Conformational epitopes of myelin oligodendrocyte glycoprotein (MOG) provide a major target for demyelinating autoantibodies in experimental autoimmune encephalomyelitis and recent studies indicate that a similar situation may exist in multiple sclerosis. We recently solved the crystal structure of the extracellular domain of MOG (MOGex) in complex with a Fab derived from the demyelinating mAb 8-18C5 and identified the conformational 8-18C5 epitope on MOG that is dominated by the surface exposed FG loop of MOG. To determine the importance of this epitope with regard to the polyclonal Ab response to MOGex we investigated the effects of mutating His103 and Ser104, the two central amino acids of the FG loop, on Ab binding. Mutation of these two residues reduced binding of a panel of eight demyelinating conformation-dependent mAbs to <20% compared with binding to wild-type MOGex, whereas substitution of amino acids that do not contribute to the 8-18C5 epitope had only a minor effect on Ab binding. The same restriction was observed for the polyclonal MOG-specific Ab response of MOG DNA-vaccinated BALB/c and SJL/J mice. Our data demonstrate that the pathogenic anti-MOG Ab response primarily targets one immunodominant region centered at the FG loop of MOG. Comparison of the structure of MOGex with the structures of related IgV-like domains yields a possible explanation for the focused Ab response.

Myelin oligodendrocyte glycoprotein (MOG)4 is a highly conserved transmembrane myelin protein that is expressed exclusively in the CNS (1, 2). It was identified originally as an immunodominant target for demyelinating autoantibodies in a guinea pig model of experimental autoimmune encephalomyelitis (EAE) (3, 4). Subsequently, MOG was shown to induce a variant of EAE that exhibits many of the clinical and pathological characteristics of multiple sclerosis (MS) in both rats and primates. In these species, inflammation and blood brain barrier dysfunction are initiated by the MOG-specific T cell response, but the formation of large demyelinating lesions is dependent on Ab-dependent effector mechanisms triggered by MOG-specific autoantibodies (5, 6, 7).

Ab-mediated demyelination in MOG-EAE reproduces many features associated with demyelination in MS including the local deposition of Igs and complement activation products in areas of active demyelination. These similarities together with the exposed location of MOG at the surface of CNS myelin and the association of MOG-specific autoantibodies with myelin debris identify MOG as a prime candidate for demyelinating Abs in MS (8, 9, 10). Yet, the frequency and clinical significance of MOG-specific autoantibodies in MS remains controversial. Some laboratories report that MS is associated with significantly elevated titers of MOG-specific autoantibodies (7, 11, 12, 13, 14, 15, 16), while others find no significant difference between MS patients, patients with other inflammatory neurological diseases, or even healthy control donors (17, 18, 19, 20, 21). Recent studies suggest that this lack of concordance between different studies is due to the inability of most assay methods to differentiate between pathogenic (demyelinating) and nonpathogenic MOG-specific Abs. Experimental animals mount a complex Ab response to the extracellular domain of MOG (MOGex) that targets both linear as well as discontinuous, conformational epitopes (22, 23, 24). Crucially, only conformation-dependent MOG-specific Abs are able to initiate demyelination in EAE (22, 23, 24). In contrast, Abs directed against linear MOGex-peptide epitopes are unable to bind to the native protein at the membrane surface and fail to induce any significant pathology in vivo. Clinical studies suggest that a similar situation exists in humans. MOG-specific Ab responses directed against discontinuous, conformational epitopes are present in a subset of MS patients (10, 16, 17, 25), while autoantibodies to linear peptide MOGex epitopes are found in both MS patients and healthy controls.

In view of the importance of conformation-dependent MOGex-specific Abs in immune-mediated demyelination, we recently determined the three-dimensional structure of the complex formed between MOGex and a Fab derived from the demyelinating mouse MOG-specific mAb 8-18C5 (26). This structural analysis confirmed that the epitope recognized by 8-18C5 is highly discontinuous and identified the surface-exposed FG loop of MOGex as a major component of the epitope.

We have now used site-directed mutagenesis to investigate the importance of interactions that involve this region of MOG within the polyclonal MOG-specific Ab repertoire. We report that >80% of Ag recognition by conformation-dependent MOGex-specific mAbs is abolished by mutation of two amino acids (His103 and Ser104) at the tip of the FG loop. This highly focused response is also seen for the MOGex-specific Ab response in BALB/c and SJL/J mice after MOG-DNA vaccination. This study identifies a key immunodominant epitope within the demyelinating MOG-specific Ab repertoire that is centered at the FG loop of MOG. Comparison of the structure of MOGex with the structures of related Ig-V like domains of other proteins reveals the lack of structural conservation in this epitope region suggesting a possible explanation for its immunodominance within the MOGex-specific Ab repertoire.

Intermolecular contacts were analyzed using programs of the CCP4 package (27) and the model building program O (28). Electrostatic potentials were calculated in GRASP (29) by using atomic charges according to Weiner et al. (30). The solvent-accessible surface of MOGex was calculated with the utility SURFACE of the CCP4 program package. Mutagenesis was conducted using the extracellular domain of rat MOG (MOGex) subcloned into the His-tag expression vector pQE-12 (26) by following the method of “QuikChange Site-Directed Mutagenesis” by Stratagene. The oligonucleotides used were: 5′-CTTCAGAGACCACGAATACCAAGAAGAAGCCGCCG-3′ (SM1), 5′-CACATGCTTCTTCAGAGACGGCGAATACCAAG-3′ (DM1), 5′-CACATGCTTCTTCAGAGACGCTGAATACCAAG-3′ (DM2), and the corresponding reverse complementary oligonucleotides. The identity of the mutations was verified by DNA sequencing of the purified plasmids.

Plasmids containing the extracellular domain of human MOG and the “humanized” rat MOGex mutant Ser42Pro were a gift of N. Ruddle (Yale University, New Haven, CT) (31). The extracellular domain of rat and human MOG and the mutant proteins were overexpressed in inclusion bodies in Escherichia coli. After disruption of the cells by sonification, the inclusion bodies were purified by repetitive steps of centrifugation and resuspension in 50 mM Tris-HCl (pH 8.0), 0.3 M NaCl, 0.5% lauryldimethylamine oxide. The inclusion bodies were solubilized in solubilization buffer (100 mM NaH2PO4, 10 mM Tris, 6 M guanidinium chloride, 40 mM mercaptoethanol (pH 8.0)). After dilution in mercaptoethanol-free solubilization buffer, the denatured MOGex was bound to Ni-NTA Superflow (Qiagen) material and refolded on the column in two steps. At first, a linear gradient from solubilization buffer (1 mM mercaptoethanol) to 100 mM NaH2PO4, 10 mM Tris, 3 mM glutathione (pH 8.0) over 10 h and 80-column volumes was applied, followed by a short linear gradient (2 h, 2-column volumes) to remove the glutathione for complete oxidation of the refolded MOGex. After elution, unfolded and aggregated MOGex was removed by a final size exclusion chromatography step. Identity and integrity of the proteins were checked by mass spectrometry and one-dimensional [1H]NMR. Protein concentrations were determined by UV/Vis spectroscopy; relative concentrations were determined by the Bradford protein assay (Bio-Rad).

