Genetics

The prominent role of genetic factors in the causation of epilepsy has been suspected for centuries based on the observations of familial aggregation but, in spite of extensive research, its definition has been elusive. A complete review of the genetic contribution to epilepsy is not possible because epilepsy is a heterogeneous condition in which many causative factors may intervene. The basic mechanisms of the epilepsies involve many of the neural processes that have been recognized in the normal brain, making the number of genes that could be involved in epileptogenesis quite high (Anderson et al., 2002). Although the study of rare epilepsy syndromes with single gene inheritance has recently permitted the discovery of several genes involved in epileptogenesis, most common genetic epilepsies do not follow familial patterns of aggregation consistent with simple inheritance.

The genetically determined epilepsies may be divided into the following two major subgroups: those that originate directly from the functional consequences of a defective gene product on neuronal excitability and those secondary to structural brain abnormalities. Many of the genetically determined epilepsies belonging to the latter category are actually associated diseases, some of which have specific inheritance mechanisms. These associated conditions may have mendelian inheritance, or they may be a derivative of chromosomal abnormality syndromes. The main mendelian disorders associated with epilepsy (Table 20.1) have already been covered in other chapters of this book (see Chapter 19); these are only mentioned briefly here.

In many circumstances, the distinction between idiopathic and symptomatic genetic epilepsies may be artificial. For example, genetic factors that are prominent in the etiology of the idiopathic epilepsies also reportedly influence the likelihood of developing posttraumatic epilepsy (Jennett, 1975). Some gene mutations may only decrease the seizure threshold, therefore increasing the individual’s susceptibility to environmental factors (Ottman, 1997). Ion channel mutations are another example. Although they are thought to affect neuronal excitability only, they can cause severe epileptic encephalopathies, such as occurs in Dravet syndrome, in which the mental retardation could suggest a symptomatic origin.

GENERAL CONSIDERATIONS

Clinical and electroencephalographic (EEG) studies in families have provided interesting evidence about the importance of the interaction between specific genetic factors and environmental determinants consistent with a polygenic inheritance model.

Studies on twins have demonstrated much higher concordance rates in monozygotic than in dizygotic pairs (Ottman, 1997). Even considering that most such studies have been performed without particular attention to the specific epilepsy type or syndrome, these rates overwhelmingly favored the presence of a shared genetic susceptibility in identical twins when idiopathic (“intact”) cases were considered (70% versus 6% of dizygotic twins) (Lennox, 1960).

First-degree relatives of individuals with epilepsy have an increased genetic susceptibility to show epileptiform EEG abnormalities and seizures (Andermann, 1982). For example, up to 50% of the siblings and offspring of patients with what has been termed benign rolandic epilepsy also have rolandic spikes; however, only 12% develop epilepsy (Bray and Wiser, 1965a, 1965b). The relatives of children with absence epilepsy (Metrakos and Metrakos, 1966) have also been reported to have a high incidence of generalized paroxysmal EEG abnormalities. Such an observation indicates that, in at least a minority of patients, nonspecific factors act by transforming the individual’s genetic predisposition into overt seizures. Annegers et al. (1982b) estimated that the standardized morbidity ratios for unprovoked seizures in the relatives of individuals with idiopathic childhood-onset epilepsy were 2.5 in siblings and 6.7 in offspring. More distant relatives did not have an increased risk of unprovoked seizures. A later study of the same patient population found that morbidity ratios for the offspring were 3.4 (Ottman, 1997). A study of siblings and children of individuals with symptomatic focal epilepsy came to complementary conclusions, as EEG abnormalities were present in 20%, seizures in 7%, and epilepsy in 4% (Andermann and Straszak, 1982). Therefore, the
inheritance of nonspecific predisposing factors may act by increasing the chances that acquired epileptogenic factors will cause seizures or epilepsy even in the relatives of patients with acquired lesions, though this occurs to a much lesser degree than that observed in the relatives of patients with idiopathic epilepsy.

