Pedigree of the episodic ataxia family with the EAAT1 186S mutation. Episodic ataxia is indicated by a filled black square or circle for males or females, respectively. C186S indicates heterozygosity for the mutation. WT indicates homozygosity for the wild-type allele.
EAAT1 C186S mutation. A, Schematic representation of the EAAT1 protein and the location of the mutated Cys186 amino acid in transmembrane segment 4b (indicated by a black dot) (the structure is adapted from Yernool et al14). B, Conservation of the mutated residue Cys186 highlighted in gray. The protein sequences were obtained from GenBank (Homo sapiens, NP_004163; Bos taurus, NM_0010803750; Mus musculus, NP_683740; Rattus norvegicus, NP_062098; salamander, O57321Danio rerio, NP_997805Drosophila melanogaster, NP_477428; human EAAT2, AY066021; human EAAT3, NP_004161; human EAAT4, NP_005062; human EAAT5, NP_006662). C, Glutamate uptake assay in COS7 cells expressing mutant EAAT1 -186S (mean [SEM], 88.2 [5.5]) or wild-type EAAT1 -186C (mean [SEM], 107.8 [6.9]). The results are the mean (SEM) of the 4 experiments, each in triplicate. The values are picomoles of glutamate transported per milligram of protein per minute of incubation. Asterisk indicates significant reduction of glutamate uptake compared with wild type (P = 0.029). Error bars indicate SEM. HP indicates helical hairpin.
de Vries B, Mamsa H, Stam AH, Wan J, Bakker SLM, Vanmolkot KRJ, Haan J, Terwindt GM, Boon EMJ, Howard BD, Frants RR, Baloh RW, Ferrari MD, Jen JC, van den Maagdenberg AMJM. Episodic Ataxia Associated With EAAT1 Mutation C186S Affecting Glutamate Reuptake. Arch Neurol. 2009;66(1):97-101. doi:10.1001/archneurol.2008.535
Copyright 2009 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2009
Episodic ataxia (EA) is variably associated with additional neurologic symptoms. At least 4 genes have been implicated. Recently, a mutation in the SLC1A3 gene encoding the glutamate transporter EAAT1 was identified in a patient with severe episodic and progressive ataxia, seizures, alternating hemiplegia, and migraine headache. The mutant EAAT1 showed severely reduced uptake of glutamate. The syndrome was designated EA6 and shares overlapping clinical features with EA2, which is caused by mutations in CACNA1A.
To test the role of the SLC1A3 gene in EA.
Genetic and functional studies. We analyzed the coding region of the SLC1A3 gene by direct sequencing.
DNA samples from 20 patients with EA (with or without interictal nystagmus) negative for CACNA1A mutations were analyzed.
Main Outcome Measures
We identified 1 novel EAAT1 mutation in a family with EA and studied the functional consequences of this mutation using glutamate uptake assay.
We identified a missense C186S mutation that segregated with EA in 3 family members. The mutant EAAT1 showed a modest but significant reduction of glutamate uptake.
We broadened the clinical spectrum associated with SLC1A3 mutations to include milder manifestations of EA without seizures or alternating hemiplegia. The severity of EA6 symptoms appears to be correlated with the extent of glutamate transporter dysfunction.
Episodic ataxias (EAs) are rare genetic disorders characterized by recurrent episodes of cerebellar ataxia variably associated with additional neurologic features. Different subtypes of EA are defined on the basis of genetic loci and clinical manifestations.1
The most common and best characterized subtypes of EA are EA1 and EA2. EA1 is caused by missense mutations in the KCNA1 gene encoding a subunit of neuronal Kv1.1 K+ channels.2 EA1 usually presents with short-lasting attacks that often are triggered by exertion, stress, or startle. Patients show persistent interictal motor unit activity (myokymia). EA2 is caused by mutations in the CACNA1A gene encoding the pore-forming subunit of neuronal Cav2.1 Ca2+ channels.3 Mostly, nonsense, frameshift, splice site, and missense mutations have been described, resulting in either a complete loss4 or partial impairment5,6 of Cav2.1 channel function. The episodes in EA2 last longer than in EA1, up to several hours,7 and are often associated with vertigo and migrainous headache and can be triggered by exercise, fatigue, and stress.8 Acetazolamide may prevent attacks.9 Between attacks, nystagmus usually occurs. Many patients have interictal ataxia in addition to the attacks. The EA3, EA4, and EA5 subtypes are rarer and less well-defined disorders compared with EA1 and EA2.1
The EA6 subtype was identified in a 10-year-old patient with a severe phenotype of episodic and progressive ataxia, seizures, alternating hemiplegia, and migraine headache.10 A heterozygous de novo P290R missense mutation was identified in the SLC1A3 gene by use of a candidate gene approach. SLC1A3 encodes the glial excitatory amino acid transporter EAAT1, which is involved in glutamate removal from the synaptic cleft.11,12 Functional analysis of the mutant EAAT1 protein showed marked reduction of glutamate uptake in vitro.10
In the present study, we performed a mutation analysis of the SLC1A3 gene (OM/M 600111) in 20 patients with EA2-like symptoms but without CACNA1A mutations. In 1 family, we found an EAAT1 mutation that segregated with the disease in 3 patients. Functional studies revealed a moderate impairment of glutamate reuptake.