Crystals of DM2-MOGex were grown at 20°C with the sitting drop vapor diffusion method by mixing 1.5 μl of protein (4 mg/ml, in 1/4 PBS) with 0.8 μl of 100 mM Tris-HCl (pH 7.2), 12% PEG8000, and 100 mM magnesium acetate. The hexagonal crystals of DM2-MOGex belong to space group P3221 and contain one molecule per asymmetric unit like the wild-type crystals. Diffraction data of the crystals frozen in mother liquor supplied with 20% butanediol as cryoprotectant were collected at the BESSY synchrotron (beamline MX14-2, λ = 0.91841 Å) using a MAR CCD 165-mm detector. Data were integrated and scaled using XDS (32). The structure of DM2-MOGex was determined by molecular replacement using the program PHASER (33) and wild-type MOGex (Protein Data Bank ID 1PKO) as a search model. Refinement was performed by alternated model building using the program O (28) and crystallographic refinement cycles with CNS (34) that included simulated annealing, positional refinement, and restrained temperature factor refinement using the parameters of Engh and Huber (35). Analysis of the structure was performed with programs of the CCP4 package (27). Data collection and refinement parameters are summarized in Table I.

Table I.

Data collection and refinement statisticsa

Data Collection  
 Space group P3221 
 Unit cell dimensions (Å) a = b = 49.24, c = 77.60 
 Resolution (Å) 1.70 (1.70–1.80) 
 Observed reflections 97,717 
 Independent reflections 12,373 
 Completeness (%) 99.3 (99.9) 
Rsym (%)b 4.1 (32.7) 
 <I>/<ς(I)> 29.4 (6.7) 
 B value of Wilson plot (Å222.8 
Refinement  
Rcryst/Rfree (%)c 19.9/24.5 
 No. of atoms: protein/water 972/139 
 Rms deviation: bonds (Å)/angles (°)/ bonded B’s (Å2)d 0.011/1.66/1.42 
 Mean B value: protein/water (Å223.10/34.2 
Data Collection  
 Space group P3221 
 Unit cell dimensions (Å) a = b = 49.24, c = 77.60 
 Resolution (Å) 1.70 (1.70–1.80) 
 Observed reflections 97,717 
 Independent reflections 12,373 
 Completeness (%) 99.3 (99.9) 
Rsym (%)b 4.1 (32.7) 
 <I>/<ς(I)> 29.4 (6.7) 
 B value of Wilson plot (Å222.8 
Refinement  
Rcryst/Rfree (%)c 19.9/24.5 
 No. of atoms: protein/water 972/139 
 Rms deviation: bonds (Å)/angles (°)/ bonded B’s (Å2)d 0.011/1.66/1.42 
 Mean B value: protein/water (Å223.10/34.2 
a

Values in parentheses correspond to the highest resolution shell.

b

Rsym = ΣhiIi(h) − <I(h)> /ΣhiIi(h).

c

Rcryst = 5 ΣhFo(h) − Fc(h)∥/ΣhFo(h). Rfree was determined from 5% of the data that were not used for refinement.

d

Root mean square deviation of bonded B’s: root mean square deviation of temperature factors of bonded atoms.

Large amounts of the expression vector pcDNA3.1(−) (Invitrogen) that was used as a control and of pcDNA3.1(−) containing the coding sequence of full-length MOG and its signal sequence (Bourquin et al. (36)) were prepared using Qiagen Endotoxin-Free Plasmid kits (Qiagen). Female SJL/J and BALB/c mice and Dark Agouti rats were purchased from Charles River Laboratories. At the age of 4–5 wk, plasmid DNA in PBS (mice, 100 μg; rats, 400 μg) was injected into the tibialis anterior muscle. Plasma was prepared by centrifugation and stored at −20°C. All animal experiments were performed according to the Bavarian state regulations for animal experimentation, and were approved by the responsible authorities.

Ab binding to MOGex and to the mutant proteins was measured by ELISA. The mouse mAbs 8-18C5 (4), Y1, Y8, Y9, Y10, Y11, Z2, Z4, Z8, and Z12 (37) were purified from hybridoma supernatants by affinity chromatography on Protein G. Their concentration was estimated by UV/Vis spectroscopy and colorimetrically by the Bradford method. Ninety-six-well plates (Maxisorb; Nunc) were coated with 100 μl of 10 μg/ml Ag in PBS (1 h, 30°C), washed three times with PBS containing 0.2% Tween 20 and blocked with PBS containing 1% w/v BSA (2 h, 30°C). After washing, the plates were incubated with the mAbs (∼0.5 μg/ml in PBS) or the plasma samples of the MOG-vaccinated mice diluted 1/250 for 1 h at 30°C. The washing procedure was repeated and anti-mouse IgG (Fab′)2, conjugated with HRP (Amersham Biosciences), that was diluted 1/10000 in PBS was added and the plates were incubated for 1 h at 30°C. Ab binding was detected by oxidation of o-phenylene diamine and quantified by measuring the absorbance at 490 nm after stopping the reaction with H2SO4. The displayed values correspond to the means of triplicate (plasma samples) and quadruplicate (hybridoma supernatants) measurements of a representative experiment.

Searches for related sequences were performed with BLAST (www.ncbi. nlm.nih.gov) and sequence alignments with GCG (Wisconsin Package version 10.3; Accelrys). The structural alignment was calculated with LSQMAN (38). Figures were drawn with MOLSCRIPT (39), RASTER3D (40), and GRASP (29). The graphical representation of the sequence alignment was prepared using Alscript (41).

The structure of the extracellular Ig-V like domain of MOG (MOGex) in complex with a chimeric Fab derived from mouse mAb 8-18C5 identified a highly discontinuous epitope on MOG that is composed of the N terminus, the three upper loops, and two amino acids of the CC′ loop (Fig. 1, a and b) (26). In agreement with the observed cross-reactivity of 8-18C5 between species, those amino acids forming the epitope are strictly conserved between human, rat, and mouse MOG. Despite the high discontinuity of the epitope, the major contribution to Ab binding is made by the FG loop and the flanking G-strand amino acids (amino acids 101–108). These amino acids not only form 10 of the 12 intermolecular hydrogen bonds involved in Ab binding but also provide ∼65% of the total contact area. In the center of the epitope, His103 and Ser104 that form the tip of the FG loop are inserted deeply into a cleft of the paratope of 8-18C5 (Fig. 1, c and d) and lose the highest amount of solvent accessible surface upon Ab binding, during which Ser104 becomes completely buried in the Ag-Ab interface (Fig. 1 a).

FIGURE 1.

