Regulation of myeloperoxidase-specific T cell responses during disease remission in antineutrophil cytoplasmic antibodyassociated vasculitisThe role of Treg cells and tryptophan degradation.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 62, No. 5, May 2010, pp 1539–1548 DOI 10.1002/art.27403 © 2010, American College of Rheumatology Regulation of Myeloperoxidase-Specific T Cell Responses During Disease Remission in Antineutrophil Cytoplasmic Antibody–Associated Vasculitis The Role of Treg Cells and Tryptophan Degradation Konstantia-Maria Chavele,1 Deepa Shukla,1 Tracey Keteepe-Arachi,1 Judith Anna Seidel,1 Dietmar Fuchs,2 Charles D. Pusey,1 and Alan D. Salama1 Objective. T lymphocytes have been implicated in the pathogenesis of antineutrophil cytoplasmic antibody–associated vasculitis (AAV). Patients with myeloperoxidase (MPO) antineutrophil cytoplasmic antibody (ANCA) experience relapses less frequently than those with proteinase 3 ANCA, suggesting greater immune regulation. This study was undertaken to investigate MPO-specific T cell reactivity during disease remission and the factors regulating their responsiveness. Methods. MPO-specific T cells were quantified by enzyme-linked immunospot assay with additional Treg cell depletion or exogenous interleukin-2. Serum tryptophan and its metabolites were measured. In vivo blockade of indoleamine 2,3-dioxygenase (IDO) was performed, and its effect on MPO reactivity was assessed. Results. During disease remission, MPO-specific interferon-␥–producing T cell frequencies were comparable with those found in healthy controls and significantly lower than those found in patients with acute disease. CD4ⴙCD25ⴙ regulatory cells did not play a role in maintaining these low MPO-specific T cell frequencies, since depletion of Treg cells did not augment MPO-specific responses, and FoxP3 levels were diminished in patients compared with controls. Treg cell function, however, was comparable in patients and controls, suggesting numerical rather than functional deficiency. We found diminished serum tryptophan levels and elevated levels of its metabolite kynurenine in patients with MPO AAV as compared with controls. To confirm the effect of tryptophan degradation on MPO responses in vivo, we inhibited degradation in MPOimmunized WKY rats and found greater immune responsiveness to MPO and a tendency to more severe glomerulonephritis. Conclusion. Our findings indicate that MPOspecific T cell frequencies are regulated during disease remission in association with tryptophan degradation. The tryptophan regulatory pathway is induced during active disease and persists during disease remission. Supported by the National Institute for Health Research (UK) Biomedical Research Centre Funding Scheme. Dr. Salama’s work was supported in part by a grant from the Vasculitis Foundation; he is also recipient of a National Institute for Health Research Clinician Scientist award. 1 Konstantia-Maria Chavele, PhD, Deepa Shukla, MSc, Tracey Keteepe-Arachi, MBBS, MRCP, Judith Anna Seidel, MSc, Charles D. Pusey, DSc, Alan D. Salama, MBBS, PhD, FRCP: Imperial College London and Hammersmith Hospital, London, UK; 2Dietmar Fuchs, PhD: Medical University of Innsbruck, Innsbruck, Austria. Address correspondence and reprint requests to Alan D. Salama, MBBS, PhD, FRCP, Renal Section, Division of Medicine, Imperial College London, Hammersmith Hospital, Du Cane Road, London W12 0NN, UK. E-mail: A.firstname.lastname@example.org. Submitted for publication June 11, 2009; accepted in revised form February 4, 2010. The antineutrophil cytoplasmic antibody– associated vasculitides (AAV) are a group of relapsing– remitting systemic diseases that result in considerable morbidity and mortality. Up to 25% of patients develop end-stage renal failure despite optimal immunosuppression. Understanding the immune mechanisms that regulate disease activity is critical for a more successful therapeutic approach to management of the disease. Recent work has highlighted the importance of both humoral and cellular effectors in disease pathogenesis. Pathogenic transplacental transfer of antimyeloperoxidase (anti-MPO) antibody to a neonate has been re1539 1540 CHAVELE ET AL ported (1), while animal studies have demonstrated the ability of anti-MPO antibodies to induce focal necrotizing crescentic glomerulonephritis (2) and influence transmigration of leukocytes (3). In vitro, antineutrophil cytoplasmic antibodies (ANCAs) have been shown to activate neutrophils and monocytes, which can then cause endothelial cell damage (4,5). ANCA-primed leukocytes demonstrate augmented adherence to and transmigration across blood vessel walls (6). However, renal biopsy sections demonstrate pauci-immune staining for immunoglobulins (7), while there is a preponderance of both macrophages and T cells in and around affected glomeruli (8). Our group and others have shown that T cells responding to the autoantigens MPO and proteinase 3 (PR3) are found in patients with active disease (9–11), and persistent T cell activation has also been reported (12). Additionally, ANCA are class-switched antibodies, which therefore require T cell help for their generation. Evidence of monocyte/macrophage activation has been shown by high levels of interleukin-6 (IL-6) and neopterin in the sera of patients with active disease (13). Our group has also demonstrated elevated levels of urinary monocyte chemoattractant protein 1 during active renal involvement (14). The mechanism underlying the