Clinical spectrum of SCN1A mutations

Summary

Mutations in the NaV1.1 neuronal sodium channel alpha-subunit (SCN1A) gene have been documented in a spectrum of epilepsy syndromes, ranging from the relatively benign generalized epilepsy with febrile seizures plus (GEFS⁺) to severe myoclonic epilepsy in infancy (SMEI), and rare cases of familial migraine. More than 300 new mutations have been identified to date, with missense mutations being the most common in GEFS⁺ and more deleterious mutations (nonsense, frameshift) representing the majority of SMEI mutations. Microchromosomal abnormalities including SCN1A deletions, amplifications, and duplications are also found in patients with SMEI. Deletions range in size from one single exon to abnormalities extending beyond SCN1A and involving contiguous genes. The majority of SCN1A mutations in SMEI arise de novo. SCN1A mutations are found throughout the protein structure, and some clustering of mutations is observed in the C-terminus and the loops between segments 5 and 6 of the first three domains of the protein. Functional studies so far show no consistent relationship between changes to channel properties and clinical phenotype.

The voltage-gated Na channel alfa1 subunit (SCN1A) (MIM# 182389) is the most clinically relevant among all the known epilepsy genes, with the largest number of epilepsy related mutations so far characterized (Mulley et al., 2005). SCN1A is constructed with a 4-fold symmetry consisting of structurally homologous domains (D1–D4), each containing six membrane-spanning segments (S1–S6) and a region (S5–S6 pore loop) controlling ion selectivity and permeation. Genetic defects of SCN1A are found throughout the protein structure, and some clustering of mutations is observed in the C-terminus and the loops between segments 5 and 6 of the first three domains of the protein. SCN1A mutations have been linked to several epilepsy syndromes with overlapping clinical characteristics but divergent clinical severity (Escayg et al., 2000; Wallace et al., 2001; Claes et al., 2003; Ceulemans et al., 2004; Mulley et al., 2005), and in rare cases of familial migraine (Dichgans et al., 2005). The majority of known mutations in SCN1A lead to severe myoclonic epilepsy in infancy (SMEI; MIM# 607208), with significant numbers also accounting for generalized epilepsy with febrile seizures plus (GEFS⁺; MIM# 604233), borderline SMEI (SMEB), and intractable childhood epilepsy with generalized tonic–clonic seizures (ICEGTCS). SCN1A mutations have also been reported in a few patients with some other rare early onset epileptic encephalopathies; such patients could probably be included in the SMEI spectrum (Harkin et al. 2007). Despite this phenotypic variability, most epilepsies associated with SCN1A mutations are characterized by fever as a trigger factor. Nonetheless, disruption of SCN1A is not a major cause of familial simple febrile seizures (FS). Moreover, despite the expression of SCN1A in the heart, cardiac disturbances seem not to occur in these SCN1A-related epilepsies.

SMEI and Borderline Phenotypes

SMEI or Dravet syndrome is a severe syndrome with onset in the first year of life, typically beginning with prolonged febrile hemiclonic or GTCS. The condition evolves with myoclonic, absence, partial, and atonic seizures developing between 1 and 4 years of age associated with developmental delay and regression. Two acronyms, “borderline SMEI (SMEB)” and ICEGTC, have been used to describe patients resembling SMEI but in which myoclonic seizures are absent with a less severe psychomotor impairment (Fujiwara et al., 2003).

SMEI is the most common (more than 70%) phenotype associated with SCN1A mutations, and if SMEB and ICEGTC are included, about 90% of individuals with SCN1A mutations have this disabling disorder (Claes et al., 2001; Fujiwara et al., 2003; Nabbout et al., 2003; Ceulemans et al., 2004; Marini et al., 2007). The majority of SCN1A mutations in SMEI arise de novo, which does not explain the family history of GEFS+ that is observed in 50% of patients with SMEI (Fujiwara et al., 2003). Most of these mutations are nonsense or frameshift mutations resulting in truncation of the protein; missense mutations are also common, occurring in about one-third of patients with SMEI, whereas deletions and splice-site mutations are more rare. Several recurrent mutations are now emerging as the number of published studies increases.

