of June 4, 2018.This information is current as

SclerosisMyelin Surface Antigen in Pediatric Multiple Age-Dependent B Cell Autoimmunity to a

Kai W. WucherpfennigKhoury, Amit Bar-Or, David A. Hafler, Brenda Banwell and Tavakoli, Jerome de Seze, Zhannat Idrissova, Samia J.O'Connor, Vissia Viglietta, Susan J. Wong, Norma P.

C.Donald Gagne, Jayne M. Ness, Silvia Tenembaum, Kevin Kuhle, Ludwig Kappos, Kevin Rostasy, Daniela Pohl,Bettina Franz, Julia Kennedy, Shannon McArdel, Jens Katherine A. McLaughlin, Tanuja Chitnis, Jia Newcombe,

http://www.jimmunol.org/content/183/6/4067doi: 10.4049/jimmunol.0801888August 2009;

2009; 183:4067-4076; Prepublished online 17J Immunol 

MaterialSupplementary

8.DC1http://www.jimmunol.org/content/suppl/2009/08/17/jimmunol.080188

Referenceshttp://www.jimmunol.org/content/183/6/4067.full#ref-list-1

, 21 of which you can access for free at: cites 58 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists, Inc. All rights reserved.Copyright © 2009 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on June 4, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on June 4, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Age-Dependent B Cell Autoimmunity to a Myelin SurfaceAntigen in Pediatric Multiple Sclerosis1

Katherine A. McLaughlin,*† Tanuja Chitnis,‡§¶ Jia Newcombe,� Bettina Franz,* Julia Kennedy,#

Shannon McArdel,† Jens Kuhle,** Ludwig Kappos,** Kevin Rostasy,†† Daniela Pohl,‡‡

Donald Gagne,§§ Jayne M. Ness,¶¶ Silvia Tenembaum,�� Kevin C. O’Connor,‡§ Vissia Viglietta,‡§

Susan J. Wong,## Norma P. Tavakoli,## Jerome de Seze,*** Zhannat Idrissova,††† Samia J. Khoury,‡§

Amit Bar-Or,§§ David A. Hafler,‡§ Brenda Banwell,# and Kai W. Wucherpfennig2*†

Multiple sclerosis (MS) typically manifests in early to mid adulthood, but there is increasing recognition of pediatric-onset MS, aidedby improvements in imaging techniques. The immunological mechanisms of disease are largely unexplored in pediatric-onset MS, inpart because studies have historically focused on adult-onset disease. We investigated autoantibodies to myelin surface Ags in a largecohort of pediatric MS cases by flow cytometric labeling of transfectants that expressed different myelin proteins. Although Abs to nativemyelin oligodendrocyte glycoprotein (MOG) were uncommon among adult-onset patients, a subset of pediatric patients had serum Absthat brightly labeled the MOG transfectant. Abs to two other myelin surface Ags were largely absent. Affinity purification of MOG Absas well as competition of binding with soluble MOG documented their binding specificity. Such affinity purified Abs labeled myelin andglial cells in human CNS white matter as well as myelinated axons in gray matter. The prevalence of such autoantibodies was highestamong patients with a very early onset of MS: 38.7% of patients less than 10 years of age at disease onset had MOG Abs, comparedwith 14.7% of patients in the 10- to 18-year age group. B cell autoimmunity to this myelin surface Ag is therefore most common inpatients with a very early onset of MS. The Journal of Immunology, 2009, 183: 4067–4076.

M ultiple sclerosis (MS)3, a chronic inflammatory demy-elinating disease of the CNS, is the most commoncause of chronic neurological disability in young

adults (1). There has been increasing recognition of pediatric MSover the past two decades, in part due to the advent of magnetic

resonance imaging (MRI) techniques, which enable sensitive de-tection of CNS white matter abnormalities (2). It is now estimatedthat 2–5% of cases of MS begin before an age of 16 (3–5), but thetrue incidence of pediatric MS remains unknown. The average ageof clinical onset in most pediatric MS cohorts is between 8 and 14years, but the disease occurs even in children 2–5 years of age. Inthe majority of pediatric MS patients (�90%), the disease courseis relapsing-remitting. Pediatric disease onset is associated with ayounger age at disease progression milestones, such as the require-ment for mobility aids, than adult-onset MS (4–7). The rate ofdisease relapses is significantly higher in pediatric-onset MS thanadult-onset disease, even when the initiation of disease-modifyingtherapies following a clinically isolated event is taken into account(8). A short interval (less than 1 year) between the first two demyeli-nating episodes, incomplete recovery after the first attack as well as asecondary progressive disease course are unfavorable prognostic fac-tors (9, 10). Cognitive impairment is common in pediatric-onset MS,and can seriously affect academic performance (11–13).

Unlike adult-onset MS, which is more common in females andpopulations with European ancestry, pediatric MS occurs in manyethnic groups and does not have a strong gender bias before pu-berty (14, 15). An analysis of MS patients cared for in Torontoshowed that parents were of non-European descent for a substan-tially larger percentage of pediatric MS patients (41.9% of mothersand 48.8% of fathers) than for adult-onset MS patients (8.1% ofmothers and 7.1% of fathers), and represented ethnic groups fromAsia, the Caribbean, and the Middle East (15). Similar to thesefindings, a recent study in the northeastern United States revealedthat only 86.2% of patients with pediatric-onset MS identifiedthemselves as Caucasian, compared with 93.1% of patients withadult-onset MS ( p � 0.014) (3). Specifically, the proportions ofAfrican-American and Hispanic- or Latin-American patients werehigher in the pediatric-onset group.

*Department of Cancer Immunology and AIDS, Dana-Farber Cancer Institute, †Pro-gram in Immunology and ‡Department of Neurology, Harvard Medical School, §Cen-ter for Neurologic Diseases, Brigham and Women’s Hospital, Boston, MA 02115,¶Partners Pediatric Multiple Sclerosis Center, Massachusetts General Hospital forChildren, Boston, MA 02114; �NeuroResource, University College London Instituteof Neurology, London, United Kingdom; #Institute of Neurology, Department of Pae-diatrics, Division of Neurology, The Hospital for Sick Children, Toronto, Canada;**Department of Research and Department of Neurology, University Hospital Basel,Basel, Switzerland; ††Department of Pediatrics and Pediatric Neurology, Georg-August-University, Goettingen, Germany; ‡‡Children’s Hospital of Eastern Ontario,Ottawa, Ontario, Canada; §§Department of Neurology and Neurosurgery, MontrealNeurological Institute, Montreal, Quebec, Canada; ¶¶Children’s Hospital of Alabama,Birmingham, AL 35233; ��Department of Pediatric Neurology, National PediatricHospital “Dr. J.P. Garrahan”, Buenos Aires, Argentina; ##Diagnostic ImmunologyLaboratory, Wadsworth Center, New York State Department of Health, Albany, NY12207; ***Department of Neurology, Strasbourg University, Alsace, France; and†††Department of Neurology, Kazak State Medical University, Almaty, Kazakhstan

Received for publication June 11, 2008. Accepted for publication July 10, 2009.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by Grant PO1 AI045757 from the National Institutes ofHealth (to K.W.W.), Grant RG 4122-A-6 from the National Multiple Sclerosis So-ciety (to K.W.W.), and Grant A B-O from the Wadsworth Foundation (to B.B.).2 Address correspondence and reprint requests to Dr. Kai W. Wucherpfennig, Dana-Farber Cancer Institute, 44 Binney Street, Room D1410, Boston, MA 02115. E-mailaddress: Kai_Wucherpfennig@dfci.harvard.edu3 Abbreviations used in this paper: MS, multiple sclerosis; ADEM, acute dissemi-nated encephalomyelitis; CIS, clinically isolated syndrome; CSF, cerebrospinal fluid;MAG, myelin-associated glycoprotein; MFI, mean fluorescence intensity; MOG, my-elin oligodendrocyte glycoprotein; NMO, neuromyelitis optica; OMG, oligodendro-cyte-myelin glycoprotein; MRI, magnetic resonance imaging; HA, hemagglutinin.

Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00

The Journal of Immunology

www.jimmunol.org/cgi/doi/10.4049/jimmunol.0801888

by guest on June 4, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

Early during the disease course, there can be substantial dif-ficulties distinguishing between pediatric MS and acute dissem-inated encephalomyelitis (ADEM) (16). ADEM is characterizedby an acute onset of widespread CNS inflammation and demy-elination and is more common in children than adults (16 –18).For a diagnosis of ADEM, new criteria established by the In-ternational Pediatric MS Study Group require a polysymptom-atic clinical presentation that includes encephalopathy, which isdefined as either a behavioral change (such as confusion, ex-cessive irritability) or alteration in consciousness (lethargy,coma) (19). ADEM typically follows a monophasic course withpartial or complete recovery. However, recurrent ADEM (de-fined as a new event of ADEM with a recurrence of the initialsymptoms 3 mo or more after the initial event) and multiphasicADEM (two or more distinct ADEM episodes, each meetingclinical criteria for ADEM but involving new areas of the CNS,separated by at least 3 mo) occur rarely (20). Because the prog-nosis and treatment regimens for MS and ADEM differ sub-stantially, it is essential to understand the biological similaritiesand differences between these conditions.

The largest reported study of pediatric demyelinating diseasesillustrates this issue (21). In this study, 296 patients were followedafter an initial demyelinating episode; 168 patients (57%) experi-enced two or more episodes of demyelination and were given adiagnosis of MS. Although MS was more commonly preceded bya clinically isolated syndrome (CIS), 20% of the patients with afinal diagnosis of MS were considered to have ADEM at clinicalonset. Given these diagnostic difficulties, biomarkers that reflectdifferences in disease pathogenesis would be valuable. For exam-ple, Abs to aquaporin-4, a water channel enriched on astrocyticfoot processes at the blood-brain barrier, are commonly found inneuromyelitis optica (NMO), an inflammatory demyelinating dis-ease that affects the optic nerves and the spinal cord (22). High-titer Abs are associated with increased disease activity (23), andthe presence of aquaporin Abs has recently been incorporated intothe diagnostic criteria for NMO (24).

Little is currently known about disease mechanisms in pediatricMS, in particular with respect to similarities and differences toadult-onset MS. Recent clinical trials with rituximab in adult MSpatients have shown that B cell depletion results in a pronouncedreduction in new demyelinating lesions (25), proving that B cellsplay a central role in the disease. Oligoclonal IgG Abs are presentin the cerebrospinal fluid (CSF) of the majority of adult-onset MSpatients, and CSF analysis of 136 patients with a disease onsetbefore age 16 identified oligoclonal IgG in 92% of cases (26), butthe specificity of these Abs remains unknown.

Pathogenic autoantibodies need to be able to bind to struc-tures on the surface of the myelin or oligodendrocytes, but theabundant myelin Ags myelin basic protein and proteolipid pro-tein are inaccessible to Abs in the intact myelin sheath. Myelinoligodendrocyte glycoprotein (MOG), a CNS-specific myelinand oligodendrocyte surface protein, is one of the relevant can-didate Ags, and a mAb (8-18C5) against MOG induces severedemyelination in mice or rats with mild experimental autoim-mune encephalomyelitis (27, 28). Although Abs able to bind tothe native conformation of the protein cause demyelination,Abs to MOG peptides that do not bind the folded protein are notpathogenic (29, 30). It is therefore critical to differentiate be-tween Abs to folded MOG and denatured or linear epitopes, atask for which traditional assays such as Western blot andELISA analyses are poorly suited.

We recently reported a radioimmunoassay with a tetramericform of MOG for the detection of MOG autoantibodies in patientswith inflammatory demyelinating CNS diseases and found Abs to

MOG in a subset of patients with ADEM (13/69 patients, 18.8%)(31). Interestingly, one of 19 pediatric MS sera gave a strong sig-nal in this radioimmunoassay, whereas only one of 109 adult-onsetMS sera was positive. This pediatric MS serum sample also la-beled a MOG-GFP transfectant, indicating that autoantibodies inthis patient indeed bound native MOG. We therefore decided toexamine a larger cohort of pediatric and adult-onset MS patients todefine the relationship between the presence of such autoantibod-ies and age at disease onset.

Materials and MethodsClinical samples

Pediatric and adult samples were collected by international neurologicalcenters that care for adults or children with MS and other demyelinatingdiseases. Pediatric MS was defined using the McDonald Criteria (32, 33).None of the cases defined as pediatric MS met the criteria for multiphasicor recurrent ADEM (19). The majority of pediatric-onset MS and pediatriccontrol serum samples were obtained from centers participating in theWadsworth International Pediatric MS Consortium, which did not makeMRI scans available for centralized review. Adult-onset MS was definedusing the Poser or McDonald criteria (32–34). Most adult-onset MS (in-cluding relapsing-remitting, secondary progressive, and primary progres-sive MS), CIS, and control serum and CSF samples were collected at thePartners Multiple Sclerosis Center at Brigham and Women’s Hospital(Boston, MA). Serum samples from patients with NMO were from theHopital Civil (Strasbourg, France), and viral encephalitis samples wereprovided by the New York State Department of Health. Each site collectedsamples using a protocol approved by their Institutional Review Board, andinformed consent was obtained from all subjects. Samples were stored at�80°C in aliquots.

Selection of MOG-GFP clones

The cloning of human MOG into the pEGFP-N1 vector (Clontech) andtransfection of Jurkat cells was previously described (31). Transfected cellswere grown in DMEM (Invitrogen) supplemented with 10% FBS (AtlantaBiologicals), 100 IU/ml penicillin, 100 �g/ml streptomycin, 35 mMHEPES, and 2 mM L-glutamine (Mediatech) with 1 mg/ml G418 (Invitro-gen) to maintain selection. MOG-GFP and GFP clones were generated bysorting single GFPhigh cells on a FACSAria (BD Biosciences). After ex-pansion, MOG clones were stained with an Ab to MOG (clone 8-18C5),and MOG-GFP and GFP clones with a matching level of GFP brightnesswere selected.

Flow cytometry

Cells were harvested, washed in PBS containing 1% BSA (Fisher) (PBS/BSA), and resuspended at a density of 1 � 106/ml. A total of 50,000 cells(50 �l) were incubated with serum at a 1/50 dilution (or other, as indicated)in V-bottom plates (Corning) for 1 h at 4°C. For CSF stains, 25,000 cellsat a density of 2 � 106/ml were incubated with 50 �l of CSF. Cells werewashed twice with 200 �l of PBS/BSA, and incubated with biotinylatedsecondary Abs to human IgG (pan-IgG clone HP-6017 (Sigma-Aldrich), orsubclass-specific IgG1 clone HP6069, IgG2 clone HP6002, IgG3 cloneHP6047, and IgG4 clone HP6023 (Calbiochem)) at a 1/1000 dilution for 30min at 4°C. Cells were washed twice and incubated with streptavidin-PE(Invitrogen) at a 1/1000 dilution for 20 min, washed once with PBS aloneand fixed with 1% formaldehyde in PBS at 4°C before analysis. A total of10,000 events per well were recorded on an LSRII instrument equippedwith a high-throughput sampler (BD Biosciences). Data analysis was per-formed in FlowJo (Tree Star) and Excel (Microsoft). Nonparametric sta-tistical tests were performed in SISA Tables (Quantitative Skills).

Affinity purification of Abs to MOG for FACS analysis

The Ig domain of MOG and the membrane proximal Ig domain of CD80were linked to a BirA biotinylation site and cloned into the pET22b vector(Promega). Proteins were expressed in Escherichia coli, refolded from in-clusion bodies, purified by HPLC, and biotinylated using the BirA enzymeusing published protocols (35, 36). A total of 100 �g of biotinylated MOGor CD80 was incubated with 50 �l of settled streptavidin-agarose beads(Sigma-Aldrich) in 200 �l of PBS/BSA overnight at 4°C on an orbitalrotator, and washed to remove unbound Ag. The 10 �l of settled beadswere incubated with 5 �l of serum and 200 �l of PBS/BSA for 5 h at 4°C.The supernatant was saved, and the beads were washed four times with 1ml of PBS/BSA, followed by elution in 200 �l of 0.1 M glycine (pH 2.5).Eluted proteins were separated from the beads by spin filtration (Spin-X 0.2

4068 MOG AUTOANTIBODIES IN PEDIATRIC MS

by guest on June 4, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

micron; Corning) and the pH was neutralized by addition of 2 M Tris base.The 60 �l of unbound or eluted proteins were incubated with 50,000 cellsin a volume of 125 �l with or without 25 �g of nonbiotinylated recom-binant MOG as a competitor, and bound Abs were detected by FACS asdescribed above.

