An SCN2A mutation in a family with infantile seizures from Madagascar reveals an increased subthreshold Na+ current

Authors

  • Stephan Lauxmann,

    1. Department of Neurology and Epileptology, Hertie Institute for Clinical Brain Research, University of Tuebingen, Tuebingen, Germany
    Search for more papers by this author
    • These authors contributed equally to the manuscript.

  • Nadia Boutry-Kryza,

    1. INSERM, U1028, CNRS, UMR5292, Lyon Neuroscience Research Center, TIGER Team, Lyon, France
    2. Department of Medical Genetics, Hospices Civils de Lyon, Lyon, France
    3. University Lyon 1, Villeurbanne, France
    Search for more papers by this author
    • These authors contributed equally to the manuscript.

  • Clotilde Rivier,

    1. Department of Pediatrics, Hopital Nord-Ouest, Villefranche sur Saône, France
    Search for more papers by this author
  • Stephan Mueller,

    1. Department of Neurology and Epileptology, Hertie Institute for Clinical Brain Research, University of Tuebingen, Tuebingen, Germany
    Search for more papers by this author
  • Ulrike B. S. Hedrich,

    1. Department of Neurology and Epileptology, Hertie Institute for Clinical Brain Research, University of Tuebingen, Tuebingen, Germany
    Search for more papers by this author
  • Snezana Maljevic,

    1. Department of Neurology and Epileptology, Hertie Institute for Clinical Brain Research, University of Tuebingen, Tuebingen, Germany
    Search for more papers by this author
  • Pierre Szepetowski,

    1. INSERM U901, Mediterranean Institute of Neurobiology (INMED), Marseille, France
    Search for more papers by this author
  • Holger Lerche,

    Corresponding author
    1. Department of Neurology and Epileptology, Hertie Institute for Clinical Brain Research, University of Tuebingen, Tuebingen, Germany
    • Address correspondence to Gaetan Lesca, Service de Génétique, Hôpital femme-Mère-Enfant, Groupement Hospitalier Est, 69677 Bron, France. E-mail: gaetan.lesca@chu-lyon.fr Or Holger Lerche, Abt. Neurologie mit Schwerpunkt Epileptologie, Hertie Institut für Klinische Hirnforschung, Universität Tübingen, Hoppe-Seyler-Str. 3, 72076 Tübingen, Germany. E-mail: holger.lerche@uni-tuebingen.de

    Search for more papers by this author
  • Gaetan Lesca

    Corresponding author
    1. INSERM, U1028, CNRS, UMR5292, Lyon Neuroscience Research Center, TIGER Team, Lyon, France
    2. Department of Medical Genetics, Hospices Civils de Lyon, Lyon, France
    3. University Lyon 1, Villeurbanne, France
    • Address correspondence to Gaetan Lesca, Service de Génétique, Hôpital femme-Mère-Enfant, Groupement Hospitalier Est, 69677 Bron, France. E-mail: gaetan.lesca@chu-lyon.fr Or Holger Lerche, Abt. Neurologie mit Schwerpunkt Epileptologie, Hertie Institut für Klinische Hirnforschung, Universität Tübingen, Hoppe-Seyler-Str. 3, 72076 Tübingen, Germany. E-mail: holger.lerche@uni-tuebingen.de

    Search for more papers by this author

Summary

Missense mutations in SCN2A, encoding the brain sodium channel NaV1.2, have been described in benign familial neonatal-infantile seizures (BFNIS), a self-limiting disorder, whereas several SCN2A de novo nonsense mutations have been found in patients with more severe phenotypes including epileptic encephalopathy. We report a family with BFNIS originating from Madagascar. Onset extended from 3 to 9 months of age. Interictal EEGs were normal. In two patients, ictal electroencephalography (EEG) studies showed partial seizure patterns with secondary generalization in one. Seizures remitted before 18 months of age, with or without medication. Intellectual development was normal. A novel missense mutation of SCN2A, c.4766A>G/p.Tyr1589Cys, was found in a highly conserved region of NaV1.2 (D4/S2-S3). Functional studies using heterologous expression in tsA201 cells and whole-cell patch clamping revealed a depolarizing shift of steady-state inactivation, increased persistent Na+ current, a slowing of fast inactivation and an acceleration of its recovery, thus a gain-of-function. Using an action potential waveform in a voltage-clamp experiment we indicated an increased inward Na+ current at subthreshold voltages, which can explain a neuronal hyperexcitability. Our results suggest that this mutation induces neuronal hyperexcitability, resulting in infantile epilepsy with favorable outcome.

