T cell epitope spreading to myelin oligodendrocyte glycoprotein in HLA-DR4 transgenic mice during experimental autoimmune encephalomyelitis

Clinical Immunology 111 (2004) 53 – 60 www.elsevier.com/locate/yclim

T cell epitope spreading to myelin oligodendrocyte glycoprotein in HLA-DR4 transgenic mice during experimental autoimmune encephalomyelitis Juliane Klehmet, a Carey Shive, a Rocio Guardia-Wolff, a Ines Petersen, b Edward G. Spack, c Bernhard O. Boehm, d Robert Weissert, b and Thomas G. Forsthuber a,* a

Institute of Pathology, School of Medicine, Case Western Reserve University, Cleveland, OH 44106, USA b Department Neurology, University of Tu¨bingen, Tu¨bingen, Germany c InterMune, Brisbane, CA 94005, USA d Division of Endocrinology, University Hospital of Ulm, Ulm, Germany Received 11 July 2003; accepted with revision 30 December 2003

Abstract Epitope spreading has been implicated in the pathogenesis of experimental autoimmune encephalomyelitis (EAE) and human multiple sclerosis (MS). T cell epitope spreading has been demonstrated in rodents for myelin basic protein (MBP) and proteolipid protein (PLP) determinants, but not for myelin oligodendrocyte glycoprotein (MOG), another important myelin antigen. Moreover, the role of human autoimmunity-associated MHC molecules in epitope spreading, including HLA-DR2 and DR4, has not been formally examined. To address these questions, we investigated epitope spreading to MOG determinants in HLA-DR4 (DRB1*0401) transgenic mice during EAE. The data show that upon induction of EAE in HLA-DR4 transgenic mice with the immunodominant HLA-DR4-restricted MOG peptide 97 – 108 (MOG97 – 108; TCFFRDHSYQEE), the T cell response diversifies over time to MOG181 – 200 (core: MOG183 – 191; FVIVPVLGP) and MBP. The spreading epitope MOG181 – 200 binds with high affinity to HLA-DRB1*0401 and is presented by human HLADRB1*0401+antigen presenting cells. Moreover, this epitope is encephalitogenic in HLA-DRB1*0401 transgenic mice. This study demonstrates intra- and intermolecular epitope spreading to MOG and MBP in ‘‘humanized’’ HLA-DR4 transgenic mice. D 2004 Published by Elsevier Inc. Keywords: Oligodendrocyte; Encephalomyelitis; Glycoprotein

Introduction Activation of CD4+ T cells is dependent on the recognition of peptides presented by MHC class II molecules on antigen presenting cells (APCs). The set of peptides recognized by T cells in the context of particular MHC class II molecules is usually limited and remains constant over time for foreign antigens [1]. However, the T cell response to selfantigens is dynamic and additional T cell epitopes may become recognized over the course of autoimmune diseases

Abbreviations: MOG, Myelin oligodendrocyte glycoprotein; MS, multiple sclerosis; MBP, myelin basic protein; CNS, central nervous system; EAE, experimental autoimmune encephalomyelitis. * Corresponding author. Institute of Pathology, Case Western Reserve University, Biomedical Research Building 936, 2109 Adelbert Road, Cleveland, OH 44106-4943. Fax: +1-216-368-1357. E-mail address: [email protected] (T.G. Forsthuber). 1521-6616/$ - see front matter D 2004 Published by Elsevier Inc. doi:10.1016/j.clim.2003.12.012

such as Type 1 diabetes or experimental autoimmune encephalomyelitis (EAE). This phenomenon has been termed epitope spreading [1,2]. T cell epitope spreading during EAE has been demonstrated for proteolipid protein (PLP) and myelin basic protein (MBP) [1,2], whereas to date, the diversification of T cell immunity to MOG has not been reported. The clinical significance of epitope spreading has remained unresolved. In animal models of autoimmunity, engagement of new T cell epitopes of self-antigens was associated with relapses of EAE [2,3], whereas other reports failed to detect a role for epitope spreading in progression and relapses of this disease [4]. Epitope spreading has also been reported in human patients with autoimmune diseases [5 –7]. However, T cell responses in patients are usually measured in the peripheral blood, and the fluctuations of T cell responses to myelin epitopes may alternatively reflect the shifting of autoreac-

