Published Ahead of Print on July 25, 2018 as 10.1212/WNL.

0000000000006035
ARTICLE OPEN ACCESS

Glutamate receptor δ2 serum antibodies in

pediatric opsoclonus myoclonus ataxia syndrome
Georgina Berridge, BSc (Hons), David A. Menassa, DPhil, Teresa Moloney, PhD, Patrick J. Waters, PhD, Correspondence
Imogen Welding, BMBCh, Selina Thomsen, Sameer Zuberi, MD, Roman Fischer, PhD, A. Radu Aricescu, PhD, Prof. Lang
Michael Pike, FRCPCH, Russell C. Dale, PhD, Benedikt Kessler, PhD, Angela Vincent, FRCPath, Ming Lim, PhD,* bethan.lang@ndcn.ox.ac.uk
Sarosh R. Irani, MRCP, DPhil,* and Bethan Lang, PhD*

®
Neurology 2018;0:e1-e10. doi:10.1212/WNL.0000000000006035

Abstract
Objective
To identify neuronal surface antibodies in opsoclonus myoclonus ataxia syndrome (OMAS)
using contemporary antigen discovery methodology.

Methods
OMAS patient serum immunoglobulin G immunohistochemistry using age-equivalent rat
cerebellar tissue was followed by immunoprecipitation, gel electrophoresis, and mass spec-
trometry. Data are available via ProteomeXchange (identifier PXD009578). This generated
a list of potential neuronal surface cerebellar autoantigens. Live cell-based assays were used to
confirm membrane-surface antigens and adsorb antigen-specific immunoglobulin Gs. The
serologic results were compared to the clinical data.

Results
Four of the 6 OMAS sera tested bound rat cerebellar sections. Two of these sera with similar
immunoreactivities were used in immunoprecipitation experiments using cerebellum from
postnatal rat pups (P18). Mass spectrometry identified 12 cell-surface proteins, of which
glutamate receptor δ2 (GluD2), a predominately cerebellar-expressed protein, was found at
a 3-fold-higher concentration than the other 11 proteins. Antibodies to GluD2 were identified
in 14/16 (87%) OMAS samples, compared with 5/139 (5%) pediatric and 1/38 (2.6%) adult
serum controls (p < 0.0001), and in 2/4 sera from patients with neuroblastoma without
neurologic features. Adsorption of positive OMAS sera against GluD2-transfected cells sub-
stantially reduced but did not eliminate reactivity toward cerebellar sections.

Conclusion
Autoantibodies to GluD2 are common in patients with OMAS, bind to surface determinants,
and are potentially pathogenic.

*These authors contributed equally to this work.

From the Oxford Autoimmune Neurology Group (G.B., D.A.M., T.M., P.J.W., I.W., S.T., M.P., A.V., S.R.I., B.L.), Nuffield Department of Clinical Neurosciences, John Radcliffe Hospital,
Oxford; Target Discovery Institute (G.B., R.F., B.K.), NDM Research Building, University of Oxford, Old Road Campus, Oxford; Paediatric Neurosciences Research Group (S.Z.), School
of Medicine, University of Glasgow; Division of Structural Biology (A.R.A.), Nuffield Department of Clinical Medicine, University of Oxford, UK; Clinical Neuroimmunology (R.C.D.),
Institute for Neuroscience and Muscle Research, University of Sydney, Australia; Children’s Neuroscience Centre (M.L.), Evelina London Children’s Hospital at St Thomas’ NHS
Foundation Trust, King’s Health Partners Academic Health Science Centre, London; and Faculty of Medicine and Life Sciences (M.L.), King’s College London, UK.

Go to Neurology.org/N for full disclosures. Funding information and disclosures deemed relevant by the authors, if any, are provided at the end of the article.

The Article Processing Charge was funded by the Wellcome Trust.

This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 (CC BY), which permits unrestricted use, distribution, and reproduction in any
medium, provided the original work is properly cited.

Copyright © 2018 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Academy of Neurology. e1
 Glossary
CBA = cell-based assay; DAPI = 4’,6-diamidino-2-phenylindole; DCN = deep cerebellar nuclei; GABA = γ-aminobutyric acid;
GluD2 = glutamate receptor δ2; HA = hemagglutinin; IgG = immunoglobulin G; OMAS = opsoclonus myoclonus ataxia
syndrome; RT = room temperature.