Immunocytochemistry on human CNS tissue sections

Abs used for immunocytochemistry were purified from sera by protein Gimmunoprecipitation, eluted with 0.1 M glycine (pH 2.5), dialyzed intoPBS, and biotinylated using sulfo-NHS-biotin (Pierce) at a 50-fold molarratio of biotin to IgG. MOG and BSA were conjugated to cyanogen bro-mide-activated Sepharose beads (Amersham) per the manufacturer’s in-structions and blocked in PBS plus 5% BSA overnight. The 200 �g of totalbiotinylated IgG was affinity purified on MOG and BSA beads in PBS plus2% BSA overnight, and unbound Abs were retained before the first wash.Bound Abs were eluted using 0.1 M glycine (pH 2.5) plus 1% BSA as acarrier protein. Bound and unbound Abs were concentrated to �200 �lusing Microcon concentrators (Millipore).

Tissue sections from the CNS of six adult tissue donors without CNSdisease were from the NeuroResource tissue bank, UCL Institute ofNeurology, London, U.K. Serial cryostat sections (10 �m) were cap-tured onto polylysine-coated slides, acetone-fixed and incubated withblocking sera. Sections were incubated for 2 h with 20 �g of totalbiotinylated IgG or Abs corresponding to 100 �g total IgG furtherpurified on MOG or BSA beads. Immunoperoxidase staining was per-formed with a Vectastain avidin-biotin peroxidase kit (Vector Labora-tories) using diaminobenzidine and nickel (II) chloride to give blackstaining (37). Staining patterns observed with patient IgG were com-pared with adjacent sections labeled with the myelin-specific anti-MOGmAb 8-18C5 and the oligodendrocyte-specific mAb 14E (38). PrimaryAbs were omitted for immunocytochemistry controls.

Cloning and transfection of hemagglutinin (HA)-tagged Ags

The pCI-neo vector (Promega) was modified to create a vector suitable forexpression of N-terminal tagged type I transmembrane proteins with afluorescent reporter under control of an internal ribosome entry site (IRES).Oligonucleotides were used to link the H2-Kb signal peptide and HAepitope tag (YPYDVPDYASL) to a unique multiple cloning site contain-ing XcmI, Bsu36I, EspI, and XbaI recognition sequences, which was in-serted at the NheI-XbaI restriction sites of pCI-neo. IRES and ZsGreensequences were amplified from the pHAGE-CMV-fullEF1�-IRES-Zs-Green vector, as created by J.-S. Lee provided by the Dana-Farber/HarvardCancer Center DNA Resource Core (Boston, MA) and inserted at a NotIsite downstream of the pCI-neo multiple cloning site.

Human myelin-associated glycoprotein (MAG) and oligodendrocyte-myelin glycoprotein (OMG) cDNA were obtained from GeneService (www.geneservice.co.uk) and used to clone the extracellular, transmembrane, andcytoplasmic domains of each Ag by PCR using Phusion polymerase (NewEngland Biolabs). These sequences, minus the endogenous signal peptide,were cloned into the modified vector with XcmI and XbaI enzymes (NewEngland Biolabs). Jurkat cells were transfected by electroporation, andstable transfectants selected with 2 mg/ml G418. Cells expressing ZsGreenwere sorted and stained with a biotinylated Ab to the HA tag (clone 3F10;Roche) and streptavidin-PE (Invitrogen). Single PEhigh, ZsGreenhigh cellswere sorted as described, and the clones with highest HA expression wereused for autoantibody detection.

ResultsDetection of high-titer autoantibodies to native MOG inpediatric MS patients

Our previous FACS experiments used a MOG-GFP construct inwhich GFP was fused to the cytoplasmic domain of full lengthhuman MOG so that the brightness of GFP fluorescence reportedon the level of MOG expression. A stably transfected Jurkat cellline had been generated, containing cells with different levels ofMOG-GFP expression. FACS analysis of serum samples fromADEM patients showed a diagonal staining pattern for IgG bind-ing vs GFP fluorescence (31), indicating that the labeling intensityclosely correlated with the density of the Ag at the cell surface.This finding suggested that the sensitivity of the assay could beimproved by single cell cloning of the cells that expressed thehighest level of MOG-GFP. We therefore sorted single cellsbased on GFP fluorescence from MOG-GFP or control GFP

transfectants and selected MOG-GFP (used as MOG throughoutstudy) and GFP control clones with similar GFP expressionlevels for our experiments.

To enable sensitive detection of autoantibodies, a three-stepstaining procedure was used in which 50,000 MOG-GFP or controlGFP cells were incubated with serum at a 1/50 dilution and boundAbs were detected using biotinylated anti-human IgG and strepta-vidin-PE. Specific autoantibody binding was expressed as the ratioof the PE mean fluorescence intensity (MFI) of the MOG vs theGFP clone (binding ratio), and ratios greater than 5 were consid-ered to be positive. This threshold is three SDs above the meanbinding ratio of all nonautoimmune and non-neurological controls(rounded off from the actual value of 5.03). Samples with bindingratios greater than 3 were analyzed for a total of two to five mea-surements per sample, depending on the volume of serum avail-able. Sera with average binding ratios greater than 5 were repro-ducibly positive (see supplemental Fig. 1).4

Fig. 1A shows that labeling of the MOG clone by a number ofpediatric MS sera was very bright (binding ratios greater than 50)and illustrates the range of staining intensities for positive pediatric

4 The online version of this article contains supplemental material.

FIGURE 1. Identification of autoantibodies to native MOG in pediatricMS patients. A, Brightness of fluorescent labeling with different positivepediatric MS sera. FACS data representing Ab binding to the MOG-GFPtransfectant (open histograms) and the control GFP transfectant (shadedhistograms) are shown for sera from six pediatric MS patients. Sera wereincubated at a dilution of 1/50 with cells, and bound Abs were detectedusing biotinylated anti-human IgG and streptavidin-PE. For each sample,the ratio of the MFI for the MOG-GFP and the control transfectant isshown (top left corner). The six depicted examples represent the entirerange of fluorescent intensities for positive sera, with binding ratios of�5–200 for the MOG vs the control transfectant. B, Ab titers measured byFACS analysis. Sera were incubated with MOG-GFP cells at the indicateddilutions. Abs to MOG were detectable in all positive sera at a dilution of1/400 and many at a dilution of 1/800.

4069The Journal of Immunology

by guest on June 4, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

MS sera. All positive sera labeled the MOG transfectant at a 1/400dilution and several sera even at a 1/800 dilution (Fig. 1B). Com-parison of the MOG-GFP and control GFP clones for each samplewas important because the sera differed in the level of nonspecificbinding to the Jurkat cells (Fig. 1A). The MOG clone gave a higherfluorescence signal than the MOG line and binding ratios that weretypically 5- to 10-fold higher than with the lines (see supplementalFig. 2).4 Because of their superior properties, these GFP and MOGclones were used in all subsequent analyses. Compared with thetetramer radioimmunoassay, the major advantages of this flow cy-

tometric assay are that it enables higher throughput and avoids theuse of radioactivity.

MOG Abs define a subset of pediatric MS patients

Sera from patients with pediatric- or adult-onset MS as well ascontrol donors from an international group of clinics (Table I)were tested for Abs to MOG at a 1/50 dilution. Abs to MOG weredetected in 28 of 131 pediatric-onset MS samples (21.3%) butrarely found in controls (Fig. 2A). In particular, all samples frompatients with juvenile diabetes (n � 28) as well as patients with

FIGURE 2. Autoantibodies to native MOG are more prevalent in pediatric-onset than in adult-onset MS patients, and infrequent in other inflammatorydemyelinating CNS diseases. A, Sera from pediatric (left) or adult individuals (right) were assayed for Abs to MOG at a dilution of 1/50. The MOG tocontrol binding ratio for each sample was calculated as described, and ratios greater than 5 were considered positive. Abs to MOG were significantly morecommon in pediatric- than adult-onset MS patients (p � 4.48 � 10�7) or control subjects and labeled the transfectant more brightly. B, Analysis of MOGAbs in patient populations from MS centers in different countries. Although anti-MOG was detectable in pediatric MS samples from all countries studied,the frequency of MOG Abs in adults varied considerably between centers.

