Variable epilepsy phenotypes associated with a familial intragenic deletion of the SCN1A gene

Summary

Deletions and duplications/amplifications of the α1-sodium channel subunit (SCN1A) gene occur in about 12% of patients with Dravet syndrome (DS) who are otherwise mutation-negative. Such genomic abnormalities cause loss of function, with severe phenotypes, reproductive disadvantage and, therefore, sporadic occurrence. Inherited mutations, occurring in ∼5% of patients with DS, are usually missense; transmission occurs from a mildly affected parent exhibiting febrile seizures (FS) or the generalized epilepsy with febrile seizures plus (GEFS+) spectrum. We identified an intragenic SCN1A deletion in a three-generation, clinically heterogeneous family. Sequence analysis of SCN9A, a putative modifier, ruled out pathogenic mutations, variants, or putative disease–associated haplotype segregating with phenotype severity. Intrafamilial variability in phenotype severity indicates that SCN1A loss of function causes a phenotypic spectrum in which seizures precipitated by fever are prominent and schematic syndrome subdivisions would be inappropriate. SCN1A deletions should be ruled out even in individuals with mild phenotypes.

The α1-sodium channel subunit (SCN1A) gene is the main genetic player in the etiology of severe myoclonic epilepsy of infancy [SMEI or Dravet syndrome (DS); Online Mendelian Inheritance in Man (OMIM) 607208]. SCN1A heterozygous de novo mutations account for most individuals with DS (Mulley et al., 2005). Chromosomal rearrangements such as deletions and duplications/amplifications of SCN1A have been detected in about 12% of patients with DS who are otherwise mutation-negative (Marini et al., 2009), with an overall frequency of 2–3%. Inherited SCN1A mutations, usually missense, are found in ∼5% of patients with DS. The family history of DS probands with inherited mutations is characterized by the occurrence of milder phenotypes, consistent with febrile seizures (FS), febrile seizure plus (FS+), or generalized epilepsy with febrile seizures plus (GEFS+; OMIM 604233) in the previous generation (Singh et al., 2001; Nabbout et al., 2003).

Possible reasons for intrafamilial and clinical heterogeneity have been widely discussed in the literature. The genetic and environmental background likely influences the variable familial expression of SCN1A mutations, whereas parental mosaicism accounts for only a few cases (Depienne et al., 2010). However, no definitive conclusion on the factors involved has been reached.

Herein we report a three-generation, clinically heterogeneous family in which we identified a genomic rearrangement leading to a partial SCN1A deletion.

Methods

Patients

The proband (III:1) was referred to our epilepsy clinic for investigations of a seizure disorder that had a familial distribution. We collected a clinical history of the family by interviewing Patient II:2, personally examined Patients II:3, III:1, and III:2; and obtained original medical records of Patient II:3. Electroencephalography (EEG) studies were performed in Patients II:3, III:1, and III:2.

The family originated from central Italy; a pedigree is shown in Fig. 1. The institutional review board approved the study protocol. Written informed consent was obtained from each subject. Blood samples were collected for DNA analysis.

Figure 1.   Pedigree of the family carrying the SCN1A deletion. The proband is indicated by the arrow. del, SCN1A deletion of exons 2–4; –, wild-type allele; ^, DNA available.

Based on the clinical history and, in particular, the seizure-precipitating role of fever, we decided to investigate the possible involvement of the SCN1A gene. After detecting a deletion within this gene, in view of the phenotypic heterogeneity of the family, we also investigated the SCN9A gene (Singh et al., 2009) to establish its possible role as genetic modifier.

Mutation detection and multiplex ligation-dependent probe amplification (MLPA) analysis

Genomic DNA was isolated from peripheral blood leukocytes using an automated DNA isolation robot, according to the manufacturer’s protocol (EZ1DNA blood, Qiagen, Hilden, Germany).