Opsoclonus myoclonus ataxia syndrome (OMAS), also known cerebellar tissue and identified autoantibodies to the extracel-
as “dancing eye syndrome,” is a rare disorder that mainly affects lular domain of glutamate receptor δ2 (GluD2) in the sera of
children. OMAS is characterized by conjugate, asynchronous, pediatric patients with OMAS.
multidirectional eye movements (opsoclonus), myoclonus,
ataxia, behavioral and sleep disturbance, and sometimes cog-
nitive decline.1–3 The clinical and imaging assessment of the Methods
disease suggest involvement of the cerebellum and/or pontine Patient material
omnipause neurons. MRI in the acute phase is usually normal, OMAS serum samples (data available from Dryad, table 1,
but recently, patients with long-standing OMAS have been doi.org/10.5061/dryad.tq61224) were collected at diagnosis
shown to have a reduction in the cerebellar gray matter volume, from 16 children (median age 2 years, range 1–8.5 years);
especially in the vermis and flocculonodular lobes, alongside further samples were available at 48 weeks in 5 of these
a more generalized reduction in cortical thickness.4 patients. Eight (53%) were male and 11 (73%) had an asso-
ciated neuroblastoma. As outlined in the table, disease control
In pediatric OMAS, the age at onset is typically within the sera were available from children with new-onset epilepsy
relatively narrow 12- to 36-month age range.2,5 Furthermore, (median age 2.2 years, range 0.5–3 years, n = 78), Rasmussen
OMAS associates with an underlying neuroblastoma in ap- encephalitis (age 8.2 years, range 1–18 years, n = 23) and
proximately 50% of pediatric patients.1,6 Neuroblastoma is the autoimmune and other forms of encephalitis (age 8.25 years,
most common solid tumor of childhood, derived from the range 0.4–15 years, n = 38), and from healthy adult controls
sympathetic nervous system, and occurs almost exclusively in (n = 37). Resected neuroblastoma tissue from one patient
infancy and early childhood, with a median peak age between (18-month-old female) was available for study. Sera from 4
18 and 24 months.7 patients with neuroblastoma but without neurologic dys-
function (absence or presence of neurologic syndrome is
While the precise pathogenesis of OMAS is undefined, the a recorded field) were obtained from the Children’s Cancer
close association with neuroblastoma strongly suggests a par- and Leukaemia and Tissue Bank, Leicester Royal Infirmary.
aneoplastic autoimmune process. B cell expansions with ele-
vated levels of B cell activating factor have been shown in the Cerebellar tissue staining
CSF of patients with OMAS,8,9 and an HLA association has Sprague-Dawley rats (P18, and adult) were perfused with saline
been established in some patients.10 Moreover, the neuro- (0.9%) under deep anesthesia. The brains were flash-frozen in
blastomas have marked lymphocytic infiltrates, akin to the isopentane at −40°C. Frozen rat brain sections (12 μm thick)
thymic histology observed in early-onset myasthenia gravis.11 were fixed with 4% paraformaldehyde (15 minutes at room
Finally, some studies describe binding of OMAS patient se- temperature [RT]). Sections were incubated with sera or
rum immunoglobulins to Purkinje cells, the surface of cere- commercial antibodies (1:100–1:200) before incubation with
bellar dendritic arborizations, and to a few candidate neuronal biotinylated goat anti-human antibody (1:200; Vector Labo-
proteins, although no reproducible antigenic targets have yet ratories, Burlingame, CA). ABC complex (Elite kit; Vector
been established.11–15 Indeed, in one recent study, serum Laboratories) and diaminobenzidine (0.5 mg/mL plus 0.03%
immunoglobulin G (IgG) precipitated 7 neuronal proteins of H2O2) were used to develop the reaction. For immunoflu-
found in neuroblastoma cell lines but none were shown to be orescence, sections were incubated with commercial antibodies
direct targets of the autoantibodies.16 or serum (1:100–1:200) before fixation (15 minutes, RT). IgG
binding was detected with a species-appropriate fluorescently
The striking overlap of symptom onset in OMAS and the peak labeled Alexa Fluor secondary antibody (ThermoFisher Sci-
age of neuroblastoma detection led us to hypothesize that this entific, Waltham, MA) and counterstained with DAPI (4’,6-
temporal juxtaposition was important in the pathogenesis of diamidino-2-phenylindole). Further details are presented in
OMAS. Furthermore, the above observations strongly impli- figure legends and data available from Dryad (Methods, doi.
cate cerebellar structures in disease etiology. Therefore, in our org/10.5061/dryad.tq61224).
search for putative pathogenic autoantibodies in OMAS, we
hypothesized an advantage to using cerebellar tissue repre- Neuroblastoma immunofluorescence
sentative of humans at approximately 2 years of age. Here, we The neuroblastoma tissue from one patient (OMAS 15) was
combine immunohistology, immunoprecipitation, mass spec- frozen in OCT (Fisher Healthcare). For immunofluorescence,
trometry, and bioinformatic techniques on age-equivalent rat neuroblastoma sections were incubated with commercial

e2 Neurology | Volume , Number  | Month 0, 2018 Neurology.org/N

Table Autoantibody specificities in patients with OMAS and in controls
No. Median age, y (range) Tumor GluD2, n (%) EAAT2, n GABAB R, n (%) Cerebellin, n

OMAS 16 2 (1–8.5) 11 14 (87.5) 0 1 (6.25) 0

Disease controls (children)

Rasmussen encephalitis 23 8.2 (1–18) 0 2 (8.7) 0 0 0

New-onset epilepsy 78 2.2 (0.5–3.8) 0 0 0 2 (2.6) 0

Autoimmune encephalitis 38 8.3 (0.4–15) 0 3 (7.8) 1 0 NP

Neuroblastoma

Children 3 2.5 (1.9–4.2) 3 2 (66) 0 0 0

Adults 1 20 1 0 0 0 0

Healthy controls

Adults 37 35 (21–70) 0 1 (3) 0 0 0

Abbreviations: EAAT2 = excitatory amino acid transporter 2; GABAB R = γ-aminobutyric acid type B receptor; GluD2 = glutamate receptor δ2; NP = not
performed; OMAS = opsoclonus myoclonus ataxia syndrome.
Mass spectrometry–derived targets (GluD2, EAAT2, the GABAB receptor, and cerebellin) were used in separate cell-based assays. Samples were considered
positive if surface binding was observed at a dilution of 1:50 (see methods). GluD2 antibodies were present in 14 of 16 patients with OMAS before treatment
and only 1 of 8 at week 48 (data available from Dryad, table 1, doi.org/10.5061/dryad.tq61224).