Table I. Demographics of all pediatric- or adult-onset MS patient groupsa

N

PercentageFemale

(%)

Age at Sample (y) MS Subtype

Median Range RRMS SPMS PPMS Marburg

Pediatric MS 131 63.3 14.6 5.0–19.4 93 3 0 1Pediatric neurological control 34 55.9 13.7 2.1–18.5Pediatric non-neurological control 37 67.6 12.3 2.0–16.0Juvenile diabetes 28 71.4 10.7 5.7–17.6Adult-onset MS 254 67.9 42 18–76.9 171 49 34 0Adult control 86 56.5 40 22–64Clinically isolated syndrome 30 80.0 37.9 22.9–54.3Neuromyelitis optica 13 84.6 41 24–55Viral encephalitis 29 n/a n/a n/a

a Sera from pediatric- or adult-onset MS donors as well as control donors from an international group of clinics were tested for Abs to MOG. Typesof MS include relapsing-remitting (RRMS), secondary progressive (SPMS), and primary progressive (PPMS).

Data not available (n/a).

4070 MOG AUTOANTIBODIES IN PEDIATRIC MS

by guest on June 4, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

non-neurological diseases (n � 37) were negative. Among patientswith other neurological diseases (n � 34), one (with a develop-mental delay) was clearly positive, whereas the other was margin-ally positive.

By comparison, MOG Abs were only detected in a small subsetof adult-onset MS patients (11/254 sera, 4.3%) and the brightnessof labeling with these serum samples was substantially lower thanwith the positive pediatric MS samples. The difference in anti-MOG frequency between all adult- and pediatric-onset MS caseswas statistically significant ( p � 4.48 � 10�7). All anti-MOG-positive pediatric MS sera were from individuals with relapsing-remitting MS, the most common MS subtype in our pediatricgroup (Table I). Abs to MOG were found in all major clinicalsubtypes of adult-onset MS, including 6/171 relapsing-remitting,3/49 primary progressive, and 2/34 secondary progressive patients.Abs to MOG were not detected in adults with NMO or viral en-cephalitis, and were rarely found at low levels in adults with a CIS(Fig. 2A). We have also initiated a prospective analysis of MOGAbs in pediatric demyelinating diseases to examine persistence ofsuch Abs. As part of this ongoing study, we obtained serial serumsamples from 39 children with demyelinating events, 5 of whomwere positive for Abs to MOG. Serial samples were available infour of the five cases, and serum anti-MOG was detectable at eachfollow-up interval. Three patients had two samples each, taken atthe first clinic visit and 4, 6, or 13 mo later. One patient had Absto MOG in each of n � 3 serial samples, the last taken 33 mo afterthe initial visit and more than 4 years after the first clinical eventof ADEM.

Matched serum and CSF samples were available from 13 pedi-atric patients; difficulties in obtaining CSF prevented analysis ofmore pediatric MS samples. Abs to MOG were detectable in theCSF of one pediatric-onset MS patient (binding ratio of 8.63 forCSF, ratio of 23.7 for paired serum) who had a first clinical eventat 13.3 years of age. In two other pediatric MS cases with pairedsamples, MOG-specific Abs were detectable in the serum (bind-ing ratios of 20.4 for CSF and 5.8 for paired serum) but not inpaired CSF. It is important to consider that the IgG concentra-tion is �1000-fold higher in the serum compared with the CSFand that detection in the serum may therefore be more sensitive.Also, it is possible that Abs in CSF are present as immunecomplexes and therefore not able to bind MOG in our assay orthat a fraction of MOG Abs are absorbed by their target Ag onoligodendrocytes/myelin.

The pediatric MS samples came from medical centers for MS insix different countries, and MOG-positive cases were identified foreach center (Fig. 2B), with some differences in the frequency ofpositive cases. Among adult-onset MS patients, the frequency ofAbs to MOG differed between centers (0–8.3% of positive sera).Although 12 of 43 pediatric MS samples from Canada containedanti-MOG, none of the 53 Canadian adult MS samples were MOG-positive (Fig. 2B). Because all these samples were analyzed by oneindividual using the same assay, differences in the frequency ofMOG-positive adult-onset MS cases reported by different groupsmay reflect differences in the studied patient populations.

MOG Abs are most prevalent in patients with a very earlydisease onset

A number of possible factors may be responsible for the presenceof MOG Abs in a subset of pediatric MS patients. No significantassociation between anti-MOG and gender or ethnicity was found(see supplemental Table I).4 However, the presence of Abs signif-icantly correlated with a younger age at disease onset ( p � 0.009):38.7% of individuals with a first clinical event before age 10 hadMOG Abs, compared with 14.7% of patients with a disease onset

between 10 to 18 years (Fig. 3A). As stated above, the frequencyof MS patients with MOG Abs was even lower in the adult-onsetpopulation (4.3%). In adult-onset MS, 3 of the 6 MOG-positivepatients for whom data were available had their first clinical eventat an age of over 40 years (Fig. 3B), and there was no correlationbetween the presence of these Abs and age at first clinical event( p � 0.1835). Abs to MOG were found in pediatric MS patientswith recent disease onset (�1 year) and also those with a longerdisease course (�5 years) (Fig. 3C), indicating that MOG canbe a target in early stages of MS, and that the Abs may persist,at least for some time. Most patients were either untreated or

FIGURE 3. MOG Abs are most prevalent in MS patients with a veryearly disease onset. A, Abs to MOG are most common in MS patients witha disease onset before 10 years of age. Pediatric-onset MS patients wereranked by age at the first clinical event. Although the majority of sampleswere from older patients (�10 years), the frequency of Abs to MOG wassignificantly higher (p � 0.009) in individuals under 10 years of age at thefirst clinical event. B, Analysis of MOG Abs in pediatric- and adult-onsetMS patients based on age at disease onset. Abs to MOG were most com-mon in the youngest pediatric-onset MS cases, whereas a correlation withage at diagnosis was not apparent among the infrequent adult-onset caseswith MOG Abs. C, Relationship between the presence of MOG Abs andthe time interval since diagnosis of MS. Abs were found in pediatric-onsetMS patients with recent disease onset and established disease.

4071The Journal of Immunology

by guest on June 4, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

had received IFNs at the time of sample collection (see supple-mental Table II),4 and the usage of specific immunomodulatorydrugs was not significantly different in the MOG-positive and-negative groups ( p � 0.636904).

Based on our previous findings of MOG Abs in ADEM and thefact that this disease is most common in children, we investigatedthe relationship between Abs to MOG and clinical presentation atthe first demyelinating event. A significant association was foundbetween anti-MOG and a first clinical event diagnosed as ADEM( p � 0.0049): 30.8% of MOG-positive MS patients had an initialdiagnosis of ADEM, compared with 8% of MOG-negative cases.Also, 50% of pediatric MS cases with an initial ADEM-like firstevent had MOG Abs (see supplemental Table I).4

Functional properties of MOG Abs from pediatric MS patients

Abs to MOG may cause demyelination and oligodendrocyte deathby either complement activation or Fc receptor-dependent mech-anisms. IgG subtypes differ in the ability to fix complement andinduce Ab-dependent cellular cytotoxicity, and we therefore sub-classified MOG-specific IgG Abs by FACS using a panel of sec-ondary Abs specific for defined IgG subtypes (IgG1-IgG4). Themajority of sera contained only MOG-specific IgG1 (7 of 12 sera),one sample had both MOG-specific IgG1 and IgG3 and anotherwas dominated by IgG2 (Fig. 4). The lower sensitivity of the sub-class-specific secondary Abs compared with the pan-IgG Ab pre-vented identification of the dominant IgG subtype in three samples.Human IgG1 is efficient in fixing complement and inducing Ab-dependent cell-mediated cytotoxicity.

The murine 8-18C5 mAb to MOG is well known to bind oli-godendrocytes in culture and induce demyelination in vivo. Wetested whether Abs from pediatric MS patients could compete with8-18C5 for binding to MOG-GFP cells. Because the secondary Abused to detect bound IgG recognizes human but not mouse pro-teins, this assay was not confounded by detection of 8-18C5 on thecells. Addition of 8-18C5 decreased the binding of patient serumAbs to MOG-GFP cells in a dose-dependent manner for all sam-ples tested (Fig. 5A). This result shows that the Abs from pediatricMS patients bind to surfaces on MOG that significantly overlapwith the 8-18C5 epitope.

