Of interest to the etiology of demyelinating autoimmune disease is the potential to aberrantly activate CD4+ T cells due to cross-recognition of multiple self-epitopes such as has been suggested for myelin oligodendrocyte glycoprotein epitope 35–55 (MOG35–55) and neurofilament medium protein epitope 15–35 (NFM15–35). NFM15–35 is immunogenic in C57BL/6 mice but fails to induce demyelinating disease by polyclonal T cells despite having the same TCR contact residues as MOG35–55, a known encephalitogenic Ag. Despite reported cross-reactivity with MOG-specific T cells, the polyclonal response to NFM15–35 did not expand threshold numbers of MOG38–49 tetramer–positive T cells. Furthermore, NFM lacked functional synergy with MOG to promote experimental autoimmune encephalomyelitis because NFM-deficient synonymous with knockout mice developed an identical disease course to wild-type mice after challenge with MOG35–55. Single-cell analysis of encephalitogenic T cells using the peptide:MHC monomer-based two-dimensional micropipette adhesion frequency assay confirmed that NFM was not a critical Ag driving demyelinating disease because NFM18–30–specific T cells in the CNS were predominantly reactive to MOG38–49. The absence of NFM contribution to disease allowed mapping of the amino acids required for encephalitogenicity and expansion of high-affinity, MOG-specific T cells that defined the polyclonal response. Alterations of N-terminal residues outside of the NFM15–35 core nonamer promoted expansion of high-affinity, MOG38–49 tetramer–positive T cells and promoted consistent experimental autoimmune encephalomyelitis induction, unlike mice challenged with NFM15–35. Although NFM15–35 is immunogenic and cross-reactive with MOG at the polyclonal level, it fails to expand a threshold level of encephalitogenic, high-affinity MOG-specific T cells.

Myelin-specific T cells exist in all individuals, but it is not clear how they initially become triggered to attack self and promote autoimmune disease (1). Multiple factors influence the autoimmune T cell response, which can be shaped by positive and negative selection pressures in the thymus and periphery (2, 3), genetic predispositions (4), environmental exposures (5), ability to migrate to the CNS, and differentiation into effector and memory subsets (6). Central to these factors is how the TCR sees self-peptides presented on MHC, with MHC being the strongest genetic susceptibility factor currently identified for multiple sclerosis (79). One hypothesis for activation of autoreactive T cells is that exposure to structurally related or cross-reactive peptides derived from self or foreign origins can break tolerance (1013). CD4+ T cell cross-recognition between myelin oligodendrocyte glycoprotein epitope 35–55 (MOG35–55) and neurofilament medium protein epitope 15–35 (NFM15–35) in demyelinating autoimmunity is of interest for disease etiology because these two self-epitopes have synergist potential as targets of autoimmune T cell attack due to identical amino acids at proposed TCR contacts (14). Conceptually, T cell recognition and responsiveness to multiple peptides is a critical feature of protective immune surveillance where a limited TCR repertoire is presented with a myriad of peptides displayed on MHC (1517). In one example, a single T cell clone specific for myelin basic protein (Ac1-11) has the potential to recognize 106 peptides generated from a combinatorial library (18).

Physical interactions between TCR and peptide:MHC (pMHC) provide another level of input for understanding the initiation of TCR signaling. T cell cross-reactivity is unique to the amino acid structure of an individual TCR, dictating how it physically recognizes peptides oriented within MHC (19). Alterations in peptides at amino acid residues interfacing with the TCR, known as altered peptide ligands, can change the functional outcome of T cell responses (12, 18, 2022) by changing affinity of the TCR:pMHC interaction as well as binding kinetics, including on/off rates and bond lifetime (23, 24). Associations between affinity and function of T cells contributing to polyclonal, demyelinating autoimmune disease have identified a breadth of high and low affinity interactions between TCR and pMHC in C57BL/6 and NOD models (2527). Our goal is to understand how cross-recognition, cross-reactivity, and biophysical interactions, defined by two-dimensional (2D) affinity, influence onset of demyelinating autoimmunity in polyclonal models.

The study of polyclonal T cell cross-reactivity to MOG35–55 and NFM15–35 is a relevant platform to study the etiology of demyelinating autoimmune disease. Currently, polyclonal studies examining T cell specificity to MOG and NFM do not paint a clear model for cross-reactivity and disease because NFM involvement is based on functional responses after MOG35–55 and not NFM15–35 priming (28). Although this experimental approach stems from earlier reports stating NFM15–35 does not induce experimental autoimmune encephalomyelitis (EAE) (14), we questioned why this would be the case and chose to comparatively track Ag-specific T cells after a MOG35–55 or NFM15–35 challenge.

We demonstrate that the lack of EAE onset after NFM15–35 challenge correlated with insufficient expansion of high-affinity, MOG38–49 tetramer–positive T cells in spleen and peripheral lymph nodes along with their absence in the CNS. 2D micropipette adhesion frequency assay was used as a monomer-based tool to characterize TCR affinity and enhance detection of Ag-specific T cells above the limits of tetramer reagents (25). After MOG35–55 challenge we found that the CNS T cell infiltrates recognizing NFM18–30 largely cross-recognized MOG38–49 but the NFM-specific T cells generally lacked the higher affinity cell populations typically seen with the MOG-specific infiltrates. The functional requirement of MOG and NFM bispecific T cells to EAE onset and severity was tested with MOG35–55 challenge of NFM-deficient (NFM−/−) mice. This design revealed MOG and not NFM to be the critical autoantigen for EAE because NFM−/− mice developed a similar disease course to wild-type (WT), NFM-sufficient mice. Furthermore, we mapped N-terminal residues of NFM15–35 that dictated poor encephalitogenicity such that, when modified, permitted EAE induction as well as detection of MOG38–49 tetramer–positive T cells. The presented data concomitant with the fact that MOG−/− mice do not develop EAE after MOG35–55 or NFM15–35 challenge (29) supported the conclusion that NFM has a minimal role in the onset of EAE and the expansion of encephalitogenic T cells from the polyclonal repertoire where pathogenicity is dictated by expansion of a critical threshold level of higher affinity tetramer-positive, MOG-specific T cells.

Female C57BL/6 mice were purchased from Charles River Frederick Facility, formerly the National Cancer Institute. With permission from Dr. H. Reid at Monash University, MOG−/− mice (29) were obtained from Dr. X.-F. Bai at The Ohio State University. NFM−/− mice were obtained from Dr. Julien and colleagues (30) and backcrossed to National Cancer Institute C57BL/6N mice due to visual dominance of the 129 background coat color and the presence of an additional nefh knockout. Completion of the backcross was confirmed by DartMouse (Lebanon, NH) and NFM protein deficiency was determined by histology performed at Emory National Institute of Neurological Disorders and Stroke Neuropathology/Histochemistry Core Facility (Atlanta, GA). IFN-γR−/− (B6.129S7-Ifngr1tm1Agt/J) mice, Thy1.1 mice (B6.PL-Thy1a/CyJ), and 2D2 transgenic mice (C57BL/6-Tg(Tcra2D2,Tcrb2D2)1Kuch/J) were initially purchased from the Jackson Laboratory. These mice along with SMARTA [Tg(TcrLCMV)Aox] mice (31) were all housed in the Division of Animal Resources at Emory University and handled in accordance with protocols approved by the Institutional Animal Care and Use Committee. Experimental mice were between 6 and 11 wk of age.

Peptides used for T cell priming and EAE induction were generated in our laboratory with the Prelude peptide synthesizer (Protein Technologies) with the following sequences: MOG35–55 (MEVGWYRSPFSRVVHLYRNGK), NFM15–35 (RRVTETRSSFSRVSGSPSSGF), NFM15–35 T20Y (RRVTEYRSSFSRVSGSPSSGF), NFM15–35 E19W, T20Y (RRVTWYRSSFSRVSGSPSSGF), NFM15–35 S23P (RRVTETRSPFSRVSGSPSSGF), NFM15–35 S28V (RRVTETRSSFSRVVGSPSSGF).

Polyclonal T cell lines were generated by footpad priming mice with emulsions of MOG35–55 or NFM15–35 (1 mg/ml) in CFA containing a final concentration of 0.5 mg/ml heat-killed Mycobacterium tuberculosis (H37 RA) in IFA (BD). Draining lymph nodes were harvested 12–14 d after priming. Single-cell suspensions were generated by passing cells through a 100 μm cell strainer and directly assessed for Ag-specific proliferation by uptake of [3H] tritiated thymidine (VWR International). Next, 6 × 105 primed lymph node cells or naive 2D2 splenocytes were cultured in a 96-well flat-bottom plate with indicated peptides and concentrations. After 48 h, 3H-thymidine (0.4 μCi per well) was added to the culture media for 24 h before the cells were harvested onto a filtermat (PerkinElmer) with the FilterMate 196 harvester (Packard). The uptake of 3H-thymidine was analyzed with the 1450 Microbeta TriLux microplate liquid scintillation counter (PerkinElmer). Cell culture media contained RPMI 1640 (Corning) supplemented with 10% heat-inactivated FBS (Life Technologies), 4 mM l-glutamine (Corning), 0.01 M HEPES (Corning), 100 μg/ml gentamicin (Corning), and 20 μM 2-ME (Sigma-Aldrich). Stimulation index was calculated as a ratio of the cpm between peptide-stimulated versus unstimulated cells.

EAE was induced with two s.c. injections of emulsion delivering 200 μg designated peptide and 0.4 mg M. tuberculosis in IFA per injection on days 0 and 7. Then 250 ng of pertussis toxin (List Biological) was concomitantly administered i.p. to mice on days 0 and 2. Mice were monitored for signs of disease including weight loss and paralytic symptoms. Mice presented with typical ascending paralysis were scored accordingly: (0.5) partial tail paralysis, (1.0) complete tail paralysis, (2.0) hindlimb weakness, (2.5) ataxia, (3.0) partial hindlimb paralysis, (3.5) complete hindlimb paralysis, (4.0) inability to right itself, and (5.0) moribund.

Passive EAE was generated by adoptive transfer of 1 × 107 cells from an NFM T cell line per C57BL/6 mouse, irradiated with 400 rad 1 d before transfer. The T cell line was generated by inducing either WT C57BL/6 mice (naturally Thy1.2) or Thy1.1 congenic C57BL/6 mice with NFM15–35 in CFA on days 0 and 7 with pertussis toxin administered on days 0 and 2 as described above. Spleens were harvested 14 d later, mashed into single-cell suspensions, and cultured as previously described but with the addition of 50 ng/ml of IL-2 and 5 ng/ml recombinant mouse IL-12 (Gemini Bio). Splenocytes were cultured for 3 d and with 10 × 106 blast cells transferred i.p. Mice were then monitored for signs of EAE.

