Epileptic Encephalopathy, Early Infantile, 6

A number sign (#) is used with this entry because early infantile epileptic encephalopathy-6 (EIEE6) is caused by heterozygous mutation in the SCN1A gene (182389) on chromosome 2q24. About 95% of the mutations are de novo (Claes et al., 2001; Vadlamudi et al., 2010).

Mutations in the SCN1A gene are also responsible for autosomal dominant generalized epilepsy with febrile seizures plus (GEFS+) (GEFSP2; 604403).

Description

Dravet syndrome, first described by Dravet (1978), is a clinical term for early-onset epileptic encephalopathy (EIEE) characterized by generalized tonic, clonic, and tonic-clonic seizures that are initially induced by fever and begin during the first year of life. Seizures are usually refractory to treatment. Later, patients also manifest other seizure types, including absence, myoclonic, and partial seizures. The EEG is often normal at first, but later characteristically shows generalized spike-wave activity. Psychomotor development stagnates around the second year of life, and affected individuals show subsequent mental decline and other neurologic manifestations (summary by Harkin et al., 2007).

Since mutation in the SCN1A gene can also cause the less severe disorder autosomal dominant generalized epilepsy with febrile seizures-plus, Dravet syndrome and migrating partial seizures of infancy (MPSI) are considered to be the most severe phenotypes within the spectrum of SCN1A-related epilepsies (Ohmori et al., 2002; Carranza Rojo et al., 2011).

Deprez et al. (2009) provided a review of the genetics of epilepsy syndromes starting in the first year of life, and included a diagnostic algorithm.

For a general phenotypic description and a discussion of genetic heterogeneity of early infantile epileptic encephalopathy, see EIEE1 (308350).

Clinical Features

Dravet syndrome, previously known as 'severe myoclonic epilepsy of infancy' (SMEI), is an epileptic syndrome characterized by normal development before onset, seizures beginning in the first year of life in the form of generalized or unilateral febrile clonic seizures, secondary appearance of myoclonic seizures, and occasionally partial seizures. It is associated with ataxia, slowed psychomotor development, and mental decline, and is often refractory to medication (Dravet et al., 1992; Sugawara et al., 2002).

Renier and Renkawek (1990) reported that an autopsy of a 19-month-old boy with SMEI showed microdysgenesis of the cerebellum and cerebral cortex as well as malformation of the spinal cord.

Doose et al. (1998) reported a large group of patients with severe intractable epilepsy of infancy or childhood with frequent generalized tonic-clonic seizures. At onset, the disorder was characterized by prolonged febrile and afebrile seizures as the only seizure type. With advancing age, the symptomatology became increasingly polymorphic due to additional seizure types, such as complex or focal. The most common triggering feature was fever or immersion in a hot bath, and most patients had severe impairment of mental development after seizure onset. Doose et al. (1998) noted the phenotypic overlap with SMEI.

Fujiwara et al. (2003) reported 25 Japanese patients with SMEI and 10 Japanese patients with what they termed 'intractable childhood epilepsy with generalized tonic-clonic seizures (ICEGTC),' which was only distinguished from SMEI by the absence of myoclonus. Twenty-two (62.8%) patients had a family history of seizures, including febrile convulsions and epilepsy consistent with GEFS+. The majority of patients had high voltage 4- to 7-Hz diffuse slow background activity on EEG. A total of 30 heterozygous mutations were identified in the SCN1A gene in this group of patients.

Buoni et al. (2006) reported a 13-year-old boy with SMEI in whom the clinical phenotype evolved to GEFS+2 in adolescence. The patient had prolonged febrile seizures at ages 6, 10, and 13 months, afebrile complex partial seizures with secondary generalization beginning at age 18 months, and 2 episodes of status epilepticus at age 2 years. He also had abnormal EEG findings and myoclonic jerks. Antiepileptic medication was unsuccessful. At age 4 years, the seizure frequency decreased in response to medication, and by age 9, he had complex partial seizures with secondary generalization. By age 13, he was treated with valproate and had a febrile seizure. There was no mental retardation. Buoni et al. (2006) emphasized the relatively benign outcome in this patient despite having SMEI. Genetic analysis identified a de novo heterozygous 1-bp deletion in the SCN1A gene (182389.0017).

