Abstract
Nitrotyrosine is widely recognized as a surrogate marker of up-regulated inducible NO synthase expression at sites of inflammation. However, the potential immunogenicity of autologous proteins containing nitrotyrosine has not previously been investigated. Herein, we used the I-EK-restricted T cell epitope of pigeon/moth cytochrome c (PCC/MCC88–103) to assess the ability of T cells to recognize ligands containing nitrotyrosine. Substitution of the single tyrosine (Y97) in PCC/MCC88–103 with nitrotyrosine abrogates recognition by the MCC88–103-specific T cell hybridoma 2B4. CBA (H2K) mice immunized with MCC88–103 or nitrated MCC88–103 peptides produce T cell responses that are mutually exclusive. Transgenic mice that constitutively express PCC under the control of an MHC class I promoter are tolerant toward immunization with MCC88–103, but exhibited a robust immune response against nitrated MCC88–103. Analysis of T cell hybridomas specific for nitrated-MCC88–103 indicated that subtle differences in TCR VDJ gene usage are sufficient to allow nitrotyrosine-specific T cells to escape the processes of central tolerance.
Protein-associated nitrotyrosine is widely recognized as a hallmark of inflammation and is often used as a surrogate marker for the up-regulation of inducible NO synthase (iNOS).3 Accumulation of protein-associated nitrotyrosine has been documented in inflammatory conditions of diverse origin such as atherosclerosis (1), respiratory disease (2), transplant rejection (3), multiple sclerosis (4), Alzheimer’s disease (5), celiac disease (6), arthritis (7, 8, 9), ischemia-reperfusion injury (10), autoimmune diabetes (11), autoimmune uveitis (12), and infectious diseases (13). Although the precise molecular mechanism(s) responsible for in vivo formation of nitrotyrosine is still not firmly established, peroxynitrite and other reactive nitrogen species generated during periods of inflammation are thought to be primary intermediates (14, 15). The conversion of tyrosine to nitrotyrosine has been shown to have considerable biological ramifications, including alteration of enzymatic activity (16). However, the effect of this modification on immunological reactivity has received only limited attention. A number of polyclonal sera (17) and mAbs (17, 18) specific for protein-associated nitrotyrosine have been generated (usually using nitrotyrosine-containing keyhole limpet hemocyanin as an immunogen). This clearly indicates that the nitrotyrosine moiety of modified proteins can contribute to an Ab epitope. These Abs have proven to be sensitive and specific tools for the detection of nitrotyrosine in vivo and in vitro. However, to the best of our knowledge, there is no prior evidence as to whether the conversion of tyrosine to nitrotyrosine can be detected by components of the cellular immune system. If conversion of tyrosine to nitrotyrosine in autologous proteins can render these proteins recognizable as immunogenic autoantigens, this could have important implications for autoimmune diseases. Recent studies have indicated that a host of post-translational modifications, including glycosylation (19, 20), phosphorylation (21, 22), and cysteinylation (23, 24), can all affect T cell immunoreactivity. Indeed, it has been proposed that these modifications may play a role in the initiation and/or maintenance of autoimmune disease by contributing to the breakdown of immunological tolerance (25, 26).
In the present study we have used an in vivo model system to determine whether the conversion of tyrosine to nitrotyrosine can affect Ag recognition by components of the cellular immune system. For this purpose we have used the well-established IEk-restricted peptide from the model Ag moth/pigeon cytochrome c (MCC88–103). The MCC88–103 peptide comprises one of the most extensively studied model systems for understanding MHC restriction and T cell recognition and continues to be widely used as a tool for understanding the processes of T cell tolerance and T cell activation (27, 28). The IEk-restricted MCC88–103 peptide contains a single tyrosine residue (Y97) that is not involved in MHC binding, but is critically important for T cell recognition (27, 29). Using this system we formally demonstrate that conversion of tyrosine to nitrotyrosine has profound consequences in terms of immunological reactivity. More importantly, we observed that self Ags containing a tyrosine to nitrotyrosine conversion can evade the process of central tolerance, rendering them potentially important autoantigens.
