seronegative patients) had a higher than the mean con-
Half of the seronegative patients with NMO had a
control group. The striking difference in the incidence of
anti-AQP4 T cell responses between the NMO and MS
groupsmayprovideamoresensitivetool for distinction
between these 2 demyelinating diseases.
Discussion. Despite the association of anti-AQP4
antibodies with NMO, there is no convincing evidence
that these antibodies are solely responsible for the
pathogenesis of the disease. A possible role of T cells in
NMO immunopathogenesis has been suggested lately.
Warabi et al.3reported a generalized upregulation of
T-cell activity in patients with NMO and Matsuya et
al.4described an increase in CD69-activated T cells tar-
geting distinct AQP4 determinants in peripheral blood
port for the role of AQP4-reactive T cells in NMO
pathogenesis comes from animal studies, in which it
was shown that an NMO-like disease after transfer of
AQP4 antibodies is greatly and crucially facilitated by
activated T cells.5–7
Two possible scenarios may explain the role of T
cells in NMO. 1) In genetically susceptible individuals,
anti-AQP4 antibodies may develop and cause astrocyte
destruction. Subsequently, AQP4 is released and di-
induced by T cells could be the initial event, followed
by the involvement of humoral autoimmune mecha-
nisms and the production of anti-AQP4 antibodies
(representing thus an epiphenomenon) that may target
the astrocytes and accelerate tissue loss.
icant involvement of T cells in human NMO. These
findings may contribute to a better understanding of
the diverse immune mechanisms involved in NMO
sponses may provide an additional and possibly more
sensitive tool for distinguishing between MS and
NMO, especially in patients who are seronegative for
From the Department of Neurology, MS Center, and Laboratory of
Neuroimmunology and the Agnes-Ginges Center for Neurogenetics,
Hadassah-Hebrew University Medical Center, Ein-Karem, Jerusa-
Author contributions: Adi Vaknin-Dembinsky: drafting/revising the
manuscript, study concept or design, acquisition of data, statistical
analysis, study supervision. Livnat Brill: drafting/revising the man-
uscript, study concept or design, analysis or interpretation of data,
acquisition of data. Ibrahim Kassis: drafting/revising the manu-
script, analysis or interpretation of data, contribution of vital re-
agents/tools/patients, acquisition of data, statistical analysis.
Panayiota Petrou: drafting/revising the manuscript. Haim Ovadia:
analysis or interpretation of data, acquisition of data. Tamir Ben-
Hur: drafting/revising the manuscript; Oded Abramsky, drafting/
revising the manuscript. Dimitrios Karussis: drafting/revising the
manuscript, study concept or design, analysis or interpretation of
data, study supervision.
Disclosure: The authors report no disclosures relevant to the manu-
script. Go to Neurology.org for full disclosures.
Received July 27, 2011. Accepted in final form April 21, 2012.
Correspondence & reprint requests to Dr. Adi Vaknin-Dembinsky:
Copyright © 2012 by AAN Enterprises, Inc.
1.Wingerchuk DM, Lennon VA, Lucchinetti CF, Pittock
SJ, Weinshenker BG. The spectrum of neuromyelitis op-
tica. Lancet Neurol 2007;6:805–815.
Wingerchuk DM, Lennon VA, Pittock SJ, Lucchinetti
CF, Weinshenker BG. Revised diagnostic criteria for neu-
romyelitis optica. Neurology 2006;66:1485–1489.
Warabi Y, Yagi K, Hayashi H, Matsumoto Y. Character-
ization of the T cell receptor repertoire in the Japanese
neuromyelitis optica: T cell activity is up-regulated com-
pared to multiple sclerosis. J Neurol Sci 2006;249:145–
Matsuya N, Komori M, Nomura K, et al. Increased T-cell
immunity against aquaporin-4 and proteolipid protein in
neuromyelitis optica. Int Immunol 2011;23:565–573.
Bradl M, Misu T, Takahashi T, et al. Neuromyelitis op-
tica: pathogenicity of patient immunoglobulin in vivo.
Ann Neurol 2009;66:630–643.
Bennett JL, Lam C, Kalluri SR, et al. Intrathecal patho-
genic anti-aquaporin-4 antibodies in early neuromyelitis
optica. Ann Neurol 2009;66:617–629.
