Breaking T cell tolerance against self type II collagen in HLA–DR4–transgenic mice and development of autoimmune arthritis




To establish a new animal model in DRB1*0401 (DR4)–transgenic mice in which T cell tolerance to self type II collagen (CII) can be broken and allow for the development of autoimmune arthritis, to investigate the role of posttranslational modifications of the CII259–273 epitope in the induction and breaking of tolerance of DR4-restricted T cells, and to characterize DR4-restricted T cell recognition of the immunodominant CII259–273 epitope.


DR4-transgenic mice expressing either the entire human CII protein (HuCII) or only the immunodominant T cell epitope of heterologous CII (MMC) in joint cartilage were established on different genetic backgrounds, and susceptibility to collagen-induced arthritis (CIA) was tested.


HuCII mice displayed stronger T cell tolerance to heterologous CII than did MMC mice. On the B10 background, arthritis developed only in MMC mice with a defective oxidative burst. However, MMC mice on the C3H background were susceptible to arthritis also with a functional oxidative burst. Significant recall responses in tolerized mice were detected only against the nonglycosylated CII259–273 epitope. Recognition of the CII259–273 epitope was heterogeneous, but the majority of T cells in DR4 mice specifically recognized the nonglycosylated side chain of lysine at position 264.


It is possible to break tolerance to self CII and induce arthritis in DR4 mice. However, arthritis susceptibility is tightly controlled by the genetic background and by the source of the transgenic element for expressing the heterologous CII peptide as a self CII protein in the joint. In contrast to CIA in Aq-expressing mice, the nonglycosylated CII259–273 epitope is clearly immunodominant in both tolerized and nontolerized DR4 mice.

Rheumatoid arthritis (RA) is a chronic inflammatory disease affecting the peripheral joints. The role of T cells in RA is supported by the occurrence of activated CD4+ T cells within the inflamed synovium and by immunogenetic data linking RA with shared epitope class II major histocompatibility complex (MHC) genes and with PTPN22, as well as by the approval of abatacept as a treatment of RA (1).

Type II collagen (CII) is the major component of hyaline cartilage and has been suggested as a possible autoantigen in RA, since autoimmunity to CII is commonly detected in RA. In addition, an RA-like disease, collagen-induced arthritis (CIA), can be induced in mice after immunization with CII. Susceptibility to CIA is associated with the murine class II MHC molecule Aq, but also mice with transgenic expression of the RA-associated class II MHC molecule HLA–DRB1*0401 (DR4) or *0101 (DR1) can develop CIA following immunization with CII. Interestingly, Aq, DR1, and DR4 molecules present almost the identical immunodominant epitope, which has been localized to amino acid positions 259–273 on CII (2–7).

Posttranslational modification has been suggested as a possible mechanism whereby immunologic tolerance to self antigens can be broken. This has also been highlighted in RA, in which autoantibodies specific for citrullinated self antigens are highly specific for RA and can be detected several years before the onset of clinical disease (1). CII is also subjected to citrullination; autoantibodies specific for citrullinated CII have been observed in 40% of RA patients (8), and antibodies specific for citrullinated CII are pathogenic in mice (9). With regard to T cells, the immunodominant CII259–273 peptide harbors 2 lysine residues, at positions 264 and 270, which can be posttranslationally modified by hydroxylation and subsequent glycosylation. Remarkably, the CII259–273 epitope has been found to be uniformly glycosylated in normal cartilage but modified by arthritis in rats and humans (10, 11).

Although the role of CII as a relevant autoantigen in RA is unclear, CII may still serve as an excellent model autoantigen for understanding how the immune system interacts with joint-derived self antigens in order to establish immunologic tolerance and for understanding how posttranslational modifications may influence these processes. We previously investigated T cell tolerance to self CII in DR4-expressing mice (12). By comparing the recall response in DR4-transgenic mice, which either did or did not express human CII in a cartilage-specific manner, we could show that DR4-restricted T cells exhibit strong tolerance to self CII, and despite considerable efforts, we were not able to induce CIA in CII-tolerized HuCII.DR4-transgenic mice. The aim of the present investigation therefore was to establish a new animal model in which the T cell tolerance to self CII could be broken in order to initiate an autoimmune response to self CII and the development of arthritis. We also wanted to investigate the role of posttranslational modifications of the CII259–273 epitope in the induction and breaking of tolerance of DR4-restricted T cells, since this has been found to be of major importance in mice expressing the murine Aq molecule (13).