Tetramer enrichment was performed as previously described (32). In brief, spleen and peripheral lymph nodes (inguinal, lumbar, mesenteric, cervical, axillary, and brachial) were harvested and processed into a single-cell suspension by passing through a 100 μm strainer. Cells were washed in cold 1× HBSS (Corning) and resuspended with 4 μg/ml of tetramer PE conjugated NFM:I-Ab and/or APC conjugated MOG:I-Ab [National Institutes of Health (NIH) tetramer core (33, 34)] in Fc block (heat-killed mouse and rat serum, Sigma-Aldrich) at a volume 2× the pellet volume. After 1 h at room temperature, cells were washed in cold FACS wash (0.1% BSA, 0.05% NaN3, 1× PBS), resuspended in L FACS wash plus 50 μl of anti-PE and/or anti-APC beads (Miltenyi Biotec), and incubated for 30 min on ice. Cells were then washed and enriched on an LS column (Miltenyi Biotec). Unbound and column-bound cell numbers were determined with AccuCheck microbeads (Invitrogen) alongside cell surface marker characterization performed with flow cytometry (LSR II; BD) and FlowJo software (Tree Star). Abs used included CD3ε-FITC (145-2C11; BD Pharmingen), CD11b-PerCP-Cy5.5 (M1/70; BD Pharmingen), CD11c-PerCP-Cy5.5 (HL3; BD Pharmingen), CD19-PerCP-Cy5.5 (1D3; Tonbo Biosciences), CD4–Brilliant Violet 510 (RM4-5; BioLegend), CD8–Brilliant Violet 785 (53-6.7; BioLegend), CD44–Alexa Fluor 700 (IM7; eBioscience).

Mice were euthanized by CO2-induced asphyxiation and perfused with 1× Dulbecco’s PBS (Corning) through the left ventricle, with drainage mediated by laceration of the inferior vena cava. From individual mice, single-cell suspensions of the CNS, brain, and spinal cord were generated by passing the tissue through a 100 μm strainer. Mononuclear cells were isolated by Percoll density centrifugation (Sigma-Aldrich). CD4+ T cells were isolated from pooled CNS mononuclear cells using MACS with L3T4 CD4-positive selecting beads (Miltenyi Biotec). Enriched T cells were assessed for Ag specificity and effective 2D affinities with the 2D micropipette adhesion frequency assay using pMHC coated RBC sensors (25, 27, 3537).

Human RBCs were isolated from healthy volunteers in accordance with the Institutional Review Board at Emory University. RBCs were prepared by first coating the cells with varying concentrations of Biotin-X-NHS (EMD Millipore) followed by 0.5 mg/ml streptavidin (Thermo Fisher Scientific) and then 1 μg of biotinylated monomers MOG38–49:I-Ab, NFM18–30:I-Ab, GP66–77:I-Ab (NIH Tetramer Core Facility). Quantitation of TCR and pMHC densities were determined by flow cytometry (LSR II; BD) after labeling with PE-conjugated TCRβ (H57-597; eBioscience) or I-A/I-E (M5/114.15.2; BD), and quantitated with PE-Quantibrite bead standards (BD).

Adhesion frequencies between T cells and pMHC RBC sensors were determined if binding, visualized by RBC membrane distension, was observed after individual T cells were brought into contact 25–50 times with movement controlled by a piezo actuator (38). Background or nonspecific adhesion frequencies are considered to be below 0.1 in this Ag-specific system (25). Adhesion frequencies along with molecule surface densities were used to calculate effective 2D affinities (AcKa) using AcKa= - ln[1-Pa(∞)}/mrml with mr and ml representing TCR and pMHC surface densities, respectively.

Testing individual T cells among multiple pMHC is done by sequential binding. First, RBC sensors with different pMHC are concentrated at opposing locations within the microscope chamber containing purified T cells. Next, one T cell is aspirated into the micropipette attached to the piezo actuator and the MOG38–49:I-Ab RBC sensor is aspirated into the second micropipette. After 25–50 touches, the MOG RBC sensor is expelled and then the NFM18–30:I-Ab RBC sensor can be aspirated and tested against the same aspirated T cell. The next individual T cell will be tested in reverse order from the previous assessment, i.e., the NFM sensor first followed by the MOG sensor to assess if the sequential order influences subsequent Ag binding.

Statistical analyses were performed with Prism version 6 (GraphPad Software). Disease onset was assessed by unpaired, nonparametric t tests with Mann–Whitney post hoc tests. Two-tailed unpaired parametric t tests with F test variances were used for analyses between two groups whereas comparison of multiple means was performed with one-way ANOVA. Paired t tests were run where single-cell detection of MOG versus NFM specificity was directly compared.

The experiments highlighted in this study are aimed at clarifying the encephalitogenic potential of NFM based on the work initially describing T cell cross-reactivity between MOG35–55 and NFM15–35 in polyclonal MOG and NFM cell lines (14). We also found both MOG35–55 and NFM15–35 to be immunogens based on the proliferation of polyclonal lymph node cells directly ex vivo, using a range of peptide doses to determine sensitivity of the T cell response (Fig. 1A). Restimulation of lymph node cells with the cognate priming Ag revealed that higher peptide concentrations (10–100 μM) were needed to see proliferation above background. Interestingly, cross-reactive proliferation to the noncognate Ag was poorly detected, if at all, despite six of the nine putative core epitopes being identical between NFM20–28 and MOG40–48. This indicated that T cell expansion via the cross-reactive, noncognate epitope was not significant on the polyclonal level, questioning the relevance of cross-reactive expansion to disease. C57BL/6 mice actively challenged with NFM15–35 using a conventional induction regimen lacked the weight loss and encephalitogenic potential exhibited by MOG35–55 (Fig. 1B) (14, 28).

FIGURE 1.

NFM is immunogenic but not encephalitogenic. (A) Lymph node cells were harvested 12–14 d after priming C57BL/6 mice with MOG35–55 or NFM15–35 in CFA. Cells were directly assessed for Ag-specific proliferation and 3H-thymidine incorporation. Cpm were assessed and reported as stimulation index to average multiple experiments. Data are averaged from at least three experiments and five or more replicates per condition. (B) EAE was induced in C57BL/6 mice using MOG35–55 (n = 7) or NFM15–35 (n = 15), with mice being monitored for paralytic severity and weight loss. Data are representative of two experiments.

FIGURE 1.

NFM is immunogenic but not encephalitogenic. (A) Lymph node cells were harvested 12–14 d after priming C57BL/6 mice with MOG35–55 or NFM15–35 in CFA. Cells were directly assessed for Ag-specific proliferation and 3H-thymidine incorporation. Cpm were assessed and reported as stimulation index to average multiple experiments. Data are averaged from at least three experiments and five or more replicates per condition. (B) EAE was induced in C57BL/6 mice using MOG35–55 (n = 7) or NFM15–35 (n = 15), with mice being monitored for paralytic severity and weight loss. Data are representative of two experiments.

Close modal

To more clearly tease apart the contribution of NFM15–35 to the polyclonal response, we used NFM−/− mice (30). Agouti mice double-deficient in NFM and NFH proteins were generously obtained from Dr. Julien and subsequently backcrossed in our laboratory with our progeny confirmed as NFM−/−, NFH+/+, and 99% C57BL/6N by speed congenics (DartMouse, Fig. 2A), PCR (Fig. 2B), and histology (Fig. 2C). NFM-deficient mice were used to eliminate potential thymic and peripheral selection pressures due to the expression of NFM by medullary and cortical thymic epithelial cells (39, 40). NFM15–35–primed lymph node cells from WT or NFM-deficient mice gave a similar ex vivo proliferative response (Fig. 2D). This showed that NFM15–35 was equally immunogenic regardless of NFM expression, which was the opposite phenotype observed with MOG−/− mice (29). The course of MOG-induced EAE was then characterized in the NFM-deficient mice to eliminate potential T cell cross-reactivity with NFM. We hypothesized that if T cell cross-reactivity to NFM contributed to EAE onset or severity then a diminished disease course would be seen in the absence of NFM. NFM−/− mice, however, developed EAE similarly to WT NFM-sufficient mice with no difference in day of symptom onset and overall disease course (Fig. 2E). Furthermore, NFM15–35 challenge was unable to induce EAE in MOG−/− mice, clearly indicating that NFM alone is not sufficient to cause EAE (Fig. 2F).

FIGURE 2.

Characterization of NFM-deficient mice suggests MOG is the critical autoantigen for induced EAE. (A) Mice were backcrossed to C57BL/6 and confirmed by DartMouse to be 99% C57BL/6N (n = 3) with analyses of one mouse shown. Single nucleotide polymorphisms from C57BL/6N are shown in green and variances are highlighted in yellow (129× 1/Sv) or pink (heterozygous for C57BL/6N and 129). (B) WT control mice or progeny of the NFM−/− backcross to C57BL/6 were individually PCR tested for the presence (Neo Primers) or absence (WT Primers) of the neomycin cassette disrupting exon 1 of Nefm; shown are two representative gels. (C) Representative light microscopy images (original magnification ×10 or ×20) of the pyramidal cell layer in the hippocampus of WT C57BL/6 miceversus NFM−/− mice. NFM expression was evaluated by immunostaining with NFM-specific Ab NN18 (Millipore MAB5254) and compared with control slides where the primary Ab was withheld. NFM (dark brown) was visualized in cell bodies and fiber tracts of WT mice (n = 1) but not NFM−/− mice (n = 2). Sections were counterstained with hematoxylin (blue/purple) to highlight nuclei. (D) Lymph node cells were harvested 12–14 d post-NFM15–35/CFA priming of C57BL/6 or NFM−/− mice. Cpm were assessed 24 h after the addition of 3H-thymidine. Data are the average of two replicates per condition. (E) MOG35–55 EAE was induced in WT and NFM−/− mice, with disease course and weight loss subsequently monitored. The data in this study represent one of three experiments, with a total of n = 10 WT miceand n = 28 NFM−/− mice. There was no significant difference in day of symptom onset, p = 0.093, using an unpaired, nonparametric t test with Mann–Whitney post hoc test. (F) MOG−/− mice were challenged with an NFM15–35 EAE induction to assess ability of T cells to recognize NFM alone and cause EAE, n = 19. WT mice were challenged with MOG35–55 as a positive control n = 5. Data shown are representative of two experiments.

FIGURE 2.