Jansen et al. (2006) reported 14 adults with Dravet syndrome who ranged in age from 18 to 47 years. All had been referred for refractory epilepsy and intellectual disability without an etiologic diagnosis. Medical history revealed seizure onset between 3 to 11 months (mean 6 months), which was associated with fever in 9 patients. During childhood, all had generalized or unilateral tonic-clonic seizures, 12 had myoclonic seizures, 11 had absence seizures, 8 had complex partial seizures, and 6 had atonic seizures. Psychomotor development slowed in all after initial normal development. Eight patients had a family history of seizures. As adults, generalized tonic-clonic seizures were the dominant type, but all other types of seizures still occurred. Ten patients had motor abnormalities, including cerebellar signs in 4, pyramidal signs in 6, and extrapyramidal signs in 4. One patient had low-average intellect, 2 had mild intellectual disability, 5 were moderately retarded, and 6 had severe impairment. Two patients lived independently but were unemployed. Genetic analysis showed that 10 patients had mutations in the SCN1A gene and 1 had a mutation in the GABRG2 gene. Jansen et al. (2006) noted that the findings indicated a poor outcome for affected individuals and emphasized that correct diagnosis in adult patients requires a knowledge of early medical history.

Riva et al. (2009) found that 2 unrelated children with genetically confirmed Dravet syndrome had progressive neurocognitive decline when longitudinally assessed from ages 11 and 23 months to 7 and 8 years, respectively. Importantly, delayed motor, intellectual, and rational development was already apparent at the time of seizure onset in both patients. One patient had a more severe seizure phenotype consistent with an epileptic encephalopathy, with numerous myoclonic seizures occurring almost daily and more frequent occurrence of tonic-clonic seizures compared to the second patient. However, both patients showed progressive deterioration in cognitive function over time, although there were differences in specific neuropsychologic functions affected. Riva et al. (2009) concluded that SCN1A mutations may play a role in early and progressive mental impairment in addition to their role in epilepsy.

Clinical Variability

Harkin et al. (2007) identified SCN1A mutations in a cohort of patients with a wide spectrum of infantile epileptic encephalopathies. Among a total of 188 patients, SCN1A mutations were found in 52 (79%) of 66 with SMEI (Dravet syndrome) and in 25 (69%) of 36 with 'severe myoclonic epilepsy of infancy-borderline (SMEB),' a phenotype lacking one or more features of SMEI, such as myoclonus or generalized spike-wave discharges on EEG. In addition, SCN1A mutations were less commonly found in patients with other forms of early-onset epilepsy, characterized as cryptogenic generalized or focal epilepsy, myoclonic-astatic epilepsy, and severe infantile multifocal epilepsy (SIMFE). Although the study indicated that a broader range of seizure phenotypes is associated with SCN1A mutations, Harkin et al. (2007) noted that the nosologic boundaries between these phenotypes is blurred. There were no apparent genotype/phenotype correlations.

'Malignant migrating partial seizures of infancy' (MPSI, MMPSI) is a clinical term for a severe form of infantile epileptic encephalopathy with seizure onset between 1 day and 6 months. EEG studies typically show migrating focal onset progressing to multifocal onset, and seizures are refractory to therapeutic intervention. Affected individuals have developmental regression after seizure onset, severe global developmental delay, and progressive microcephaly. Early death often occurs. The phenotype is considered to be more severe than that of typical Dravet syndrome (summary by Freilich et al., 2011 and Carranza Rojo et al., 2011). Freilich et al. (2011) reported a female infant with EIEE6 manifest clinically as MPSI associated with a heterozygous mutation in the SCN1A gene (A1669E; 182389.0023). She had a severe phenotype, with onset of seizures at age 10 weeks, progression to refractory recurrent seizures by age 5 months, status epilepticus, EEG evidence of migrating focal onset progressing to multifocal seizures, progressive microcephaly, and profound psychomotor delay. She died at age 9 months.

