Presence of autoantibodies to the glycolytic enzyme ╨Ю┬▒-enolase in sera from patients with early rheumatoid arthritis.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 46, No. 5, May 2002, pp 1196–1201 DOI 10.1002/art.10252 © 2002, American College of Rheumatology Presence of Autoantibodies to the Glycolytic Enzyme ␣-Enolase in Sera From Patients With Early Rheumatoid Arthritis Vincent Saulot,1 Olivier Vittecoq,2 Roland Charlionet,1 Patrice Fardellone,3 Catherine Lange,4 Laure Marvin,4 Nadine Machour,1 Xavier Le Loët,2 Danièle Gilbert,1 and François Tron1 Objective. To identify a new autoantigen/ autoantibody population in rheumatoid arthritis (RA) sera. Methods. Following a population-based recruitment effort, 255 patients with very early arthritis (median disease duration 4 months) were studied using different clinical, biologic, and radiologic assessments. After a followup period of 1 year, patients were classified as having RA (n ⴝ 145), non-RA rheumatic diseases (n ⴝ 70), and undifferentiated arthritis (n ⴝ 40). Patients’ sera were analyzed by one-dimensional (1D) and 2D Western blotting. The recognized 50-kd protein was analyzed by matrix-assisted laser desorption ionization–time-of-flight (MALDI-TOF) mass spectrometry (MS). RA serum reactivities were evaluated against the recombinant protein synthesized by an in vitro coupled transcription–translation system. Results. On 1D Western blots, 36 of the 145 RA sera bound to a 50-kd polypeptide. On 2D Western blots, anti–50-kdⴙ RA sera recognized a triplet of isoelectric point 6.5–7.0 and a molecular mass of 50 kd. The 3 spots of the triplet were analyzed by MALDI-TOF MS and were shown to correspond to human ␣-enolase. A goat anti-enolase antiserum, which recognized a band comigrating with the 50-kd antigen on 1D Western blots, gave a labeling pattern on 2D Western blots similar to that observed with anti–50-kdⴙ RA sera. Among the 36 RA sera that identified ␣-enolase in protein maps, only 8 recognized the recombinant (unmodified) ␣-enolase. The specificity of anti–␣-enolase antibodies for RA was 97.1%. Half of the anti–␣-enolase–positive RA patients were negative for both rheumatoid factor and antifilaggrin antibodies. The presence of anti–␣-enolase antibodies was the greatest predictive factor of radiologic progression in the first 66 RA patients included. Conclusion. Autoantibodies to ␣-enolase, an enzyme of the glycolytic pathway, are present in the sera of patients with very early RA and have potential diagnostic and prognostic value for RA. Rheumatoid arthritis (RA) has long been thought to be an autoimmune disease because various autoantigens have been shown to be the targets of synovial T cells and autoantibodies present in patients’ synovial fluid and/or sera (1). Although the autoimmune nature of RA remains controversial, these autoantibodies have received much attention because their characterization, namely, identification of their target antigen, may shed some light on the nature and origin of the immune process and, in fact, their presence might constitute a marker of the disease (2). Several autoantibody populations have been detected during the course of RA (1). For example, rheumatoid factor (RF) has long been considered to be the most sensitive and best predictive diagnostic marker of RA but has relatively poor specificity. Recently, Supported by INSERM, the Fondation de la Recherche Médicale, and the Programme Hospitalier de Recherche Clinique VERA-O (Very Early Rheumatoid Arthritis Outcome). Mr. Saulot is the recipient of a fellowship from the Association de Recherche sur la Polyarthrite. 1 Vincent Saulot, MS, Roland Charlionet, PhD, Nadine Machour, PhD, Danièle Gilbert, PhD, François Tron, MD, PhD: Institut National de la Santé et de la Recherche Médicale, Rouen, France; 2 Olivier Vittecoq, MD, PhD, Xavier Le Loët, MD: Institut National de la Santé et de la Recherche Médicale, and Centre Hospitalier Universitaire de Rouen, Rouen, France; 3Patrice Fardellone, MD, PhD: Centre Hospitalier Universitaire d’Amiens, Amiens, France; 4Catherine Lange, PhD, Laure Marvin, PhD: Centre de Spectrométrie de Masse, Mont Saint Aignan, France. Address correspondence and reprint requests to Danièle Gilbert, PhD, INSERM Unité 519, IFRMP-23, Faculté Mixte de Médecine et Pharmacie, 22, Boulevard Gambetta, 76183 Rouen, France. E-mail: email@example.com. Submitted for publication May 31, 2001; accepted in revised form January 11, 2002. 