The discontinuous 8-18C5 epitope on MOG centered at His103 and Ser104. a, Bar diagram of the solvent-accessible surface of MOG per amino acid in Å2. The total height of each bar corresponds to the solvent-accessible surface of unbound MOG; the part of the surface that is covered by 8-18C5 upon binding (regions A–E) corresponds to the black fraction of the bars. Below the diagram the sequence alignment of rat, mouse, and human MOG and the secondary structural elements of rMOG are shown. b, Stereo view of the MOG-(8-18C5) interface. Amino acids of the 8-18C5 epitope on MOG and amino acids of 8-18C5 that contact His103 and Ser104 are shown in ball-and-stick representation. Apart from MOG amino acids 101–108 (region E) in the epitope centre, the N terminus (A) and three other sequence stretches of MOG (B–D) contribute to binding of 8-18C5. c, Surface-and-ribbon representation of MOGex. Amino acids of the 8-18C5 epitope are colored green except for His103 (blue) and Ser104 (orange). In addition, Ser42 that corresponds to proline in human MOG is shown in pink, Tyr96 (hMOG: Phe) in yellow, and Val111 (hMOG: Met) in turquoise. d, Binding of His103 and Ser104 (green) at the tip of the FG loop (amino acids 102–105) of MOGex (yellow) to the paratope of 8-18C5. The electrostatic potential of 8-18C5 is mapped onto the molecular surface of the paratope region of 8-18C5 (red: negative charge, blue: positive charge). In the mutated proteins SM, DM1, and DM2, Ser104 and His103 are exchanged for Glu and Gly/Ala, respectively.

FIGURE 1.

The discontinuous 8-18C5 epitope on MOG centered at His103 and Ser104. a, Bar diagram of the solvent-accessible surface of MOG per amino acid in Å2. The total height of each bar corresponds to the solvent-accessible surface of unbound MOG; the part of the surface that is covered by 8-18C5 upon binding (regions A–E) corresponds to the black fraction of the bars. Below the diagram the sequence alignment of rat, mouse, and human MOG and the secondary structural elements of rMOG are shown. b, Stereo view of the MOG-(8-18C5) interface. Amino acids of the 8-18C5 epitope on MOG and amino acids of 8-18C5 that contact His103 and Ser104 are shown in ball-and-stick representation. Apart from MOG amino acids 101–108 (region E) in the epitope centre, the N terminus (A) and three other sequence stretches of MOG (B–D) contribute to binding of 8-18C5. c, Surface-and-ribbon representation of MOGex. Amino acids of the 8-18C5 epitope are colored green except for His103 (blue) and Ser104 (orange). In addition, Ser42 that corresponds to proline in human MOG is shown in pink, Tyr96 (hMOG: Phe) in yellow, and Val111 (hMOG: Met) in turquoise. d, Binding of His103 and Ser104 (green) at the tip of the FG loop (amino acids 102–105) of MOGex (yellow) to the paratope of 8-18C5. The electrostatic potential of 8-18C5 is mapped onto the molecular surface of the paratope region of 8-18C5 (red: negative charge, blue: positive charge). In the mutated proteins SM, DM1, and DM2, Ser104 and His103 are exchanged for Glu and Gly/Ala, respectively.

Close modal

To investigate whether this region is important for Ag recognition by other MOG-specific Abs, we generated the following mutated versions of MOGex: a single mutant SM1 (Ser104Glu) and two double mutants DM1 (His103Gly, Ser104Glu) and DM2 (His103Ala, Ser104Glu) (Fig. 1 d). Substituting Ser104 by glutamate should significantly perturb binding of 8-18C5 to MOG as it is unlikely that the bulky glutamate side chain will fit into the small binding pocket of 8-18C5 occupied by Ser104 of wild-type MOG. In addition, contacts between the anionic glutamate side chain and the partially negatively charged surface of 8-18C5 in the vicinity of Ser104 should be unfavorable. The additional mutation of His103 in the double mutants results in the loss of two buried hydrogen bonds between MOG and amino acids of 8-18C5. The introduction of bulkier amino acids at position 103 was avoided because the side chain of this amino acid determines the strained conformation of the FG loop (26, 42) and larger side chains might complicate refolding of MOG. In contrast to other amino acids that contribute to the 8-18C5 epitope, the side chains of His103 and Ser104 are not involved in intramolecular hydrogen bonds and are thus suitable targets for mutations that will not perturb the overall structural integrity and conformation of the protein.

All three mutant proteins could be expressed and refolded under the same conditions as wild-type MOGex and eluted as single peaks with nearly the same retention time as the wild-type protein from a size exclusion chromatography column. In addition, one-dimensional [1H]NMR spectra of the wild-type and the mutant proteins are virtually identical and all exhibit typical characteristics of folded proteins (43), i.e., a large signal dispersion in the aliphatic region between 1.0 and −1.0 ppm as well as a signal dispersion beyond 8.5 ppm (Fig. 2,a). Thus, as intended, the mutations neither destabilize the three-dimensional structure of MOGex nor provoke global structural changes. To make sure that the mutations do not induce minor structural changes, for example in the conformation of loops near the site of mutagenesis, we crystallized the most strongly mutated MOGex protein DM2 and solved the three-dimensional structure of DM2 at a resolution of 1.7 Å (Table I). The overall structure and the loop conformations correspond to wild-type MOGex (Fig. 2,b). Apart from the two mutated amino acids, the only slight structural changes are found at the N and C terminus of MOG that are flexible in solution and become ordered in the crystal due to crystal contacts that differ in wild-type and mutant MOG crystals because of different crystal packing. For the same reason, the CC′ loop is flexible in the wild-type but ordered in the mutant structure and the (8-18C5)-Fab-MOGex structure (Fig. 2 b).

FIGURE 2.

Structure of mutated MOGex proteins. a, Amide region (>8 ppm) and aliphatic regions (1.0 to −1.0 ppm) of one-dimensional proton NMR spectra of wild-type MOGex and of the mutant proteins SM, DM1, and DM2. b, Stereo view of the overlaid Cα backbone of the MOG mutant DM2 (red), wild-type MOGex (blue, 1PKO) and wild-type MOGex complexed with (8-18C5)-Fab (yellow, 1PKQ).

FIGURE 2.

Structure of mutated MOGex proteins. a, Amide region (>8 ppm) and aliphatic regions (1.0 to −1.0 ppm) of one-dimensional proton NMR spectra of wild-type MOGex and of the mutant proteins SM, DM1, and DM2. b, Stereo view of the overlaid Cα backbone of the MOG mutant DM2 (red), wild-type MOGex (blue, 1PKO) and wild-type MOGex complexed with (8-18C5)-Fab (yellow, 1PKQ).

Close modal

Analyzing the interaction of the MOG mutants with 8-18C5 by ELISA revealed that the single mutation Ser104Glu reduces the binding of MOG to 8-18C5 by >40% (Fig. 3 a). The additional exchange of His103 by glycine in DM1 and alanine in DM2 resulted in reduction of the ELISA signal by 97 and 100%, respectively. This confirms the central role of MOG amino acids 103/104 in binding 8-18C5 as deduced from the crystal structure of the MOGex-(8-18C5) Fab complex. We then tested whether these amino acids also play a significant role in Ag recognition by a panel of MOG-specific mAbs that were derived from BALB/c mice immunized with MOG or myelin proteins (37). All these mAbs recognize native glycosylated as well as deglycosylated MOG from different species in vitro and bind to native human MOG expressed on fibroblasts (22, 37). Except for the two mAbs Y8 and Z8, all mAbs that we tested enhanced demyelination in vivo (37).

FIGURE 3.