Carranza Rojo et al. (2011) found that 2 of 15 unrelated infants with a clinical diagnosis of MPSI had defects in the SCN1A gene. One had a de novo missense mutation (R862G; 182389.0024) and the other had a de novo 11.06-Mb deletion of chromosome 2q24.2-q31.1 encompassing more than 40 genes that included SCN1A. The patient with the R862G mutation had onset of multifocal hemiclonic seizures at age 2 weeks with status epilepticus. She had acquired microcephaly, developmental regression, and severe intellectual disability. These reports expanded the severity of the epileptic phenotype associated with SCN1A mutations to include MPSI. Moreover, the lack of SCN1A mutations in 13 patients with a similar diagnosis by Carranza Rojo et al. (2011) indicated genetic heterogeneity for the MPSI entity.

Inheritance

Approximately 95% of patients with Dravet syndrome have de novo heterozygous mutations, which explains the unaffected status of many sibs and parents (Vadlamudi et al., 2010).

Fujiwara et al. (1990) reported a pair of monozygotic male twins who both had SMEI and showed a similar phenotype with regard to seizure onset, seizure symptomatology, and EEG expression.

Of 12 unrelated patients with Dravet syndrome, Singh et al. (2001) found that 11 had a family history of seizures and the twelfth was the offspring of a consanguineous marriage. A total of 39 related affected individuals were identified and the phenotypes included febrile seizures, partial seizures, and several unclassified seizures. Singh et al. (2001) suggested that Dravet syndrome is the most severe form of generalized epilepsy with febrile seizures plus (see 604233).

Selmer et al. (2009) reported a family of Norwegian origin in which a mother with a history of migraine was somatic mosaic for a truncating SCN1A mutation that she transmitted, through 2 different husbands, to her 2 daughters who had Dravet syndrome. The mother had attacks of migraine without aura since age 12 to 14 years; the mutation was estimated to be present in approximately 5% of the mother's blood and inferred to be present in a proportion of her germ cells. Selmer et al. (2009) postulated that migraine in the mother may represent the mildest end of the phenotypic spectrum caused by SCN1A mutations.

Of 44 SCN1A mutations that occurred de novo in patients with Dravet syndrome, Heron et al. (2010) found that 75% were of paternal origin and 25% were of maternal origin. The cohort included 1 set of affected sibs, whose originating parent was thought to have gonadal mosaicism. The average age of parents did not differ from that of the general population. The findings indicated that de novo SCN1A mutations originated most commonly, but not exclusively, from the paternal chromosome. Heron et al. (2010) suggested that the greater frequency of paternally derived SCN1A mutations was likely due to the greater chance of mutational events because of the increased number of mitoses during spermatogenesis compared to oogenesis, with a greater susceptibility to mutagenesis of methylated DNA characteristic of sperm cells.

Depienne et al. (2010) studied 19 families in which at least 1 individual had Dravet syndrome due to an inherited SCN1A mutation. In 12 cases, the transmitting parent was mosaic for the mutation, and the proportion of each mutation in parental blood cells ranged from 0.4 to 85%. The mutation was inherited from the mother in 6 cases and from the father in 6 cases. Six of the parents who were mosaic had mild features, including febrile seizures and tonic-clonic seizures, and the seizure phenotype correlated partially with increasing mutation load in blood cells. In the 6 remaining families, an SCN1A missense mutation segregated with Dravet syndrome and with autosomal dominant GEFS+ (GEFSP2; 604403). The findings indicated that some families with SCN1A mutations show wide phenotypic variability, with Dravet syndrome at the severe end of the spectrum.