54

J. Klehmet et al. / Clinical Immunology 111 (2004) 53–60

tive T cells from peripheral tissues back to the vascular circulation. To overcome the limitations of human studies, we have taken advantage of ‘‘humanized’’ transgenic mice expressing HLA-DR4 (DRB1*0401) molecules in the absence of endogenous mouse MHC class II [8]. These transgenic mice permit the identification of immunodominant peptides presented by a single, autoimmunity-associated HLA-DR allele [9 –11]. Previously, we showed that MOG97 – 108 was the immunodominant HLA-DR4-restricted MOG epitope in these transgenic mice, and that immunization with this peptide induced EAE [12]. In the present study, we used this system to investigate epitope spreading to MOG restricted by the human HLA-DRB1*0401 molecule. The results show that the T cell response in MOG97 – 108immunized HLA-DR4 transgenic mice was initially restricted to MOG97 – 108. However, over time, we found strong T cell immunity to a second MOG epitope, MOG181 – 200, and to MBP in the transgenic mice. Together, the data demonstrate intra- and intermolecular epitope spreading induced by MOG in HLA-DR4 transgenic mice.

Materials and methods Mice, antigens, and treatments HLA-DR4 (DRB1*0401) transgenic mice were generated as described previously [8] and bred at CWRU under specific pathogen-free conditions. All animal procedures were conducted according to guidelines of the Institutional Care and Use Committee (IACUC) of CWRU. Female transgenic mice were injected at 6 –10 weeks of age with MOG97 – 108 (TCFFRDHSYQEE) in complete Freund’s adjuvant (CFA). 200 ng Pertussis toxin (PT; List Biological Laboratories, Campbell, CA) was injected ip at 0 and 24 h after the immunization. Overlapping MOG peptides were obtained from Princeton Biomolecules (Langhorne, PA). CFA was prepared by mixing incomplete Freund’s adjuvant (IFA; Gibco BRL, Grand Island, NY) with Mycobacterium tuberculosis H37RA at 5 mg/ml (Difco Laboratories, Detroit, MI). MBP was prepared as described [13]. Antigens were mixed with the adjuvant to yield a 2 mg/ml emulsion, of which 50 Al was injected sc as specified. Cell preparations and T cell separation Single cell suspensions were prepared from HLA-DR4 popliteal lymph node or spleen cells as previously described [12] and plated in HL-1 serum-free medium (BioWhittaker, Walkersville, MD), together with antigen. CD4+ T cells were obtained by passing the cells through a murine CD4+ T cell enrichment column (R&D Systems, Minneapolis, MN) following the manufacturer’s suggested protocol. FACS analysis showed more than 95% enrichment for CD4+ T cells.

EBV-transfected HLA-DR4 (DRB1*0401) homozygous B cells [14] were added at 1  105 cells per well as previously described [12]. Cytokine measurements by ELISPOT and computer-assisted ELISPOT image analysis Cytokine ELISPOT assays were performed as described [15]. ELISPOT plates (ImmunoSpot, Cellular Technology, Cleveland, OH) were coated overnight with IFN-g-specific capture antibody (R46A2, 4 Ag/ml) diluted in 1 PBS. The plates were blocked with 1% BSA in PBS, for 1 h at room temperature, then washed four times with PBS. Cells from draining lymph nodes were plated at 5  105 cells/well alone or with MOG peptides (7 AM) in HL-1 medium supplemented with 1% L-glutamine and cultured for 24 h. Subsequently, the cells were removed by washing 4 with PBS then 4 with PBS/Tween and the biotinylated detection antibody XMG1.2-biotin (2 Ag/ml) for IFN-g was added and incubated overnight. The plate-bound second antibody was then visualized by adding streptavidin-alkaline phosphatase (SAV-AP, Dako, Carpinteria, CA) and NBT/BCIP substrate (Biorad, Hercules, CA/Sigma, St. Louis, MO). Image analysis of ELISPOT assays was performed on a Series 1 ImmunoSpotk Image Analyzer (Cellular Technology) as described previously [15,16]. In brief, digitized images of individual wells of the ELISPOT plates were analyzed for cytokine spots based on the comparison of experimental wells (containing T cells and APC with antigen) and control wells (T cells and APC, no antigen). After separation of spots that touched or partially overlapped, nonspecific noise was gated out by applying spot size and circularity analysis as additional criteria. Spots that fell within the accepted criteria were highlighted and counted. Stimulation index (SI) was calculated by dividing the number of cytokine spots detected in wells pulsed with relevant antigen by the number of cytokine spots in wells without antigen (medium only). The spot number in unimmunized or control mice (irrelevant antigen) was in the same range as the medium controls shown. Evaluation of clinical disease Mice were monitored daily for 30 days, and on alternate days thereafter. A mean clinical score was assigned for each group using the following scale [17]: 0, no abnormality; 1, limp tail; 2, moderate hind limb weakness; 3, complete hind limb paralysis; 4, quadriplegia or premoribund state; 5, death. Binding predictions of MOG sequences for HLA-DRB1*0401 Potential core peptides with high binding affinity for HLA-DRB1*0401 were identified as described previously [12,18]. Briefly, a computer program was written that parsed the MOG sequence into successive 9-mers, each

J. Klehmet et al. / Clinical Immunology 111 (2004) 53–60

beginning one amino acid after the start of the previous 9mer. The contribution towards binding of each amino acid was summed based on the matrix published by Hammer et al. [18], yielding a peptide score. Higher peptide scores indicate relatively higher affinity binding to HLA-DRB1* 0401. Several immunodominant peptides documented in the literature yielded relative affinity scores in the range of 4 –6 [18].