Characterization of NFM-deficient mice suggests MOG is the critical autoantigen for induced EAE. (A) Mice were backcrossed to C57BL/6 and confirmed by DartMouse to be 99% C57BL/6N (n = 3) with analyses of one mouse shown. Single nucleotide polymorphisms from C57BL/6N are shown in green and variances are highlighted in yellow (129× 1/Sv) or pink (heterozygous for C57BL/6N and 129). (B) WT control mice or progeny of the NFM−/− backcross to C57BL/6 were individually PCR tested for the presence (Neo Primers) or absence (WT Primers) of the neomycin cassette disrupting exon 1 of Nefm; shown are two representative gels. (C) Representative light microscopy images (original magnification ×10 or ×20) of the pyramidal cell layer in the hippocampus of WT C57BL/6 miceversus NFM−/− mice. NFM expression was evaluated by immunostaining with NFM-specific Ab NN18 (Millipore MAB5254) and compared with control slides where the primary Ab was withheld. NFM (dark brown) was visualized in cell bodies and fiber tracts of WT mice (n = 1) but not NFM−/− mice (n = 2). Sections were counterstained with hematoxylin (blue/purple) to highlight nuclei. (D) Lymph node cells were harvested 12–14 d post-NFM15–35/CFA priming of C57BL/6 or NFM−/− mice. Cpm were assessed 24 h after the addition of 3H-thymidine. Data are the average of two replicates per condition. (E) MOG35–55 EAE was induced in WT and NFM−/− mice, with disease course and weight loss subsequently monitored. The data in this study represent one of three experiments, with a total of n = 10 WT miceand n = 28 NFM−/− mice. There was no significant difference in day of symptom onset, p = 0.093, using an unpaired, nonparametric t test with Mann–Whitney post hoc test. (F) MOG−/− mice were challenged with an NFM15–35 EAE induction to assess ability of T cells to recognize NFM alone and cause EAE, n = 19. WT mice were challenged with MOG35–55 as a positive control n = 5. Data shown are representative of two experiments.

Close modal

The absence of NFM-induced EAE was similar to our previously published reports with the MOG 45D variant peptide, which like NFM shares all primary TCR contact residues with WT MOG but differs at MHC anchor residues (33, 41). Interestingly, MOG 45D could induce EAE and expand MOG38–49–specific, tetramer-positive T cells when IFN-γ signaling was deficient in mice (33). We therefore tested whether NFM15–35 challenge of IFN-γR−/− would induce EAE and found NFM15–35 was still not encephalitogenic (Fig. 3A). Because the MOG 45D data suggested a threshold level of MOG tetramer–positive T cells was needed for onset of polyclonal EAE, we monitored Ag-specific T cell expansion in C57BL/6 mice after an MOG35–55 or NFM15–35 challenge to assess why NFM15–35 does not actively induce EAE.

FIGURE 3.

MOG-specific T cell expansion is poor after NFM challenge. (A) EAE was induced in IFN-γ receptor KO mice after MOG35–55 (n = 6) or NFM15–35 (n = 14) challenge. Disease course is representative of two experiments. (B) WT mice were induced with an EAE challenge of MOG35–55 or NFM15–35 in CFA plus pertussis toxin. MOG38–49 or NFM18–30 tetramer pulldowns were performed on days 14 (n = 6–7 mice per group) or 22 (n = 8–15 mice per group) postchallenge. MOG38–49 detection of CD4+ CD44hi T cells differed significantly between MOG35–55 and NFM15–35 induction 14 or 22 d postinjection. CD4+ T cells were identified with an initial lymphocyte gate (forward light scatter [FSC]-A/side scatter [SSC]-A) followed by a singlet gate (SSC-W/SSC-H), CD3+ CD11b CD11c CD19 gate, and CD4+ versus CD8+ gate. (C) Day 22 T cell activation status was evaluated by observing the number of activated CD44hi versus unactivated CD44low MOG-specific T cell numbers. NFM15–35 induction did not significantly induce the numbers of CD44hi MOG-specific T cells (p < 0.0001) expanded by MOG35–55 induction. (D) WT or NFM−/− mice were challenged with NFM15–35 in CFA, as in (A). Then 22 d later, no significant differences were found in expansion of MOG38–49 T cells between WT and deficient mice (n = 10 per strain, p = 0.41). (E) No significant differences in NFM18–30 detection were seen between MOG35–55 versus NFM15–35 induction groups at day 14 (p = 0.09) or day 22 (p = 0.25). Representative flow plots from the bound fraction of the tetramer enrichment at day 14 postchallenge are shown (F). (G) MOG38–49 tetramer was used to identify Ag-specific T cells in the CNS of five to six mice per group on day 22 after induction. There was no significant difference seen in the CD3+ CD4+ CD44+ population, p = 0.06, despite significant differences in enumerating MOG and NFM tetramer–positive cells. All statistical analyses were done using two-tailed unpaired, parametric t tests assuming both populations had equal SDs.

FIGURE 3.

MOG-specific T cell expansion is poor after NFM challenge. (A) EAE was induced in IFN-γ receptor KO mice after MOG35–55 (n = 6) or NFM15–35 (n = 14) challenge. Disease course is representative of two experiments. (B) WT mice were induced with an EAE challenge of MOG35–55 or NFM15–35 in CFA plus pertussis toxin. MOG38–49 or NFM18–30 tetramer pulldowns were performed on days 14 (n = 6–7 mice per group) or 22 (n = 8–15 mice per group) postchallenge. MOG38–49 detection of CD4+ CD44hi T cells differed significantly between MOG35–55 and NFM15–35 induction 14 or 22 d postinjection. CD4+ T cells were identified with an initial lymphocyte gate (forward light scatter [FSC]-A/side scatter [SSC]-A) followed by a singlet gate (SSC-W/SSC-H), CD3+ CD11b CD11c CD19 gate, and CD4+ versus CD8+ gate. (C) Day 22 T cell activation status was evaluated by observing the number of activated CD44hi versus unactivated CD44low MOG-specific T cell numbers. NFM15–35 induction did not significantly induce the numbers of CD44hi MOG-specific T cells (p < 0.0001) expanded by MOG35–55 induction. (D) WT or NFM−/− mice were challenged with NFM15–35 in CFA, as in (A). Then 22 d later, no significant differences were found in expansion of MOG38–49 T cells between WT and deficient mice (n = 10 per strain, p = 0.41). (E) No significant differences in NFM18–30 detection were seen between MOG35–55 versus NFM15–35 induction groups at day 14 (p = 0.09) or day 22 (p = 0.25). Representative flow plots from the bound fraction of the tetramer enrichment at day 14 postchallenge are shown (F). (G) MOG38–49 tetramer was used to identify Ag-specific T cells in the CNS of five to six mice per group on day 22 after induction. There was no significant difference seen in the CD3+ CD4+ CD44+ population, p = 0.06, despite significant differences in enumerating MOG and NFM tetramer–positive cells. All statistical analyses were done using two-tailed unpaired, parametric t tests assuming both populations had equal SDs.

Close modal

Cell numbers of tetramer-positive T cells were examined using the tetramer pulldown technique to enhance detection and quantitation of T cells within the polyclonal repertoire (32) (Fig. 3). Days 14 and 22 postchallenge with MOG35–55 or NFM15–35 were assessed as time points typical of onset and peak paralytic symptoms in our laboratory (33, 4143). At day 14 after MOG35–55 challenge, the MOG-specific T cell population expanded to 16,000 cells ± 2200 (mean ± SEM) with only a subtle decline in numbers to 9200 cells ± 1800 by day 22 (p = 0.02, Fig. 3B). NFM15–35 challenge maximally expanded cross-reactive MOG38–49–specific T cells to a number five times lower than MOG35–55 challenge at day 14, respectively 3200 cells ± 770, which further diminished 34% by day 22 to 1100 cells ± 410, p = 0.018. Breakdown of the day 22 MOG38–49 tetramer–positive populations into CD4+ CD44hi versus CD4+ CD44low indicated how well T cells were being activated by the Ag-priming regimen (Fig. 3C). MOG35–55 challenge significantly expanded a greater number of CD44hi (9200 cells ± 1800) than CD44low cells (1360 cells ± 290), whereas NFM priming did not, with 1100 cells ± 410 versus 4300 cells ± 2000 respectively.

We considered that endogenous NFM expression could potentially regulate the expansion of MOG38–49 tetramer–positive T cells after NFM15–35 challenge. We used the tetramer pulldown method to enumerate any changes in T cell expansion between NFM-deficient and WT mice 22 d postchallenge. We found a similar expansion (p = 0.41) of MOG38–49–specific T cells after NFM15–35 challenge between WT C57BL/6 mice (1500 cells ± 580) and NFM−/− mice (890 cells ± 460) (Fig. 3D), suggesting that endogenous expression of NFM was not negatively regulating expansion of MOG38–49 tetramer–positive cells in peripheral spleen and lymph nodes at this time point.

The identification of polyclonal NFM18–30–specific T cells in the spleen and lymph nodes was difficult without enriching for the cells by tetramer pulldown. Low level detection was possible with this method, most notably at day 14 postchallenge (Fig. 3E, 3F). There was no significant difference in the detection of NFM18–30 tetramer–positive cells between MOG35–55 versus NFM15–35 challenge at days 14 (respectively 6100 cells + 1000 and 3500 cells ± 980, p = 0.09) and 22 (p = 0.25 at 1400 cells ± 490 and 760 cells ± 210 respectively). NFM18–30–specific T cell numbers were significantly diminished by day 22 when compared with the numbers seen at day 14 following MOG35–55 challenge (p = 0.0004) and NFM15–35 challenge (p = 0.008).

The notable difference in peripheral detection of MOG38–49 T cells identified 22 d after NFM15–35 challenge (Fig. 3B) spurred the question whether the missing peripheral MOG-specific T cells had migrated to the CNS by this time point (Fig. 3G). CD4+ CD44hi MOG38–49–specific T cells were found in the CNS after MOG35–55 (760 cells ± 240) but not NFM15–35 challenge (2.7 cells ± 1.7), despite a similar number of CD3+ CD4+ CD44hi T cell infiltrates (25,000 cells ± 7200 and 9800 cells ± 3000 respectively). There was limited detection of tetramer–positive NFM18–30–specific T cells after MOG35–55 challenge (32.8 cells ± 7.0) with no detection of this population in the CNS following NFM15–35 challenge (2.2 cells ± 1.5). The Ag specificity of the remaining CD3+ infiltrates could not be determined by MOG38–49 or NFM18–30 tetramer.