To further verify the specificity of these Abs, we affinity purifiedpolyclonal Abs from sera using a MOG protein preparation rep-resenting the extracellular domain that had been refolded from E.coli inclusion bodies. The protein carried a C-terminal BirA tag forsite-specific biotinylation, enabling capture onto streptavidinbeads. The MOG extracellular domain contains a single Ig domainand we used a protein of similar size (membrane proximal Ig do-main of CD80) as a control. Biotinylated MOG or CD80 werebound to streptavidin-coated agarose beads and incubated with pe-diatric MS sera. Both bound and unbound Abs were used to stainMOG and GFP clones. Abs capable of staining the MOG clone didnot bind to control beads, but were isolated from all four sera usingMOG beads (Fig. 5B). Addition of soluble recombinant MOG pro-tein as a competitor to the staining reaction substantially reducedthe level of specific binding to the MOG clone. The recombinantMOG protein used for affinity purification and competition of Ab

FIGURE 4. The majority of pediatric MS patients have MOG IgG1Abs, an IgG subtype that can fix complement and bind to Fc receptors. IgGsubclass-specific secondary Abs were used to detect the binding of serumAbs to the MOG transfectant. In the example shown, only MOG-specificIgG1 Abs were detected. IgG1 was the most common subtype among thesepediatric MS patients, found in 8 of 12 tested sera. IgG2 and IgG3 wererarely found, and IgG4 was not detected in any sample.

FIGURE 5. Abs in pediatric MS sera are specific for MOG. A, A mAbto MOG competes with serum Abs for Ag binding. Sera and indicatedamounts of the MOG-specific Ab 8-18C5 were incubated together withMOG-GFP cells for 1 h. A human-specific secondary Ab was used todetect bound IgG. Addition of increasing amounts of 8-18C5 decreased theamount of serum Ab bound to cells. B, Affinity-purified Abs specificallybind to MOG transfectants. Biotinylated recombinant MOG (rMOG, ex-tracellular domain) or an Ig superfamily control protein (membrane-prox-imal Ig domain of CD80) were captured onto streptavidin beads for affinityisolation of MOG-specific Abs from 5 �l of serum. Unbound and elutedAbs were used to stain MOG and control transfectants. Binding ratiosobtained with purified Abs from four sera are summarized in the table. Abspurified on MOG beads bound MOG on the cell surface, and addition of 25�g of soluble recombinant MOG inhibited binding of eluted MOG-specificAbs for all four sera tested.

4072 MOG AUTOANTIBODIES IN PEDIATRIC MS

by guest on June 4, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

binding was not glycosylated, indicating that at least a subset ofMOG-specific Abs in pediatric MS sera do not require the N-linked glycan for binding. These experiments firmly establish thatthe Abs detected with our flow cytometric assay are indeed specificfor MOG.

MOG-specific Abs label human white matter and myelinatedaxons

We next sought to determine whether MOG-specific Abs frompediatric MS patients can bind MOG in the CNS and thereforeperformed immunocytochemistry on acetone-fixed human brainsections containing both white and gray matter using biotinylatedAbs detected with avidin-peroxidase. As a positive control, myelinwas stained with the anti-MOG mAb 8-18C5 (Fig. 6A) (and seesupplemental Fig. 3A)4 and Abs directed against myelin basic pro-tein and Wolfgram protein (data not shown). Glial cells with the

morphology of oligodendrocytes were immunostained on adjacentsections with the oligodendrocyte-specific Abs 14E (Fig. 6A) (andsee supplemental Fig. 3A),4 CNPase, and carbonic anhydrase (datanot shown).

Total IgG was isolated from anti-MOG positive and negativesera, biotinylated and then further purified using beads withimmobilized MOG or BSA as a control protein. In samples fromtwo MOG-positive pediatric MS cases (W52 and W24), Absaffinity purified on MOG beads but not control BSA beads la-beled CNS white matter, whereas no staining was observed withaffinity purified Abs from control donor D14 (Fig. 6B). Absfrom patient W52, and more weakly with W24, also labeled cellbodies in the white matter similar in morphology to cells iden-tified with the oligodendrocyte-specific Ab 14E and otheroligodendrocyte-specific Abs.

Myelinated axon bundles in the subpial cortical gray matterwere stained with the anti-MOG mAb 8-18C5 and serum IgG froma previously studied ADEM patient (R4) with high-titer MOG Abs(31) (Fig. 6C). Similar small bundles of myelinated axons and glialsatellite cells in the gray matter were stained with Abs affinitypurified on MOG beads from pediatric MS patient W52, but nostaining was seen with Abs eluted from BSA beads (Fig. 6C).These glial cells were also 14E-immunopositive.

We also performed staining with total biotinylated serum IgGnot purified on MOG or BSA beads (see supplemental Fig. 3).4

IgG from ADEM patient R4 stained white matter in a myelin-likepattern (see supplemental Fig. 3B).4 IgG from pediatric MS patientW52 labeled myelin and glial cells, whereas IgG from pediatric

FIGURE 6. Abs to MOG stain myelin in human brain. A, MOG proteinand oligodendrocytes in the white matter were detected with mAbs 8-18C5and 14E, respectively. Scale bar represents 25 microns. B, Immunocyto-chemistry with affinity-purified Abs from pediatric MS patients. Biotinyl-ated total serum IgG was purified on MOG or BSA-coated beads, andaffinity purified Ab corresponding to 100 �g of total IgG was used perstain. MOG-specific Abs from pediatric MS patients W52 and W24 labeledwhite matter in a myelin and oligodendrocyte-like pattern. No staining wasseen with Abs isolated on BSA control beads or Abs from a control donor(D14). Scale bar represents 25 microns. C, MOG-specific Abs label my-elinated axons and satellite glia in the subpial gray matter. Regions of graymatter in the same tissue sections as shown above were assessed for Abbinding. Small bundles of myelinated axons were labeled by the MOGmAb 8-18C5 and total IgG from a MOG-positive ADEM patient (R4), andsatellite glia cell bodies were identified with Ab 14E (data not shown).Pediatric MS serum Abs (patient W52) purified on MOG but not BSAbeads also bound myelinated axons. Scale bar represents 50 microns.

FIGURE 7. Abs to other myelin surface proteins in pediatric MS sera.A, Generation of transfectants that express other myelin proteins on the cellsurface. Jurkat cells were transfected with vectors that drove expression ofZsGreen as well as MOG, MAG, or OMG. Transfectants were cloned bysingle cell sorting and surface expression was verified using an Ab to theHA tag attached to the N terminus of each protein. B, Examples of labelingof these transfectants with MOG-GFP-positive pediatric MS sera. Sampleswere incubated at a 1/50 dilution with Jurkat cells transfected with HA-tagged Ags, and bound IgG was detected with biotinylated anti-human IgGand streptavidin-PE. C, Pediatric MS sera with MOG Abs do not showbroad anti-myelin reactivity. Ab binding for 13 serum samples is shown asthe MFI of PE. Although Abs to MAG and OMG were detectable at lowlevels in a few sera, the labeling of MOG transfectants was considerablybrighter in all cases. Abs to OMG and MAG were not detected in MOG-negative pediatric MS sera or pediatric controls (data not shown).

4073The Journal of Immunology

by guest on June 4, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

MS patient W24 primarily labeled cells with a glial morphology(see supplemental Fig. 3B).4 Only weak cell body staining wasobserved with IgG from a control donor (D14) and MOG-negativepediatric MS and juvenile diabetes samples (data not shown).Overall, labeling with MOG affinity purified Abs was stronger be-cause the specific Abs were more concentrated. Also, affinity pu-rification reduced nonspecific staining. The finding that affinitypurified MOG-specific Abs label CNS myelin, myelinated axons,and glial cells with oligodendrocyte morphology demonstrates thatthese serum Abs recognize their CNS target Ag.

Evaluation of other myelin surface Ags

To determine whether patients with Abs to MOG also have circu-lating Abs to other myelin surface proteins, we created a set ofJurkat transfectants expressing MAG, OMG, or MOG. The pro-teins carried an N-terminal HA epitope tag to verify expression onthe cell surface and clones expressing high levels of the HAepitope tag on the surface and a fluorescent reporter (ZsGreen) inthe cytosol (Fig. 7A) were sorted and used to detect specific Abs inpediatric MS sera by FACS. Examples of five anti-MOG-positivesera are shown in Fig. 7B. Abs to MAG and OMG were not foundin most sera, but when detected only weakly stained the respectivetransfectant (Fig. 7C). Sera from 25 anti-MOG-negative pediatricMS patients and 25 pediatric controls did not contain Abs to OMGor MAG (data not shown). MOG is therefore a more frequenttarget for autoantibodies in pediatric MS than MAG or OMG.

