Rasmussen encephalitis associated with SCN1A mutation

Authors


  • Iori Ohmori and Mamoru Ouchida contributed equally to this work.

Address correspondence to Iori Ohmori, Department of Cellular Physiology, Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama University, 5-1 Shikata-cho, 2-chome, Okayama 700-8558, Japan. E-mail: iori@md.okayama-u.ac.jp

Summary

Mutations in the SCN1A gene, encoding the neuronal voltage-gated sodium channel α1 subunit, cause SMEI, GEFS+, and related epileptic syndromes. We herein report the R1575C-SCN1A mutation identified in a patient with Rasmussen encephalitis. R1575C were constructed in a recombinant human SCN1A and then heterologously expressed in HEK293 cells along with the human β1 and β2 sodium channel accessory subunits. Whole-cell patch-clamp recording was used to define biophysical properties. The R1575C channels exhibited increased channel availability and an increased persistent sodium current in comparison to the wild-type. These defects of electrophysiological properties can result in neuronal hyperexitability. The seizure susceptibility allele may influence the pathogenesis of Rasmussen encephalitis in this case.

Rasmussen encephalitis (RE) is an autoimmune-mediated brain disorder associated with intractable focal epilepsy, slowly progressive neurological deterioration, and progressive unilateral hemispheric atrophy (Rasmussen et al., 1958). Serum samples of patients with RE contain antibodies to glutamate receptor GluR3, and immunization of animals with GluR3 induces a disorder resembling the human disease (Roger et al., 1994). Histopathological studies have shown an infiltration of T lymphocytes and microglia cells, astrocytosis, and neuronal loss. A viral infection, humoral autoimmunity and T-cell-mediated cytotoxicity have been postulated as the possible mechanisms of RE (Bien et al., 2005). Although the pathogenesis of RE has not yet been completely elucidated, it is widely accepted that environmental factors largely contribute to RE.

In contrast to RE, the inherited form of epilepsy is caused by genetic factors. Generalized epilepsy with febrile seizures plus (GEFS+) is a familial epilepsy syndrome characterized by heterogeneous phenotypes, including febrile seizures, febrile seizures plus, and mild to severe epilepsies (Scheffer & Berkovic, 1997). The SCN1A gene, which encodes the neuronal voltage-gated sodium channel α1 subunit, has been identified in patients with GEFS+, severe myoclonic epilepsy of infancy (SMEI, or Dravet syndrome) and related epileptic syndromes. Other undiscovered modifier genes and environmental factors may lead to the wide spectrum of SCN1A-associated epilepsies.

We herein report the SCN1A-R1575C mutation identified in a patient with RE. We found this mutation fortuitously in the process of the SCN1A mutational screening of 98 patients having various types of epileptic syndromes. The R1575C mutant channels transiently expressed in human embryonic kidney (HEK) 293 cells exhibited defects of the electrophysiological properties. Since de novo mutations of the SCN1A gene have been reported in patients with alleged vaccine encephalopathy (Berkovic et al., 2006), we should carefully assess this new aspect of genetic-environmental interaction in an autoimmune-mediated brain disorder.

Materials and Methods

Patient A

The patient was a 10-year-old girl and her representative clinical data are shown in Fig. 1. Regarding the family history, a febrile seizure occurred in her paternal aunt and one of her maternal cousins, respectively (Fig. 1A). She was born after an uneventful 39-week-long pregnancy with a weight of 3,422 g.

Figure 1.


Clinical data of patient A. (A) The pedigree of patient A. (B) Awake EEG record at 8 years 6 months of age. The background activity is slow in the left hemisphere with irregular spike-and-wave complexes. (C) MRI showed a left hemispheric mild cerebral atrophy with a gray matter T2/FLAR hyperintense signal at 8 years 10 months of age (C-1, T1-weighted image; C-2, T2-weighted image, C-3, FLAIR). (D) SPECT was unremarkable at 2 years 10 months of age (D-1) and showed hypoperfusion in the left parietotemporal lobes at 9 years 8 months of age (D-2). (E) A mutation analysis of the SCN1A gene. Patient A and her father had the R1575C mutation. Numbering of amino acids is given in GenBank accession number AB093548.

She had febrile seizures at 1 year 1 month of age and 1 year 7 months of age. In the initial prodromal phase, her development and EEG were both age appropriate. Afebrile generalized tonic–clonic seizures (GTCS) began to occur weekly at 2 years of age. Her mental development delayed thereafter. Spikes were found in the left frontal region on EEG and valproate was started. She had varicella at 2 years 5 months of age. She has suffered from myoclonic seizures since 2 years 6 months of age. Her EEG showed many spike-and-wave complexes. A battery of examinations, including magnetic resonance imaging (MRI), computed tomography (CT), single-photon emission computed tomography (SPECT) with 99mTc-ECD (ethyl cysteinate dimmer) (Fig. 1 D-1), CSF, and analyses of chromosome, urinary organic acids, and lysosomal enzymes, were all normal. The administration of clonazepam suppressed the myoclonic seizures and potassium bromide reduced the GTCS.