breakdown in T cell tolerance to MPO or PR3 is unclear, as is the mechanism regulating autoreactive T cells in healthy individuals. Previous studies in another systemic autoimmune disease, Goodpasture’s disease, demonstrated that while central thymic deletion was involved, regulation was maintained by a population of Treg cells, which suppressed T cell responses to the Goodpasture autoantigen (15). In the present study, we sought to identify the regulatory factors that accounted for alteration in T cell reactivity with MPO in patients with MPO AAV. PATIENTS AND METHODS Patients. Patients with confirmed AAV who were followed up at the Hammersmith Renal Unit and were currently or previously positive for anti-MPO antibodies were selected and provided consent for participation in the study. Healthy controls were recruited from 2 different hospitals. The study was approved by the Hammersmith, Queen Charlotte’s and Chelsea Ethics Committee (ethics reference number 04/ Q0406/25). Leukocyte isolation and enrichment. Peripheral blood mononuclear cells (PBMCs) were obtained by separating whole blood using Lymphoprep (Nycomed) and were resuspended in the appropriate volume of complete RPMI medium, containing 10% human serum (London Transfusion Services), penicillin, streptomycin, and 2 mM L-glutamine (all from Gibco). Enzyme-linked immunospot (ELISpot) analysis. The frequency of cytokine-producing MPO-specific T cells was measured using ELISpot analysis using 20 g/ml of MPO, as previously described (15). Optimal MPO concentrations for use in the assay were established in preliminary studies. Viable PBMCs were enumerated using the vital dye trypan blue (BioWhittaker). The resulting spots were counted on a computer-assisted ELISpot Image Analyzer (Autoimmun Diagnostika). The frequencies were then expressed as interferon-␥ (IFN␥)–producing cells per million PBMCs. In additional experiments, putative CD25⫹ regulatory cells were depleted from PBMCs as previously described (15). Depletion was confirmed by flow cytometry using directly conjugated anti-CD4 and anti-CD25 monoclonal antibodies (both from PharMingen) using a FACSCalibur (Becton Dickinson) and analyzed with CellQuest software (Becton Dickinson). In further experiments, one fraction of cells was treated with exogenous IL-2 (30 units/ml; Roche Diagnostics) for the duration of the assay, as previously described (16). Finally, Treg cells were isolated with CD4CD25 beads (Miltenyi), and purified CD4CD25⫹ cells were used to suppress the response of CD4CD25⫺ cells stimulated (in a ratio of 1 Treg cell to 10 effector cells) with CD3CD28 expansion beads (Dynal). Tryptophan, kynurenine, and neopterin measurements. Sera were analyzed by high-performance liquid chromatography to quantify tryptophan and kynurenine levels as previously described (17), while measurement of neopterin was performed by enzyme-linked immunosorbent assay (ELISA; ELItest Neopterin Screening), according to the recommendations of the manufacturer (Brahms Diagnostica) and as previously described (18). Effect of indoleamine 2,3-dioxygenase (IDO) inhibition by 1-methyl-DL-tryptophan (1-MT) on in vivo anti-MPO responses. Wistar Kyoto (WKY) rats (n ⫽ 10) were immunized intramuscularly with human MPO (400 g/kg; Calbiochem) or with 1,000 g/kg of purified human MPO extracted from HL-60 cells in Freund’s complete adjuvant (Sigma). After 4 weeks (or 2 weeks for the higher dose), animals were assessed for urinary abnormalities and were matched for degree of hematuria. They were subsequently randomized to receive 50 mg of 1-MT in phosphate buffered saline (PBS) (n ⫽ 5) or placebo (n ⫽ 5) daily by intraperitoneal injection. Four weeks later (or 2 weeks later for the higher dose group), animals were killed. Sera were assayed for anti-MPO antibody production by ELISA, and for tryptophan levels by highperformance liquid chromatography. Splenocytes were isolated, and their proliferative responses to MPO in the presence or absence of the tryptophan metabolites quinolinic acid, 3-hydroxyanthranilic acid, or kynurenine (all at 100 M; Sigma) were tested using tritiated thymidine incorporation. Statistical analysis. Frequencies obtained before and after CD25 depletion as well as with and without IL-2 were analyzed using a paired t-test. All other comparisons between patients and controls, and between animal cohorts were analyzed using the Mann-Whitney U test. If 3 groups were compared, one-way analysis of variance (ANOVA) was used. T CELLS IN AAV 1541 Table 1. Demographic and clinical features of the patients with MPO AAV and healthy controls* Sex, % male Age, mean ⫾ SD years Diagnosis, % of patients MPA CSS Immunosuppression, no. (%) Prednisolone alone AZA alone Prednisolone and AZA Prednisolone and CYC MPO antibody titer, mean ⫾ SD IU/ml BVAS, median Serum creatinine level, median (range) moles/liter Patients with acute disease (n ⫽ 9) Patients with disease in remission (n ⫽ 23) 55 68 ⫾ 10.3 39 63.2 ⫾ 14.8 100 0 83 17 0 (0) 0 (0) 1 (11) 2 (22) 65.6 ⫾ 45.7 15 285 (94–1,136) 2 (9) 1 (4) 9 (39) 0 (0) 33.4 ⫾ 30.3 0 124 (63–567) Healthy controls (n ⫽ 50) 50 57.8 ⫾ 19.9 NA NA NA NA NA NA 6.9 ⫾ 0.9 0 77 (47–117) * MPO AAV ⫽ myeloperoxidase antineutrophil cytoplasmic antibody–associated vasculitis; MPA ⫽ microscopic polyangiitis; NA ⫽ not applicable; CSS ⫽ Churg-Strauss syndrome; AZA ⫽ azathioprine; CYC ⫽ cyclophosphamide; BVAS ⫽ Birmingham Vasculitis Activity Score. RESULTS Patient characteristics. We invited consecutive patients with known MPO AAV