Materials and Methods
Experimental animals
Wild-type CBA/J mice were obtained from The Jackson Laboratory (Bar Harbor, ME). PCC-transgenic mice (28) bred onto a B10.A background (line 119) (30) were a gift from Dr. R. Schwartz (National Institutes of Health, Bethesda, MD). All animals were used at 6–12 wk of age.
Reagents
Human IL-2 was provided by Dr. C. Reynolds (National Cancer Institute, Bethesda, MD). IL-2 production was assessed using the IL-2-dependent cell line CTLL-2 (31) obtained from American Type Culture Collection (Manassas, VA). For ELISA experiments, IFN-γ capture Ab (R4-6A2) and biotinylated anti-mouse IFN-γ (XMG1.2) were purchased from BD PharMingen (San Diego, CA). The T cell hybridoma 2B4 and the IEk-expressing CHO cells have been described previously (29) and were gifts from Dr. M. Davis (Stanford, CA). The MCC88–103 peptide was synthesized by the Peptide Synthesis Facility at Queen’s University (Kingston, Canada), and the nitrated MCC88-103 (nMCC88–103) peptide was synthesized by Sigma-Genosys (Woodlawns, TX).
Cell preparation and culture
Mice (CBA/J and PCC-transgenic) were immunized in the rear footpad with 25 μg of peptide (MCC88–103 or nMCC88–103) mixed with IFA. Ten days after immunization, mice were euthanized, and the draining popliteal lymph node cells were recovered. Mononuclear cells were prepared from lymph nodes, and cells (2 × 105 cells/well in 96-well plates) were cultured for 3 days in RPMI 1640 (Life Technologies, Gaithersburg, MD) supplemented with 10% FBS (Sigma-Genosys), 25 mM HEPES buffer, 50 μM 2β-ME, penicillin/streptomycin, and l-glutamine (Life Technologies) in the presence or the absence of the indicated peptides. On day 3 of culture half the culture supernatant was removed for analysis of IFN-γ secretion by ELISA, and the remaining cells were pulsed overnight with [3H]thymidine (1 μCi/well) to measure peptide-induced proliferation. Lymph node cells from mice (CBA/J and PCC-transgenic) immunized with nMCC88–103 peptide were also stimulated with nMCC88–103 peptide in bulk (1 × 106 cells/well in six-well plates). After 10-day stimulation cells were fused with BW5147 thymoma cells (obtained from American Type Culture Collection) using previously described protocols (32) to generate T cell hybridomas. Hybridomas were cloned in 96-well plates by limiting dilution and were assessed for CD3 surface expression by flow cytometry and for IL-2 production in response to IEk-expressing CHO cells in the presence or the absence of MCC88–103 or nMCC88–103 peptide.
TCR β-chain gene usage
119-1F5 T hybridoma cells were lysed using TRIzol reagent (Life Technologies), and total RNA was used for analysis of TCR β-chain gene usage employing a degenerate primer PCR approach as previously described (33). Briefly, cDNA was prepared by standard methodology using oligo(dT) and reverse transcriptase. The TCR β-chain sequence was then amplified by PCR using a degenerate TCR β-chain variable region consensus primer 5′-TAAGCGGCCGCATGSLYTGGTAYWXXCAG-3′ (S = G/T; L = A/G/T; Y = C/T; W = A/C; X = A/G) and a β-chain constant region primer 5′-CAGCTCAGCTCCACGTGG-3′. The PCR reaction was performed using three cycles with a low stringency annealing step (1 min at 94°C, 1 min at 37°C, 1 min at 72°C), followed by 27 cycles using a high stringency annealing step (1 min at 94°C, 1 min at 55°C, 1 min at 72°C). The resulting PCR product was cloned into the TA vector (Invitrogen, San Diego, CA) and was subjected to automated nucleotide sequence analysis at the University of Ottawa Biotechnology Research Institute.