1196 AUTOANTIBODIES TO ␣-ENOLASE IN EARLY RHEUMATOID ARTHRITIS antifilaggrin antibodies (AFA) were shown to be more specific to RA (3). Furthermore, the antigenic determinants recognized by AFA and present in RA sera were shown to be generated posttranslationally at various sites on profilaggrin by arginine-residue deimination, changing them into citrulline residues (4,5). These observations prompted us to search for new autoantibody populations, including those directed against posttranslationally modified proteins, in the sera of RA patients using Western blotting of a 2-dimensional (2D) protein map and mass spectrometry. PATIENTS AND METHODS Patients and sera. Two hundred fifty-five patients with very early arthritis were recruited prospectively by rheumatologists and by general practitioners in private practice in 2 French regions: upper Normandy and the metropolitan area of Amiens. For inclusion criteria, patients were required to have swelling in at least 2 joints which had persisted for ⱖ4 weeks and evolved for ⬍6 months (median disease duration 4 months), and to have taken no local or systemic corticosteroids or disease-modifying antirheumatic drugs before inclusion. Exclusion criteria were inflammatory back pain, pregnancy, or breastfeeding. Patients were followed up for 1 year after the appearance of the first symptoms, at which time diagnoses were established by a committee of 5 expert rheumatologists. Three subgroups of patients were thus defined: 1) 145 patients with RA (subgroup 1), 2) 70 patients with non-RA rheumatic diseases (subgroup 2), and 3) 40 patients with undifferentiated arthritis (subgroup 3) (Table 1). The following markers were evaluated as previously described (3): RF measured by agglutination tests (latex fixation test and Rose-Waaler test) and by enzyme-linked immunosorbent assay (ELISA) (IgM, IgA, and IgG isotypes), AFA detected by indirect immunofluorescence, and anticalpastatin antibodies detected by ELISA. Radiographs of the hands and feet were carried out at inclusion and at the end of followup. They were read chronologically (in order of patient inclusion) by 2 experienced rheumatologists (OV and PF), using the van der Heijde–modified Sharp method of scoring for radiographic damage (6), for the first 66 RA patients included in subgroup 1. The criterion for severity of the disease was defined by progression of radiologic damage during the followup period. Sera were obtained at the time of inclusion from all patients enrolled in the cohort, as well as from 26 patients with systemic lupus erythematosus (SLE), 13 with systemic sclerosis (SSc), and 14 with primary biliary cirrhosis. Antibodies. Polyclonal goat IgG raised against a carboxy-terminal–region peptide common to ␣, ␤, and ␥ isoforms of mouse, rat, and human enolase (200 g/ml; Santa Cruz Biotechnology, Santa Cruz, CA) was used at a dilution of 1:100. Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting. Sample preparation. Fresh human placenta was minced and was resuspended (1 gm/ml) in lysis buffer (50 mM Tris HCl, pH 8.0, 100 mM 1197 Table 1. Frequencies of autoantibodies recognizing ␣-enolase in the cohort of 255 patients with very early arthritides* Disease RA Non-RA rheumatic diseases Spondylarthropathies Psoriatic arthritis Other Osteoarthritis Crystal-induced arthritis Paramalignant arthritis Parvovirus B19 arthritis Sarcoidosis Connective tissue disease Primary Sjögren’s syndrome Systemic lupus erythematosus Polymyositis Mixed connective tissue disease Unclassified Other Wegener’s granulomatosis Behçet’s disease Eosinophilic fasciitis Polymyalgia rheumatica Undifferentiated arthritis Anti–␣enolase– positive patients No. of patients No. % 145 70 24 11 13 13 8 3 3 3 12 4 1 1 1 5 36 2 0 0 0 0 0 0 0 1 0 0 0 0 0 0 24.8 2.8 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 40 0 0 1 0 4 0 0 0 0 10 * RA ⫽ rheumatoid arthritis. NaCl, 0.02% NaN3, 1% Triton X-100, 1 mM ethylenediaminetetraacetic acid, 1% protease inhibitor