Epitope specificity of MOG-specific monoclonal mouse Abs. a, Interaction with MOGex and MOGex with the mutated FG loop, as determined by ELISA. Except for Y11, the binding affinity of all Abs to the double mutants is drastically decreased. b, Interaction with human MOGex and the “humanized” rat MOGex mutant Ser42Pro. Ab binding is only slightly reduced compared with wild-type rat MOGex.

FIGURE 3.

Epitope specificity of MOG-specific monoclonal mouse Abs. a, Interaction with MOGex and MOGex with the mutated FG loop, as determined by ELISA. Except for Y11, the binding affinity of all Abs to the double mutants is drastically decreased. b, Interaction with human MOGex and the “humanized” rat MOGex mutant Ser42Pro. Ab binding is only slightly reduced compared with wild-type rat MOGex.

Close modal

Analysis of the epitope specificity of these mAbs revealed that binding of the mutant MOGex proteins to eight of the nine mAbs was very similar to that of 8-18C5 (Fig. 3 a). Only the interaction with mAb Y11 was not influenced by any of the mutations. For all the other mAbs, binding to the double mutants was reduced to 0–25% in the ELISA. The strongest effect was seen for Y9 and Y10 whose binding to DM1 as well as DM2 was reduced to ≤8% compared with wild-type MOGex. The single mutant was still recognized by all eight mAbs although the ELISA signal was weaker than that obtained using the wild-type protein, yielding ELISA signals in the range of 47–79%. Only Y10 showed no statistically significant reduction in binding.

The amino acid sequences of the IgV-like domains of human and rat MOG exhibit a sequence identity of 92%. The most significant differences at the molecular surface of the IgV-like domains are the substitution of rat Gly9 by arginine and rat Ser42 by proline in human MOGex (Fig. 1,a). Ser42 is positioned directly adjacent to the 8-18C5 epitope on MOG whereas Gly9 is located at a site not directly involved in the binding of 8-18C5 (Fig. 1,c). To investigate the effects of these structural changes on Ab binding, we determined the ability of the mAbs to bind to the extracellular domain of human MOG (hMOG) and the rat MOGex mutant Ser42Pro (S42P) (31) by ELISA. In agreement with the structure of the MOGex-(8-18C5)Fab complex, substitution of Ser42 by proline had only a relatively minor effect on Ab binding. The ELISA signal was reduced by 19% relative to wild-type MOGex for 8-18C5, 7% for Z2 and only minimally for the remaining mAbs (Fig. 3 b). The additional substitutions present in hMOG further reduced Ab binding. In the case of 8-18C5, the ELISA signal was decreased by 38%, while the binding of five further mAbs (Z2, Z4, Z8, Z12, Y8) was reduced by 10 to 28%.

To confirm that the FG loop plays a major role in Ab recognition of MOGex, we investigated the epitope specificity patterns of the conformation-dependent polyclonal Ab response to MOG induced by DNA vaccination in BALB/c and SJL/J mice (36). Whereas active immunization with MOGex generates a mixed repertoire of conformation-dependent and peptide-specific anti-MOG Abs that greatly complicates analysis of individual components of the repertoire, and furthermore results in severe EAE, a situation in which pathogenic conformation-dependent Abs may be preferentially bound within the CNS compartment, MOG-DNA vaccination generates a strictly conformation-dependent polyclonal Ab response to native membrane-bound MOG without inducing EAE (36).

In BALB/c as well as in SJL/J mice, MOG-specific Abs were detected 2 wk after vaccination. The Ab titer reached a maximum 4–6 wk after vaccination and then fell to ∼40% of this value by 11 wk (Fig. 4,a). Analysis of epitope specificity by ELISA using the mutated MOGex proteins revealed that both strains developed a specificity pattern that closely resembles that of the mAbs (Fig. 4,b, Table II). Two weeks after vaccination, the mean ELISA values obtained using the double mutants were reduced to 19 ± 8% (SJL/J) and 17 ± 11% (BALB/c) compared with the wild-type protein. At this early time point, these data match the averaged value for the binding of all 10 mAbs to the double mutants relative to wild-type protein which was 18%. At the same time the ELISA signals obtained using the single mutant, the Ser42Pro mutant and the human protein, were significantly reduced. Among these proteins, binding to the single mutant was affected most, resulting in ELISA signals ranging from 22 to 94% relative to wild-type MOGex. However, with time, Ab binding to MOGex mutant proteins exhibited a definite trend to increase. After 11 wk, the averaged absorption values for the double mutants had increased to 28 ± 9% (SJL/J) and 32 ± 10% (BALB/c) of the wild-type values (Table II). Similarly, cross-reactivity of hMOG and S42P MOG with wild-type MOGex significantly increased between 2 and 11 wk postvaccination. These observations suggest that maturation of the MOG-specific Ab response is associated with epitope diversification, although the FG loop remains the dominant target. Analysis of sera obtained from MOG-DNA-vaccinated Dark Agouti rats indicates that the dominant role of the FG loop as a target for MOGex-specific Abs is not restricted to the mouse. The specificity pattern of the polyclonal response to MOGex in the DA rat resembled that of the mouse mAbs; binding to the single mutant was reduced to 70% and the average binding to the double mutants was reduced to 30% of wild-type MOGex (Fig. 4 b).

FIGURE 4.

Epitope specificity of the Ab repertoire of MOG-DNA vaccinated rodents. a, Time course of the MOG-specific Ab response after vaccination of SJL/J and BALB/c mice. MOG-specific Abs can be detected by ELISA using diluted plasma samples during the whole observation period (14–73 days after vaccination) with a maximum titer detected after 4–6 wk (•/▴: MOG-DNA-vaccinated SJL/J/BALB/c mice; ○/▵: SJL/J/BALB/c mice) vaccinated with control vector. Data of all mice were normalized to the 4-wk plasma sample and values from all mice of one group were averaged with the SD accounting for the individual variability in the development of the MOG-specific Ab response. b, Epitope specificity of two representative mice (SJL/J, BALB/c; 2 wk after vaccination) and Dark Agouti rats (4 wk after vaccination) determined by ELISA. Data correspond to the mean of three measurements with SDs A490 ≤ 0.03. Binding to the double mutants DM1 and DM2 is strongly reduced indicating a very focused Ab response.

FIGURE 4.

Epitope specificity of the Ab repertoire of MOG-DNA vaccinated rodents. a, Time course of the MOG-specific Ab response after vaccination of SJL/J and BALB/c mice. MOG-specific Abs can be detected by ELISA using diluted plasma samples during the whole observation period (14–73 days after vaccination) with a maximum titer detected after 4–6 wk (•/▴: MOG-DNA-vaccinated SJL/J/BALB/c mice; ○/▵: SJL/J/BALB/c mice) vaccinated with control vector. Data of all mice were normalized to the 4-wk plasma sample and values from all mice of one group were averaged with the SD accounting for the individual variability in the development of the MOG-specific Ab response. b, Epitope specificity of two representative mice (SJL/J, BALB/c; 2 wk after vaccination) and Dark Agouti rats (4 wk after vaccination) determined by ELISA. Data correspond to the mean of three measurements with SDs A490 ≤ 0.03. Binding to the double mutants DM1 and DM2 is strongly reduced indicating a very focused Ab response.