antibodies (1:200) and stained with a species-appropriate acquisition on a Thermo Q Exactive mass spectrometer (data
fluorescently labeled Alexa Fluor secondary antibody and available from Dryad, Methods, doi.org/10.5061/dryad.
counterstained with DAPI. tq61224). Methods for filtering of protein hits are illustrated
in figure 1 and data available from Dryad (table 2, doi.org/10.
Isolation of autoantigens 5061/dryad.tq61224).
The 2 OMAS sera with the strongest binding to the cerebellum
and the deep cerebellar nuclei (DCN) and pooled healthy Cell-based assays
control serum were used for the discovery of autoantigens via Complementary DNA encoding the full-length human GluD2
immunoprecipitation and analysis by mass spectrometry. Post- mature polypeptide (GenBank ID NM_001510; Asp24-
natal day 18 rat cerebellum was gently triturated and washed Ile1007) was cloned into the pHLsec vector (PMID:
with phosphate-buffered saline, then incubated with 85 to 100 17001101), immediately downstream of the secretion signal
μL undiluted patient or control serum for 60 minutes with sequence and an external hemagglutinin (HA) peptide
occasional inversion before the addition of solubilization buffer (YPYDVPDYA), and was used in a live cell-based assay (CBA)
(150 mM NaCl, 10 mM Tris HCl, pH 7.4, 1% Triton X-100, to detect antibody binding. Culture and staining procedures for
protease inhibitor cocktail [P8340; Sigma-Aldrich, St. Louis, the live CBAs were performed and scored as previously
MO]) for 60 minutes on ice. The suspension was harvested after described.17,18 Sera (from 1:50) or commercial antibodies
2 rounds of centrifugation (2,000g for 5 minutes). Protein (1:750) were applied to live transfected cells for 1 hour at RT
G Sepharose beads (Sigma) were added to the supernatant followed by 4% paraformaldehyde fixation, washing, and in-
(3 hours at 4°C) to bind the bound antibody-antigen complexes, cubation with unlabeled goat anti-human Fc-specific antibody
then extensively washed stepwise (150 mM through 1 M NaCl (1:750, Fisher A31125), and finally a third antibody layer with
in solubilization buffer). The IgG-bound proteins were eluted by Alexa Fluor 568 donkey anti-goat IgG (1:750, Fisher A11057).
heating the Protein G Sepharose beads to 90°C in Laemmli Binding of commercial antibodies was detected using the ap-
sample buffer and the eluted proteins were electrophoresed propriate species-specific secondary antibody (Alexa Fluor 568
(4%–12% sodium dodecyl sulfate gradient gel [WG1402; rabbit anti-mouse A-11061 and Alexa Fluor 568 goat anti-rabbit
Invitrogen, Carlsbad, CA]). The protein bands were visualized A-11011).
with Imperial blue stain (ThermoFisher).
Standard protocol approvals, registrations,
Analysis by mass spectrometry and patient consents
Eluates from the immunoprecipitation were prepared for Ethics for this study was covered by the research ethics com-
liquid chromatography–tandem mass spectrometry by pool- mittee (REC) (16/YH/0013) and London-Fulham (13/LO/
ing the Laemmli sample buffer eluted immunoprecipitate 0706 NRES), Children’s Hospital at Westmead (12/SCHN/
fractions followed by chloroform-methanol precipitation. 395 and 09/SCHN/56), and Glasgow ERUK study (REC
Mass spectrometry was performed using data-dependent reference: 13/WS/0299). All animal work was conducted

Neurology.org/N Neurology | Volume , Number  | Month 0, 2018 e3

 Figure 1 Autoantigen identification in OMAS

(A) Serum immunoglobulin G (1:100) binding to rat cerebellar sections (12 μm) from HCs (A.a, A.c) or patients with OMAS (A.b, A.d). Strong staining is observed
with OMAS sera in the granular layer (open arrow) and also in areas of the deep cerebellar nuclei (boxed area in upper panels, filled arrow in lower panels).
Scale bars: 500 μm (A.a, A.b) and 200 μm (A.c, A.d). (B) Gel electrophoresis of postnatal day 20 rat cerebellum tissue immunoprecipitate. Eluted samples after
immunoprecipitation with OMAS1 sera and pooled HC sera were run on a 4% to 12% sodium dodecyl sulfate precast gradient gel (WG1402; Invitrogen).
(A) Unique band, approximate molecular weight of 100 to 110 kDa was seen exclusively in OMAS samples. The boxed area was excised for mass spectrometry.
(C) Flow diagram for filters in mass spectrometry experiments. The mass spectrometry proteomics data have been deposited to the ProteomeXchange
Consortium via the PRIDE partner repository31,32 with the dataset identifier PXD009578. (D) Identification of GluD2 as a putative autoantigen target in OMAS.
Relative amounts of 12 surface-expressed neuronal proteins immunoprecipitated by 2 different OMAS sera (red/white bars) and not by HC (blue bars). For full
description of the identified proteins, see data available from Dryad (table 2, doi.org/10.5061/dryad.tq61224). GluD2 = glutamate receptor δ2; HC = healthy
control; MW = molecular weight; OMAS = opsoclonus myoclonus ataxia syndrome.

according to British Home Office regulations and under license was characterized by widespread IgG immunoreactivity of the
(Home Office: 4003581). cerebellar cortex, especially within the granular layer, and
strong IgG binding to the paravermal zone, where the DCN
Data availability are located; the plane of the section includes the interposed
Supplementary data are available from Dryad, doi.org/10. nucleus. There was no evident staining in the white matter of
5061/dryad.tq61224. The mass spectrometry proteomics the cerebellum (figure 1A). The 2 OMAS sera with the largest
data have been deposited to the ProteomeXchange Consor- volume of serum available, which showed this pattern (5-year-
tium via the PRIDE partner repository19,20 with the dataset old female and 2-year-old male; both with neuroblastoma),
identifier PXD009578. were used in the antigen discovery experiments.