The poor expansion and sustainability of cross-reactive, MOG38–49–specific T cells after NFM15–35 challenge allowed us to probe which amino acids within the NFM20–28 core nonamer would expand encephalitogenic T cells that cross-recognized MOG38–49. Published data indicated NFM poorly binds I-Ab, therefore we first modified NFM20–28 at proposed MHC anchor residues (positions P1, P4, and P9 within the NFM20–28 core), and tested their ability to promote EAE. The P6 anchor residue was not modified because serine is found at this position in both MOG35–55 and NFM15–35 peptides. The NFM-specific amino acids were swapped with the respective residues found within MOG35–55 (Table I). P1 (40Y) of MOG35–55 is a critical MHC anchor residue shown by I-Ab peptide-competition assays such that, functionally, an alanine substitution at this position results in a loss of MOG-specific T cell responsiveness (42, 4446). We therefore made NFM variant T20Y at P1, challenged mice, and found that encephalitogenic potential increased to 33% incidence compared to a challenge with WT NFM15–35 (Fig. 4A, Table II), confirming a previous report (40). We generated additional NFM15–35 MHC variant peptides at P4 and P9, respectively S23P and S28V, but these peptides did not promote EAE (Fig. 4A). The suboptimal induction of disease with the P1 change alone suggested that P1 along with residues outside of the core nonamer may be important for inducing 100% incidence of EAE.

Table I.
Amino acid sequences of MOG35–55, NFM15–35, and NFM15–35 variant peptides
-1123456789
NFM15–35 
MOG3555 M E V G W Y R S P F S R V V H L Y R N G K 
NFM T20Y Y 
NFM E19W, T20Y W Y 
NFM S23P P 
NFM S28V V 
-1123456789
NFM15–35 
MOG3555 M E V G W Y R S P F S R V V H L Y R N G K 
NFM T20Y Y 
NFM E19W, T20Y W Y 
NFM S23P P 
NFM S28V V 

The nine core amino acids (P1–P9) are denoted along with P-1. Substitutions in NFM15–35 were made at P-1, P1, P4, and P9 with the respective amino acids found in MOG35-55 (indicated in bold and italics).

FIGURE 4.

NFM variants at amino residues P1 and P-1 restored MOG3840 tetramer detection and encephalitogenic incidence to 100%. (A) Synthesized variant peptides of NFM15–35 were used to induce EAE in C57BL/6 mice and disease course was monitored [(A) and (B), Table II]. (A) is the graphical version of experiment 1 and (B) is the graphical version of experiment 5 detailed in Table II. The experiments represented in (A) and (B) were performed two times. (C) 2D2 T cell dose response curves for NFM15–35, MOG35–55, and NFM variants. This experiment was performed one time in duplicate. (D) MOG-specific T cell numbers from the periphery at day 22 postchallenge were enumerated with MOG38–49 and NFM18–30 tetramers. Statistics between indicated peptide challenges were performed with two-tailed unpaired, parametric t tests with the following designations: **p < 0.0001 with F test variance, p < 0.0001, *p < 0.0001 with F test variance of p = 0.01, and T20Y versus NFM15–35 was not significant with p = 0.245 and F test variance p = 0.857. MOG35–55 versus NFM E19W, T20Y has p = 0.455 and exhibited no significant differences in means or F test variances. Expansion of NFM-specific T cell numbers in the periphery was also assessed and no overall differences were seen by ordinary one-way ANOVA, p = 0.63.

FIGURE 4.

NFM variants at amino residues P1 and P-1 restored MOG3840 tetramer detection and encephalitogenic incidence to 100%. (A) Synthesized variant peptides of NFM15–35 were used to induce EAE in C57BL/6 mice and disease course was monitored [(A) and (B), Table II]. (A) is the graphical version of experiment 1 and (B) is the graphical version of experiment 5 detailed in Table II. The experiments represented in (A) and (B) were performed two times. (C) 2D2 T cell dose response curves for NFM15–35, MOG35–55, and NFM variants. This experiment was performed one time in duplicate. (D) MOG-specific T cell numbers from the periphery at day 22 postchallenge were enumerated with MOG38–49 and NFM18–30 tetramers. Statistics between indicated peptide challenges were performed with two-tailed unpaired, parametric t tests with the following designations: **p < 0.0001 with F test variance, p < 0.0001, *p < 0.0001 with F test variance of p = 0.01, and T20Y versus NFM15–35 was not significant with p = 0.245 and F test variance p = 0.857. MOG35–55 versus NFM E19W, T20Y has p = 0.455 and exhibited no significant differences in means or F test variances. Expansion of NFM-specific T cell numbers in the periphery was also assessed and no overall differences were seen by ordinary one-way ANOVA, p = 0.63.

Close modal
Table II.
Summary table of EAE experiments with NFM variant peptides compared with MOG3555
Experiment No.Peptiden% IncidenceAverage Day of OnsetAverage Maximal Score
MOG 3555 50 13 
NFM T20Y 50 15 2.5 
NFM S23P — — 
NFM S28V — — 
MOG 3555 100 >14.5 ± 0.7 >3 ± 0.7 
NFM T20Y 25 32 2.5 
NFM S23P — — 
NFM S28V — — 
Combined MOG 35–55 75   
Experiments 1 and 2 NFM T20Y 33.3   
MOG 3555 100 >10 ± 3.5 >3.7 ± 0.29 
NFM E19W, T20Y 100 >10.6 ± 1.5 >2.5 ± 1.2 
MOG 3555 100 >8.2 ± 1.1 >3.9 ± 0.22 
 NFM E19W, T20Y 100 >10.3 ± 2.0 >3.3 ± 0.46 
Experiment No.Peptiden% IncidenceAverage Day of OnsetAverage Maximal Score
MOG 3555 50 13 
NFM T20Y 50 15 2.5 
NFM S23P — — 
NFM S28V — — 
MOG 3555 100 >14.5 ± 0.7 >3 ± 0.7 
NFM T20Y 25 32 2.5 
NFM S23P — — 
NFM S28V — — 
Combined MOG 35–55 75   
Experiments 1 and 2 NFM T20Y 33.3   
MOG 3555 100 >10 ± 3.5 >3.7 ± 0.29 
NFM E19W, T20Y 100 >10.6 ± 1.5 >2.5 ± 1.2 
MOG 3555 100 >8.2 ± 1.1 >3.9 ± 0.22 
 NFM E19W, T20Y 100 >10.3 ± 2.0 >3.3 ± 0.46 

Synthesized variant peptides of NFM15–35 were used to induce EAE in C57BL/6 mice. Disease was monitored in experimental mice (n) and percent incidence of disease along with average day of symptom onset and maximal disease score ± SD are reported. Disease incidences after challenge with MOG35–55 versus T20Y were summarized in the combined section (Experiments 1 and 2). Dashes (—) appear in instances where there was no onset of EAE.

Published data showed that I-Ab can tolerate changes in the peptide MHC anchor positions through predicted hydrogen bonding with peptide N-terminal residues at P-1 and P-2 (45) and that alanine substitutions at P-1 reduce proliferation of MOG-specific T cell lines (44). We subsequently engineered NFM15–35 with the amino acids found at MOG35–55 P-1 concomitantly with the T20Y mutation at P1 generating the double mutant NFM E19W, T20Y (Table I). Introduction of aromatic side chains at P-1 and P1 restored encephalitogenicity to 100% incidence compared with NFM15–35 (Fig. 4B, Table II). The 2D2 TCR transgenic model used to initially define MOG and NFM cross-reactivity showed the dose response curve for NFM E19W, T20Y to be most similar to MOG35–55, exhibiting maximal proliferation at 10 μM peptide. The NFM T20Y proliferative response was most similar to NFM15–35 and marked by maximal proliferation at 1 μM peptide (Fig. 4C). This reiterates the importance of W at P-1 for the characteristic MOG responsiveness.

We appreciated that the N-terminal substitutions were influencing polyclonal encephalitogenicity as well as MOG versus NFM responsiveness in 2D2 T cells, and wanted to further clarify how these substitutions affected expansion of MOG38–49 and NFM18–30 tetramer–positive cells within the polyclonal EAE response (Fig. 4D). High-affinity, MOG38–49 tetramer–positive cells expanded similarly (p = 0.45) between NFM E19W, T20Y (7300 cells ± 1400) and WT MOG35–55–induced EAE (9200 cells ± 1800) (Fig. 4D, left panel). Expansion of MOG38–49–specific T cells was significantly reduced after challenge with NFM variant T20Y (2000 cells ± 540) or WT NFM15–35 (1100 cells ± 410) when compared with the numbers expanded by MOG35–55 (p = 0.006, p < 0.0001 respectively) or NFM E19W, T20Y (p = 0.003, p < 0.0001, respectively). There was no significant difference between expansion of MOG38–49–specific T cells by T20Y and NFM15–35, p = 0.244, which could relate to inconsistent EAE onset with this Ag. Detection of NFM18–30–specific T cells by tetramer was not enhanced by any amino acid substitution examined with no significant difference among the groups (p = 0.68, Fig. 4D, right panel). Overall, the encephalitogenicity of NFM-variant peptides was dependent on expansion of higher-affinity, MOG38–49 tetramer–positive T cells and not NFM18–30–specific T cells.

We previously reported that MOG tetramer enriches for high-affinity T cells, so reduced detection of higher-affinity, NFM18–30–specific T cells by tetramer suggested these T cells display TCR with generally low affinity or avidity interactions with pMHC (25). The 2D micropipette adhesion frequency assay provides a more sensitive technology to sample Ag specificity and effective 2D affinity interactions among individual T cells within the polyclonal response without using avidity-based tetramers. We previously reported that tetramer poorly detected MOG38–49: I-Ab or MOG42–55: I-Ag7–specific T cells when compared with the 2D micropipette adhesion frequency assay (25, 26). Similarly, tetramer also underestimated the percent detection of NFM18–30: I-Ab–specific T cells ∼100 times less when compared with detection by the 2D micropipette adhesion frequency assay, respectively 0.5% versus 55% (Fig. 5A). Specificity of NFM18–30:I-Ab monomeric detection was demonstrated by the lack of binding to CD4+ SMARTA T cells, which are specific for lymphocytic choriomeningitis virus glycoprotein epitope 61–80 (Fig. 5B).

FIGURE 5.