DiscussionThese results show that circulating high-titer Abs to MOG arepresent in a substantial subset of pediatric-onset MS patients, inparticular children with a very early disease onset. Stringent cri-teria establish their Ag specificity: the Abs label MOG but notcontrol transfectants, they can be affinity purified with recombinantMOG but not a structurally related control protein, and addition ofsoluble MOG inhibits Ab binding. Abs to MOG from pediatric-onset MS patients also bound myelin in normal human white mat-ter and myelinated axons in subpial gray matter. The majority ofMOG-specific Abs have the IgG1 subtype that can fix complementand bind to Fc receptors, indicating that they have the potential todamage myelin or oligodendrocytes if they gain access to the CNS.Such Abs were rarely detected in pediatric controls or patients withadult-onset demyelinating conditions, including MS. In the pedi-atric patient population, their presence strongly correlated with ageat disease onset and an initial ADEM-like presentation, but notwith gender or ethnicity. Future prospective studies will examinethe relationship between such an Ab response and clinical diseasecourse, MRI features, response to treatment, and prognosis.

The pronounced therapeutic response to B cell depletion withrituximab shows that B cells play a central role in the pathogenesisof adult-onset MS. It is likely that B cells contribute to diseaseprogression not only as a source of autoantibodies, but also bypresenting myelin-derived Ags to autoreactive T cells and drivinginflammation through production of cytokines and chemokines.Ag-specific B cells are highly effective as APCs (39, 40), and theirelimination could substantially decrease T cell priming and acti-vation. The relative importance of these different mechanisms tothe pathogenesis of MS remains unresolved. Also, the specificitiesof the involved B cells and their Ab products remain largelyunknown.

Previous studies of autoantibodies in adult-onset MS yieldedconflicting results. Although some found Abs to MOG in patientswith MS or CIS (41, 42), others reported no difference in the fre-quency of anti-MOG between MS and other neurological diseases(43–45) or healthy controls (46). This controversy surrounding the

presence of MOG Abs in MS patients is likely attributable to dif-ferences in detection methods. Our flow cytometric approachtherefore emphasized detection of Abs to native, properly foldedMOG protein.

Three other groups also generated MOG-expressing cells to ex-amine the presence of autoantibodies in adult-onset MS patients.Haase et al. (47) generated a stable MOG transfectant in LTK3�

cells and found that only one of 17 serum samples from adult MSpatients labeled this transfectant, even though all patients as wellas healthy control subjects had Abs that detected linear MOG pep-tides in an ELISA. Lavile et al. compared serum Ab staining of aMOG transfected Chinese hamster ovary cell line to nontrans-fected Chinese hamster ovary cells and concluded that IgG Absspecific for native MOG were most frequently found in the serumof patients with CIS and relapsing-remitting MS (48). However,the binding ratios of the MOG and control cells were less than twofor all positive samples, even though high serum concentrationswere used for staining (1/10 dilution, compared with the 1/50 di-lution used in this study). We defined binding ratios over 5 aspositive, and obtained binding ratios as high as 200.2 in pediatricMS samples.

In another study, Zhou et al. (49) transduced a human glioblas-toma cell line with a lentivirus encoding MOG and stained thesecells at a serum dilution of 1/36. The MOG-specific Ab responsewas calculated by subtracting median fluorescence intensity valuesobtained with MOG and control transfectants. The authors re-ported that 32% of adult-onset MS patients and 4% of controlsubjects had detectable Abs, but the difference in MFI between theMOG and control cells was rather small (�50) for most sera. Bycomparison, we observed differences in MFI of 516 to 22,102 forthe six representative examples shown in Fig. 1A. We prefer toexpress the data as a binding ratio between the MOG and controltransfectants rather than as a difference in MFI because the level ofbackground differs greatly between sera. For example, a samplewith a high background and a small MOG-specific increase inbinding can have a large change in MFI (for example, MFI valuesof 600 and 500 for MOG and GFP transfectants, respectively),even though the binding ratio is small (1.2 in this example). Al-though both analysis methods permit the identification of sampleswith high levels of MOG-specific Abs, using MFI differences canresult in the classification of samples with a high background aspositive.

In this study, we have analyzed the largest collection of adult-onset MS sera to date (n � 254), and the results show differencesin the frequency of MOG-positive cases between samples fromMS centers in Canada (0/53, 0%), the US (5/129, 3.9%), and Swit-zerland (6/72, 8.3%). These apparent differences may be caused bythe overall low frequency of Abs to MOG in adult-onset MS, butdifferences in patient selection or genetic and environmental fac-tors cannot be excluded. Sample handling may also be a contrib-uting factor, but appears unlikely because we observed that MOGAbs remain detectable after multiple freeze-thaw cycles.

Our findings support the conclusion that Abs to MOG are ratheruncommon among adult-onset MS patients, whereas the anti-MOGreactivity identified in a subset of pediatric MS cases is among thestrongest autoantibody responses observed so far in MS. Usingmultimeric forms of folded and glycosylated MOG protein, wepreviously detected similar Abs to MOG in a subset of ADEMpatients (31). In the current study of pediatric MS, the presence ofMOG Abs strongly correlated with an initial ADEM-like onset,although all children subsequently experienced two or more non-ADEM demyelinating events as required for a diagnosis of MS(19). The 50% of pediatric MS patients with an initial diagnosis of

4074 MOG AUTOANTIBODIES IN PEDIATRIC MS

by guest on June 4, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

ADEM had Abs to MOG, whereas 92% of MOG negative pedi-atric MS patients did not have an ADEM-like initial event. Pro-spective studies are required to further define the relationship be-tween MOG Abs and ADEM as well as pediatric MS.

Why have we only detected Abs to MOG, but not MAG orOMG, which are also myelin surface Ags with large extracellulardomains? Chronic inflammation can result in immune responses tomultiple self-Ags, a phenomenon referred to as epitope spreading(50). However, significant immune responses are not mounted toevery Ag in the target structure. MOG is a highly encephalitogenicprotein in immunization-based animal models of MS and is theonly known myelin component to induce pathogenic Ab and T cellresponses (51, 52). MOG also has substantial sequence similaritywith the milk protein butyrophilin, and Ab as well as T cell cross-reactivity between these Ags has been demonstrated (53, 54).

Why does the prevalence of these autoantibodies change withage at disease onset in pediatric populations? Several general andpossibly interrelated factors may be involved: genetic susceptibil-ity, the kinetics and magnitude of the autoimmune response, en-vironmental factors, and the biology of myelination during child-hood. In type I diabetes, children with particular combinations ofMHC class II genes are more likely to become diabetic at a youngage and thus appear to carry a higher genetic risk (55). Similarly,children who develop MS at a young age may carry combinationsof genes that raise susceptibility to MS to a higher level than inindividuals who develop MS later in life. The frequency and func-tionality of MOG-specific T cells could also affect production ofMOG Abs. The importance of T cell–B cell collaboration washighlighted by recent studies which showed that genetically engi-neered mice with a high frequency of both MOG-specific T cellsand B cells spontaneously develop severe CNS inflammation,whereas mice with only a high frequency of MOG-specific B cellsremain healthy (56–58).

This study shows that MOG Abs identify a subset of pediatricMS patients with a very early disease onset. In contrast, MOG Absare less common in patients with adult-onset disease. Future stud-ies will address the pathogenetic significance of the emergence ofsuch autoantibodies in pediatric demyelinating diseases.

AcknowledgmentsWe acknowledge the contributions of the site investigators who contributedto the collection of the clinical samples. Members of the Wadsworth Pe-diatric Multiple Sclerosis project include: Silvia Tenembaum (Hospital dePediatria, Buenos Aires, Argentina), Anita Belman (SUNY, Stony Brook,NY), Alexei Boiko and Olga Bykova (Russian State Medical University,Moscow, Russia), Jayne Ness (Children’s Hospital of Alabama, Birming-ham, AL), Jean Mah and Cristina Stoian (University of Calgary, Calgary,Ontario, Canada), Marcelo Kremenchutzky (London Health Sciences, Lon-don, Ontario, Canada), Mary Rensel (Cleveland Clinic Foundation, Cleve-land, OH), Jin Hahn (Stanford University, Stanford, CA), Bianca Wein-stock-Guttman and Ann Yeh (SUNY, Buffalo, NY), Kevin Farrell(University British Columbia, Vancouver, BC, Canada), Mark Freedman(University of Ottawa, Ottawa, Ontario, Canada), Emmanuelle Waubant,University of California, San Francisco, CA), Martino Ruggieri (Universityof Catania, Catania, Italy), Matti Iivanainen (University of Helsinki, Hel-sinki, Finland), Virender Bhan (Dalhousie University, Halifax, Nova Sco-tia, Canada), and Marie-Emmanuelle Dilenge (Montreal Children’s Hos-pital, Montreal, Quebec, Canada).

