The recognition of multiple ligands by a single TCR is an intrinsic feature of T cell biology, with important consequences for physiological and pathological processes. Polyspecific T cells targeting distinct self-antigens have been identified in healthy individuals as well as in the context of autoimmunity. We have previously shown that the 2D2 TCR recognizes the myelin oligodendrocyte glycoprotein epitope (MOG)35–55 as well as an epitope within the axonal protein neurofilament medium (NF-M15–35) in H-2b mice. In this study, we assess whether this cross-reactivity is a common feature of the MOG35–55-specific T cell response. To this end, we analyzed the CD4 T cell response of MOG35–55-immunized C57BL/6 mice for cross-reactivity with NF-M15–35. Using Ag recall responses, we established that an important proportion of MOG35–55-specific CD4 T cells also responded to NF-M15–35 in all mice tested. To study the clonality of this response, we analyzed 22 MOG35–55-specific T cell hybridomas expressing distinct TCR. Seven hybridomas were found to cross-react with NF-M15–35. Using an alanine scan of NF-M18–30 and an in silico predictive model, we dissected the molecular basis of cross-reactivity between MOG35–55 and NF-M15–35. We established that NF-M F24, R26, and V27 proved important TCR contacts. Strikingly, the identified TCR contacts are conserved within MOG38–50. Our data indicate that due to linear sequence homology, part of the MOG35–55-specific T cell repertoire of all C57BL/6 mice also recognizes NF-M15–35, with potential implications for CNS autoimmunity.

The recognition of multiple distinct ligands by individual T and B lymphocytes is an inherent property of the adaptive immune response that can be referred to as polyspecificity (1, 2). At the T cell level, polyspecificity is relevant for thymic development and for homeostasis in the periphery. For these processes, T cells receive signals through the TCR from ligands different from those that will result in full-blown activation during a T cell response (36). Conceptual and mathematical models argue that polyspecificity is a necessary compromise to cope with the limited number of T cells one’s immune system can harbor and with the vast number of ligands T cells should respond to (7). Experimental data using a variety of TCR, including self-reactive TCR, have confirmed that a large number of distinct peptides can act as agonists for a given T cell (810).

Owing to imperfect allelic exclusion at the TCRα locus, a proportion of T cells can express one rearranged TCRβ-chain and two functional TCRα-chains and therefore express two distinct TCR bearing distinct specificities (1113). This dual TCR expression can either broaden dominant tolerance when the dual TCR are expressed on Foxp3+ regulatory T cells (Tregs) (14) or predispose to autoimmunity (1517). In particular, dual TCR expression accounted for the triggering of autoimmunity by viral infection in a mouse model of CNS autoimmunity (18). Nevertheless, the presence of dual TCR expression is not necessary to explain the polyspecificity of a given T cell and polyspecific T cell clones expressing a single TCR have been identified (1924).

One of the facets of polyspecificity in the context of autoimmune diseases is molecular mimicry, defined as sequence similarities between foreign and self-antigens that result in the cross-activation of autoreactive T and B cells by microbial Ags (2537). Originally, an important degree of sequence homology was considered necessary to explain molecular mimicry, but subsequent extensive dissection with synthetic peptide combinatorial libraries revealed that TCR Ag recognition is quite degenerate, and Ag similarity, resulting in the productive triggering of a given TCR, goes beyond sequence homology (10, 38, 39).

Some experimental data suggest that the concept of molecular mimicry can be extended to the activation of T cells by distinct self-antigens during an autoimmune response. For example, Avery et al. (40) described the simultaneous recognition of corneal Ags and an allotype-bearing peptide derived from IgG2a by T cell clones in a model of autoimmune stromal keratitis. Interestingly, mouse strains expressing the cross-reactive IgG2a allotype displayed full resistance to keratitis, as a result of IgG2a-derived peptides tolerizing the cornea-specific autoreactive T cell clones. These results suggest that molecular mimicry between circulating and sequestered self-antigens might be important in maintenance of self-tolerance (40). Conversely, self molecular mimicry could have deleterious consequences when two self-antigens are coexpressed in the target organ of an autoimmune response (41, 42).

We previously reported that T cell polyspecificity can result in cumulative autoimmunity in a spontaneous mouse model of multiple sclerosis. The 2D2 transgenic mice express a public TCR specific for the myelin oligodendrocyte glycoprotein (MOG) 35–55 peptide presented by I-Ab and develop spontaneous optic neuritis and encephalomyelitis driven by activated transgenic T cells (43). Unexpectedly, spontaneous autoimmune disease persisted, at the same frequency, in 2D2 mice that had been crossed on a MOG−/− background, implying the existence of an alternative antigenic target for 2D2 T cells in the CNS. This Ag was identified as neurofilament medium (NF-M), an intermediate filament of the axonal cytoskeleton (44). Indeed, 2D2 T cells are able to react to both MOG35–55 and NF-M15–35 in vitro, with NF-M15–35 eliciting a heteroclitic response. Importantly, the transfer of activated 2D2 T cells into recipient mice expressing both or either MOG and NF-M revealed a cumulative impact of the two self-antigens as targets of an autoimmune attack.

The peptides MOG35–55 and NF-M15–35 present an important degree of sequence identity. Within the minimal core epitope of MOG35–55 for CD4 T cells of H-2b mice (45, 46), six amino acids are identical at corresponding positions within MOG40–48 (YRSPFSRVV) and NF-M20–28 (TRSSFSRVS). Because the MOG residues R41, F44, R46, and V47 are all important TCR contacts for the 2D2 TCR (45) and are shared with NF-M (44), we hypothesized that the observed bispecificity would rely on the ability of the 2D2 TCR to recognize the same residues in the context of NF-M. Moreover, because the core epitope MOG40–48 appears to be necessary and sufficient to the stimulation of MOG35–55-reactive clones other than the 2D2, we investigated whether bispecificity for MOG35–55 and NF-M15–35 is unique to the 2D2 T cell clone or more widespread among MOG35–55-specific T cells of C57BL/6 mice.

C57BL/6 mice where purchased from Charles River Laboratories (L'Arbresle, France). MOG-specific 2D2 TCR-transgenic mice on a C57BL/6 background (43), have been backcrossed with CD45.1 congenic (2D2) or RAG-2−/− (2D2 RAG-2−/−) animals. OVA-specific OT-II TCR-transgenic mice (47) were kindly provided by Dr. Sylvie Guerder (Centre de Physiopathologie Toulouse-Purpan). TCRα+/− mice on a C57BL/6 background (48) were kindly provided by Dr. Bernard Ryffel (Immunologie et Neurogénétique Experimentales et Moléculaires, Orléans, France).

All animals have been housed under specific pathogen-free conditions with the exception of TCRα+/− mice that were housed in a conventional facility. All experimental protocols were approved by the local ethics committee and are in compliance with European Union guidelines.

The MOG35–55 (MEVGWYRSPFSRVVHLYRNGK), NF-M15–35 (RRVTETRSSFSRVSGSPSSGF), and OVA323–339 (ISQAVHAAHAEINEAGR) peptides were purchased from Polypeptide Laboratories (San Diego, CA) with a purity grade >95%. The NF-M18–30 (TETRSSFSRVSGS) peptide and the alanine-mutated variants NF-M (Supplemental Table I), as well as the control peptide mRNA capping enzyme (mRCE; ENKHQRRLVKLLL), were purchased from GeneCust (Dudelange, Luxembourg), and purity was >85%.

Seven- to 12-wk-old C57BL/6 or TCRα+/− mice were immunized s.c. at the base of the tail with 100 μg MOG35–55 or 100 μg NF-M15–35 in 200 μl CFA (BD Difco, Franklin Lakes, NJ) containing 500 μg M. tuberculosis H37 RA (BD Difco). Nine days postimmunization, spleens and inguinal lymph nodes were collected, and CD4 T cells were purified from single-cell suspensions by negative selection using the Dynabeads Untouched Mouse CD4 Cells kit (Invitrogen, Grand Island, NY). CD4 T cells were then cultured in RPMI 1640 medium supplemented with 10% FCS in 96-well round-bottom plates with 30 Gy-irradiated splenocytes (2.5 × 105 CD4 T cells for 1 × 106 irradiated splenocytes) and stimulated with increasing concentrations of MOG35–55, NF-M15–35, or OVA323–339 peptides. Unfractioned spleen and lymph node cell suspensions were also put in culture in 96-well round-bottom plates (2 × 106 cells/well) with concentrations of MOG35–55 or NF-M15–35 ranging from 0.1 to 100 μg/ml. Where indicated, the pan anti–MHC class II (MHC-II) Ab (clone M5/114) was added to the culture medium at a concentration of 60 μg/ml. Cell cultures were incubated at 37°C with 5% CO2, and after 72 h, supernatants were harvested and stored at −80°C.

Most MOG35–55-specific T cell hybridomas were previously generated from draining lymph node cells and CNS-infiltrating mononuclear cells of MOG35–55-immunized C57BL/6 mice (49). The 2G4.3, 2G4.1, and 1E7.21 hybridomas were obtained from draining lymph node cells of MOG35–55-immunized C57BL/6 mice as previously described (50). Briefly, cells were stimulated with MOG35–55 for 10 d in HL-1 medium then expanded with IL-2 for an additional 2 d. Blastic cells were isolated after Ficoll (GE Healthcare, Waukesha, WI) centrifugation and fused with the TCR-negative BW5147 thymoma cell line using polyethylene glycol. Fused cells were cloned by limiting dilution and selected for MOG35–55 reactivity.

Lymph nodes and spleens were collected from 2D2 RAG-2−/− mice exhibiting no clinical signs of spontaneous experimental autoimmune encephalomyelitis (EAE). CD4 T cells were magnetically sorted via positive selection using mouse CD4 L3T4 MicroBeads (Miltenyi Biotec, Auburn, CA) according to the manufacturer’s instructions. CD4 T cells were then cultured in RPMI 1640 medium supplemented with 10% FCS in 96-well round-bottom plates with 30 Gy-irradiated splenocytes (2.5 × 105 CD4 T cells for 1 × 106 irradiated splenocytes) and stimulated with increasing concentration of MOG35–55, NF-M15–35, or NF-M18–30 peptides. Cell cultures were incubated at 37°C with 5% CO2, and after 72 h, supernatants were stored.

For antigenic stimulation, hybridoma cells were cultured with 30-Gy–irradiated splenocytes at a 1:4 ratio and stimulated with MOG35–55, MOG38–50, NF-M15–35, and NF-M18–30 as well as each of the alanine-substituted NF-M peptides as indicated in the figure legends. After 24 h, supernatants were harvested and stored.

The cytokine concentrations in culture supernatants were assessed after 72 h (IFN-γ and IL-17A) or 24 h (IL-2) by sandwich ELISA. Assays were performed with DuoSet kits from R&D Systems (Minneapolis, MN), according to the manufacturer’s instructions. Plates were developed with an HRP colorimetric reaction, and cytokine concentrations were determined from absorbance values at 450 nm that were corrected for the background absorbance at 540 nm.

