A Glucose-6-Phosphate Isomerase Peptide Induces T and B Cell–Dependent Chronic Arthritis in C57BL/10 Mice
2013; Elsevier BV; Volume: 183; Issue: 4 Linguagem: Inglês
10.1016/j.ajpath.2013.06.019
ISSN1525-2191
AutoresAngela Pizzolla, Kajsa Wing, Rikard Holmdahl,
Tópico(s)Adenosine and Purinergic Signaling
ResumoImmunization with human glucose-6-phosphate isomerase (hG6PI) protein or with several of its peptides induces arthritis in DBA/1 mice. We investigated G6PI peptide–induced arthritis in C57BL/10 mice and the effect of oxidative burst on disease. To study the arthritogenicity of G6PI peptides and its immune dependency, we used genetically modified and congenic mice on the C57BL/10 background and in vitro T- and B-cell assays. hG6PI325-339 peptide induced arthritis in C57BL/10 mice. The disease was associated with major histocompatibility complex class II and was dependent on T cells, B cells, and complement C5. Th1 and Th17 cells primed with the hG6PI325-339 peptide cross-reacted with the murine G6PI protein. The severity of the disease increased in mice carrying a mutation in Ncf1 (Ncf1∗/∗), which abolishes the NADPH oxidase 2 complex oxidative burst. Ncf1∗/∗ mice developed arthritis also on immunization with the mouse G6PI325-339 peptide and in the absence of C5. The antibody responses to the G6PI protein and peptides were minimal in both Ncf1∗/∗ and wild-type mice. Herein is described G6PI peptide as the first peptide to induce arthritis in C57BL/10 mice. The differences between the wild-type and Ncf1∗/∗ mice suggest that an alternative complement-independent arthritogenic pathway could be operative in the absence of oxidative burst. Immunization with human glucose-6-phosphate isomerase (hG6PI) protein or with several of its peptides induces arthritis in DBA/1 mice. We investigated G6PI peptide–induced arthritis in C57BL/10 mice and the effect of oxidative burst on disease. To study the arthritogenicity of G6PI peptides and its immune dependency, we used genetically modified and congenic mice on the C57BL/10 background and in vitro T- and B-cell assays. hG6PI325-339 peptide induced arthritis in C57BL/10 mice. The disease was associated with major histocompatibility complex class II and was dependent on T cells, B cells, and complement C5. Th1 and Th17 cells primed with the hG6PI325-339 peptide cross-reacted with the murine G6PI protein. The severity of the disease increased in mice carrying a mutation in Ncf1 (Ncf1∗/∗), which abolishes the NADPH oxidase 2 complex oxidative burst. Ncf1∗/∗ mice developed arthritis also on immunization with the mouse G6PI325-339 peptide and in the absence of C5. The antibody responses to the G6PI protein and peptides were minimal in both Ncf1∗/∗ and wild-type mice. Herein is described G6PI peptide as the first peptide to induce arthritis in C57BL/10 mice. The differences between the wild-type and Ncf1∗/∗ mice suggest that an alternative complement-independent arthritogenic pathway could be operative in the absence of oxidative burst. Glucose-6-phosphate isomerase (G6PI) is a ubiquitously expressed cytosolic enzyme that has an essential role in glycolysis. It can also be secreted and function as an extracellular signaling molecule, in which context it is known as tumor autocrine mobility factor, neuroleukine, or maturation factor.1Haga A. Niinaka Y. Raz A. Phosphohexose isomerase/autocrine motility factor/neuroleukin/maturation factor is a multifunctional phosphoprotein.Biochim Biophys Acta. 2000; 1480: 235-244Crossref PubMed Scopus (74) Google Scholar G6PI became relevant in arthritis research when it was found that T-cell receptor (TCR) transgenic K/BxN mice on a mixed non-obese diabetic (NOD) and B6 background developed spontaneous arthritis.2Matsumoto I. Staub A. Benoist C. Mathis D. Arthritis provoked by linked T and B cell recognition of a glycolytic enzyme.Science. 