At 5 years of age, she had idiopathic thrombocytopenic purpura (ITP). She frequently had focal motor seizures with asymmetric tonic posture. She had a mild gait abnormality with less movement in the right arm than in the left at around 5 years of age.

At 8 years of age, right hemiparesis was evident. EEG showed asymmetric slowing in the left hemisphere (Fig. 1B) and a cerebral atrophy of the left hemisphere was found in MRI (Fig. 1C). She was suspected to have RE, so autoantibodies against GluR2 and GluR3 in the serum were analyzed using ELISA. Autoantibodies against GluR3 were detected. Her condition has since deteriorated and her IQ (Tanaka-Binet) was 29 at 9 years of age. SPECT showed hypoperfusion in the left temporoparietal lobes (Fig. 1D-2).

Molecular genetic analysis

SCN1A mutations were analyzed by the previously reported methods (Ohmori et al., 2002).

The study was approved by the Ethics Committee of Okayama University Graduate School of Medicine, Dentistry, and Pharmaceutical Sciences. Written informed consent was obtained from the parents.

Mutagenesis and heterologous expression of SCN1A

Mutagenesis of full-length human SCN1A cDNA (a generous gift from Dr. George, Vanderbilt University) was performed as described previously (Lossin et al., 2002). In all of the experiments, recombinant SCN1A was heterologously coexpressed with the human β1 and β2 accessory subunits in the HEK293 cells.

Electrophysiology and data analysis

Whole-cell voltage-clamp recordings were used to characterize the functional properties of the wild-type (WT)-SCN1A and R1575C mutant sodium channel, as described previously (Ohmori et al., 2006). Sodium channel currents were recorded at room temperature, 24–72 h after transfection. Patch pipettes were pulled (1.3–1.9 MΩ) by multistage P-97 Flaming–Brown micropipette puller. The pipette solution consisted of (in mM): 110 CsF, 10 NaF, 20 CsCl, 2 ethylene glycol tetraacetic acid (EGTA), 10 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), with a pH of 7.35 and osmolarity of 310 mOsmol/kg. The bath solution consisted of (in mM): 145 NaCl, 4 KCl, 1.8 CaCl2, 1 MgCl2, 10 HEPES, with a pH of 7.35 and osmolarity of 310 mOsmol/kg. Specific voltage-clamp protocols assessing channel activation, steady-state fast inactivation, recovery from fast inactivation and persistent current were used as described previously (Ohmori et al., 2006). The results are presented as the mean ± SEM, and statistical comparisons were made in reference to WT-SCN1A by using the unpaired Student's t-test. The data analysis was performed using the Clampfit 8.2 (Axon Instruments, Union City, CA, U.S.A.) and OriginPro 7.5 (OriginLab, Northampton, MA, U.S.A.) software programs.

Results

SCN1A mutational analysis

The novel R1575C mutation was identified in patient A and her unaffected father (Fig. 1E). No other family members agreed to undergo any genetic tests. We did not find the R1575C mutation in the healthy individuals (n = 96).

Electrophysiological properties

We examined the R1575C mutation for functional defects by using a heterologous expression system. Fig. 2A illustrates the representative whole-cell currents recorded from the HEK293 cells expressing WT-SCN1A, or R1575C. Cells expressing the R1575C mutant exhibited a voltage-dependent inward current that resembled the cells transfected with WT channels. The inactivation proceeded in a rapid biexponential manner, with no apparent difference between WT and R1575C (Fig. 2B). However, the presence of a persistent sodium current in R1575C was observed. Fig. 2C illustrates the representative tetrodotoxin (TTX)-subtracted current traces for both WT-SCN1A and R1575C in response to a 100-ms voltage step to −10 mV from a holding potential of −120 mV. The cells expressing R1575C exhibited a significantly greater persistent current that was expressed as percentage of the peak current in comparison to WT-SCN1A (Fig. 2D).

Figure 2.


Sodium currents were recorded from HEK293 cells expressing either WT-SCN1A or R1575C. (A) Representative whole-cell sodium currents. The currents were activated by voltage steps to between −80 and +60 mV in 20 mV increments from a holding potential of −120 mV. (B) Voltage dependence of fast inactivation time constants for WT-SCN1A and R1575C. The fast and slow time constants were plotted versus voltage. The whole-cell current inactivation of R1575C is similar toWT-SCN1A. (C) Representative tetrodotoxin (TTX)-subtracted whole-cell sodium currents at −10mV recorded from HEK293 cells expressing WT-SCN1A or R1575C. Each current trace has been normalized to its peak sodium current. (D) Average persistent sodium current expressed as a percentage of the peak current for WT-SCN1A and R1575C. Persistent currents were evaluated by 100-ms depolarization steps to level between −30mV and +10mV from a holding potential of −120mV in the presence and absence of 10 μM TTX. R1575C exhibited a significantly increased persistent current ranged from −30 mV to 0 mV in comparison to WT-SCN1A (p < 0.05).