who were followed up in the vasculitis clinic at Hammersmith Hospital to participate. In addition, we enrolled a small number of patients who were newly diagnosed as having MPO AAV as positive controls and a group of healthy volunteers. In total, 32 patients with MPO AAV (of whom 23 had disease in remission and 9 had acute disease) and 50 healthy controls were recruited. Their demographic and clinical characteristics are shown in Table 1. Samples were obtained from patients with acute disease prior to the initiation of therapy, except for 3 of them: 1 who experienced a disease relapse during treatment with prednisolone and azathioprine, and 2 who had begun immunotherapy within the preceding 7 days. Patients with disease in remission were defined as those whose disease was in combined clinical and biochemical remission. Twelve of these 23 patients were receiving maintenance immunosuppression: 9 were taking combined prednisolone and azathioprine, 2 were taking prednisolone alone, and 1 was taking azathioprine alone. Patients with acute disease had a higher anti-MPO antibody titer and mean serum creatinine level than either of the other groups. However, the 3 groups were well matched for age (P ⬍ 0.05 by one-way ANOVA). While all of the patients with acute disease were diagnosed as having microscopic polyangiitis, 83% of the patients with disease in remission had microscopic polyangiitis, and 17% had Churg-Strauss syndrome. At the time of recruitment, only 7 of the 23 patients with disease in remission were MPO ANCA negative by ELISA. T cell frequencies in patients with MPO AAV during acute disease and remission. Using the IFN␥ ELISpot assay, we measured the frequencies of IFN␥producing T cells reactive with MPO in PBMCs obtained from patients during disease remission (n ⫽ 11), from patients with active disease (positive controls; n ⫽ 4), and from healthy controls (n ⫽ 14). Our data demonstrated that the frequency of MPO-reactive T cells was diminished to levels found in healthy controls during disease remission (mean ⫾ SEM spots/million PBMCs 39.5 ⫾ 8.3 in patients with disease in remission versus 24.7 ⫾ 8.7 in healthy controls; P not significant [NS]), while patients with acute AAV had significantly elevated T cell frequencies (mean ⫾ SEM spots/million PBMCs 172 ⫾ 39.6; P ⬍ 0.05 versus controls, by one-way ANOVA) (Figure 1) as has been described previously in proliferation experiments (9). No significant differences were found in T cell frequencies between patients with disease in remission who were taking maintenance immunosuppression and those who were not taking any therapy (mean ⫾ SEM spots/million PBMCs 30.9 ⫾ 28 in treated patients versus 28.7 ⫾ 25 in untreated patients), suggesting that the hyporesponsiveness was not solely due to immunosuppressive drugs. In addition, no differences were found between the MPO ANCA–positive and MPO ANCA–negative patients with disease in remission with respect to T cell frequencies (mean ⫾ SEM spots/million 1542 CHAVELE ET AL Figure 1. Autoreactive T cell frequencies in patients and controls. Myeloperoxidase (MPO)–specific interferon-␥ (IFN␥) frequencies were measured by enzyme-linked immunospot analysis in healthy controls (n ⫽ 14), patients with MPO antineutrophil cytoplasmic antibody–associated vasculitis (AAV) in remission (n ⫽ 11), and patients with acute MPO AAV (n ⫽ 4). Values are the mean and SEM number of spots/million peripheral blood mononuclear cells (PBMCs). ⴱⴱ ⫽ P ⬍ 0.05 versus controls. PBMCs 32.4 ⫾ 15 in ANCA-negative patients versus 30.1 ⫾ 12.9 in ANCA-positive patients). To investigate the factors regulating low T cell frequencies in patients with disease in remission, we repeated T cell assays with depletion of potential Treg cells or with the addition of exogenous IL-2 to reverse T cell anergy, as previously described (15,16). Responses to tetanus or mumps antigen were used as controls. CD25 cells were depleted as previously described (15) using magnetic bead depletion, which achieved a median reduction in CD25high cells (containing the putative regulatory cells) of 93%. Overall, in the cohort of patients with MPO AAV in remission who were tested (n ⫽ 7), we found no change in the MPO-specific IFN␥ frequency following CD25 depletion (median MPOspecific T cell frequency 16 spots/million PBMCs [range 0–114] predepletion and 39 spots/million PBMCs [range 0–65] postdepletion; P NS by paired t-test) (Figure 2A). When considering individual patients, we found none in whom the T cell frequency was statistically increased following CD25 cell removal, demonstrating that CD25⫹ Treg cells were not suppressing MPO-specific T cell responses, unlike our previous findings in patients with Goodpasture’s disease (15). In addition, using real-time polymerase chain reaction, we measured in whole blood samples, the quantity of messenger RNA specific for the Treg transcription factor FoxP3, normalized to the housekeeping gene GAPDH. We found significantly lower FoxP3 expression in patients with disease in remission (n ⫽ 10) than in controls (n ⫽ 10), suggesting that no significant population of FoxP3 regulatory cells had emerged fol- Figure 2. No evidence of Treg cells influencing MPO-specific T cell frequencies. A, Enzyme-linked immunospot frequencies of MPOspecific IFN␥ in patients with MPO AAV in remission (n ⫽ 7), before and after depletion (depl) of CD25⫹ cells. There was no overall increase in T cell frequencies following CD25 cell depletion (P not significant by paired t-test). B, Real-time