies of rheumatoid arthritis: Diagnostic and pathogenetic implications. Semin Arthritis Rheum 1994;23:379 87. 28. Gratacós J, Collado A, Filella X et al. Serum cytokines (IL-6, TNF, IL1 and IFN) in ankylosing spondylitis: A close correlation between serum IL-6 and disease activity and severity. Br J Rheumatol 1994;33:927 31.