Nelson PA, Khodadoust M, Prodhomme T, et al. Immu-
nodominant T cell determinants of aquaporin-4, the au-
toantigen associated with neuromyelitis optica. PLoS One
MILD PAROXYSMAL KINESIGENIC
DYSKINESIA CAUSED BY PRRT2 MISSENSE
MUTATION WITH REDUCED PENETRANCE
Paroxysmal kinesigenic dyskinesia (PKD) is an uncom-
mon disorder characterized by brief episodes of invol-
untary dystonia or choreoathetosis triggered by sudden
voluntary movement.1Recently, several groups identi-
transmembrane protein 2 (PRRT2) gene.2–7We report
a missense c.913G?A (p.Gly305Arg) change in
PRRT2 in 4 siblings with PKD, 2 with infrequent
symptoms. The association of a milder phenotype
and reduced penetrance with this missense mutation
Case Reports. Four Caucasian siblings with PKD8
as well as an unaffected sibling and both unaffected
parents were genotyped (figure). Institutional ap-
proval and informed consent were obtained.
Jennifer Friedman, MD
Jesus Olvera, MS
Jennifer L. Silhavy, MS
Stacey B. Gabriel, PhD
Joseph G. Gleeson, MD
Neurology 79August 28, 2012
II-1 is a 24 year-old man with brief episodes of
paroxysmal dyskinesia triggered by sudden move-
ment, beginning at age 16. Typically spells were rare,
but during stressful times occurred up to 10 times
per day and improved dramatically with carbamaz-
epine up to 400 mg. daily. Currently, spells are
nearly resolved with medication taken only at times
of increased stress.
II-3, II-4, and II-5 are 19-year-old fraternal trip-
lets. Brief dystonic, tensing, or choreoathetotic
movements began between ages 12 and 14. Spells
occurred up to 3 times per day (II-4) or, rarely, pri-
marily in association with athletic competitions (II-3
and II-5). Spells were triggered by initiation of move-
ment with greater likelihood of events at times of
increased stress or excitement (during athletic com-
petitions or walking to the front of the class to pres-
ent). Caffeine exacerbated spells in II-3 and II-5. II-3
also noted eye-blinking tics. II-5 had a history of
congenital torticollis attributed to intrauterine po-
sitioning. His examination showed right-sided
shoulder depression and hypertrophy of the para-
spinal muscles without scoliosis. Results of exami-
nations of others were normal. In all 3, there was
resolution of spells with low-dose carbamazepine
(100 mg daily) taken by II-3 and II-5 only rarely
Missense change c.913G?A (p.Gly305Arg) in
the PRRT2 gene was identified in II-1 and II-4 by
whole exome sequencing. Sanger sequencing of the
remainder identified the c.913G?A change in II-3
and II-5 and in the unaffected mother (figure).
Discussion. We report a missense change in the
PRRT2 gene associated with familial PKD with mild
phenotype and reduced penetrance. Although the
pathogenicity of this alteration cannot be proven, it
changes a highly conserved amino acid residue (fig-
ure),9was predicted by PolyPhen-2 to be deleterious,
and was not identified across 1,000 individuals in our
in-house exome database, in 1,092 individuals se-
quenced in the 1000 Genomes database,10or among
variants identified by Chen et al.2in 500 Han Chinese
been identified by Liu et al.5in a patient with sporadic
PKD, consistent with its pathogenicity.
Previous studies reported PKD attack frequency
of greater than 20 per day in 31% of patients and at
least once per day in 72%.8Comparatively, the phe-
notype in this family is mild with 2 siblings with
infrequent symptoms requiring carbamazepine only
rarely. In addition, the mother is a nonsymptomatic
carrier. Although the mutation identified in this sin-
gle family is insufficient to establish a clear genotype-
phenotype correlation, it is possible that the missense
mutation leads to partial loss of function and a
milder phenotype. Nevertheless, additional modifi-
ers are probably at play, because PRRT2 kindreds can
display phenotypic variability that may include in-
fantile convulsions.6Other neurologic symptoms
have also been reported in patients and kindreds with
(A) Family pedigree: filled symbols denote individuals with
paroxysmal kinesigenic dyskinesia and unfilled symbols in-
dicate unaffected individuals. DNA from all individuals
shown was sequenced. Asterisks indicate individuals with
the c.913G?A (p.Gly305Arg) change in the PRRT2 gene.
(B) Sequence conservation (highlighted in red) in the PRRT2
gene in the region of the changed amino acid across
multiple vertebrate species suggests that this region is
part of an important functional or structural motif. Adapted
from UCSC Genome Browser—Human, February 2009
(GRCh37/hg19 Assembly) (http://genome.ucsc.edu).9
Neurology 79August 28, 2012
PKD although no definitive association has been es-
tablished.8These symptoms include dystonia and
tics, which were also noted in our patients.
The PRRT2 protein contains 2 extracellular do-
mains, 2 transmembrane domains, and 1 cytoplas-
mic domain.2The majority of reported mutations to
date lead to premature transcription termination and
truncated protein products. In vitro, truncated prod-
ucts have been found to mis-localize from membrane
to cytoplasm2or, alternatively, to be poorly ex-
pressed, resulting in haploinsufficiency.7Reports of
patients with possible PKD with deletions encom-
passing the PRRT2 allele11,12support the idea that
haploinsufficiency can result in the PKD phenotype.