All the 26 coding exons and flanking intronic sequences of the SCN1A gene (GeneID: 6323; accession: AB093548.1) were amplified and then analyzed by denaturing high-performance liquid chromatography (dHPLC) on a Wave automated instrument (Transgenomic Inc., Cheshire, United Kingdom). Abnormal profiles observed on dHPLC were subsequently analyzed by sequencing, using the BigDye Terminator v.1.1chemistry (Applied Biosystems, Foster City, CA, U.S.A.). The products were analyzed on a 3130 XL ABI Prism DNA sequencer (Applied Biosystems). Coding exons and flanking intronic sequences of the SCN9A gene (GeneID: 6335; accession: NM_002977.2) were analyzed directly by sequencing. Primer sequences, polymerase chain reaction (PCR)/dHPLC, and sequencing conditions are available upon request.

The MLPA SCN1A kit (SALSA P137 version A2; MRC-Holland, Amsterdam, The Netherlands) contains 26 paired probes from the SCN1A region (covering all the exons) and 14 control probes to detect sequences located in other chromosomal regions. Deletion/duplication screening was performed according to the manufacturer’s protocol. Patients were tested as described previously (Marini et al., 2007). Confirmation of deletion was shown by repeatability in a second MLPA reaction and by observation that all deleted exons were contiguous. The SCN1A full-length isoform (AB093548.1) was studied in silico to assess whether the SCN1A intragenic deletion was “in” or “out” of frame.

Results

Patients

The family pedigree is shown in the Figure, and a clinical summary of affected family members is presented in Table 1.

Table 1.   Clinical features of patients with SCN1A deletion

Patient ID/Gender	Age	1st seizure, type	Seizure frequency/type during follow-up	Current seizure frequency	Cognitive level	Brain MRI	Current treatment	Diagnosis
CLB, clobazam; CNZ, clonazepam; DQ, developmental quotient; DS, Dravet syndrome; F, female; FS, febrile seizures; FS+, febrile seizures plus; FSIQ, Full Scale Intelligence Quotient; GC, generalized clonic seizures; GMDS-R, Griffiths Mental Developmental Scale-Revised; GTC, generalized tonic–clonic seizures; M, male; m, months; mod, moderate; MR, mental retardation; MRI, magnetic resonance imaging; NA, not available; PB, phenobarbital; S, seizures, TPM, topiramate; VPA, valproate; WISC-R, Wechsler Intelligence Scale for Children-Revised; y, year(s).
I:1/M	70 y	5 y, FS	Rare: FS, FS+, GTC until 12 y	S free	Normal	NA	None	FS+
II:2/F	40 y	3 m, FS	Rare: FS, FS+, Focal (visual) S until 9 y	S free	Normal	NA	None	FS+/Focal S
II:3/M	37 y	5 m, F status	Numerous FS, Unilateral, absence, atonic until 20 y, GC, GTC still present	4–5/year	Autism, Severe MR	Normal	VPA, CNZ, PB	DS
III:1/F	11 y	4 m, F status	Rare: FS, FS+, F status, Focal (motor) S	2–3/year	Mild-mod MR (FSIQ = 50, WISC-R)	Normal	TPM, VPA, CLB	FS+/Focal S
III:2/M	20 m	5 m, F status	Rare: FS, F status	2/year	Normal (DQ = 105, GMDS-R)	Normal	TPM, VPA	FS+

Mutation detection and MLPA analysis

dHPLC and sequence analysis of the SCN1A gene failed to reveal any point mutation. MLPA analysis on the genomic DNA of the proband revealed a deletion involving exons 2–4. This deletion affects the reading frame and is predicted to introduce a premature stop codon. Subsequent analysis revealed a complete segregation of the rearrangement in all affected family members (Fig. 1). Sequencing analysis of the SCN9A gene revealed no pathogenic mutations or variations.