CNS T cell infiltrates that recognize NFM1830 are largely specific for MOG3849. (A) Splenocytes from a mouse primed with NFM15–35 were cultured for 1 wk and tested for NFM18–30 specificity by tetramer (A) or the 2D micropipette adhesion frequency assay with n = 28 cells. (B) NFM specificity of the 2D micropipette adhesion frequency assay was shown via comparison with lymphocytic choriomeningitis virus specific SMARTA T cells cultured on GP66–80 (n = 16 cells). Densities of the NFM18–30 or GP66–77 I-Ab monomers coated on the RBC sensor are reported on the x-axis and each symbol represents one T cell. (C) EAE was induced with MOG35–55 and CD4+ T cells were isolated from the CNS at indicated time points where onset designates days 12–16 (n = 56 cells), peak days 20–23 (n = 70 cells), and chronic days 28–32 (n = 53 cells) postinduction. Each dot represents one T cell. Log-transformed affinities were analyzed by two-tailed, unpaired parametric t tests with assumption of equal SDs were used to compare MOG38–49 versus NFM18–30 specific T cells at each time point; onset (p = 0.51), peak (p = 0.058), and chronic (p = 0.55). Asterisks mark significant differences in population breadth between MOG38–49– and NFM18–30–specific T cells at onset (p = 0.004) and chronic (p = 0.029) disease using unpaired, parametric t tests with F tests to compare variances. (DF) Breakdown of the data collected in (C). (D) Bar graphs indicate the average percent of detection among the individual T cells analyzed per 2D micropipette experiment; average experiments include seven for onset, four for peak, and five for chronic time points. No significant difference was seen between MOG- and NFM-specific detection using a two-tailed paired t tests, p > 0.1864. (E) Each dotted line links one individual T cell between its affinities for MOG38–49 versus NFM18–30. Zero binding of a T cell to a given Ag could not be graphed on a log-scale and were given an arbitrary value of 1 × 10−7 and are graphically visualized below the solid y-intercept line. (F) Breakdown of individual T cell specificity for one or both Ags using the data in (D). Double binder is a category of cells that recognize both MOG38–49 and NFM18–30.

FIGURE 5.

CNS T cell infiltrates that recognize NFM1830 are largely specific for MOG3849. (A) Splenocytes from a mouse primed with NFM15–35 were cultured for 1 wk and tested for NFM18–30 specificity by tetramer (A) or the 2D micropipette adhesion frequency assay with n = 28 cells. (B) NFM specificity of the 2D micropipette adhesion frequency assay was shown via comparison with lymphocytic choriomeningitis virus specific SMARTA T cells cultured on GP66–80 (n = 16 cells). Densities of the NFM18–30 or GP66–77 I-Ab monomers coated on the RBC sensor are reported on the x-axis and each symbol represents one T cell. (C) EAE was induced with MOG35–55 and CD4+ T cells were isolated from the CNS at indicated time points where onset designates days 12–16 (n = 56 cells), peak days 20–23 (n = 70 cells), and chronic days 28–32 (n = 53 cells) postinduction. Each dot represents one T cell. Log-transformed affinities were analyzed by two-tailed, unpaired parametric t tests with assumption of equal SDs were used to compare MOG38–49 versus NFM18–30 specific T cells at each time point; onset (p = 0.51), peak (p = 0.058), and chronic (p = 0.55). Asterisks mark significant differences in population breadth between MOG38–49– and NFM18–30–specific T cells at onset (p = 0.004) and chronic (p = 0.029) disease using unpaired, parametric t tests with F tests to compare variances. (DF) Breakdown of the data collected in (C). (D) Bar graphs indicate the average percent of detection among the individual T cells analyzed per 2D micropipette experiment; average experiments include seven for onset, four for peak, and five for chronic time points. No significant difference was seen between MOG- and NFM-specific detection using a two-tailed paired t tests, p > 0.1864. (E) Each dotted line links one individual T cell between its affinities for MOG38–49 versus NFM18–30. Zero binding of a T cell to a given Ag could not be graphed on a log-scale and were given an arbitrary value of 1 × 10−7 and are graphically visualized below the solid y-intercept line. (F) Breakdown of individual T cell specificity for one or both Ags using the data in (D). Double binder is a category of cells that recognize both MOG38–49 and NFM18–30.

Close modal

CD4+ T cells isolated from the CNS were monitored by the 2D micropipette adhesion frequency assay for MOG38–49 and NFM18–30 specificity and affinity throughout the course of MOG35–55–induced EAE in C57BL/6 mice, particularly at onset of paralytic symptoms (days 12–16 post–MOG35–55 induction), peak (days 20–23 postinduction), and more chronic time points (days 28–32 postinduction) (Fig. 5C). NFM18–30: I-Ab and MOG38–49: I-Ab RBC sensors were placed at opposing ends of the media chamber to prevent mixing, and individual T cells were tested in a random sequence with the sensors. This means one cell was tested for MOG38–49 specificity first, followed by NFM18–30 assessment whereas the second T cell was tested in the opposite order to rule out potential biasing of TCR:pMHC recognition based on memory of a previous pMHC (47). The order of TCR contact with MOG or NFM pMHC did not alter the overall adhesion frequency (Supplemental Fig. 1). Geometric mean affinities of MOG38–49–specific T cells at onset, peak, and chronic time points are reported as 1.1 × 10−5 μm4, 6.2 × 10−6 μm4, and 1.1 × 10−5 μm4 respectively. NFM18–30–specific T cells exhibited means of 1.4 × 10−5 μm4 at onset, 1.1 × 10−5 μm4 at peak, and 9.5 × 10−6 μm4 at chronic time points. Unpaired parametric t test comparisons of the mean MOG and NFM affinities per time point indicated no significant differences between the groups assuming equal SDs (p ≥ 0.058).

Although the means were similar, the breadth of affinities seen within the MOG38–49–specific population differed from NFM18–30–specific cells (Fig. 5C). Significant differences in affinity ranges between these two populations were seen at onset [F(52,30)=2.7, p = 0.004] and chronic [F(48,36)=2.0, p = 0.029] time points as measured by the F-test to compare variances. The greatest breadth in maximal and minimal 2D affinities were seen at onset, with breadths of 1927 for MOG38–49: I-Ab and 158 for NFM18–30: I-Ab; a finding we also reported during EAE onset in NOD mice (26). NFM18–30–specific T cells with high 2D affinities were absent from detection at peak and chronic time points using 1.1 × 10−4 μm4 as a cutoff for high affinity previously published for MOG38–49: I-Ab T cells (25). At onset, the few high-affinity NFM18–30–specific T cells found remained close to this cutoff, with a geometric mean of 1.3 × 10−4 μm4, an affinity below the MOG38–49–specific population at 6.1 × 10−4 μm4. Overall, there is a deficit in the number of higher-affinity, NFM-specific T cells detected by both the 2D micropipette adhesion frequency assay and tetramer (Fig. 3).

The total percentage of cells that recognize MOG38–49 versus NFM18–30 on average (SD) is not significantly different at onset, peak, or chronic time points (p ≥ 0.1864, Fig. 5D). This suggested that there was significant T cell cross-recognition between MOG and NFM. Analysis of individual T cells from the CNS indicated that each cell has a unique, intrinsic capacity to cross-recognize MOG38–49 and NFM18–30. Importantly, individual T cells tested against multiple pMHC revealed that TCR affinity for one Ag did not dictate strength of recognition to the second Ag, meaning high affinity for MOG does not indicate high affinity for NFM or vice versa (Fig. 5E). Breakdown of single-cell cross-recognition (Fig. 5F) indicated that among all time points it was rare to find T cells in the CNS that recognized NFM18–30 only and not also MOG38–49, occurring at 5.7–11.3% of the population. The majority of T cells in the CNS at peak (58.6%) or chronic (56.6%) disease actually recognized both MOG38–49 and NFM18–30, denoted as double binders. Therefore, the NFM18–30–specific T cells at peak and chronic time points are predominantly MOG38–49 specific at 91.1 and 83.3% respectively. It should be noted that the NFM-only T cells could be functionally MOG reactive but below the limit of detection by the 2D micropipette adhesion frequency assay, a phenotype exhibited by 2D2 T cells (34). At onset (Fig. 5E) it was rare to find T cells displaying a 2D2-like phenotype of measurable affinity for NFM and not MOG, only 4 of the 56 cells tested at onset. Overall, 12 of the 56 T cells displayed heteroclitic affinities for NFM over MOG compared with 26 of the 56 cells displaying affinities dominant for MOG over NFM.

The dominant cross-reactivity of NFM-specific cells with MOG led us to track MOG specificity in a NFM T cell line previously reported to promote EAE (14). Krishnamoorthy et al. (14) showed an NFM line could proliferate to both MOG and NFM, and we confirm this cross-recognition with the 2D micropipette adhesion frequency assay (Fig. 6A). A Thy1.1+ NFM15–35 T cell line was adoptively transferred into irradiated C57BL/6 Thy1.2+ recipients with no additional administration of CFA, peptide Ag, or pertussis toxin. After EAE onset (Fig. 6B), we tested Ag specificity of the CNS infiltrates and found MOG38–49 but not NFM18–30 tetramer–positive cells present in the CNS (Fig. 6C). The CNS cell population was a mix of Thy1.1 positive and negative cells, with the Thy1.1 NFM cell line being the source of the CD44hi MOG tetramer positive T cells (Fig. 6D). NFM18–30–specific T cell contributions to EAE are strongly associated with MOG38–49–specific cross-recognition and not standalone NFM reactivity. These data in total point to MOG35–55 as the dominant autoantigen for polyclonal T cell–mediated EAE in C57BL/6 mice even when heteroclitic responsiveness to NFM is present.

FIGURE 6.

Encephalitogenicity of an NFM T cell line is concomitant with infiltration of tetramer-positive, MOG-specific T cells in the CNS. (A) Lymph nodes from NFM15–35 primed C57BL/6 WT mice were harvested and cultured for 1 wk on NFM15–35. A 2D micropipette adhesion frequency assay was used to assess MOG versus NFM specificity. This figure represents one of two experiments, displaying data from 36 cells. (B) Irradiated (400 rad) C57BL/6 mice received 10 × 106 cells from a NFM T cell line (i.p.). Then 25 d post-transfer, the CNS of mice with scores from 1.0 to 3.5 were harvested and Ag specificity was quantitated with MOG38–49 or NFM18–30 tetramer (C). (D) Representative flow plots from adoptive transfer of a Thy1.1+ NFM T cell line in to a Thy1.2 recipient. (C) and (D) are data representative from one of two experiments. All statistical analyses were done using two-tailed unpaired, parametric t tests assuming both populations had equal SDs.

FIGURE 6.

Encephalitogenicity of an NFM T cell line is concomitant with infiltration of tetramer-positive, MOG-specific T cells in the CNS. (A) Lymph nodes from NFM15–35 primed C57BL/6 WT mice were harvested and cultured for 1 wk on NFM15–35. A 2D micropipette adhesion frequency assay was used to assess MOG versus NFM specificity. This figure represents one of two experiments, displaying data from 36 cells. (B) Irradiated (400 rad) C57BL/6 mice received 10 × 106 cells from a NFM T cell line (i.p.). Then 25 d post-transfer, the CNS of mice with scores from 1.0 to 3.5 were harvested and Ag specificity was quantitated with MOG38–49 or NFM18–30 tetramer (C). (D) Representative flow plots from adoptive transfer of a Thy1.1+ NFM T cell line in to a Thy1.2 recipient. (C) and (D) are data representative from one of two experiments. All statistical analyses were done using two-tailed unpaired, parametric t tests assuming both populations had equal SDs.