The c.913G?A mutation is a missense substitu-
tion, located in exon 3, that codes for a change in the
cytoplasmic region of the protein. Although we lack
functional data, it is likely that this mutation leads to
partial loss of function of the affected allele through
an as yet undefined mechanism. It has been proposed
that PKD is an ionic channelopathy and that PRRT2
may complex with or regulate key properties of ion
channels2or that that PKD may result from dysfunc-
tion of synaptic regulation due to interaction with
SNAP25.7However, the precise role PRRT2 muta-
tions play in PKD remains unknown.
Further study of this and other families will be
necessary to establish genotype-phenotype correla-
tions, to confirm the pathogenicity and mechanism
of action of the c.913G?A (p.Gly305Arg) change,
to identify the spectrum of genetic variations in
PRRT2 associated with PKD, and to determine the
possible association of PRRT2 mutations with other
movement disorders including dystonia and tics.
From the Department of Neurosciences and Pediatrics, Rady Chil-
dren’s Hospital (J.F., J.O., J.L.S., J.G.G.), University of California,
San Diego; Howard Hughes Medical Institute (J.O., J.L.S.,
J.G.G.), La Jolla, CA; and Broad Institute of MIT and Harvard
(S.B.G.), Cambridge, MA.
Author contributions: Jennifer Friedman designed and conceptual-
ized the study, analyzed and interpreted data. and drafted and re-
vised the manuscript. Jesus Olvera analyzed and interpreted data
and contributed to revision of the manuscript. Jennifer Silhavy ana-
lyzed and interpreted data.; Stacey B. Gabriel analyzed and inter-
preted data. Joseph Gleeson designed and conceptualized the study,
analyzed and interpreted data, and contributed to revision of the
to Eric Lander) for sequencing support and analysis.
Study funding: Supported by the National Institutes of Health
(R01NS048453 and R01NS052455 to J.G.G.), the Simons Foun-
dation Autism Research Initiative, and the Howard Hughes Medi-
cal Institute (J.G.G.).
Disclosure: J. Friedman has received consulting fees from Biomarin
Pharmaceuticals. J. Olevera, J. Silhavy, S. Gabriel, and J. Gleeson
report no disclosures. Go to Neurology.org for full disclosures.
Received December 9, 2011. Accepted in final form March 28, 2012.
Correspondence & reprint requests to Dr. Friedman: jrfriedman@
Copyright © 2012 by AAN Enterprises, Inc.
1.Spacey S, Adams P. Familial paroxysmal kinesigenic dyski-
nesia. In: Pagon RA, Bird TD, Dolan CR, Stephens K,
eds. GeneReviews. Seattle: University of Washington;
2005 (updated 2011).
Chen WJ, Lin Y, Xiong ZQ, et al. Exome sequencing
identifies truncating mutations in PRRT2 that cause par-
oxysmal kinesigenic dyskinesia. Nat Genet 2011;43:1252–
Wang JL, Cao L, Li XH, et al. Identification of PRRT2 as
the causative gene of paroxysmal kinesigenic dyskinesias.
Li J, Zhu X, Wang X, et al. Targeted genomic sequencing
identifies PRRT2 mutations as a cause of paroxysmal kine-
sigenic choreoathetosis. J Med Genet 2012;49:76–78.
Liu Q, Qi Z, Wan XH, et al. Mutations in PRRT2 result
in paroxysmal dyskinesias with marked variability in clini-
cal expression. J Med Genet 2012;49:79–82.
Heron SE, Grinton BE, Kivity S, et al. PRRT2 mutations
cause benign familial infantile epilepsy and infantile con-
vulsions with choreoathetosis syndrome. Am J Hum Genet
Lee H, Huang H, Bruneau N, et al. Mutations in the gene
PRRT2 cause paroxysmal kinesigenic dyskinesia with in-
fantile convulsions. Cell Reports 2012;1:2–12.
agnostic criteria. Neurology 2004;63:2280–2287.
Kent WJ, Sugnet CW, Furey TS, et al. The human genome
browser at UCSC. Genome Res 2002;12:996–1006.
1000 Genomes Project Consortium. A map of human ge-
nome variation from population-scale sequencing. Nature
Lipton J, Rivkin MJ. 16p11.2-related paroxysmal kinesi-
genic dyskinesia and dopa-responsive parkinsonism in a
child. Neurology 2009;73:479–480.
Dale RC, Grattan-Smith P, Fung VS, Peters GB. Infan-
tile convulsions and paroxysmal kinesigenic dyskinesia
with 16p11.2 microdeletion. Neurology 2011;77:1401–
Neurology 79August 28, 2012