cocktail; Sigma, St. Louis, MO). The suspension was sonicated for 5 minutes, boiled for 5 minutes, and centrifuged for 20 minutes at 15,000g. IgG were eliminated by incubating the supernatant with a mixture of Sepharose beads coupled to protein A (Sigma) for 1 hour at 4°C. After decantation, aliquots of supernatant were stored at ⫺80°C. Electrophoresis and Western blotting. Proteins were separated in SDS-PAGE according to the method of Laemmli, and electrotransferred onto a nitrocellulose membrane. The filters were then incubated with an RA serum (1:300) or goat anti-enolase antibodies (1:100) in Tris buffered saline–0.05% Tween 20 (TBST)–5% dry milk for 2 hours. After washing, the filters were incubated for 1 hour with 1:15,000-diluted peroxidase-conjugated goat anti-human IgG (Sigma) or 1:2,000-diluted peroxidase-conjugated donkey anti-goat IgG (Santa Cruz Biotechnology) in 0.05% TBST–milk. The filters were washed and revealed by a chemiluminescence reaction (Supersignal; Pierce, Rockford, IL). 2D-PAGE and immunoblotting. MC-F12 cells (mammary epithelial cell line CRL 10782; American Type Culture Collection, Rockville, MD) were scraped from plastic dishes, centrifuged, immediately resuspended in acetone (⫺18°C), 10% trichloroacetic acid, 0.12% dithiothreitol (DTT), and kept at ⫺18°C overnight. After centrifugation (35,000g for 30 minutes at 4°C), the supernatant was then discarded and the pellet resuspended in acetone, 0.2% DTT. After 1 hour at 1198 SAULOT ET AL Figure 1. Analysis of rheumatoid arthritis (RA) sera by immunoblotting on an MC-F12 protein map. a, MC-F12 extract separated by 2-dimensional polyacrylamide gel electrophoresis and stained with Coomassie blue (MC-F12 protein map). The following antisera were incubated with the filters: b, serum from an RA patient that bound to the 50-kd polypeptide; c, anti–50-kd antibodies affinity-purified from the same serum used in b, by elution from the nitrocellulose strip containing the 50-kd band; d, serum from a patient with systemic lupus erythematosus that bound to the 50-kd polypeptide. Molecular mass is indicated on the right and the isoelectric point (pI) is indicated at the top of each figure. The 3 isoforms of ␣-enolase, ␣1, ␣2, and ␣3, are indicated by the arrows. ⫺18°C, the sample was spun again at 35,000g for 30 minutes. The supernatant was discarded and the pellet was dried under vacuum. The pellet was then resuspended in a solution of 8M urea, 2% CHAPS, 1% (weight/volume) dithioerythritol, 7% spermine (Sigma), 1% protease-inhibitor cocktail (Sigma), and incubated for 1 hour at 4°C. After centrifugation, the supernatant was collected and stored at ⫺80°C. The lysate was subjected to 2D-PAGE according to the system previously described (7) using nonlinear, pH 3–10 gradients for the first dimension and 10% SDS-PAGE for the second dimension. Finally, gels were either stained with Coomassie blue to obtain a protein map or transferred onto polyvinylidene fluoride (PVDF) membranes. The immunoreactive spots blotted onto the PVDF membranes were detected as described above. Autoradiographic films were superimposed on Coomassie blue–stained protein maps to localize the recognized protein spots. Trypsin digestion and mass spectrometry. The protein spots were excised from polyacrylamide gels digested by trypsin, and peptides were analyzed with the matrix-assisted laser desorption ionization–time-of-flight (MALDI-TOF) Reflector spectrometer as previously described (7). The peptide masses were matched within the largest window possible for isoelectric point (pI) and molecular mass, and with species specificity (Homo sapiens). Antibody elution. Proteins extracted from human placenta were separated by SDS-PAGE and transferred onto a nitrocellulose membrane as described above. After incubation with the RA serum and visualization of the polypeptide of interest, the band was cut out of the remaining membrane and incubated for 20 minutes in 1 ml of elution buffer (0.2M glycine, pH 2.8) and the supernatant was neutralized with 0.1 volume of 1M Tris. The eluted antibodies were used for immunoblotting on 2D protein maps. In vitro antigen expression. The complete complementary DNA (cDNA) encoding human ␣-enolase was