TABLE 20.1. Main mendelian disorders and chromosomal abnormalities associated with epilepsy

Disorder	Gene	Chromosomal Mapping and/or Chromosome(s) Involved	Epilepsy (%)
Lafora disease	EPM2A	6q23.25	100
Unverricht-Lundborg disease	EPM1	21q22.3	100
Ceroid lipofuscinosis (recessive)	CLN1, CLN2, CLN3, CLN	1p32, 11p15.5, 513q21.1-32, 16p12.1	100
Galactosialidosis	GLB2	20q13.1	100
Gangliosidosis GM1 type 1	GLB1	3p21.33	100
Type III Gaucher disease	GBA	1q21	100
Aicardi	—	Xp22	100
Miller-Dieker/ILS	LIS1	17p13.3	100
XLAG and XL infantile spasms	ARX	X	100
Wolf-Hirschhorn syndrome (4p-)	—	4p-	100
Ring chromosome 20	—	20 ring	100
Ring chromosome 14	—	14 ring	100
Angelman syndrome	UBE3A	15q11-q13	90-100
Inv Dup (15) syndrome	—	15q tetrasomy	90-100
SBH/XL lissencephaly	DCX	Xq22.3	90-100
Periventricular nodular heterotopia	FLN1	Xq28	80-100
Tuberous sclerosis	TSC1/TSC2	9q24/16p13.3	60-100
Rett syndrome	MECP2	Xq28	70
Familial cavernous angiomas	unknown	7q	70
Alpha-thalassemia mental retardation	XH2	Xq13.3	45
Fragile X	FMR1	Xq27.3	28-45
Trisomy 21 (Down syndrome)	—	21	12-40
Abbreviations: ILS, isolated lissencephaly sequence; SBH, subcortical band heterotopia; XL, X linked; XLAG, X-linked lissencephaly with ambiguous genitalia.

The relatives of patients with early onset epilepsy have a higher risk of developing seizures than do the relatives of those with later onset epilepsy (Anderson et al., 1991), which is in keeping with the high frequency of epilepsies of genetic origin in young children. The possible presence of a gradient of risk in first-degree relatives was suggested by Lennox (1947), with the highest risk being present in the relatives of probands with an onset before 4 years of age. Other factors that enhance the risk of epilepsy in offspring include a mother who is affected, which confers a higher risk than an affected father (9% versus 3%), and an age at onset of epilepsy in the affected parent of younger than 20 years (Ottman, 1997).

Hauser et al. (1986) observed that the risk for epilepsy in relatives of probands with febrile convulsions is increased to an extent similar to that in the relatives of probands with epilepsy. A history of both febrile convulsions and epilepsy in the proband implies an even greater risk for relatives, suggesting a higher genetic load that favors the cooccurrence of both manifestations. However, potentially epileptogenic environmental and behavioral factors may be equally shared within the same family, making many studies of familial aggregation of limited use in estimating the specific role of genetic factors (Ottman, 1997).

IDIOPATHIC GENERALIZED EPILEPSIES WITH COMPLEX INHERITANCE

A multifactorial inheritance seems to be operative in most of the common forms of epilepsy. This type of genetic influence can act in conjunction with environmental factors in determination of the phenotype. Studies in which the specific epilepsy syndromes in probands and relatives have been taken into account have provided a more precise estimate of the magnitude of the genetic influences. A high incidence of epilepsy was observed in the siblings of probands with myoclonic astatic epilepsy (MAE) (13% to 20%), childhood absence epilepsy (CAE) (5% to 10%), and juvenile myoclonic epilepsy (JME) (5% to 7%) (Beck-Mannaggetta et al., 1989). The risk for offspring of individuals who had CAE, JME, and grand mal on awakening was estimated to be at about 7%. The offspring of individuals with a history of absence
epilepsy were also found to carry a higher risk than the offspring of parents with other types, not only for absence seizures but also for other seizure types in the study of Ottman et al. (1989).