polymerase chain reaction expression of FoxP3 relative to GAPDH in whole blood from patients with MPO AAV (n ⫽ 10) and healthy controls (n ⫽ 10). There was significantly lower FoxP3 expression in patients compared with controls. ⴱ ⫽ P ⬍ 0.05 by Mann-Whitney U test. C, Suppression of proliferation of CD25⫺ effector cells by purified CD25⫹ Treg cells obtained from patients with disease in remission (n ⫽ 11) or controls (n ⫽ 5), demonstrating an equal capacity of Treg cells from both patients and controls to suppress proliferation. Data are presented as box plots, where the boxes represent the interquartile range, the lines within the boxes represent the median, and the lines outside the boxes represent the range. See Figure 1 for other definitions. T CELLS IN AAV Figure 3. Evidence of T cell anergy. Enzyme-linked immunospot frequencies of MPO-specific IFN␥ were determined in the presence or absence of exogenous interleukin-2 (IL-2) in patients with MPO AAV in remission (n ⫽ 10). There was no overall increase in T cell frequencies following addition of IL-2 (P not significant by paired t-test). Data are presented as box plots, where the boxes represent the interquartile range, the lines within the boxes represent the median, and the lines outside the boxes represent the range. See Figure 1 for other definitions. lowing disease remission (Figure 2B). These data demonstrate the lack of a Treg cell population in patients with AAV, potentially explaining the tendency for these patients to experience disease relapse more than patients with Goodpasture’s disease (15). To further assess the functional capacity of Treg cells in patients, we purified Treg cells using positive CD25 bead selection and assessed their ability to suppress a nonspecific proliferative response in CD25⫺ effector cells, with a 1:10 ratio of Treg cells to effector cells. Overall, we found that Treg cells isolated from patients (n ⫽ 11) were equally capable of suppressing an effector response as Treg cells isolated from healthy controls (n ⫽ 5) (Figure 2C), although there was a wider spread in the patients than in the controls (median suppression 49% [range 16–92%] in patients with disease in remission, and 57% [range 42–61%] in controls; P NS). Taken together, these data suggest that there is a deficiency of Treg cells in patients with MPO AAV in remission but that those cells that are present are equally suppressive. To investigate T cell anergy, we performed similar ELISpot assays, but with the addition of IL-2. Overall, in the patient cohort (n ⫽ 10) there was no difference in MPO-specific T cell frequencies following the addition of IL-2 (median spots/million PBMCs 13.7 [range 0–193] without IL-2 and 42 [range 0–276] following the addition of IL-2; P NS) (Figure 3). However, in 2 individual patients there was an increase in T cell 1543 frequencies, which was statistically significant in only 1 patient (patient 11) (mean ⫾ SEM spots/million PBMCs 68.3 ⫾ 27.5 without IL-2 and 164.3 ⫾ 12.5 following the addition of IL-2; P ⫽ 0.012 by paired t-test). Therefore, for the majority of patients, we found that T cell anergy was not an important regulatory mechanism inducing MPO-specific T cell hyporesponsiveness. Levels of serum tryptophan and kynurenine in patients and controls. In the presence of inflammatory cytokines (IFN␥ and tumor necrosis factor), IDO is up-regulated and catabolizes the essential amino acid tryptophan. Since tryptophan depletion and elevation of its main metabolite kynurenine are associated with T cell hyporesponsiveness, we measured levels of tryptophan and kynurenine in sera from patients with acute disease and patients with disease in remission and compared them with those in healthy controls. We found that patients with acute disease had the lowest levels of serum tryptophan, followed by patients with disease in remission, and then controls (median 37.5 moles/liter [range 14.9–72] in patients with acute disease, 50.9 moles/liter [range 27–104.5] in patients with disease in remission, and 59.9 moles/liter [range 30.4–81.0] in healthy controls; P ⬍ 0.01 for patients with acute disease versus controls and P ⬍ 0.05 for patients with acute disease versus patients with disease in remission). Patients also had higher levels of kynurenine (median 3.2 moles/liter [range 1.7–5.4] in patients with acute disease, 2.84 moles/liter [range 1.1–5.6] in patients with disease in remission, and 1.87 moles/liter [range 1.04– 3.6] in healthy controls; P ⬍ 0.001 for patients with acute disease versus controls and for patients with disease in remission versus controls, by one-way ANOVA). There was also a significant difference between patients and controls with regard to the ratio of kynurenine to tryptophan (median 122.3 [range 40–130] in patients with acute disease, 52.2 [range 16.9–160.9] in patients with disease in remission, and 32.2 [range 19.1–79.5] in healthy controls; P ⬍ 0.0001 for patients with acute disease versus controls and P ⬍ 0.05 for patients with disease in remission versus controls, by one-way ANOVA) (Figure 4A). Since levels of tryptophan and its metabolites are known to be increased in patients with renal impairment (19), we normalized tryptophan to creatinine and still found significantly lower levels in patients compared with controls (median ratio of tryptophan to creatinine 0.08 moles/mole [range 0.05–0.45] in patients with acute disease, 0.43 moles/mole [range 0.05–1.17] in patients with disease in remission, and 0.72 moles/mole [range 0.58–0.905] 1544 Figure 4. Tryptophan degradation and neopterin production in patients with MPO AAV in remission. A, Serum ratio of kynurenine to tryptophan in healthy controls (n ⫽ 50), patients with MPO AAV in remission (n ⫽ 23), and patients with acute MPO AAV (n ⫽ 9). ⴱⴱ ⫽ P ⬍ 0.05; ⴱⴱⴱ ⫽ P ⬍ 0.0001 versus controls, by one-way analysis of variance (ANOVA). B, Serum tryptophan levels corrected for creatinine in healthy controls (n ⫽ 14), patients with MPO AAV in remission (n ⫽ 23), and patients with acute MPO AAV (n ⫽ 9). ⴱ ⫽ P ⬍ 0.05; ⴱⴱⴱ ⫽ P ⬍ 0.001 versus controls, by one-way ANOVA. C, Serum neopterin levels in healthy controls, patients with MPO AAV in remission, and patients with acute MPO AAV. ⴱ ⫽ P ⬍ 0.05; ⴱⴱ ⫽ P ⬍ 0.001 versus controls, by one-way ANOVA. D, Correlation between serum neopterin and the ratio of kynurenine to tryptophan. There was a significant correlation by Spearman’s rank analysis. In A–C, data are presented as box plots, where the boxes represent the interquartile range, the lines within the boxes represent the median, and the lines outside the boxes represent the range. See Figure 1 for other definitions. CHAVELE ET AL in healthy controls; P ⬍ 0.001 for patients with acute disease versus controls and P ⬍ 0.05 for patients with disease in remission versus controls) (Figure 4B). Finally, we examined whether immunosuppressive therapy could account for the differences in tryptophan and kynurenine levels. We found no significant differences in levels of kynurenine or tryptophan or in the calculated ratio between patients taking maintenance immunotherapy and those not taking maintenance immunotherapy. Serum neopterin levels. The notion that tryptophan degradation was due to monocyte/macrophage up-regulation of IDO through the action of proinflammatory cytokines was supported by the presence of elevated levels of serum neopterin in patients with acute disease and patients with disease in remission. Neopterin is a low molecular mass product of monocyte/macrophages, the up-regulation of which is similarly influenced by proinflammatory cytokines such as IFN␥ (13). In patients with acute MPO AAV and patients with MPO AAV in remission, serum neopterin levels were significantly elevated compared with those in healthy controls (median 30.2 nmoles/liter [range 9.8–93.8] in patients with acute disease, 9.9 nmoles/liter [range 3.3– 38.2] in patients with disease in remission, and 6.0 nmoles/liter [range 1.8–20] in healthy controls; P ⬍ 0.001 for patients with acute disease versus controls and P ⬍ 0.05 for patients with disease in remission versus controls, by one-way ANOVA) (Figure 4C). Serum neopterin levels remained elevated in patients compared with controls when corrected for renal function. Importantly, a close correlation existed between kynurenine-totryptophan ratios and neopterin concentrations (r2 ⫽ 0.644, P ⫽ 0.0007 by Spearman’s rank test) (Figure 4D). Similar to the tryptophan and kynurenine levels, maintenance immunosuppression did not alter neopterin levels. These data confirm and extend the data in the literature (20) demonstrating elevated neopterin and diminished tryptophan levels during acute inflammation, but critically suggest that patients with disease in remission have a degree of ongoing subclinical inflammation that leads to up-regulation of IDO (and neopterin), and hence degradation of tryptophan, which persists from the time of acute disease presentation. Interestingly, conventional maintenance immunosuppression does not seem to influence this ongoing enzymatic degradation of tryptophan or up-regulation of neopterin. Correlation of tryptophan degradation with immunoreactivity with MPO. We correlated the levels of tryptophan and kynurenine corrected for creatinine with T CELLS IN AAV 1545 Figure 5. Effect of in vivo indoleamine 2,3-dioxygenase (IDO) inhibition on myeloperoxidase (MPO) immunity. A, Serum tryptophan levels in rats left untreated or treated with 1-methyl-DL-tryptophan (1-MT) 8 weeks following immunization with MPO. The 1-MT–treated cohort had elevated tryptophan levels compared with controls, demonstrating effective IDO blockade. ⴱ ⫽ P ⫽ 0.026 by Mann-Whitney U test. B, Serum anti-MPO antibody titer, as assessed by anti-MPO enzyme-linked immunosorbent assay, in rats treated with placebo or 1-MT, demonstrating a nonsignificant increase in antibody titer. Values are the optical density (OD) at 450 nm. Data in A and B are presented as box plots, where the boxes represent the interquartile range, the lines within the boxes represent the median, and the lines outside the boxes represent the range. C, T cell proliferative responses to MPO, assessed using splenocytes from placebo-treated and 1-MT–treated animals, demonstrating a significant increase in MPO-specific T cell responses following prolonged IDO blockade. Values are the mean and SEM proliferation counts. ⴱⴱ ⫽ P ⫽ 0.006 by Mann-Whitney U test. D, Glomerular fibrinoid necrosis in animals immunized with high-dose MPO and treated with either placebo (n ⫽ 5) or 1-MT (n ⫽ 5), demonstrating a tendency for worse disease in the 1-MT group. Values are the mean and SEM. E, Suppression of MPO T cell proliferation with the tryptophan metabolites 3-hydroxyanthranilic acid (HIAA), quinolinic acid (QA), and kynurenine (kyn) in 5 animals immunized with MPO and treated with 1-MT, demonstrating significant suppression of proliferation by kynurenine and 3-hydroxyanthranilic acid. Values are the mean and SEM. ⴱ ⫽ P ⬍ 0.05; ⴱⴱ ⫽ P ⬍ 0.01 