Benign familial neonatal-infantile seizures (BFNIS; OMIM#607745), is a self-limiting autosomal dominant disorder characterized by clusters of partial or secondarily generalized seizures with onset within the first year of life, spontaneous remission with few recurrences later on, and a generally good neurologic outcome (Berkovic et al., 2004). BFNIS are caused by mutations in the SCN2A (OMIM*182390) gene encoding the voltage-gated Na+ channel NaV1.2 (Heron et al., 2002). Herein, we report a comprehensive clinical, genetic, and pathophysiologic study of a large family from Madagascar.

Patients and Methods

Clinical evaluation

The proband (III-12) was referred to the department of pediatrics of Villefranche-sur-Saone. The family originated from Madagascar. Clinical information was obtained from direct interviews and examinations of affected individuals and their relatives. Medical records were available for III-2, IV-3, and III-12, and electroencephalography (EEG) records for III-12 and IV-3. Written informed consent was obtained from all participants or their parents according to the French bioethics law.

Mutation screening

Coding exons and intronic boundaries of SCN2A (GenBank accession number [Hg19]: NM_021007.2) were sequenced according to standard protocols. Consequences of the mutation on protein function were evaluated using the Alamut-1.42 mutation prediction software (Interactive Biosoftware, Mont Saint-Aignan, France) taking into account the Grantham scale, SIFT (Sorting Intolerant From Tolerant), and Polyphen algorithms. dbSNP (http://www.ncbi.nlm.nih.gov/projects/SNP/), 1,000 genomes (http://browser.1000genomes.org/index.html), and exome variant server (EVS, http://evs.gs.washington.edu/EVS/) databases were used to check for the mutation in controls.

Mutagenesis, transfection, and electrophysiology

All experimental procedures have been described previously (Liao et al., 2010b). The c.4766A>G mutation was introduced into the cDNA of the adult NaV1.2 human splice variant. Wild-type or mutant Na+ channel α-subunits were transfected alone or together with hβ1- and hβ2-subunits in tsA201 cells via “TransIT-LT1” (Mirus, Madison, WI, U.S.A.) reagent and functionally characterized using whole-cell patch-clamping. Data are shown as means ± standard error of the mean (SEM); Student's t-test was applied for statistical analysis. For details see Data S1.

Results

Clinical data

The pedigree is shown in Fig. 1A, and all clinical data are summarized in Table 1. Eight patients experienced nonfebrile seizures with onset between 3 and 9 months of age and remission before 18 months, with or without antiepileptic therapy. Partial onset was documented in patient III-12 with ictal EEG recordings showing temporocentral spikes (Fig. S1). None of the patients experienced seizures later in life. They all had normal psychomotor and cognitive development with good school performance.

Table 1. Summary of the clinical data of the family members with infantile epilepsy
PatientsII-1II-3II-5III-2III-6III-7III-12IV-3
  1. G, generalized; P, partial; UK, unknown; ND, not done; CTS, centrotemporal spikes; CBZ, carbamazepine; CLZ, clonazepam; PHT, phenytoin; VPA, sodium valproate.

Age at onset (months)3<9<643.554.54
Seizure typeGGGGGGPG
Seizure durationShortUK<2 min2–4 min<2 min1 min<2 min15 min (first seizure)
Cluster (duration)UKUKYesYes (12 h)Yes (24 h)NoYes (2–4 days)Yes (1 day)
Number of clustersUK+++++++++3052
Age at offset (months)UK156187.5UK5.54.5
Interictal EEGNDNDNDNormalNormalNormalBilateral CTSRight CTS with occasional contralateral diffusion
Brain MRINDNDNDNDNormalNDNormalRight temporal lesion
Initial treatmentNoneNoneNoneVPAPHBNoneVPA, then + CLZ, then + PHTCLZ, then + PHT, then + VPA
Long-term treatment (duration)NoneNoneNoneVPA (7 months)VPA (1 year)NoneCBZ at 5 months of age due to recurrent seizures (2 years)VPA (1 year)
Figure 1.