The relatives of probands with specific idiopathic syndromes have a tendency to develop the same type of epilepsy (Beck-Mannaggetta and Janz, 1991), although not necessarily the same syndrome. In the Italian League Against Epilepsy concordance study (Italian League Against Epilepsy Genetic Collaborative Group, 1993), which was conducted on 72 families of probands with idiopathic generalized epilepsy in which more than three individuals were affected, multiple types of idiopathic generalized epilepsy were observed in 75% of the families. However, the concordance for the epilepsy syndrome in the identical twin pair study of Berkovic et al. (1998) was 94%.

A major difficulty that arises in the study of the genetic basis of the epilepsies derives from the poor correspondence between genotype and phenotype. Studies on monogenic epilepsies have discovered that reduced penetrance is common with epilepsy genes, thus implying that mutation carriers may be unaffected (Ottman, 1997). Reduced penetrance is even more difficult to demonstrate in polygenic epilepsies. Locus (or genic) heterogeneity (i.e., the possibility that mutations at different loci may underlie the same syndrome) implies that different families with the same syndrome may harbor mutations of different genes. For example, the generalized epilepsy with febrile seizures plus (GEFS+) spectrum of phenotypes has been linked to mutations of the SCN1A, SCN1B, and GABR1A genes in different families (Meisler et al., 2001). Allelic heterogeneity is observed when different alleles at the same locus cause different syndromes, so families or individuals with different syndromes have different mutations of the same gene. For example, many children with Dravet syndrome with mutations of the SCN1A gene have truncating mutations (Nabbout et al., 2003; Claes et al., 2001), whereas missense mutations of the same gene have been detected in individuals with milder phenotypes within the GEFS+ spectrum.

TABLE 20.2. Genetic epilepsy syndromes with simple inheritance (gene unknown)

Mode of Transmission		Linkage
Autosomal dominant
	Adolescent-onset idiopathic generalized epilepsy	8p12, 18q12, 5p
	Autosomal dominant cortical myoclonus and epilepsy	2p11.1-q12.2
	Autosomal dominant nocturnal frontal lobe epilepsy	15q24
	Autosomal dominant rolandic epilepsy with speech dyspraxia	—
	Benign familial infantile convulsions	19q, 16p, 2q24
	Benign rolandic epilepsy	15q24
	Childhood absence epilepsy	8q24
	Familial mesial temporal lobe epilepsy	—
	Familial adult myoclonic epilepsy	8q23.3-24.1
	Familial partial epilepsy with variable foci	22q11-q12
	Febrile seizures	8q13, 19p, 5q14-15
	Idiopathic generalized epilepsy	3q26, 14q23, 2q36, 16p12-p11
	Infantile convulsions and paroxysmal coreoathetosis	16p12-11.2
	Juvenile myoclonic epilepsy	6p21, 5q14
	Partial epilepsy with pericentral spikes	4p15
Autosomal recessive
	Familial idiopathic myoclonic epilepsy	16p13
	Rolandic epilepsy-exercise-induced dystonia-writer’s cramp	16p12-11.2

EPILEPSIES WITH SINGLE GENE INHERITANCE

Over the last several years, the recognition of families in which epilepsy phenotypes, which are comparable to those observed in the common idiopathic epilepsies but which are segregated in a mendelian fashion, has led to genetic linkage studies and the identification of a number of loci (Table 20.2). Causative genes, mostly those for the ion channels, have been identified for a number of these conditions (Meisler et al., 2001; Lerche et al., 2001) (Table 20.3). Although single gene epilepsies account for only about 1% of the epilepsy population (Ottman, 1997), they have been instrumental to understanding the molecular genetic mechanisms of epileptogenesis that may also be operative in the more common idiopathic epilepsies with complex inheritance. Some nonprogressive syndromes of epilepsy associated with other paroxysmal neurologic disorders, such as myoclonus and dystonia (Guerrini et al., 1999b, 2001b; Szepetowski et al., 1997), suggest that the
same genetic mechanisms of channel dysfunction known to be operative in the single gene epilepsies may be part of the cause in these cases as well.