The phenotypic range of infantile epileptic encephalopathies associated with SCN1A mutations has been recently expanded to also include cryptogenic focal and generalized epilepsies, myoclonic–astatic, and Lennox Gastaut syndrome (Harkin et al., 2007; Zucca et al., 2008) and the so-called vaccine encephalopathy (Berkovic et al., 2006). Overall, these data indicate that SCN1A mutation should be considered in any epileptic encephalopathy with seizure onset before 1 year of age, even if cognitive decline does not occur for several years thereafter. Nonetheless, the full phenotypic spectrum of SCN1A mutations is still undefined, as there are controversial aspects to the diagnosis of infantile encephalopathies, especially SMEI, in which it remains questionable whether myoclonic seizures are essential or whether intellectual disability must be severe.

Recent studies have also illustrated that about 12% of mutation-negative SMEI patients have microchromosomal abnormalities—deletions, amplifications, or duplications— involving SCN1A; these genomic anomalies are best detected with multiplex ligation-dependent probe amplification (Marini et al., 2007). Large deletions extend beyond SCN1A and involve other contiguous genes. The involvement of other GEFS⁺-related genes SCN1B, SCN2A, GABRG2, and the subunit SCN2B in SMEI has been examined, with these genes excluded as having a major role in the development of SMEI (Harkin et al., 2002).

Clinical and Genetic Aspects of GEFS⁺

GEFS⁺ is a familial epilepsy syndrome with autosomal dominant inheritance and incomplete penetrance in some large pedigrees, and a polygenic inheritance in the majority of small families that shows a marked phenotypic heterogeneity. The most common phenotypes are FS or FS plus, when seizures with fever persist beyond the age of 6 years or they are associated with afebrile tonic–clonic seizures (Scheffer & Berkovic, 1997). Less frequent phenotypes seen in GEFS⁺ include mild generalized epilepsies, severe epileptic encephalopathies such as myoclonic–astatic epilepsy and, more rarely, SMEI, which represents the very severe end of the spectrum. Temporal lobe epilepsy with or without hippocampal sclerosis has been increasingly recognized in the GEFS⁺ spectrum associated with SCN1A mutations (Mantegazza et al., 2005).

SCN1A mutations are present in about 10% of familial GEFS⁺ patients (Escayg et al., 2000;Wallace et al., 2001); they are spread throughout the gene but most frequently outside the pore-forming region and are typically missense mutations (Kanai et al., 2004; Mulley et al., 2005). Mutations in other epilepsy-associated ion channel subunit genes SCN1B and GABRG2 also account for a small percentage of GEFS⁺, leaving the majority of familial cases of GEFS⁺ unexplained.

SCN1A Mutations and Other Phenotypes

The spectrum of phenotypes associated with SCN1A mutations has also been enlarged to include other infantile epileptic encephalopathies, namely, cryptogenic generalized epilepsy, cryptogenic focal epilepsy, and a subgroup of patients designated as severe infantile multifocal epilepsy (Harkin et al. 2007).This phenotype is characterized by early onset multifocal seizures and later cognitive decline. Single patients with myoclonic–astatic epilepsy (Ebach et al., 2005), Lennox-Gastaut syndrome (Harkin et al. 2007), infantile spasm (Wallace et al., 2003), Rasmussen encephalitis (Ohmori et al., 2008), and Panayiotopoulos (Grosso et al., 2007) carrying SCN1A changes are also on record.

The Mutational Spectrum of SCN1A and Genotype–Phenotype Correlation

A comprehensive list of SCN1A mutations has been recently reported (Lossin, 2009), and the related website (http://web.scn1a.info) is being updated.