Vadlamudi et al. (2010) reviewed the effect of timing of de novo mutagenesis in the SCN1A gene and described a discordant monozygotic twin pair, in which 1 SMEI-affected sib carried a heterozygous SCN1A truncation mutation. Detailed mutation analysis of various tissues from the affected twin identified a truncating SCN1A mutation (182389.0008) in lymphocytes, hair, buccal cells, skin fibroblasts, and cell lines derived from neuroepithelium, but not in tissues taken from the unaffected twin, the parents, or an unaffected sib. No evidence of somatic mosaicism was detected in the unaffected twin or the parents. Since the mutation was found in all tissues from the affected twin but not in tissues from the unaffected twin, Vadlamudi et al. (2010) concluded that the SCN1A mutation occurred in the premorula stage, most likely at the 2-cell stage.

Cytogenetics

Suls et al. (2010) reported a 4-generation Bulgarian family with epilepsy transmitting a heterozygous 400-kb deletion on chromosome 2q24 encompassing the SCN1A and TTC21B (612014) genes. The phenotype was variable, but all had onset of generalized tonic-clonic seizures around the first year of life (range, 8 to 14 months), and some had myoclonic or absence seizures. Three of 4 patients had febrile seizures in infancy. One patient had mild mental retardation, 1 had psychomotor slowing, and 1 had mental retardation from early infancy; all had reduced seizures on medication. The fourth patient died of status epilepticus at age 13 months. Thus, 2 patients had a phenotype reminiscent of Dravet syndrome, whereas the phenotype in the other 2 was more consistent with GEFS+2. The unaffected father in the first generation was found to be somatic mosaic for the deletion. Suls et al. (2010) noted that deletions involving SCN1A usually result in Dravet syndrome, in which affected individuals cannot raise a family and thus do not transmit the mutation. The report of this family with a deletion of SCN1A in which 2 affected individuals were able to raise a family suggested the presence of genetic modifiers and showed intrafamilial variability.

Molecular Genetics

Mutations in the SCN1A Gene

In 7 patients with Dravet syndrome, Claes et al. (2001) found heterozygous mutations in the SCN1A gene, including 3 deletions and 1 insertion that resulted in premature stop codons, a nonsense, a splice donor site, and a missense mutation; see, e.g., 182389.0007-182389.0009. The mutations were absent in all parents, suggesting that de novo mutations are a major cause of SMEI. Claes et al. (2001) noted that most of the mutations resulted in early termination of translation, producing a truncated SCN1A protein.

In 14 patients, including a pair of monozygotic twins, with classic symptoms of Dravet syndrome, Sugawara et al. (2002) identified 10 heterozygous mutations in the SCN1A gene. There were 3 frameshift mutations which resulted in intragenic stop codons and truncated channels, and 7 nonsense mutations which also resulted in truncated channels. In 4 patients, no mutations were detected in either the SCN1A or SCN1B (600235) genes.

In 24 of 29 patients with Dravet syndrome, Ohmori et al. (2002) found heterozygous de novo mutations in SCN1A, mutations in which have been identified also in GEFS+. That mutations in the SCN1A gene can cause severe myoclonic epilepsy in infancy supports the suggestion of Singh et al. (2001) that Dravet is part of the GEFS+ spectrum. Indeed, Dravet syndrome and GEFS+ have been observed in the same family.

Among 93 patients with Dravet syndrome, Nabbout et al. (2003) identified 29 different mutations in the SCN1A gene in 33 patients (35%). All cases were sporadic, but a history of febrile seizures and epilepsy was found in the families of 32% and 12% of the probands, respectively. Three of the mutations were inherited from a parent. The authors concluded that the disorder is genetically heterogeneous and may also exhibit complex inheritance.

In 7 of 10 unrelated Japanese patients with intractable childhood epilepsy with generalized tonic-clonic seizures, Fujiwara et al (2003) identified mutations in the SCN1A gene (see, e.g., 182389.0013; 182389.0014). All of the mutations were missense. Two unrelated affected children had mothers with the mutation who had a phenotype consistent with GEFS+. Fujiwara et al. (2003) concluded that myoclonus is not a necessary feature of the disorder.