Rat CII was prepared from Swarm rat chondrosarcoma by pepsin digestion or from lathyritic rat chondrosarcoma and was further purified as described previously (14). We used the following CII peptides, containing the 259–273 sequence of rat CII, with or without modifications of the K264 and/or K270 residue: K264 (unmodified lysine at position 264); K264R, K270R, and K264/270R (where K264, K270, and both residues are substituted for arginine); HyK ([5R]-5-hydroxy-L-lysine at position 264); and Gal264, Gal270, and Gal264/270 ([β]-D-galactopyranosyl residue on L-hydroxylysine at position 264, position 270, or both residues, respectively). The CII peptides were synthesized, purified, and characterized as previously described (7,15,16). Both CII and CII peptides were dissolved and stored in 0.1M acetic acid at 4°C. Purified protein derivative (PPD) was from Statens Serum Institut.


Transgenic mice coexpressing DR4 and human CD4 as well as human CII (HuCII) while lacking endogenous murine class II MHC have been described previously (12,17–19). These humanized mice were backcrossed for 10 generations onto the C3H.Q and B10.Q backgrounds and subsequently intercrossed before being used in experiments. Before intercrossing, the HuCII transgene was also exchanged for the MMC transgene (20) by crossing with MMC.C3H.Q or MMC.B10.Q.Ncf1*/* mice (21,22) (i.e., MMC-transgenic mice that also lack a functional Ncf1 gene, therefore lacking a functional phagocytic NADPH oxidase complex 2). The MMC transgene is a mutated mouse CII gene in which the amino acid at position 266 has been changed from an aspartic acid to a glutamic acid, thereby expressing the rat/human CII259–273 epitope in a CII-specific manner. Thus, while the HuCII transgene will lead to expression of the entire human CII protein, MMC-transgenic mice only express the heterologous (rat/human) CII259–273 epitope in the cartilage. The strains described above were maintained by repeated intercrossing at the animal facility at the Division of Medical Inflammation Research, Lund University. MMC- and HuCII-expressing mice were kept as heterozygotes, and only nontransgenic littermates were used as controls. All animal experiments were approved by the Lund–Malmö laboratory animal ethics committee.

CII immunization and arthritis evaluation.

For arthritis experiments, mice were immunized at the base of the tail with 100 μg of CII emulsified in 100 μl Freund's complete adjuvant (CFA; Difco). Mice were boosted 5 weeks later with 50 μg of CII in 50 μl Freund's incomplete adjuvant (Difco). Arthritis development was followed through visual scoring 3 times a week, starting 2 weeks after the primary immunization. Arthritis was evaluated using an extended protocol (23); each paw received a score of 1–15, with a maximum score of 60 per mouse.

Antibody responses.

Blood samples were collected 5 and 10 weeks after primary immunization, and isolated sera were investigated for total anti-CII IgG, IgG1, IgG2b, and IgG2a/c levels through quantitative enzyme-linked immunosorbent assay (ELISA) as described previously (24), where serum was titrated (1:10–1:106) in parallel to the standard, and titer values were interpolated within the linear range and related to the standard curve. Biotinylated rat anti-mouse IgGκ (prepared in-house; clone 187.1) or mouse anti-mouse IgG2c (BD Biosciences) or peroxidase-conjugated goat anti-mouse antibodies specific for IgG1, IgG2a, or IgG2b (Southern Biotechnology) were used as detecting antibodies. Binding of biotinylated antibodies was revealed by ExtrAvidin Peroxidase (Sigma-Aldrich). Of note, mice of the C3H and B10 background express different, but related, IgG2 alleles: IgG2a and IgG2c, respectively (25). Plates were developed using ABTS (Roche Diagnostic Systems) as substrate and measured at 405 nm (Synergy-2; BioTek Instruments). Total anti-CII IgG levels were measured as μg/ml using purified polyclonal anti-CII IgG antibodies of a known concentration as a standard. Isotype levels were measured as arbitrary concentrations using our purified monoclonal anti-CII antibodies of the IgG1 (clone CB20), IgG2a (clone CII-CI), and IgG2b (clone M2139) isotypes or pooled sera from arthritic mice for the IgG2c ELISA.

Cellular assays.