Results and Discussion
We first assessed the consequence of tyrosine to nitrotyrosine conversion in terms of recognition by the MCC88–103-specific T cell hybridoma 2B4. As previously described (29), 2B4 responds to MCC88–103 peptide presented in the context of IEk by synthesizing IL-2 in a dose-dependent manner (Fig. 1⇓C). In contrast, 2B4 T cells were completely nonresponsive to stimulation with a synthetic peptide (nMCC88–103) containing a nitrotyrosine in place of Y97, even at the highest concentrations tested. Although the loss of recognition of nMCC88–103 by the T cell hybridoma 2B4 constituted a negative result, this observation provided the first clear evidence that modification of tyrosine to nitrotyrosine could have an impact on the process of T cell recognition. Previous studies have indicated that Y97 of MCC88–103 is not involved in MHC binding (29); however, we wanted to eliminate the possibility that conversion of Y97 to nitrotyrosine disrupted presentation of the nMCC88–103 peptide by MHC. To test this, we immunized CBA (H2k) mice with either MCC88–103 or nMCC88–103 peptides and assessed their cellular immune response using in vitro recall assays. Draining lymph node cells from mice immunized with MCC88–103 peptide secreted IFN-γ in a dose-dependent manner in response to in vitro stimulation with MCC88–103 peptide (Fig. 1⇓D). These same cells also secreted IFN-γ in response to stimulation with nMCC88–103 peptide; however, the response to the latter was significantly weaker than the response to MCC88–103 peptide. Strikingly, mice immunized with nMCC88–103 peptide had the opposite pattern of recognition. Cells from nMCC88–103-immunized animals secreted IFN-γ in response to in vitro stimulation with nMCC88–103 peptide, but were completely unresponsive to MCC88–103 peptide. The presence of a robust nMCC88–103-specific immune response provided strong evidence that conversion of the single tyrosine residue of MCC88–103 to nitrotyrosine does not significantly impact on the ability of this peptide to be presented by IEk. This is consistent with previous results showing that Y97 does not make direct contact with the peptide binding groove of IEk (27, 29). Furthermore, these results demonstrate unequivocally that conversion of tyrosine to nitrotyrosine has potentially profound consequences in terms of T cell recognition.
A, Alignment of PCC88–104, MCC88–103, and nMCC88–103 peptides showing the position of the single tyrosine residue (Y97), which is converted to nitrotyrosine (nY97) in nMCC88–103. B, Schematic comparison of the side chains of tyrosine and nitrotyrosine. C, Response of the hybridoma 2B4 (29 ) to peptides MCC88–103 and nMCC88–103. 2B4 T cell hybridoma cells (5 × 104) were mixed with 5 × 104 IEk-transfected CHO cells plus the indicated concentration of MCC88–103 or nMCC88–103 peptides. After 24 h, supernatants were recovered and assessed for IL-2 production using proliferation of the IL-2-dependent CTLL-2 cell line. Conversion of Y97 to nY97 in peptide nMCC88–103 completely abrogates recognition by 2B4 T cells. D, T cell specificity in mice immunized with MCC88–103 or nMCC88–103 peptides. CBA mice were immunized with MCC88–103 (left panel) or nMCC88–103 peptides (right panel) in IFA. Ten days postimmunization, draining lymph node cells were recovered and stimulated in vitro with the indicated concentrations of MCC88–103 or nMCC88–103 peptides. After 3 days of stimulation supernatants were recovered and assayed for the presence of secreted IFN-γ by ELISA. Data are plotted as the average IFN-γ produced in replicate wells ± SEM.
We next examined whether nitration of an autologous protein might be capable of rendering it immunogenic and potentially recognizable as an autoantigen. To address this question, we used transgenic mice that constitutively express PCC under the control of the MHC class I promoter (28). These mice are unresponsive to immunization with MCC88–103 due to the process of central tolerance whereby potentially autoreactive T cells are eliminated during maturation in the thymus via the process of negative selection. In agreement with earlier findings (28), we could not detect an MCC88–103-specific cellular immune response in PCC transgenic mice after immunization with MCC88–103 peptide (Fig. 2⇓). However, immunization of PCC transgenic mice with nMCC88–103 peptide elicited a robust cellular immune response against nMCC88–103, as measured by Ag-specific in vitro proliferation and IFN-γ production. Together these results indicate that autologous proteins containing a tyrosine to nitrotyrosine conversion may not be subject to the constraints of immunological tolerance. Accordingly, we propose that the high levels of tissue nitrotyrosine observed in various types of inflammation may actually be provoking a chronic inflammatory condition by acting as potential autoantigens.
