Seven novel SCN1A mutations in Chinese patients with severe myoclonic epilepsy of infancy

To the Editors:

Severe myoclonic epilepsy of infancy (SMEI) is an intractable epileptic encephalopathy, beginning with prolonged generalized or hemiclonic seizures associated with fever in the first year of life. Subsequently, it evolves to include multiple seizure types, such as myoclonic, absences, and partial seizures. The regression of psychomotor development appears after onset. Pyramidal signs and ataxia are observed in some patients. The seizures are refractory to antiepileptic drugs (AEDs) (ILAE, 1989). The SCN1A gene, encoding the neuronal voltage-gated sodium channel α1-subunit dominantly expressed in the central nervous system (CNS), is one of the most important pathogenic “epilepsy genes” identified to date. Almost 170 SCN1A mutations (including missense, nonsense, deletion, and splice site mutations) have been identified in SMEI and related epilepsies among different populations (Claes et al., 2001; Harkin et al., 2007). However, it is still unknown whether SCN1A is the major pathogenic gene in Chinese SMEI patients. We have now performed mutation analysis of the SCN1A gene in a population of Chinese SMEI patients.

Clinical analysis of the patients is presented in Table 1. All patients have different degrees of mental retardation and are intractable to AEDs. Because these patients have different types of myoclonic seizures, they have been defined as typical SMEI patients.

Table 1.  Clinical features of 12 SMEI patients

No.	Sex	Present age (years)	Family history	First seizure			Seizure type					Status epilepticus	Temperature sensitivity	EEG	MR	Ataxia	Pyramidal syndrome	MRI
				Age (months)	Precipitating factors	Seizure type	GTCS (GCS)	AS	MS	HS	PS
+, present; –, absent. Abbreviations: FS, Febrile seizures; GTCS, generalized tonic–clonic seizures; GCS, generalized clonic seizures; PS, partial seizures; HS, hemiclonic seizures; AS, absence seizures; MS, myoclonic seizures; MR, mental retardation; F, frontal; and sp-w, spike-wave complex; IPS, intermittent photic stimulation.
1	M	4	Maternal uncle: epilepsy	3	Hot bath	PS	−	+	+	+	+	−	+	Multifocal spikes, F-dominant sp-w	+	+	−	−
2	M	5	Father: FS	6	Fever	GTCS	+	+	+	−	+	+	+	F-dominant sp-w	+	−	−	−
3	M	4	Maternal uncle: FS	6	Fever	HS	+	+	+	+	+	+	+	Diffuse and focal sp-w	+	−	−
4	M	5	Sister: FS	5	Vaccination	GTCS	+	+	+	−	+	−	+	F-dominant sp-w	+	−	−	Lateral ventricle enlargement, L>R
5	F	5	Mother: FS	5	Fever	HS	+	+	+	+	+	+	+	Diffuse and multifocal spikes	+	−	−	−
6	M	1	Second paternal grandfather: FS	2	−	HS	+	+	+	+	+	−	+	Diffuse and multifocal spikes	+	−	−	−
7	M	3	Paternal sister:FS	5	Fever	GTCS	+	+	+	+	+	−	+	Multifocal and diffuse sp-w	+	−	−	−
8	M	14	−	9	Fever	PS	+	−	+	+	+	+	+	Mutifocal sp-w	+	+	+	−
9	F	9	Mother, paternal aunt and grandfather: FS	9	−	PS	+	+	+	+	+	+	+	Rare spikes	+	−	−	Mild atrophy
10	M	3	−	6	Fever	HS	+	+	+	+	+	+	+	Rare multifocal spikes IPS(+)	+	+	+	−
11	M	5	Father: FS	4	Fever	GTCS	+	+	+	+	+	+	+	Diffuse and focal sp-w	+	+	−	−
12	M	5	−	6	Fever	GTCS	+	+	+	−	−	+	+	Diffuse sp-w,	−	−	−	−

Genomic DNA was extracted from peripheral blood lymphocytes of the 12 SMEI patients, their parents, and 50 normal controls. Mutation analysis for SCN1A was performed by polymerase chain reaction and direct sequencing. Different heterozygous coding variants were found in 10 unrelated cases of the 12 SMEI patients (Table 2). This is the first confirmation of the significant correlation of SCN1A mutation and SMEI in a Chinese patient  population.

Table 2. SCN1A mutations in 10 SMEI patients

Patient	Exon	Mutation		Mechanism	Location in protein	Conservative analysis	The same mutations in		Inheritance
		cDNA (c.)	Protein (p.)				Father	Mother
+, present; –, absent; and ND, not determined. The numbering of the nucleotide acid and amino acids is according to the full-length isoform containing the full length of exon11 (SCN1A Genebank Accession Number, CDS_AB093548; P35498). The numbers of the patients in the Table 2 correspond to those in the Table 1.
1	2	269T>C	F90S	Transition	N-terminal	Highly	ND	−	ND
2	2	272T>C	I91T	Transition	N-terminal	Moderately	−	−	De novo
3	6	715G>A	A239T	Transition	D I S4-S5 linker	Highly	−	−	De novo
4	11	1667delT	L556fsX557	Frame shift	D I/II linker	Highly	ND	−	ND
5	11	1834C>T	R612X	Transition	D I-II linker	Highly	−	−	De novo
6	15	2854T>G	W952G	Transversion	D II S5-S6 linker	Highly			De novo
7	21	4003G>A	V1335M	Transition	D III S4-S5 linker	Highly	−	−	De novo
8	21	4168G>A	V1390M	Transition	D III S5-S6 linker	Highly	−	−	De novo
9	21	4223G>A	W1408X	Transition	D III S5-S6 linker	Highly	−	−	De novo
10	22	4298G>A	G1433E	Transition	D III S5-S6 linker	Highly	−	−	De novo