1999; 286: 1732-1735Crossref PubMed Scopus (494) Google Scholar TCR transgenic T cells activated B cells to produce arthritogenic antibodies2Matsumoto I. Staub A. Benoist C. Mathis D. Arthritis provoked by linked T and B cell recognition of a glycolytic enzyme.Science. 1999; 286: 1732-1735Crossref PubMed Scopus (494) Google Scholar, 3Korganow A.S. Ji H. Mangialaio S. Duchatelle V. Pelanda R. Martin T. Degott C. Kikutani H. Rajewsky K. Pasquali J.L. Benoist C. Mathis D. From systemic T cell self-reactivity to organ-specific autoimmune disease via immunoglobulins.Immunity. 1999; 10: 451-461Abstract Full Text Full Text PDF PubMed Scopus (559) Google Scholar that bound to the G6PI deposited on the cartilage surface of the joints.4Matsumoto I. Maccioni M. Lee D.M. Maurice M. Simmons B. Brenner M. Mathis D. Benoist C. How antibodies to a ubiquitous cytoplasmic enzyme may provoke joint-specific autoimmune disease.Nat Immunol. 2002; 3: 360-365Crossref PubMed Scopus (291) Google Scholar B cells contributed to the disease by producing antibodies and presenting the antigen to T cells.5Tanaka-Watanabe Y. Matsumoto I. Iwanami K. Inoue A. Goto D. Ito S. Tsutsumi A. Sumida T. B cells play a crucial role as antigen-presenting cells and collaborate with inflammatory cytokines in glucose-6-phosphate isomerase-induced arthritis.Clin Exp Immunol. 2009; 155: 285-294Crossref PubMed Scopus (18) Google Scholar Immunization with human recombinant G6PI protein induced arthritis in DBA/1,6 C3H.Q, and C57BL/10.Q (B10.Q) mice.7Bockermann R. Schubert D. Kamradt T. Holmdahl R. Induction of a B-cell-dependent chronic arthritis with glucose-6-phosphate isomerase.Arthritis Res Ther. 2005; 7: R1316-R1324Crossref PubMed Google Scholar All of these strains carried major histocompatibility complex class II (MHC II) of haplotype q, which together with the genetic background determined the outcome of the disease.7Bockermann R. Schubert D. Kamradt T. Holmdahl R. Induction of a B-cell-dependent chronic arthritis with glucose-6-phosphate isomerase.Arthritis Res Ther. 2005; 7: R1316-R1324Crossref PubMed Google Scholar Using published information about arthritogenic T-cell epitopes in DBA/1 mice and predicting the binding motif to Aq molecules, Iwanami et al8Iwanami K. Matsumoto I. Tanaka Y. Inoue A. Goto D. Ito S. Tsutsumi A. Sumida T. Arthritogenic T cell epitope in glucose-6-phosphate isomerase-induced arthritis.Arthritis Res Ther. 2008; 10: R130Crossref PubMed Scopus (27) Google Scholar selected G6PI peptides that were thought to bind Aq molecules. One of the peptides, human (h) G6PI325-339, induced arthritis in DBA/1 mice in an IL-17–dependent manner.8Iwanami K. Matsumoto I. Tanaka Y. Inoue A. Goto D. Ito S. Tsutsumi A. Sumida T. Arthritogenic T cell epitope in glucose-6-phosphate isomerase-induced arthritis.Arthritis Res Ther. 2008; 10: R130Crossref PubMed Scopus (27) Google Scholar Bruns et al9Bruns L. Frey O. Morawietz L. Landgraf C. Volkmer R. Kamradt T. Immunization with an immunodominant self-peptide derived from glucose-6-phosphate isomerase induces arthritis in DBA/1 mice.Arthritis Res Ther. 2009; 11: R117Crossref PubMed Scopus (20) Google Scholar confirmed this discovery in a hypothesis-free screening of the full G6PI protein and found two additional G6PI peptides that induced arthritis in DBA/1 mice. We tested all of these peptides in B10.Q mice to determine whether peptide-induced arthritis depended on similar pathogenic pathways as other protein-induced arthritis models such as collagen-induced arthritis (CIA) and G6PI protein–induced arthritis. We found that hG6PI325-339 peptide was the most efficient inducer of arthritis in B10.Q mice. Using congenic and