55

nM) were incubated with fixed amounts of the tracer peptide (CLIP97 – 120) in the presence of various concentrations (10fold dilutions between 1 nM and 100 AM) of the unlabeled MOG peptides. The concentration yielding 50% inhibition of binding of the tracer peptide (IC50) was measured by plotting the percentage of inhibition versus the concentration of MOG peptide. Peptides were tested in two independent experiments.

In vitro peptide binding studies Results In vitro binding competition assays were performed as described previously [12,19]. In brief, biotinylated tracer peptides were used in a direct binding assay to establish optimal binding conditions for the purified HLA-DR4 (DRB1*0401) molecules. Relative affinities of MOG peptides for the HLA-DR4 molecules were assessed by an inhibition ELISA assay based on a dissociation-enhanced lanthanide fluoro-immunoassay (DELFIA; Wallac, Turku, Finland). In the inhibition assay, HLA-DR4 molecules (10

Intra- and intermolecular epitope spreading after induction of EAE with MOG97 – 108 in HLA-DR4 transgenic mice Previously, we showed that MOG97 – 108 was the immunodominant extracellular MOG epitope in HLA-DR4 (DRB1*0401) transgenic mice [12]. To investigate if epitope spreading occurs in the response of HLA-DR4 transgenic mice immunized with MOG during EAE, we induced disease

Fig. 1. Mapping of HLA-DR4 (DRB1*0401) restricted T cell epitopes of MOG. 6 – 8 wk old HLA-DR4 (DRB1*0401) transgenic mice were immunized with MOG97 – 108 in CFA sc and PT ip. After 9 days (A, D), 30 days (B), or 60 days (C, E) frequencies of antigen-specific IFN-g-producing T-cells were measured by cytokine ELISPOT assay in single cell suspensions of popliteal lymph node cells (A, D) or spleen cells (B, C, E) as outlined in Materials and methods. Shown are the mean F SD of the number of IFN-g producing cells per 5  105 lymph node cells (A, D) or 1  106 spleen cells (C, B, E) averaged for all mice per group (n = 4 mice) after recall with overlapping 20-mer MOG peptides (A, B, C) or MBP (D, E). The means were calculated from the results of triplicate wells, with the background subtracted (usually <5 spots). Similar results were obtained in three independent experiments.

56

J. Klehmet et al. / Clinical Immunology 111 (2004) 53–60

Table 1 Side chain scanning of MOG for HLA-DR4 (DRBI*0401) binding sequencesa AA position

MOG peptide sequence

Binding score (Hammer et al. [18])

183 – 191 99 – 107 127 – 135 178 – 186 15 – 23 83 – 91 136 – 144 195 – 103 120 – 128 182 – 190 137 – 145 40 – 48

FVIVPVLGP FFRDHSYQE LVLLAVLPV WKITLFVIV LVGDEVELP LRIRNVRFS LLLQITVGL LIICYNWLH YWVSPGVLV LFVIVPVLG LLQITVGLV YRPPFSRVV

4.2 3.9 3.5 3.1 3.0 2.8 2.8 2.7 2.5 2.5 2.3 2.1

a Scanning of the MOG sequence and calculation of peptide scores and estimated IC50s was performed as outlined in Materials and methods. Sequences with combined peptide binding scores above 2 (Hammer et al. [18]) were arranged according to their binding score (third column).

with MOG97 – 108 and tested T cell recall responses over time to overlapping MOG peptides covering the whole MOG sequence, or with MBP protein by cytokine ELISPOT assay. As shown in Fig. 1, we detected early on day 9 after immunization with MOG97 – 108 vigorous IFN-g production by T cells to MOG97 – 108, but not to any other MOG epitope (Fig. 1A) or MBP (Fig. 1D). However, when MOG97 – 108injected mice were tested 30 to 60 days after immunization, we found T cell responses to both MOG97 – 108, and to a new MOG epitope, MOG181 – 200 (Figs. 1B, C). Similar results were obtained by measuring IL-2 responses to MOG peptides by cytokine ELISPOT assay (data not shown). In contrast, only weak IL-5 production to the MOG181 – 200 epitope but not to any other MOG peptide was noted, arguing against engagement of Th2 immunity during epitope spreading in this model (data not shown). Importantly, 60 days after immunization, we also detected strong IFN-g production by T cells to a second myelin antigen, myelin basic protein (MBP; Fig. 1E). As shown in Fig. 4, T cell responses to the MOG epitopes were mediated by CD4+ T cells. Taken together, the data show that upon induction of EAE with the immunodominant HLA-DR4 restricted peptide MOG97 – 108 the autoimmune Th1 cell response diversified over time to MOG181 – 200 (intra-molecular spreading) and to MBP (inter-molecular spreading).