Unfractionated splenocytes from MOG35–55-immunized mice were incubated on polyvinylidene difluoride ELISPOT plates (Millipore, Billerica, MA) precoated with anti–IFN-γ or anti–IL-17A Abs from Mabtech (Nacka Strand, Sweden) or eBioscience (San Diego, CA) ELISPOT kits. Cells were restimulated with concentrations of MOG35–55 or NF-M15–35 ranging from 1 to 100 μg/ml and incubated at 37°C with 5% CO2 for 48 h. Cytokine spots were revealed with biotinylated anti–IFN-γ and anti–IL-17 mAbs, and streptavidin coupled to HRP according to the manufacturer’s instructions. Cytokine spots were then quantified on an ImmunoSpot S6 Ultra-V with CTL-ImmunoSpot software version 5.0 (CTL-Europe, Bonn, Germany).

Additionally, unfractionated draining lymph node cell suspensions were cultured overnight in RPMI 1640 medium supplemented with 10% FCS and stimulated with MOG35–55, NF-M15–35, or OVA323–339. On the following day, cells were incubated for 4 h with GolgiPlug (containing brefeldin A; BD Biosciences). Cells were then stained with a viability marker (Fixable Viability Dye eFluor 780; eBioscience) and the following mix of Abs: PE-Cyanine 7-CD90.2 (clone 53-2.1; eBioscience), Pacific Blue–CD4 (clone RM4-5; BD Pharmingen), PerCP-Cy5.5-CD8β (clone YTS156.7.7; BioLegend, San Diego, CA), Alexa Fluor 700–CD44 (clone IM7; eBioscience), allophycocyanin-anti–IFN-γ (clone XMG1.2; BD), and FITC-anti–IL-17 (clone TC11-18H10; BD Biosciences) Abs using a Cytofix/cytoperm PLUS kit (BD Biosciences). Stained cells were acquired on a BD Biosciences LSR II flow cytometer, and data were analyzed with BD Biosciences FACSDiva software version 6.2.

The 30-Gy–irradiated splenocytes from C57BL/6 mice were coincubated with the stimulatory peptide OVA323–339 (0.25, 2.5, 25, and 250 μM) and with the competitor MOG35–55, NF-M15–35, or alanine-substituted NF-M peptides (150, 30, and 6 μM) in PBS with 25% RPMI 1640 for 2 h. Peptide-pulsed splenocytes were washed three times, and 106 cells was incubated for 6 h with 2.5 × 105 CD4 OT-II cells purified by negative selection using the Dynabeads Untouched Mouse CD4 Cells kit. Cells were then stained with the following combination of Abs: PerCP-Cy5.5-CD69 (clone H1.2F3; BioLegend), Pacific Blue–CD4 (clone RM4-5; BD Pharmingen), PE-TCRβ (clone H57-597; BD Pharmingen), Alexa Fluor 700–CD45.1 (clone A20; BioLegend), allophycocyanin-CD62L (clone MEL-14; BD Pharmingen), and a viability marker (Fixable Viability Dye eFluor 780; eBioscience). Stained cells were acquired on a BD Biosciences LSR II flow cytometer, and data were analyzed with BD Biosciences FACSDiva software version 6.2.

The biochemical peptide-binding assay was based on binding inhibition of the radiolabeled ROIV peptide (sequence YAHAAHAAHAAHAAHAA) to I-Ab, as previously described (51). Purified I-Ab molecules (5–500 nM) were coincubated for 48 h with various concentrations (120 nM to 120 μM) of unlabeled peptide inhibitors and 1–10 nM [125I] radiolabeled ROIV. The reaction was carried out at pH 5.5 in PBS containing 0.7% digitonin and in the presence of a protease inhibitor mixture. Following incubation, I-Ab/peptide complexes were separated from free peptide by size-exclusion gel filtration chromatography on TSK200 columns (TosoHaas 16215; Tosoh, Montgomeryville, PA). In competitive assays, the concentration of peptide yielding 50% inhibition of the binding of the radiolabeled peptide was calculated. Under the conditions used, the measured IC50 values are reasonable approximations of true KD values.

RNA from MOG35–55-specific hybridomas was isolated using TRIzol (Invitrogen) and reverse-transcribed with the SuperScript III First-Strand kit for RT-PCR (Invitrogen). cDNA samples were then amplified using either Cα- or Cβ-specific reverse primers and a panel of Vα or Vβ-specific forward primers corresponding to unique 5′ coding regions of the different variable regions (Supplemental Table II). Vβ primers were named according to Wilson et al. (52), whereas Vα primers were named according to the International Immunogenetic Information System (www.imgt.org).

The PCR products were run on a 1.5% agarose gel containing ethidium bromide at 500 ng/ml and purified with the QIAquick Gel Extraction Kit (Qiagen) according to the manufacturer’s instructions. The Vα-Cα and Vβ-Cβ amplicons were then sequenced with the same primers at the local Genomics facility (http://get.genotoul.fr) on a 3130xl Genetic Analyzer (Applied Biosystems, Foster City, CA). The obtained sequences were analyzed with 4Peaks Version 1.7.2 and Bl2seq (NCBI-BLAST) and then aligned with the database of the International Immunogenetic Information System (http://www.imgt.org).

Minimal core epitopes were detected computationally as previously described (46). Briefly, the experimental structure of I-Ab [Protein Data Bank code 1lnu (53)] was analyzed, and a 4 × 20 binding preference matrix was constructed, which tabulated estimates of the tendency of an amino acid to bind in each of the specificity determining pockets P1, P4, P6, and P9. The binding estimates were based on the physical properties of the pockets such as size and charge. Highly and moderately preferred residues were given a score of −2 and −1, respectively; residues that would neither contribute nor hinder binding scored 0, and residues that were likely to hinder binding weakly or strongly (e.g., by being too large) were given a score of 1 or 2. The binding ability of overlapping 9-aa segments of NF-M15–35 was estimated as the sum of scores for the four pockets. Previous calculations with this in-house computer program successfully predicted minimal core epitopes, all of which scored −5 or lower (46, 54, 55). Initial model structures of the I-Ab peptide-binding domain in complex with 12-aa peptides that include the predicted core epitopes were constructed based on the experimental structure. Each initial model underwent a restrained molecular dynamics simulation in water in which the I-Ab molecule Cα atoms were restrained, whereas the peptide and the solvent were free to move. We used the Accelrys package (Accelrys Software, San Diego, CA) for these computations and for figure preparation.

The MOG35–55-specific CD4 T-cell repertoire of C57BL/6 mice is focused on the MOG40–48 core peptide (46). When presented within I-Ab, this MOG epitope provides four TCR contact residues that are all conserved in the NF-M20–28 peptide. This suggests that MOG35–55-specific CD4 T cells other than the 2D2 T cell clone might respond to NF-M15–35. To assess whether this is a general feature of C57BL/6 mice, we analyzed whether MOG35–55 immunization elicits T cells that react to NF-M15–35 Ag-recall stimulation ex vivo. Purified CD4 T cells from draining lymph nodes and spleens of MOG35–55-immunized mice were restimulated with MOG35–55, NF-M15–35, or a control peptide. Importantly, Ag-recall responses to NF-M15–35 by purified CD4 T cells could be detected in eight out of eight individual mice and in three pools of two mice (Fig. 1A, 1B). The recall response to the NF-M15–35 peptide was weaker than that generated with MOG35–55, the immunizing peptide, suggesting that not all MOG-reactive CD4 T cells cross-react with NF-M and/or that they exhibit a lower functional avidity for the NF-M peptide. These responses were Ag specific, as no cytokine release was detected in response to the control peptide. The Ag specificity was further proven upon immunization of C57BL/6 mice with OVA323–339, which led to strong ex vivo recall responses to OVA323–339, without any detectable response to MOG35–55 or NF-M15–35 (Fig. 1C, 1D). This excludes that the observed T cell response to NF-M could be elicited by Mycobacterium-derived Ags or by non–TCR-mediated activation. Conversely, following immunization of C57BL/6 mice with NF-M15–35, CD4 T cells from the draining lymph nodes produced only a modest amount of IL-17 in response to either NF-M15–35 or MOG35–55 (Supplemental Fig. 1). Of note, the recall response to NF-M15–35 following MOG35–55 immunization was I-Ab dependent, as it was completely abolished by addition of an anti–MHC-II Ab but not an isotype-matched control Ab (data not shown). Moreover, no ex vivo response to NF-M15–35 was elicited from CD4 T cells purified from MOG35–55-immunized NOD mice (data not shown), indicating that bispecificity did not extend to the I-Ag7 allele. Collectively, these data indicate that MOG35–55 immunization activates CD4 T cells that also recognize the NF-M15–35 peptide in the context of I-Ab and that this is a common feature of C57BL/6 mice.

FIGURE 1.

CD4 T cells primed in vivo with MOG35–55 consistently respond to NF-M15–35 restimulation ex vivo. C57BL/6 mice were immunized with MOG35–55 (A and B) or OVA323–339 (C and D). Nine days after immunization, CD4 T cells were purified from the spleen and the production of IFN-γ (A and C) and IL-17 (B and D) was assessed following stimulation with MOG35–55, NF-M15–35, or OVA323–339. (A and B) Data represent the mean + SEM of background-subtracted values from 11 observations (8 individual mice and 3 pools of two mice) from 3 independent experiments. (C and D) Data are the mean + SEM values from six observations (three independent mice and three pools of two mice) from three independent experiments. Statistical comparison of the magnitude of the cytokine response to NF-M peptide versus OVA peptide: *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 1.

CD4 T cells primed in vivo with MOG35–55 consistently respond to NF-M15–35 restimulation ex vivo. C57BL/6 mice were immunized with MOG35–55 (A and B) or OVA323–339 (C and D). Nine days after immunization, CD4 T cells were purified from the spleen and the production of IFN-γ (A and C) and IL-17 (B and D) was assessed following stimulation with MOG35–55, NF-M15–35, or OVA323–339. (A and B) Data represent the mean + SEM of background-subtracted values from 11 observations (8 individual mice and 3 pools of two mice) from 3 independent experiments. (C and D) Data are the mean + SEM values from six observations (three independent mice and three pools of two mice) from three independent experiments. Statistical comparison of the magnitude of the cytokine response to NF-M peptide versus OVA peptide: *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

We next enumerated the MOG35–55 and NF-M15–35 bispecific T cells elicited by immunization with MOG35–55. First, we quantified IFN-γ and IL-17–producing cells by ELISPOT assays after stimulation with either MOG35–55 or NF-M15–35 (Fig. 2A, 2B). The number of IFN-γ–producing cells per spleen was ∼60,000 ± 8000 and 15,000 ± 3000 after restimulation with 100 μg/ml MOG35–55 or NF-M15–35, respectively. The number of IL-17–producing cells per spleen was 40,000 ± 8000 in response to MOG35–55 and 16,000 ± 3000 in response to NF-M15–35. Then, we determined the frequency of lymph node CD4 T cells from MOG35–55-immunized C57BL/6 mice that produced either IFN-γ, or IL-17 or both cytokines after stimulation with MOG35–55 or NF-M15–35 by intracellular FACS staining (Fig. 2C, 2D). MOG35–55 stimulation at a dose of 100 μg/ml revealed that on average 2.3% of activated CD4 T cells have acquired a Th1 (IFN-γ+) phenotype, whereas 3% were Th17 (IL-17+) and 1% produced both IFN-γ and IL-17. After NF-M15–35 stimulation, these values were on average 1.3% for Th1, 0.9% for Th17, and 0.3% for the double producers. Whereas the mean fluorescence intensity for IL-17 was similar in NF-M– versus MOG-stimulated Th17 cells, a 2-fold reduction in IFN-γ mean fluorescence intensity was detected in Th1 cells following NF-M versus MOG restimulation. Collectively, these results reveal that after immunization with MOG35–55, an important proportion of activated MOG35–55-specific CD4 T cells is capable of responding to NF-M15–35 by producing proinflammatory cytokines. Depending on the dose of the stimulus and the cytokine produced, this proportion ranges from 25 to 60%.

