Autoantibodies to myelin oligodendrocyte glycoprotein (MOG) can induce demyelination and oligodendrocyte loss in models of multiple sclerosis (MS). Whether anti-MOG Abs play a similar role in patients with MS or inflammatory CNS diseases by epitope spreading is unclear. We have therefore examined whether autoantibodies that bind properly folded MOG protein are present in the CNS parenchyma of MS patients. IgG was purified from CNS tissue of 14 postmortem cases of MS and 8 control cases, including cases of encephalitis. Binding was assessed using two independent assays, a fluorescence-based solid-phase assay and a solution-phase RIA. MOG autoantibodies were identified in IgG purified from CNS tissue by solid-phase immunoassay in 7 of 14 cases with MS and 1 case of subacute sclerosing panencephalitis, but not in IgG from noninflamed control tissue. This finding was confirmed with a solution-phase RIA, which measures higher affinity autoantibodies. These data demonstrate that autoantibodies recognizing MOG are present in substantially higher concentrations in the CNS parenchyma compared with cerebrospinal fluid and serum in subjects with MS, indicating that local production/accumulation is an important aspect of autoantibody-mediated pathology in demyelinating CNS diseases. Moreover, chronic inflammatory CNS disease may induce autoantibodies by virtue of epitope spreading.

Multiple sclerosis (MS)4 is an inflammatory disease of the CNS myelin thought to be of autoimmune origin. The human disease is modeled in mice by experimental autoimmune encephalomyelitis (EAE) induced by immunization with several distinct myelin Ags, including myelin oligodendrocyte glycoprotein (MOG). The presentation of peptides derived from these Ags to CD4+ T cells is a minimal requirement for the development of inflammatory foci in the CNS (1). B cells and Abs are not absolutely required for the development of EAE, as the disease can be induced in B cell-deficient mice (2), and spontaneous lesions develop in myelin basic protein (MBP) TCR transgenic RAG-1−/− mice, which lack B cells (3). However, autoantibodies to myelin Ags may nevertheless be highly relevant in CNS demyelinating diseases in the more complex setting of an intact immune system. Moreover, it has been demonstrated that even in primary experimental viral diseases of the CNS, epitope spreading occurs such that activated myelin-reactive T cells are induced with the potential for inducing further CNS damage (4).

The potential of MOG autoantibodies to induce severe demyelination and oligodendrocyte loss was convincingly demonstrated by passive transfer of anti-MOG mAb (8-18C5) into mice or rats with mild clinical signs of EAE (5, 6), and demyelination was shown to be dependent on the coordinate action of myelin-specific T cells and autoantibodies. Abs to MOG also appear to play an important role in the chronic, relapsing-remitting disease process in the marmoset model of EAE induced by immunization with the extracellular (EC) domain of MOG (7, 8). Large demyelinated lesions resembling MS plaques were found in this model, characterized by deposition of complement components and Ig (including those directed to MOG) and the uptake of myelin debris by macrophages. MOG-specific Abs participate in attacking the myelin membrane by triggering complement- and Ab-dependent cellular cytotoxicity-dependent effector mechanisms (9, 10).

Substantial evidence for an abnormal humoral immune response in the CNS of patients with MS (11, 12) includes intrathecal Ig synthesis (13) and the presence of IgG oligoclonal bands in the cerebrospinal fluid (14, 15). In addition, B cells and plasma cells have been identified in MS lesions (16, 17), along with Abs (9, 18), Ig transcripts (19), and evidence of B cell differentiation (20). Molecular studies designed to characterize the V regions (VH) of IgG expressed in MS plaques (21, 22, 23, 24) and cerebrospinal fluid (25, 26, 27) revealed that the V regions are biased in family representation, extensively mutated, and oligoclonal. These characteristics strongly indicate that the B cells in the cerebrospinal fluid and brain lesions of patients with MS have undergone T cell-mediated, Ag-driven clonal expansion.

Little is known regarding the specificity of anti-myelin Abs derived from the CNS plaque tissue of patients with MS or subjects with encephalitis. One elegant study examined deposition of MOG Abs in MS plaque tissue. By using gold-labeled peptides of MOG and MBP, Abs against both myelin Ags could be visualized in vesicular degenerating myelin in active MS plaques and in lesions of the marmoset model of MS (10). However, this technique could not detect Abs to conformation-sensitive epitopes on folded MOG protein (rather than linear peptide epitopes) in MS lesions. Several recent studies in human MS (28) and the EAE model (29, 30, 31) highlight the importance of protein conformation for MOG autoantibody binding and go on to demonstrate that pathogenic Abs bind to conformation-dependent epitopes. Thus, the tertiary structure of the autoantigen needs to be considered in an evaluation of MOG autoantibodies in CNS inflammatory disease.

Essential criteria for identifying pathogenic autoantibodies are that they can be found at the site of injury and that they recognize the native form of the target Ag. In this study, we examined the hypothesis that autoantibodies recognizing MOG are produced or accumulate in the CNS of patients with MS and possibly other chronic inflammatory CNS diseases. IgG was isolated from the CNS parenchyma of autopsy cases with MS, other inflammatory CNS diseases, and noninflammatory controls and examined for autoantibody binding with two different preparations of folded MOG protein. Serum and cerebrospinal fluid were also examined from patients with MS and control subjects. Our data are the first to demonstrate that Abs recognizing folded MOG protein can be found in the CNS parenchyma of patients with chronic CNS inflammation.

All specimens were obtained between 1999 and 2004 and were stored at −80°C until use. A diagnosis of MS based on the criteria of McDonald (32) was confirmed in the subjects from whom they were taken. All studies were approved by the Brigham and Women’s Hospital governing Institutional Review Board committees. Eight of the MS CNS tissue samples were collected at the Department of Medicine, Multiple Sclerosis Patient Care and Research Clinic, University of Alberta. All of these patients had secondary-progressive MS; five were female and three male. Two other sets of samples were collected at the Brigham and Women’s Hospital. Both were from female patients with progressive MS. The remaining three MS CNS IgG samples and the two samples of subacute sclerosing panencephalitis (SSPE) CNS IgG were provided by D. Gilden and G. Owens of the University of Colorado Health Science Center (Denver, CO). Details regarding these specimens have been previously reported (33, 34). Normal CNS tissue was collected in the Department of Pathology at Brigham and Women’s Hospital and consisted of normal-appearing white matter obtained postmortem from individuals without neurological disease. Also included were two other inflammatory neurological disease control CNS tissue samples from a patient with cerebral aspergillosis and from another with acute and chronic meningoencephalitis.

Serum and cerebrospinal fluid samples were collected at the Brigham and Women’s Hospital’s Multiple Sclerosis Clinic or the Department of Pathology. Serum and cerebrospinal fluid were from sources distinct from the sources of CNS tissue. Serum was collected from patients with MS (n = 37). Of the 37 serum samples, 20 were from individuals with relapsing-remitting MS (RRMS), 12 were from patients with secondary progressive MS, and 5 were from patients with primary progressive MS (PPMS). Nineteen matching cerebrospinal fluid samples were included: 17 from patients diagnosed with RRMS, and 2 with PPMS. Also included were 19 additional unmatched cerebrospinal fluid samples: from 17 individuals diagnosed with RRMS and 2 with PPMS. Normal control sera were included (n = 46): 33 from normal healthy donors, and 13 from patients from whom cerebrospinal fluid was also collected. These 13 matched serum/cerebrospinal fluid and an additional 16 unmatched cerebrospinal fluid samples were from individuals who had diagnostic lumbar punctures; measured components were within normal limits, and no diagnoses were confirmed. Twenty-five matched cerebrospinal fluid and serum samples from patients diagnosed with encephalitis were provided by the Wadsworth Center, New York State Department of Health. These samples were collected during the acute phase of the illness. All sera and spinal fluids from encephalitis patients were tested under conditions approved by the Institutional Review Board of the New York State Department of Health.