Immune responses in PCC transgenic mice after immunization with MCC88–103 or nMCC88–103 peptides. PCC transgenic mice of an H-2k haplotype (line 119) (28 ) were immunized with MCC88–103 (upper panels) or nMCC88–103 peptides (lower panels) in IFA. Ten days postimmunization, draining lymph node cells were recovered and stimulated in vitro with the indicated concentrations (micromolar) of MCC88–103 or nMCC88–103 peptides. After 3 days of stimulation cells were pulsed with tritiated thymidine to determine the level of peptide-induced proliferation (left panels), and supernatants were recovered and assayed for the presence of secreted IFN-γ by ELISA (right panels). Data are plotted as the average stimulation index ([3H]thymidine incorporation in the presence of Ag/[3H]thymidine incorporation in the presence of medium only) of replicate wells plus the SEM (left panels) or the average IFN-γ produced in replicate wells ± SEM (right panels).
To further characterize the response of PCC transgenic mice against nMCC88–103 peptide we generated a panel of nMCC88–103-specific T cell hybridomas from mice that had been immunized with nMCC88–103 peptide (three from wild-type CBA mice and nine from PCC transgenic mice). Although the T cell hybridomas varied in terms of the absolute amount of IL-2 produced in response to Ag stimulation, all 12 hybridomas showed exquisite sensitivity and specificity for nMCC88–103 compared with MCC88–103 peptide; only two of the hybridomas (119-4C9 and CBA-4C8) exhibited weak responsiveness to nonmodified MCC88–103 peptide (Fig. 3⇓A). Once again, the reactivity of T cell hybridomas in response to stimulation with nMCC88–103 peptide plus IEk-transfected CHO cells confirms that nitration of Y97 does not abrogate the ability of the nMCC88–103 peptide to bind MHC, but, rather, selectively activates a distinct repertoire of IEk-restricted T cells. To gain a better understanding of this repertoire, TCR β-chain usage was evaluated for one of the hybridomas derived from PCC transgenic mice (119-1F5). The sequence of the TCR β-chain of 119-1F5 was very similar to that of a previously described MCC88–103-specific T cell clone (6.9R.D6) (34). Both clones use the combination of Vβ1 and Jβ1.2; however, they differ in the area of the D region (Fig. 3⇓B). Specifically, the TCR β-chain of the nMCC88–103-reactive T cell hybridoma contains a positively charged arginine residue within the D region, whereas the β-chain of the MCC88–103-reactive clone 6.9R.D6 does not. The presence of a positively charged residue in this position of the TCR might be expected based upon the critical role of the Y97 for T cell recognition and the addition of a negative charge upon conversion of tyrosine to nitrotyrosine. In addition, a recent description of the crystal structure of MCC88–103 bound to IEk indicates that the aromatic ring of Y97 is in close contact with the neighboring residue K99, and that Y97 plays a role in proper positioning of K99, which is a critical TCR contact residue (27). By introducing a negative charge on Y97, it is conceivable that the interaction betweenY97 and K99 has been disrupted. Regardless of the mechanism, it is interesting to note that the very subtle alteration of TCR β-chain usage found on nMCC88–103-specific T cells is sufficient to allow their escape from negative selection in PCC transgenic mice. Once again, this finding provides evidence that nitrated peptides are unlikely to participate in the process of thymic negative selection and that T cells bearing TCRs capable of recognizing nitrated self-peptides escape this process and enter the periphery.
A, Responses of a panel of T cell hybridomas raised against nMCC88–103. PCC transgenic mice of an H-2k haplotype (line 119) (28 ) and wild-type CBA mice were immunized with nMCC88–103 peptide in IFA. Ten days postimmunization, splenocytes were recovered and stimulated in vitro with nMCC88–103 peptide (10 μM) for 10 days. Activated cells were then fused with BW5147 thymoma cells to generate T cell hybridomas, which were subsequently cloned by limiting dilution on 96-well plates. T cell hybridomas (5 × 104) were mixed with 5 × 104 IEk-transfected CHO cells in the presence or the absence of MCC88–103 or nMCC88–103 peptides (10 μM). After 24 h supernatants were recovered and assessed for IL-2 production using proliferation of the IL-2-dependent CTLL-2 cell line. Data are plotted as the average [3H]thymidine incorporation of replicate wells of CTLL-2 cells. B, Characterization of the TCR β-chain V-D-J gene usage by the nMCC88–103-specific T cell hybridoma 119-1F5. Shown is the nucleotide and amino acid sequence of the β-chain V-D-J junction region for the 119-1F5 hybridoma aligned with the same region from the MCC88–103-specific T cell clone 6.9R.D6 (34 ). Identical nucleotides are indicated as dots; absent nucleotides are indicated by dashes.