isolated from a cDNA expression library derived from synoviocytes obtained from an RA patient (Stratagene, La Jolla, CA) and immunoscreened with goat anti-enolase antibodies. This cDNA was subcloned in frame in the pSPUTK in vitro translation vector (Stratagene) using the Apa I and Bam HI restriction sites. The translation product was synthesized as a AUTOANTIBODIES TO ␣-ENOLASE IN EARLY RHEUMATOID ARTHRITIS Figure 2. Western blot analysis of rheumatoid arthritis (RA) sera using the biotinylated in vitro–synthesized recombinant ␣-enolase as the antigen. Filters were incubated with the following reagents: lane 1, peroxidase-conjugated streptavidin; lane 2, goat anti-enolase antibodies; lane 3, secondary anti-goat IgG antibodies conjugated to peroxidase (negative control); lanes 4 and 5, RA sera that bound to ␣-enolase in the 2-dimensional (2D) protein map (only the sera in lane 4 bound to the recombinant product); lane 6, systemic lupus erythematosus serum that bound to ␣-enolase in the 2D protein map. biotinylated polypeptide, purified by SoftLink Soft Release Avidin Resin (Promega, Madison, WI), migrated in SDSPAGE, and subjected to Western blotting experiments as described above. RESULTS 1D Western blot analysis of RA sera. Two hundred fifty-five sera obtained from patients with early arthritis were screened by Western blot on human placenta extract. Among the sera obtained from the 145 patients diagnosed 1 year later as having RA, 36 (24.8%) recognized a 50-kd band. Only 2 of the 70 sera obtained from patients with early non-RA inflammatory arthritides reacted with the 50-kd protein (Table 1). Thus, the anti–50-kd reactivity was mainly observed in early RA sera with a specificity of 97.1% compared with sera from patients with non-RA rheumatic diseases (subgroup 2). Moreover, 17 of the 36 positive sera (47.2%) reacted with the 50-kd band only, i.e., they were AFA and RF negative. These results suggest that the 50-kd antigen– anti–50-kd antibody system might contribute to the early diagnosis of RA, and prompted us to identify the 50-kd polypeptide. 2D Western blot analysis of RA sera. To characterize the 50-kd polypeptide recognized by the 36 RA sera, we first verified its presence in extracts of the mammary epithelial cell line, MC-F12, currently used 1199 for 2D-PAGE in our laboratory. All 36 anti–50-kd⫹ RA sera were then analyzed by immunoblotting on MC-F12 proteins separated by 2D-PAGE (Figure 1a). All sera recognized the same protein triplet, which consisted of components with a molecular mass of ⬃50 kd and pI between 6.5 and 7.0 (Figure 1b). Among RA and non-RA sera that did not react with the 50-kd band in 1D Western blot from the cohort of patients with very early arthritides, 30 were randomly chosen and analyzed by immunoblotting on the MC-F12 protein map obtained by 2D-PAGE. None of them bound to the identified triplet. Immunoblotting with eluted antibodies. To demonstrate the identity between the 50-kd polypeptide recognized by RA sera in 1D placenta-extract immunoblots and the triplet recognized in the MC-F12 protein map, the antibodies reacting with the former were eluted and allowed to react with the MC-F12 protein map. Figure 1c shows that eluted antibodies bound exclusively to the same triplet. Mass spectrometry analysis of the immunoreactive MC-F12 triplet. To characterize the triplet recognized by RA sera, the 3 protein spots (␣1, ␣2, and ␣3) were analyzed by MALDI-TOF mass spectrometry. The comparison of the mass spectra with those contained in other databases (MS-FIT and SWISS-PROT) allowed us to identify, with high probability (7 ⫻ 104), the 3 isoforms of human ␣-enolase. The database-matched peptides of the spectrum of the most basic and major protein spot (␣3) covered 80% of the protein sequence of ␣-enolase. Comparison of the 3 very similar spectra given by ␣1, ␣2, and ␣3 revealed that the 3 isoforms of ␣-enolase had slightly different pI, mainly attributed to differences in their phosphorylation (data not shown). Reactivity of RA sera with recombinant human ␣-enolase. Recombinant human ␣-enolase was produced by an in vitro transcription–translation system to test the reactivities of sera with the recombinant polypeptide. First, correct expression of the recombinant human ␣-enolase was verified by Western blot, in which the translated polypeptide was visualized with peroxidase-conjugated streptavidin at ⬃50 kd, in accordance with the expected size of the recombinant product. Second, goat anti-enolase antibodies incubated in Western blot recognized the in vitro–synthesized polypeptide, as expected. Then, the 36 RA sera that bound to ␣-enolase in the 2D protein map were analyzed by Western blot using the recombinant enolase as the antigen. Only 8 of them recognized the recombinant protein, while the other 28 RA sera were negative (Figure 2). 