CNS tissue sections 5–7 μm thick were fixed in Formalin, embedded in paraffin, and stained by H&E and luxol fast blue. Immunocytochemistry was performed on paraffin sections using anti-CD20, anti-CD138, anti-CD4, and anti-CD8 mAbs from DakoCytomation. Binding of these primary mAbs was detected with an anti-mouse Ig Ab conjugated to the HRP, using diaminobenzidine as the HRP substrate (3,3′-diaminobenzidine). Slides were washed with water, counterstained with hematoxylin, dehydrated, and mounted with Poly Mount (Poly Scientific).

For isolation of Igs, frozen blocks of CNS tissue from MS and control cases were finely cut, washed extensively in PBS, and homogenized in ice-cold PBS buffer (PBS containing 1.0% Nonidet P-40 and a mixture of protease inhibitors) using a glass Dounce apparatus (0.1 g of tissue per milliliter of buffer). Insoluble material was removed by centrifugation at 20,000 × g for 20 min. The pellet was rehomogenized and centrifuged two additional times. The supernatants were pooled and clarified by centrifugation at 60,000 × g for 60 min. The clarified supernatant was loaded onto a column containing 1 ml of protein A-Sepharose (Pharmacia) fitted with a 3-ml glycine-blocked Sepharose CL4B precolumn. Following loading, the column was washed with 40 bed volumes of PBS, and the IgG was then eluted with an acetic acid buffer (100 mM acetic acid, 150 mM sodium chloride). The eluted material was immediately neutralized by addition of 500 mM sodium bicarbonate (pH 8.5). The IgG was further purified by application to 400 μl of protein A Poros cartridge (Applied Biosystems) connected to an HPLC. Samples were loaded in running buffer (50 mM sodium bicarbonate, pH 8.5), and the IgG was eluted by injection of a solution containing 12 mM HCl and 150 mM sodium chloride. The eluted samples were immediately neutralized by addition of 500 mM sodium bicarbonate, aliquoted, and frozen at −80°C. The purity of the isolated IgG was assessed by SDS-PAGE under reducing conditions. The IgG purification procedure for the three MS and two SSPE specimens provided by D. Gilden and G. Owens has been reported previously (34).

IgG was measured using a sandwich ELISA to enable normalization in the autoantibody detection assays. The wells of 96-well microtiter plates (Immulon 4HBX; ThermoLabsystems) were coated for 1 h at room temperature with 50 μl of affinity-purified goat anti-human Ig (Chemicon International) adjusted to 10 μg/ml in 100 mM carbonate buffer (pH 8.3). Plates were blocked for 1 h at room temperature with PBS, 5% BSA, and 0.1% Tween 20. Samples were serially diluted in blocking solution, added to the plates, and incubated overnight at 4°C. A peroxidase-conjugated goat Ab to human IgG (Chemicon International) was added and incubated for 1 h at room temperature. Substrate solution containing tetramethylbenzidine (Microwell; Kirkegaard & Perry Laboratories) was added to the plates, and the color was allowed to develop for 5 min before being stopped by the addition of phosphoric acid. The IgG concentration in each sample was determined by relating the OD to that of the human IgG standard.

Lysozyme isolated from human neutrophils (Sigma-Aldrich), hepatitis C virus core Ag aa 2–192 (Research Diagnostics), and HIV gp160 (Chemicon International) were used as control Ags.

The EC (aa 1–121) of human MOG (35) used in the solid-phase dissociation-enhanced lanthanide fluorescence immunoassay (DELFIA) assay was expressed in Escherichia coli and was refolded from inclusion bodies. The construct represented the EC domain of human MOG (aa 1–121), with addition of an N-terminal methionine and an 8-residue C-terminal segment representing a XhoI restriction site (Leu-Glu) and a His6-tag. The construct was cloned into the NdeI-XhoI sites of pET-22b (Novagen), and this plasmid was used to transform BL21 DE3 E. coli cells. Protein expression was induced by addition of isopropyl β-d-thiogalactoside (1 mM final concentration) into log-phase cultures, and cells were collected by centrifugation 4 h following induction. The His6-tagged protein was isolated by Ni-NTA chromatography and refolded by rapid dilution to 50 μg/ml into a buffer composed of 100 mM Tris-HCl, 400 mM arginine, 2 mM EDTA, pH 8.0, supplemented with 5 mM reduced glutathione, 0.5 mM oxidized glutathione, and 0.5 mM PMSF. Following incubation for 24 h at 4°C, the protein was concentrated by ultrafiltration (Centricon Plus-80; Amicon), and the buffer was changed to 20 mM Tris, pH 8.0. Further purification was performed by HPLC with a POROS HQ/M strong anion exchange column with a linear NaCl gradient (0–1 M) at pH 8.0 (20 mM Tris). The protein was then dialyzed against cold PBS, and its purity was verified by SDS-PAGE.

The EC domain of human MOG used in the solution-phase RIA was expressed in an in vitro translation system. Posttranslational modifications and processing were achieved through targeting of the translated MOG to endoplasmic reticulum (ER) membranes, as described (36). The EC domain of human MOG was subcloned into a vector derived from pSP64 (Promega) for the generation of RNA transcripts via the T7 promoter. The construct included a Kozak consensus sequence (GCC-GCC-ACC) engineered immediately upstream of the initiation codon to enhance translation (37). This was followed by the endogenous signal peptide for human MOG (MASLSRPSLPSCLCSFLLLLLLQVSSSYA), immediately followed by two copies of the EC of human MOG (residues 1–121; GQFR… PFYW) (38) that were joined by a flexible linker (SRGGGGSGGGGSGGGGSEL) (39).

All in vitro translation reactions were performed at 30°C. Each 25-μl reaction contained 15 μl of nuclease-treated rabbit reticulocyte lysate (Promega), 0.5 μl of amino acid mixture minus methionine (Promega), 0.5 μl of SUPERase-In RNase inhibitor (Ambion), 2.0 μl of 35S-labeled methionine (Amersham), 2.0 μl of EBV-transformed human B cell ER microsomes, and 200 ng of RNA. Reactions were performed with an initial translation period of 15 min at 30°C under reducing conditions (rabbit reticulocyte lysate contains DTT), followed by a 1-h incubation period after addition of oxidized glutathione to 4 mM, during which disulfide bonds form.

CD spectra were recorded with a Jasco spectropolarimeter at 25°C at the Department of Biochemistry, Tufts University School of Medicine. The cell path length was 0.1 cm. rMOG was analyzed at 0.2 mg/ml in PBS at a neutral pH. CD results, reported as mean residue weight ellipticity (θ), were the mean of four spectra.

For detection of human autoantibodies, Maxisorb low fluorescence 96-well plates (PerkinElmer-Wallac) were coated overnight at 4°C with 250 ng of Ag in 50 μl of PBS. After coating, wells were washed in 50 mM Tris, pH 7.8, 150 mM NaCl containing 0.05% Tween 20, 20 μM EDTA, and 0.5% Triton X-100, and were blocked for 1 h at room temperature with DELFIA assay buffer (PerkinElmer-Wallac) supplemented with 1% BSA and 0.5% Triton X-100. For measurement of autoantibodies to MOG, purified IgG was normalized to 10 μg/ml, serum was diluted 1/1200, and cerebrospinal fluid was normalized for IgG concentration (10 μg/ml); dilutions were made in blocking buffer, and Abs were allowed to bind to the Ag-coated wells at 4°C overnight. Bound Abs were detected by incubation with a biotin-conjugated polyclonal goat anti-human IgG (Chemicon International) for 1 h at room temperature, followed by incubation with streptavidin-labeled europium (PerkinElmer-Wallac) for 1 h at room temperature. The time-resolved fluorescence signal was measured following addition of enhancement solution (PerkinElmer-Wallac) and was expressed as counts per second.

All RIAs were performed in duplicate. Samples being compared were analyzed on the same assay plate, which also included positive and negative controls. In each Ab analysis, 35S-labeled Ag (15,000–20,000 cpm) was incubated with IgG (concentrations, as indicated in Results) overnight at 4°C in buffer (20 mM Tris, pH 7.4, 150 mM NaCl, 1.0% BSA, and 0.1% Tween 20) in a total volume of 50 μl. Ig or Ig-Ag complexes were then precipitated with 50 μl of a 50% protein-A Sepharose suspension in a MultiScreen-DP opaque 96-well filtration plate (Millipore) or in microcentrifuge tubes. The mixtures were shaken for 1 h at 4°C and washed with cold buffer for two cycles (each cycle with three washes and 5 min of shaking at 4°C between cycles) using the Millipore vacuum-operated 96-well plate washer. After washing, 100 μl of scintillation liquid (Microscint-20; Packard Instrument) was added to each well or tube, and the amount of radioactivity was determined.