DisclosuresThe authors have no financial conflict of interest.

References1. Hauser, S. L., and J. R. Oksenberg. 2006. The neurobiology of multiple sclerosis:

genes, inflammation, and neurodegeneration. Neuron 52: 61–76.2. Hanefeld, F. 2007. Pediatric multiple sclerosis: a short history of a long story.

Neurology 68: S3–6.

3. Chitnis, T., B. Glanz, S. Jaffin, and B. Healy. 2009. Demographics of pediatric-onset multiple sclerosis in an MS center population from the Northeastern UnitedStates. Multiple Sclerosis 15: 627–631.

4. Duquette, P., T. J. Murray, J. Pleines, G. C. Ebers, D. Sadovnick, P. Weldon,S. Warren, D. W. Paty, A. Upton, W. Hader, R. Nelson, A. Auty, B. Neufeld, andC. Meltzer. 1987. Multiple sclerosis in childhood: clinical profile in 125 patients.J. Pediatrics 111: 359–363.

5. Ghezzi, A., V. Deplano, J. Faroni, M. Grasso, M. Liguori, G. Marrosu,C. Pozzilli, I. Simone, and M. Zaffaroni. 1997. Multiple sclerosis in childhood:clinical features of 149 cases. Multiple Sclerosis 3: 43–46.

6. Boiko, A., G. Vorobeychik, D. Paty, V. Devonshire, D. Sadovnick, and the UBCMS Clinic Neurologists. 2002. Early onset multiple sclerosis: a longitudinalstudy. Neurology 59: 1006–1010.

7. Renoux, C., S. Vukusic, Y. Mikaeloff, G. Edan, M. Clanet, B. Dubois,M. Debouverie, B. Brochet, C. Lebrun-Frenay, J. Pelletier, et al. 2007. Naturalhistory of multiple sclerosis with childhood onset. N. Engl. J. Med. 356:2603–2613.

8. Gorman, M. P., B. C. Healy, M. Polgar-Turcsanyi, and T. Chitnis. 2009. In-creased relapse rate in pediatric-onset compared with adult-onset multiple scle-rosis. Arch. Neurol. 66: 54–59.

9. Ruggieri, M., P. Iannetti, A. Polizzi, et al. 2004. Multiple sclerosis in childrenunder 10 years of age. Neurol. Sci. 25(Suppl. 4): S326–S335.

10. Ness, J. M., D. Chabas, A. D. Sadovnick, D. Pohl, B. Banwell,B. Weinstock-Guttman, and the International Pediatric MS Study Group. 2007.Clinical features of children and adolescents with multiple sclerosis. Neurology68: S37–S45.

11. Banwell, B. L., and P. E. Anderson. 2005. The cognitive burden of multiplesclerosis in children. Neurology 64: 891–894.

12. MacAllister, W. S., A. L. Belman, M. Milazzo, D. M. Weisbrot,C. Christodoulou, W. F. Scherl, T. E. Preston, C. Cianciulli, and L. B. Krupp.2005. Cognitive functioning in children and adolescents with multiple sclerosis.Neurology 64: 1422–1425.

13. MacAllister, W. S., J. R. Boyd, N. J. Holland, M. C. Milazzo, L. B. Krupp, andthe International Pediatric MS Study Group. 2007. The psychosocial conse-quences of pediatric multiple sclerosis. Neurology 68: S66–S69.

14. Banwell, B., A. Ghezzi, A. Bar-Or, Y. Mikaeloff, and M. Tardieu. 2007. Multiplesclerosis in children: clinical diagnosis, therapeutic strategies, and future direc-tions. Lancet Neurol. 6: 887–902.

15. Kennedy, J., P. O’Connor, A. D. Sadovnick, M. Perara, I. Yee, and B. Banwell.2006. Age at onset of multiple sclerosis may be influenced by place of residenceduring childhood rather than ancestry. Neuroepidemiology 26: 162–167.

16. Wingerchuk, D. M. 2003. Postinfectious encephalomyelitis. Curr. Neurol. Neu-rosci. Rep. 3: 256–264.

17. Dale, R. C., C. de Sousa, W. K. Chong, T. C. Cox, B. Harding, and B. G. Neville.2000. Acute disseminated encephalomyelitis, multiphasic disseminated enceph-alomyelitis and multiple sclerosis in children. Brain 123: 2407–2422.

18. Tenembaum, S., N. Chamoles, and N. Fejerman. 2002. Acute disseminated en-cephalomyelitis: a long-term follow-up study of 84 pediatric patients. Neurology59: 1224–1231.

19. Krupp, L. B., B. Banwell, S. Tenembaum, and the International Pediatric MSStudy Group. 2007. Consensus definitions proposed for pediatric multiple scle-rosis and related disorders. Neurology 68: S7–S12.

20. Tenembaum, S., T. Chitnis, J. Ness, J. S. Hahn, and the International PediatricMS Study Group. 2007. Acute disseminated encephalomyelitis. Neurology 68:S23–36.

21. Mikaeloff, Y., S. Suissa, L. Vallee, C. Lubetzki, G. Ponsot, C. Confavreux,M. Tardieu, and K. S. Group. 2004. First episode of acute CNS inflammatorydemyelination in childhood: prognostic factors for multiple sclerosis and disabil-ity. J. Pediatrics 144: 246–252.

22. Lennon, V. A., D. M. Wingerchuk, T. J. Kryzer, S. J. Pittock, C. F. Lucchinetti,K. Fujihara, I. Nakashima, and B. G. Weinshenker. 2004. A serum autoantibodymarker of neuromyelitis optica: distinction from multiple sclerosis. Lancet 364:2106–2112.

23. Takahashi, T., K. Fujihara, I. Nakashima, T. Misu, I. Miyazawa, M. Nakamura,S. Watanabe, Y. Shiga, C. Kanaoka, J. Fujimori, S. Sato, and Y. Itoyama. 2007.Anti-aquaporin-4 antibody is involved in the pathogenesis of NMO: a study onantibody titre. Brain 130: 1235–1243.

24. Wingerchuk, D. M., V. A. Lennon, S. J. Pittock, C. F. Lucchinetti, andB. G. Weinshenker. 2006. Revised diagnostic criteria for neuromyelitis optica.Neurology 66: 1485–1489.

25. Hauser, S. L., E. Waubant, D. L. Arnold, T. Vollmer, J. Antel, R. J. Fox,A. Bar-Or, M. Panzara, N. Sarkar, S. Agarwal, A. Langer-Gould, C. H. Smith,and the HERMES Trial Group. 2008. B-cell depletion with rituximab in relaps-ing-remitting multiple sclerosis. N. Engl. J. Med. 358: 676–688.

26. Pohl, D., K. Rostasy, H. Reiber, and F. Hanefeld. 2004. CSF characteristics inearly-onset multiple sclerosis. Neurology 63: 1966–1967.

27. Schluesener, H. J., R. A. Sobel, C. Linington, and H. L. Weiner. 1987. A mono-clonal antibody against a myelin oligodendrocyte glycoprotein induces relapsesand demyelination in central nervous system autoimmune disease. J. Immunol.139: 4016–4021.

28. Linington, C., M. Bradl, H. Lassmann, C. Brunner, and K. Vass. 1988. Augmen-tation of demyelination in rat acute allergic encephalomyelitis by circulatingmouse monoclonal antibodies directed against a myelin/oligodendrocyte glyco-protein. Am. J. Pathol. 130: 443–454.

29. Brehm, U., S. J. Piddlesden, M. V. Gardinier, and C. Linington. 1999. Epitopespecificity of demyelinating monoclonal autoantibodies directed against the hu-man myelin oligodendrocyte glycoprotein (MOG). J. Neuroimmunol. 97: 9–15.

4075The Journal of Immunology

by guest on June 4, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

30. von Budingen, H. C., S. L. Hauser, J. C. Ouallet, N. Tanuma, T. Menge, andC. P. Genain. 2004. Epitope recognition on the myelin/oligodendrocyte glycop-rotein differentially influences disease phenotype and antibody effector functionsin autoimmune demyelination. Eur. J. Immunol. 34: 2072–2083.