Lastly, we began to assess the requirements for the production of nitrated ligands in vivo. We used the nMCC88–103-specific T cell hybridoma 119-1F5 as a tool to detect the conversion of MCC88–103 to nMCC88–103 (Figs. 1⇑ and 2⇑ suggest that 0.1 μM is the limit of detection for this system). When APC (peritoneal exudate cells or splenic mononuclear cells) from IEk-expressing mice were stimulated with IFN-γ plus LPS, they accumulated nitrite in the culture supernatant, consistent with up-regulation of iNOS (Table I⇓). However, when these same cells were pulsed with MCC88–103 while being treated with IFN-γ plus LPS, they did not convert MCC88–103 to nMCC88–103 at levels detectable by 119-1F5 T cells. This is consistent with the proposed role of the intermediate peroxynitrite in the formation of nitrotyrosine. Peroxynitrite is formed by the combination of NO plus superoxide, the latter of which is unlikely to be produced in the present system. Peroxynitrite itself is very toxic, and addition of low levels of peroxynitrite (1–2 μM) directly to the assay resulted in almost complete cell death (data not shown). However, when MCC88–103 peptide was treated with peroxynitrite and then added to the assay, 119-1F5 T cells responded robustly, suggesting that if peroxynitrite is produced at low concentrations in vivo, then it is likely that immunoreactive nitrated ligands could accumulate at sites of chronic inflammation. The results presented herein formally demonstrate that autologous proteins containing nitrotyrosine can elicit a robust anti-self-immune response. This phenomenon may be of broad potential significance in the field of autoimmune disease, as nitrotyrosine-containing proteins are often present in tissues that are targeted during autoimmunity (35). Thus, we hypothesize that nitrotyrosine-containing autologous proteins may be recognized by the immune system as foreign and may therefore contribute to the process of autoimmunity. Continuous activation of autoreactive nitrotyrosine-specific T cells would be expected to result in inflammation and an influx of inflammatory cells, such as neutrophils and macrophages, that produce superoxide and NO, leading to further production of nitrotyrosine-containing autoantigens. Indeed, treatment with inhibitors of iNOS has been repeatedly shown to have a dramatic beneficial effect in models of autoimmunity (36) or in clinical settings such as transplantation (37, 38). We are currently engaged in the process of identifying nitrated proteins at sites of inflammation associated with autoimmune disease to assess whether these proteins constitute targets of the cellular immune response.
Treatment of PEC with IFN-γ/LPS induces nitrite production, but does not render PCC recognizable by nPCC-specific T cells
Acknowledgments
We thank Dr. Mark Davis (Stanford, CA) for providing the 2B4 and IEk-transfected CHO cell lines, Dr. Ron Schwartz (National Institutes of Health) for providing the PCC transgenic mice, and Dr. S. Hedrick (University of California, San Diego) for critical reading of the manuscript.
Footnotes
↵1 This work was supported by grants (to H.C.B. and R.G.) from the Canadian Arthritis Network and the Arthritis Society. J.R.W. was supported by Canadian Institutes of Health Research.
↵2 Address correspondence and reprint requests to Dr. John R. Webb, Department of Biochemistry, Microbiology, and Immunology, Faculty of Medicine, University of Ottawa, 451 Smyth Road, Ottawa, Canada K1H 8M5. E-mail address: jwebb{at}uottawa.ca
↵3 Abbreviations used in this paper: iNOS, inducible NO synthase; PCC/MCC88–103, pigeon/moth cytochrome c; nMCC88-103, nitrated MCC88-103.
- Received February 26, 2003.
- Accepted May 19, 2003.
- Copyright © 2003 by The American Association of Immunologists
References
In this issue