Close modal

Given the identical TCR contact residues of NFM15–35 and MOG35–55, it was surprising that NFM did not induce EAE (Fig. 1). Ag-specific T cell recognition is critical for generating a robust immune response where increased cell numbers influence the onset and magnitude of the response (48, 49). The idea that autoreactive T cell expansion is influenced by TCR cross-recognition and reactivity to peptides with alterations at TCR or MHC residues introduced through amino acid substitutions has been well documented, particularly in multiple sclerosis and EAE research (5, 15, 19, 42, 5053). In one example, expansion of tetramer–positive, MOG-specific T cells in the polyclonal repertoire has been reported in mice challenged with foreign-derived peptides that are structurally similar to MOG40–48 (19). These mice developed EAE with varying degrees of severity and the weaker disease courses were associated with slightly diminished expansion of higher affinity, MOG tetramer–positive T cells (19). In the case of NFM15–35 challenge, there were reduced numbers of CD44hi MOG38–49 tetramer–positive T cells in the spleen and peripheral lymph nodes, which were not sustained long term when compared with MOG35–55 challenge (Fig. 3). Expansion of a reduced number of MOG-specific cells ultimately failed to promote T cell enrichment in the CNS and EAE after NFM15–35 challenge (Figs. 1, 3, 4).

Important differences between the core nonamers of NFM15–35 and MOG35–55 that influence T cell reactivity are the amino acid residues proposed to interface with I-Ab molecule. In fact, replacement of the P1 MHC anchor residue partially restored the disease potential of the NFM peptide. We previously reported that an MHC variant peptide of MOG35–55 at peptide position P6, MOG 45D, also resulted in poor encephalitogenic potential despite sharing five TCR contact residues with MOG35–55 (33). Of interest was the finding that IFN-γ–deficient signaling rendered mice permissive to 45D-mediated EAE and that disease was concomitant with enhanced detection of MOG38–49–specific, tetramer-positive T cells. These data suggested that a threshold of high-affinity, MOG38–49 tetramer–positive T cells was needed for polyclonal T cells to establish EAE in the IFN-γ–deficient mice. Because IFN-γ deficiency did not restore encephalitogenic potential of NFM15–35 as it did for 45D (Fig. 3), we addressed the role of additional amino acid differences between NFM15–35 and MOG35–55 that hindered T cell encephalitogenicity.

We found that amino acid substitutions within NFM15–35 enabled us to map the residues critical for encephalitogenicity and expansion of MOG38–49 tetramer–positive T cells in a polyclonal setting (Fig. 4). For MOG35–55, evidence suggested that amino acid residues P-1 and P-2 outside of the core nonamer contribute to MHC anchoring and to TCR binding in individual clones (44, 45, 54, 55), with implications for T cell recognition of NFM15–35 (28). We focused on modulating NFM at P1 (T20Y) to mirror MOG at that position because Petersen et al. (44) reported that MOG Y40 was important for MHC binding and T cell responsiveness; S23P and S28V MHC anchor substitutions were tested in comparison. S23P and S28V did not promote EAE in C57BL/6 mice and T20Y exhibited a low incidence of disease (Table II), a phenotype published during this time (40). We further showed that dual alterations at NFM15–35 P-1 (E19W) and P1 (T20Y) enhanced EAE incidence to 100% (Table II). It was interesting that EAE incidence of NFM E19W, T20Y was concomitant with expansion of polyclonal MOG38–49 tetramer–positive T cells and a MOG-like proliferation profile by 2D2 monoclonal T cells (Fig. 4). Expansion of high numbers of polyclonal MOG38–49 tetramer–positive T cells was clearly dependent on the Y20 and W19 amino acids at P1 and P-1, respectively. Mapping of the amino acids required for expansion of tetramer-positive T cells suggested that a threshold number of MOG38–49–specific T cells was required for consistent onset of disease, namely an average of 7324–9244 cells seen with NFM E19W, T20Y, and MOG35–55.

NFM-deficient mice were used to indicate whether Ag-specific selection pressures influenced encephalitogenicity. Although it was possible that T cells educated in these mice would be hyperresponsive to NFM based on reports showing enhanced Ag-specific T cell responses in MBP−/− and MOG−/− mice (29, 56), we saw no significant difference in NFM-specific T cell proliferation between WT and NFM-deficient mice (Fig. 2). This could likely relate to the findings that peptides with low affinity for MHC do not mediate thymocyte negative selection, which first came to prominence with acetylated MBP (57). Our group also found this true for a model Ag system (58). It is not totally unexpected that deletion failed to alter the NFM-specific T cell response considering the MHC alterations inherent in NFM15–35 elicit a weaker association for I-Ab than MOG35–55 (28), which is consistent with the high concentration of NFM required to tolerize EAE compared with MOG (59). In support of this, our studies revealed that cross-reactivity was not altered between WT and NFM−/− mice because MOG35–55 induced EAE similarly in both mouse strains (Fig. 2). Furthermore, a report published during this time showed that ex vivo T cell cross-reactivity between MOG and NFM could be enhanced in MOG−/− but not NFM−/− mice (40). These data combined with the finding that NFM15–35 challenge of MOG−/− mice still did not induce EAE through cross-reactivity (Fig. 2) supported the finding that NFM is not a critical T cell Ag in polyclonal EAE.

The equivalent disease severity exhibited by our NFM−/− mice after MOG35–55 challenge differed somewhat from another report showing reduced EAE severity in the absence of NFM (59). Our mice (30) backcrossed to C57BL/6N had fulminant EAE and it is not known whether environmental or genetic factors caused this point of phenotypic divergence. However, our data still support the observations of Ramadan et al. (59) that NFM expression was not required for T cell–mediated EAE. This is consistent with the reports that proliferation and cytokine production by MOG-specific T cells were markedly diminished by restimulation with NFM compared with cognate MOG (28, 40, 59). We support that CNS T cell infiltrates are functionally responsive to NFM during EAE (14, 28), because pMHC monomers detected T cell cross-recognition of MOG38–49 and NFM18–30 by the majority of infiltrates at peak and chronic time points (Fig. 5).

Dissecting the role of individual cross-reactive clones in EAE lends to observations of heteroclitic responses where there is enhanced recognition and responsiveness to a second epitope above the initial priming Ag (Fig. 5). Heteroclitic behavior is best documented with clonal TCR in response to peptides with amino acid substitutions at TCR or MHC contacts but how one clone reflects on the bulk polyclonal response is less clear and dependent on previous antigenic exposures (12, 57, 6062). Furthermore, the priming Ag, whether NFM15–35 or MOG35–55, could lead to asymmetry or skewedness of the immune response based on divergent expansion of unique T cell clones. Certainly there are reported differences in percentage detection of individual TCRα and TCRβ usage after priming with these respective peptides (14), yet overlap exists with divergence being dictated by a few amino acids at P1 and P-1 at the N terminus of the core nonamer (Fig.4). The sequence similarities between MOG35–55 and NFM15–35 and the current data in the field support that T cells are largely bispecific between these two peptides. We would argue that our analyses question the functional significance of NFM15–35 on a polyclonal level because NFM-monospecific T cells are rare in the population (Fig. 5). Even in the presence of heteroclitic recognition of NFM, we found MOG to be the dominant autoantigen for disease onset in the absence of standalone encephalitogenicity by NFM15–35 (Fig. 2).

High-affinity, MOG tetramer–positive T cells are a relatively small population within the polyclonal repertoire (2527, 63), and our data suggest they are significant contributors to EAE onset. These data together with the report that expansion or engraftment numbers have been associated with spontaneous EAE in retrogenic models (49) provide insight as to why NFM does not actively induce EAE. Reduced expansion of MOG tetramer–positive T cells by NFM, when compared to MOG, is further complicated by the inability of these cells to migrate to the CNS and cause EAE. We would argue that NFM is not providing a strong enough stimulus to evoke de novo EAE based on the weak interaction with I-Ab, which can be overcome with amino acid substitutions (Fig. 4). Culturing NFM-primed splenocytes with NFM was a means to enhance T cell recognition and activation on NFM to enhance expansion encephalitogenic NFM-or MOG-specific T cell clones (14, 64, 65). As previously reported, NFM-expanded T cells were able to passively transfer EAE, but we found this was associated with transfer of cross-reactive MOG38–49 tetramer–positive T cells capable of enrichment in the CNS (Fig. 6). This again showcased NFM without standalone encephalitogenic potential and supported MOG as the critical determinant in the bispecific model required for the onset of demyelinating autoimmune disease. Overall, our experiments in this study indicate that expansion of a threshold number of high-affinity, MOG tetramer–positive T cells within the polyclonal response is a readout of encephalitogenic potential.

We acknowledge the NIH Tetramer Core Facility at Emory University for providing pMHC tetramers and biotinylated pMHC monomers for use in the tetramer pulldown and 2D micropipette adhesion frequency assays. We thank Dr. Marla Gearing from the Emory National Institute of Neurological Disorders and Stroke Neuropathology/Histochemistry Core Facility for collaborative spirit and helpful comments on presenting the histological data. We thank Jennifer Cosby for reviewing the manuscript.

This work was supported by funding from the National Institutes of Health and the National Multiple Sclerosis Society. Grants supporting this research include National Institute of Neurological Disorders and Stroke Grant R01 NS071518 and National Institute of Allergy and Infectious Diseases Grant R01 AI110113 (to B.D.E.). L.B. was supported by a postdoctoral fellowship from the National Multiple Sclerosis Society (Grant FG1963A1/1). J.J.S. is currently funded by the National Institutes of Health (Grant R25 NS070680) and the National Multiple Sclerosis Society (Grant 127992A).

The online version of this article contains supplemental material.

Abbreviations used in this article:

2D

two-dimensional

EAE

experimental autoimmune encephalomyelitis

MOG35–55

myelin oligodendrocyte glycoprotein epitope 35–55

NFM15–35

neurofilament medium protein epitope 15–35

NIH

National Institutes of Health

pMHC

peptide:MHC

WT

wild-type.