Mice were immunized with CII in CFA (3 injections of 60 μg at 3 different locations around the base of the tail), and 10 days later cells were prepared from the draining lymph nodes and restimulated with CII peptides in vitro in flat-bottomed 96-well plates (700,000 cells/well) for 96 hours, whereupon the culture supernatants were assayed for interferon-γ (IFNγ) and interleukin-17 (IL-17) content by ELISA. Cells were cultured in Dulbecco's modified Eagle's medium plus Glutamax I (Gibco) supplemented with 5% heat-inactivated fetal calf serum and penicillin/streptomycin. For IFNγ detection, R46A2 (5 μg/ml) and biotin-conjugated AN18.17.24 (0.6 μg/ml; Mabtech) were used as capture and detection antibodies, respectively. For IL-17 detection, TC11-18H10 (1 μg/ml; BD Biosciences) and biotin-conjugated TC11-8H4.1 (0.5 μg/ml; BD Biosciences) were used as capture and detection antibodies, respectively. Binding of biotinylated antibodies was revealed by europium-labeled streptavidin, and plates (Costar) were analyzed using a Victor 1420 multilabel counter (PerkinElmer). Cytokine levels were measured as arbitrary concentrations using supernatant from concanavalin A (Con A)–stimulated B10.Q mouse splenocytes as standard.

For enzyme-linked immunospot (ELISpot) assays, plates (Millipore) were used according to the manufacturer's instructions, using the same capture and detection antibodies as for the ELISA, although at higher concentrations (10 μg/ml and 2 μg/ml, respectively, for IFNγ and 5 μg/ml and 1 μg/ml, respectively, for IL-17A). Five hundred thousand cells per well were incubated with antigen for 24 hours, and cytokine spots were visualized using Sigma Fast BCIP/nitroblue tetrazolium and subsequently enumerated using an ImmunoScan ELISpot Analyzer (CTL Europe). Establishment of T cell hybridoma clones and determination of antigen specificity were performed essentially as described previously (7) by immunization of mice with CII in CFA followed 10 days later by in vitro restimulation of lymph node cells with lathyritic rat CII (pepsin free) before fusing the cells with the T cell receptor (TCR)–negative BW5147 thymoma cell line variant (26).

MHC peptide binding assay.

Peptide binding assays were performed by measuring the ability of synthetic peptides to inhibit the binding of biotinylated CLIP peptide to purified DR4 molecules as described (27). Briefly, a mixture of a fixed concentration of purified soluble recombinant DR4 molecules (0.1 μM), biotinylated CLIP peptide (2.5 μM), and increasing concentrations of competitor peptides was incubated for 48 hours at room temperature in the presence of a protease inhibitor cocktail (Complete; Roche Diagnostics). Recombinant DR4 molecules were then captured on microtiter assay plates that had been precoated with L243 monoclonal antibody and blocked with phosphate buffered saline (PBS) containing 2% low-fat milk. To remove the excess of nonbound peptides, plates were washed with PBS, and the amount of biotinylated CLIP peptide bound to the recombinant DR4 molecules was measured with europium-labeled streptavidin.


B10 and C3H backgrounds have different influences on arthritis susceptibility in mice displaying T cell tolerance to self CII.

To study T cell tolerance to self CII and its impact on arthritis susceptibility in DR4-expressing mice, we backcrossed DR4 mice to MMC-transgenic mice on the B10 background (yielding MMC.B10.DR4 mice) and subsequently immunized these mice with CII. MMC-transgenic mice express a mutated mouse CII protein in joint cartilage that differs from the CII protein in their nontransgenic littermates at a single amino acid within the DR4-restricted immunodominant T cell epitope (D266 in mouse CII versus E266 in heterologous CII). Hence, the MMC transgene allows CII-specific T cells to interact with self CII and acquire tolerance to the heterologous T cell epitope in the naive mouse. However, it was not possible to address tolerance to self CII in these mice, since both MMC.B10.DR4 mice and B10.DR4 mice were found to be highly resistant to CIA, with only few mice developing very mild arthritis (Table 1 and Figure 1A).

Table 1.  CIA susceptibility in DR4 mice on different genetic backgrounds*
Strain, geneArthritis, no. with/no. without (%)Maximum score, mean ± SEMDay of onset, mean ± SEM
  • *

    Collagen-induced arthritis (CIA) susceptibility in DR4 mice expressing either the wild-type or mutated Ncf1 gene (Ncf1*) and the human type II collagen (CII) transgene (HuCII) or the MMC transgene (MMC). Lm = transgenic-negative littermates.

  • Includes all mice that developed arthritis among all animals throughout the experiment.

  • Arthritic mice only.

  • §

    P < 0.05 versus nontolerized littermates, by Fisher's exact test.

  • P < 0.05 versus nontolerized littermates, by Mann-Whitney U test.

 Lm3/20 (15)2 ± 144 ± 3
 MMC1/20 (5)142
 Lm6/11 (55)34 ± 1045 ± 3
 HuCII0/11 (0)§
 Lm25/34 (74)15 ± 338 ± 2
 MMC16/29 (55)7 ± 244 ± 3
 Lm6/11 (55)19 ± 645 ± 6
 MMC3/15 (20)8 ± 266 ± 4
Figure 1.