knockout mice, we found that the disease was dependent on MHC II, T and B cells, and complement factor C5. We also tested hG6PI325-339 peptide–induced arthritis on a mutated strain of B10.Q mice, which was previously found in this laboratory to be susceptible to CIA and other autoimmune disease models.10Hultqvist M. Olofsson P. Holmberg J. Bäckström B.T. Tordsson J. Holmdahl R. Enhanced autoimmunity, arthritis, and encephalomyelitis in mice with a reduced oxidative burst due to a mutation in the Ncf1 gene.Proc Natl Acad Sci U S A. 2004; 101: 12646-12651Crossref PubMed Scopus (275) Google Scholar These mice carry a mutation in the Ncf1 gene (B10.Q.Ncf1∗/∗, named Ncf1∗/∗ mice), which leads to a defective NADPH oxidase 2 (NOX2) complex–dependent oxidative burst.11Huang C.K. Zhan L. Hannigan M.O. Ai Y. Leto T.L. P47(phox)-deficient NADPH oxidase defect in neutrophils of diabetic mouse strains, C57BL/6J-m db/db and db/+.J Leukoc Biol. 2000; 67: 210-215Crossref PubMed Scopus (92) Google Scholar Ncf1∗/∗ mice developed severe arthritis after immunization with hG6PI325-339 peptide and the corresponding homologous self-peptide (murine G6PI325-339 peptide) and induced activation of autoreactive T cells. Contrary to B10.Q mice, Ncf1∗/∗ mice developed hG6PI325-339 peptide–induced arthritis without complement C5. The induced antibody response in both Ncf1∗/∗ and wild-type (WT) mice was minimal, with barely detectable response to the G6PI protein and peptides. Thus, we established a new peptide-induced arthritis model operating through a T and B cell–dependent pathway but without antibody effector mechanisms. All mice used were genetically controlled and shared the C57BL/10 background. The MHC congenic C57BL/10.P/rhd, C57BL/10.Q/rhd, and C57BL/10.RIII/rhd strains originated from the Jan Klein mouse colony (Tübingen, Germany) and were maintained in our laboratory. C57BL/10.P/rhd (B10.P) mice express MHC II H2-Ap encoded by a congenic fragment from the P/J strain on chromosome 17; MHC II congenic C57BL/10.Q/rhd (B10.Q) mice express MHC II H2-Aq encoded by a fragment from the DBA/1 strain on chromosome 17; and C57BL/10.RIII/rhd (B10.RIII) mice express MHC II H2-Ar encoded by a congenic fragment from the RIII strain on chromosome 17. B10.Q mice with a mutated Ncf1 mutation (Ncf1m1J/m1J, denoted as Ncf1∗/∗) have been described previously.10Hultqvist M. Olofsson P. Holmberg J. Bäckström B.T. Tordsson J. Holmdahl R. Enhanced autoimmunity, arthritis, and encephalomyelitis in mice with a reduced oxidative burst due to a mutation in the Ncf1 gene.Proc Natl Acad Sci U S A. 2004; 101: 12646-12651Crossref PubMed Scopus (275) Google Scholar, 11Huang C.K. Zhan L. Hannigan M.O. Ai Y. Leto T.L. P47(phox)-deficient NADPH oxidase defect in neutrophils of diabetic mouse strains, C57BL/6J-m db/db and db/+.J Leukoc Biol. 2000; 67: 210-215Crossref PubMed Scopus (92) Google Scholar B10.Q μMT mice were generated by backcrossing the original μMT founder to B10.Q mice for >10 generations. TCR knockout animals (Tcrbtm1Mom) were purchased from The Jackson Laboratory (Bar Harbor, ME) and backcrossed to B10.Q mice for >10 generations. Congenic B10.Q.C5.NOD mice contain a fragment from NOD.Q mice on chromosome 2 that harbors the defective C5 molecule.12Johansson A.C. Sundler M. Kjellén P. Johannesson M. Cook A. Lindqvist A.K. Nakken B. Bolstad A.I. Jonsson R. Alarcón-Riquelme M. Holmdahl R. Genetic control of collagen-induced arthritis in a cross with NOD and C57BL/10 mice is dependent on gene regions encoding complement factor 5 and FcgammaRIIb and is not associated with loci controlling diabetes.Eur J Immunol. 