Analysis of immunoprecipitated Ag by SDS-PAGE was performed by boiling the washed protein A beads in 20 μl of 1× SDS-PAGE sample buffer, followed by loading onto a 12% gel. Separated proteins were transferred to a polyvinylidene difluoride membrane, followed by exposure of a phosphor imaging screen. Subsequent analysis with a Storm phosphor image analyzer (Amersham Pharmacia Biotech) allowed quantification of immunoprecipitated MOG.

A challenge with the examination of MS brain tissue for Ig is the limited availability of histologically well-characterized postmortem MS tissue, because relatively large quantities (>1 g) are required for IgG purification. IgG was isolated from tissue blocks that had been frozen at −80°C. In the MS cases, lesions could be identified on the surface of these blocks. Extensive histological analysis of MS CNS tissue sample MS11 (Table I) is shown. Histological examination confirmed the presence of a plaque with substantial demyelination (loss of luxol fast blue staining; Fig. 1,A) and hallmark perivascular cuffing (Fig. 1,B). Immunochemical characterization of the demyelinated area demonstrated CD19+ B cells at a high density within the perivascular infiltrates (Fig. 1,C) and CD138+ plasma cells in the parenchyma (Fig. 1,D). Both CD4+ and CD8+ T cells were present at a modest density (Fig. 1, E and F). Postmortem pathology reports on the remaining MS samples confirmed the presence of characteristic MS lesions, but no staining for B cell subsets or other immunohistochemistry was available.

Table I.

Yield of IgG isolated from CNS Samplesa

Sample NumberMacroscopic Tissue CharacteristicsTissue Weight (g)IgG Amount Isolated (μg)IgG (μg/g) of Tissue
MS cases     
 MS 1 75% WM 128 128 
 MS 2 >90% WM 2.72 480 176.5 
 MS 3 >50% WM 3.12 380 121.9 
 MS 4 >90% WM 3.2 650 203.1 
 MS 5 (all) >75% WM 3.9 220 97 
 MS 6 Visible lesion WM 3.7 680 183.8 
 MS 7 WM 3.7 504 136.1 
 MS 8 (all) 75% WM multiple lesions 3.2 1978 618 
 MS 9 WM multiple lesions 4.2 238 56 
 MS 10 (all) 75% WM 7.2 426 59 
 MS 11 WM and some gray matter 0.39 116 296 
 MS 12 WM possible lesion 6.61 192 29 
 MS 13 WM lesion — — — 
 MS 14 WM lesion — — — 
 MS 15 WM lesion — — — 
Normal controls     
 N3 WM 2.7 72 36.2 
 N4 WM 3.2 80 25 
 N5.2 >90% WM 7.7 160 20.8 
 N5.3 >90% WM 7.7 180 23 
 N6.1 WM 8.5 568 66 
 N6.2 >75% WM 9.6 960 100 
Other inflammatory CNS diseases     
 OIND1 WM/GM 4.74 135 28.5 
 OIND2 WM heavy vasculature 0.67 462 683.4 
 SSPE1 WM MV detected — — — 
 SSPE2 WM MV detected — — — 
Sample NumberMacroscopic Tissue CharacteristicsTissue Weight (g)IgG Amount Isolated (μg)IgG (μg/g) of Tissue
MS cases     
 MS 1 75% WM 128 128 
 MS 2 >90% WM 2.72 480 176.5 
 MS 3 >50% WM 3.12 380 121.9 
 MS 4 >90% WM 3.2 650 203.1 
 MS 5 (all) >75% WM 3.9 220 97 
 MS 6 Visible lesion WM 3.7 680 183.8 
 MS 7 WM 3.7 504 136.1 
 MS 8 (all) 75% WM multiple lesions 3.2 1978 618 
 MS 9 WM multiple lesions 4.2 238 56 
 MS 10 (all) 75% WM 7.2 426 59 
 MS 11 WM and some gray matter 0.39 116 296 
 MS 12 WM possible lesion 6.61 192 29 
 MS 13 WM lesion — — — 
 MS 14 WM lesion — — — 
 MS 15 WM lesion — — — 
Normal controls     
 N3 WM 2.7 72 36.2 
 N4 WM 3.2 80 25 
 N5.2 >90% WM 7.7 160 20.8 
 N5.3 >90% WM 7.7 180 23 
 N6.1 WM 8.5 568 66 
 N6.2 >75% WM 9.6 960 100 
Other inflammatory CNS diseases     
 OIND1 WM/GM 4.74 135 28.5 
 OIND2 WM heavy vasculature 0.67 462 683.4 
 SSPE1 WM MV detected — — — 
 SSPE2 WM MV detected — — — 
a

MS samples 5, 8, and 10 include IgG isolated from distinct blocks from the same brain. The yield reflects the results from the collective blocks (indicated as “all”). For several of the samples IgG yield was not determined (—). WM, white matter; MV, measles virus; SSPE, subacute sclerosising panencephalitis; OIND1, aspergillosis; OIND2, chronic meningoencephalitis.

FIGURE 1.

Evidence of humoral immunity in MS lesions. Histological characterization of MS CNS tissue (case MS11) reveals the presence of T cells, B cells, and plasma cells. IgG was isolated from MS CNS tissue for analysis in Ab-binding assays. A, Luxol fast blue staining of CNS tissue from case MS11 demonstrates severe demyelination and demarcation of the lesion boundary. B, H&E staining reveals perivascular cuffing of infiltrating cells within the demyelinated area. C, CD20-positive B cells are among the perivascular inflammatory infiltrate within the demyelinated area. D, Ab-producing CD138-positive plasma cells are among the inflammatory infiltrate within the demyelinated area. CD4 (E)- and CD8 (F)-positive T cells are among the inflammatory infiltrate within the demyelinated area. The black bar in the lower left of each figure represents 100 μm. G, HPLC protein A affinity column trace of IgG derived from MS CNS tissue. The IgG eluted from a preparative protein A column was injected at time 0, followed by a 0.5 M NaCl wash at 10 min, and IgG elution at 21 min by injection of 12 mM HCl, 150 mM NaCl. A second elution with 20% acetic acid at 30 min demonstrated that the majority of IgG had been collected in the initial elution. H, SDS-PAGE analysis of purified IgG from MS CNS tissue. IgG samples (1.75 μg) were reduced and separated, and the Ab H and L chains were visualized with Coomassie blue stain. Lanes 2–9, Represent IgG isolated from MS CNS samples 2–9, respectively.

FIGURE 1.

Evidence of humoral immunity in MS lesions. Histological characterization of MS CNS tissue (case MS11) reveals the presence of T cells, B cells, and plasma cells. IgG was isolated from MS CNS tissue for analysis in Ab-binding assays. A, Luxol fast blue staining of CNS tissue from case MS11 demonstrates severe demyelination and demarcation of the lesion boundary. B, H&E staining reveals perivascular cuffing of infiltrating cells within the demyelinated area. C, CD20-positive B cells are among the perivascular inflammatory infiltrate within the demyelinated area. D, Ab-producing CD138-positive plasma cells are among the inflammatory infiltrate within the demyelinated area. CD4 (E)- and CD8 (F)-positive T cells are among the inflammatory infiltrate within the demyelinated area. The black bar in the lower left of each figure represents 100 μm. G, HPLC protein A affinity column trace of IgG derived from MS CNS tissue. The IgG eluted from a preparative protein A column was injected at time 0, followed by a 0.5 M NaCl wash at 10 min, and IgG elution at 21 min by injection of 12 mM HCl, 150 mM NaCl. A second elution with 20% acetic acid at 30 min demonstrated that the majority of IgG had been collected in the initial elution. H, SDS-PAGE analysis of purified IgG from MS CNS tissue. IgG samples (1.75 μg) were reduced and separated, and the Ab H and L chains were visualized with Coomassie blue stain. Lanes 2–9, Represent IgG isolated from MS CNS samples 2–9, respectively.