Development of autoimmune collagen-induced arthritis depends on transgenic element, reactive oxygen species production, and genetic background. Mice with either a functional Ncf1 genotype (A and B) or a mutated Ncf1 (Ncf1*) genotype (C and D) and expressing either the MMC transgene (AC) or the HuCII transgene (D) were immunized with rat type II collagen (CII) in Freund's complete adjuvant on day 0 and boosted 5 weeks later with rat CII in Freund's incomplete adjuvant and followed up for arthritis development. HuCII- and MMC-transgenic mice were bred in a heterozygotic manner, and non–HuCII/MMC-transgenic littermates (Lm) were used as controls. Graphs show the severity of arthritis and represent only mice that had developed disease before the termination of the experiment (arthritic animals only [AAO]). Values are the mean and SEM. Numbers in parentheses are the number of arthritic mice/total number of mice.

We have previously reported that MMC-expressing B10.Q mice are completely protected from CIA (28), while MMC-derived tolerance to self CII was only partial in MMC.C3H.Q mice. We therefore backcrossed DR4 mice to the C3H background. In contrast to B10.DR4 mice, >50% of the C3H.DR4 mice developed severe arthritis after immunization with heterologous CII (Table 1 and Figure 1B). Notably, 3 weeks later, arthritis also became apparent in 20% of the MMC.C3H.DR4 mice. Hence, in contrast to B10.DR4 mice, C3H.DR4 mice are readily susceptible to CIA, and the C3H background therefore appears to be superior for investigating T cell tolerance to self CII in MMC-expressing mice.

Introduction of a single autoimmunity-promoting gene allows arthritis resistance to be broken in MMC.B10.DR4 mice, but not in HuCII.B10.DR4 mice.

Many genetically modified animals are available on the 2 related C57BL/6 and C57BL/10 backgrounds. It would therefore be advantageous to establish a CII-tolerized animal model on the C57BL background to avoid time-consuming backcrossing of the gene of interest onto the C3H background. To determine whether it would be possible to overcome arthritis resistance in B10.DR4 and MMC.B10.DR4 mice, we introduced the previously identified autoimmunity-promoting Ncf1 mutation in these mice to yield B10.DR4.Ncf1* mice. This mutation results in a truncated NCF-1 protein and a lower production of reactive oxygen species (ROS) and surprisingly leads to an increased susceptibility to CIA (22). Indeed, the Ncf1 mutation had a pronounced autoimmunity-promoting effect when introduced to yield B10.DR4.Ncf1* mice, resulting in severe arthritis in 74% of the mice (Table 1 and Figure 1C). Most importantly, introduction of the Ncf1 mutation in MMC.B10.DR4 mice (to yield MMC.B10.DR4.Ncf1* mice) could break tolerance and resulted in a penetrant (55%) but relatively mild arthritis compared with B10.DR4.Ncf1* mice (P = 0.0346).

We previously demonstrated T cell tolerance in HuCII-transgenic B10.DR4 mice (12) but were unable to induce arthritis in these mice. We therefore tested whether the presence of the autoimmunity-promoting Ncf1 mutation would also allow arthritis to occur in these mice. However, HuCII.B10.DR4.Ncf1* mice, which express the entire HuCII protein as a self antigen in joint cartilage, were found to be completely resistant to CIA upon immunization with CII (Table 1 and Figure 1D). Hence, the HuCII transgene mediates strong protection from CIA that, in contrast to the MMC transgene, cannot be subdued by deficient Ncf1 expression.

Development of arthritis in MMC-expressing mice is associated with an increased IgG2 response.

Five weeks after immunization, all mice expressing the MMC transgene had significantly lower antibody levels compared with those in littermate controls (Figures 2A–C). For MMC.B10.DR4.Ncf1* mice, this was also confirmed for all IgG subclasses (IgG1, IgG2c, and IgG2b; data not shown for IgG1 and IgG2b). MMC.C3H.DR4 mice displayed a significantly increased antibody response on day 70 as compared with day 35, reaching levels indistinguishable from those in C3H.DR4 littermates (Figure 2C). Despite the striking arthritis-promoting effect of the Ncf1 mutation, only a modest increase in anti-CII levels was observed in B10.DR4.Ncf1* mice compared with B10.DR4 mice (Figures 2A and B), and at 10 weeks, antibody levels had declined in both MMC.B10.DR4.Ncf1* mice and MMC-negative littermates (Figure 2B).