31. O’Connor, K. C., K. A. McLaughlin, P. L. De Jager, T. Chitnis, E. Bettelli, C. Xu,W. H. Robinson, S. V. Cherry, A. Bar-Or, B. Banwell, et al. 2007. Self-antigentetramers discriminate between myelin autoantibodies to native or denatured pro-tein. Nat. Med. 13: 211–217.

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,M. Sandberg-Wollheim, et al. 2001. Recommended diagnostic criteria for mul-tiple sclerosis: guidelines from the international panel on the diagnosis of mul-tiple sclerosis. Ann. Neurol. 50: 121–127.

33. Polman, C. H., S. C. Reingold, G. Edan, M. Filippi, H.-P. Hartung, L. Kappos,F. D. Lublin, L. M. Metz, H. F. McFarland, P. W. O’Connor, et al. 2005. Diag-nostic criteria for multiple sclerosis: 2005 revisions to the “McDonald Criteria”.Ann. Neurol. 58: 840–846.

34. Poser, C. M., D. W. Paty, L. Scheinberg, W. I. McDonald, F. A. Davis,G. C. Ebers, K. P. Johnson, W. A. Sibley, D. H. Silberberg, andW. W. Tourtellotte. 1983. New diagnostic criteria for multiple sclerosis: guide-lines for research protocols. Ann. Neurol. 13: 227–231.

35. Jang, M.-H., N. P. Seth, and K. W. Wucherpfennig. 2003. Ex vivo analysis ofthymic CD4 T cells in nonobese diabetic mice with tetramers generated fromI-Ag7/class II-associated invariant chain peptide precursors. J. Immunol. 171:4175–4186.

36. 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, and K. W. Wucherpfennig.2005. Antibodies from inflamed central nervous system tissue recognize myelinoligodendrocyte glycoprotein. J. Immunol. 175: 1974–1982.

37. Newcombe, J., A. Uddin, R. Dove, B. Patel, L. Turski, Y. Nishizawa, andT. Smith. 2008. Glutamate receptor expression in multiple sclerosis lesions. BrainPathol. 18: 52–61.

38. Newcombe, J., N. Naik, and M. L. Cuzner. 1992. Monoclonal antibody 14Erecognizes an antigen common to human oligodendrocytes, Schwann cells, Berg-mann glia, and a subpopulation of reactive glia. Neurochem. Res. 17: 933–938.

39. Lanzavecchia, A. 1985. Antigen-specific interaction between T and B cells. Na-ture 314: 537–539.

40. Rivera, A., C.-C. Chen, N. Ron, J. P. Dougherty, and Y. Ron. 2001. Role of Bcells as antigen-presenting cells in vivo revisited: antigen-specific B cells areessential for T cell expansion in lymph nodes and for systemic T cell responsesto low antigen concentrations. Int. Immunol. 13: 1583–1593.

41. Berger, T., P. Rubner, F. Schautzer, R. Egg, H. Ulmer, I. Mayringer, E. Dilitz,F. Deisenhammer, and M. Reindl. 2003. Antimyelin antibodies as a predictor ofclinically definite multiple sclerosis after a first demyelinating event. N. Engl.J. Med. 349: 139–145.

42. Kennel De March, A., M. De Bouwerie, M. N. Kolopp-Sarda, G. C. Faure,M. C. Bene, and C. C. A. Bernard. 2003. Anti-myelin oligodendrocyte glycop-rotein B-cell responses in multiple sclerosis. J. Neuroimmunol. 135: 117–125.

43. Mantegazza, R., P. Cristaldini, P. Bernasconi, F. Baggi, R. Pedotti, I. Piccini,N. Mascoli, L. L. Mantia, C. Antozzi, O. Simoncini, et al. 2004. Anti-MOGautoantibodies in Italian multiple sclerosis patients: specificity, sensitivity andclinical association. Int. Immunol. 16: 559–565.

44. Markovic, M., V. Trajkovic, J. Drulovic, S. Mesaros, N. Stojsavljevic,I. Dujmovic, and M. M. Stojkovic. 2003. Antibodies against myelin oligoden-

drocyte glycoprotein in the cerebrospinal fluid of multiple sclerosis patients.J. Neurol. Sci. 211: 67–73.

45. Reindl, M., C. Linington, U. Brehm, R. Egg, E. Dilitz, F. Deisenhammer,W. Poewe, and T. Berger. 1999. Antibodies against the myelin oligodendrocyteglycoprotein and the myelin basic protein in multiple sclerosis and other neuro-logical diseases: a comparative study. Brain 122: 2047–2056.

46. Lampasona, V., D. Franciotta, R. Furlan, S. Zanaboni, R. Fazio, E. Bonifacio,G. Comi, and G. Martino. 2004. Similar low frequency of anti-MOG IgG andIgM in MS patients and healthy subjects. Neurology 62: 2092–2094.

47. 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 finespecificity of the myelin oligodendrocyte glycoprotein autoantibody response inpatients with multiple sclerosis and normal healthy controls. J. Neuroimmunol.114: 220–225.

48. Lalive, P. H., T. Menge, C. Delarasse, B. Della Gaspera, D. Pham-Dinh,P. Villoslada, H.-C. von Budingen, and C. P. Genain. 2006. Antibodies to nativemyelin oligodendrocyte glycoprotein are serologic markers of early inflammationin multiple sclerosis. Proc. Natl. Acad. Sci. USA 103: 2280–2285.

49. Zhou, D., R. Srivastava, S. Nessler, V. Grummel, N. Sommer, W. Bruck, H.-P.Hartung, C. Stadelmann, and B. Hemmer. 2006. Identification of a pathogenicantibody response to native myelin oligodendrocyte glycoprotein in multiple scle-rosis. Proc. Natl. Acad. Sci. USA 103: 19057–19062.

50. Sercarz, E. E. 2000. Driver clones and determinant spreading. J. Autoimmunity14: 275–277.

51. Von Budingen, H.-C., N. Tanuma, P. Villoslada, J.-C. Ouallet, S. L. Hauser, andC. P. Genain. 2001. Immune responses against the myelin/oligodendrocyte gly-coprotein in experimental autoimmune demyelination. J. Clin. Immunol. 21:155–170.

52. Bernard, C. C. A., T. G. Johns, A. Slavin, M. Ichikawa, C. Ewing, J. Liu, andJ. Bettadapura. 1997. Myelin oligodendrocyte glycoprotein: a novel candidateautoantigen in multiple sclerosis. J. Mol. Med. 75: 77–88.

53. Guggenmos, J., A. S. Schubart, S. Ogg, M. Andersson, T. Olsson, I. H. Mather,and C. Linington. 2004. Antibody cross-reactivity between myelin oligodendro-cyte glycoprotein and the milk protein butyrophilin in multiple sclerosis. J. Im-munol. 172: 661–668.

54. Stefferl, A., A. Schubart, M. Storch, A. Amini, I. Mather, H. Lassmann, andC. Linington. 2000. Butyrophilin, a milk protein, modulates the encephalitogenicT cell response to myelin oligodendrocyte glycoprotein in experimental autoim-mune encephalomyelitis. J. Immunol. 165: 2859–2865.

55. Hathout, E. H., N. Hartwick, O. R. Fagoaga, A. R. Colacino, J. Sharkey,M. Racine, S. Nelsen-Cannarella, and J. W. Mace. 2003. Clinical, autoimmune,and HLA characteristics of children diagnosed with type 1 diabetes before 5 yearsof age. Pediatrics 111: 860–863.

56. Bettelli, E., D. Baeten, A. Jager, R. A. Sobel, and V. K. Kuchroo. 2006. Myelinoligodendrocyte glycoprotein–specific T and B cells cooperate to induce a Devic-like disease in mice. J. Clin. Invest. 116: 2393–2402.

57. Krishnamoorthy, G., H. Lassmann, H. Wekerle, and A. Holz. 2006. Spontaneousopticospinal encephalomyelitis in a double-transgenic mouse model of autoim-mune T cell/B cell cooperation. J. Clin. Invest. 116: 2385–2392.

58. Litzenburger, T., R. Fassler, J. Bauer, H. Lassmann, C. Linington, H. Wekerle,and A. Iglesias. 1998. B lymphocytes producing demyelinating autoantibodies:development and function in gene-targeted transgenic mice. J. Exp. Med. 188:169–180.

4076 MOG AUTOANTIBODIES IN PEDIATRIC MS

by guest on June 4, 2018http://w

ww

.jimm

unol.org/D

ownloaded from