2001; 31: 1847-1856Crossref PubMed Scopus (81) Google Scholar The mice were kept and bred in a climate-controlled specific pathogen–free (Federation of Laboratory Animal Science Associations II) environment with 12-hour light-dark cycles, housed in polystyrene cages containing wood shavings, and provided with standard rodent chow and water ad libitum in the animal house of the Division of Medical Inflammation Research, Karolinska Institutet (Stockholm, Sweden). All experiments were performed in 8- to 10-week-old mice under standard conditions (littermates, scored blindly by A.P., mixed in cages, and age- and sex-matched). Experiments were approved by the Stockholm Ethical Committee under license Nos. M107/07, M109/07, and N66/10. All peptides were purchased from Schafer-N ApS (Copenhagen, Denmark). Peptides used for arthritis induction and in vitro stimulation of T cells were dissolved in dimethyl sulfoxide (Prolabo; VWR International, Ltd., Leicestershire, UK) and 1.2 mol/L urea (Bio-Rad Laboratories, Inc., Hercules, CA). The aliquoted stocks at 21 mg/mL were stored at −80°C until use. From the stock, the peptides were then diluted in PBS or complete medium for in vitro stimulation of cells. The peptides used for MHC binding and enzyme-linked immonusorbent assay (ELISA), N-terminal biotinylated and otherwise, were dissolved in dimethyl sulfoxide, 1.2 mol/L urea, and 9 mmol/L Tris (2-carboxyethyl) phosphine at a concentration of 21 mg/mL and stored in 10-μL aliquots at −80°C until use. Recombinant human (h) and mouse (m) G6PI proteins were a gift. Rat type II collagen (CII) was purified from Swarm rat chondrosarcoma as previously described.13Smith B.D. Martin G.R. Miller E.J. Dorfman A. Swarm R. Nature of the collagen synthesized by a transplanted chondrosarcoma.Arch Biochem Biophys. 1975; 166: 181-186Crossref PubMed Scopus (205) Google Scholar Murine class II–associated invariant chain peptide (CLIP) (sequence) was diluted in PBS and stored at 4°C until use. The rat CII256-270 peptide (H-GIAGFKGEQGPKGETG-OH) was synthesized, purified, and characterized as previously described.14Michäelsson E. Andersson M. Engström A. Holmdahl R. Identification of an immunodominant type-II collagen peptide recognized by T cells in H-2q mice: self tolerance at the level of determinant selection.Eur J Immunol. 1992; 22: 1819-1825Crossref PubMed Scopus (119) Google Scholar Both CII and CII peptide were dissolved and stored in 0.1 mol/L acetic acid at 4°C. Arthritis was induced by injecting an emulsion composed of complete Freund's adjuvant (Difco Laboratories, Inc., Detroit, MI) and 10 μg peptides or 300 μg protein diluted in PBS (about 5 mmol/L final concentration for both). The total volume of emulsion injected was 150 μL, divided into two intradermal injections of 75 μL each at the two sides of the base of the tail. Arthritis development was monitored using a macroscopic scoring system; one point was given for each swollen or red toe or joint, and five points for a swollen ankle or wrist, for a maximum score of 60 points per mouse. At day 30, paws were dissected and fixed in 4% paraformaldehyde and decalcified in solution containing EDTA, polyvinylpyrolidone, Tris HCl, and potassium HCl until the bone was dissolved. Paws were then dehydrated and embedded in paraffin. Sections (5 μm thick) were cut and stained with H&E. Lymph nodes were conferred to a single cell suspension; 106 cells per well were plated in 96-well cell culture plates (Nunc; Thermo Fisher Scientific, Inc., Waltham, MA) and cultured for 96 hours in Dulbecco's modified Eagle's medium (Gibco; Life Technologies Corp., Carlsbad, CA) supplemented with 10% heat-inactivated fetal bovine serum (PAA Laboratories, GE Healthcare Bio-Sciences Corp., Piscataway, NJ), 10 mmol/L HEPES, penicillin/streptomycin, and 5 and 0 μmol/L β-mercaptoethanol (Sigma, St. Louis, MO). Cells were stimulated using various concentrations of hG6PI325-339, mG6PI325-339, hG6PI