Close modal

Following detergent solubilization and removal of insoluble material by high-speed centrifugation, the IgG fraction was isolated by protein A affinity chromatography. A separate protein A column was used for each tissue sample to prevent cross-contamination, and these IgG samples were concentrated and further purified on an HPLC protein A cartridge (Fig. 1,G). SDS-PAGE demonstrated that these preparations were pure, because only the IgG H and L chains were visualized by staining with Coomassie blue (Fig. 1,H). Consistent with other published studies (40), the yield of IgG/g tissue was considerably higher from the MS cases than the normal controls and ranged from 29 to 618 μg/g tissue in MS cases compared with 20.8–100 μg/g in the noninflammatory control cases (Table I). Five of the MS cases (MS2, MS4, MS6, MS8, and MS11) clearly stood out in terms of the IgG yield (176.5–618 μg/g tissue) and may have represented the cases with the most active lesions in the analyzed tissue blocks. One of the inflammatory control tissues also yielded relatively large amounts of IgG (683.4 μg/g tissue). However, this sample was highly vascularized, and consequently a fraction of the IgG may not have been synthesized within the CNS.

For detection of autoantibodies using a solid-phase method, we chose a DELFIA because this fluorescence-based assay provides higher sensitivity and a wider linear range than traditional ELISAs (41, 42). The assay does not rely on colorometric detection stopped by the user, which can introduce a subjective element and assay-to-assay variability, but rather a time-resolved fluorescence readout. Our adaptation for detection of Abs used a stringent assay buffer with 0.5% Triton X-100, which significantly reduced background and eliminated weak false positives. These conditions did not prevent the binding of physiologically relevant Abs, as we were able to detect and titer robust Ab responses to hepatitis C virus and HIV Ags in the serum and cerebrospinal fluid of individuals infected by these viruses (data not shown).

The rMOG used for the solid-phase assay was expressed in E. coli and refolded using a redox system that permits formation of the disulfide bond that stabilizes the Ig domain fold. The protein was further purified by anion-exchange HPLC and shown to represent a single band by SDS-PAGE (data not shown). First, to test whether the rMOG used in the solid-phase assay was folded, we examined its secondary structure by CD spectroscopy. The far-UV CD spectra included a minimum at 218 nm and a positive peak near 195 nm (Fig. 2). These spectral features indicate the presence of β-sheets (43), a characteristic feature of the Ig-fold.

FIGURE 2.

Postexpression folding and disulfide arrangement of rMOG generate a native secondary structure. The far-UV CD spectra of rMOG include a minima at 218 nm and a positive peak near 195 nm, characteristic of β-sheets. The curve shown is the mean of four individual spectra and is reported as mean residue weight ellipticity.

FIGURE 2.

Postexpression folding and disulfide arrangement of rMOG generate a native secondary structure. The far-UV CD spectra of rMOG include a minima at 218 nm and a positive peak near 195 nm, characteristic of β-sheets. The curve shown is the mean of four individual spectra and is reported as mean residue weight ellipticity.

Close modal

The solid-phase DELFIA assay was used for initial measurement of the levels of MOG-specific autoantibodies in purified IgG from MS and non-MS tissue. Tissue preparations positive for binding to MOG were identified as those that provided a signal at least 4-fold over two control Ags, HIV gp160 and lysozyme, in the same assay. IgG binding to MOG could be detected in 7 of 14 cases with MS (Table II and Fig. 3 A). To assess interassay reliability, IgG from MS8 was collected in five fractions, four of which had detectable levels of autoantibodies to MOG. In contrast, MOG-specific autoantibodies were not detected in the IgG isolated from the noninflammatory control CNS tissue. Using the solid-phase assay, Abs to MOG were detected in the IgG isolated from one of four cases with chronic CNS inflammation, which was a case of SSPE, a disease caused by measles virus.

Table II.

Measurement of MOG autoantibodies in IgG derived from CNS tissuea

Samplegp160 Binding (cps)MOG Binding (cps)IgG MOG Binding Titer (μg/ml)
MS cases    
 MS 2 479 ± 17 4,836 ± 253 8.0 
 MS 3 13,414 ± 209 6,171 ± 119 — 
 MS 4 1,403 ± 25 9,046 ± 118 4.0 
 MS 5.1 2,511 ± 28 2,458 ± 52 — 
 MS 5.2 360 ± 11 9,575 ± 829 2.3 
 MS 6 613 ± 40 2,196 ± 74 — 
 MS 7 610 ± 6 1,333 ± 38 — 
 MS 8.1 2,311 ± 57 1,426 ± 41 >10 
 MS 8.2 2,390 ± 45 19,958 ± 877 0.7 
 MS 8.4 2,303 ± 29 16,107 ± 237 1.2 
 MS 8.5 6,620 ± 405 28,977 ± 460 0.3 
 MS 8.6 5,793 ± 362 23,718 ± 245 0.4 
 MS 9 2,717 ± 80 4,306 ± 139 — 
 MS 10.1 2,010 ± 41 1,720 ± 26 >10 
 MS 10.2 2,624 ± 111 3,385 ± 43 — 
 MS 11 733 ± 74 2,996 ± 205 >10 
 MS 12 1,707 ± 169 6,430 ± 803 >10 
 MS 13 574 ± 8 752 ± 25 >10 
 MS 14 272 ± 55 1,908 ± 218 >10 
 MS 15 379 ± 30 2,827 ± 246 >10 
Normal Controls    
 N3 201 ± 70 692 ± 70 — 
 N4 1,443 ± 110 1,628 ± 278 — 
 N5.2 1,331 ± 88 3,993 ± 440 >10 
 N5.3 1,106 ± 54 2,716 ± 125 >10 
 N6.1 2,105 ± 32 2,150 ± 660 >10 
 N6.2 1,729 ± 91 1,622 ± 179 >10 
Other inflammatory CNS diseases    
 OIND1 1,065 ± 185 2,236 ± 280 >10 
 OIND2 933 ± 149 1,531 ± 306 >10 
 SSPE1 563 ± 9 173 ± 20 >10 
 SSPE2 324 ± 32 2,578 ± 84 >10 
Samplegp160 Binding (cps)MOG Binding (cps)IgG MOG Binding Titer (μg/ml)
MS cases    
 MS 2 479 ± 17 4,836 ± 253 8.0 
 MS 3 13,414 ± 209 6,171 ± 119 — 
 MS 4 1,403 ± 25 9,046 ± 118 4.0 
 MS 5.1 2,511 ± 28 2,458 ± 52 — 
 MS 5.2 360 ± 11 9,575 ± 829 2.3 
 MS 6 613 ± 40 2,196 ± 74 — 
 MS 7 610 ± 6 1,333 ± 38 — 
 MS 8.1 2,311 ± 57 1,426 ± 41 >10 
 MS 8.2 2,390 ± 45 19,958 ± 877 0.7 
 MS 8.4 2,303 ± 29 16,107 ± 237 1.2 
 MS 8.5 6,620 ± 405 28,977 ± 460 0.3 
 MS 8.6 5,793 ± 362 23,718 ± 245 0.4 
 MS 9 2,717 ± 80 4,306 ± 139 — 
 MS 10.1 2,010 ± 41 1,720 ± 26 >10 
 MS 10.2 2,624 ± 111 3,385 ± 43 — 
 MS 11 733 ± 74 2,996 ± 205 >10 
 MS 12 1,707 ± 169 6,430 ± 803 >10 
 MS 13 574 ± 8 752 ± 25 >10 
 MS 14 272 ± 55 1,908 ± 218 >10 
 MS 15 379 ± 30 2,827 ± 246 >10 
Normal Controls    
 N3 201 ± 70 692 ± 70 — 
 N4 1,443 ± 110 1,628 ± 278 — 
 N5.2 1,331 ± 88 3,993 ± 440 >10 
 N5.3 1,106 ± 54 2,716 ± 125 >10 
 N6.1 2,105 ± 32 2,150 ± 660 >10 
 N6.2 1,729 ± 91 1,622 ± 179 >10 
Other inflammatory CNS diseases    
 OIND1 1,065 ± 185 2,236 ± 280 >10 
 OIND2 933 ± 149 1,531 ± 306 >10 
 SSPE1 563 ± 9 173 ± 20 >10 
 SSPE2 324 ± 32 2,578 ± 84 >10 
a