Our previous attempts to identify putative antigens using

Results tissue derived from embryonic or postnatal rat tissue (<P6)
OMAS serum IgG binding and downstream had proven unsuccessful (data not shown). Therefore, cere-
proteomic analyses bellar tissue of postnatal rat pups (P17-20), considered to be
Initially, adult rat brain sections were used to look for anti- age-equivalent to 18- to 24-month-old humans,21 were used
bodies in OMAS sera. This revealed a distinct pattern of instead. Precipitated OMAS IgG-antigen complexes were
immunoreactivity in 4 of the 6 samples tested. This pattern subjected to sodium dodecyl sulfate–polyacrylamide gel

e4 Neurology | Volume , Number  | Month 0, 2018 Neurology.org/N

electrophoresis analysis and compared to a control sera (n = GluD2-specific autoantibodies
5). A band of approximately 100 to 110 kDa was identified This identification of GluD2 as an autoantigen was confirmed
from OMAS patient gels (figure 1B). This region was excised using a CBA.17 HEK293T cells were transfected with com-
from the OMAS and control gels and subjected to analysis by plementary DNA encoding GluD2 fused to an extracellular
tandem mass spectrometry. The excised bands identified HA tag. Expression of GluD2 was verified with a commercial
GluD2 in the patient but not control samples. The eluates antibody against the intracellular C-terminus of GluD2 (figure
typically contained approximately 18,000 peptides matching 2A; permeabilized CBA) and with a commercial antibody to
to several hundred proteins were isolated from gel bands. To the extracellular HA tag on the surface of live (non-
identify targets of potentially pathogenic autoantibodies, permeabilized) GluD2-transfected cells (figure 2B). Having
stringent filters were applied to filter out proteins present in established surface expression of GluD2, we observed that the
the controls, and from the unique ones, to identify cerebellar- OMAS sera (1:50 dilution) used in the antigen-discovery
specific membrane proteins with an extracellular domain. This program, but not healthy control sera, contained IgG that
reduced the putative target pool to 12 proteins (figure 1, C bound GluD2-transfected live HEK cells (figure 2, C and D).
and D; and data available from Dryad, table 2, doi.org/10.
5061/dryad.tq61224). Of these 12 proteins, GluD2, which Overall, 14 of 16 (87.5%) OMAS samples bound the extracel-
shows high cerebellar/Purkinje cell specificity,22 was detected lular domain of GluD2-transfected cells with endpoint titrations
at approximately 3-fold-higher levels than any of the others between 1:50 and 1:400 (figure 2E). The 2 seronegative patients
proteins and was enriched by 20-fold as compared to healthy were male (11 and 18 months old), both with an associated
control samples. neuroblastoma. In addition, 2 of the 4 (50%) sera from patients

Figure 2 GluD2 live cell-based assay

HEK 293T cells were transfected with complementary DNA encoding full-length GluD2, which had an extracellular HA tag. (A) Commercial antibody to GluD2
(1:200, D13266; Frontier Institute Japan, directed against an intracellular epitope) binds to permeabilized GluD2-transfected cells and surface expression of
GluD2 tagged with HA was confirmed using an anti-HA antibody (B). Sera (1:50 dilution) from patients with OMAS (C) but not HCs (D) bound the surface of the
GluD2-transfected cells. GluD2-reactive immunoglobulin Gs were removed after adsorption against GluD2-transfected (E) but not when adsorbed against
mock-transfected (F) HEK cells. Scale bar = 10 μm. Graph (G) showing endpoint titration of all samples. Samples considered positive (solid line) if signal is
observed at a titration of 1:40 or above.17 There is a significant difference between the groups (p < 0.0001; Kruskal-Wallis test). AE = autoimmune encephalitis;
GluD2 = glutamate receptor δ2; HA = hemagglutinin; HC = healthy control; NB = neuroblastoma; NOE = new-onset epilepsy; OMAS = opsoclonus myoclonus
ataxia syndrome; RE = Rasmussen encephalitis.

Neurology.org/N Neurology | Volume , Number  | Month 0, 2018 e5

 with neuroblastoma without neurologic features showed GluD2 doi.org/10.5061/dryad.tq61224). However, γ-aminobutyric
antibodies (table). By contrast, in the control groups, only 5 of acid type B (GABAB)-receptor antibodies were detected in 1 of
139 (3.6%) of the pediatric neurologic controls had GluD2 15 OMAS and 2 of 139 disease controls. Samples at 48-week
antibodies (Rasmussen encephalitis 2/23, new-onset epilepsy follow-up were available post immunotherapy from 8 patients
0/78, autoimmune encephalitis 3/38; p < 0.0001, Fisher exact with OMAS, 7 of which had been GluD2 antibody–positive at
test) (data available from Dryad, table 3, doi.org/10.5061/ presentation. Only one sample remained GluD2 antibody–
dryad.tq61224). One of the healthy controls showed binding positive at 48 weeks in an asymptomatic patient (data available
(table). from Dryad, table 1, doi.org/10.5061/dryad.tq61224).