Using multiplex ligation-dependent probe amplification (MLPA), Mulley et al. (2006) identified exon deletions in the SCN1A gene (182389.0018; 182389.0019) in 2 (15%) of 13 unrelated SMEI patients who did not have point or splice site mutations in the SCN1A gene. The findings provided a new molecular mechanism for the disorder.

Depienne et al. (2009) identified pathogenic mutations or deletions, including 161 novel point mutations, in the SCN1A gene in 242 (73%) of 333 patients with Dravet syndrome. The most common mutations were missense (42%), and 14 patients had microrearrangements in or deletions of the gene. Thus, the disease mechanism appeared to be haploinsufficiency of the SCN1A gene. Mutations were scattered throughout the gene, and there were no apparent genotype/phenotype correlations.

Orrico et al. (2009) identified 21 mutations, including 14 novel mutations, in the SCN1A gene in 22 (14.66%) of 150 Italian pediatric probands with epilepsy. SCN1A mutations were found in 21.2% of patients with GEFS+ (604233) and in 75% of patients with SMEI from the overall patient cohort. Only 1 potentially pathogenic mutation was identified in the SCN1B gene (600235), and no mutations were found in the GABRG2 gene (137164).

Sun et al. (2010) identified pathogenic mutations in the SCN1A gene in 49 (77.8%) of 63 Chinese probands with Dravet syndrome. The majority of mutations were truncating (61.2%). The mutations included 19 missense, 14 frameshift, 6 nonsense, and 8 splice site alterations. MLPA analysis identified deletions or duplications of SCN1A in 2 (12.5%) of 16 patients who were negative by sequencing. Forty mutations were de novo, and 1 was inherited from a mother who was mosaic for the mutation and had a phenotype consistent with GEFS+. Ten of 12 de novo mutations studied were of paternal origin, and 2 were of maternal origin. Sun et al. (2010) emphasized that MLPA analysis is essential for correct diagnosis in sequencing-negative patients with Dravet syndrome.

Potential Modifier Genes

Harkin et al. (2002) reported a family with GEFS+ (604233) caused by a heterozygous mutation in the GABRG2 gene (Q351X; 137164.0003); 1 family member had a more severe phenotype, consistent with Dravet syndrome. However, Ohmori et al. (2002) found no mutations of the GABRG2 gene in 29 patients with Dravet syndrome. They also found no mutations in SCN1B (600235), the other gene that had been related to generalized epilepsy with febrile seizures.

In 2 patients diagnosed with Dravet syndrome, Singh et al. (2009) identified a heterozygous mutation in the SCN9A gene (K655R; 603415.0019); one of the patients also had a mutation in the SCN1A gene (182389). The K655R mutation was also identified in a patient with GEFSP7 (see 604233). Singh et al. (2009) also presented evidence that the SCN9A gene on chromosome 2q24 may be a modifier of Dravet syndrome; 9 (8%) of 109 patients with Dravet syndrome were found to have an SCN9A mutation, including 6 patients who were double heterozygous for SCN9A and SCN1A mutations and 3 patients with only heterozygous SCN9A mutations, consistent with multifactorial inheritance.

Population Genetics

From an analysis of data on children with seizures from a national database, Hurst (1990) determined that the incidence of SMEI is 1 in 40,000.

Nomenclature

Dravet syndrome has also been referred to as 'severe myoclonic epilepsy of infancy' (SMEI), but the term 'Dravet syndrome' is preferred because not all affected individuals show myoclonus (Engel, 2001). Dravet syndrome is also referred to here as early infantile epileptic encephalopathy-6 (EIEE6).