Figure 2.

Serum anti-CII antibody levels in non–MMC-expressing and MMC-expressing DR4 mice on different genetic backgrounds. Total anti-CII IgG levels (AC) and IgG2a/c levels (D) were compared between MMC-transgenic mice and nontransgenic littermates at 5 weeks (AD) and 10 weeks (B and C) after primary immunization with CII. Open symbols represent healthy animals; solid symbols represent animals that developed arthritis during the experiment (see Figure 1 and Table 1). Horizontal lines in AC represent the mean. In D, IgG2a/c levels stratified according to clinical outcome (H = healthy; A = arthritis) are shown as box plots, where the boxes represent the 25th to 75th percentiles, the lines within the boxes represent the median, and the lines outside the boxes represent the minimum and maximum values. P values were obtained using the Mann-Whitney U test. See Figure 1 for other definitions.

Stratification of antibody levels according to the outcome of arthritis showed no significant difference in non–MMC-expressing DR4 mice. However, MMC.B10.DR4.Ncf1* and MMC.C3H.DR4 mice affected by arthritis displayed significantly increased levels of the IgG2a/IgG2c subclass as compared with healthy mice (Figure 2D). Arthritic MMC.C3H.DR4 mice also had significantly higher IgG1 levels (P = 0.0364) (data not shown) and total IgG levels (P = 0.0173) (Figure 2C) as compared with unaffected MMC.C3H.DR4 mice on day 35. Thus, arthritis susceptibility is correlated with an increased anti-CII IgG2a/c response in both MMC.C3H.DR4 and MMC.B10.DR4.Ncf1* mice. However, only in MMC.C3H.DR4 mice do antibody levels increase and reach levels comparable with those in their nontransgenic littermates around the same time that clinical arthritis develops.

Autoreactive T cells in MMC-expressing DR4 mice are specific for the nonglycosylated CII peptide.

We next investigated CII-specific recall responses in MMC.DR4 mice and nontransgenic littermates of different genetic backgrounds. Irrespective of the background, nontransgenic littermates mounted the strongest IL-17 and IFNγ responses against the nonglycosylated CII peptide, as determined by ELISpot assay (Figure 3) and ELISA (data not shown). These mice also produced a significant, although much weaker, response against the Gal264 peptide. One striking feature observed was the almost deficient IL-17 response in mice of the C3H background (Figure 3C), which was also confirmed after stimulation with PPD and the mitogen Con A (data not shown).

Figure 3.

In vitro recall response against the CII259–273 peptides in MMC-transgenic or HuCII-transgenic DR4 mice and in nontransgenic littermates, as determined by enzyme-linked immunospot assays for interferon-γ (IFNγ) and interleukin-17 (IL-17). The indicated numbers of mice were immunized 10 days previously with rat CII in Freund's complete adjuvant before lymph node cells (LNCs) were restimulated in vitro with the unmodified CII peptide (K264) or with the CII peptide with a galactose moiety on hydroxylysine either at position 264 (Gal264) or at both position 264 and position 270 (Gal264/270). Values are the mean and SEM number of spots recorded. ∗ = P < 0.05; ∗∗ = P < 0.01 versus unstimulated cells (Bg), by Mann-Whitney U test. See Figure 1 for other definitions.

T cell tolerance to the CII259–273 epitope was evident in all MMC-expressing mice. However, although it was strongly reduced compared with that in their nontransgenic littermates, MMC mice still mounted a significant recall response to the nonglycosylated CII peptide. Again, despite the arthritis-promoting effect of the Ncf1 mutation, only minimal differences were observed in IL-17 responses between MMC.B10.DR4 and MMC.B10.DR4.Ncf1* mice. The immunodominance of the nonglycosylated CII peptide in MMC mice initially surprised us, since we have previously reported an almost completely deficient T cell response in HuCII-transgenic DR4 mice (12). However, by investigating the recall response in CII-primed HuCII.B10.DR4.Ncf1* mice, we could confirm our previous data, since HuCII.B10.DR4.Ncf1* mice also failed to mount a significant response to both nonglycosylated and glycosylated self CII peptides (Figure 3D).

Collectively, these data show that the nonglycosylated CII peptide is immunodominant in DR4-expressing mice and that the choice of transgenic element strongly affects the level of achieved T cell tolerance to self CII. Moreover, tolerance to self CII in MMC-expressing DR4 mice is strong but incomplete, and the remaining autoreactive T cells are exclusively specific for the nonglycosylated CII peptide and appear to be critical for autoimmune arthritis to develop in these mice.