protein, or mG6PI or were left unstimulated. Anti-cytokine ELISAs were performed by coating MaxiSorp plates (Nunc) with the specific purified antibody in PBS overnight at 4°C: 5 μg/mL TC11-18H10 (BD Biosciences, San Jose, CA) for IL-17 and 5 μg/mL R46A2 for interferon (IFN)-γ. After blocking with 2% powdered fat-free milk in water and washing with PBS plus 0.1% Tween (Sigma), the supernatant from cell culture and recombinant cytokine as standard were added to the plates for 2 hours at room temperature. For IL-17 cytokine, the standard was purchased from e-Bioscience (San Diego, CA), and for IFN-γ, homemade supernatants from concanavalin A–stimulated lymph node cells were used. After washing, the biotinylated detection antibody was added for 1 hour at room temperature: 0.5 μg/mL TC11-8H4.1 (BD Biosciences) for IL-17 and 0.6 μg/mL AN18.17.26 (Mabtech AB, Nacka Strand, Sweden) for IFN-γ. After washing, the biotinylated antibody was detected using europium-labeled streptavidin diluted in assay buffer (PerkinElmer, Inc., Waltham, MA) for 30 minutes at room temperature. After a final wash, the enhancement solution (PerkinElmer) was added, and the luminescence emitted was measured using a Wallac Victor 1420 multilabel counter (PerkinElmer). The results were analyzed using the SoftMax Pro version 2.6.1 (Molecular Devices Corp., Sunnyvale, CA) to determine the concentration of the cytokines in the supernatant. Mice were terminally bled via the eye at the end of the disease or via the cheek at day 14 after immunization, and the heparinized blood was centrifuged to obtain plasma. For detection of antibodies at day 14, the serum was then serially diluted from 1:1000 to 1:10,000 for anti-hG6PI IgG and from 1:25 to 1:50 for anti-CII IgG. The plasma for anti-peptide immunoglobulin detection was concentrated by filtration through centrifugal filter units with a 100,000 MW cutoff (Amicon Ultra; Millipore Corp., Billerica, MA) and washed three times with PBS. The amount of protein was quantified using the DC protein Assay (Bio-Rad Laboratories) according to the manufacturer's instructions. Avidin (Thermo Scientific Pierce Protein Biology Products, Rockford, IL) (10 μg/mL) and approximately 0.16 μmol/L recombinant mG6PI and hG6PI proteins were diluted in PBS and rat CII in carbonate buffer [15 mmol/L Na2CO3, 35 mmol/L NaHCO3, and 11 mmol/L mgCl2 (pH 9.6) with 1 mol/L NaOH]. The proteins were then coated onto MaxiSorp plates (Nunc) for 2 hours at 37°C. After blocking with 2% fat-free milk powder for 45 minutes at 37°C, the plates were washed in PBS and 0.1% Tween (Sigma). Biotinylated peptides were diluted to a concentration of 250 μg/mL in dimethyl sulfoxide and then further diluted in carbonate buffer to a final concentration of 0.5 μg/mL (approximately 0.27 μmol/L). The final concentrations of dimethyl sulfoxide, urea, and Tris (2-carboxyethyl) phosphine in the carbonate buffer were 0.2% v/v, 28.5 μmol/L, and 21.4 nmol/L, respectively. The biotinylated peptides or no peptide control in carbonate buffer were added to the wells coated with avidin for 2 hours at room temperature. The plates were washed, and plasma from immunized mice diluted in PBS was added to the plates at various concentrations for 2 hours at room temperature. After washing, peroxidase-conjugated anti-immunoglobulin antibodies were diluted in PBS, 0.1% bovine serum albumin, and 0.1% Tween and added to the plates for 1 hour at room temperature. To detect the antibody, horseradish peroxidase–conjugated AffiniPure goat anti-mouse IgG (H+L) (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) was added to the plates. The horseradish peroxidase was detected using ABTS [2,2′-azino-bis(3-ethylbenzo-thiazoline-6-sulfonic acid)] (Roche Diagnostics GmbH, Mannheim, Germany) as