Animal Model

Yu et al. (2006) found that Scn1a -/- mice developed severe ataxia and seizures and died on postnatal day 15. Scn1a +/- mice had spontaneous seizures and sporadic deaths beginning after postnatal day 21, with a notable dependence on genetic background. Loss of Scn1a did not change voltage-dependent activation or inactivation of sodium channels in hippocampal neurons. However, the sodium current density was substantially reduced in inhibitory interneurons of Scn1a -/- and +/- mice. The findings suggested that reduced sodium currents in GABAergic inhibitory interneurons resulting from heterozygous SCN1A mutations may cause the hyperexcitability that leads to epilepsy in patients with SMEI.

Oakley et al. (2009) generated a mouse model of SMEI by targeted heterozygous deletion of the Scn1a gene. Mutant mice developed seizures induced by elevated core body temperature, whereas wildtype mice were unaffected. In 3 age groups studied, none of postnatal day (P) 17 to 18 mutant mice had temperature-induced seizures, but nearly all P20 to P22 and P30 to P46 mutant mice developed myoclonic seizures followed by generalized seizures caused by elevated core body temperature. There was an age-related susceptibility to seizures at lower temperatures as well as a general increase in severity of seizures with increasing age. Spontaneous seizures were only observed in mice older than P32, suggesting that mutant mice become susceptible to temperature-induced seizures before spontaneous seizures. Interictal EEG spike activity was seen at normal body temperature in most P30 to P46 mutant mice, but not in P20 to P22 or P17 to P18 mutant mice, indicating that interictal epileptic activity correlates with seizure susceptibility. Most P20 to P22 mutant mice had interictal spike activity with elevated body temperature. Oakley et al. (2009) concluded that their results defined a critical developmental transition for susceptibility to seizures in SMEI, demonstrated that body temperature elevation alone is sufficient to induce seizures in mutation carriers, and revealed a close correspondence between human and mouse SMEI in the temperature and age dependence of seizure frequency and severity.

Martin et al. (2007) showed that the seizure severity of heterozygous Scn1a +/- mice (see Yu et al., 2006), which is a mouse model for SMEI, was ameliorated by a heterozygous point mutation (med-jo) in the Scn8a gene (600702). Double-heterozygous Scn1a +/- and Scn8a +/(med-jo) mice had seizure thresholds that were comparable to wildtype littermates, and the Scn8a(med-jo) allele was also able to rescue the premature lethality of Scn1a +/- mice and extended the life span of Scn1a -/- mice. The authors hypothesized that the opposing effects of Scn1a and Scn8a dysfunction on seizure thresholds result from differences in the cell types that are influenced by the respective sodium channel subtypes. Scn1a mutants result in reduced sodium currents in inhibitory GABAergic interneurons of the hippocampus and cortex, whereas Scn8a mutants affect excitatory pyramidal cells of the hippocampus and cortex, suggesting that reduced excitability of these cells may underlie the elevated seizure resistance of Scn8a-mutant mice. Martin et al. (2007) suggested that their results demonstrated that genetic interactions can alter seizure severity, and supported the hypothesis that genetic modifiers, including the SCN8A gene, contribute to the clinical variability observed in SMEI and GEFS+.

Han et al. (2012) reported that mice with Scn1a haploinsufficiency exhibit hyperactivity, stereotyped behaviors, social interaction deficits, and impaired context-dependent spatial memory. Olfactory sensitivity is retained, but novel food odors and social odors are aversive to Scn1a +/- mice. GABAergic neurotransmission is specifically impaired by this mutation, and selective deletion of Na(v)1.1 channels in forebrain interneurons is sufficient to cause these behavioral and cognitive impairments. Remarkably, treatment with low-dose clonazepam, a positive allosteric modulator of GABA(A) receptors, completely rescued the abnormal social behaviors and deficits in fear memory in the mouse model of Dravet syndrome, demonstrating that they are caused by impaired GABAergic neurotransmission and not by neuronal damage from recurrent seizures. Han et al. (2012) concluded that their results demonstrated a critical role for Na(v)1.1 channels in neuropsychiatric functions and provided a potential therapeutic strategy for cognitive deficit and autism spectrum behaviors in Dravet syndrome.