T cell recognition of the nonglycosylated CII259–273 epitope is heterogeneous, but K264 is a critical TCR contact point for the majority of responding cells.

Both DR4 and Aq molecules present the CII259–273 peptide to CII-specific T cells. However, while the glycosylated CII peptide is immunodominant in Aq mice, our present data show that the nonglycosylated peptide dominates in DR4 mice. To investigate the reason for this discrepancy, we wanted to explore to what extent K264 and K270 would influence T cell recognition of the heterologous CII259–273 peptide. We therefore generated T cell clones from CII-immunized B10.DR4.Ncf1* mice in order to test these against CII peptides in which K264 and/or K270 had been substituted for arginine (R), since this has been found to be the preferred amino acid at position P2 for peptides binding to DRB1*0401 (29). We also performed binding studies, which confirmed that substitutions of lysine for arginine residues had minimal consequences for class II MHC peptide binding (Figure 4A). Similarly, binding studies of nonglycosylated, hydroxylated, and glycosylated CII peptides showed that all posttranslationally modified peptides, with the possible exception of Gal264/270, bound with affinity comparable to that of the unmodified CII259–273 peptide (Figure 4B).

Figure 4.

HLA–DR4 binding, clonotypic T cell recognition, and DR4-restricted antigenicity of wild-type and modified type II collagen 259–273 (CII259–273) peptides. A and B, Inhibition of the biotinylated CLIP peptide in binding to soluble HLA–DR4 molecules with arginine-substituted peptides at position 264 (K264R) or position 270 (K270R) or position 264 and position 270 (K264/270R) (A) or with CII259–273 peptides harboring a hydroxylysine at position 264 (HyK264) or with a galactose moiety on hydroxylysine either at position 264 (Gal264) or at both position 264 and position 270 (Gal264/270) (B). The binding of the nonbiotinylated CLIP peptide as well as that of the unmodified CII259–273 peptide (K264/270) is included as a reference. C, Specific recognition of modified CII259–273 variants by the DR4-restricted T cell hybridoma clone 7C4H1 (group IV in Table 2), as determined by interleukin-2 (IL-2) production. D, Investigation of antigenicity of selected variants of the CII259–273 peptides and of cross-reactivity between these variants in B10.DR4.Ncf1* mice by immunization with the wild-type (K264/270)– or arginine-substituted versions of the peptide, followed by in vitro restimulation 10 days later with the same peptides as well as with the Gal264 peptide at 10 μg/ml. After 4 days of stimulation, supernatants were collected and investigated for interferon-γ (IFNγ) content. Values are the mean and SEM. Bg = unstimulated cells.

All T cell clones tested responded to the nonglycosylated K264 peptide. However, these could be divided into 4 groups according to their stringency in terms of additional peptides recognized (Table 2). Three of 4 groups were highly dependent on K264, since the subtle alteration from K264 to HyK264 resulted either in an abolished (group I) or in a reduced (group II–III) response, which could not be attributed to a reduced binding of the HyK264 peptide to the DR4 molecule (Figure 4B). In contrast, only 1 group (group IV) displayed a more promiscuous pattern, with responses to the K264R and K264/270R peptides as well as to the corresponding mouse CII peptide (E266D) (Figure 4C), which has low binding affinity to the DR4 molecule (4, 30).

Table 2. T cell hybridoma responses to modified type II collagen 259–273 (CII259–273) peptides*
  • *

    Shown is semiquantitative scoring of the T cell hybridoma response following stimulation with the wild-type heterologous nonglycosylated CII259–273 collagen peptide (K264) or with modified versions thereof, in which lysine residues at position 264, position 270, or both positions had been substituted for arginines (R), hydroxylysine (HyK), or a galactosylated hydroxylysine (Gal), as well as with the corresponding mouse peptide (E266D). Clones are categorized into 4 groups, where each group signifies a unique response pattern in terms of peptides recognized. Sensitivity of the T cell hybridomas was determined by the amount of peptide required for the induction of interleukin-2 production above background levels: − = no response; + = ≥50 μg/ml; ++ = ≥10 μg/ml; +++ = ≥2 μg/ml; ++++ = ≥0.4 μg/ml; +++++ = ≥0.08 μg/ml.

No. of clones2132
E266D (mouse)+++

Importantly, galactosylation of K264 (Gal264) resulted in a completely abrogated response, even for clones belonging to group IV and which appeared to be less dependent on K264 for their activation. Similarly, galactosylation of K270 (Gal270) inhibited or severely impaired the response in 2 of the 4 groups, even though all clones could respond to the K270R peptide. This suggests that the galactose moiety on either hydroxylysine residue could potentially act as a steric hindrance for T cells that are more dependent on residues other than either of the 2 lysine residues for their activation.