We investigated 20 patients who were referred for molecular confirmation of EA2 in whom no mutations were found in the CACNA1A gene. These patients showed typical EA2-like symptoms, including interictal nystagmus but no myokymia, attacks of mild ataxia with a duration of several hours, and a positive response to acetazolamide. Except for 2 patients from the United States, all patients came from Europe, mostly the Netherlands. Family members of the proband with the SLC1A3 mutation (Figure 1) underwent neurologic examination by experienced neurologists (S.L.M.B. and A.H.S). All patients gave informed consent, and the study was approved by the local review board.
Genomic DNA was isolated from peripheral leukocytes using a standard salting out extraction method.13 All exons and flanking intronic regions of the SLC1A3 gene were amplified by polymerase chain reaction (PCR), using genomic DNA as a template. Direct sequencing was performed by cycle sequencing (Prism Big Dye Terminators Cycle Sequencing kit; Applied Biosystems, Foster City, California) using the dideoxy termination method and an ABI3700 automated sequencer (Applied Biosystems). Two hundred healthy controls were screened for the mutation by PCR analysis of exon 5 and subsequent restriction digestion of PCR products with restriction enzyme AluI.
Functional studies10 on glutamate uptake of wild-type and mutant EAAT1 were performed as described previously. In brief, full-length wild-type complementary DNA (EAAT1-WT) was cloned into a mammalian expression vector pcDNA3.1 (Invitrogen; Carlsbad, California). The mutant construct (EAAT1 186S) was generated by performing site-directed mutagenesis (QuikChange; Stratagene; La Jolla, California). For functional analyses of the SLC1A3 C186S mutation, 2 μg of wild-type (EAAT1-WT) or mutant (EAAT1-186S) EAAT1 complementary DNA constructs were transfected into COS7 cells. One day after transfection, the cells were dissociated and plated onto 60-mm-diameter tissue culture dishes. The cells were incubated with 1.5 mL of 1μM L-glutamic acid containing 1 μCi/mL of L-[3,4-3H]-glutamic acid for 2 minutes at room temperature. A total of 4 independent and masked experiments were performed, each in triplicate.
Mutation analysis of the SLC1A3 gene in 20 patients revealed in 1 patient a heterozygous c.556 T>A substitution (SLC1A3 reference sequence; GenBank NM 004172) that changed a cysteine to a serine at position 186 (C186S) of the EAAT1 protein (Figure 2A and 2B). The mutation was absent in 200 Dutch control individuals. C186S was identified in the proband (III-3), clinically affected family members II-3 and III-4, and 1 asymptomatic family member (III-2) (Figure 1).
Clinical information of the affected family members is summarized in the Table. The proband (III-3) is a 35-year-old man who has had episodes of ataxia since early childhood. Attacks gradually changed over time. Initially, vertigo, nausea, and vomiting were the most bothersome symptoms. Later in life, truncal and gait ataxia during the attacks became more prominent. Attacks are often associated with nausea, vomiting, photophobia, phonophobia, vertigo, diplopia, slurred speech, and blurred vision. No headache was reported. Typically, attacks were provoked by emotional stress, fatigue, or consuming alcohol or caffeine. Attack duration was usually between 2 and 3 hours. Currently, his average attack frequency is once a month. Interictal neurologic examination revealed a horizontal gaze-evoked nystagmus without gait or truncal ataxia. Interictal electroencephalographic recording revealed no epileptic activity, and magnetic resonance imaging revealed no abnormalities (data not shown).
His mother (II-3) and sister (III-4) were also diagnosed as having EAs. The 56-year-old mother (II-3) has had episodes of ataxia similar to those of the proband since elementary school. Her attacks are also associated with vertigo, nausea, vomiting, photophobia, phonophobia, and slurred speech. The attacks were not associated with headache. She now has approximately 10 attacks per year, which may last for several hours and can be triggered by stress. The 28-year-old sister (III-4) has had episodes of ataxia since the age of 14 years. Associated symptoms include vertigo, nausea, vomiting, and mild photophobia. Sometimes, the day after an attack, she experiences bilateral headache not associated with nausea, vomiting, phonophobia, or photophobia. Reported triggers are exercise, fatigue, and stress. Currently, she has on average 6 attacks a year. Typically, attacks last several hours. Acetazolamide significantly reduced the frequency of attacks in all 3 affected family members.
His 40-year-old cousin (III-2) is an asymptomatic carrier of the C186S EAAT1 mutation. He experienced 4 attacks of migraine without aura and has tension-type headache, but does not exhibit signs or symptoms related to ataxia. Individuals I-1, I-2, and II-2 were considered healthy based on limited heteroanamnestic information. His grandfather had died at the age of 98 years. His grandmother had complained about dizziness, but no neurologic examination was performed during her lifetime. No relevant clinical information is available for individual II-2, who died of an unrelated cause. Non–mutation carrier III-1 is asymptomatic.