substrate. Absorbance was measured at 405 nm using a Wallac Victor 1420 multilabel counter (Molecular Devices Corp.). The assay was performed in duplicate, and the mean value was reported. A peptide-binding assay was used to indirectly measure the affinity of binding of peptides to MHC by determining the ability of the peptides to inhibit binding of biotinylated CLIP peptide to purified soluble recombinant empty Aq MHC molecules, as described previously.15Kjellén P. Brunsberg U. Broddefalk J. Hansen B. Vestberg M. Ivarsson I. Engström Å. Svejgaard A. Kihlberg J. Fugger L. Holmdahl R. The structural basis of MHC control of collagen-induced arthritis; binding of the immunodominant type II collagen 256-270 glycopeptide to H-2Aq and H-2Ap molecules.Eur J Immunol. 1998; 28: 755-767Crossref PubMed Scopus (71) Google Scholar The mix of MHC molecules (final concentration 0.8 μmol/L), biotinylated murine CLIP peptide (final concentration 3 μmol/L), and the competitive peptide at various concentrations was incubated for 48 hours at room temperature in PBS and protease inhibitors cocktail (Complete; Roche Diagnostics) in U-shaped PP-microplates (Greiner Bio-One, Stonehouse Gloucestershsire, UK). The amount of biotinylated CLIP peptide bound to MHC was then quantified via ELISA. Each mix was loaded onto precoated and preblocked anti-His tag plates (HisSorp; Qiagen NV, Venlo, The Netherlands) in triplicate for 2 hours at room temperature to link the MHC molecule to the plate. After washing in PBS plus 0.1% Tween (Sigma), the biotinylated CLIP bound to MHC was detected using europium-labeled streptavin diluted in assay buffer and enhancement solution according to the manufacturer's instructions (PerkinElmer). The luminescence emitted was read using a Wallac Victor 1420 multilabel counter (Molecular Diagnostics). The mean of the triplicates is given. The concentration of peptide that prevented 50% of the biotinylated CLIP from binding was defined as IC50 value. Statistical analysis was performed using commercially available software (PRISM version 5.0c; GraphPad Software, Inc., San Diego, CA). The two-tailed U test with 95% confidence interval was used when comparing two groups, and the Kruskal-Wallis test followed by Dunn's comparison posttest when comparing more than two groups. Only biological replicates are given. To test whether B10.Q mice were susceptible to arthritis induced by hG6PI peptides, we selected three hG6PI peptides identified as arthritogenic in DBA/1 mice.8Iwanami K. Matsumoto I. Tanaka Y. Inoue A. Goto D. Ito S. Tsutsumi A. Sumida T. Arthritogenic T cell epitope in glucose-6-phosphate isomerase-induced arthritis.Arthritis Res Ther. 2008; 10: R130Crossref PubMed Scopus (27) Google Scholar, 9Bruns L. Frey O. Morawietz L. Landgraf C. Volkmer R. Kamradt T. Immunization with an immunodominant self-peptide derived from glucose-6-phosphate isomerase induces arthritis in DBA/1 mice.Arthritis Res Ther. 2009; 11: R117Crossref PubMed Scopus (20) Google Scholar Two peptides, hG6PI85-99 (H-FNGEKINYTEGRAVL-OH) and hG6PI469-483 (H-EGNRPTNSIVFTKLT-OH), as well as the full recombinant hG6PI protein, induced very mild disease in both B10.Q and B10.Q.Ncf1∗/∗ mice (Figure 1, A, C, and D), whereas only hG6PI325-339 peptide (H-IWYINCFGCETHAML-OH) induced disease with high severity and incidence (Figure 1, B and G–J). A similar molar concentration of protein and peptides was administered. The harmful effect of a dysfunctional NOX2-dependent oxidative burst, caused by a mutation in the Ncf1 gene (Figure 1B), was apparent only in severe hG6PI325-339 peptide–induced arthritis, consistent with previous observations in other induced and spontaneous arthritis models.10Hultqvist M. Olofsson P. Holmberg J. Bäckström B.T. Tordsson J. Holmdahl R. Enhanced autoimmunity, arthritis, and encephalomyelitis in mice with a reduced oxidative burst due to a mutation in the Ncf1 gene.Proc Natl Acad Sci U S A. 2004; 101: 12646-12651Crossref PubMed Scopus (275) Google Scholar, 16Hagenow K. Gelderman K.A. Hultqvist M. Merky P. Bäcklund J. Frey O. Kamradt T. Holmdahl R. Ncf1-associated reduced oxidative burst promotes IL-33R+ T cell-mediated adjuvant-free arthritis in mice.J Immunol. 