The preferential recognition of K264 over K270 was also confirmed at the polyclonal level (Figure 4D), where immunization of DR4 mice with either the wild-type CII259–273 peptide or the K270R peptide resulted in an almost identical recall response against both peptides. In contrast, immunization with K264-modified peptides (K264R or K264/270R) resulted in a strongly reduced recall response against the wild-type CII259–273 peptide. Finally, no recall response was detected against the Gal264 peptide, irrespective of the immunizing peptide. Thus, DR4-restricted T cell recognition of the CII259–273 epitope is heterogeneous, but the majority of T cells specific for the nonglycosylated CII peptide are critically dependent on an unmodified K264 for their activation.


The first objective of the current investigation was to establish an animal model in DR4 mice in which T cell tolerance to self CII could be broken. Classic CIA in Aq mice is induced by immunization with CII of non-mouse origin, and the developing arthritis is highly dependent on activation of T cells specific for the heterologous CII259–273 peptide. Therefore, to study tolerance to self CII, we used the MMC-transgenic mouse, which expresses the immunodominant T cell epitope of heterologous CII in a mutated mouse CII protein. Because of this point mutation (D→E at position 266), only autoreactive T cells will become activated in MMC mice upon immunization with rat CII, and differences observed between MMC and control mice will be either directly or indirectly related to T cell tolerance.

We show herein that T cell tolerance is clearly present, since the recall response to rat CII is strongly reduced in MMC-transgenic DR4 mice. However, T cell tolerance is not complete, since MMC mice can still mount a significant recall response. Furthermore, we show that incomplete T cell tolerance to self CII may render mice susceptible to development of CIA, but this is highly dependent on the genetic background, since MMC.C3H.DR4 mice, but not MMC.B10.DR4 mice, are susceptible. Still, by introducing the Ncf1 mutation, it is possible to increase susceptibility even in MMC mice on the B10 background. The exact mechanism by which ROS regulates autoreactive T cells in the CIA model is currently unknown, but it is believed to involve oxidation by antigen-presenting cells of the cell membrane and its associated proteins on interacting T cells (for review, see ref.31). In the absence of a normal ROS production, the threshold for activating autoreactive T cells may be lower and/or cells will survive longer.

We observed a deficient IL-17 response in MMC.C3H.DR4 mice as compared with both MMC.B10.DR4 and MMC.B10.DR4.Ncf1* mice. Our C3H mice are Toll-like receptor 4 competent (32), and the weak IL-17 response is probably an intrinsic feature of the C3H background and has also been reported previously in other models (33). Still, the reduced IL-17 response in C3H mice is unlikely to explain the difference in arthritis susceptibility between B10.DR4 and C3H.DR4 mice, since IL-17 responses are generally assumed to promote autoimmunity.

MMC.C3H.DR4 mice also displayed an altered anti-CII antibody response as compared with MMC.B10.DR4.Ncf1* mice, where the anti-CII IgG response increased over time in MMC.C3H.DR4 mice to levels indistinguishable from those in C3H.DR4 littermate controls. Meanwhile, antibody levels tended to decrease in MMC.B10.DR4.Ncf1* mice. The MMC-derived reduction in anti-CII antibody responses is not likely explained by B cell tolerance to self CII, since B cell epitopes are identical between MMC mice and littermate controls. In fact, most anti-CII antibodies in both rats and mice are germline encoded (34–36). They also cross-react with CII of other species, but not with other collagen types, such as type IV or type I collagen. Instead, low-affinity CII-specific T cells that have escaped tolerance or that have acquired tolerance but have not been deleted become sufficiently activated to help B cells to produce anti-CII antibodies. However, such T cells will be less frequent and less efficient in supplying B cell help, thereby delaying the anti-CII antibody response in reaching arthritogenic levels.

A stronger anti-CII IgG response was also observed in nontransgenic C3H.DR4 mice as compared with B10.DR4 and B10.DR4.Ncf1* mice. This has also been observed in the C3H.Q strain (28) and probably reflects an intrinsic phenotype of the genetic background. It could be speculated that introduction of the Ncf1 mutation will lead to a more efficient T cell response, possibly via IL-17, that could stimulate the anti-CII B cell response to reach arthritogenic levels earlier in MMC.B10.DR4.Ncf1* mice compared with MMC.C3H.DR4 mice. Alternatively, the Ncf1 mutation may lower the critical level of arthritogenic antibodies required for arthritis to develop. Such a hypothesis can now be investigated using these models.