Functional studies reveal pronounced gain-of-function changes for the p.Tyr1589Cys (p.Y1589C) mutation. (A) pedigree. Filled black symbols, patients with infantile seizures; centered black symbol, obligate carrier without any reported history of seizure; arrow, proband; wt, wild-type allele; m, mutated allele. Slashes across circles and squares indicate deaths. (B, C) Families of whole-cell Na+ currents recorded from tsA201 cells transfected with either WT (B) or Y1589C mutant (C) channels. Na+ currents were elicited by step depolarizations ranging from −105 to +67.5 mV from a holding potential of −140 mV. (D) Voltage dependence of steady-state Na+ channel activation and fast inactivation revealing an increase in window current (area under the overlap of both curves) for Y1589C channels compared to the WT. Lines represent fits of Boltzmann functions. (E) Time course of recovery from fast inactivation determined at −100 mV revealing an acceleration for Y1589C mutant channels. Lines represent fits of exponential functions. (F) Voltage dependence of the time constant of recovery from fast inactivation, τ rec, which was significantly different between WT and mutant channels. (G) Voltage dependence of the fast inactivation time constant, τh, for WT and Y1589C mutant channels revealing a slowing of fast inactivation for the mutant channels. (H) Voltage dependence of the persistent currents showing a largely increased persistent current for the mutation. Current amplitudes were recorded at the end of a 70 msec depolarization to 0 mV and are normalized to the peak amplitude (steady state current/initial peak current). (I) Top: action potential used as a voltage stimulus; Bottom: recorded current density (current size divided by cell capacity). Here we show the first sweeps representatively as mean of all recorded cells; without leak subtraction; red line: Y1589C (n = 14), black line: WT Nav1.2 (n = 13). With the solutions used, the reversal potential was at 41.8 mV (35.5 + 6.3 mV junction potential) explaining the large positive peak in the bottom graph. (J) Top: currents shown in I normalized to the maximum peak current; Bottom: subtraction of currents carried by WT from those by mutant channels. (K) Top: Negative peak currents of mutant channels were significantly delayed in the subthreshold range. Bottom: bar graph showing the significantly increased negative peak currents for mutant channels (red column bar for Y1589C and black column bar for WT Nav1.2). The values of the electrophysiologic results, numbers of experiments, and p-values are given in Table S1. All values are shown as means ± SEM. *p < 0.05; ***p < 0.001.

Mutation screening

A heterozygous c.4766A>G mutation was found in exon 26 of the SCN2A gene, predicting mutation p.Tyr1589Cys (or p.Y1589C) in the highly conserved D4/S2-S3 region of NaV1.2 (Fig. S2A). It was found in all eight affected individuals and in the obligate carrier II-6 (Fig. 1A), but absent in 200 chromosomes from French controls and has not been reported in dbSNP, 1,000 genomes, and EVS databases. In silico analysis showed a high degree of nucleotide conservation for Tyr1589 (Fig. S2B,C) and an important physical and chemical gap between the tyrosine and cysteine residues with a Grantham score of 194 (0–215). SIFT and Polyphen algorithms both predicted a deleterious functional effect.

Functional studies

Functional characterization of mutant and wild-type NaV1.2 α-subunits with or without both β-subunits revealed significant differences between wild-type and mutant channels, which were a bit more pronounced in the presence of β-subunits, indicating that the mutation affects intrinsic gating properties of the α-subunit that are further influenced by auxiliary subunits (Fig. 1B–K). The mutation caused an increased window current due to a depolarizing shift of the steady-state fast inactivation curve, whereas steady-state activation was not significantly changed (Fig. 1D, Table S1). Consistent with a destabilization of the fast inactivated state, the fast inactivation time constants were significantly prolonged for mutant channels and the persistent current was increased (Fig. 1H), a noninactivating current, which could be blocked by 10 μm tetrodotoxin. Furthermore, we found a two- to threefold acceleration of the recovery from fast inactivation. Such alterations indicate a gain of function and predict an increase in neuronal excitability via a membrane depolarization due to an increased window and persistent Na+ current and a shortening of the refractory period after an action potential (Fig. 1E,F). In addition to these gain-of-function alterations, mutant channels exhibited a significantly lower peak current density compared to wild-type, which, however, only reached statistical significance in the presence of β-subunits (Table S1).