Discussion

This family exhibits remarkable clinical heterogeneity, comprising various phenotypes, all of which had been associated previously with SCN1A abnormalities. Heterogeneity involved both seizure severity and cognitive impairment. Clinical variability highlights the difficulties in correlating SCN1A mutations with phenotype severity, even when dealing with a loss-of-function mutation.

Inherited loss-of-function mutations of SCN1A (including missense mutations in the pore-forming regions, frameshifts, nonsense, splice-site mutations, and genomic rearrangements) are rare, as they are usually associated with the most severe phenotypes and reproductive disadvantage (Nabbout et al., 2003; Marini et al., 2007; Gökben et al., 2009; Depienne et al., 2010). Two nonsense mutations, the p.Glu289X (Nabbout et al., 2003) and the p.Arg1645X (Gökben et al., 2009), as well as a deletion of one base pair resulting in the p.Asp1293ValfsX7 mutation (Marini et al., 2007), were inherited by patients with DS whose carrier parents had experienced isolated seizures (Marini et al., 2007; Gökben et al., 2009) or mild epilepsy (Nabbout et al., 2003).

Deletions involving the SCN1A gene vary in size from the megabase range, involving contiguous genes, down to one exon (Marini et al., 2009). Because SCN1A deletions result in loss of function, the associated phenotypes would be expected to be severe. Indeed, with very few exceptions, deletion carriers present as severe, sporadic cases (Marini et al., 2009; Suls et al., 2010). Conversely, in the family we are reporting, phenotypic heterogeneity, also including milder phenotypes, has permitted transmission through three generations. Patient I:1 had had FS+. Patient II:2, exhibiting FS+ and focal seizures with normal cognition, has a much milder phenotype with respect to her brother (II:3), whose clinical features are compatible with a severe form of DS. The proband (III:1) exhibits a clinical syndrome, the severity of which is intermediate between that observed in her mother (II:2) and uncle (II:3), with mild to moderate cognitive impairment, FS, and febrile status, as well as afebrile generalized and focal seizures. The proband’s brother (III:2) appears to be less severely affected, although he is still too young to be assigned a severity rate with confidence. Our observation confirms that by Suls et al. (2010), who described a clinically heterogeneous four-generation Bulgarian family with a deletion encompassing the SCN1A and TTC21B genes.

The phenotypic expression of SCN1A can be altered by genetic modifiers or somatic mosaicism. In particular, it has been indicated that SCN9A might be a putative modifier of SCN1A in DS (Singh et al., 2009). In addition, a mild phenotype in parents of severely affected individuals has been related to somatic mosaicism in rare cases (Gennaro et al., 2006; Morimoto et al., 2006; Depienne et al., 2009, 2010; Marini et al., 2009; Singh et al., 2009). Sequence analysis of the SCN9A gene revealed no pathogenic mutations, variants, or putative disease–associated haplotype segregating with the phenotype severity within the family we are reporting. Furthermore, somatic mosaicism was not identified in individual I:1.

Remarkable variability in phenotype severity within a family favors the concept that SCN1A loss of function causes a spectrum of epilepsy phenotypes in which seizures precipitated by fever are the prominent feature and schematic subdivisions would be inappropriate. For example, individuals II:2 and III:1 (mother and daughter) might have been classified as having either borderline DS or a combination of focal seizure and GEFS+, with or without febrile status. However, cognitive level was normal in individual II:2, but clearly impaired in her daughter.

Our findings demonstrate that the search for SCN1A abnormalities should rule out a deletion even in individuals with particularly mild phenotypes. Remarkable intrafamilial variability is in line with hypotheses favoring a crucial role of, as yet unknown, modifiers genes in modulating the phenotypic counterpart of SCN1A mutations.

Acknowledgments

This work was supported by the sixth Framework thematic priority Life sciences, Genomics and Biotechnology for Health, contract number: LSH-CT-2006-037315 (EPICURE) (to R.G.).

Disclosure

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. None of the authors has any conflict of interest to disclose.