FIGURE 2.

Following immunization with MOG35–55, a large proportion of T cells respond to NF-M15–35 ex vivo. (A and B) Ag-specific cells among splenocytes of MOG35–55-immunized mice were enumerated at day 9 postimmunization by IFN-γ or IL-17 ELISPOT after restimulation with MOG35–55, NF-M15–35, or OVA323–339. The panels represent IFN-γ (A) or IL-17 (B) spot formation after restimulation of 200,000 splenocytes per condition, in duplicate. Only occasional spots were observed without stimulation (0.6 ± 0.3 spots for IL-17; 0.6 ± 0.3 spots for IFN-γ) or in the presence of OVA323–339 (0.3 ± 0.2 spots for IL-17; 0.5 ± 0.3 spots for IFN-γ). The graphs represent the quantification of cytokine-producing cells per spleen, showing the mean plus SEM of eight individual mice from two independent experiments. (C and D) Draining lymph node cells from MOG35–55-immunized mice were restimulated overnight with MOG35–55, NF-M15–35, or OVA323–339. The frequency of Th1 and Th17 cells and IFN-γ/IL-17 double producers was assessed by intracellular cytokine staining at day 9. We gated on CD4 CD44hi T cells because only very few cytokine-producing cells were present in the CD44lo gate. (C) Representative FACS plots for each ex vivo restimulation condition. (D) Frequency of IFN-γ producers, IL-17 producers, and double producers among CD4 CD44hi T cells in response to different peptides. Data show the mean plus SEM of nine individual mice from two independent experiments. NS, no stimulatory Ag added.

FIGURE 2.

Following immunization with MOG35–55, a large proportion of T cells respond to NF-M15–35 ex vivo. (A and B) Ag-specific cells among splenocytes of MOG35–55-immunized mice were enumerated at day 9 postimmunization by IFN-γ or IL-17 ELISPOT after restimulation with MOG35–55, NF-M15–35, or OVA323–339. The panels represent IFN-γ (A) or IL-17 (B) spot formation after restimulation of 200,000 splenocytes per condition, in duplicate. Only occasional spots were observed without stimulation (0.6 ± 0.3 spots for IL-17; 0.6 ± 0.3 spots for IFN-γ) or in the presence of OVA323–339 (0.3 ± 0.2 spots for IL-17; 0.5 ± 0.3 spots for IFN-γ). The graphs represent the quantification of cytokine-producing cells per spleen, showing the mean plus SEM of eight individual mice from two independent experiments. (C and D) Draining lymph node cells from MOG35–55-immunized mice were restimulated overnight with MOG35–55, NF-M15–35, or OVA323–339. The frequency of Th1 and Th17 cells and IFN-γ/IL-17 double producers was assessed by intracellular cytokine staining at day 9. We gated on CD4 CD44hi T cells because only very few cytokine-producing cells were present in the CD44lo gate. (C) Representative FACS plots for each ex vivo restimulation condition. (D) Frequency of IFN-γ producers, IL-17 producers, and double producers among CD4 CD44hi T cells in response to different peptides. Data show the mean plus SEM of nine individual mice from two independent experiments. NS, no stimulatory Ag added.

Close modal

Dual TCR T cells express two TCR comprising the same β-chain paired with two distinct α-chains (12). The presence of a NF-M15–35 recall response following MOG35–55 immunization could be due to CD4 T cells that express two distinct TCR, one specific for MOG35–55/I-Ab and the other for NF-M15–35/I-Ab. In order to test this possibility, we took advantage of TCRα+/− mice because their T cells can only express one rearranged TCR (48, 56). Immunization of TCRα+/− C57BL/6 mice resulted in sizeable recall responses to MOG35–55, NF-M15–35 but not to OVA323–339 (Fig. 3A, 3B). The presence of NF-M15–35 cross-reactive T cells was confirmed by enumeration of cytokine-producing cells. Indeed, ELISPOT assays detected 17,000 ± 4000 IFN-γ+ cells and 9000 ± 2000 IL-17+ cells after MOG35–55 restimulation and 6000 ± 1000 IFN-γ+ cells and 4000 ± 2000 IL-17+ cells after restimulation with NF-M15–35. These results indicate that the bispecificity displayed by the MOG35–55-reactive repertoire of C57BL/6 mice persists in the absence of dual TCR-expressing T cells, implying that individual TCR are able to recognize the two self-antigens.

FIGURE 3.

The bispecificity of CD4 T cells for MOG35–55 and NF-M15–35 does not depend on the expression of dual TCR. Production of IFN-γ (A) and IL-17 (B) by splenic CD4 T cells from MOG35–55-immunized TCRα+/− mice evaluated by ELISA following restimulation with MOG35–55, NF-M15–35, or OVA323–339. Data represent the mean + SEM values from seven individual mice from two independent experiments. Statistical comparison of the magnitude of the cytokine response to NF-M peptide versus OVA peptide: **p < 0.01. (C and D) Ag-specific T cells among splenocytes of MOG35–55-immunized mice were enumerated by IFN-γ or IL-17 ELISPOT after restimulation with MOG35–55, NF-M15–35, or OVA323–339. The graphs represent the quantification of IFN-γ– (C) or IL-17–producing (D) cells per spleen. Data are the mean + SEM of eight individual mice from two independent experiments for IFN-γ and four individual mice from one experiment for IL-17.

FIGURE 3.

The bispecificity of CD4 T cells for MOG35–55 and NF-M15–35 does not depend on the expression of dual TCR. Production of IFN-γ (A) and IL-17 (B) by splenic CD4 T cells from MOG35–55-immunized TCRα+/− mice evaluated by ELISA following restimulation with MOG35–55, NF-M15–35, or OVA323–339. Data represent the mean + SEM values from seven individual mice from two independent experiments. Statistical comparison of the magnitude of the cytokine response to NF-M peptide versus OVA peptide: **p < 0.01. (C and D) Ag-specific T cells among splenocytes of MOG35–55-immunized mice were enumerated by IFN-γ or IL-17 ELISPOT after restimulation with MOG35–55, NF-M15–35, or OVA323–339. The graphs represent the quantification of IFN-γ– (C) or IL-17–producing (D) cells per spleen. Data are the mean + SEM of eight individual mice from two independent experiments for IFN-γ and four individual mice from one experiment for IL-17.

Close modal

To directly test whether bispecificity is a feature of single TCR and assess whether different TCR share this property, we analyzed 22 MOG35–55-specific T cell hybridomas generated from MOG35–55-immunized mice (Table I). We sequenced their TCR and showed that all of the hybridomas included in subsequent analyses are derived from independent T cell clones (Table I). The MOG35–55-specific T cell hybridomas were stimulated with serial dilutions of MOG35–55, NF-M15–35, or an unrelated peptide, and IL-2 production was measured as a readout of TCR triggering. All 22 unique T cell hybridomas as well as the 2D2 RAG−/− T cells responded to MOG35–55 but not to the control peptide (Fig. 4). Fifteen of these did not react to NF-M15–35 and were classified as MOG35–55 monospecific (Fig. 4, bottom left panel). The remaining seven hybridomas consistently responded to both MOG35–55 and NF-M15–35; four produced higher concentrations of IL-2 when stimulated with MOG35–55 (Fig. 4, bottom middle panel), whereas the other three responded more vigorously to NF-M15–35 than to MOG35–55 (Fig. 4, bottom right panel), as seen for 2D2 T cells (Fig. 4, top panel). Taken together, these data demonstrate that among the distinct MOG35–55-specific T cells tested, more than one-third cross-reacts with NF-M15–35. Importantly, these T cells use distinct TCR, suggesting that MOG35–55/I-Ab–specific TCR have an intrinsic propensity for recognition of the NF-M15–35 peptide.