To confirm antigenic specificity, GluD2-reactive OMAS sera GluD2 expression in cerebellum and
were adsorbed either against GluD2-transfected or untrans- neuroblastoma tissue
fected HEK cells. Only GluD2 adsorption eliminated the In light of these findings, we revisited the cerebellar staining
binding (figure 2, E and F). Furthermore, all OMAS sera were (figure 3). Application of GluD2-adsorbed sera to rat cere-
negative for IgG binding to EAAT2 (excitatory amino acid bellar sections revealed a marked reduction of staining in the
transporter 2), and cerebellin, identified at lower levels by the granular area and at the site of the interposed nucleus of the
mass spectrometry (table; data available from Dryad, table 2, DCN. However, residual staining was still observed in the 2

Figure 3 Binding of OMAS sera to GluD2 and other targets in cerebellar tissue

Binding patterns of HC and patient with OMAS. OMAS sera (green) bound the DCN area (B) and granular cells (C) this binding was partially reduced after serum
absorption against GluD2-transfected HEK cells (E and F) and unchanged after adsorption against untransfected HEK cells (H and I). The plane of the section
that has been consistently cut usually included the interposed nucleus of the DCN area outlined by the dashed line; as no staining was seen in sections A, D, or
G, no line was drawn. Similar results are shown for the bind serum dilution 1:100. Antibody binding was visualized by Alexa Fluor 488 goat anti-human (1:750,
Fisher A-11013) and counterstained with DAPI. Scale bar = 100 μm. DAPI = 4’,6-diamidino-2-phenylindole; DCN = deep cerebellar nuclei; GluD2 = glutamate
receptor δ2; HC = healthy control; OMAS = opsoclonus myoclonus ataxia syndrome.

e6 Neurology | Volume , Number  | Month 0, 2018 Neurology.org/N

sera in which sufficient quantities remained for further testing pediatric OMAS. However, the IgG cerebellar reactivities ob-
(OMAS 5, OMAS 14). Also, a neuroblastoma from one of the served after GluD2-IgG absorption suggest it is not the sole
GluD2-positive OMAS patients (OMAS 15) bound the potentially pathogenic agent in OMAS, and future studies should
commercial anti-GluD2 antibody indicating the presence of aim to define these other autoantigens. Nevertheless, links be-
GluD2 within that tumor (figure 4). tween GluD2 and several aspects of OMAS offer intriguing
insights and are discussed in more detail below.
Taken together, these results indicate that autoantibodies to
GluD2 are frequently present in OMAS sera and target in- The ionotropic GluD2 is a cerebellar-specific receptor involved
terposed nuclei of the DCN and other cerebellar structures. in synaptic organization and thus is an appropriate target for
However, residual staining after GluD2-specific adsorption antibodies in OMAS. Children with mutations in the GluD2
implies the presence of additional, as yet unidentified, anti- gene (GRID2) show developmental delay, a loss of acquired
bodies that target similar brain regions. motor skills, ocular apraxia, cerebellar ataxia, and cerebellar
atrophy.23,24 GluD2 is highly expressed on the dendritic spines
of Purkinje cells. These cells project GABAergic neurons into
Discussion the vermis and DCN, the output cells of the cerebellum.
Several convergent datasets strongly suggest OMAS has an au- Modulation of these projections may alter circuitry of the
toimmune basis.1,8,9,12–16 However, despite several efforts to cerebellum (vermis and fastigial nuclei), the inferior olives, and
date, target antigens have remained elusive. In this study, mass the brainstem saccade premotor neurons (excitatory and in-
spectrometry and bioinformatic techniques using age-equivalent hibitory burst neurons, and omnipause neurons).25 Indeed,
cerebellar tissue were used to identify GluD2 as an autoantigen GluD2-deficient mice, with fewer functional synapses between
in OMAS. The expression of GluD2 in neuroblastoma tissue the parallel fibers and Purkinje cells, have involuntary sponta-
taken from a GluD2 sera–positive patient with OMAS was neous eye and limb movements.26
confirmed by immunofluorescence. Given data implicating cer-
ebellar structures in disease pathogenesis, antigen-specific GluD2 is especially highly expressed at the parallel fiber-
modulation of GluD2 may underlie some features of OMAS. Purkinje cell synapse. At this synapse, GluD2 interacts with
The results support that antibodies that bind the extracellular cerebellin,27 a molecule that we also found in the immuno-
domain of GluD2 may be a potentially pathogenic antibody in precipitates from patient IgG-GluD2 complexes. Indeed, by

Figure 4 Expression of GluD2 in OMAS neuroblastoma tissue

Sections of neuroblastoma from a child with OMAS who had serum GluD2 antibodies were incubated in commercial antibodies (A) (rabbit anti-GluD2,
D13266; Frontier Institute Japan) or mouse anti-HA (B) (H3663, Sigma-Aldrich) at a dilution of 1:200. The sections were fixed (15 minutes in 3% PFA) and stained
with a species-appropriate secondary antibody (Alexa Fluor 568 rabbit anti-mouse A-11061; or Alexa Fluor 568 goat anti-rabbit A-11011) and counterstained
with DAPI. The sections demonstrate the presence of GluD2 in the tumor. Scale bar: 25 μm. DAPI = 4’,6-diamidino-2-phenylindole; GluD2 = glutamate receptor
δ2; HA = hemagglutinin; OMAS = opsoclonus myoclonus ataxia syndrome; PFA = paraformaldehyde.