The source of the CII-transgenic element had a dramatic effect on the robustness of tolerance, where the HuCII transgene induced considerably stronger T cell tolerance to the CII259–273 peptide than did the MMC transgene. This difference probably relates to the expression levels and availability of the transgenic CII for tolerance induction in vivo. This has also been shown for other model autoantigens (37, 38) and may be due to differences in transgenic promoter expression or number of transgene copies inserted. In fact, by comparing different lines (founders) of the HuCII transgene in Aq mice, an expression level–dependent T cell tolerance to the CII259–273 epitope could be observed (18).

A second objective of the study was to investigate the extent to which T cells recognize posttranslational modifications of CII in DR4 mice. We show that the nonglycosylated CII259–273 epitope is clearly immunodominant in both MMC.DR4 mice and littermate controls. This is in strong contrast to Aq-expressing mice, in which the galactosylated peptide dominates. Our earlier data show that the K264 residue within the immunodominant CII259–273 epitope is uniformly O-glycosylated when CII is isolated from healthy cartilage of humans and rats (10, 11). Thus, T cell tolerance should primarily affect T cells specific for glycosylated CII if it is regulated through peripheral presentation of self CII derived from joint cartilage, as suggested previously (39).

There is also evidence for ectopic expression of CII in the thymus (40, 41), but it is not known whether this includes presentation of either nonglycosylated or glycosylated CII peptides or both. Biased T cell tolerance against the nonglycosylated T cell epitope in MMC-expressing mice of the H-2q haplotype could be explained by a sufficient presentation of this peptide, leading to strong central tolerance. In contrast, peripheral tolerization of glycopeptide-specific T cells may be less efficient, possibly as a result of an inferior processing and presentation of the glycosylated CII peptide by macrophages and dendritic cells, as has been shown in DR1-transgenic mice (42, 43). It is reasonable to assume that the MMC transgene operates similarly in both DR4- and Aq-expressing mice. However, nonphysiologic expression (low or toxic) of the DR4 transgene could influence the selection of the nonglycosylated CII peptide as immunodominant in these mice, perhaps due to a reduced presentation of the nonglycosylated CII peptide within the thymus.

The Aq and DR4 molecules present the CII259–273 peptide in a slightly different manner, in which residues I260 and F263 bind to the P1 and P4 pockets, respectively, of Aq and in which residues F263 and E266 bind to the P1 and P4 pockets, respectively, of DR4. Hence, the galactose moiety of K264 will be more centrally located (i.e., P5) on the Aq molecule than on the DR4 molecule (P2). Meanwhile, the lysine at position 270 will be shifted toward the center (P8) on DR4, as compared with its peripheral placement (P11) on Aq. The K264 residue clearly constitutes a critical TCR contact point for Aq-restricted CII-specific T cells (44, 45). However, the importance of the K264 residue as a DR4-restricted TCR contact or class II MHC anchor point has been questioned, whereas K270 should be exposed toward the TCR (2, 4, 46, 47).

It could be speculated that the nonglycosylated CII peptide is immunodominant in DR4 mice because the majority of CII-specific T cells do not rely on the K264 residue, but rather, on residues located between P3 and P7 of the peptide–DR4 complex. However, our data show that this is not the case and that in fact K264 appears to be more important than the K270 residue for interaction with the TCR. Instead, our data suggest the possibility that the physiologic and uniform glycosylation of self CII at K264 (10, 11) could act to sterically hinder the acquisition of tolerance by T cells specific for the nonglycosylated epitope. If so, then traumatic alterations in the level of glycosylation of self CII could lead to peripheral presentation of nonglycosylated CII peptides. These could then act as neoepitopes and promote autoimmunity toward joint-specific antigens in an inflamed environment.


All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. J. Bäcklund had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Batsalova, Dzhambazov, Merky, J. Bäcklund.

Acquisition of data. Batsalova, Dzhambazov, Merky, A. Bäcklund, J. Bäcklund.

Analysis and interpretation of data. Batsalova, Dzhambazov, Merky, A. Bäcklund, J. Bäcklund.

Project supervision. J. Bäcklund.


We thank Rikard Holmdahl for support and discussions enabling the study and for critically reading the manuscript. We also thank Carlos and Kristina Palestro, Johanna Ekelund, and Isabella Bohlin for taking care of the animals, Emma Mondoc and Malin Neptin for CII preparation, and Katrin Dilja Jonsdottir and Erik Jansson for technical support of cell culture assays.