Table I.
TCR sequencing of MOG35–55-specific T cell hybridomas and 2D2 T cells
Clone Identification No.SpecificityTRAVTRAJCDR3αTRBVTRBDTRBJCDR3β
4MOGN602 MOG 6D-6 16 C A L S P T S S G Q K L V F 19 2-4 C A S S P Q G G Q N T L Y F 
  5D-4 11 C A A S P S G Y N K L T F     
1E7.21 MOG 7-4 12 C A A S E P T G G Y K V V F 19 1/2 1-5 C A S S M D N Q A P L F 
1MOG26 MOG 9D-3 17 C A L S I G A N S A G N K L T F 12-2 2-7 C A S S L E T A G Y E Q Y F 
1MOG210 MOG 12D-2 18 C A L S D R G S A L G R L H F 19 2-3 C A S S I A D S S A E T L Y F 
1MOG233 MOG 9D-4 30 C V L T D D T N A Y K V I F 20 1-6 C G A R G D S Y S P L Y F 
3MOGN208 MOG 12-3 40 C A V N T G N Y K Y V F 23 2-5 C S S G T G G E D T Q Y F 
1MOG237 MOG 3-1 42 C A V S F S G G S N A K L T F 13-2 2-2 C A S G D L G N T G Q L Y F 
1MOG202 MOG 7-4 37 C A A T L T G N T G K L I F 12-2 2-7 C A S S L E G Q G F E Q Y F 
1MOG6 MOG 9D-4 47 C V S P M D Y A N K M I F 13-1 1/2 2-1 C A S S D G E L Y A E Q F F 
4MOGN604 MOG 14-2 21 C A A N Y N V L Y F 1-4 C A S S P L R D R G D E R L F 
1MOG35 MOG 6-6 31 C A L D N S N N R I F F 19 1/2 2-5 C A S S I T W D S Q D T Q Y F 
1MOG231 MOG 6-6 31 C A L D N S N N R I F F 13-3 2-5 C A S S D A G A G N Q D T Q Y F 
1MOG206 MOG 8-1 23 C A T D V N Y N Q G K L I F 1/2 2-4 C A S S Q D R G G Q N T L Y F 
2G4.1 MOG 12-3 23 C A L R N Y N Q G K L I F 13-3 1-2 C A S S D A G T K S N S D Y T F 
1MOG204 MOG 7D-2 23 C A A S N Y N Q G K L I F 20 1/2 2-5 C G A R D T Q D T Q Y F 
2MOG60 MOG/NF-M 9D-3 23 C A L S G Y N Q G K L I F 14 1-2 C A S R D S A N S D Y T F 
1MOG27 MOG/NF-M 7D-2 23 C A A R G Y N Q G K L I F 13-1 2-5 C A S S P D R G H Q D T Q Y F 
2G4.3 MOG/NF-M 7D-2 23 C A A I G Y N Q G K L I F 19 2-4 C A S S P P T G G G G D T L Y F 
3MOGN204 MOG/NF-M 3-1 23 C A V S G Y N Q G K L I F 13-2 2-3 C A S G G G L G G T S A E T L Y F 
2D2 MOG/NF-M 9N-3 23 C A V R S Y N Q G K L I F 16 1/2 1-1 C A S S L D P G A N T E V F F 
1MOG244 MOG/NF-M 3-1 43 C A V S G Y N N N A P R F 13-2 2-7 C A S G D A G T G Y E Q Y F 
1MOG213 MOG/NF-M 3-3 39 C A V S S Y N A G A K L T F 20 1-6 C G A R G T G G Y N S P L Y F 
1MOG9 MOG/NF-M 12D-3 17 C A L S A A N S A G N K L T F 13-2 2-4 C A S G D W G G E D T L Y F 
Clone Identification No.SpecificityTRAVTRAJCDR3αTRBVTRBDTRBJCDR3β
4MOGN602 MOG 6D-6 16 C A L S P T S S G Q K L V F 19 2-4 C A S S P Q G G Q N T L Y F 
  5D-4 11 C A A S P S G Y N K L T F     
1E7.21 MOG 7-4 12 C A A S E P T G G Y K V V F 19 1/2 1-5 C A S S M D N Q A P L F 
1MOG26 MOG 9D-3 17 C A L S I G A N S A G N K L T F 12-2 2-7 C A S S L E T A G Y E Q Y F 
1MOG210 MOG 12D-2 18 C A L S D R G S A L G R L H F 19 2-3 C A S S I A D S S A E T L Y F 
1MOG233 MOG 9D-4 30 C V L T D D T N A Y K V I F 20 1-6 C G A R G D S Y S P L Y F 
3MOGN208 MOG 12-3 40 C A V N T G N Y K Y V F 23 2-5 C S S G T G G E D T Q Y F 
1MOG237 MOG 3-1 42 C A V S F S G G S N A K L T F 13-2 2-2 C A S G D L G N T G Q L Y F 
1MOG202 MOG 7-4 37 C A A T L T G N T G K L I F 12-2 2-7 C A S S L E G Q G F E Q Y F 
1MOG6 MOG 9D-4 47 C V S P M D Y A N K M I F 13-1 1/2 2-1 C A S S D G E L Y A E Q F F 
4MOGN604 MOG 14-2 21 C A A N Y N V L Y F 1-4 C A S S P L R D R G D E R L F 
1MOG35 MOG 6-6 31 C A L D N S N N R I F F 19 1/2 2-5 C A S S I T W D S Q D T Q Y F 
1MOG231 MOG 6-6 31 C A L D N S N N R I F F 13-3 2-5 C A S S D A G A G N Q D T Q Y F 
1MOG206 MOG 8-1 23 C A T D V N Y N Q G K L I F 1/2 2-4 C A S S Q D R G G Q N T L Y F 
2G4.1 MOG 12-3 23 C A L R N Y N Q G K L I F 13-3 1-2 C A S S D A G T K S N S D Y T F 
1MOG204 MOG 7D-2 23 C A A S N Y N Q G K L I F 20 1/2 2-5 C G A R D T Q D T Q Y F 
2MOG60 MOG/NF-M 9D-3 23 C A L S G Y N Q G K L I F 14 1-2 C A S R D S A N S D Y T F 
1MOG27 MOG/NF-M 7D-2 23 C A A R G Y N Q G K L I F 13-1 2-5 C A S S P D R G H Q D T Q Y F 
2G4.3 MOG/NF-M 7D-2 23 C A A I G Y N Q G K L I F 19 2-4 C A S S P P T G G G G D T L Y F 
3MOGN204 MOG/NF-M 3-1 23 C A V S G Y N Q G K L I F 13-2 2-3 C A S G G G L G G T S A E T L Y F 
2D2 MOG/NF-M 9N-3 23 C A V R S Y N Q G K L I F 16 1/2 1-1 C A S S L D P G A N T E V F F 
1MOG244 MOG/NF-M 3-1 43 C A V S G Y N N N A P R F 13-2 2-7 C A S G D A G T G Y E Q Y F 
1MOG213 MOG/NF-M 3-3 39 C A V S S Y N A G A K L T F 20 1-6 C G A R G T G G Y N S P L Y F 
1MOG9 MOG/NF-M 12D-3 17 C A L S A A N S A G N K L T F 13-2 2-4 C A S G D W G G E D T L Y F 

For each clone, the V, (D), and J gene usage is presented for the α- and β-chain of the TCR, as well as the sequences of the CDR3α and CDR3β.

FIGURE 4.

Distinct MOG35–55-specific T cells also recognize NF-M15–35. Twenty-two T cell hybridomas obtained from MOG35–55-immunized C57BL/6 mice (bottompanel) were stimulated overnight with MOG35–55, NF-M15–35, or the unrelated peptide HA110–119 in the presence of irradiated splenocytes. CD4 T cells from 2D2 RAG−/− TCR-transgenic mice (top panel) were included as a positive control for reactivity to both MOG35–55 and NF-M15–35. Based on the IL-2 released in the supernatants, the clones could be classified into three categories, of which a representative profile is shown: monospecific clones responding only to MOG35–55 (bottom left panel), bispecific clones that respond more vigorously to MOG35–55 than to NF-M15–35 (bottom middle panel), and bispecific clones that respond more vigorously to NF-M15–35 than to MOG35–55 (bottom right panel). Data correspond to the mean + SEM of duplicate assessments from a representative experiment. Each T cell hybridoma was tested at least three times with consistent results.

FIGURE 4.

Distinct MOG35–55-specific T cells also recognize NF-M15–35. Twenty-two T cell hybridomas obtained from MOG35–55-immunized C57BL/6 mice (bottompanel) were stimulated overnight with MOG35–55, NF-M15–35, or the unrelated peptide HA110–119 in the presence of irradiated splenocytes. CD4 T cells from 2D2 RAG−/− TCR-transgenic mice (top panel) were included as a positive control for reactivity to both MOG35–55 and NF-M15–35. Based on the IL-2 released in the supernatants, the clones could be classified into three categories, of which a representative profile is shown: monospecific clones responding only to MOG35–55 (bottom left panel), bispecific clones that respond more vigorously to MOG35–55 than to NF-M15–35 (bottom middle panel), and bispecific clones that respond more vigorously to NF-M15–35 than to MOG35–55 (bottom right panel). Data correspond to the mean + SEM of duplicate assessments from a representative experiment. Each T cell hybridoma was tested at least three times with consistent results.

Close modal

These hybridomas use a single α/βTCR, except for 4MOGN602 that expresses one TCRβ and two in-frame TCRα mRNA sequences (Table I, Supplemental Table III). As previously reported (46), the MOG35–55/I-Ab complex can be recognized by a diverse TCR repertoire. Indeed, these 23 MOG-specific TCR (the 2D2-TCR and 22 hybridomas) use 20 different Vα-Jα and 21 Vβ-Jβ combinations. However, consistent with previous reports (57, 58), the Vβ8 (TRBV13-2), Vβ6 (TRBV19), and Jβ2.5 gene segments were overrepresented, as well as, to a minor extent, the Vβ1 (TRBV5), Vβ15 (TRBV20), and Jβ2.4 segments. For the TCRα-chain, a frequent usage of Vα5 (TRAV3), Vα3 (TRAV9D), Vα4 (TRAV6D), and Vα1 (TRAV7D) segments was observed. Similarly, the Jα17, Jα31, and Jα43 segments were each used twice, and, strikingly, the Jα23 gene segment was used by eight (35%) of the MOG-reactive TCR. Unexpectedly, no shared motif was detected in the CDR3β sequences of the MOG-reactive TCR (Table I) despite their common specificity to MOG40–48 and NF-M20–28, apparently in the same register, that shared the same TCR-contact residues (see below). Interestingly, CDR3α motifs appeared to differ between MOG mono- versus bispecific TCR. Indeed, the 1MOG35 and 1MOG231 monospecific hybridomas, while using different TCRβ-chains, shared the same TCRα-chain with the NSN junction. Moreover, 4 of the 15 MOG monospecific TCR harbored a NYN junctional motif. In striking contrast, 5 of the 8 bispecific TCR exhibited the GYN motif, in which the G resulted from 3 different codons generated by junctional diversity (Supplemental Table III), as compared with only 1 of the 15 monospecific TCR (p = 0.009; Fisher exact test). Collectively, these data indicate that different TCR sequences can lead to MOG and NF-M bispecificity and suggest that the shared GYN CDR3α motif, frequently associated with Jα23 gene usage, contributes to this bispecificity (Table I).

The importance of each amino acid within MOG40–48 (YRSPFSRVV) for I-Ab binding and for TCR stimulation has been previously assessed, experimentally and computationally, for 2D2 T cells and for polyclonal MOG35–55-reactive T cell lines (45, 46). The residues Y40, P43, S45, and V48 were proposed to be accommodated within the P1, P4, P6, and P9 pockets of I-Ab, respectively, whereas residues R41, S42, F44, R46, and V47, which are at least partially exposed in the MHC/peptide complex (pMHC), are likely sites for TCR binding. I-Ab pocket P1 is large, hydrophobic, and can accommodate the side chains of F and Y and of smaller aliphatic hydrophobic or polar amino acids. The pocket P4 can bind aliphatic hydrophobic or polar side chains but not charged side chains; the large aromatic side chains are less likely to bind in this pocket. Pocket P6 is restrictive and can only bind small side chains such as V or N, whereas the larger pocket P9 can also accommodate other aliphatic side chains. Notably, the small amino acids G, A, C, and S can, in principle, bind in each of the four pockets; however, their contribution to the peptide binding affinity is likely to be small. Based on these analyses, the computations suggested three possible minimal core epitopes within NF-M15–35: NF-M17–25 (VTETRSSFS), NF-M20–28 (TRSSFSRVS), and NF-M24–32 (FSRVSGSPS). One of these epitopes, NF-M20–28, exposes the same residue types as MOG40–48 for TCR contact. Thus, R21, S22, F24, R26, and V27 in NF-M20–28 correspond to R41, S42, F44, R46, and V47, respectively, in MOG40–48. NF-M20–28 is not an optimal I-Ab binding peptide because small residues are located in the specificity determining pockets P1 and P4, and therefore, its I-Ab binding is expected to be weaker than that of MOG40–48.