versus MPO control, by one-way analysis of variance. ccpm ⫽ cell counts per minute. the MPO ANCA titer in our cohort of patients. A negative correlation between the tryptophan-tocreatinine ratio and the MPO ANCA titer was found (r2 ⫽ ⫺0.43, P ⫽ 0.03), but no correlation was found for kynurenine levels. These data suggest that those patients with the greatest level of tryptophan degradation may have the highest MPO ANCA titer. This relationship was lost when patients with acute disease were omitted from the analysis, suggesting that in patients with disease in remission the anti-MPO antibody titer does not directly correlate with the degree of tryptophan degra- dation. Similarly, we did not find a correlation between tryptophan or kynurenine levels and absolute MPOspecific IFN␥-producing T cell frequencies, but these were already suppressed to control levels. Effect of IDO on MPO immunologic responses in vivo. In order to investigate the causal role of IDO in immune responsiveness in AAV, we immunized WKY rats with MPO, with or without IDO inhibition using 1-MT. WKY rats immunized with human MPO and treated with 1-MT (n ⫽ 5) had elevated levels of serum and urinary tryptophan as compared with placebo-treated animals (n ⫽ 1546 CHAVELE ET AL 5). The median serum tryptophan level was 155.2 moles/ liter (range 139–173) in the 1-MT–treated group versus 136.7 moles/liter (range 121.5–147.5) in the placebo group (P ⫽ 0.026 by t-test) (Figure 5A). The median urinary tryptophan level was 19.4 moles/liter (range 8.5– 40) in the 1-MT–treated group versus 4.2 moles/liter (range 0.8–8.8) in the control group (P ⫽ 0.0043 by Mann-Whitney U test). Of note, 1-MT–treated animals exhibited greater anti-MPO antibody titers (although the difference was not statistically significant) compared with the control group (Figure 5B). Splenocyte proliferation in response to MPO was significantly greater in the 1-MT– treated group (P ⫽ 0.006 by Mann-Whitney U test) (Figure 5C), demonstrating increased immunogenicity in animals in which the IDO pathway was antagonized. At a low dose of immunizing MPO, little glomerulonephritis was induced (21), and no differences were observed in renal pathology between the groups. However, with higher immunizing doses of MPO, we found similarly augmented anti-MPO responses and a tendency for more severe glomerulonephritis in the 1-MT group, with a greater number of glomeruli demonstrating crescents or fibrinoid necrosis (mean ⫾ SEM number of glomeruli with fibrinoid necrosis 1.6 ⫾ 1 in the control group versus 4.8 ⫾ 2.3 in the 1-MT–treated group) (Figure 5D). The effect of tryptophan metabolites on antiMPO T cell responses was examined in 5 animals immunized with MPO and treated with 1-MT. We found that kynurenine, quinolinic acid, and 3-hydroxyanthranilic acid all suppressed responsiveness to MPO, but only kynurenine and 3-hydroxyanthranilic acid did so significantly (P ⬍ 0.01 for 3-hydroxyanthranilic acid versus MPO control and P ⬍ 0.05 for kynurenine versus MPO control, by one-way ANOVA) (Figure 5E).These data demonstrate that not only kynurenine, but also other tryptophan metabolites, may contribute to T cell unresponsiveness. DISCUSSION In AAV, disease relapses are common, with ⬎30% of patients experiencing a relapse within 5 years of followup (22). The factors that regulate disease activity in patients with AAV have not been well defined. Debate continues as to whether ANCA levels per se correlate well with disease activity. Patients may be persistently antibody positive while their disease is in clinical and biochemical remission, and although some disease relapses are preceded by ANCA positivity or by an increase in ANCA titers, this is not true in all cases (23–25). However, T cells are known to play a role in AAV, as helpers for B cells as well as direct immune effectors. As such, they act as orchestrators of the autoimmune response. Hence, the factors that control T cell reactivity should impact disease phenotype and could provide useful therapeutic strategies for inducing longer-lasting disease-free remissions. We studied the regulation of T cell responses in patients with MPO AAV, since these patients have been shown to have fewer disease relapses than patients with PR3 AAV (22,26), suggesting a more robust immunologic regulation mechanism. We demonstrated that T cell responsiveness to MPO does diminish with disease activity, with patients with disease in remission having levels comparable with those in healthy controls. T cell responses are not suppressed by a population of antigenspecific Treg cells, unlike the effector T cells in Goodpasture’s disease, a “one hit” autoimmune disease (15). Similar findings demonstrating ineffective Treg cell suppression have been obtained in patients with Wegener’s granulomatosis (WG), in whom functional defects in Treg cells have been demonstrated during disease remission, despite augmented numbers of circulating CD4⫹CD25⫹ FoxP3-expressing Treg cells (27). However, unlike the Treg cells in WG patients, we found that the isolated Treg cells from MPO AAV patients were equally capable of suppressing nonspecific responses. This is more in keeping with the numerical deficit of Treg cells found in patients with ANCA-associated Churg-Strauss syndrome, which differentiates them from patients with eosinophilic pneumonia and asthma (28). In only 1 patient did we find that T cell hyporesponsiveness was due to anergy, reversed by exogenous IL-2. T cell anergy may result from T cell