lying epitope spreading to MOG181 – 200, we investigated the binding affinity of this MOG sequence for HLA-DR4 (DRB1*0401). Screening of the region MOG181 – 200 with a peptidebinding algorithm [12,18] showed that it contained the sequence MOG183 – 191 (FVIVPVLGP), which was predicted to have the highest overall binding affinity for HLA-DR4 with a relative binding score of 4.2 (Table 1). In comparison, the immunodominant MOG97 – 108 epitope (core 99– 107) had a predicted binding score of 3.9. Based on this algorithm, MOG183 – 191 was considered the minimal core of the sequence MOG181 – 200. To formally determine the binding affinity of this region, we synthesized the peptide MOG179 – 191 (KITLFVIVPV LGP), a 13-mer peptide containing the minimal core region MOG183 – 191. As shown in Fig. 2, the results of the binding studies confirmed the binding predictions, demonstrating that this peptide bound with high affinity to HLA-DR4 with an IC50 of 1.6 AM. Consistent with the binding predictions (Table 1), its binding affinity for HLA-DR4 was higher than that of MOG97 – 108 (2.9 AM). To summarize, the results show that epitope spreading occurred to the MOG region with the highest binding affinity for HLA-DR4. MOG179 – 191-specific T cells are encephalitogenic The pathogenic potential of T cells specific for myelin antigens cannot be directly tested in patients. Moreover, epitope spreading in multiple sclerosis (MS) patients does not necessarily correlate with disease progression or relapses [20,21] and it has remained unresolved if T cells specific for these myelin epitopes are beneficial or pathogenic.

MOG181 – 200 has a high-binding affinity for HLA-DR4 (DRB1*0401) It is not known why T cell responses diversify to particular determinants of self-antigens, but peptide affinity for MHC may influence T cell priming and activation. During antigen processing, the affinity by which peptides bind to HLA-DR4 molecules contributes to the stability of the MHC –peptide complexes and their ability to induce T cell responses. Hence, to elucidate the mechanism(s) under-

Fig. 2. In vitro binding affinity of MOG179 – 191 to HLA-DR4 (DRB1*0401). MOG peptides MOG21 – 40, MOG97 – 108, MOG121 – 140 and MOG179 – 191 were probed for their binding affinity to purified HLA-DR4 (DRB1*0401) molecules. 1/IC50 values for each of the peptides were derived from the inhibition curves of the biotinylated tracer peptide CLIP obtained by dissociation-enhanced lanthanide fluoro-immunoassay (DELFIA) as described in Materials and methods.

J. Klehmet et al. / Clinical Immunology 111 (2004) 53–60

To test the pathogenic potential of T cells specific for the MOG spreading epitope, we immunized ‘‘humanized’’ HLA-DR4 transgenic mice with MOG179 – 191 and PT and observed for EAE. Shown in Fig. 3 are the disease scores for individual animals. The data show that MOG179 – 191 induced severe clinical EAE in the mice, paralleled by weight loss. Histologic examination revealed inflammatory infiltrates predominantly centered around vessels and ventricles and predominantly consisting of CD4+ T cells and macrophages/ microglia similar to the CNS pathology observed in MOG97 – 108 injected animals ([12]; data not shown). Similar results were obtained upon immunization with MOG181 – 200 (data not shown). Incidence of disease was 84% (16/19 mice), with average day of disease onset on day 8.6 F 2.2 and average highest disease score of 2.6 F 1.2. Control mice immunized with the irrelevant antigen hen egg white lysozyme (HEL) did not develop clinical disease and showed no inflammatory infiltrates in the CNS (data not shown). Together, the results showed that the spreading-epitope MOG179 – 191 induced EAE in HLA-DR4 transgenic mice. The spreading-epitope MOG179 – 191 is presented by human APCs We showed previously that T cell responses to MOG in MS patients were preferentially directed against epitopes in