1
Bielekova
,
B.
,
M. H.
Sung
,
N.
Kadom
,
R.
Simon
,
H.
McFarland
,
R.
Martin
.
2004
.
Expansion and functional relevance of high-avidity myelin-specific CD4+ T cells in multiple sclerosis.
J. Immunol.
172
:
3893
3904
.
2
Yu
,
W.
,
N.
Jiang
,
P. J.
Ebert
,
B. A.
Kidd
,
S.
Müller
,
P. J.
Lund
,
J.
Juang
,
K.
Adachi
,
T.
Tse
,
M. E.
Birnbaum
, et al
.
2015
.
Clonal deletion prunes but does not eliminate self-specific αβ CD8(+) T lymphocytes.
Immunity
42
:
929
941
.
3
Kieback
,
E.
,
E.
Hilgenberg
,
U.
Stervbo
,
V.
Lampropoulou
,
P.
Shen
,
M.
Bunse
,
Y.
Jaimes
,
P.
Boudinot
,
A.
Radbruch
,
U.
Klemm
, et al
.
2016
.
Thymus-derived regulatory T cells are positively selected on natural self-antigen through cognate interactions of high functional avidity.
Immunity
44
:
1114
1126
.
4
Isobe
,
N.
,
L.
Madireddy
,
P.
Khankhanian
,
T.
Matsushita
,
S. J.
Caillier
,
J. M.
Moré
,
P. A.
Gourraud
,
J. L.
McCauley
,
A. H.
Beecham
,
L.
Piccio
, et al
International Multiple Sclerosis Genetics Consortium
.
2015
.
An ImmunoChip study of multiple sclerosis risk in African Americans.
Brain
138
:
1518
1530
.
5
Su
,
L. F.
,
B. A.
Kidd
,
A.
Han
,
J. J.
Kotzin
,
M. M.
Davis
.
2013
.
Virus-specific CD4(+) memory-phenotype T cells are abundant in unexposed adults.
Immunity
38
:
373
383
.
6
Korn
,
T.
,
A.
Kallies
.
2017
.
T cell responses in the central nervous system.
Nat. Rev. Immunol.
17
:
179
194
.
7
Trowsdale
,
J.
2011
.
The MHC, disease and selection.
Immunol. Lett.
137
:
1
8
.
8
Riedhammer
,
C.
,
R.
Weissert
.
2015
.
Antigen presentation, autoantigens, and immune regulation in multiple sclerosis and other autoimmune diseases.
Front. Immunol.
6
:
322
.
9
Schubert
,
D. A.
,
S.
Gordo
,
J. J.
Sabatino
Jr.
,
S.
Vardhana
,
E.
Gagnon
,
D. K.
Sethi
,
N. P.
Seth
,
K.
Choudhuri
,
H.
Reijonen
,
G. T.
Nepom
, et al
.
2012
.
Self-reactive human CD4 T cell clones form unusual immunological synapses.
J.Exp. Med.
209
:
335
352
.
10
Wucherpfennig
,
K. W.
,
J. L.
Strominger
.
1995
.
Molecular mimicry in T cell-mediated autoimmunity: viral peptides activate human T cell clones specific for myelin basic protein.
Cell
80
:
695
705
.
11
Legoux
,
F. P.
,
J. B.
Lim
,
A. W.
Cauley
,
S.
Dikiy
,
J.
Ertelt
,
T. J.
Mariani
,
T.
Sparwasser
,
S. S.
Way
,
J. J.
Moon
.
2015
.
CD4+ T cell tolerance to tissue-restricted self antigens is mediated by antigen-specific regulatory T cells rather than deletion.
Immunity
43
:
896
908
.
12
Cole
,
D. K.
,
A. M.
Bulek
,
G.
Dolton
,
A. J.
Schauenberg
,
B.
Szomolay
,
W.
Rittase
,
A.
Trimby
,
P.
Jothikumar
,
A.
Fuller
,
A.
Skowera
, et al
.
2016
.
Hotspot autoimmune T cell receptor binding underlies pathogen and insulin peptide cross-reactivity.
J. Clin. Invest.
126
:
2191
2204
.
13
Sethi
,
D. K.
,
S.
Gordo
,
D. A.
Schubert
,
K. W.
Wucherpfennig
.
2013
.
Crossreactivity of a human autoimmune TCR is dominated by a single TCR loop.
Nat. Commun.
4
:
2623
.
14
Krishnamoorthy
,
G.
,
A.
Saxena
,
L. T.
Mars
,
H. S.
Domingues
,
R.
Mentele
,
A.
Ben-Nun
,
H.
Lassmann
,
K.
Dornmair
,
F. C.
Kurschus
,
R. S.
Liblau
,
H.
Wekerle
.
2009
.
Myelin-specific T cells also recognize neuronal autoantigen in a transgenic mouse model of multiple sclerosis.
Nat. Med.
15
:
626
632
.
15
Wooldridge
,
L.
,
J.
Ekeruche-Makinde
,
H. A.
van den Berg
,
A.
Skowera
,
J. J.
Miles
,
M. P.
Tan
,
G.
Dolton
,
M.
Clement
,
S.
Llewellyn-Lacey
,
D. A.
Price
, et al
.
2012
.
A single autoimmune T cell receptor recognizes more than a million different peptides.
J. Biol. Chem.
287
:
1168
1177
.
16
Birnbaum
,
M. E.
,
J. L.
Mendoza
,
D. K.
Sethi
,
S.
Dong
,
J.
Glanville
,
J.
Dobbins
,
E.
Ozkan
,
M. M.
Davis
,
K. W.
Wucherpfennig
,
K. C.
Garcia
.
2014
.
Deconstructing the peptide-MHC specificity of T cell recognition.
Cell
157
:
1073
1087
.
17
Zarnitsyna
,
V. I.
,
B. D.
Evavold
,
L. N.
Schoettle
,
J. N.
Blattman
,
R.
Antia
.
2013
.
Estimating the diversity, completeness, and cross-reactivity of the T cell repertoire.
Front. Immunol.
4
:
485
.
18
Maynard
,
J.
,
K.
Petersson
,
D. H.
Wilson
,
E. J.
Adams
,
S. E.
Blondelle
,
M. J.
Boulanger
,
D. B.
Wilson
,
K. C.
Garcia
.
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
.
19
Nelson
,
R. W.
,
D.
Beisang
,
N. J.
Tubo
,
T.
Dileepan
,
D. L.
Wiesner
,
K.
Nielsen
,
M.
Wüthrich
,
B. S.
Klein
,
D. I.
Kotov
,
J. A.
Spanier
, et al
.
2015
.
T cell receptor cross-reactivity between similar foreign and self peptides influences naive cell population size and autoimmunity. [Published erratum appears in 2015 Immunity 42: 1212–1213.]
Immunity
42
:
95
107
.
20
Evavold
,
B. D.
,
P. M.
Allen
.
1991
.
Separation of IL-4 production from Th cell proliferation by an altered T cell receptor ligand.
Science
252
:
1308
1310
.
21
Sloan-Lancaster
,
J.
,
B. D.
Evavold
,
P. M.
Allen
.
1993
.
Induction of T-cell anergy by altered T-cell-receptor ligand on live antigen-presenting cells.
Nature
363
:
156
159
.
22
Hsu
,
B. L.
,
B. D.
Evavold
,
P. M.
Allen
.
1995
.
Modulation of T cell development by an endogenous altered peptide ligand.
J. Exp. Med.
181
:
805
810
.
23
Hong
,
J.
,
S. P.
Persaud
,
S.
Horvath
,
P. M.
Allen
,
B. D.
Evavold
,
C.
Zhu
.
2015
.
Force-regulated in situ TCR-peptide-bound MHC class II kinetics determine functions of CD4+ T cells.
J. Immunol.
195
:
3557
3564
.
24
Liu
,
B.
,
W.
Chen
,
B. D.
Evavold
,
C.
Zhu
.
2014
.
Accumulation of dynamic catch bonds between TCR and agonist peptide-MHC triggers T cell signaling.
Cell
157
:
357
368
.
25
Sabatino
,
J. J. Jr.
,
J.
Huang
,
C.
Zhu
,
B. D.
Evavold
.
2011
.
High prevalence of low affinity peptide-MHC II tetramer-negative effectors during polyclonal CD4+ T cell responses.
J. Exp. Med.
208
:
81
90
.
26
Kersh
,
A. E.
,
L. J.
Edwards
,
B. D.
Evavold
.
2014
.
Progression of relapsing-remitting demyelinating disease does not require increased TCR affinity or epitope spread.
J. Immunol.
193
:
4429
4438
.
27
Hood
,
J. D.
,
V. I.
Zarnitsyna
,
C.
Zhu
,
B. D.
Evavold
.
2015
.
Regulatory and T effector cells have overlapping low to high ranges in TCR affinities for self during demyelinating disease.
J. Immunol.
195
:
4162
4170
.
28
Lucca
,
L. E.
,
S.
Desbois
,
A.
Ramadan
,
A.
Ben-Nun
,
M.
Eisenstein
,
N.
Carrié
,
J. C.
Guéry
,
A.
Sette
,
P.
Nguyen
,
T. L.
Geiger
, et al
.
2014
.
Bispecificity for myelin and neuronal self-antigens is a common feature of CD4 T cells in C57BL/6 mice.
J. Immunol.
193
:
3267
3277
.
29
Liñares
,
D.
,
P.
Mañá
,
M.
Goodyear
,
A. M.
Chow
,
C.
Clavarino
,
N. D.
Huntington
,
L.
Barnett
,
F.
Koentgen
,
R.
Tomioka
,
C. C.
Bernard
, et al
.
2003
.
The magnitude and encephalogenic potential of autoimmune response to MOG is enhanced in MOG deficient mice.
J. Autoimmun.
21
:
339
351
.
30
Jacomy
,
H.
,
Q.
Zhu
,
S.
Couillard-Després
,
J. M.
Beaulieu
,
J. P.
Julien
.
1999
.
Disruption of type IV intermediate filament network in mice lacking the neurofilament medium and heavy subunits.
J. Neurochem.
73
:
972
984
.
31
Oxenius
,
A.
,
M. F.
Bachmann
,
R. M.
Zinkernagel
,
H.
Hengartner
.
1998
.
Virus-specific MHC-class II-restricted TCR-transgenic mice: effects on humoral and cellular immune responses after viral infection.
Eur. J. Immunol.
28
:
390
400
.
32
Moon
,
J. J.
,
H. H.
Chu
,
J.
Hataye
,
A. J.
Pagán
,
M.
Pepper
,
J. B.
McLachlan
,
T.
Zell
,
M. K.
Jenkins
.
2009
.
Tracking epitope-specific T cells.
Nat. Protoc.
4
:
565
581
.
33
Sabatino
,
J. J. Jr.
,
J.
Shires
,
J. D.
Altman
,
M. L.
Ford
,
B. D.
Evavold
.
2008
.
Loss of IFN-gamma enables the expansion of autoreactive CD4+ T cells to induce experimental autoimmune encephalomyelitis by a nonencephalitogenic myelin variant antigen.