stimulation in the absence of adequate costimulation, which induces a particular pattern of gene activation, and results in specific effects, including dampened T cell receptor signaling and T cell cytokine production. However, in the presence of IL-2 signaling through its receptor, anergy is reversed, T cell proliferation is induced, and the particular anergic gene pattern is modulated. Recent data suggest that this effect of IL-2 occurs via mammalian target of rapamycin, JAK-3, and alterations in activated activator protein 1 levels (29). We then examined the tryptophan/kynurenine pathway as an explanation for lymphocyte hyporesponsiveness, and demonstrated an elevated kyneurenine to tryptophan ratio both in patients with acute MPO AAV and in patients with MPO AAV in remission. We found evidence that this was related to monocyte/macrophage activation, since increased levels of the monocyte prod- T CELLS IN AAV uct neopterin were detected in a similar pattern. Elevated levels of neopterin, as well as other monocyte/ macrophage cytokines, in patients with AAV during active disease and remission have also been reported by other groups (13,20,30). Alterations in tryptophan and its metabolites have been shown to contribute to immune regulation in maternal–fetal tolerance (31) and in renal transplant recipients (32), as well as in other autoimmune diseases (33,34). Tryptophan degradation and kynurenine generation are associated with T cell hyporesponsiveness and apoptosis. Additionally, under certain circumstances, tryptophan metabolism has been shown to induce a positive feedback loop augmenting FoxP3-expressing Treg cells (35,36). IDO is upregulated by proinflammatory cytokines in an attempt to modulate immune responsiveness and therefore represents a natural negative regulatory pathway. Additionally, CTLA-4 signaling through B7 in antigen-presenting cells may also up-regulate IDO (37). In our experimental model, immunization with MPO resulted in greater immune responsiveness to MPO when IDO was chronically inhibited, and resulted in a trend toward more severe glomerulonephritis. These data confirmed that modulation of the autoimmune MPO-specific response can be achieved by up-regulation of IDO and tryptophan degradation. We did not find an augmented Treg cell population in the presence of tryptophan degradation, unlike some previous studies (35,36), and this requires further investigation, since it may suggest another defect in regulation in MPO AAV patients. That disease relapses may still occur in MPO AAV demonstrates that tryptophan degradation alone may not always be sufficient to limit immune responsiveness in this disease, as is the case in transplant rejection, which can occur despite significant tryptophan degradation (32). These data additionally suggest that even patients whose disease is in remission experience a degree of subclinical inflammation, leading to IDO up-regulation, which may serve as a more effective marker for tailoring immunosuppressive therapy. ACKNOWLEDGMENTS We are grateful to Dr. John Meeks (Department of Biochemistry, Imperial College Healthcare Trust) for help with creatinine assays and to Dr. Sarah Zaher (Imperial College London) for help with tryptophan metabolite studies. AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved 1547 the final version to be published. Dr. Salama had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Chavele, Shukla, Pusey, Salama. Acquisition of data. Chavele, Shukla, Keteepe-Arachi, Seidel, Fuchs, Salama. Analysis and interpretation of data. Chavele, Shukla, Keteepe-Arachi, Fuchs, Salama. REFERENCES 1. Bansal PJ, Tobin MC. Neonatal microscopic polyangiitis secondary to transfer of maternal myeloperoxidase-antineutrophil cytoplasmic antibody resulting in neonatal pulmonary hemorrhage and renal involvement. Ann Allergy Asthma Immunol 2004;93: 398–401. 2. Xiao H, Heeringa P, Hu P, Liu Z, Zhao M, Aratani Y, et al. Antineutrophil cytoplasmic autoantibodies specific for myeloperoxidase cause glomerulonephritis and vasculitis in mice. J Clin Invest 2002;110:955–63. 3. Little MA, Smyth CL, Yadav R, Ambrose L, Cook HT, Nourshargh S, et al. Antineutrophil cytoplasm antibodies directed against myeloperoxidase augment leukocyte-microvascular interactions in vivo. Blood 2005;106:2050–8. 4. Ewert BH, Jennette JC, Falk RJ. The pathogenic role of antineutrophil cytoplasmic autoantibodies. Am J Kidney Dis 1991;18: 188–95. 5. Lu X, Garfield A, Rainger GE, Savage CO, Nash GB. Mediation of endothelial cell damage by serine proteases, but not superoxide, released from antineutrophil cytoplasmic antibody–stimulated neutrophils. Arthritis Rheum 2006;54:1619–28. 6. Calderwood JW, Williams JM, Morgan MD, Nash GB, Savage CO. ANCA induces ␤2 integrin and CXC chemokine-dependent neutrophil-endothelial cell interactions that mimic those of highly cytokine-activated endothelium. J Leukoc Biol 2005;77:33–43. 7. Haas M, Eustace JA. Immune complex deposits in ANCAassociated crescentic glomerulonephritis: a study of 126 cases. Kidney Int 2004;65:2145–52. 8. Cunningham MA, Huang XR, Dowling JP, Tipping PG, Holdsworth SR. Prominence of cell-mediated immunity effectors in “pauci-immune” glomerulonephritis. J Am Soc Nephrol 1999;10: 499–506. 9. Griffith ME, Coulthart A, Pusey CD. T cell responses to myeloperoxidase (MPO) and proteinase 3 (PR3) in patients with systemic vasculitis. Clin Exp Immunol 1996;103:253–8. 