In order to investigate whether the recognition of the NF-M peptide by the bireactive T cells requires the same TCR contact residues as the recognition of the MOG peptide, we performed an alanine (A) scan on NF-M18–30 (Supplemental Table I). First, we ascertained for each hybridoma that the response to NF-M18–30 was strong enough to identify A-substitutions that induced a reduction of the response. The IL-2 response to NF-M18–30 was considered workable for all the hybridomas except for one (1MOG9) with greater reactivity to MOG35–55 than NF-M15–35. We then tested the remaining seven bispecific T cells (six hybridomas and 2D2 T cells) with the panel of NF-M18–30 peptides containing single A-substitution to evaluate the impact of these changes on the agonist activity of the peptides relative to that of the native NF-M18–30 (Table II). Based on the computational analysis, we expected that A-mutations of essential I-Ab or TCR binding residues would strongly affect the agonist activity, in particular mutations of the large residues E19, R21, F24, and R26 (position P-1, P2, P5, and P7 in the minimal core epitope NF-M20–28), all of which are predicted TCR contact. Notably, the predicted I-Ab binding residues are generally small (e.g., S23 [in P4] or S28 [in P9]), and their mutation is not expected to affect strongly the I-Ab binding and TCR activation. The S23A substitution induced an unchanged response in five clones and an augmented one for 2MOG60 and MOG244.2 cells. Interestingly, the S22A, S25A, and G29A substitutions caused increased agonist activity for the majority of clones with one notable exception: S25A had no agonist activity for 2MOG60 cells. A-substitution of S28 led to a decrease in agonist activity for two TCR (2D2 and 2MOG60) and was neutral for the others. A-substitution of the residues T18, T20, and S30 strongly reduced the production of IL-2 by the majority of clones tested (Table II). V27A (conserved in MOG38–50 at position V47) was an extremely poor agonist or even a null peptide for all the tested clones. Importantly, E19, R21, F24, and R26 are critical amino acids, as NF-M18–30 peptides with A-substitutions at these positions failed to elicit any consistent IL-2 production by the bireactive T cells (Table II).

Table II.
IL-2 response of the bispecific T cells to a panel of alanine-substituted NF-M18–30 peptides expressed as percentage of control values
Peptide CloneT18AE19AT20AR21AS22AS23AF24AS25AR26AV27AS28AG29AS30A
2D2 21 19 290 151 233 63 82 38 
3MOGN204 34 16 98 88 1187 167 355 47 
1MOG27 19 346 93 2092 96 107 25 
1MOG213 16 17 282 164 815 87 157 18 
2G4.3 64 39 209 169 272 126 190 70 
1MOG244 11 1424 478 4308 91 5478 30 
2MOG60 31 181 6000 7284 18 7803 
Peptide CloneT18AE19AT20AR21AS22AS23AF24AS25AR26AV27AS28AG29AS30A
2D2 21 19 290 151 233 63 82 38 
3MOGN204 34 16 98 88 1187 167 355 47 
1MOG27 19 346 93 2092 96 107 25 
1MOG213 16 17 282 164 815 87 157 18 
2G4.3 64 39 209 169 272 126 190 70 
1MOG244 11 1424 478 4308 91 5478 30 
2MOG60 31 181 6000 7284 18 7803 

Each clone was stimulated with 2.5, 25, or 250 μg/ml of native and each mutated peptides. For each of the three doses, the ratio of the response to a given mutated peptide divided by the response to native NF-M18–30 was calculated, and data represent the mean of the three ratios as percentages. Each clone was tested at least twice. A 0 was attributed to responses within background or inferior to 10% with no dose-response effect.

As mentioned above, the abrogation of the stimulatory capacity of A-mutated peptides following single position mutation can result from a compromised MHC binding, preventing the formation of stable pMHC complexes, or from mutation of a key TCR contact residue, pMHC complexes are formed but do not permit TCR triggering. To distinguish between these two possibilities, we assessed the capacity of the poorly stimulatory A-substituted NF-M peptides to compete with OVA323–339 for binding to I-Ab and consequently inhibit OT-II CD4 T cell activation. In the absence of competitor peptide or in the presence of an I-Ab nonbinder peptide [mRCE (59)], OT II CD4 T cells upregulated the activation marker CD69 in a dose-dependent manner in response to OVA323–339-pulsed APCs (Fig. 5A). Coincubation of APCs with OVA323–339 and the native MOG35–55 or NF-M15–35 peptide resulted in a significant reduction in the proportion of OT-II T cells expressing CD69 (Fig. 5A, 5B), validating this indirect method to determine the I-Ab binding capacity of a peptide. We then used this assay to predict the I-Ab binding of the nonstimulatory E19A, R21A, F24A, and R26A peptides. Two patterns were displayed in the competitive binding test. E19A and R21A failed to significantly inhibit CD69 expression by OT-II cells, suggesting that these A-substituted peptides exhibit very weak or no binding to I-Ab. One explanation could be that E19 and R21 contribute to the stable anchoring of NF-M18–30 to I-Ab. Conversely, F24A and R26A significantly inhibited CD69 induction, indicating that these two peptides bind to I-Ab molecules. The I-Ab binding of the A-substituted peptides was also tested with a biochemical competition assay using purified I-Ab molecules and an I-Ab-binding radiolabeled peptide. Consistent with the previous results, the E19A and R21A peptides displayed an IC50 of 10.7 and 36.3 μM, respectively, suggestive of very poor affinity for I-Ab (Fig. 5C). S28A, despite a weak affinity for I-Ab (IC50 >10 μM), proved a strong agonist for five out of seven clones. In contrast, all other A-substituted peptides, including F24A, R26A, and V27A exhibited IC50 within 5-fold change as compared with the parental NF-M18–30 peptide, implying that these peptides bind to I-Ab. Collectively, these observations strongly suggest that, whereas E19 and R21 may be I-Ab–binding residues, F24, R26, and V27 are key TCR contact residues of the NF-M18–30 peptide for all of the T cell clones tested, which is reminiscent of the contribution of the corresponding identical residues on MOG38–50. The importance of F24, R26, and V27 for TCR contact was predicted by the computational model (Fig. 6), as these residues are exposed in the I-Ab/NF-M20–28 complex. The importance of E19 and R21 for I-Ab binding is seemingly contradictory to the model, which suggests that E19 and R21 are partially exposed and therefore likely TCR contact residues. However, A-mutation experiments only indicate that a given residue is important for binding; they do not provide indication regarding the location of this residue in the complex. Thus, it is possible that although E19 and R21 contribute to I-Ab binding, they are not essential anchors and are not important for determining the specificity motif of I-Ab.

FIGURE 5.

The F24, R26, and V27 residues of the NF-M18–30 peptide are essential TCR contacts for the bispecific T cells. Competition assays to test the binding of selected NF-M alanine-substituted peptides to I-Ab. (A and B) OT II CD4 T cells were stimulated with irradiated splenocytes pulsed with OVA323–339 and different competitor peptides. Inhibition of CD69 induction on OT II CD4 T cells following stimulation with OVA323–339 and a competitor peptide, as compared with stimulation with OVA323–339 alone, was interpreted as binding of the competitor peptide to I-Ab. (A) Representative FACS histograms showing the expression of CD69 from CD62L negative OT II cells under the indicated stimulation conditions. The mRCE peptide is an I-Ab nonbinder (www.iedb.org). (B) Percentage of inhibition obtained for the OVA323–339 concentration of 2.5 μM and the competitor peptide concentration of 150 μM. Data are the mean ± SEM of five independent experiments (*p < 0.05, **p < 0.01, two-tailed paired Student t test using the no competitor value as reference). (C) Competitor peptides were coincubated with the radiolabeled ROIV peptide and purified I-Ab molecules. The concentration required to inhibit 50% of the binding of the ROIV peptide to I-Ab (IC50) was calculated for each peptide and related to the IC50 of native NF-M18–30. Values >1 indicate stronger binding, whereas values <1 reflect weaker binding. The A-substituted peptides resulting in a 5- or 10-fold reduction in binding to I-Ab are highlighted in dark and light gray, respectively. The IC50 values for MOG35–55, NF-M15–35, and NF-M18–30 were 247, 685, and 1760 nM, respectively.

FIGURE 5.

The F24, R26, and V27 residues of the NF-M18–30 peptide are essential TCR contacts for the bispecific T cells. Competition assays to test the binding of selected NF-M alanine-substituted peptides to I-Ab. (A and B) OT II CD4 T cells were stimulated with irradiated splenocytes pulsed with OVA323–339 and different competitor peptides. Inhibition of CD69 induction on OT II CD4 T cells following stimulation with OVA323–339 and a competitor peptide, as compared with stimulation with OVA323–339 alone, was interpreted as binding of the competitor peptide to I-Ab. (A) Representative FACS histograms showing the expression of CD69 from CD62L negative OT II cells under the indicated stimulation conditions. The mRCE peptide is an I-Ab nonbinder (www.iedb.org). (B) Percentage of inhibition obtained for the OVA323–339 concentration of 2.5 μM and the competitor peptide concentration of 150 μM. Data are the mean ± SEM of five independent experiments (*p < 0.05, **p < 0.01, two-tailed paired Student t test using the no competitor value as reference). (C) Competitor peptides were coincubated with the radiolabeled ROIV peptide and purified I-Ab molecules. The concentration required to inhibit 50% of the binding of the ROIV peptide to I-Ab (IC50) was calculated for each peptide and related to the IC50 of native NF-M18–30. Values >1 indicate stronger binding, whereas values <1 reflect weaker binding. The A-substituted peptides resulting in a 5- or 10-fold reduction in binding to I-Ab are highlighted in dark and light gray, respectively. The IC50 values for MOG35–55, NF-M15–35, and NF-M18–30 were 247, 685, and 1760 nM, respectively.

Close modal
FIGURE 6.

Model structure of the I-Ab/MOG38–50 and I-Ab/NF-M18–30 interaction. I-Ab is shown as a yellow solvent accessible surface; the peptides are shown as ball-and-stick models with carbon, nitrogen, and oxygen in green, blue, and red, respectively. Shown are side views of the I-Ab interaction with peptides. A part of the I-Ab surface was omitted to expose the peptide-binding groove. Peptide residues (indicated by letters) that bind in I-Ab pockets P1, P4, P6, and P9 are indicated under the peptide. The predictive I-Ab/MOG40–48 interaction was previously modeled (46).

FIGURE 6.

Model structure of the I-Ab/MOG38–50 and I-Ab/NF-M18–30 interaction. I-Ab is shown as a yellow solvent accessible surface; the peptides are shown as ball-and-stick models with carbon, nitrogen, and oxygen in green, blue, and red, respectively. Shown are side views of the I-Ab interaction with peptides. A part of the I-Ab surface was omitted to expose the peptide-binding groove. Peptide residues (indicated by letters) that bind in I-Ab pockets P1, P4, P6, and P9 are indicated under the peptide. The predictive I-Ab/MOG40–48 interaction was previously modeled (46).