To investigate the functional consequences of the EAAT1 C186S mutation, radioactive glutamate uptake assays were performed in COS7 cells. The low level of endogenous glutamate uptake activity has long established the COS7 cells as being well suited for functional studies of glutamate transporters.15 We measured glutamate uptake in COS7 cells transfected with the wild-type (EAAT1-186C) or the mutant construct (EAAT1-186S). An 18% reduction in glutamate uptake was observed in cells expressing the mutant (mean [SEM], 88.2 [5.5]) compared with the wild-type (mean [SEM], 107.8 [6.9]) EAAT1, measured in picomoles per milligram of total protein per minute of incubation (P = 0.029; Figure 2C).
We scanned the SLC1A3 gene for mutations in 20 patients with EA2-like symptoms without CACNA1A mutations because of overlapping clinical features between EA2 and EA6. We found a novel nucleotide change c.556T>A in the SLC1A3 gene, resulting in EAAT1 mutation C186S, in a family with EA and interictal nystagmus but without migraine, seizures, cerebellar atrophy, or alternating hemiplegia.
Our genetic and functional data suggest that mutation C186S is pathogenic. First, the mutation C186S segregated with all 3 symptomatic family members but was not identified in a large panel of controls. The asymptomatic mutation carrier (III-2) had migraine without aura, but given the relatively high prevalence of migraine it is unlikely that these attacks are caused by the EAAT1 mutation. Therefore, he likely represents a nonpenetrant case of EA. Second, Cys186 is highly conserved among species (Figure 2B). Our functional studies revealed a reduced glutamate reuptake for the mutant EAAT1 (Figure 2C). Cys186 resides in transmembrane segment 4B (Figure 2A) on the outer perimeter of the human EAAT1 transporter protein that is implicated in intersubunit contact.15 The 4B-4C loop was recently shown to undergo substrate-dependent conformational changes and has been hypothesized to be important in stabilizing the trimeric structure of the transporter and coordinating the cooperativity for sodium binding.16
Clinical severity of EA6 appears to be well correlated with glutamate reuptake capability of mutant EAAT1. The P290R mutation leads to a complete loss of glutamate reuptake and is associated with a severe EA phenotype with months-long attacks, seizures, and alternating hemiplegia.10 In contrast, the C186S mutation has a mild effect on glutamate reuptake and is correlated with a milder EA phenotype. Although it is hard to predict from cellular studies how a mild increase in extracellular glutamate will affect cerebellar functioning in patients, it is well known that ion and neurotransmitter pathways are complex and tightly regulated. Subtle changes in these pathways have been associated with clinical manifestations.17,18
Since we found a mutation in only 1 of 20 patients with CACNA1A-negative EA2-like symptoms, other genes must be involved. Likely candidate genes are components of ion and neurotransmitter pathways involved in the regulation of cerebellar neuronal excitability.
Correspondence: Arn M. J. M. van den Maagdenberg, PhD, Department of Human Genetics, Leiden University Medical Centre, 2300 RC Leiden, the Netherlands (firstname.lastname@example.org).
Accepted for Publication: June 19, 2008.
Author Contributions: Ms de Vries, Ms Mamsa, and Dr Stam contributed equally to this study and all are considered first authors. Study concept and design: de Vries, Mamsa, Wan, Bakker, Haan, Frants, Baloh, Ferrari, Jen, and van den Maagdenberg. Acquisition of data: de Vries, Mamsa, Stam, Boon, and Howard. Analysis and interpretation of data: de Vries, Mamsa, Wan, Vanmolkot, Haan, Terwindt, Baloh, Jen, and van den Maagdenberg. Drafting of the manuscript: de Vries, Mamsa, Haan, Howard, Baloh, Jen, and van den Maagdenberg. Critical revision of the manuscript for important intellectual content: Stam, Bakker, Vanmolkot, Terwindt, Boon, Frants, Baloh, Ferrari, Jen, and van den Maagdenberg. Statistical analysis: Mamsa and Jen. Obtained funding: Baloh, Jen, and van den Maagdenberg. Administrative, technical, and material support: de Vries, Mamsa, Stam, Wan, Bakker, Vanmolkot, Jen, and van den Maagdenberg. Study supervision: Haan, Terwindt, Howard, Frants, Baloh, Ferrari, Jen, and van den Maagdenberg.
Financial Disclosure: None reported.
Funding/Support: This work was supported by grants U54 NS059065 and P50 DC05224 from the National Institutes of Health (Dr Baloh), grants 903-52-291 (Drs Ferrari and Frants) and Vici 918.56.602 (Dr Ferrari) from the Netherlands Organization for Scientific Research, The Migraine Trust (Drs Ferrari and Frants), grant LSHM-CT-2004-504837 from the European Union “EUROHEAD” (Drs Ferrari, Frants, and van den Maagdenberg), and the Center for Medical System Biology in the framework of the Netherlands Genomics Initiative.