To explore Na+ currents in response to more physiologic stimuli, we recorded action potentials (APs) from pyramidal cells in acute mouse brain slices and used those as stimuli for voltage clamp experiments in tsA201 cells expressing NaV1.2 channels (Fig. 1I). In comparison to the wild-type, mutant channels showed an increase of Na+ currents in the slowly depolarizing subthreshold phase before the fast peak (Fig. 1I–K), consistent with the increased window and persistent currents. Although the nonphysiologic reversal potential complicated this analysis, the Na+ current during the main AP peak was smaller for mutant channels, corresponding to the observed decrease in current density, but not changed in shape (Fig. 1J,K). In addition, pulse trains at 80 Hz and a train of APs revealed a trend to an increased availability of mutant channels (Figs. S3 and S4).

Discussion

We present herein the first BFNIS family of non-European origin exhibiting a homogeneous age at onset, partial-onset, short-lasting seizures mostly occurring in clusters and stereotyped for each individual, but different between family members, and a favorable outcome despite discontinuation of treatment. Patient IV-3 had a temporal lesion and a particularly long, 15-min lasting seizure, so that both the SCN2A mutation and the lesion could have contributed to epileptogenesis and explain an unusual duration of seizures.

The p.Tyr1589Cys mutation affects a highly conserved amino acid in the D4/S2-S3 loop of NaV1.2. It segregated with the phenotype in the family, was absent in French controls, and although controls from Madagascar were not available, the mutation has not been reported in public databases. Because this tyrosine site has not been described as a predicted phosphorylation motif (according to PhosphoSitePlus, http://www.phosphosite.org/), the observed altered ion channel behavior is likely due to the substituted amino acid itself and not the consequence of altered phosphorylation. Up to now, 12 different SCN2A mutations have been reported in families with BFNIS, including this one (Heron et al., 2002; Berkovic et al., 2004; Striano et al., 2006; Herlenius et al., 2007; Liao et al., 2010b). All mutations associated with BFNIS are missense involving highly conserved amino acids. An additional SCN2A missense mutation was described in a sporadic case with severe and difficult to treat neonatal infantile epilepsy resolving only at month 13 of age and with later onset episodes of ataxia, myoclonus, and pain (Liao et al., 2010a).

So far, functional testing has been performed for six BFNIS mutations. Most of them exhibited a gain of function, but some also a reduced current density, which was discussed as loss of function (Scalmani et al., 2006; Xu et al., 2007; Misra et al., 2008; Liao et al., 2010b). The most severe phenotype showed the most prominent gain of function (Liao et al., 2010a). Our findings revealed mainly gain-of-function effects, increasing the sodium inward current in the critical subthreshold range of an action potential as revealed by an AP clamp experiment using physiologic stimuli. We consider this as the pathophysiologic hallmark for this mutation that can best explain the occurrence of epileptic seizures.

The spontaneous remission of seizures, also observed in the family presented here, could be due to developmental changes of sodium channel expression at axon initial segments (AIS) of principal excitatory neurons. As we previously described, NaV1.2 channels are transiently highly expressed at the AIS and NaV1.6 channels, which are not mutated, are up-regulated later during development and represent the predominant channel in adult ages (Liao et al., 2010b).

For three other SCN2A mutations causing severe epilepsy and cognitive impairment, complete or severe loss-of-function mechanisms have been described (Kamiya et al., 2004; Ogiwara et al., 2009), which may induce complex network changes in contrast to the small gain-of-function changes we and others observed for BFNIS mutations. This might explain the large differences in the clinical presentation of SCN2A-associated epileptic disorders.

Acknowledgments

We thank all patients and their relatives for participating in this study and Raphaelle Lamy for technical assistance. This study was supported by grants from the German Research Foundation and the European Science Foundation (DFG, Le1030/10-2 and ESF (EuroEPINOMICS)/DFG Le1030/11-1), and the Federal Ministry for Education and Research (BMBF, NGFNplus/EMINet: 01GS08123) (all to HL).

Disclosures

HL serves on scientific advisory boards for the following companies: Eisai, GSK (GlaxoSmithKline), Pfizer, UCB (Union Chimique Belge), Valeant; receives industry-funded travel costs: GSK, Pfizer, UCB; receives honoraria for speaking engagement or educational activities from Desitin, Eisai, GSK, Pfizer, UCB; receives research support from Sanofi-Aventis, UCB. The other authors have no conflict of interest to disclose. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

Ancillary