Close modal

We report in this study that the capacity of I-Ab/MOG35–55-specific 2D2-TCR transgenic CD4 T cells to also recognize the I-Ab/NF-M15–35 complex is shared by an important proportion of polyclonal MOG35–55-specific CD4 T cells from regular C57BL/6 mice. Moreover, these bispecific autoreactive CD4 T cells are present in all mice tested and represent, therefore, a shared feature of C57BL/6 mice. In addition, we demonstrate that cross-recognition of the I-Ab/MOG35–55 and I-Ab/NF-M15–35 complexes occurs through a single TCR. The bispecific TCR may share CDR3α sequences, and, importantly, all bispecific TCR crucially interact with at least three residues shared by the MOG and NF-M peptides.

Bispecificity due to dual TCR expression on T cells is quite widespread and can potentially favor autoimmunity (1518). This is, however, not the basis for the bispecificity described in the current study as the bispecific T cell recognition persisted in TCRα+/− mice. Moreover, the analysis of independent MOG35–55-specific T cell hybridomas revealed that bispecificity exists at the level of a single TCR. In the same line, epitope spreading involving different cohorts of T cells cannot explain the bispecificity described in this study because recognition of both MOG35–55 and NF-M15–35 was observed early after immunization, in the absence of CNS inflammation, and occurred at the clonal T cell level. However, the polyspecificity intrinsically exhibited by T cells could contribute to the epitope spreading phenomenon. Indeed, CD4 T cells recognizing multiple self-antigens, such as the CD4 T cells studied in this paper, could preferentially be recruited during an ongoing autoimmune process. However, we cannot formally rule out that some of the hybridomas might originate from Tregs because Foxp3+ cells were not depleted from MOG-stimulated mononuclear cells prior to fusion. Nevertheless, using the very same protocol, no T cell hybridoma could be generated using fusion with sorted GFP+Foxp3+ (Treg), whereas we reliably succeeded in obtaining hybridomas from GFPFoxp3 effector T cells (data not shown). These results indirectly suggest that the panel of hybridomas studied in this paper is devoid of Treg-derived TCRs.

In our panel of 22 T cell hybridomas, we could identify 7 bispecific TCR that are different from each other and from the 2D2 TCR in term of Vα–Jα and Vβ–Jβ gene usage. One Vα4-Jα31 (TRAV6D-TRAJ31) rearrangement was fully shared by two MOG-monospecific TCR (1MOG35 and 1MOG231), resulting in a DNSNNRI CDR3α sequence resembling the public RNSNNRI sequence detected in MOG-reactive T cell lines and CNS-infiltrating T cells (60, 61). The 1MOG35 and 1MOG231 TCR also used the same Jβ gene segment but with different Vβ, resulting in different CDR3β lengths and sequences. More strikingly, the Jα23 gene segment was overrepresented among MOG-reactive TCR. Specifically, 5 out of 8 bispecific TCR use the Jα23 gene segment, as compared with 3 out of 15 monospecific. The germline-encoded motif YNQGKL is present in all Jα23+ TCR but mono- and bispecific TCR seem to differ in the identity of the amino acid preceding this motif. Indeed, a G was present at this position for four of the five bispecific and an N for the three monospecific TCR. Interestingly, whereas the N is germline-encoded, the G results from junctional diversity, suggesting a nonrandom association between this CDR3 residue and the bispecificity for MOG35–55 and NF-M15–35. Of note, the Jα23 gene segment is used in public rearrangements with the Vα9 segment, translating into NYNQGKL, SYNQGKL, or GYNQGKL CDRα motifs identical to the ones reported in this study (60, 61).

We investigated how bispecific TCR interact with NF-M15–35 in the context of I-Ab using A-substituted NF-M peptides. None of the bispecific T cells could be activated with substituted peptides at residues E19, R21, F24, or R26. The V27A peptide was at best a very weak agonist for all of the bispecific T cells tested. Analysis of the binding of I-Ab of the A-substituted NF-M peptides revealed that all but R21A retain binding capacity in the range of that of the parental NF-M18–30 peptide. E19 and R21, located in positions P1 and P2, likely contribute to I-Ab binding, yet they are exposed and their importance for TCR binding cannot be ruled out. Therefore, F24 and R26 are very likely key contact residues for all studied TCR, whereas V27 may be either an essential or important TCR contact residue depending on the TCR. Importantly, the same amino acids have been implicated in the TCR interaction with the MOG35–55/I-Ab complex, suggesting that the bispecificity described in this study relies on the sequence identity between MOG and NF-M at precise positions. Based on this shared TCR recognition pattern, we have searched using the BlastP program to identify self-antigens, other than MOG and NF-M, that bear this motif. Eight were identified but none was stimulatory toward the 2D2 T cells. In retrospect, because the MOG35–55-specific response is mainly focused on the core epitope MOG40–48 (45, 46), it was not unexpected that T cells other than the 2D2 clone recognize a similar core epitope within a distinct protein (NF-M). Why some MOG35–55-specific TCR exhibit this bispecificity, whereas others fail to respond to NF-M15–35 will be best addressed by structural studies. In addition, the somewhat different patterns of reactivity exhibited by the bispecific T cells to the mutant peptides suggest that their TCR exhibit several modes of I-Ab/NF-M15–35 complex recognition.

Experimental data have demonstrated that a large number of different peptides can stimulate self-reactive TCR with functional sensitivities comparable or superior to that of the self-peptide (9, 62). This applies to both MHC class I and MHC-II–restricted TCR. The highest estimate to date is in the order of a million different decamer peptides being recognized by a given self-reactive TCR (10). There is no doubt that this large level of cross-reactivity at the TCR level offers ample opportunity for molecular mimicry between the large world of foreign peptides and the more limited set of self-peptides. However, although more difficult to quantify given the bias imposed by thymic (and peripheral) negative selection, this wide TCR cross-reactivity provides room for the phenomenon of self-molecular mimicry described by several groups including ours (38, 40, 42, 44, 63).

Following immunization with MOG35–55, bispecific T cells can be revealed ex vivo by functional assays based on their production of IFN-γ and IL-17. Interestingly, bispecific CD4 T cells appear to be enriched in IL-17–producing T cells as compared with the global MOG35–55 response, consistent with previous reports that the Th1/Th17 ratio can be influenced by the functional avidity of the interaction between the TCR and the pMHC complexes (64). Our description of a consistent proportion of myelin-reactive CD4 T cells that responds to a neuronal Ag raises the question of the consequences of this phenomenon for CNS autoimmunity. NF-M is expressed in both the central and peripheral nervous system and thus is not as secluded as MOG (65). However, the fact that a response to MOG35–55 can be detected in the periphery argues against a scenario of cross-tolerization by NF-M, as in the autoimmune stromal keratitis model (40). Conversely, the self-molecular mimicry described in this study could confer unique pathogenic properties to the bispecific CD4 T cells during MOG35–55-induced EAE. For example, bispecific T cells may be more prone to infiltrate the CNS gray matter because NF-M, but not MOG, is strongly expressed in this region. This may result in enhanced pathogenicity as previously described for the 2D2 T cells using adoptively transfer EAE experiments (44). Concerning the possible involvement in EAE of the bispecific T cell clones studied in this paper, it is of note that two of them, 1MOG244.2 and 1MOG9, have already been shown in a retrogenic model to drive EAE following active immunization with MOG35–55, as well as to cause spontaneous disease (49). Future studies will thus aim at deciphering the pathogenic potential of MOG35–55 monospecific versus MOG35–55/NF-M15–35 bispecific CD4 T cells in the widely used EAE model induced by immunization of C57BL/6 mice with MOG35–55 (66). We have referred to the property of TCR that can recognize multiple distinct self-peptides and thereby increase the pathogenicity of autoreactive T cells as cumulative autoimmunity (41). Importantly, this feature is not the prerogative of inbred experimental animals. Indeed, human polyspecific T cell clones showing recognition of more than one self-antigens have been described both in healthy individuals (67) and in patients with autoimmune diseases (38, 6870).

Significant work is still needed to determine the contribution of autoreactive T cells recognizing multiple self-antigens and of cumulative autoimmunity in the initiation and progression of self-tissue destruction. This issue is particularly difficult to address in humans. Experimental models may help delineate the potential importance of this phenomenon for the development of an autoimmune process.

Note added in proof.

When this manuscript was undergoing the review process, a very thorough study explored the structural bases of TCR cross-reactivity. Consistent with our data, the authors concluded that the different peptides recognized by a given TCR share TCR-binding motifs but may differ at residues outside the TCR interface (71).

We thank Drs. A. Saoudi, N. Fazilleau, and S. Guerder for input on the project and critical reading of the manuscript, the staff of the UMS006 animal facility for care of the mice, J. Sidney and P.-E. Paulet for valuable assistance, F. L’Faqihi and V. Duplan for running the core flow cytometry facility, and the the GeT-Purpan Genomics Facility (Genopole Toulouse, Midi-Pyrenees, France) for help in sequencing.

This work was supported by INSERM, the Centre National de la Recherche Scientifique, the Association pour la Recherche sur la Sclérose rn Plaques, the Medical Research Foundation (DEQ20090515409 to R.S.L.), and the Italian Federation for Multiple Sclerosis (2012/B/6 to L.E.L.).

The online version of this article contains supplemental material.

Abbreviations used in this article:

EAE

experimental autoimmune encephalomyelitis

MHC-II

MHC class II

MOG

myelin oligodendrocyte glycoprotein

mRCE

mRNA capping enzyme

NF-M

neurofilament medium

pMHC

MHC/peptide complex

Treg

regulatory T cell.