Neurology.org/N Neurology | Volume , Number  | Month 0, 2018 e7

 linking GluD2, neuroblastomas, and the cerebellar nuclei, our to children with OMAS but were also found in 2 of 4 children
data generate several hypotheses offering potential insights with neuroblastoma but without neurologic disease. We spec-
into OMAS etiology and pathogenicity. First, OMAS is a very ulate that these antibodies, which may have increased in re-
rare condition and pediatric onset is most often within a very sponse to the underlying tumor are nonpathogenic, not at
narrow temporal window of 12 to 48 months. It is known that a critical threshold for that individual, or unable to gain anti-
in this early period, GluD2 expression rises in the cerebel- genic access through the blood-brain barrier. Nevertheless, this
lum,22 and concurrently, neuroblastomas, which we show can may be biologically plausible as the antibody may have been
express GluD2, are also maturing. It may be this ectopic ex- induced by the presence of an underlying tumor. The antibody
pression in the neuroblastoma that breaks immunologic tol- was also found in a small number (3.6%) of neurologic con-
erance and leads to GluD2 autoantibodies, which can auto trols; the tumor status in all but one of these patients was
react with brain structures. Second, given the IgG staining unknown. However, the significantly increased occurrence of
pattern observed with the OMAS sera, the brain structures these antibodies vs controls (87.5% vs 3.6%) offers a potentially
that would be targeted by the antibodies include focal cere- useful diagnostic test in OMAS. Finally, despite a marked re-
bellar nuclei. These nuclei have roles in saccadic eye move- duction in cerebellar staining after GluD2-reactive IgG re-
ments, omnipause neuron function, and ataxia.25 The origin moval, some staining remained when reapplied to cerebellar
of the myoclonus in OMAS is not well explained on the basis sections, indicating the presence of further autoantibodies to
of a purely cerebellar dysfunction and this aspect requires additional antigenic targets.
further investigation.
Taken together, our findings provide possible mechanistic
GluD2 autoantibodies have been reported previously, al- explanations for the site of the lesion in OMAS, the charac-
though largely using methodology that favors detection of teristic age at OMAS onset, and the relationship between the
autoantibodies against intracellular epitopes. Several single or tumor and the immune system. Antibodies to surface-expressed
small case reports have described antibodies to GluD2, and GluD2 could identify a therapy-responsive disorder that would
other glutamate-receptor subtypes, mainly in adult patients benefit from early treatment and tumor surveillance.
with cerebellitis and encephalitis. 28–30 However, the peptide-
based ELISAs used are unlikely to detect antibodies that react Author contributions
with the surface of native neuronal proteins. By contrast, in Georgina Berridge: drafting/revising the manuscript, study
one patient with transverse myelitis following allogenic stem concept or design, analysis or interpretation of data, accepts
cell transplantation, patient serum IgG stained both the cer- responsibility for conduct of research and will give final approval,
ebellar molecular layer and GluD2-transfected HEK cells. In acquisition of data, statistical analysis. David A. Menassa:
this study, GluD2 antibodies were not detected in approxi- drafting/revising the manuscript, study concept or design,
mately 300 disease and healthy controls.31 analysis or interpretation of data, accepts responsibility for
conduct of research and will give final approval, acquisition of
The selection of patient sera and the starting material were data. Teresa Moloney: analysis or interpretation of data, accepts
both critical in this study. The chosen sera bound to specific responsibility for conduct of research and will give final ap-
areas of the cerebellum, particularly the DCN, while the proval, acquisition of data. Patrick J. Waters: drafting/revising
cerebellar tissue used for the mass spectrometry experiment the manuscript, study concept or design, analysis or interpre-
was obtained from rats at an age equivalent to 18 to 24 human tation of data, accepts responsibility for conduct of research and
months. Previous experiments using fetal material had been will give final approval, contribution of vital reagents/tools/
unsuccessful, which is consistent with the very low expression patients. Imogen Welding: analysis or interpretation of data,
of GluD2 before birth, in both rodents and humans, and its accepts responsibility for conduct of research and will give final
rapid increase post partum. approval, acquisition of data. Selina Thomsen: analysis or in-
terpretation of data, accepts responsibility for conduct of re-
Despite being the first autoantigen with pathogenic potential search and will give final approval, acquisition of data. Sameer
described at high frequency in a substantial cohort of patients Zuberi: drafting/revising the manuscript, accepts responsibility
with OMAS, our study has several limitations. First, albeit only for conduct of research and will give final approval, acquisition of
studied in a subset of patients, the antibody frequently dis- data. Roman Fischer: drafting/revising the manuscript, study
appeared rapidly following immunotherapy. However, we are concept or design, analysis or interpretation of data, accepts
aware of one patient in whom it persisted for >18 years of active responsibility for conduct of research and will give final ap-
disease.32 The small sample size (n = 8) and later serial sam- proval, contribution of vital reagents/tools/patients, acquisition
pling (48 weeks from disease onset) did not permit evaluation of data, study supervision. A. Radu Aricescu: drafting/revising
of correlation of antibodies to disease activity. Second, the the manuscript, accepts responsibility for conduct of research
serum GluD2 antibody levels were not very high (maximum 1: and will give final approval, contribution of vital reagents/tools/
400), although, in general, this can depend on the relative and patients. Michael Pike: drafting/revising the manuscript, accepts
native cell-surface expression of antigenic proteins. In addition, responsibility for conduct of research and will give final ap-
this study only examined serum. CSF may have offered addi- proval, acquisition of data. Russell C. Dale: analysis or in-
tional information. Third, GluD2 antibodies were not unique terpretation of data, accepts responsibility for conduct of