1200 SAULOT ET AL Anti–␣-enolase antibodies in other systemic diseases. Sera from 26 patients with SLE, 13 with SSc, and 14 with primary biliary cirrhosis were screened by 1D Western blot on human placenta extract. Five of the 26 SLE sera (19%) and 2 of the 13 SSc sera (15%) reacted with the 50-kd protein. All of these sera tested on the MC-F12 protein map bound to the different isoforms of ␣-enolase (Figure 1d). All of these sera also reacted with the recombinant protein obtained by in vitro transcription–translation (Figure 2). Prognostic value of anti–␣-enolase antibodies for RA. Sixty-six RA patients (subgroup 1) who were enrolled in the cohort and who underwent evaluation of radiologic damage were analyzed to determine the prognostic value of anti–␣-enolase antibodies. The independent variable was dichotomized into the presence (n ⫽ 26) or absence (n ⫽ 40) of radiologic progression. Dependent variables were all of the autoantibodies tested (see Patients and Methods). Using a univariate analysis (Fisher’s 2-tailed exact test), only positive results on the latex fixation test (P ⫽ 0.01) or presence of anti–␣-enolase antibodies at inclusion (P ⫽ 0.01) was predictive of the progression of radiologic damage. Using a multivariate stepwise logistic regression analysis, positivity for anti–␣-enolase antibodies was the strongest predictor of radiologic progression (P ⫽ 0.004). DISCUSSION This study showed that autoantibodies directed against human ␣-enolase are present at an early stage in the sera of patients who are subsequently diagnosed as having RA. Indeed, almost 25% of RA sera recognized a 50-kd polypeptide on Western blot using human placenta extract and a protein triplet in MC-F12 extracts separated by 2D-PAGE. This protein triplet, identified as human ␣-enolase by mass spectrometry, corresponded to the 50-kd band recognized by RA sera, because 1) anti–50-kd antibodies eluted from the 50-kd band blotted onto a nitrocellulose strip gave the same immunoreactive pattern, and 2) goat anti–␣-enolase antibodies that reacted with the 50-kd band comigrating with the polypeptide bound by RA sera also recognized the triplet, whose 3 protein spots were shown to correspond to 3 isoforms of human ␣-enolase (data not shown). Thus, human ␣-enolase is a new putative target autoantigen in RA. Autoantibodies directed against ␣-enolase have previously been described in various pathologic situations, including autoimmune diseases such as SLE (8) and discoid lupus erythematosus (9), SSc, primary bili- ary cirrhosis, and autoimmune hepatitis (10), chronic inflammatory diseases such as primary sclerosing cholangitis, inflammatory bowel diseases, primary membranous nephropathy, and cancer-associated retinopathy (10). Thus, at first glance, the anti–␣-enolase response does not seem to be restricted to RA, but rather, seems to occur in several chronic disorders, which limits its potential significance and diagnostic value in RA. However, properties of the anti–␣-enolase antibodies detected in RA patients’ sera might encourage a reconsideration of their role in the pathogenesis and diagnosis of RA. First, Western blot analysis of the 255 sera obtained from patients with early arthritis showed that the 50-kd response was observed almost exclusively in sera obtained from patients subsequently shown to have RA. Of note, they were not detected in the more common non-RA rheumatic diseases, i.e., spondylarthropathies and primary Sjögren’s syndrome. Moreover, half of the anti–␣-enolase–positive RA patients were negative for both RF and AFA. Second, the presence of