57

the transmembranous/intracellular regions of the protein [22]. Similarly, in the present study, we found that epitope spreading occurred to a MOG peptide in the transmembranous region of this protein (MOG179 – 191). To support the hypothesis that the results obtained in the HLA-DR4 transgenic mice could have relevance for human MS, we investigated if MOG179 – 191 could be presented by human HLA-DR4+ APCs. As shown in Fig. 4, HLA-DR4 transgenic mice were immunized with MOG179 – 191 or MOG97 – 108, and 10 days later, CD4+ T cells were purified and incubated with the respective MOG peptides and human HLA-DR4+ B cells [14]. The results show that human B cells pulsed with MOG 179 – 191 induced vigorous IFN-g production by MOG179 – 191 primed T cells (Fig. 4, left panel, black bars), but not by MOG97 – 108-primed T cells (Fig. 4, left panel, grey bars). In contrast, MOG97 – 108-specific T cells responded to B cells pulsed with MOG97 – 108, but not to B cells incubated with MOG179 – 191 (Fig. 4, right panel, black bars). No T cell response was detected in wells containing HLA-DR4+ B cells incubated with medium alone, or with irrelevant control antigen (data not shown). To conclude, human HLA-DR4+ B cells efficiently presented MOG179 – 191 to antigen-specific T cells. Furthermore, the data suggested that epitope spreading in MS patients could similarly involve determinants in the trans-

Fig. 3. The spreading epitope MOG179 – 191 induces EAE in HLA-DR4 (DRB1*0401) transgenic mice. Transgenic mice were immunized with MOG179 – 191 in CFA sc and PT ip. Mice were observed at regular intervals for clinical signs of disease (A) and weight (B) and scored accordingly (see Materials and methods). Shown are individual animals from one experiment (n = 7 mice). Similar results were obtained in three independent experiments.

58

J. Klehmet et al. / Clinical Immunology 111 (2004) 53–60

Fig. 4. Human B cells present MOG179 – 191 to CD4+ T cells. HLA-DR4 (DRB1*0401) transgenic mice were immunized with MOG179 – 191 (black bars) or MOG97 – 108 (grey bars). 10 days later, draining lymph node cells were isolated, pooled, and CD4+ T cells purified by column separation as outlined in Materials and methods. The purified CD4+ T cells were tested in IFN-g ELISPOT assays using a human B cell line homozygous for HLADR4 (DRBI*0401) as APCs. Shown is a representative experiment of three experiments performed.

membranous region of MOG with high affinity for MHC class II molecules.

Discussion In this study, we show epitope spreading to MOG181 – 200 and MBP upon immunization of HLA-DR4 transgenic mice with the peptide MOG 97 – 108 . The results show that MOG181 – 200 has a high binding affinity for HLA-DR4 and induces EAE in HLA-DR4 transgenic mice. Previously, determinant spreading was reported in the T cell response to MBP and PLP in rodent models of EAE [1– 3]. Similarly, T cell responses to islet-cell antigens in NOD mice diversified over the course of autoimmune diabetes [23 –25]. In these models, the specificity and kinetic of the newly engaged autoantigenic determinants was predictable and consistent. The role of determinant spreading in human autoimmune diseases, such as MS, has been more difficult to dissect. Several groups reported epitope spreading for PLP or MBP determinants over the course of disease in MS patients [5,7,21]. However, T cell responses to individual PLP or MBP epitopes in MS patients were inconsistent and fluctuated over time. Moreover, epitope spreading did not necessarily correlate with disease activity in MS patients [20,21,26]. Similarly, it has been reported in a TCR transgenic mouse model of EAE that disease progression and relapses occurred in the absence of epitope spreading [4]. In contrast, other animal studies showed that the induction of tolerance to spreading epitopes with soluble myelin peptides or myelin coupled to splenocytes prevented disease relapses,