The SCN1A mutation rates vary markedly (33.3–100%) for SMEI, depend presumably on the numbers of the patients, the case selection criteria, and ethnic differences (Mulley et al., 2005; Fujiwara 2006). The detection rate (10/12, 83.3%) in our data is similar to that found in Japanese patients (82.7%), suggesting the homogeneity of ethnic and geographic conditions between Chinese and Japanese populations (Ohmori et al., 2002). It is possible that the high rate of SCN1A mutations in our data is due to the small number of the patients and the typical characteristics of the SMEI patients. The mutations within SCN1A are randomly distributed throughout the gene, without obvious “hot spots.” There were seven missense mutations, two nonsense mutations, and one deletion mutation generating a frame shift. To the best of our knowledge, mutations p.F90S, p.I91T, p.L556fsX557, p.R612X, p.W952G, p.V1335M, and p.G1433E (Table 2) are novel, which reinforces the mutational heterogeneity characteristic of SCN1A (Harkin et al., 2007).

The pathogenic function of the missense mutations is supported by the fact that none of them were present in 100 control chromosomes. Conservative analysis of these mutations suggests that they affect conserved domains of the protein in human alpha channels (and other vertebrates and invertebrates), consistent with their pathogenic function. All the missense mutations changed the charge state of the residues, which may alter the structural motif and function of the sodium channel. For example, in the p.G1433E mutation, the polar glutamine acid (147 Da) is significantly larger than the nonpolar amino acid glycine (75 Da). Three mutations—p.W952G, p.V1390M, and p.G1433E—were located in the pore region, whereas two—p.A239T and p.V1335M—were positioned near the voltage sensor. These missense mutations, then, were located in regions that may influence the ion selectivity and alter the functional activity of the channel (Yu and Catteral, 2003). Two mutations—p.F90S and p.I91T—were in the N-terminus, which would influence the localization and expression level of the α1 subunit in the neuron membrane. The truncation mutations in our study—p.L556fsX557, p.R612X, and p.W1408X—may lead to the predictable terminations at amino acids 557, 612, and 1408, respectively. These truncated proteins are likely to function as a dominant negative mutation, whereas the remaining normal genes show haploid insufficiency. Functional studies have demonstrated that almost all SCN1A truncation mutations associated with SMEI markedly reduced inward sodium currents and cause loss of function of SCN1A (Ohmori et al., 2006). Biophysical analysis will be required to clarify the exact pathogenic mechanism of these mutations on channel function.

A family history of febrile seizures or epilepsy was positive in 9 of the 10 patients associated with SCN1A mutations. Of the 10 families with SCN1A mutations, DNA was analyzed from 8 sets of parents (not available in the other two), and no mutation was found in the parental DNA—i.e., all the mutations appeared to be de novo. The fact that a majority of known SCN1A mutations associated with SMEI are de novo suggests not only that SCN1A mutations play a significant role in the pathogenesis of this disorder, but also that SMEI is likely to be a monogenic condition. Interestingly, the truncation mutation p.W1408X was detected in patient 9 with a bilineal family history; there was with no identifiable SCN1A defect in the parents. Relatives of SMEI patients appear to have febrile seizures and other types of epilepsy at a higher rate than the general population, although a significant difference was not observed (Mulley et al., 2005; Mancardi et al., 2006). The high frequency of antecedent febrile seizure or epilepsy observed in SMEI patients seems to complicate the interpretation of molecular studies that suggest that most SCN1A mutations in SMEI are of de novo origin.

Patients with SMEI are usually diagnosed after 2 years during routine investigation. When the patients have prolonged febrile seizures within the first year of life and later develop multiple seizure types associated with fever sensitivity, the diagnosis of SMEI should be considered (Caraballo & Fejerman, 2006). The positive result of a SCN1A mutation analysis would reinforce the diagnosis. The early diagnosis at the molecular level is of importance not only for genetic counseling, but to allow the anticipatory guidance and specific treatment (Korff et al., 2007; Striano et  al., 2007). Our study strengthens the view that SCN1A is commonly involved in the etiology of SMEI, and suggests that the distribution of SCN1A mutations in Chinese SMEI patients is potentially different from other patient populations. Further studies with a larger sample are warranted to establish genotype–phenotype correlations. Recognition of SCN1A mutations as the basis of this devastating epileptic encephalopathy avoids further potentially invasive investigations and assists in targeting therapy. The development of individualized treatments tailored to each patient with mutation is warranted.

Acknowledgments

We are very grateful to all members of the families who participated in our work. This research was supported by grants from Beijing Natural Science Foundation of China (No. 7072083).

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.