J. Immunol.
180
:
4451
4457
.
34
Rosenthal
,
K. M.
,
L. J.
Edwards
,
J. J.
Sabatino
Jr.
,
J. D.
Hood
,
H. A.
Wasserman
,
C.
Zhu
,
B. D.
Evavold
.
2012
.
Low 2-dimensional CD4 T cell receptor affinity for myelin sets in motion delayed response kinetics.
PLoS One
7
:
e32562
.
35
Chesla
,
S. E.
,
P.
Selvaraj
,
C.
Zhu
.
1998
.
Measuring two-dimensional receptor-ligand binding kinetics by micropipette.
Biophys. J.
75
:
1553
1572
.
36
Huang
,
J.
,
V. I.
Zarnitsyna
,
B.
Liu
,
L. J.
Edwards
,
N.
Jiang
,
B. D.
Evavold
,
C.
Zhu
.
2010
.
The kinetics of two-dimensional TCR and pMHC interactions determine T-cell responsiveness.
Nature
464
:
932
936
.
37
Blanchfield
,
J. L.
,
S. K.
Shorter
,
B. D.
Evavold
.
2013
.
Monitoring the dynamics of T cell clonal diversity using recombinant peptide:MHC technology.
Front. Immunol.
4
:
170
.
38
Zarnitsyna
,
V. I.
,
C.
Zhu
.
2011
.
Adhesion frequency assay for in situ kinetics analysis of cross-junctional molecular interactions at the cell-cell interface.
J. Vis. Exp.
57
:
e3519
.
39
St-Pierre
,
C.
,
S.
Brochu
,
J. R.
Vanegas
,
M.
Dumont-Lagacé
,
S.
Lemieux
,
C.
Perreault
.
2013
.
Transcriptome sequencing of neonatal thymic epithelial cells.
Sci. Rep.
3
:
1860
.
40
Lucca
,
L. E.
,
P. P.
Axisa
,
M.
Aloulou
,
C.
Perals
,
A.
Ramadan
,
P.
Rufas
,
B.
Kyewski
,
J.
Derbinski
,
N.
Fazilleau
,
L. T.
Mars
,
R. S.
Liblau
.
2016
.
Myelin oligodendrocyte glycoprotein induces incomplete tolerance of CD4(+) T cells specific for both a myelin and a neuronal self-antigen in mice.
Eur. J. Immunol.
46
:
2247
2259
.
41
Wasserman
,
H. A.
,
C. D.
Beal
,
Y.
Zhang
,
N.
Jiang
,
C.
Zhu
,
B. D.
Evavold
.
2008
.
MHC variant peptide-mediated anergy of encephalitogenic T cells requires SHP-1.
J. Immunol.
181
:
6843
6849
.
42
Ford
,
M. L.
,
B. D.
Evavold
.
2003
.
Regulation of polyclonal T cell responses by an MHC anchor-substituted variant of myelin oligodendrocyte glycoprotein 35-55.
J. Immunol.
171
:
1247
1254
.
43
Bettini
,
M.
,
K.
Rosenthal
,
B. D.
Evavold
.
2009
.
Pathogenic MOG-reactive CD8+ T cells require MOG-reactive CD4+ T cells for sustained CNS inflammation during chronic EAE.
J. Neuroimmunol.
213
:
60
68
.
44
Petersen
,
T. R.
,
E.
Bettelli
,
J.
Sidney
,
A.
Sette
,
V.
Kuchroo
,
B. T.
Bäckström
.
2004
.
Characterization of MHC- and TCR-binding residues of the myelin oligodendrocyte glycoprotein 38-51 peptide.
Eur. J. Immunol.
34
:
165
173
.
45
Liu
,
X.
,
S.
Dai
,
F.
Crawford
,
R.
Fruge
,
P.
Marrack
,
J.
Kappler
.
2002
.
Alternate interactions define the binding of peptides to the MHC molecule IA(b).
Proc. Natl. Acad. Sci. USA
99
:
8820
8825
.
46
Ben-Nun
,
A.
,
I.
Mendel
,
R.
Bakimer
,
M.
Fridkis-Hareli
,
D.
Teitelbaum
,
R.
Arnon
,
M.
Sela
,
N.
Kerlero de Rosbo
.
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
(
4
Suppl. 1
):
S14
S22
.
47
Zarnitsyna
,
V. I.
,
J.
Huang
,
F.
Zhang
,
Y. H.
Chien
,
D.
Leckband
,
C.
Zhu
.
2007
.
Memory in receptor-ligand-mediated cell adhesion.
Proc. Natl. Acad. Sci. USA
104
:
18037
18042
.
48
Moon
,
J. J.
,
H. H.
Chu
,
M.
Pepper
,
S. J.
McSorley
,
S. C.
Jameson
,
R. M.
Kedl
,
M. K.
Jenkins
.
2007
.
Naive CD4(+) T cell frequency varies for different epitopes and predicts repertoire diversity and response magnitude.
Immunity
27
:
203
213
.
49
Alli
,
R.
,
P.
Nguyen
,
T. L.
Geiger
.
2008
.
Retrogenic modeling of experimental allergic encephalomyelitis associates T cell frequency but not TCR functional affinity with pathogenicity.
J. Immunol.
181
:
136
145
.
50
Karin
,
N.
,
D. J.
Mitchell
,
S.
Brocke
,
N.
Ling
,
L.
Steinman
.
1994
.
Reversal of experimental autoimmune encephalomyelitis by a soluble peptide variant of a myelin basic protein epitope: T cell receptor antagonism and reduction of interferon gamma and tumor necrosis factor alpha production.
J. Exp. Med.
180
:
2227
2237
.
51
Nicholson
,
L. B.
,
J. M.
Greer
,
R. A.
Sobel
,
M. B.
Lees
,
V. K.
Kuchroo
.
1995
.
An altered peptide ligand mediates immune deviation and prevents autoimmune encephalomyelitis.
Immunity
3
:
397
405
.
52
Bielekova
,
B.
,
B.
Goodwin
,
N.
Richert
,
I.
Cortese
,
T.
Kondo
,
G.
Afshar
,
B.
Gran
,
J.
Eaton
,
J.
Antel
,
J. A.
Frank
, et al
.
2000
.
Encephalitogenic potential of the myelin basic protein peptide (amino acids 83-99) in multiple sclerosis: results of a phase II clinical trial with an altered peptide ligand.
Nat. Med.
6
:
1167
1175
.
53
Kappos
,
L.
,
G.
Comi
,
H.
Panitch
,
J.
Oger
,
J.
Antel
,
P.
Conlon
,
L.
Steinman
;
The Altered Peptide Ligand in Relapsing MS Study Group
.
2000
.
Induction of a non-encephalitogenic type 2 T helper-cell autoimmune response in multiple sclerosis after administration of an altered peptide ligand in a placebo-controlled, randomized phase II trial.
Nat. Med.
6
:
1176
1182
.
54
Ben-Nun
,
A.
,
N.
Kerlero de Rosbo
,
N.
Kaushansky
,
M.
Eisenstein
,
L.
Cohen
,
J. F.
Kaye
,
I.
Mendel
.
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
.
55
Udyavar
,
A.
,
R.
Alli
,
P.
Nguyen
,
L.
Baker
,
T. L.
Geiger
.
2009
.
Subtle affinity-enhancing mutations in a myelin oligodendrocyte glycoprotein-specific TCR alter specificity and generate new self-reactivity.
J. Immunol.
182
:
4439
4447
.
56
Harrington
,
C. J.
,
A.
Paez
,
T.
Hunkapiller
,
V.
Mannikko
,
T.
Brabb
,
M.
Ahearn
,
C.
Beeson
,
J.
Goverman
.
1998
.
Differential tolerance is induced in T cells recognizing distinct epitopes of myelin basic protein.
Immunity
8
:
571
580
.
57
Anderton
,
S. M.
,
C. G.
Radu
,
P. A.
Lowrey
,
E. S.
Ward
,
D. C.
Wraith
.
2001
.
Negative selection during the peripheral immune response to antigen.
J. Exp. Med.
193
:
1
11
.
58
McNeil
,
L. K.
,
B. D.
Evavold
.
2002
.
Dissociation of peripheral T cell responses from thymocyte negative selection by weak agonists supports a spare receptor model of T cell activation.
Proc. Natl. Acad. Sci. USA
99
:
4520
4525
.
59
Ramadan
,
A.
,
L. E.
Lucca
,
N.
Carrié
,
S.
Desbois
,
P. P.
Axisa
,
M.
Hayder
,
J.
Bauer
,
R. S.
Liblau
,
L. T.
Mars
.
2016
.
In situ expansion of T cells that recognize distinct self-antigens sustains autoimmunity in the CNS.
Brain
139
:
1433
1446
.
60
Petrova
,
G.
,
A.
Ferrante
,
J.
Gorski
.
2012
.
Cross-reactivity of T cells and its role in the immune system.
Crit. Rev. Immunol.
32
:
349
372
.
61
Nicholson
,
L. B.
,
H.
Waldner
,
A. M.
Carrizosa
,
A.
Sette
,
M.
Collins
,
V. K.
Kuchroo
.
1998
.
Heteroclitic proliferative responses and changes in cytokine profile induced by altered peptides: implications for autoimmunity.
Proc. Natl. Acad. Sci. USA
95
:
264
269
.
62
Madura
,
F.
,
P. J.
Rizkallah
,
C. J.
Holland
,
A.
Fuller
,
A.
Bulek
,
A. J.
Godkin
,
A. J.
Schauenburg
,
D. K.
Cole
,
A. K.
Sewell
.
2015
.
Structural basis for ineffective T-cell responses to MHC anchor residue-improved “heteroclitic” peptides.
Eur. J. Immunol.
45
:
584
591
.
63
Martinez
,
R. J.
,
R.
Andargachew
,
H. A.
Martinez
,
B. D.
Evavold
.
2016
.
Low-affinity CD4+ T cells are major responders in the primary immune response.
Nat. Commun.
7
:
13848
.
64
Nguyen
,
P.
,
W.
Liu
,
J.
Ma
,
J. N.
Manirarora
,
X.
Liu
,
C.
Cheng
,
T. L.
Geiger
.
2010
.
Discrete TCR repertoires and CDR3 features distinguish effector and Foxp3+ regulatory T lymphocytes in myelin oligodendrocyte glycoprotein-induced experimental allergic encephalomyelitis.
J. Immunol.
185
:
3895
3904
.
65
Zhao
,
Y.
,
P.
Nguyen
,
J.
Ma
,
T.
Wu
,
L. L.
Jones
,
D.
Pei
,
C.
Cheng
,
T. L.
Geiger
.
2016
.
Preferential use of public TCR during autoimmune encephalomyelitis.
J. Immunol.
196
:
4905
4914
.

The authors have no financial conflicts of interest.

Supplementary data