suggesting that these secondary epitopes may play a role in the pathogenesis of EAE [2,27]. The differing requirements for epitope spreading in these EAE models need to be reconciled before one can come to a conclusion regarding its role in disease pathogenesis. One possible explanation is that the high frequencies of autoreactive T cells in the TCR transgenic mice may have been sufficient to perpetuate the disease; in contrast, WT mice with diverse TCR repertoires may require a contribution of T cells with additional myelinspecificities for relapsing – remitting EAE. In human MS, the role of epitope spreading is even more complex. It is not known whether or not the fluctuations and shifts in epitope-reactivity by T cells in the peripheral blood of human patients were truly the result of epitope spreading. Alternatively, the changes in T cell reactivity observed in patients over time could be the result of shifts of primed autoreactive T cells from the blood to the CNS and vice versa. Moreover, it may take longer for subdominant autoreactive T cells to reach the clonal size necessary to become detectable in the peripheral blood of MS patients, as compared with immunodominant clones. Hence, the reported changes in the T cell responses to autoantigens in patients may not be truly epitope spreading, but rather reflect the migration or population dynamics of autoreactive T cells. Both of these issues are difficult to formally address in humans. Using a ‘‘humanized’’ HLA-DR transgenic mouse model, we show that HLA-DR4-restricted T cell responses induced by immunization with a single MOG peptide can diversify over the course of EAE to a second MOG peptide (intramolecular epitope spreading), and to a different myelin antigen (MBP, intermolecular epitope spreading). Moreover, the data show that epitope spreading can occur in the context of a single HLA-DR allele (i.e., HLA-DR4). In this transgenic mouse model that we have tested, determinant spreading occurred consistently to MOG181 – 200 (core 183 –191), and T cell responses to this epitope were detected as early as 3 weeks after immunization (data not shown). MOG181 – 200 is in the transmembranous region of the MOG molecule. Interestingly, T cell responses to MOG in humans are also frequently found in this region [22]. Hence, we suggest that the T cell responses to the transmembranous epitopes of MOG found in MS patients could be the result of epitope spreading. However, T cell responses to MOG epitopes were also detected in healthy control patients, raising the question as to their role in the pathogenesis of MS [22]. We found that immunization with the spreading-epitope MOG179 – 191 (and the longer peptide MOG181 – 200) induced EAE in the HLA-DR4 transgenic mice. Thus, this peptide was presented in the CNS and stimulated encephalitogenic Th1 cells. Importantly, human APCs presented this peptide to HLA-DR4 restricted T cells (Fig. 4). Therefore, our results suggest that this MOG region could also induce pathogenic T cells in MS patients. However, it is alternatively possible that T cells specific for these epitopes in MS patients and healthy individuals could be beneficial, for example, by secreting

J. Klehmet et al. / Clinical Immunology 111 (2004) 53–60

Th2 cytokines or by acting as regulatory T cells [28 – 30]. Finally, these cells could participate in CNS repair [31]. To our knowledge, the presented results show for the first time intra- and intermolecular epitope spreading induced by MOG in mice. The in vitro binding studies showed that MOG179 – 191 bound with high affinity to HLA-DR4. Hence, the data indicated that MOG peptides with high affinity for HLA-DR4 were favored during epitope spreading. This interpretation is consistent with other reports showing epitope spreading to high-affinity MHC-binding self-peptides [25]. However, most of the autoreactive T cells in this system were of low functional avidity [25]. This finding may reflect the negative selection of high-avidity self-reactive T cells in the thymus by the presentation of high affinity MHC binding self-antigens [32,33]. In contrast, it has been suggested that MOG is not expressed in the thymus [34], which is consistent with the observation that MOG97 – 108-reactive T cells in HLA-DR4 transgenic mice have a high functional avidity for their antigen (T.G.F., unpublished). Thus, we propose that in the absence of thymic negative selection, HLA-DR4dependent T cell epitope spreading is driven towards higher affinity binding self-peptides and induces T cells with higher functional avidity [35]. However, if the respective self-antigens are expressed in the thymus and induce negative selection, T cells with specificity for low-affinity MHC binding self-peptides may escape negative selection [36]. In this scenario, we would predict that self-peptides with lower affinity for MHC become the target for epitope spreading. It must be emphasized, however, that epitope spreading does not necessarily reflect the pathogenicity of the induced T cell response. For example, epitope spreading of Th2 responses to islet cell antigens has been shown to protect from insulindependent diabetes mellitus in NOD mice [37]. In summary, in the present study, we demonstrate intraand intermolecular epitope spreading induced by MOG in an HLA-DR4 transgenic mouse model. The results suggest that epitope spreading needs to be taken into account in devising antigen-specific therapies for human patients. Acknowledgments This work was supported by grants AI-41609-01, 1RO1AR45918 and NS-42809 from the National Institute of Health, the Harry Weaver Neuroscience Scholarship JF2092-A-1 and RG3322A2 from the National Multiple Sclerosis Society; a grant from the Wadsworth Foundation to T.G.F, the Deutsche Forschungsgemeinschaft to R.W. (We 1947/3-2 and 4-1, and SFB510 project D6) and B.O.B. (SFB 518), and a fellowship of the Studienstiftung des Deutschen Volkes to J.K. References [1] P.V. Lehmann, T. Forsthuber, A. Miller, E.E. Sercarz, Spreading of Tcell autoimmunity to cryptic determinants of an autoantigen, Nature 358 (6382) (1992) 155 – 157.