TABLE 20.3. Known idiopathic epilepsy genes in 2003

Gene		Function	Locus	Inheritance/MIM	Type of Mutations	Epilepsy Syndrome	Seizure Types
GABRA1		Partial inhibition of GABA-activated currents	5q34	AD/606904	Missense	AD JME	TCS, myoclonic, absence
	GABA_A α₁-receptor subunit	Partial inhibition of GABA-activated currents
GABRG2		Rapid Inhibition of GABAergic neurons	5q31	AD/604233	Missense, truncation	FS, CAE, GEFS+	Febrile, absence, TCS, myoclonic, clonic, partial
	GABA_A receptor γ subunit	Rapid Inhibition of GABAergic neurons
SCN2A		Fast sodium influx initiation and propagation of action potential	2q24	AD/604233	Missense	GEFS+ BFNIC	Febrile, afebrile generalized tonic and TCS
	Sodium channel α₂ subunit						Febrile, afebrile generalized tonic and TCS
SCNIA		Somatodendritic sodium influx	2q24	AD/604233	Missense	GEFS+ SMEI	Febrile, absence, myoclonic, TCS, partial
	Sodium channel α₁ subunit	Somatodendritic sodium influx
SCNIB		Coadjuvate and modulate a subunit	19q13	AD/604233	Missense	GEFS+	Febrile, absence, tonic-clonic, myoclonic
	Sodium channel β₁ subunit	Coadjuvate and modulate a subunit					Febrile, absence, tonic-clonic, myoclonic
KCNQ2		M current interacts with KCNQ3	20q13	AD/602235	Missense	BFNC	Neonatal convulsions
	Potassium channel
KCNQ3		M current interacts with KCNQ2	8q24	AD/121201	Missense	BFNC	Neonatal convulsions
	Potassium channel	M current interacts with KCNQ2
CHRNA4		Presynaptic; nicotinic current modulation; interacts with β₂ subunit	20q13	AD/600513	Missense	ADNFLE	Sleep related focal seizures
	Acetylcholine receptor α₄ subunit
CHRNB2		Presynaptic; nicotinic current modulation; interacts with α₂ subunit	1p21	AD/605375	Missense	ADNFLE	Sleep related focal seizures
	Acetylcholine receptor β₂ subunit
LGI1		Disregulates homeostasis, interactions between neurons and glia?	10q24	AD/600512	Missense	ADPEAF	Partial seizures with auditory or visual hallucinations
	Leucine-rich, glioma activated						Partial seizures with auditory or visual hallucinations
CLCN2		Neuronal chloride efflux	3q26	AD	Stop codon, splicing, missense	IGEs	TCS, myoclonic, absence
	Voltage-gated chloride channel				Stop codon, splicing, missense
Abbreviations: AD, autosomal dominant; ADNFLE, autosomal dominant nocturnal frontal lobe epilepsy; ADPEAF, autosomal dominant partial epilepsy with auditory features; BFNC, benign familial neonatal convulsions; BFNIC, benign familial neonatal-infantile convulsion; GEFS+, generalised epilepsy with febrile seizures plus; MAE, myoclonic astatic epilepsy; SMEI, severe myoclonic epilepsy of infancy; TCS, tonic-clonic seizures; XL, X linked.
Modified from Guerrini et al., 2003b.

Generalized Epilepsy with Febrile Seizures Plus and Mutations of the Voltage-Gated Sodium Channel Subunit Genes

The collection of large dominant pedigrees with about 60% penetrance that feature heterogeneous epilepsy phenotypes, including febrile seizures (FSs) that often persist beyond the age of 6 years (FS+), generalized tonic-clonic (GTC) seizures, absence, and myoclonic seizures, led to the concept of GEFS+ (Wallace et al., 1998; Scheffer and Berkovic, 1997). These types of seizures may occur in isolation, most often in FS+ individuals who have no other seizure types, or they may be combined in variable phenotypes that are often mild but that, in very rare cases (Berkovic and Scheffer, 2001), may also include intractable syndromes, such as MAE and severe myoclonic epilepsy (Dravet syndrome).