The great majority of mutations lead to SMEI and related early onset epileptic encephalopathies, suggesting that there is a genotype–phenotype correlation. A substantial number of SCN1A mutations are also associated with the much milder GEFS+ phenotypes. There are clinical similarities between SMEI and GEFS⁺, including recurrent, often prolonged, seizures provoked by fever in infancy, the family history, and the shared molecular genetic etiology, raising the hypothesis that these disorders represent two extremes in clinical presentation of the same condition. Molecular evidences of a shared etiology include the finding that the R1648H mutation can give rise to both conditions: The mutation caused GEFS⁺ in a family (Escayg et al., 2000) whereas it caused SMEI in another patient (Ohmori et al., 2002). Nevertheless a clinical diversity and a mutational diversity between SMEI/SMEB and GEFS⁺ are observed.

In general, the milder phenotypes of GEFS⁺ or ICEGTC are associated with missense mutations, whereas many SMEI and SMEB patients have truncation mutations (Ceulemans et al., 2004). The severity may also be explained by the position of these mutations, as there seems to be an association between SMEI and missense mutations in the pore region (Fujiwara et al., 2003; Kanai et al., 2004). It is also possible that germline and somatic SCN1A mutational mosaicism may contribute to individual and familial phenotypic variability (Marini et al., 2007).

Despite a vast amount of data, however, genotype–phenotype correlation is not yet clear. Moreover, the fact that GEFS⁺ families show a large phenotypic variability that ranges from mild phenotypes to very severe ones (Scheffer & Berkovic, 1997) indicates that a more complex interaction between genetic and acquired factors modulate disease severity of produced phenotypes.

Functional Consequences of SCN1A Mutations

Various epileptogenic mutations of SCN1A have been reported that cause both loss- and gain-of-function of Nav1.1 in vitro (Spampanato et al., 2003; Mantegazza et al., 2005). Therefore, the downstream result of opposite alterations in sodium currents presumably has different effects on inhibitory and excitatory neuronal networks, leading to the common final pathway of epileptogenesis. It is not clear yet how a loss of function of sodium channel can give rise to a network hyperexcitability that underlie epileptic seizures.

A better understanding of the underlying epileptogenic mechanism caused by SCN1A could come from animal models. In (haploinsufficient-equivalent) heterozygote Scn1a+/− knockout mice, reduced sodium current density was noted only in hippocampal inhibitory interneurons, and not in excitatory pyramidal neurons (which upregulated other sodium channel expression), which the authors suggested might explain hyperexcitability underlying seizures in Dravet syndrome (Yu et al., 2006).

Recent studies provided an intriguing potential mechanism for a specific loss-of-function SCN1A mutation that could also cause defective protein trafficking and protein–protein interactions, which may modulate the effect of mutation and, so, underlie phenotypic variation in some epileptic families with SCN1A mutations (Hirose, 2006; Rusconi et al., 2007). Aberrant protein trafficking has been already demonstrated for several γ-aminobutyric acid (GABA)-receptor subunits and the SCN5A sodium channel gene causing inherited cardiac arrhythmias. So, it is tempting to speculate that defective protein trafficking comprises a common motif for genetic epilepsies, especially channelopathies.

At present, there is no unifying mechanism that can explain how the spectrum of the observed functional effects of epileptogenic mutations relates to the epilepsy syndromes seen in patients. Certainly, simple loss-of-function of SCN1A mutations does not account for the clinical distinction between SMEI and GEFS⁺.

Conclusions

Mutations in SCN1A are associated with a wide range of human epilepsy syndromes and rare cases of familial migraine. Functional defects range from loss- to gain-of-function. The relationship between clinical syndrome and biophysical phenotypes is complex, and other genetic, developmental, and environmental factors may influence the clinical expression of SCN1A mutations. A better understanding of the pathophysiologic basis for Na-channel dysfunction may help with the design of targeted therapies.

Acknowledgments

This work was supported by the Genetic Commission of the Italian League Against Epilepsy.

We have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

Disclosure: The authors have no conflicts of interest to declare.