Close modal
Table II.

Epitope specificity of the Ab response of MOG-DNA-vaccinated mice over time, detected by ELISAa

WeeksWTSMDM1DM2S42PhMBSA
SJL 100 49 (±22) 17 (±9) 21 (±8) 81 (±8) 63 (±17) 4 (±5) 
 100 67 (±13) 16 (±13) 17 (±12) 96 (±4) 84 (±5) 5 (±1) 
 11 100 70 (±7) 27 (±10) 28 (±10) 101 (±4) 97 (±16) 4 (±6) 
BALB/c 100 33 (±10) 16 (±12) 19 (±10) 84 (±23) 71 (±9) 3 (±2) 
 100 63 (±13) 21 (±7) 19 (±8) 101 (±3) 83 (±9) 4 (±1) 
 11 100 76 (±10) 35 (±8) 29 (±11) 95 (±5) 89 (±8) 1 (±1) 
WeeksWTSMDM1DM2S42PhMBSA
SJL 100 49 (±22) 17 (±9) 21 (±8) 81 (±8) 63 (±17) 4 (±5) 
 100 67 (±13) 16 (±13) 17 (±12) 96 (±4) 84 (±5) 5 (±1) 
 11 100 70 (±7) 27 (±10) 28 (±10) 101 (±4) 97 (±16) 4 (±6) 
BALB/c 100 33 (±10) 16 (±12) 19 (±10) 84 (±23) 71 (±9) 3 (±2) 
 100 63 (±13) 21 (±7) 19 (±8) 101 (±3) 83 (±9) 4 (±1) 
 11 100 76 (±10) 35 (±8) 29 (±11) 95 (±5) 89 (±8) 1 (±1) 
a

Dates correspond to the time after DNA vaccination. Absorption values are normalized to the Ab response targeting wild-type rat MOG and represent the mean of seven mice with the SD accounting for the variability in the epitope specificity of individual mice.

Conformation-dependent MOG-specific autoantibodies mediate demyelination both in vitro and in vivo and are implicated in the pathogenesis of MS. However, the epitope specificity of this response is unknown and the clinical significance of MOG-specific autoantibodies in human disease remains controversial. We now report that Ag recognition by demyelinating MOGex-specific autoantibodies in rodents is focused on the two amino acids His103 and Ser104 that are located at the tip of the surface-exposed FG loop of MOG.

Analysis of the fine specificity of MOG-specific mAbs indicates that these two amino acids are part of a broader antigenic region located at the “top” of MOGex centered on the FG loop. All the mAbs used in this study bind to native, glycosylated MOGex as expressed at the cell surface with six of them (8-18C5, Y8, Y9, Z2, Z4, Z8, and Z12) recognizing purely discontinuous epitopes (22, 37). However, although the binding of mAbs Y1 and Y10 to MOGex also involves a dominant interaction with the FG loop, Y1 and Y10 additionally recognize linear epitopes delineated by the MOG-derived peptides 63–87 and 76–100 (22). The overlapping region of these peptides contains the sequence of the tip of the exposed DE loop (amino acids 76–79), which is directly adjacent to the 8-18C5 epitope (26) (Fig. 1,c). Besides, the binding of both mAbs is not influenced by the Ser42Pro mutation (Fig. 3 b). Merging these data generates a probable epitope of Y1 and Y10 that includes the FG loop amino acids His103 and Ser104 and thus overlaps with the 8-18C5 epitope, but which is shifted toward the back sheet of the Ig domain to cover the tip of the DE loop. Interestingly, the combined antigenic region includes the N-glycosylation site (Asn31) indicating that glycosylation may also influence the composition of the demyelinating MOG-specific Ab repertoire.

It should be mentioned that we used rat MOGex to study binding of mouse Abs. The amino acid sequences of rat and mouse MOGex are >95% identical (Fig. 1 a). All three nonconserved surface residues that are well-accessible to Abs are replaced by chemically similar, slightly larger residues in mouse compared with rat MOGex. Therefore, it is most likely that mouse MOGex-specific Abs also bind to rat MOGex which is confirmed by the cross-reactivity of all 10 mouse mAbs (37), especially of Y1 and Y10 that seem to bind to the DE loop that harbors two of the three mutations (Ser75Thr; Gly77Ser).

To confirm that the results obtained using MOG-specific mAbs were representative of the polyclonal response to MOGex, we investigated the ability of polyclonal MOG-specific antisera to bind to the mutated proteins. We found that mutation of His103 and Ser104 had a dramatic effect on Ag recognition by polyclonal MOGex-specific Abs induced by MOG-DNA vaccination. In two different mice strains, >80% of binding was abolished by mutation of His103 and Ser104. These results confirm that the antigenic region centered on the tip of the FG loop is the immunodominant target of the MOGex-specific Ab response in these experimental animals. However, other minor epitopes contribute to the polyclonal response to MOGex that may overlap with the immunodominant region, or be structurally distinct. One of the latter is defined by mAb Y11. This was the only Ab for which Ag recognition was independent of contributions from amino acids His103 and Ser104. However, Y11 does bind to amino acids 76–100 (22) that encompass the EF loop (amino acids 86–93), suggesting that its epitope is located at the “bottom” membrane proximal surface of MOGex.

It should be noted that despite the high level of sequence identity between rat and human MOG, we observed differences in their ability to bind MOG-specific Abs. The surface-exposed amino acids comprising the 8-18C5 epitope are fully conserved between rodent and human MOG. This identity extends to the immediate vicinity of the 8-18C5 epitope with the exception of Ser42 that is substituted by proline in the human protein. This single substitution is sufficient to significantly influence binding of 8-18C5 as demonstrated using the “humanized” Ser42Pro mutant of rat MOG. Binding is further decreased when human MOG is used as an Ag. This pattern of relative binding was also observed for five other mAbs as well as for the polyclonal Ab response to MOG. These observations suggest that substitutions affecting internal amino acids such as Tyr96Phe and Val111Met that are located in proximity to Ser/Pro42 lead to slight but significant changes of the surface structure of MOG—a hypothesis that can only be definitely proven by a three-dimensional structure of human MOG.

In general, any surface patch of a protein can be targeted by Abs (44, 45). Yet, in many cases, the autoantibody repertoire is focused on restricted parts of the corresponding autoantigens in vivo (46, 47, 48, 49). Some of these immunodominant regions of nonsequestered autoantigens were identified to be cryptotopes that are normally not exposed to the immune response but become accessible, e.g., after disassembly or proteolytic cleavage of the autoantigen, others were found to be epitopes newly generated by posttranslational modifications or oxidative damage during apoptosis (50, 51, 52, 53, 54). The restricted epitope specificity of the Ab response to MOGex is surprising as this autoantigen is sequestered within the CNS where it is unable to influence the selection of the immune repertoire (55, 56, 57). In this case, naive B cells with a wide range of potential MOG specificities should evade Ag-driven selection and become integral components of the repertoire. This leads to the obvious question of why then is the epitope specificity of conformation-dependent MOGex-specific Abs focused on one main immunogenic region of the protein? Several observations suggest that this may be due to the deletion of certain epitope specificities from the Ab repertoire by components of self proteins that mimic the corresponding MOG epitopes (58, 59).