2009; 183: 874-881Crossref PubMed Scopus (27) Google Scholar No significant effect of the mutated Ncf1 was observed in arthritis induced by the G6PI protein or the G6PI peptides, inducing only mild arthritis (Figure 1, A, C, and D). Immunization with hG6PI peptides generated only barely detectable titers of anti-hG6PI protein IgG, in contrast to immunization with G6PI protein, which induced high titers (Figure 1E). The anti-hG6PI IgG titers in individual mice were not associated with arthritis. Anti-CII antibodies are present in serum samples from many patients with rheumatoid arthritis,17Andriopoulos N.A. Mestecky J. Miller E.J. Bennett J.C. Antibodies to human native and denatured collagens in synovial fluids of patients with rheumatoid arthritis.Clin Immunol Immunopathol. 1976; 6: 209-212Crossref PubMed Scopus (41) Google Scholar and their presence was checked in the G6PI-immunized animals (Figure 1F). The antibody response to CII was detectable but at low levels and did not reflect arthritis severity. Histopathologic analysis of the paws from mice immunized with hG6PI325-339 peptide 30 days earlier showed that B10.Q.Ncf1∗/∗ mice developed severely inflamed synovium and erosive destruction of bone and cartilage (Figure 1L), whereas B10.Q mice developed mild disease with only moderate synovium infiltration (Figure 1M). To determine the MHC association of hG6PI325-339 peptide–induced arthritis, we used congenic B10 mice expressing MHC of different haplotypes. Both B10.Q (H-2q) and B10.P (H-2p) mice were susceptible to hG6PI325-339 peptide–induced arthritis, whereas B10.RIII (H-2r) mice were resistant (Figure 2). As in the B10.Q mice, also in the susceptible B10.P strain, the Ncf1 mutation increased the severity of the disease (Figure 2, A and B). In conclusion, hG6PI325-339 peptide induces arthritis in B10 mice expressing MHC of q or p, but not r, haplotype. To check the antigen specificity and cytokine profile of T cells primed by immunization with hG6PI325-339 peptide, 30 days after immunization, draining lymph node cells were stimulated in vitro using the same peptide or the homologous mG6PI325-339 peptide, the full hG6PI protein, or the full mG6PI protein. Lymph node cells stimulated with hG6PI325-339 peptide produced high amounts of IL-17 and IFN-γ (Figure 3, A and B, respectively) but no IL-4 (data not shown). In contrast, very low cytokine production was detectable after repeat stimulation with mG6PI325-339 peptide (Figure 3, A and B). hG6PI protein stimulated high cytokine production (Figure 3, C and D). mG6PI protein stimulated low but significant production of IFN-γ and IL-17 (Figure 3, C and D), indicating that immunization with hG6PI325-339 peptide leads to priming of cross-reactive T cells to the murine protein. Compared with B10.Q cells, B10.Q.Ncf1∗/∗ cells produced a similar level of cytokines in response to human and murine G6PI325-339 peptides and G6PI proteins (Figure 3, A–D). Similar results were observed at days 7 and 13 after disease induction (data not shown). B10.RIII mice were resistant to hG6PI325-339 peptide–induced arthritis (Figure 2C). Nevertheless, both Ncf1∗/∗ and Ncf1 WT mice on a B10.RIII background mounted a moderate Th1 response to hG6PI325-339 peptide (Figure 3, E and F). B10.P mice were susceptible to hG6PI325-339 peptide–induced arthritis (Figure 