1
Sewell
A. K.
2012
.
Why must T cells be cross-reactive?
Nat. Rev. Immunol.
12
:
669
677
.
2
Wucherpfennig
K. W.
,
Allen
P. M.
,
Celada
F.
,
Cohen
I. R.
,
De Boer
R.
,
Garcia
K. C.
,
Goldstein
B.
,
Greenspan
R.
,
Hafler
D.
,
Hodgkin
P.
, et al
.
2007
.
Polyspecificity of T cell and B cell receptor recognition.
Semin. Immunol.
19
:
216
224
.
3
Ignatowicz
L.
,
Kappler
J.
,
Marrack
P.
.
1996
.
The repertoire of T cells shaped by a single MHC/peptide ligand.
Cell
84
:
521
529
.
4
Ignatowicz
L.
,
Rees
W.
,
Pacholczyk
R.
,
Ignatowicz
H.
,
Kushnir
E.
,
Kappler
J.
,
Marrack
P.
.
1997
.
T cells can be activated by peptides that are unrelated in sequence to their selecting peptide.
Immunity
7
:
179
186
.
5
Ernst
B.
,
Lee
D. S.
,
Chang
J. M.
,
Sprent
J.
,
Surh
C. D.
.
1999
.
The peptide ligands mediating positive selection in the thymus control T cell survival and homeostatic proliferation in the periphery.
Immunity
11
:
173
181
.
6
Stefanova
I.
,
Dorfman
J. R.
,
Tsukamoto
M.
,
Germain
R. N.
.
2003
.
On the role of self-recognition in T cell responses to foreign antigen.
Immunol. Rev.
191
:
97
106
.
7
Mason
D.
1998
.
A very high level of crossreactivity is an essential feature of the T-cell receptor.
Immunol. Today
19
:
395
404
.
8
Kersh
G. J.
,
Allen
P. M.
.
1996
.
Structural basis for T cell recognition of altered peptide ligands: a single T cell receptor can productively recognize a large continuum of related ligands.
J. Exp. Med.
184
:
1259
1268
.
9
Ishizuka
J.
,
Grebe
K.
,
Shenderov
E.
,
Peters
B.
,
Chen
Q.
,
Peng
Y.
,
Wang
L.
,
Dong
T.
,
Pasquetto
V.
,
Oseroff
C.
, et al
.
2009
.
Quantitating T cell cross-reactivity for unrelated peptide antigens.
J. Immunol.
183
:
4337
4345
.
10
Wooldridge
L.
,
Ekeruche-Makinde
J.
,
van den Berg
H. A.
,
Skowera
A.
,
Miles
J. J.
,
Tan
M. P.
,
Dolton
G.
,
Clement
M.
,
Llewellyn-Lacey
S.
,
Price
D. A.
, et al
.
2012
.
A single autoimmune T cell receptor recognizes more than a million different peptides.
J. Biol. Chem.
287
:
1168
1177
.
11
Casanova
J. L.
,
Romero
P.
,
Widmann
C.
,
Kourilsky
P.
,
Maryanski
J. L.
.
1991
.
T cell receptor genes in a series of class I major histocompatibility complex-restricted cytotoxic T lymphocyte clones specific for a Plasmodium berghei nonapeptide: implications for T cell allelic exclusion and antigen-specific repertoire.
J. Exp. Med.
174
:
1371
1383
.
12
Padovan
E.
,
Casorati
G.
,
Dellabona
P.
,
Meyer
S.
,
Brockhaus
M.
,
Lanzavecchia
A.
.
1993
.
Expression of two T cell receptor alpha chains: dual receptor T cells.
Science
262
:
422
424
.
13
He
X.
,
Janeway
C. A.
 Jr.
,
Levine
M.
,
Robinson
E.
,
Preston-Hurlburt
P.
,
Viret
C.
,
Bottomly
K.
.
2002
.
Dual receptor T cells extend the immune repertoire for foreign antigens.
Nat. Immunol.
3
:
127
134
.
14
Olivares-Villagómez
D.
,
Wang
Y.
,
Lafaille
J. J.
.
1998
.
Regulatory CD4(+) T cells expressing endogenous T cell receptor chains protect myelin basic protein-specific transgenic mice from spontaneous autoimmune encephalomyelitis.
J. Exp. Med.
188
:
1883
1894
.
15
Elliott
J. I.
,
Altmann
D. M.
.
1995
.
Dual T cell receptor alpha chain T cells in autoimmunity.
J. Exp. Med.
182
:
953
959
.
16
Zal
T.
,
Weiss
S.
,
Mellor
A.
,
Stockinger
B.
.
1996
.
Expression of a second receptor rescues self-specific T cells from thymic deletion and allows activation of autoreactive effector function.
Proc. Natl. Acad. Sci. USA
93
:
9102
9107
.
17
Sarukhan
A.
,
Garcia
C.
,
Lanoue
A.
,
von Boehmer
H.
.
1998
.
Allelic inclusion of T cell receptor alpha genes poses an autoimmune hazard due to low-level expression of autospecific receptors.
Immunity
8
:
563
570
.
18
Ji
Q.
,
Perchellet
A.
,
Goverman
J. M.
.
2010
.
Viral infection triggers central nervous system autoimmunity via activation of CD8+ T cells expressing dual TCRs.
Nat. Immunol.
11
:
628
634
.
19
Chen
J.
,
Eisen
H. N.
,
Kranz
D. M.
.
2003
.
A model T-cell receptor system for studying memory T-cell development.
Microbes Infect.
5
:
233
240
.
20
Maynard
J.
,
Petersson
K.
,
Wilson
D. H.
,
Adams
E. J.
,
Blondelle
S. E.
,
Boulanger
M. J.
,
Wilson
D. B.
,
Garcia
K. C.
.
2005
.
Structure of an autoimmune T cell receptor complexed with class II peptide-MHC: insights into MHC bias and antigen specificity.
Immunity
22
:
81
92
.
21
Takahashi
H.
,
Merli
S.
,
Putney
S. D.
,
Houghten
R.
,
Moss
B.
,
Germain
R. N.
,
Berzofsky
J. A.
.
1989
.
A single amino acid interchange yields reciprocal CTL specificities for HIV-1 gp160.
Science
246
:
118
121
.
22
Borbulevych
O. Y.
,
Piepenbrink
K. H.
,
Gloor
B. E.
,
Scott
D. R.
,
Sommese
R. F.
,
Cole
D. K.
,
Sewell
A. K.
,
Baker
B. M.
.
2009
.
T cell receptor cross-reactivity directed by antigen-dependent tuning of peptide-MHC molecular flexibility.
Immunity
31
:
885
896
.
23
Colf
L. A.
,
Bankovich
A. J.
,
Hanick
N. A.
,
Bowerman
N. A.
,
Jones
L. L.
,
Kranz
D. M.
,
Garcia
K. C.
.
2007
.
How a single T cell receptor recognizes both self and foreign MHC.
Cell
129
:
135
146
.
24
Scott
D. R.
,
Borbulevych
O. Y.
,
Piepenbrink
K. H.
,
Corcelli
S. A.
,
Baker
B. M.
.
2011
.
Disparate degrees of hypervariable loop flexibility control T-cell receptor cross-reactivity, specificity, and binding mechanism.
J. Mol. Biol.
414
:
385
400
.
25
Damian
R. T.
1965
.
Molecular mimicry in biological adaptation.
Science
147
:
824
.
26
Fujinami
R. S.
,
Oldstone
M. B.
.
1989
.
Molecular mimicry as a mechanism for virus-induced autoimmunity.
Immunol. Res.
8
:
3
15
.
27
Wucherpfennig
K. W.
,
Strominger
J. L.
.
1995
.
Molecular mimicry in T cell-mediated autoimmunity: viral peptides activate human T cell clones specific for myelin basic protein.
Cell
80
:
695
705
.
28
Zhao
Z. S.
,
Granucci
F.
,
Yeh
L.
,
Schaffer
P. A.
,
Cantor
H.
.
1998
.
Molecular mimicry by herpes simplex virus-type 1: autoimmune disease after viral infection.
Science
279
:
1344
1347
.
29
Gautam
A. M.
,
Liblau
R.
,
Chelvanayagam
G.
,
Steinman
L.
,
Boston
T.
.
1998
.
A viral peptide with limited homology to a self peptide can induce clinical signs of experimental autoimmune encephalomyelitis.
J. Immunol.
161
:
60
64
.
30
Albert
L. J.
,
Inman
R. D.
.
1999
.
Molecular mimicry and autoimmunity.
N. Engl. J. Med.
341
:
2068
2074
.
31
Mycko
M. P.
,
Waldner
H.
,
Anderson
D. E.
,
Bourcier
K. D.
,
Wucherpfennig
K. W.
,
Kuchroo
V. K.
,
Hafler
D. A.
.
2004
.
Cross-reactive TCR responses to self antigens presented by different MHC class II molecules.
J. Immunol.
173
:
1689
1698
.
32
Ercolini
A. M.
,
Miller
S. D.
.
2007
.
Molecular mimics can induce novel self peptide-reactive CD4+ T cell clonotypes in autoimmune disease.
J. Immunol.
179
:
6604
6612
.
33
Harkiolaki
M.
,
Holmes
S. L.
,
Svendsen
P.
,
Gregersen
J. W.
,
Jensen
L. T.
,
McMahon
R.
,
Friese
M. A.
,
van Boxel
G.
,
Etzensperger
R.
,
Tzartos
J. S.
, et al
.
2009
.
T cell-mediated autoimmune disease due to low-affinity crossreactivity to common microbial peptides.
Immunity
30
:
348
357
.
34
Wucherpfennig
K. W.
,
Sethi
D.
.
2011
.
T cell receptor recognition of self and foreign antigens in the induction of autoimmunity.
Semin. Immunol.
23
:
84
91
.
35
Berer
K.
,
Mues
M.
,
Koutrolos
M.
,
Rasbi
Z. A.
,
Boziki
M.
,
Johner
C.
,
Wekerle
H.
,
Krishnamoorthy
G.
.
2011
.
Commensal microbiota and myelin autoantigen cooperate to trigger autoimmune demyelination.
Nature
479
:
538
541
.
36
Varrin-Doyer
M.
,
Spencer
C. M.
,
Schulze-Topphoff
U.
,
Nelson
P. A.
,
Stroud
R. M.
,
Cree
B. A.
,
Zamvil
S. S.
.
2012
.
Aquaporin 4-specific T cells in neuromyelitis optica exhibit a Th17 bias and recognize Clostridium ABC transporter.
Ann Neurol.
72
:
53
64
.
37
Sethi
D. K.
,
Gordo
S.
,
Schubert
D. A.
,
Wucherpfennig
K. W.
.
2013
.
Crossreactivity of a human autoimmune TCR is dominated by a single TCR loop.
Nat. Commun.
4
:
2623
.
38
Quaratino
S.
,
Thorpe
C. J.
,
Travers
P. J.
,
Londei
M.
.
1995
.
Similar antigenic surfaces, rather than sequence homology, dictate T-cell epitope molecular mimicry.
Proc. Natl. Acad. Sci. USA
92
:
10398
10402
.
39
Gran
B.
,
Hemmer
B.
,
Vergelli
M.
,
McFarland
H. F.
,
Martin
R.
.
1999
.
Molecular mimicry and multiple sclerosis: degenerate T-cell recognition and the induction of autoimmunity.
Ann. Neurol.
45
:
559
567
.
40
Avery
A. C.
,
Zhao
Z. S.
,
Rodriguez
A.
,
Bikoff
E. K.
,
Soheilian
M.
,
Foster
C. S.
,
Cantor