The butyrophilin (BTN) gene family provides a large number of structural homologs of MOGex, like BTN, BTN-like proteins, or erythroid membrane-associated protein (ERMAP) with amino acid sequence identities in the range of 35–50% that can result in molecular mimicry with MOGex. Intriguingly, multiple sequence alignment of MOGex and related structures reveals that those amino acids that form the 8-18C5 epitope as defined by x-ray crystallography are among the least conserved between MOGex and its homologs expressed outside the CNS (Fig. 5). This correlation becomes even more obvious in three-dimensional space when the degree of conservation between MOGex and its homologs is mapped onto the molecular surface of MOGex (Fig. 5). The largest surface patch of nonconserved amino acids corresponds strikingly to the 8-18C5 epitope centered about the FG loop. In addition, it contains the tip of the DE loop recognized by mAbs Y1 and Y10. Though it has to be kept in mind that in general single amino acid exchanges can be sufficient for rendering a protein region antigenic, the demonstrated correspondence between antigenicity and lack of conservation is striking and may represent a simple explanation for the observed epitope specificity of conformation-dependent MOGex-specific Abs. It should be stressed that this argument applies to the development of the Ab response to the intact folded protein, not to responses to short linear epitopes that may be generated during protein degradation. Whether this correlation between “molecular foreignness” and antigenicity also applies to other sequestered autoantigens remains to be investigated.

FIGURE 5.

Concordance of MOG amino acids of the immunodominant epitope region with amino acids unique to MOG. The amino acid sequence alignment contains sequences of human Ig-V domains related to MOG that were aligned to the MOG sequence. Green arrows indicate amino acids characteristic for the Ig-V fold. Below the alignment the secondary structure elements of rat MOG are shown. Regions that form the 8-18C5 epitope on MOG are enclosed by blue frames. On the left, the molecular surface of MOGex is displayed with amino acids contacting the Ab 8-18C5 colored from orange to green according to their position in the MOG sequence. In addition, the surface of MOGex is shown with the conservation of amino acid sequences of human BTN (BTN S1(A1)) and hERMAP compared with rat MOG, mapped onto it. Conserved amino acids are not colored, the order yellow–orange–red corresponds to decreasing conservation.

FIGURE 5.

Concordance of MOG amino acids of the immunodominant epitope region with amino acids unique to MOG. The amino acid sequence alignment contains sequences of human Ig-V domains related to MOG that were aligned to the MOG sequence. Green arrows indicate amino acids characteristic for the Ig-V fold. Below the alignment the secondary structure elements of rat MOG are shown. Regions that form the 8-18C5 epitope on MOG are enclosed by blue frames. On the left, the molecular surface of MOGex is displayed with amino acids contacting the Ab 8-18C5 colored from orange to green according to their position in the MOG sequence. In addition, the surface of MOGex is shown with the conservation of amino acid sequences of human BTN (BTN S1(A1)) and hERMAP compared with rat MOG, mapped onto it. Conserved amino acids are not colored, the order yellow–orange–red corresponds to decreasing conservation.

Close modal

Although recent studies emphasize the importance of conformation-dependent MOGex-specific Abs in MS, consistent data regarding the clinical significance of MOG-specific Abs in MS do not exist. Apart from methodological difficulties concerning, for example, the preparation of the Ag MOG or the selection of patients, assays are often complicated by low sensitivity and specificity. Recently, Zhou et al. (25) reported that the mAb 8-18C5 competes with the binding of sera from MS patients to native MOG expressed on the cell surface which suggests that the 8-18C5 epitope on MOG is also recognized by human MOG-specific Abs. In case the specificity of conformation-dependent MOGex-specific Abs in humans is restricted to the same immunodominant epitope region that we report for rodents, testing the differential Ab reactivity of sera to mutant vs wild-type MOG may represent an alternative way to analyze MOG-specific Abs in MS.

We thank Nancy Ruddle for providing the plasmids of hMOG and Ser42Pro rat MOG. We are grateful to Tad Holak and Loyola d’Silva for recording the one-dimensional proton NMR spectra and to Piotr Neumann for data collection of DM2 mutant diffraction data.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

The atomic coordinates and structure factors of DM2-MOGex presented in this article have been deposited in the Protein Data Bank (PDB; www.rcsb.org) with the PDB ID code 3CSP.

2

This work was supported by the Gemeinnützige Hertie Stiftung (1.01.1/04/008, to U.J.).

4

Abbreviations used in this paper: MOG, myelin oligodendrocyte glycoprotein; EAE, experimental autoimmune encephalomyelitis; MS, multiple sclerosis; MOGex, extracellular domain of MOG; hMOG, human MOG; BTN, butyrophilin; ERMAP, erythroid membrane-associated protein.