A fluorescence-based assay (DELFIA) was used to evaluate Abs to MOG and a control Ag (HIV gp160) in three subject groups, patients with MS (MS), normal controls (N), and other inflammatory CNS diseases (OIND), which included SSPE. Triplicate samples of IgG were adjusted to 10 μg/ml in assay buffer before application to Ag-coated wells. Values 4-fold above the signal recorded for gp160 and lysozyme (not shown) were considered positive (values in bold) for binding to MOG. Several samples were examined further to determine the MOG-binding titer. Triplicate samples of IgG (10 μg/ml) were serially diluted and applied to Ag-coated wells. The titer is reported as the concentration of IgG (μg/ml) required to produce a signal of 4000 fluorescence counts per second (cps). Measurements not performed due to limited sample availability are indicated by a dash (—). MS samples 5, 8, 10, and control 5 and 6 include IgG isolated from distinct blocks from the same brain. Abbreviations are indicated in Table I.

FIGURE 3.

Autoantibodies to MOG are enriched in IgG derived from MS CNS tissue. A solid-phase fluorescence assay (DELFIA) was used to compare the levels of MOG-specific autoantibodies among normal donors, patients with MS, and patients with encephalitis. Each data point represents the mean from triplicate experiments. Before application to Ag-coated wells, tissue-derived IgG samples (A) were adjusted to 10 μg/ml, cerebrospinal fluid (B) was normalized such that the IgG concentration was 10 μg/ml, and serum (C) was diluted 1/1200 to approximate 10 μg/ml IgG. Samples were evaluated by calculating a cutoff equal to the mean + 3 SD of the normal donor samples. Values exceeding this were considered positive. Ab titration against immobilized MOG was performed to further characterize autoantibodies to MOG in MS CNS tissue. Serial dilutions of IgG from patients with MS (D) and non-MS controls (E) were applied to plates coated with rMOG. Due to its limited quantity, sample MS 8.2 was used at a lower initial concentration than the other samples.

FIGURE 3.

Autoantibodies to MOG are enriched in IgG derived from MS CNS tissue. A solid-phase fluorescence assay (DELFIA) was used to compare the levels of MOG-specific autoantibodies among normal donors, patients with MS, and patients with encephalitis. Each data point represents the mean from triplicate experiments. Before application to Ag-coated wells, tissue-derived IgG samples (A) were adjusted to 10 μg/ml, cerebrospinal fluid (B) was normalized such that the IgG concentration was 10 μg/ml, and serum (C) was diluted 1/1200 to approximate 10 μg/ml IgG. Samples were evaluated by calculating a cutoff equal to the mean + 3 SD of the normal donor samples. Values exceeding this were considered positive. Ab titration against immobilized MOG was performed to further characterize autoantibodies to MOG in MS CNS tissue. Serial dilutions of IgG from patients with MS (D) and non-MS controls (E) were applied to plates coated with rMOG. Due to its limited quantity, sample MS 8.2 was used at a lower initial concentration than the other samples.

Close modal

To determine whether autoantibodies to MOG were enriched in the CNS parenchyma, as compared with cerebrospinal fluid or sera, levels of MOG-specific autoantibodies in these different immune compartments were evaluated with the same assay (Fig. 3, B and C). Serum and cerebrospinal fluid were from sources distinct from the sources of CNS tissue. To compare these data with those derived from the CNS IgG, the cerebrospinal fluid was diluted to an IgG concentration of 10 μg/ml, and serum samples were diluted 1/1200 to obtain a concentration of ∼10 μg/ml IgG. A cutoff equal to the mean + 3 SD of the normal samples was used to identify positive samples. The cerebrospinal fluid and serum samples from the normal or the encephalitis group did not yield a signal above background. Two of the 38 cerebrospinal fluid samples (5.2%) and 1 of 37 serum samples (2.7%) from MS patients were positive, but the corresponding serum or cerebrospinal fluid samples from the same patients were negative. Moreover, these marginally positive samples gave significantly weaker signals than the MS CNS-derived IgG.

All IgG samples of sufficient quantity that were identified as positive for binding to MOG were further evaluated to determine the end-point titer of the autoantibodies. This also provided further validation of the results obtained with the single dilution experiments (Table II and Fig. 3, D and E). The titer of samples tested that were positive for binding to MOG ranged from 0.3 to 4.0 μg/ml and correlated with the strength of the signal in the single dilution experiment. Using these criteria, 7 of the 14 tissue samples from patients with MS and 1 of 4 tissue samples from subjects with CNS inflammation exhibited binding to MOG. None of the IgG from the noninflamed CNS tissue controls bound MOG in the solid-phase assay at equivalent Ab concentrations. IgG tissue samples providing a weaker signal in the single dilution experiment did not yield a detectable signal with up to 10 μg/ml IgG (Table II).

We and others have previously reported that solid-phase assays can detect low affinity interactions (44) and that a solution-phase assay can be used to discern which Abs may bind with higher affinity (45, 46). Thus, to further assess the nature of the binding of these IgG Abs to MOG, we used a second Ag preparation in which protein folding, disulfide bond formation, and glycosylation were induced by the relevant human enzymes. We generated 35S-labeled MOG using an in vitro translation system in which radiolabeled proteins with a signal peptide are inserted into ER microsomes isolated from a human cell line of Β cell origin. We had previously demonstrated that complex folding/assembly events occur in this system, because the TCR-CD3 complex composed of six different chains could be assembled in such microsomes (36). In the expression construct, two copies of the extracellular domain of MOG were coupled via a flexible linker to increase the avidity of Ag binding in a solution-phase assay (Fig. 4,A). In vivo, MOG expressed on the surface of myelin creates a multivalent surface for autoantibody binding. SDS-PAGE analysis of the radiolabeled protein demonstrated that MOG was indeed targeted to ER microsomes and that digestion with Endo H indicated that the protein was glycosylated (Fig. 4,B). Increased IgG autoantibody binding to MOG was observed for MS CNS-derived IgG MS 8.4, but not for IgG isolated from non-MS tissue (Fig. 5). This analysis demonstrated dose-dependent binding by the MS brain-derived IgG and a low level of background for the two controls. Only a limited number of CNS IgG samples could be analyzed in these experiments, because a relatively large fraction of the Abs had already been used. IgG autoantibody binding to MOG in serum or cerebrospinal fluid from a subset of normal controls and from patients with MS or encephalitis was not detected by this assay (data not shown).

FIGURE 4.

Cell-free expression in the presence of ER membranes produces glycosylated MOG. A, Map of the MOG dimer construct for in vitro transcription. The EC domain of human MOG was expressed as a dimer, as described in Materials and Methods. The cassette was inserted into the polylinker of the pSP64 poly(A) vector. SP6, SP6 transcription promoter; KCS, kozak consensus sequence; SP, human MOG signal peptide; ecMOG, EC domain of human MOG; FL, 19-aa flexible linker. B, Following expression, the MOG dimer translocated into ER membranes was denatured in buffer containing SDS and 2-ME, and then half was digested with endoglycosidase H. Equal amounts of the digested and nondigested material were then separated by SDS-PAGE for visualization.

FIGURE 4.

Cell-free expression in the presence of ER membranes produces glycosylated MOG. A, Map of the MOG dimer construct for in vitro transcription. The EC domain of human MOG was expressed as a dimer, as described in Materials and Methods. The cassette was inserted into the polylinker of the pSP64 poly(A) vector. SP6, SP6 transcription promoter; KCS, kozak consensus sequence; SP, human MOG signal peptide; ecMOG, EC domain of human MOG; FL, 19-aa flexible linker. B, Following expression, the MOG dimer translocated into ER membranes was denatured in buffer containing SDS and 2-ME, and then half was digested with endoglycosidase H. Equal amounts of the digested and nondigested material were then separated by SDS-PAGE for visualization.