H.
.
1995
.
Resistance to herpes stromal keratitis conferred by an IgG2a-derived peptide.
Nature
376
:
431
434
.
41
Liblau
R. S.
,
Wekerle
H.
,
Tisch
R. M.
.
2011
.
Cumulative autoimmunity: T cell clones recognizing several self-epitopes exhibit enhanced pathogenicity.
Front. Immunol.
2
:
47
.
42
Li
L.
,
Wang
B.
,
Frelinger
J. A.
,
Tisch
R.
.
2008
.
T-cell promiscuity in autoimmune diabetes.
Diabetes
57
:
2099
2106
.
43
Bettelli
E.
,
Pagany
M.
,
Weiner
H. L.
,
Linington
C.
,
Sobel
R. A.
,
Kuchroo
V. K.
.
2003
.
Myelin oligodendrocyte glycoprotein-specific T cell receptor transgenic mice develop spontaneous autoimmune optic neuritis.
J. Exp. Med.
197
:
1073
1081
.
44
Krishnamoorthy
G.
,
Saxena
A.
,
Mars
L. T.
,
Domingues
H. S.
,
Mentele
R.
,
Ben-Nun
A.
,
Lassmann
H.
,
Dornmair
K.
,
Kurschus
F. C.
,
Liblau
R. S.
,
Wekerle
H.
.
2009
.
Myelin-specific T cells also recognize neuronal autoantigen in a transgenic mouse model of multiple sclerosis.
Nat. Med.
15
:
626
632
.
45
Petersen
T. R.
,
Bettelli
E.
,
Sidney
J.
,
Sette
A.
,
Kuchroo
V.
,
Bäckström
B. T.
.
2004
.
Characterization of MHC- and TCR-binding residues of the myelin oligodendrocyte glycoprotein 38-51 peptide.
Eur. J. Immunol.
34
:
165
173
.
46
Ben-Nun
A.
,
Kerlero de Rosbo
N.
,
Kaushansky
N.
,
Eisenstein
M.
,
Cohen
L.
,
Kaye
J. F.
,
Mendel
I.
.
2006
.
Anatomy of T cell autoimmunity to myelin oligodendrocyte glycoprotein (MOG): prime role of MOG44F in selection and control of MOG-reactive T cells in H-2b mice.
Eur. J. Immunol.
36
:
478
493
.
47
Barnden
M. J.
,
Allison
J.
,
Heath
W. R.
,
Carbone
F. R.
.
1998
.
Defective TCR expression in transgenic mice constructed using cDNA-based alpha- and beta-chain genes under the control of heterologous regulatory elements.
Immunol. Cell Biol.
76
:
34
40
.
48
Mombaerts
P.
,
Clarke
A. R.
,
Rudnicki
M. A.
,
Iacomini
J.
,
Itohara
S.
,
Lafaille
J. J.
,
Wang
L.
,
Ichikawa
Y.
,
Jaenisch
R.
,
Hooper
M. L.
, et al
.
1992
.
Mutations in T-cell antigen receptor genes alpha and beta block thymocyte development at different stages.
Nature
360
:
225
231
.
49
Alli
R.
,
Nguyen
P.
,
Geiger
T. L.
.
2008
.
Retrogenic modeling of experimental allergic encephalomyelitis associates T cell frequency but not TCR functional affinity with pathogenicity.
J. Immunol.
181
:
136
145
.
50
Guéry
J. C.
,
Sette
A.
,
Appella
E.
,
Adorini
L.
.
1995
.
Constitutive presentation of dominant epitopes from endogenous naturally processed self-beta 2-microglobulin to class II-restricted T cells leads to self-tolerance.
J. Immunol.
154
:
545
554
.
51
Sidney
J.
,
Southwood
S.
,
Moore
C.
,
Oseroff
C.
,
Pinilla
C.
,
Grey
H. M.
,
Sette
A.
.
2013
.
Measurement of MHC/peptide interactions by gel filtration or monoclonal antibody capture.
Curr. Prot. Immunol.
DOI: 10.1002/0471142735.im1803s100.
52
Wilson
R. K.
,
Lai
E.
,
Concannon
P.
,
Barth
R. K.
,
Hood
L. E.
.
1988
.
Structure, organization and polymorphism of murine and human T-cell receptor alpha and beta chain gene families.
Immunol. Rev.
101
:
149
172
.
53
Liu
X.
,
Dai
S.
,
Crawford
F.
,
Fruge
R.
,
Marrack
P.
,
Kappler
J.
.
2002
.
Alternate interactions define the binding of peptides to the MHC molecule IA(b).
Proc. Natl. Acad. Sci. USA
99
:
8820
8825
.
54
Kaushansky
N.
,
Eisenstein
M.
,
Oved
J. H.
,
Ben-Nun
A.
.
2008
.
Activation and control of pathogenic T cells in OSP/claudin-11-induced EAE in SJL/J mice are dominated by their focused recognition of a single epitopic residue (OSP58M).
Int. Immunol.
20
:
1439
1449
.
55
de Rosbo
N. K.
,
Kaye
J. F.
,
Eisenstein
M.
,
Mendel
I.
,
Hoeftberger
R.
,
Lassmann
H.
,
Milo
R.
,
Ben-Nun
A.
.
2004
.
The myelin-associated oligodendrocytic basic protein region MOBP15-36 encompasses the immunodominant major encephalitogenic epitope(s) for SJL/J mice and predicted epitope(s) for multiple sclerosis-associated HLA-DRB1*1501.
J. Immunol.
173
:
1426
1435
.
56
Mostoslavsky
R.
,
Alt
F. W.
,
Rajewsky
K.
.
2004
.
The lingering enigma of the allelic exclusion mechanism.
Cell
118
:
539
544
.
57
Mendel
I.
,
Kerlero de Rosbo
N.
,
Ben-Nun
A.
.
1995
.
A myelin oligodendrocyte glycoprotein peptide induces typical chronic experimental autoimmune encephalomyelitis in H-2b mice: fine specificity and T cell receptor V beta expression of encephalitogenic T cells.
Eur. J. Immunol.
25
:
1951
1959
.
58
Ben-Nun
A.
,
Mendel
I.
,
Bakimer
R.
,
Fridkis-Hareli
M.
,
Teitelbaum
D.
,
Arnon
R.
,
Sela
M.
,
Kerlero de Rosbo
N.
.
1996
.
The autoimmune reactivity to myelin oligodendrocyte glycoprotein (MOG) in multiple sclerosis is potentially pathogenic: effect of copolymer 1 on MOG-induced disease.
J. Neurol.
243
(
Suppl 1
):
S14
S22
.
59
Sette
A.
,
Grey
H.
,
Oseroff
C.
,
Peters
B.
,
Moutaftsi
M.
,
Crotty
S.
,
Assarsson
E.
,
Greenbaum
J.
,
Kim
Y.
,
Kolla
R.
, et al
.
2009
.
Definition of epitopes and antigens recognized by vaccinia specific immune responses: their conservation in variola virus sequences, and use as a model system to study complex pathogens.
Vaccine
27
(
Suppl 6
):
G21
G26
.
60
Fazilleau
N.
,
Delarasse
C.
,
Motta
I.
,
Fillatreau
S.
,
Gougeon
M. L.
,
Kourilsky
P.
,
Pham-Dinh
D.
,
Kanellopoulos
J. M.
.
2007
.
T cell repertoire diversity is required for relapses in myelin oligodendrocyte glycoprotein-induced experimental autoimmune encephalomyelitis.
J. Immunol.
178
:
4865
4875
.
61
Fazilleau
N.
,
Delarasse
C.
,
Sweenie
C. H.
,
Anderton
S. M.
,
Fillatreau
S.
,
Lemonnier
F. A.
,
Pham-Dinh
D.
,
Kanellopoulos
J. M.
.
2006
.
Persistence of autoreactive myelin oligodendrocyte glycoprotein (MOG)-specific T cell repertoires in MOG-expressing mice.
Eur. J. Immunol.
36
:
533
543
.
62
Kuchroo
V. K.
,
Anderson
A. C.
,
Waldner
H.
,
Munder
M.
,
Bettelli
E.
,
Nicholson
L. B.
.
2002
.
T cell response in experimental autoimmune encephalomyelitis (EAE): role of self and cross-reactive antigens in shaping, tuning, and regulating the autopathogenic T cell repertoire.
Annu. Rev. Immunol.
20
:
101
123
.
63
Sospedra
M.
,
Muraro
P. A.
,
Stefanová
I.
,
Zhao
Y.
,
Chung
K.
,
Li
Y.
,
Giulianotti
M.
,
Simon
R.
,
Mariuzza
R.
,
Pinilla
C.
,
Martin
R.
.
2006
.
Redundancy in antigen-presenting function of the HLA-DR and -DQ molecules in the multiple sclerosis-associated HLA-DR2 haplotype.
J. Immunol.
176
:
1951
1961
.
64
Stromnes
I. M.
,
Cerretti
L. M.
,
Liggitt
D.
,
Harris
R. A.
,
Goverman
J. M.
.
2008
.
Differential regulation of central nervous system autoimmunity by T(H)1 and T(H)17 cells.
Nat. Med.
14
:
337
342
.
65
Delarasse
C.
,
Daubas
P.
,
Mars
L. T.
,
Vizler
C.
,
Litzenburger
T.
,
Iglesias
A.
,
Bauer
J.
,
Della Gaspera
B.
,
Schubart
A.
,
Decker
L.
, et al
.
2003
.
Myelin/oligodendrocyte glycoprotein-deficient (MOG-deficient) mice reveal lack of immune tolerance to MOG in wild-type mice.
J. Clin. Invest.
112
:
544
553
.
66
Gold
R.
,
Linington
C.
,
Lassmann
H.
.
2006
.
Understanding pathogenesis and therapy of multiple sclerosis via animal models: 70 years of merits and culprits in experimental autoimmune encephalomyelitis research.
Brain
129
:
1953
1971
.
67
Cai
G.
,
Hafler
D. A.
.
2007
.
Multispecific responses by T cells expanded by endogenous self-peptide/MHC complexes.
Eur. J. Immunol.
37
:
602
612
.
68
Hemmer
B.
,
Fleckenstein
B. T.
,
Vergelli
M.
,
Jung
G.
,
McFarland
H.
,
Martin
R.
,
Wiesmüller
K. H.
.
1997
.
Identification of high potency microbial and self ligands for a human autoreactive class II-restricted T cell clone.
J. Exp. Med.
185
:
1651
1659
.
69
Hemmer
B.
,
Kondo
T.
,
Gran
B.
,
Pinilla
C.
,
Cortese
I.
,
Pascal
J.
,
Tzou
A.
,
McFarland
H. F.
,
Houghten
R.
,
Martin
R.
.
2000
.
Minimal peptide length requirements for CD4(+) T cell clones—implications for molecular mimicry and T cell survival.
Int. Immunol.
12
:
375
383
.
70
Sospedra
M.
,
Zhao
Y.
,
zur Hausen
H.
,
Muraro
P. A.
,
Hamashin
C.
,
de Villiers
E. M.
,
Pinilla
C.
,
Martin
R.
.
2005
.
Recognition of conserved amino acid motifs of common viruses and its role in autoimmunity.
PLoS Pathog.
1
:
e41
.
71
Birnbaum
M. E.
,
Mendoza
J. L.
,
Sethi
D. K.
,
Dong
S.
,
Glanville
J.
,
Dobbins
J.
,
Ozkan
E.
,
Davis
M. M.
,
Wucherpfennig
K. W.
,
Garcia
K. C.
.
2014
.
Deconstructing the peptide-MHC specificity of T cell recognition.
Cell
157
:
1073
87
.

The authors have no financial conflicts of interest.

Supplementary data