1
Gardinier, M. V., P. Amiguet, C. Linington, J. M. Matthieu.
1992
. Myelin/oligodendrocyte glycoprotein is a unique member of the immunoglobulin superfamily.
J. Neurosci. Res.
33
:
177
-187.
2
Johns, T. G., C. C. A. Bernard.
1999
. The structure and function of myelin oligodendrocyte glycoprotein.
J. Neurochem.
72
:
1
-9.
3
Lebar, R., C. Lubetzki, C. Vincent, P. Lombrail, J. M. Boutry.
1986
. The M2 autoantigen of central nervous system myelin, a glycoprotein present in oligodendrocyte membrane.
Clin. Exp. Immunol.
66
:
423
-434.
4
Linnington, C., M. Webb, P. L. Woodhams.
1984
. A novel myelin-associated glycoprotein defined by a mouse monoclonal antibody.
J. Neuroimmunol.
6
:
387
-396.
5
Adelmann, M., J. Wood, I. Benzel, P. Fiori, H. Lassmann, J. M. Matthieu, M. V. Gardinier, K. Dornmair.
1995
. The N-terminal domain of the myelin oligodendrocyte glycoprotein (MOG) induces acute demyelinating experimental autoimmune encephalomyelitis in the Lewis rat.
J. Neuroimmunol.
63
:
17
-27.
6
Genain, C. P., M. H. Nguyen, N. L. Letvin, R. Pearl, R. L. Davis, M. Adelman, M. B. Lees, C. Linington, S. L. Hauser.
1995
. Antibody facilitation of multiple sclerosis-like lesions in a nonhuman primate.
J. Clin. Invest.
96
:
2966
-2974.
7
Iglesias, A., J. Bauer, T. Litzenburger, A. Schubart, C. Linington.
2001
. T- and B-cell responses to myelin oligodendrocyte glycoprotein in experimental autoimmune encephalomyelitis and multiple sclerosis.
Glia
36
:
220
-234.
8
Cross, A. H., J. L. Trotter, J. A. Lyons.
2001
. B cells and antibodies in CNS demyelinating disease.
J. Neuroimmunol.
112
:
1
-14.
9
Genain, C. P., B. Cannella, S. L. Hauser, C. S. Raine.
1999
. Identification of autoantibodies associated with myelin damage in multiple sclerosis.
Nat. Med.
5
:
170
-175.
10
O'Connor, K. C., H. Appel, L. Bregoli, M. E. Call, I. Catz, J. A. Chan, N. H. Moore, K. G. Warren, S. J. Wong, D. A. Hafler, K. W. Wucherpfennig.
2005
. Antibodies from inflamed central nervous system tissue recognize myelin oligodendrocyte glycoprotein.
J. Immunol.
175
:
1974
-1982.
11
De March, A. K., M. De Bouwerie, M. N. Kolopp-Sarda, G. C. Faure, M. C. Bene, C. C. A. Bernard.
2003
. Anti-myelin oligodendrocyte glycoprotein B-cell responses in multiple sclerosis.
J. Neuroimmunol.
135
:
117
-125.
12
Gaertner, S., K. de Graaf, B. Greve, R. Weissert.
2004
. Antibodies against glycosylated native MOG are elevated in patients with multiple sclerosis.
Neurology
63
:
2381
-2383.
13
Sun, J., H. Link, T. Olsson, B. G. Xiao, G. Andersson, H. P. Ekre, C. Linington, P. Diener.
1991
. T and B cell responses to myelin-oligodendrocyte glycoprotein in multiple sclerosis.
J. Immunol.
146
:
1490
-1495.
14
Berger, T., P. Rubner, F. Schautzer, R. Egg, H. Ulmer, I. Mayringer, E. Dilitz, F. Deisenhammer, M. Reindl.
2003
. Antimyelin antibodies as a predictor of clinically definite multiple sclerosis after a first demyelinating event.
N. Engl. J. Med.
349
:
139
-145.
15
Vojdani, A., E. Vojdani, E. Cooper.
2003
. Antibodies to myelin basic protein, myelin oligodendrocytes peptides, α-β-crystallin, lymphocyte activation and cytokine production in patients with multiple sclerosis.
J. Intern. Med.
254
:
363
-374.
16
Lalive, P. H., T. Menge, C. Delarasse, B. Della Gaspera, D. Pham-Dinh, P. Villoslada, H. C. von Budingen, C. P. Genain.
2006
. Antibodies to native myelin oligodendrocyte glycoprotein are serologic markers of early inflammation in multiple sclerosis.
Proc. Natl. Acad. Sci. USA
103
:
2280
-2285.
17
Haase, C. G., J. Guggenmos, U. Brehm, M. Andersson, T. Olsson, M. Reindl, J. M. Schneidewind, U. K. Zettl, F. Heidenreich, T. Berger, et al
2001
. The fine specificity of the myelin oligodendrocyte glycoprotein autoantibody response in patients with multiple sclerosis and normal healthy controls.
J. Neuroimmunol.
114
:
220
-225.
18
Lampasona, V., D. Franciotta, R. Furlan, S. Zanaboni, R. Fazio, E. Bonifacio, G. Comi, G. Martino.
2004
. Similar low frequency of anti-MOG IgG and IgM in MS patients and healthy subjects.
Neurology
62
:
2092
-2094.
19
Lim, E. T., T. Berger, M. Reindl, C. M. Dalton, K. Fernando, G. Keir, E. J. Thompson, D. H. Miller, G. Giovannoni.
2005
. Anti-myelin antibodies do not allow earlier diagnosis of multiple sclerosis.
Mult. Scler.
11
:
492
-494.
20
Reindl, M., C. Linington, U. Brehm, R. Egg, E. Dilitz, F. Deisenhammer, W. Poewe, T. Berger.
1999
. Antibodies against the myelin oligodendrocyte glycoprotein and the myelin basic protein in multiple sclerosis and other neurological diseases: a comparative study.
Brain
122
:
2047
-2056.
21
Markovic, M., V. Trajkovic, J. Drulovic, S. Mesaros, N. Stojsavljevic, I. Dujmovic, M. M. Stojkovic.
2003
. Antibodies against myelin oligodendrocyte glycoprotein in the cerebrospinal fluid of multiple sclerosis patients.
J. Neurol. Sci.
211
:
67
-73.
22
Brehm, U., S. J. Piddlesden, M. V. Gardinier, C. Linington.
1999
. Epitope specificity of demyelinating monoclonal autoantibodies directed against the human myelin oligodendrocyte glycoprotein (MOG).
J. Neuroimmunol.
97
:
9
-15.
23
von Budingen, H. C., S. L. Hauser, A. Fuhrmann, C. B. Nabavi, J. I. Lee, C. P. Genain.
2002
. Molecular characterization of antibody specificities against myelin/oligodendrocyte glycoprotein in autoimmune demyelination.
Proc. Natl. Acad. Sci. USA
99
:
8207
-8212.
24
von Budingen, H. C., S. L. Hauser, J. C. Ouallet, N. Tanuma, T. Menge, C. P. Genain.
2004
. Frontline: epitope recognition on the myelin/oligodendrocyte glycoprotein differentially influences disease phenotype and antibody effector functions in autoimmune demyelination.
Eur. J. Immunol.
34
:
2072
-2083.
25
Zhou, D., R. Srivastava, S. Nessler, V. Grummel, N. Sommer, W. Bruck, H. P. Hartung, C. Stadelmann, B. Hemmer.
2006
. Identification of a pathogenic antibody response to native myelin oligodendrocyte glycoprotein in multiple sclerosis.
Proc. Natl. Acad. Sci. USA
103
:
19057
-19062.
26
Breithaupt, C., A. Schubart, H. Zander, A. Skerra, R. Huber, C. Linington, U. Jacob.
2003
. Structural insights into the antigenicity of myelin oligodendrocyte glycoprotein.
Proc. Natl. Acad. Sci. USA
100
:
9446
-9451.
27
Collaborative Computational Project.
1994
. The CCP4 suite–programs for protein crystallography.
Acta Crystallogr. D
50
:
760
-763.
28
Jones, T. A., J. Y. Zou, S. W. Cowan, M. Kjeldgaard.
1991
. Improved methods for building protein models in electron-density maps and the location of errors in these models.
Acta Crystallogr. A