59

[2] B.L. McRae, C.L. Vanderlugt, C.M. Dal, S.D. Miller, Functional evidence for epitope spreading in the relapsing pathology of experimental autoimmune encephalomyelitis, J. Exp. Med. 182 (1995) 75 – 85. [3] M. Yu, J.M. Johnson, V.K. Tuohy, A predictable sequential determinant spreading cascade invariably accompanies progression of experimental autoimmune encephalomyelitis: a basis for peptide-specific therapy after onset of clinical disease, J. Exp. Med. 183 (4) (1996) 1777 – 1788. [4] R.E. Jones, D. Bourdette, N. Moes, A. Vandenbark, A. Zamora, H. Offner, Epitope spreading is not required for relapses in experimental autoimmune encephalomyelitis, J. Immunol. 170 (4) (2003) 1690 – 1698. [5] V.K. Tuohy, M. Yu, B. Weinstock-Guttman, R.P. Kinkel, Diversity and plasticity of self recognition during the development of multiple sclerosis, J. Clin. Invest. 99 (7) (1997) 1682 – 1690. [6] N. Goebels, H. Hofstetter, S. Schmidt, C. Brunner, H. Wekerle, R. Hohlfeld, Repertoire dynamics of autoreactive T cells in multiple sclerosis patients and healthy subjects: epitope spreading versus clonal persistence, Brain 123 (Pt 3) (2000) 508 – 518. [7] G. Mazza, M. Ponsford, P. Lowrey, M.J. Campbell, J. Zajicek, D.C. Wraith, Diversity and dynamics of the T-cell response to MBP in DR2 + ve individuals, Clin. Exp. Immunol. 128 (3) (2002) 538 – 547. [8] K. Ito, H.J. Bian, M. Molina, et al., HLA-DR4-IE chimeric class II transgenic, murine class II-deficient mice are susceptible to experimental allergic encephalomyelitis, J. Exp. Med. 183 (1996) 2635 – 2644. [9] M.G. Marrosu, F. Muntoni, M.R. Murru, et al., Sardinian multiple sclerosis is associated with HLA-DR4: a serologic and molecular analysis, Neurology 38 (11) (1988) 1749 – 1753. [10] M.G. Marrosu, M.R. Murru, G. Costa, et al., Multiple sclerosis in Sardinia is associated and in linkage disequilibrium with HLA-DR3 and -DR4 alleles [letter], Am. J. Hum. Genet. 61 (2) (1997) 454 – 457. [11] P.A. Muraro, M. Vergelli, M. Kalbus, et al., Immunodominance of a low-affinity major histocompatibility complex-binding myelin basic protein epitope (residues 111 – 129) in HLA-DR4 (B1*0401) subjects is associated with a restricted T cell receptor repertoire, J. Clin. Invest. 100 (2) (1997) 339 – 349. [12] T.G. Forsthuber, C.L. Shive, W. Wienhold, et al., T cell epitopes of human myelin oligodendrocyte glycoprotein identified in HLA-DR4 (DRB1*0401) transgenic mice are encephalitogenic and are presented by human B cells, J. Immunol. 167 (12) (2001) 7119 – 7125. [13] H.C. Rauch, E.R. Einstein, J. Csejtey, W.J. Davis, Protective action of the encephalitogen and other basic proteins in experimental allergic encephalomyelitis, Immunochemistry 5 (1968) 567 – 575. [14] M. Congia, S. Patel, A.P. Cope, S. De Virgiliis, G. Sonderstrup, T cell epitopes of insulin defined in HLA-DR4 transgenic mice are derived from preproinsulin and proinsulin, Proc. Natl. Acad. Sci. U. S. A. 95 (7) (1998) 3833 – 3838. [15] C.L. Shive, H. Hofstetter, L. Arredondo, C. Shaw, T.G. Forsthuber, The enhanced antigen-specific production of cytokines induced by pertussis toxin is due to clonal expansion of T cells and not to altered effector functions of long-term memory cells, Eur. J. Immunol. 30 (8) (2000) 2422 – 2431. [16] A.Y. Karulin, M.D. Hesse, M. Tary-Lehmann, P.V. Lehmann, Singlecytokine-producing CD4 memory cells predominate in type 1 and type 2 immunity, J. Immunol. 164 (4) (2000) 1862 – 1872. [17] M. Rocken, M. Racke, E.M. Shevach, IL-4 induced immune deviation as antigen-specific therapy for inflammatory autoimmune disease. Immunol. Today 17:275. [18] J. Hammer, E. Bono, F. Gallazzi, C. Belunis, Z. Nagy, F. Sinigaglia, Precise prediction of major histocompatibility complex class II-peptide interaction based on peptide side chain scanning, J. Exp. Med. 180 (6) (1994) 2353 – 2358. [19] R. Weissert, K.L. de Graaf, M.K. Storch, et al., MHC class II-regulated central nervous system autoaggression and T cell responses in peripheral lymphoid tissues are dissociated in myelin oligodendrocyte glycoprotein-induced experimental autoimmune encephalomyelitis, J. Immunol. 166 (12) (2001) 7588 – 7599.