e8 Neurology | Volume , Number  | Month 0, 2018 Neurology.org/N

research and will give final approval, contribution of vital Action Medical Research, DES Society, GOSH Charity, NIHR,
reagents/tools/patients, acquisition of data. Benedikt Kessler: MS Society, SPARKS charity; receives research support grants
drafting/revising the manuscript, analysis or interpretation of from the London Clinical Research Network and Evelina Ap-
data, accepts responsibility for conduct of research and will give peal; and has received consultation fees from CSL Behring,
final approval, acquisition of data, study supervision. Angela travel grants from Merck Serono, and educational grants to
Vincent: drafting/revising the manuscript, accepts responsibility organize meetings by Novartis, Biogen Idec, Merck Serono,
for conduct of research and will give final approval, study su- and Bayer. S. Irani is a coapplicant and receives royalties on
pervision, obtaining funding. Ming Lim: drafting/revising the patent application WO/2010/046716 titled “Neurological
manuscript, study concept or design, analysis or interpretation Autoimmune Disorders.” The patent has been licensed to
of data, accepts responsibility for conduct of research and will EUROIMMUN AG for the development of assays for LGI1
give final approval. Sarosh R. Irani: drafting/revising the man- and other VGKC-complex antibodies. S.R.I. has received
uscript, study concept or design, analysis or interpretation of speaker fees from MedImmune. B. Lang is a coapplicant and
data, accepts responsibility for conduct of research and will give receives royalties on patent application WO/2010/046716 ti-
final approval, contribution of vital reagents/tools/patients, ac- tled “Neurological Autoimmune Disorders.” The patent has
quisition of data, statistical analysis, study supervision, obtaining been licensed to EUROIMMUN AG for the development of
funding. Bethan Lang: drafting/revising the manuscript, study assays for LGI1 and other VGKC-complex antibodies. Go to
concept or design, analysis or interpretation of data, accepts Neurology.org/N for full disclosures.
responsibility for conduct of research and will give final ap-
proval, contribution of vital reagents/tools/patients, acquisition Received December 19, 2017. Accepted in final form May 18, 2018.
of data, statistical analysis, study supervision, obtaining funding.
References
1. Pranzatelli MR. The immunopharmacology of the opsoclonus-myoclonus syndrome.
Study funding Clin Neuropharmacol 1996;19:1–47.
B.L. is supported by the Pablove Foundation (United States). 2. Pike M. Opsoclonus myoclonus syndrome. Handb Clin Neurol 2013;112:
1209–1211.
S.R.I. is supported by the Wellcome Trust (104079/Z/14/Z), 3. Brunklaus A, Pohl K, Zuberi SM, de Sousa C. Outcome and prognostic features in
the UCB-Oxford University Alliance, BMA research grants: opsoclonus-myoclonus syndrome from infancy to adult life. Pediatrics 2011;128:
388–394.
2013 Vera Down and 2017 Margaret Temple grants, Epilepsy 4. Anand G, Bridge H, Rackstraw P, et al. Cerebellar and cortical abnormalities in paediatric
Research UK (P1201), and by the Fulbright UK-US Commis- opsoclonus-myoclonus syndrome. Dev Med Child Neurol 2015;57:265–272.
5. Gorman MP. Update on diagnosis, treatment, and prognosis in opsoclonus-myoclonus-
sion (MS-Research Society Award). For clinical and biobanking ataxia syndrome. Curr Opin Pediatr 2010;22:745–750.
component of this study, M.L. and B.L. were supported by the 6. Tate ED, Allison TJ, Pranzatelli MR, et al. Neuroepidemiologic trends in 105 US cases
GOSH (V1214) and SPARKS (11OUH01) charities. B.K. and of pediatric opsoclonus-myoclonus syndrome. J Pediatr Oncol Nurs 2005;22:8–19.
7. Brunklaus A, Pohl K, Zuberi SM, et al. Investigating neuroblastoma in childhood
R.F. are supported by the Kennedy Trust Fund. The authors opsoclonus-myoclonus syndrome. Arch Dis Child 2012;97:461–463.
thank the CCLG Tissue Bank for access to samples, and con- 8. Pranzatelli MR, Tate ED, Hoefgen ER, Swan JA, Colliver JA. Therapeutic down-
regulation of central and peripheral B-cell-activating factor (BAFF) production in
tributing CCLG Centres, including members of the ECMC pediatric opsoclonus-myoclonus syndrome. Cytokine 2008;44:26–32.
Paediatric Network. The CCLG Tissue Bank is funded by 9. Fühlhuber V, Bick S, Kirsten A, et al. Elevated B-cell activating factor BAFF, but not
APRIL, correlates with CSF cerebellar autoantibodies in pediatric opsoclonus-
Cancer Research UK and CCLG. The research was funded/ myoclonus syndrome. J Neuroimmunol 2009;210:87–91.
supported by the National Institute for Health Research 10. Hero B, Radolska S, Gathof BS. Opsomyoclonus syndrome in infancy with or without
neuroblastoma is associated with HLA DRB101. Pediatr Blood Cancer 2005:45. Abstract.
(NIHR) Oxford Biomedical Research Centre. The views 11. Raffaghello L, Conte M, De Grandis E, et al. Immunological mechanisms in
expressed are those of the author(s) and not necessarily those of opsoclonus-myoclonus associated neuroblastoma. Eur J Paediatr Neurol 2009;13:
219–223.
the NHS, the NIHR, or the Department of Health. 12. Connolly AM, Pestronk A, Mehta S, et al. Serum autoantibodies in childhood
opsoclonus-myoclonus syndrome: an analysis of antigenic targets in neural tissues.
J Pediatr 1997;130:878–884.
Disclosure 13. Blaes F, Fühlhuber V, Preissner KT. Identification of autoantigens in pediatric
G. Berridge, D. Menassa, and T. Moloney report no disclosures opsoclonus-myoclonus syndrome. Expert Rev Clin Immunol 2007;3:975–982.
14. Blaes F, Pike MG, Lang B. Autoantibodies in childhood opsoclonus-myoclonus
relevant to the manuscript. P. Waters is a coapplicant and syndrome. J Neuroimmunol 2008;201-202:221–226.
receives royalties on patent application WO/2010/046716 ti- 15. Panzer JA, Anand R, Dalmau J, et al. Antibodies to dendritic neuronal surface antigens
tled “Neurological Autoimmune Disorders.” The patent has 16.
in opsoclonus myoclonus ataxia syndrome. J Neuroimmunol 2015;286:86–92.
Torres-Vega E, Durán-Moreno M, Sánchez Del Pino M, et al. Immunoproteomic studies
been licensed to EUROIMMUN AG for the development of on paediatric opsoclonus-myoclonus associated with neuroblastoma. J Neuroimmunol
assays for LGI1 and other VGKC-complex antibodies. P.W. has 2016;297:98–102.
17. Irani SR, Alexander S, Waters P, et al. Antibodies to Kv1 potassium channel-complex
received speaker honoraria from Biogen Idec and EURO- proteins leucine-rich, glioma inactivated 1 protein and contactin-associated protein-2
IMMUN AG, and travel grants from the Guthy-Jackson in limbic encephalitis, Morvan’s syndrome and acquired neuromyotonia. Brain 2010;
133:2734–2748.
Charitable Foundation. I. Welding, S. Thomsen, S. Zuberi, 18. Matsuda K, Yuzaki M. Cbln1 and the δ2 glutamate receptor: an orphan ligand and an
R. Fischer, A. Radu Aricescu, M. Pike, R. Dale, and B. Kessler orphan receptor find their partners. Cerebellum 2012;11:78–84.
19. Vizcaı́no JA, Csordas A, del-Toro N, et al. 2016 update of the PRIDE database and
report no disclosures relevant to the manuscript. A. Vincent related tools. Nucleic Acids Res 2016;44:D447–D456.
is a coapplicant and receives royalties on patent application 20. Deutsch EW, Csordas A, Sun Z, et al. The ProteomeXchange Consortium in 2017:
supporting the cultural change in proteomics public data deposition. Nucleic Acids
WO/2010/046716 titled “Neurological Autoimmune Dis- Res 2017;54:D1100–D1106.
orders.” The patent has been licensed to EUROIMMUN AG 21. Sengupta P. The laboratory rat: relating its age with human’s. Int J Prev Med 2013;4:
624–630.
for the development of assays for LGI1 and other VGKC- 22. Hepp R, Hay YA, Aguado C, et al. Glutamate receptors of the delta family are widely
complex antibodies. M. Lim has received research grants from expressed in the adult brain. Brain Struct Funct 2015;220:2797–2815.