anti–␣-enolase antibodies was correlated with the severity of the articular destruction and, in addition, our preliminary data suggest that anti–␣-enolase antibodies might constitute a better predictive marker of radiologic progression than does RF. Third, the lack of reactivity against human recombinant ␣-enolase in most RA patients’ sera, compared with the reactivity in SLE patients’ sera, suggests that RA sera predominantly bind to posttranslationally modified or appropriately folded epitopes of the protein. The cytoplasmic and ubiquitous glycolytic enzyme ␣-enolase catalyzes the formation of phosphoenolpyruvate from 2-phosphoglycerate, the second of the two high-energy intermediates that generate ATP in glycolysis (10). Of note, other glycolytic enzymes, such as glucose-6-phosphate isomerase (GPI) (11) and aldolase (12), were recently identified as target antigens of RA sera, and GPI is the target of the arthritogenic IgG antibodies produced in the K/BxN mouse model, which shares several features with RA (13). Thus, antibodies directed against glycolytic enzymes might very well be implicated in the pathophysiology of RA, at least in terms of dysregulation of glucose metabolism in hyperplastic synoviocytes in RA, as has been previously described (14). REFERENCES 1. Blä␤ S, Engel JM, Burmester GR. The immunologic homunculus in rheumatoid arthritis. Arthritis Rheum 1999;42:2499–506. 2. Desprès N, Boire G, Lopez-Longo FJ, Ménard HA. The Sa AUTOANTIBODIES TO ␣-ENOLASE IN EARLY RHEUMATOID ARTHRITIS 3. 4. 5. 6. 7. system: a novel antigen–antibody system specific for rheumatoid arthritis. J Rheumatol 1994;21:1027–33. Vittecoq O, Jouen-Beades F, Krzanowska K, Bichon-Tauvel I, Ménard JF, Daragon A, et al. Rheumatoid factors, anti-filaggrin antibodies, and low in vitro interleukin 2 and interferon-␥ production are useful immunological markers for early diagnosis of community cases of rheumatoid arthritis: preliminary study. Joint Bone Spine 2001;68:144–53. Girbal-Neuhauser E, Durieux JJ, Arnaud M, Dalbon P, Sebbag M, Vincent C, et al. The epitopes targeted by the rheumatoid arthritis-associated antifilaggrin autoantibodies are posttranslationally generated on various sites of (pro)filaggrin by deimination of arginine residues. J Immunol 1999;162:585–94. Schellekens GA, de Jong BAW, van den Hoogen FHJ, van de Putte LBA, van Venrooij WJ. Citrulline is an essential constituent of antigenic determinants recognized by rheumatoid arthritisspecific autoantibodies. J Clin Invest 1998;101:273–81. van der Heijde DM, van Riel PL, Nurver-Zwart IH, Gribnau FW, van de Putte LB. Effects of hydroxychloroquine and sulphasalazine on progression of joint damage in rheumatoid arthritis. Lancet 1989;1:1036–8. Thébault S, Gilbert D, Machour N, Marvin, L, Lange C, Tron F, et al. Two-dimensional electrophoresis and mass spectrometry identification of proteins bound by a murine monoclonal anticardiolipin antibody: a powerful technique to characterize the 8. 9. 10. 11. 12. 13. 14. 1201 cross-reactivity of a single autoantibody. Electrophoresis 2000;21: 2531–9. Pratesi F, Moscato S, Sabbatini A, Chimenti D, Bombardieri S, Migliorini P. Autoantibodies specific for ␣-enolase in systemic autoimmune disorders. J Rheumatol 2000;27:109–15. Gitlits VM, Sentry JW, Matthew MLSM, Smith AI, Toth BH. Autoantibodies to evolutionarily conserved epitopes of enolase in a patient with discoid lupus erythematosus. Immunology 1997;92: 362–8. Pancholi V. Multifunctional ␣-enolase: its role in diseases. Cell Mol Life Sci 2001;58:902–20. Schaller M, Burton DR, Ditzel HJ. Autoantibodies to GPI in rheumatoid arthritis: linkage between an animal model and human disease. Nature Immunol 2001;2:746–53. Ukaji F, Kitajima I, Kubo T, Shimizu C, Nakajima T, Maruyama I. Serum samples of patients with rheumatoid arthritis contain a specific autoantibody to “denatured” aldolase A in the osteoblastlike cell line, MG-63. Ann Rheum Dis 1999;58:169–74. Matsumoto I, Staub A, Benoist C, Mathis D. Arthritis provoked by linked T and B cell recognition of a glycolytic enzyme. Science 1999;286:1732–5. Henderson B, Bitensky L, Chayen J. Glycolytic activity in human synovial lining cells in rheumatoid arthritis. Ann Rheum Dis 1979;38:63–7.