Close modal
FIGURE 5.

IgG derived from MS CNS tissue binds to a glycosylated MOG dimer. Binding was measured using a solution-phase RIA. Titration of CNS-derived IgG against glycosylated MOG. Serial dilutions of IgG isolated from CNS tissue were incubated with radiolabeled glycosylated MOG, and bound Ag was measured through immunoprecipitation. Samples from patients with MS (MS) and non-MS controls (N) are indicated in the graph. Data points represent the mean of duplicate experiments.

FIGURE 5.

IgG derived from MS CNS tissue binds to a glycosylated MOG dimer. Binding was measured using a solution-phase RIA. Titration of CNS-derived IgG against glycosylated MOG. Serial dilutions of IgG isolated from CNS tissue were incubated with radiolabeled glycosylated MOG, and bound Ag was measured through immunoprecipitation. Samples from patients with MS (MS) and non-MS controls (N) are indicated in the graph. Data points represent the mean of duplicate experiments.

Close modal

A critical question in understanding the pathophysiology of MS and inflammatory CNS diseases is whether autoantibodies are present in the inflamed tissue. In this study, this was directly examined by eluting IgG from plaque tissue of patients with MS and subjects with other inflammatory CNS diseases. Using folded Ag, we found that a subset of cases with MS and one case of chronic virus-induced CNS inflammation had autoantibodies against MOG; in the remaining control tissues, these autoantibodies were absent. Anti-MOG autoantibodies were generally not found in the cerebrospinal fluid or serum samples of patients with MS, and those that were positive demonstrated marginal binding, indicating that such autoantibodies were present at low concentrations and/or have a low affinity. The majority of CNS samples from patients with MS yielded substantially greater quantities of IgG than those from control cases, indicative of an ongoing inflammatory process. Collectively, these data demonstrate that IgG derived from MS plaque tissue contains autoantibodies that recognize the folded human MOG protein. These anti-myelin autoantibodies, present at the site of chronic inflammation, may be produced locally by CNS-infiltrating B cells and plasma cells or might accumulate in the white matter due to binding to MOG on the surface of myelin.

In other human autoimmune diseases, it is well established that autoantibodies recognize conformational epitopes of self Ags; relevant examples are GAD65, transglutaminase, and the acetylcholine receptor (47, 48, 49). Human MOG Abs identified with denatured protein or synthetic peptides may not bind to the native protein and thus may not be capable of inducing demyelination. Indeed, the importance of binding to the native structure is highlighted by studies in the EAE model demonstrating that pathogenic Abs bind to conformation-dependent epitopes (29, 30, 31, 50). The crystal structure of MOG with the Fab of the demyelinating 8-18C5 Ab indicates that the Ab binds to a conformational epitope created by three loops on the membrane-distal side of MOG (51). The glycosylation site of MOG, Asn31, is located in a loop at the top of the membrane-distal side. The contribution of this posttranslational modification therefore needs to be considered in an evaluation of MOG autoantibodies in MS. To test our hypothesis, we used MOG protein that was either efficiently refolded from inclusion bodies or was expressed using an in vitro translation system with human ER microsomes. Both systems allow the generation of properly folded proteins and the creation of the disulfide bond that stabilizes the Ig-fold; the latter also includes glycosylation. Thus, our Ag preparations used for detection of MOG autoantibodies in serum/cerebrospinal fluid samples differed from those used by other investigators, who used MOG denatured by SDS-PAGE for Western blot analysis or MOG protein that was not refolded following isolation from E. coli under denaturing conditions. Using this properly folded MOG Ag, we could detect autoantibodies in parenchymal tissue, but not in serum or cerebrospinal fluid.

Examination of the humoral response in primary experimental viral diseases of the CNS revealed epitope spreading, such that myelin-reactive T cells are activated with the potential of causing further CNS damage (4, 52). Thus, it can be hypothesized that epitope spreading with the induction of lower affinity autoantibody will also occur with chronic viral infection. Chronic EAE is associated with extensive intra- and intermolecular epitope spreading of autoreactive B cell responses (53). Moreover, this increased diversity of autoantibody responses in acute EAE predicts a more severe clinical course. Thus, it was predicted that autoantibodies to myelin Ags would be found in the CNS tissue of patients with MS. However, the clear demonstration of epitope spreading in the Theiler’s virus model raised the question of whether lower affinity autoantibodies would be detected in subjects with chronic CNS inflammation, in which CD80 costimulatory molecules are present and tissue degradation occurs (54). We detected anti-MOG autoantibodies using the solid-phase DELFIA assay in patients with MS and in a patient with SSPE. However, using the solution-phase assay that detects higher affinity autoantibody, we found anti-MOG autoantibodies in the CNS tissue of a patient with MS, not in control subjects, including those with viral encephalitis. These data suggest that while epitope spreading may occur with CNS inflammation, higher affinity Abs may evolve with autoimmune disease. This hypothesis awaits further characterization of Ab affinity with a panel of new tissue samples.

There is considerable evidence for Ab production by clonally expanded Β cells in the CNS of MS patients, but it is not known whether the Abs are produced primarily in the CNS parenchyma or the cerebrospinal fluid space. If autoantibodies are synthesized primarily in the parenchyma and then diffuse into the cerebrospinal fluid, considerable dilution is likely, given the large volumes of cerebrospinal fluid that are produced daily. The human nervous system contains an estimated 120 ml of cerebrospinal fluid at any one time, and ∼500 ml is produced daily. The cerebrospinal fluid is transported into the venous circulation primarily through arachnoid villi, which would again result in substantial dilution of autoantibodies synthesized within the CNS (55). We therefore reasoned that it may be important to assess the presence of autoantibodies in the CNS parenchyma, because Abs relevant to the disease may be more dilute and thus more difficult to detect in the cerebrospinal fluid or serum. In fact, no signal or signals only slightly above background were detected in the serum and cerebrospinal fluid samples.

Our approach to detection of MOG autoantibodies in serum/cerebrospinal fluid differs from that taken by other investigators, who used MOG protein that was not refolded. In those studies, Ab binding was noted in a subpopulation of MS patients, as well as in a considerable fraction of patients with other neurological diseases or healthy controls (56, 57, 58). Berger et al. (59) have reported a series of studies in which they examined serum and cerebrospinal fluid samples using Western blot analysis of the rMOG extracellular domain. Abs to MOG were detected in 38% of MS patients, 53% of patients with other inflammatory CNS diseases, and 3% of patients with noninflammatory CNS diseases. Detection of Abs to MBP was also not disease specific, because they were detected in 28% of MS patients, 47% of patients with other inflammatory CNS diseases, and 60% of patients with rheumatoid arthritis. The major limitations of this method are that large amounts of recombinant protein were loaded per lane, and that no control proteins were included on these blots to assess the level of background binding, an issue made relevant by the subjective nature of colorimetric detection. A recent study by Berger et al. reported that the development of clinically definite MS can be predicted based on the presence of serum IgM Abs to MOG in patients with clinically isolated syndrome. However, a recent study by another group failed to confirm these findings (60). Differences in patient populations are often cited as the cause for such discrepancies, but the basis for incongruity in these results is likely to lie in the Ag preparation and/or the assay conditions. Our inability to detect autoantibodies to MOG using both the DELFIA and a solution-phase assay is consistent with the importance of using specific and sensitive assays of autoantibodies in subjects with autoimmune disease.

Our study is the first in which IgG isolated from CNS tissue has been examined for binding to MOG, and in which brain-derived IgG was compared with serum and cerebrospinal fluid samples. The most common patterns of MS demyelination are characterized by perivenular accumulation of T cells, macrophages, and plasma cells and perivenular demyelination (61). Because prominent IgG deposition has been observed in only a subset of MS lesions (pattern II), our finding that MOG Abs could be identified in 7 of 14 cases was not unexpected. Future studies that include extensive immunohistology (not currently available to us) will attempt to correlate the MS subtype with anti-MOG Ab reactivity. Our data also suggest that the CNS parenchyma may be one site in which myelin-specific autoantibodies are synthesized. B cells and plasma cells have been identified in active MS lesions, and there is considerable evidence for clonal expansion and differentiation of B cells in the CNS of MS patients. An alternative explanation for our findings is that serum and/or cerebrospinal fluid-derived autoantibodies accumulate in the CNS over time. If the CNS parenchyma is indeed an important site of autoantibody synthesis in MS, molecular characterization of tissue-infiltrating B cells and plasma cells by single-cell techniques may provide insights into the mechanisms of demyelination.