Neurology.org/N Neurology | Volume , Number  | Month 0, 2018 e9

 23. Hills LB, Masri A, Konno K, et al. Deletions in GRID2 lead to a recessive syndrome of 28. Shimokaze T, Kato M, Yoshimura Y, et al. A case of acute cerebellitis accompa-
cerebellar ataxia and tonic upgaze in humans. Neurology 2013;81:1378–1386. nied by autoantibodies against glutamate receptor delta2. Brain Dev 2007;29:
24. Utine GE, Haliloğlu G, Salanci B, et al. A homozygous deletion in GRID2 causes 224–226.
a human phenotype with cerebellar ataxia and atrophy. J Child Neurol 2013;28: 29. Kubota M, Tkahashi Y. Steroid-responsive chronic cerebellitis with positive glutamate
926–932. receptor delta 2 antibody. J Child Neurol 2008;23:228–230.
25. Optican LM, Pretegiani E. A GABAergic dysfunction in the olivary-cerebellar- 30. Hosaka T, Nakamagoe K, Takahashi Y, Mamada N, Tamaoka A. Opsoclonus
brainstem network may cause eye oscillations and body tremor. II: model simulations associated with autoantibodies to glutamate receptors δ2. Neurol Sci 2015;36:
of saccadic eye oscillations. Front Neurol 2017;8:372. 1741–1742.
26. Yoshida T, Katoh A, Ohtsuki G, et al. Oscillating Purkinje neuron activity causing 31. Miske R, Hahn S, Rosenkranz T, et al. Autoantibodies against glutamate receptor δ2
involuntary eye movement in a mutant mouse deficient in the glutamate receptor after allogenic stem cell transplantation. Neurol Neuroimmunol Neuroinflamm 2016:
delta2 subunit. J Neurosci 2004;24:2440–2448. 3:e255.
27. Hirai H, Pang Z, Bao D, et al. Cbln1 is essential for synaptic integrity and plasticity in 32. Amini A, Lang B, Heaney D, et al. Multiple sequential antibody-associated syndromes
the cerebellum. Nat Neurosci 2005;8:1534–1541. with a recurrent mutated neuroblastoma. Neurology 2016;87:634–636.

e10 Neurology | Volume , Number  | Month 0, 2018 Neurology.org/N

 Glutamate receptor δ2 serum antibodies in pediatric opsoclonus myoclonus ataxia
syndrome
Georgina Berridge, David A. Menassa, Teresa Moloney, et al.
Neurology published online July 25, 2018
DOI 10.1212/WNL.0000000000006035

This information is current as of July 25, 2018

Updated Information & including high resolution figures, can be found at:
Services http://n.neurology.org/content/early/2018/07/25/WNL.0000000000006
035.full
Subspecialty Collections This article, along with others on similar topics, appears in the
following collection(s):
Autoimmune diseases
http://n.neurology.org/cgi/collection/autoimmune_diseases
Permissions & Licensing Information about reproducing this article in parts (figures,tables) or in
its entirety can be found online at:
http://www.neurology.org/about/about_the_journal#permissions
Reprints Information about ordering reprints can be found online:
http://n.neurology.org/subscribers/advertise

Neurology ® is the official journal of the American Academy of Neurology. Published continuously since
1951, it is now a weekly with 48 issues per year. Copyright Copyright © 2018 The Author(s). Published by
Wolters Kluwer Health, Inc. on behalf of the American Academy of Neurology.. All rights reserved. Print
ISSN: 0028-3878. Online ISSN: 1526-632X.