2B), and their lymph node cells responded to the same peptide and to a lesser extent to the murine peptide, similarly as B10.Q cells (Figure 3, G and H). No difference between Ncf1∗/∗ and Ncf1 WT mice was observed. In conclusion, the susceptible B10.Q and B10.P strains could prime Th1 and Th17 cells that were specific for hG6PI325-339 peptide and cross-reacted with mG6PI protein. To determine the dependency of hG6PI325-339–induced-arthritis from the adaptive immune system, we used B10.Q congenic mice deficient in either B or T cells. Both B cell– and T cell–deficient mice on B10.Q.Ncf1∗/∗ background demonstrated resistance to hG6PI325-339 peptide–induced arthritis (Figure 4, A and B). To understand the role of complement in this disease, mice deficient in C5 due to a congenic fragment from the NOD strain on the B10.Q background12Johansson A.C. Sundler M. Kjellén P. Johannesson M. Cook A. Lindqvist A.K. Nakken B. Bolstad A.I. Jonsson R. Alarcón-Riquelme M. Holmdahl R. Genetic control of collagen-induced arthritis in a cross with NOD and C57BL/10 mice is dependent on gene regions encoding complement factor 5 and FcgammaRIIb and is not associated with loci controlling diabetes.Eur J Immunol. 2001; 31: 1847-1856Crossref PubMed Scopus (81) Google Scholar were immunized with hG6PI325-339 peptide. Mice deficient in C5 were not susceptible to the disease, whereas C5 WT littermates were (Figure 4C). When C5-deficient mice carried the Ncf1 mutation, they were as susceptible to arthritis as were C5-sufficient Ncf1∗/∗ mice (Figure 4D). Although B and T cells are necessary for disease in Ncf1∗/∗ mice, C5 does not seem to be essential. Lymph node cells from hG6PI325-339 peptide–immunized mice (Figure 4) were stimulated again in vitro with hG6PI325-339 peptide, and IL-17 and IFN-γ production were measured. T cells from B cell– and C5-deficient mice (Figure 5, A, B, E, and F) produced similar levels of cytokines as did their WT littermates, suggesting that neither B cells nor C5 are necessary for priming of hG6PI325-339 peptide–specific Th1 and Th17 cells. Very low levels of both cytokines were detected in T cell–deficient mice (Figure 5, C and D), suggesting that T cells were the major producers of IFN-γ and the exclusive producers of IL-17 in this assay. Although anti-hG6PI protein IgG was detectable in low titers in plasma from hGPI325-339–immunized animals (Figure 1E), anti-hG6PI325-339 peptide IgG titers were undetectable at both days 14 and 30 after immunization (data not shown). To ascertain the presence of anti-hG6PI325-339 peptide IgG, plasma from pooled immunized Ncf1∗/∗ mice was concentrated using centrifugal filter units and tested via ELISA on plates coated with avidin and biotinylated peptides. Low anti-hG6PI325-339 peptide IgG titers were detectable in concentrated plasma from immunized Ncf1∗/∗ mice but not in plasma from naïve animals (Figure 6A). Similarly, very low titers of anti-mG6PI325-339 peptide IgG were detectable in plasma from immunized mice but not in plasma from naïve mice (Figure 6B). Both anti-hG6PI325-339 IgG and anti-mG6PI325-339 peptide IgG were of IgG2b and IgG2c isotypes (data not shown). Low levels of anti-mG6PI protein IgG were also detected (Figure 6B). mG6PI325-339 peptide (H-IWYINCYGCETHALL-OH) differs from human peptide by only two amino acids (underlined in the sequence). We wanted to test whether immunizing the mice with mG6PI325-339 peptide could induce arthritis. The murine peptide induced substantial but mild arthritis in B10.Q.Ncf1∗/∗ mice and failed to induce arthritis in B10.Q mice (Figure 7, A and B). Of the two amino acids that differ between human and murine peptide, of particular interest is the amino acid in position 331, a phenylalanine (F) in the human peptide and a
Referência(s)