We thank Drs. James Baleja and Gillian D. Henry of Tufts University Sackler School of Biomedical Research for use of their CD spectropolarimeter. Several well-characterized CNS tissue specimens were kindly provided by Dr. Cynthia A. Lemere of the Center for Neurologic Disease, Brigham and Women’s Hospital, and Drs. Donald H. Gilden and Gregory P. Owens of the University of Colorado Health Science Center. We also thank Dr. Angela Vincent of the John Radcliffe Hospital in Oxford and Dr. Amit Bar-Or of the Montreal Neurological Institute for critical reading of this manuscript, and Paul Guttry of the Brigham and Women’s Hospital Editorial Service for editorial assistance.

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

These studies were supported by grants to K.W.W. (National Institutes of Health, P01 AI045757) and D.A.H. (National Institutes of Health, R01 AI39229, R01 AI44447, and P01 AI045757) and in part by a Career Transition Fellowship awarded to K.C.O. from the National Multiple Sclerosis Society (TA 3000A2/1).

4

Abbreviations used in this paper: MS, multiple sclerosis; CD, circular dichroism; DELFIA, dissociation-enhanced lanthanide fluorescence immunoassay; EAE, experimental autoimmune encephalomyelitis; EC, extracellular; ER, endoplasmic reticulum; MBP, myelin basic protein; MOG, myelin oligodendrocyte glycoprotein; PPMS, primary progressive MS; RRMS, relapsing-remitting MS; SSPE, subacute sclerosing panencephalitis.