A locus for the GEFS+ spectrum has been mapped to chromosome 19q13 (Wallace et al., 1998) to which the voltage-gated sodium channel β₁ subunit gene (SCN1B) had been assigned. A missense mutation of this gene (C121W) has been found in two unrelated families (Wallace et al., 1998, 2002). In vitro functional studies (voltage-clamp recording on Xenopus laevis oocytes) showed changes consistent with the loss of function. In particular, the mutant protein failed to accelerate the recovery of the associated α subunit from inactivation (Wallace et al., 1998). Coexpression of mutant and wild type β subunits with the α subunits caused an intermediate inactivation rate (Moran and Conti 2001), which arose from the competitive binding of the inactive mutant subunit with the α subunit, accounting for dominant inheritance. In heterozygotes, the association of inactive β subunits with α subunits generates a persistent sodium current, rendering the neurons hyperexcitable and making them apt to initiate firing with small depolarizations (Meisler et al., 2001).

Two GEFS+ families with linkage to chromosome 2q24-33 were found to harbor missense mutations of the sodium channel α subunit gene SCN1A (Escayg et al., 2000). Both mutations affected highly conserved residues coding for the putative voltage sensor of the transmembrane region of the channel. Functional studies have shown that these mutations have different functional consequences. One (R1648H) accelerated the recovery from inactivation, with consequent neuronal hyperexcitability. The other (T875M) increased the slow inactivation mode, with consequent reduction of the whole channel protein accessible for opening (Escayg et al., 2000).

The most common mutations found in GEFS+ families are estimated to be those involving SCN1A, which accounted for 5.6% of the cases in a large series (Escayg et al., 2000).

An R187W mutation of the SCN2A gene has been reported in a single family (Sugawara et al., 2001).

The γ-aminobutyric acid A (GABA_A) receptor γ₂ subunit gene has also been shown to be involved in the pathogenesis of GEFS+ (Baulac et al., 2001; Wallace et al., 2001a), confirming the locus heterogeneity for this spectrum of epilepsy phenotypes.

Dravet Syndrome (Severe Myoclonic Epilepsy of Infancy) and Mutations of the Voltage-Gated Sodium Channel Subunit Genes

In severe myoclonic epilepsy of infancy (SMEI) (Dravet et al., 1992a), the relatives of the affected individuals have an elevated incidence of epilepsy (Benlounis et al., 2001; Singh et al., 2001), although a clear familial distribution for the syndrome is rare. The association of seizures with febrile episodes suggested that it may be analogous to GEFS+ and prompted mutation screening for SCN1A in a small series of affected children (Claes et al., 2001). All were shown to carry de novo frameshift or nonsense mutations leading to null alleles with complete loss of function. In two subsequent studies, Ohmori et al. (2002) detected mutations of the SCN1A gene in 24 of 29 patients with SMEI and Sugawara et al. (2002) in 10 of 14; these included deletion, insertion, missense, and nonsense mutations. These were de novo mutations because none of the parents were carriers. In a large series including approximately 100 patients, mutations of the SCN1A gene were observed in one-third of the children and, surprisingly, in a few unaffected or mildly affected parents as well (Nabbout et al., 2003). The variable frequency of mutations in various series and the high frequency of epilepsies of various types that is found in the families of these patients are as yet unexplained (Fujiwara et al., 2003) The genotype—phenotype correlation in this syndrome is probably much more complex than has been previously thought, and mutations in additional genes or strong environmental factors may be needed for the phenotype to manifest. SMEI has also been observed in association with an inherited GABRG2 mutation within a GEFS+ pedigree (Harkin et al., 2002).

Missense mutations in three sodium channel genes (SCN1A, SCN1B and SCN2A) most often cause the same spectrum of usually mild epilepsy phenotypes,
while loss-of-function mutations of one of these genes (SCN1A) usually causes a more severe phenotype. However, the phenotypes associated with missense mutations of these genes may sometimes be severe, including some cases with Dravet syndrome or a picture reminiscent of MAE. Genetic heterogeneity may result from either variable expression of some genes or allelic heterogeneity (Meisler et al., 2001).

Only gold members can continue reading. Log In or Register to continue