The current density, voltage-dependence channel activation, inactivation, and recovery from fast inactivation were examined (Fig. 3). R1575C showed no difference for the current density, voltage-dependence channel activation, and recovery from fast inactivation in comparison to WT. R1575C displayed the most significant defect in fast inactivation. R1575C exhibited a slight depolarizing shift in half-maximal steady-state inactivation (Fig. 3C).

Figure 3.


Electrophysical properties of R1575C. (A) Peak current density of whole-cell currents elicited by test pulses to various potentials and normalized to cell capacitance. Current density of R1575C is similar to WT-SCN1A. (B) Voltage dependence of channel activation measured during voltage steps to between −80 and +20 mV from a holding potential of −120 mV. Half-maximal activation occurred at −25.9 ± 1.5 mV with a slope factor of −6.2 ± 0.3 (n = 17) for WT-SCN1A and at −24.6 ± 1.1 mV with a slope factor of −6.2 ± 0.2 (n = 12) for R1575C. (C) Voltage dependence of fast inactivation measured at −10 mV in response to a 100-ms prepulse voltage step to between −140 and −10 mV from a holding potential of −120 mV. The membrane potentials for half-maximal inactivation and slope factors were as follows: WT, −62.1 ± 1.3 mV and −5.8 ± 0.2, n = 15; R1575C, −58.5 ± 1.0 mV (p= 0.043) and −6.2 ± 0.3, n = 12. (D) Time-dependent recovery from fast inactivation measured at −10 mV between 1 and 3,000 ms immediately after a 100-ms prepulse voltage step to −10mV from a holding potential of −120 mV. The time constants and fractional amplitudes were as follows: WT, τf= 1.8 ± 0.1 ms (82%± 1%), τs= 40.3 ± 5.4 ms (18%± 1%), n = 14; R1575C, τf= 1.8 ± 0.2 ms (79%± 2%), τs= 40.8 ± 4.0 ms (21%± 2%), n = 13. There was no statistically significant difference between WT and R1575C.

Discussion

The early clinical features of febrile and afebrile convulsions and family history are likely to be caused by the SCN1A mutation. The later clinical features also met the diagnostic criteria of RE (Bien et al., 2005). She had focal seizures, slowly progressive right hemiparesis, mental deterioration, right hemispheric slowing with epileptiform activity on EEG, and right hemispheric atrophy with a gray matter T2/FLAR hyperintense signal. There is the possibility that an association between the two conditions may be coincidental.

Regarding the etiology of RE, a direct viral infection, humoral autoimmunity and T-cell-mediated cytotoxicity have been postulated as the possible mechanisms (Bien et al., 2005). The association of RE plus a low-grade tumor, cortical dysplasia, tuberous sclerosis, or vascular abnormalities have also been described (Bien et al., 2005). More than 40% of patients with RE have prior infections, inflammatory episodes, or vaccinations before the onset of epilepsy (Takahashi et al., 2006). Patient A had varicella and ITP around the periods of worsening epilepsy. However, we could not identify the exact influence of infectious diseases and ITP because she originally had epileptic seizures and the course of the neurological deterioration was slowly progressive.

The R1575C-SCN1A mutation in patient A was carried by her unaffected father. However, it was not detected in healthy individuals (n = 96). An electrophysiological study showed that the R1575C channels had increased channel availability and increased persistent sodium current, which can lead to neuronal hyperexcitability. Increased persistent current is a common functional abnormality in SCN1A-associated epileptic syndromes (Lossin et al., 2002; Ohmori et al., 2006). Taken together, these findings therefore suggest that R1575C may be a seizure susceptibility allele. The seizure susceptibility allele may influence the pathogenesis of RE.

The spectrum of SCN1A-associated brain disorders has been expanding. The mildest form is simple febrile seizure (Mantegazza et al., 2005) and the most severe form is SMEI. SCN1A mutations have also been identified in familial hemiplegic migraines (Dichgans et al., 2005) and autism (Weiss et al., 2003). Moreover, differences in clinical severity among individuals carrying the same mutation are often observed in the GEFS+ family and the SMEI family. Factors that are believed to contribute to variable expressivity include intrinsic stochastic variability during the development of neuronal connectivity in the CNS, the accumulation of somatic mutations over a patient's lifetime, environmental insults, and differences in genetic backgrounds and modifier genes. Although the presence of such a susceptibility allele alone is insufficient to cause RE, it may exacerbate epileptic seizures when environmental factors are added. The combined effect of the SCN1A mutation, yet undiscovered susceptibility genes and the autoimmune-response caused by infection or vaccination should be carefully considered.

Acknowledgments

We thank Dr. K. Hongo (University of Toyama) for analyzing the anti-GluR3 antibodies. This work was supported by Grants-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology (No.15591154 to I.O. and No. 15591110 to M.O.). 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.

Disclosure of conflicts of interest: We have no conflicts of interest.

Ancillary