1
Kuchroo, V. K., A. C. Anderson, H. Waldner, M. Munder, E. Bettelli, L. B. Nicholson.
2002
. T cell response in experimental autoimmune encephalomyelitis (EAE): role of self and cross-reactive antigens in shaping, tuning, and regulating the autopathogenic T cell repertoire.
Annu. Rev. Immunol.
20
:
101
-123.
2
Wolf, S. D., B. N. Dittel, F. Hardardottir, C. A. Janeway, Jr.
1996
. Experimental autoimmune encephalomyelitis induction in genetically B cell-deficient mice.
J. Exp. Med.
184
:
2271
-2278.
3
Lafaille, J. J., K. Nagashima, M. Katsuki, S. Tonegawa.
1994
. High incidence of spontaneous autoimmune encephalomyelitis in immunodeficient anti-myelin basic protein T cell receptor transgenic mice.
Cell
78
:
399
-408.
4
McMahon, E. J., S. L. Bailey, C. V. Castenada, H. Waldner, S. D. Miller.
2005
. Epitope spreading initiates in the CNS in two mouse models of multiple sclerosis.
Nat. Med.
11
:
335
-339.
5
Schluesener, H. J., R. A. Sobel, C. Linington, H. L. Weiner.
1987
. A monoclonal antibody against a myelin oligodendrocyte glycoprotein induces relapses and demyelination in central nervous system autoimmune disease.
J. Immunol.
139
:
4016
-4021.
6
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.
7
Raine, C. S., B. Cannella, S. L. Hauser, C. P. Genain.
1999
. Demyelination in primate autoimmune encephalomyelitis and acute multiple sclerosis lesions: a case for antigen-specific antibody mediation.
Ann. Neurol.
46
:
144
-160.
8
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.
9
Storch, M. K., S. Piddlesden, M. Haltia, M. Iivanainen, P. Morgan, H. Lassmann.
1998
. Multiple sclerosis: in situ evidence for antibody- and complement-mediated demyelination.
Ann. Neurol.
43
:
465
-471.
10
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.
11
O’Connor, K. C., A. Bar-Or, D. A. Hafler.
2001
. The neuroimmunology of multiple sclerosis: possible roles of T and B lymphocytes in immunopathogenesis.
J. Clin. Immunol.
21
:
81
-92.
12
Cross, A. H., J. L. Trotter, J. Lyons.
2001
. B cells and antibodies in CNS demyelinating disease.
J. Neuroimmunol.
112
:
1
-14.
13
Kabat, E. A., D. A. Freedman, J. P. Murray, V. Knaub.
1950
. A study of the crystalline albumin, γ globulin and the total protein in the cerebrospinal fluid of one hundred cases of multiple sclerosis and other diseases.
Am. J. Med. Sci.
219
:
55
-64.
14
Lowenthal, A., M. Van Sande, D. Karcher.
1960
. The differential diagnosis of neurological diseases by fractionating electrophoretically the CSF G-globulins.
J. Neurochem.
235
:
229
-233.
15
Johnson, K. P., S. C. Arrigo, B. J. Nelson, A. Ginsberg.
1977
. Agarose electrophoresis of cerebrospinal fluid in multiple sclerosis: a simplified method for demonstrating cerebrospinal fluid oligoclonal immunoglobulin bands.
Neurology
27
:
273
-277.
16
Esiri, M. M..
1977
. Immunoglobulin-containing cells in multiple-sclerosis plaques.
Lancet
2
:
478
-480.
17
Prineas, J. W., R. G. Wright.
1978
. Macrophages, lymphocytes, and plasma cells in the perivascular compartment in chronic multiple sclerosis.
Lab. Invest.
38
:
409
-421.
18
Mehta, P. D., S. Frisch, H. Thormar, W. W. Tourtellotte, H. M. Wisniewski.
1981
. Bound antibody in multiple sclerosis brains.
J. Neurol. Sci.
49
:
91
-98.
19
Lock, C., G. Hermans, R. Pedotti, A. Brendolan, E. Schadt, H. Garren, A. Langer-Gould, S. Strober, B. Cannella, J. Allard, et al
2002
. Gene-microarray analysis of multiple sclerosis lesions yields new targets validated in autoimmune encephalomyelitis.
Nat. Med.
8
:
500
-508.
20
Corcione, A., S. Casazza, E. Ferretti, D. Giunti, E. Zappia, A. Pistorio, C. Gambini, G. L. Mancardi, A. Uccelli, V. Pistoia.
2004
. Recapitulation of B cell differentiation in the central nervous system of patients with multiple sclerosis.
Proc. Natl. Acad. Sci. USA
101
:
11064
-11069.
21
Owens, G. P., H. Kraus, M. P. Burgoon, T. Smith-Jensen, M. E. Devlin, D. H. Gilden.
1998
. Restricted use of VH4 germline segments in an acute multiple sclerosis brain.
Ann. Neurol.
43
:
236
-243.
22
Baranzini, S. E., M. C. Jeong, C. Butunoi, R. S. Murray, C. C. Bernard, J. R. Oksenberg.
1999
. B cell repertoire diversity and clonal expansion in multiple sclerosis brain lesions.
J. Immunol.
163
:
5133
-5144.
23
Smith-Jensen, T., M. P. Burgoon, J. Anthony, H. Kraus, D. H. Gilden, G. P. Owens.
2000
. Comparison of immunoglobulin G heavy-chain sequences in MS and SSPE brains reveals an antigen-driven response.
Neurology
54
:
1227
-1232.
24
Williamson, R. A., M. P. Burgoon, G. P. Owens, O. Ghausi, E. Leclerc, L. Firme, S. Carlson, J. Corboy, P. W. Parren, P. P. Sanna, et al
2001
. Anti-DNA antibodies are a major component of the intrathecal B cell response in multiple sclerosis.
Proc. Natl. Acad. Sci. USA
98
:
1793
-1798.
25
Owens, G. P., A. M. Ritchie, M. P. Burgoon, R. A. Williamson, J. R. Corboy, D. H. Gilden.
2003
. Single-cell repertoire analysis demonstrates that clonal expansion is a prominent feature of the B cell response in multiple sclerosis cerebrospinal fluid.
J. Immunol.
171
:
2725
-2733.
26
Qin, Y., P. Duquette, Y. Zhang, P. Talbot, R. Poole, J. Antel.
1998
. Clonal expansion and somatic hypermutation of VH genes of B cells from cerebrospinal fluid in multiple sclerosis.
J. Clin. Invest.
102
:
1045
-1050.
27
Colombo, M., M. Dono, P. Gazzola, S. Roncella, A. Valetto, N. Chiorazzi, G. L. Mancardi, M. Ferrarini.
2000
. Accumulation of clonally related B lymphocytes in the cerebrospinal fluid of multiple sclerosis patients.
J. Immunol.
164
:
2782
-2789.
28
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.
29
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.
30
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.
31
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.
32
McDonald, W. I., A. Compston, G. Edan, D. Goodkin, H. P. Hartung, F. D. Lublin, H. F. McFarland, D. W. Paty, C. H. Polman, S. C. Reingold, et al
2001
. Recommended diagnostic criteria for multiple sclerosis: guidelines from the International Panel on the diagnosis of multiple sclerosis.
Ann. Neurol.
50
:
121
-127.
33
Burgoon, M. P., G. P. Owens, S. Carlson, A. L. Maybach, D. H. Gilden.
2001
. Antigen discovery in chronic human inflammatory central nervous system disease: panning phage-displayed antigen libraries identifies the targets of central nervous system-derived IgG in subacute sclerosing panencephalitis.
J. Immunol.
167
:
6009
-6014.
34
Owens, G. P., M. P. Burgoon, M. E. Devlin, D. H. Gilden.
1997
. Extraction and purification of active IgG from SSPE and MS brain.
J. Virol. Methods
68
:
119
-125.
35
Hilton, A. A., A. J. Slavin, D. J. Hilton, C. C. Bernard.
1995
. Characterization of cDNA and genomic clones encoding human myelin oligodendrocyte glycoprotein.
J. Neurochem.
65
:
309
-318.
36
Call, M. E., J. Pyrdol, M. Wiedmann, K. W. Wucherpfennig.
2002
. The organizing principle in the formation of the T cell receptor-CD3 complex.
Cell
111
:
967
-979.
37
Kozak, M..
1987
. At least six nucleotides preceding the AUG initiator codon enhance translation in mammalian cells.
J. Mol. Biol.
196
:
947
-950.
38
Pham-Dinh, D., M. G. Mattei, J. L. Nussbaum, G. Roussel, P. Pontarotti, N. Roeckel, I. H. Mather, K. Artzt, K. F. Lindahl, A. Dautigny.
1993
. Myelin/Oligodendrocyte glycoprotein is a member of a subset of the immunoglobulin superfamily encoded within the major histocompatibility complex.
Proc. Natl. Acad. Sci. USA
90
:
7990
-7994.
39
Skerra, A., A. Pluckthun.
1988
. Assembly of a functional immunoglobulin Fv fragment in Escherichia coli.
Science
240
:
1038
-1041.
40
Glynn, P., H. M. Gilbert, J. Newcombe, M. L. Cuzner.
1982
. Analysis of immunoglobulin G in multiple sclerosis brain: quantitative and isoelectric focusing studies.
Clin. Exp. Immunol.
48
:
102
-110.
41
Butcher, H., W. Kennette, O. Collins, J. Demoor, J. Koropatnick.
2003
. A sensitive time-resolved fluorescent immunoassay for metallothionein protein.
J. Immunol. Methods
272
:
247
-256.
42
Ogata, A., H. Tagoh, T. Lee, T. Kuritani, Y. Takahara, T. Shimamura, H. Ikegami, M. Kurimoto, K. Yoshizaki, T. Kishimoto.
1992
. A new highly sensitive immunoassay for cytokines by dissociation-enhanced lanthanide fluoroimmunoassay (DELFIA).
J. Immunol. Methods
148
:
15
-22.
43
Yang, J. T., C. S. Wu, H. M. Martinez.
1986
. Calculation of protein conformation from circular dichroism.
Methods Enzymol.
130
:
208
-269.
44
O’Connor, K. C., T. Chitnis, D. E. Griffin, S. Piyasirisilp, A. Bar-Or, S. Khoury, K. W. Wucherpfennig, D. A. Hafler.
2003
. Myelin basic protein-reactive autoantibodies in the serum and cerebrospinal fluid of multiple sclerosis patients are characterized by low-affinity interactions.
J. Neuroimmunol.
136
:
140
-148.
45
Kawasaki, E., G. S. Eisenbarth.
2000
. High-throughput radioassays for autoantibodies to recombinant autoantigens.
Front. Biosci.
5
:
E181
-E190.
46
Yu, L., G. Eisenbarth, E. Bonifacio, J. Thomas, M. Atkinson, C. Wasserfall.
2003
. The second murine autoantibody workshop: remarkable interlaboratory concordance for radiobinding assays to identify insulin autoantibodies in nonobese diabetic mice.
Ann. NY Acad. Sci.
1005
:
1
-12.
47
Sblattero, D., F. Florian, E. Azzoni, T. Zyla, M. Park, V. Baldas, T. Not, A. Ventura, A. Bradbury, R. Marzari.
2002
. The analysis of the fine specificity of celiac disease antibodies using tissue transglutaminase fragments.
Eur. J. Biochem.
269
:
5175
-5181.
48
Richter, W., Y. Shi, S. Baekkeskov.
1993
. Autoreactive epitopes defined by diabetes-associated human monoclonal antibodies are localized in the middle and C-terminal domains of the smaller form of glutamate decarboxylase.
Proc. Natl. Acad. Sci. USA
90
:
2832
-2836.
49
Psaridi-Linardaki, L., A. Mamalaki, S. J. Tzartos.
2003
. Future therapeutic strategies in autoimmune myasthenia gravis.
Ann. NY Acad. Sci.
998
:
539
-548.
50
Von Budingen, H. C., S. L. Hauser, J. C. Ouallet, N. Tanuma, T. Menge, C. P. Genain.
2004
. Epitope recognition in the myelin/oligodendrocyte glycoprotein differentially influences disease phenotype and antibody effector functions in autoimmune demyelination.
Eur. J. Immunol.
34
:
2072
-2083.
51
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.
52
Katz-Levy, Y., K. L. Neville, J. Padilla, S. Rahbe, W. S. Begolka, A. M. Girvin, J. K. Olson, C. L. Vanderlugt, S. D. Miller.
2000
. Temporal development of autoreactive Th1 responses and endogenous presentation of self myelin epitopes by central nervous system-resident APCs in Theiler’s virus-infected mice.
J. Immunol.
165
:
5304
-5314.
53
Robinson, W. H., P. Fontoura, B. J. Lee, H. E. de Vegvar, J. Tom, R. Pedotti, C. D. DiGennaro, D. J. Mitchell, D. Fong, P. P. Ho, et al
2003
. Protein microarrays guide tolerizing DNA vaccine treatment of autoimmune encephalomyelitis.
Nat. Biotechnol.
21
:
1033
-1039.
54
Olson, J. K., J. L. Croxford, M. A. Calenoff, M. C. Dal Canto, S. D. Miller.
2001
. A virus-induced molecular mimicry model of multiple sclerosis.
J. Clin. Invest.
108
:
311
-318.
55
Rosenberg, G. A..
1990
.
Brain Fluids and Metabolism
Oxford University Press, New York. .
56
Karni, A., R. Bakimer-Kleiner, O. Abramsky, A. Ben-Nun.
1999
. Elevated levels of antibody to myelin oligodendrocyte glycoprotein is not specific for patients with multiple sclerosis.
Arch. Neurol.
56
:
311
-315.
57
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.
58
Lindert, R. B., C. G. Haase, U. Brehm, C. Linington, H. Wekerle, R. Hohlfeld.
1999
. Multiple sclerosis: B- and T-cell responses to the extracellular domain of the myelin oligodendrocyte glycoprotein.
Brain
122
:
2089
-2100.
59
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.
60
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.
61
Lucchinetti, C., W. Bruck, J. Parisi, B. Scheithauer, M. Rodriguez, H. Lassmann.
2000
. Heterogeneity of multiple sclerosis lesions: implications for the pathogenesis of demyelination.
Ann. Neurol.
47
:
707
-717.