47
:
110
-119.
29
Nicholls, A., K. A. Sharp, B. Honig.
1991
. Protein folding and association–insights from the interfacial and thermodynamic properties of hydrocarbons.
Proteins
11
:
281
-296.
30
Weiner, S. J., P. A. Kollman, D. A. Case, U. C. Singh, C. Ghio, G. Alagona, S. Profeta, P. Weiner.
1984
. A new force-field for molecular mechanical simulation of nucleic-acids and proteins.
J. Am. Chem. Soc.
106
:
765
-784.
31
Oliver, A. R., G. M. Lyon, N. H. Ruddle.
2003
. Rat and human myelin oligodendrocyte glycoproteins induce experimental autoimmune encephalomyelitis by different mechanisms in C57BL/6 mice.
J. Immunol.
171
:
462
-468.
32
Kabsch, W..
1993
. Automatic processing of rotation diffraction data from crystals of initially unknown symmetry and cell constants.
J. Appl. Crystallogr.
26
:
795
-800.
33
McCoy, A. J., R. W. Grosse-Kunstleve, P. D. Adams, M. D. Winn, L. C. Storoni, R. J. Read.
2007
. Phaser crystallographic software.
J. Appl. Crystallogr.
40
:
658
-674.
34
Brunger, A. T., P. D. Adams, G. M. Clore, W. L. DeLano, P. Gros, R. W. Grosse-Kunstleve, J. S. Jiang, J. Kuszewski, M. Nilges, N. S. Pannu, et al
1998
. Crystallography and NMR system: a new software suite for macromolecular structure determination.
Acta Crystallogr. D
54
:
905
-921.
35
Engh, R. A., R. Huber.
1991
. Accurate bond and angle parameters for x-ray protein-structure refinement.
Acta Crystallogr. A
47
:
392
-400.
36
Bourquin, C., A. Iglesias, T. Berger, H. Wekerle, C. Linington.
2000
. Myelin oligodendrocyte glycoprotein-DNA vaccination induces antibody-mediated autoaggression in experimental autoimmune encephalomyelitis.
Eur. J. Immunol.
30
:
3663
-3671.
37
Piddlesden, S. J., H. Lassmann, F. Zimprich, B. P. Morgan, C. Linington.
1993
. The demyelinating potential of antibodies to myelin oligodendrocyte glycoprotein is related to their ability to fix complement.
Am. J. Pathol.
143
:
555
-564.
38
Kleywegt, G. J., T. A. Jones.
1997
. Detecting folding motifs and similarities in protein structures.
Methods Enzymol.
277
:
525
-545.
39
Kraulis, P. J..
1991
. Molscript–a program to produce both detailed and schematic plots of protein structures.
J. Appl. Crystallogr.
24
:
946
-950.
40
Merritt, E. A., M. E. P. Murphy.
1994
. Raster3d version-2.0–a program for photorealistic molecular graphics.
Acta Crystallogr. D
50
:
869
-873.
41
Barton, G. J..
1993
. ALSCRIPT: a tool to format multiple sequence alignments.
Protein Eng.
6
:
37
-40.
42
Clements, C. S., H. H. Reid, T. Beddoe, F. E. Tynan, M. A. Perugini, T. G. Johns, C. C. A. Bernard, J. Rossjohn.
2003
. The crystal structure of myelin oligodendrocyte glycoprotein, a key autoantigen in multiple sclerosis.
Proc. Natl. Acad. Sci. USA
100
:
11059
-11064.
43
Rehm, T., R. Huber, T. A. Holak.
2002
. Application of NMR in structural proteomics: screening for proteins amenable to structural analysis.
Structure
10
:
1613
-1618.
44
Davies, D. R., G. H. Cohen.
1996
. Interactions of protein antigens with antibodies.
Proc. Natl. Acad. Sci. USA
93
:
7
-12.
45
Newman, M. A., C. R. Mainhart, C. P. Mallett, T. B. Lavoie, S. J. Smith-Gill.
1992
. Patterns of antibody specificity during the BALB/c immune response to hen egg white lysozyme.
J. Immunol.
149
:
3260
-3272.
46
Henriksson, E. W., M. Wahren-Herlenius, I. Lundberg, E. Mellquist, I. Pettersson.
1999
. Key residues revealed in a major conformational epitope of the U1–70K protein.
Proc. Natl. Acad. Sci. USA
96
:
14487
-14492.
47
Hudson, B. G., K. Tryggvason, M. Sundaramoorthy, E. G. Neilson.
2003
. Alport’s syndrome, Goodpasture’s syndrome, and type IV collagen.
N. Engl. J. Med.
348
:
2543
-2556.
48
Mahler, M., M. Bluthner, K. M. Pollard.
2003
. Advances in B-cell epitope analysis of autoantigens in connective tissue diseases.
Clin. Immunol.
107
:
65
-79.
49
Zillikens, D., P. A. Rose, S. D. Balding, Z. Liu, M. Olaguemarchan, L. A. Diaz, G. J. Giudice.
1997
. Tight clustering of extracellular Bp180 epitopes recognized by bullous pemphigoid autoantibodies.
J. Invest. Dermatol.
109
:
573
-579.
50
Plotz, P. H..
2003
. The autoantibody repertoire: searching for order.
Nat. Rev. Immunol.
3
:
73
-78.
51
Cloos, P. A. C., S. Christgau.
2004
. Post-translational modifications of proteins: implications for aging, antigen recognition, and autoimmunity.
Biogerontology
5
:
139
-158.
52
Schellekens, G. A., H. Visser, B. A. de Jong, F. H. van den Hoogen, J. M. Hazes, F. C. Breedveld, W. J. van Venrooij.
2000
. The diagnostic properties of rheumatoid arthritis antibodies recognizing a cyclic citrullinated peptide.
Arthritis Rheum.
43
:
155
-163.
53
Schellekens, G. A., B. A. de Jong, F. H. van den Hoogen, L. B. van de Putte, W. J. van Venrooij.
1998
. Citrulline is an essential constituent of antigenic determinants recognized by rheumatoid arthritis-specific autoantibodies.
J. Clin. Invest.
101
:
273
-281.
54
Schachna, L., F. M. Wigley, S. Morris, A. C. Gelber, A. Rosen, L. Casciola-Rosen.
2002
. Recognition of granzyme B-generated autoantigen fragments in scleroderma patients with ischemic digital loss.
Arthritis Rheum.
46
:
1873
-1884.
55
Litzenburger, T., R. Fassler, J. Bauer, H. Lassmann, C. Linington, H. Wekerle, A. Iglesias.
1998
. B lymphocytes producing demyelinating autoantibodies–development and function in gene-targeted transgenic mice.
J. Exp. Med.
188
:
169
-180.
56
Fazilleau, N., C. Delarasse, C. H. Sweenie, S. M. Anderton, S. Fillatreau, F. A. Lemonnier, D. Pham-Dinh, J. M. Kanellopoulos.
2006
. Persistence of autoreactive myelin oligodendrocyte glycoprotein (MOG)-specific T cell repertoires in MOG-expressing mice.
Eur. J. Immunol.
36
:
533
-543.
57
Delarasse, C., P. Daubas, L. T. Mars, C. Vizler, T. Litzenburger, A. Iglesias, J. Bauer, B. Della Gaspera, A. Schubart, L. Decker, et al
2003
. Myelin/oligodendrocyte glycoprotein-deficient (MOG-deficient) mice reveal lack of immune tolerance to MOG in wild-type mice.
J. Clin. Invest.
112
:
544
-553.
58
Litzenburger, T., H. Bluthmann, P. Morales, D. Pham-Dinh, A. Dautigny, H. Wekerle, A. Iglesias.
2000
. Development of myelin oligodendrocyte glycoprotein autoreactive transgenic B lymphocytes: receptor editing in vivo after encounter of a self-antigen distinct from myelin oligodendrocyte glycoprotein.
J. Immunol.
165
:
5360
-5366.
59
Krishnamoorthy, G., H. Lassmann, H. Wekerle, A. Holz.
2006
. Spontaneous opticospinal encephalomyelitis in a double-transgenic mouse model of autoimmune T cell B cell cooperation.
J. Clin. Invest.
116
:
2385
-2392.