60

J. Klehmet et al. / Clinical Immunology 111 (2004) 53–60

[20] G. Ristori, F. Giubilei, D. Giunti, et al., Myelin basic protein intramolecular spreading without disease progression in a patient with multiple sclerosis, J. Neuroimmunol. 110 (1 – 2) (2000) 240 – 243. [21] M. Vergelli, B. Mazzanti, E. Traggiai, et al., Short-term evolution of autoreactive T cell repertoire in multiple sclerosis, J. Neurosci. Res. 66 (3) (2001) 517 – 524. [22] R. Weissert, J. Kuhle, K.L. de Graaf, et al., High immunogenicity of intracellular myelin oligodendrocyte glycoprotein epitopes, J. Immunol. 169 (1) (2002) 548 – 556. [23] D.L. Kaufman, M. Clare-Salzler, J. Tian, et al., Spontaneous loss of Tcell tolerance to glutamic acid decarboxylase in murine insulin-dependent diabetes [see comments], Nature 366 (1993) 69 – 72. [24] M.A. Zechel, M.D. Krawetz, B. Singh, Epitope dominance: evidence for reciprocal determinant spreading to glutamic acid decarboxylase in non-obese diabetic mice, Immunol. Rev. 164 (1998) 111 – 118. [25] J. Tian, S. Gregori, L. Adorini, D.L. Kaufman, The frequency of high avidity T cells determines the hierarchy of determinant spreading, J. Immunol. 166 (12) (2001) 7144 – 7150. [26] A. Uccelli, G. Ristori, D. Giunti, et al., Dynamics of the reactivity to MBP in multiple sclerosis, J. Neurovirol. 6 (Suppl. 2) (2000) S52 – S56. [27] V.K. Tuohy, M. Yu, L. Yin, et al., The epitope spreading cascade during progression of experimental autoimmune encephalomyelitis and multiple sclerosis, Immunol. Rev. 164 (1998) 93 – 100. [28] A. Miller, O. Lider, A.B. Roberts, M.B. Sporn, H.L. Weiner, Suppressor T cells generated by oral tolerization to myelin basic protein suppress both in vitro and in vivo immune responses by the release of transforming growth factor beta after antigen-specific triggering, Proc. Natl. Acad. Sci. U. S. A. 89 (1) (1992) 421 – 425. [29] S. Sakaguchi, N. Sakaguchi, Organ-specific autoimmune disease in-

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

duced in mice by elimination of T cell subsets. V. Neonatal administration of cyclosporin A causes autoimmune disease, J. Immunol. 142 (2) (1989) 471 – 480. S. Sakaguchi, N. Sakaguchi, M. Asano, M. Itoh, M. Toda, Immunologic self-tolerance maintained by activated T cells expressing IL-2 receptor alpha-chains (CD25). Breakdown of a single mechanism of self-tolerance causes various autoimmune diseases, J. Immunol. 155 (3) (1995) 1151 – 1164. H.H. Hofstetter, D.L. Sewell, F. Liu, et al., Autoreactive T cells promote post-traumatic healing in the central nervous system, J. Neuroimmunol. 134 (1 – 2) (2003) 25 – 34. L. Klein, M. Klugmann, K.A. Nave, B. Kyewski, Shaping of the autoreactive T-cell repertoire by a splice variant of self protein expressed in thymic epithelial cells [see comments], Nat. Med. 6 (1) (2000) 56 – 61. J. Derbinski, A. Schulte, B. Kyewski, L. Klein, Promiscuous gene expression in medullary thymic epithelial cells mirrors the peripheral self, Nat. Immunol. 2 (11) (2001) 1032 – 1039. R. Bruno, L. Sabater, M. Sospedra, et al., Multiple sclerosis candidate autoantigens except myelin oligodendrocyte glycoprotein are transcribed in human thymus, Eur. J. Immunol. 32 (10) (2002) 2737 – 2747. O.S. Targoni, P.V. Lehmann, Endogenous myelin basic protein inactivates the high avidity T cell repertoire, J. Exp. Med. 187 (12) (1998) 2055 – 2063. S.M. Anderton, C.G. Radu, P.A. Lowrey, E.S. Ward, D.C. Wraith, Negative selection during the peripheral immune response to antigen, J. Exp. Med. 193 (1) (2001) 1 – 11. J. Tian, P.V. Lehmann, D.L. Kaufman, Determinant spreading of T helper cell 2 (Th2) responses to pancreatic islet autoantigens, J. Exp. Med. 186 (12) (1997) 2039 – 2043.