10. King WJ, Brooks CJ, Holder R, Hughes P, Adu D, Savage CO. T lymphocyte responses to anti-neutrophil cytoplasmic autoantibody (ANCA) antigens are present in patients with ANCA-associated systemic vasculitis and persist during disease remission. Clin Exp Immunol 1998;112:539–46. 11. Popa ER, Franssen CF, Limburg PC, Huitema MG, Kallenberg CG, Cohen Tervaert JW. In vitro cytokine production and proliferation of T cells from patients with anti–proteinase 3– and antimyeloperoxidase–associated vasculitis, in response to proteinase 3 and myeloperoxidase. Arthritis Rheum 2002;46:1894–904. 12. Lamprecht P. Off balance: T-cells in antineutrophil cytoplasmic antibody (ANCA)-associated vasculitides. Clin Exp Immunol 2005;141:201–10. 13. Muller Kobold AC, Kallenberg CG, Tervaert JW. Monocyte activation in patients with Wegener’s granulomatosis. Ann Rheum Dis 1999;58:237–45. 14. Tam FW, Sanders JS, George A, Hammad T, Miller C, Dougan T, et al. Urinary monocyte chemoattractant protein-1 (MCP-1) is a marker of active renal vasculitis. Nephrol Dial Transplant 2004; 19:2761–8. 15. Salama AD, Chaudhry AN, Holthaus KA, Mosley K, Kalluri R, 1548 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. Sayegh MH, et al. Regulation by CD25⫹ lymphocytes of autoantigen-specific T-cell responses in Goodpasture’s (anti-GBM) disease. Kidney Int 2003;64:1685–94. Ng WF, Hernandez-Fuentes M, Baker R, Chaudhry A, Lechler RI. Reversibility with interleukin-2 suggests that T cell anergy contributes to donor-specific hyporesponsiveness in renal transplant patients. J Am Soc Nephrol 2002;13:2983–9. Widner B, Werner ER, Schennach H, Wachter H, Fuchs D. Simultaneous measurement of serum tryptophan and kynurenine by HPLC. Clin Chem 1997;43:2424–6. Murr C, Widner B, Wirleitner B, Fuchs D. Neopterin as a marker for immune system activation. Curr Drug Metab 2002;3:175–87. Walser M, Hill SB. Free and protein-bound tryptophan in serum of untreated patients with chronic renal failure. Kidney Int 1993; 44:1366–71. Nassonov E, Samsonov M, Beketova T, Semenkova L, Wachter H, Fuchs D. Serum neopterin concentrations in Wegener’s granulomatosis correlate with vasculitis activity. Clin Exp Rheumatol 1995;13:353–6. Little MA, Smyth L, Salama AD, Mukherjee S, Smith J, Haskard D, et al. Experimental autoimmune vasculitis: an animal model of anti-neutrophil cytoplasmic autoantibody-associated systemic vasculitis. Am J Pathol 2009;174:1212–20. Booth AD, Almond MK, Burns A, Ellis P, Gaskin G, Neild GH, et al. Outcome of ANCA-associated renal vasculitis: a 5-year retrospective study. Am J Kidney Dis 2003;41:776–84. Finkielman JD, Merkel PA, Schroeder D, Hoffman GS, Spiera R, St.Clair EW, et al. Antiproteinase 3 antineutrophil cytoplasmic antibodies and disease activity in Wegener granulomatosis. Ann Intern Med 2007;147:611–9. Sanders JS, Stassen PM, van Rossum AP, Kallenberg CG, Stegeman CA. Risk factors for relapse in anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis: tools for treatment decisions? Clin Exp Rheumatol 2004;22 Suppl 36:S94–101. Terrier B, Saadoun D, Sene D, Ghillani-Dalbin P, Amoura Z, Deray G, et al. Antimyeloperoxidase antibodies are a useful marker of disease activity in ANCA-associated vasculitides. Ann Rheum Dis 2009;68:1564–71. Jayne D, Rasmussen N, Andrassy K, Bacon P, Tervaert JW, Dadoniene J, et al. A randomized trial of maintenance therapy for CHAVELE ET AL 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. vasculitis associated with antineutrophil cytoplasmic autoantibodies. N Engl J Med 2003;349:36–44. Abdulahad WH, Stegeman CA, van der Geld YM, Doornbos-van der Meer B, Limburg PC, Kallenberg CG. Functional defect of circulating regulatory CD4⫹ T cells in patients with Wegener’s granulomatosis in remission. Arthritis Rheum 2007;56:2080–91. Saito H, Tsurikisawa N, Tsuburai T, Akiyama K. Involvement of regulatory T cells in the pathogenesis of Churg-Strauss syndrome. Int Arch Allergy Immunol 2008;146 Suppl 1:73–6. Dure M, Macian F. IL-2 signaling prevents T cell anergy by inhibiting the expression of anergy-inducing genes. Mol Immunol 2009;46:999–1006. Ohlsson S, Wieslander J, Segelmark M. Circulating cytokine profile in anti-neutrophilic cytoplasmatic autoantibody-associated vasculitis: prediction of outcome? Mediators Inflamm 2004;13: 275–83. Munn DH, Zhou M, Attwood JT, Bondarev I, Conway SJ, Marshall B, et al. Prevention of allogeneic fetal rejection by tryptophan catabolism. Science 1998;281:1191–3. Brandacher G, Margreiter R, Fuchs D. Clinical relevance of indoleamine 2,3-dioxygenase for alloimmunity and transplantation. Curr Opin Organ Transplant 2008;13:10–5. Kwidzinski E, Bunse J, Aktas O, Richter D, Mutlu L, Zipp F, et al. Indolamine 2,3-dioxygenase is expressed in the CNS and down-regulates autoimmune inflammation. FASEB J 2005;19: 1347–9. Torres MI, Lopez-Casado MA, Lorite P, Rios A. Tryptophan metabolism and indoleamine 2,3-dioxygenase expression in coeliac disease. Clin Exp Immunol 2007;148:419–24. Manches O, Munn D, Fallahi A, Lifson J, Chaperot L, Plumas J, et al. HIV-activated human plasmacytoid DCs induce Tregs through an indoleamine 2,3-dioxygenase-dependent mechanism. J Clin Invest 2008;118:3431–9. Brenk M, Scheler M, Koch S, Neumann J, Takikawa O, Hacker G, et al. Tryptophan deprivation induces inhibitory receptors ILT3 and ILT4 on dendritic cells favoring the induction of human CD4⫹CD25⫹ Foxp3⫹ T regulatory cells. J Immunol 2009;183: 145–54. Munn DH, Sharma MD, Mellor AL. Ligation of B7-1/B7-2 by human CD4⫹ T cells triggers indoleamine 2,3-dioxygenase activity in dendritic cells. J Immunol 2004;172:4100–10.