Genetics of EEG and Epilepsy

James D. Geyer

Paul R. Carney

L. John Greenfield Jr

INTRODUCTION

A number of epilepsy syndromes and their associated electroencephalographic findings have genetic linkages. An understanding of these EEG findings and their genetic underpinnings can assist with appropriate interpretation, diagnosis, and management. In addition to the EEG patterns associated with certain epilepsy syndromes, there are EEG features that can be inherited as a trait without associated clinical findings. It is important to recognize that a particular EEG feature may not correspond to a specific clinical syndrome. Moreover, in patients with the same genetic mutation, even within the same family, there can be a wide phenotypic variability in terms of seizure semiology, severity, and comorbid features. Genetics can also influence the response to specific medications and the risk of seizure-related death.

While many of the typical findings and “normal variants” discussed in an EEG report may have a genetic basis, we do not fully understand the inheritance patterns of background EEG activities including the alpha rhythm, mu rhythm, and sawtooth waves, among others. Indeed, the fact that these features are “normal,” found in the majority of the population, makes it difficult to determine what genes are responsible for their properties. The appearance of uncommon EEG features associated with neurologic diseases and their segregation within families and individuals provide the basis for discovering what genes are involved in both the electrographic presentation and the clinical syndrome.

MECHANISMS OF GENETIC EPILEPSIES

Epilepsy can result from problems in a wide variety of genes. Those defects arise from various errors in DNA replication, including chromosomal rearrangements such as ring chromosomes, balanced or unbalanced translocations, monosomies and trisomies, as well as single nucleotide mutations that result in missense (with substitution of an incorrect amino acid or premature termination) or insertion of a nucleotide resulting in frameshift or nonsense mutations. Replication errors can result in copy number variations (CNV) in which multiple copies of a gene are present, sometimes causing a gain in function, or gene deletions in which a gene is entirely or functionally absent. Some epilepsies result from single-gene mutations, particularly those involving ion channels, such as the sodium channel SCN1A mutations associated with Dravet syndrome (severe myoclonic epilepsy of infancy).¹ In addition to the channelopathies, other gene families associated with epilepsy are those regulating neurotransmission (on either side of the synapse) and second messenger systems, as well as transcription factors like ARX²^,³ that are essential for normal brain development, and other “housekeeping” genes. Alternatively, some epilepsies are not dependent on a single gene but have a polygenic heritable component that determines epilepsy susceptibility. In many cases, developing epilepsy may require a “second hit” such as fever, infection, or brain injury that triggers epileptogenesis.⁴ Genetic abnormalities associated with epilepsy can be dominant or recessive, autosomal, X-linked, or mitochondrial and may be inherited or arise de novo during meiosis or as a somatic mutation. Gene regulation by methylation, mosaicism, and other epigenetic modifying factors may explain the marked phenotypic diversity in patients in the same family with the same genetic lesion.⁵

GENETICS IN EPILEPSY AND SEIZURE CLASSIFICATION SYSTEMS

The classification of seizure disorders uses clinical seizure semiology, comorbid symptoms, and both interictal and ictal EEG findings to categorize epileptic events into three categories: focal, generalized (convulsive or nonconvulsive), and unclassified.⁶^,⁷ The updated categorization proposed by the ILAE in 2017⁸^,⁹ creates a threetiered classification system based on seizure type (focal, generalized, or unknown), epilepsy type (focal, generalized, combined generalized and focal, or unknown), and epilepsy syndromes. A complete diagnosis involves all three levels and emphasizes the importance of the underlying etiology of the epilepsy including structural, genetic, infectious, metabolic, immune, or unknown causes.

The 2017 system recognizes the important genetic contribution to many epilepsy syndromes. In particular, the generalized epilepsies previously considered “cryptogenic” or idiopathic, having normal EEG background (with superimposed 3 Hz or faster generalized spike-and-wave) and relatively normal cognition, are now recognized as “genetic generalized epilepsies” (GGEs) even though a specific genetic etiology is known for very few of these disorders. The GGEs include childhood absence epilepsy, juvenile absence epilepsy, juvenile myoclonic epilepsy, and generalized tonicclonic seizures alone (formerly generalized tonic-clonic seizures on awakening).

Another major category of genetic epilepsies are the developmental and epileptic encephalopathies, formerly known as symptomatic generalized epilepsies, which include disorders such as West syndrome and other causes of epileptic (infantile) spasms, Lennox-Gastaut syndrome, myoclonic-astatic epilepsy, and myoclonic absences. For these disorders, the EEG background may show generalized, often, rhythmic spike-and-slow-wave discharges at <3 Hz, and background slowing is more common.

While both benign IGE/GGE and epileptic encephalopathies are linked to mono- or polygenic etiology, it is important to remember that “genetic” does not necessarily mean “inherited,” since epilepsy-causing mutations frequently arise de novo during meiosis, and there may be no family history of epilepsy. Focal epilepsies are less likely to be of genetic etiology since a focal seizure onset suggests a localized structural defect (stroke, tumor, arteriovenous malformation, etc.), though many epilepsies presenting with focal seizures may have an underlying genetic etiology that produces a focal lesion or localized network problem (tuberous sclerosis, focal cortical dysplasias, autosomal dominant nocturnal frontal lobe epilepsy [ADNFLE], etc.). Hence, obtaining a thorough family history is always important in the workup for epilepsy. On the other hand, the absence of a family history of epilepsy does not mean that there is not a genetic etiology, and genetic testing can sometime reveal new approaches to disease management as well as the ramifications for childbearing.

REVIEW

21.2: Which of the following statements about genetic epilepsies is NOT true?

a. West and Lennox-Gastaut syndromes are now considered developmental and epileptic encephalopathies.

b. Genetic epilepsies require a familial pattern of inheritance.

c. Childhood absence and juvenile myoclonic epilepsies are now among the genetic generalized epilepsies.

d. The ILAE 2017 classification categorizes epilepsies by seizure type, epilepsy type, and etiology.

e. Focal epilepsies are less likely to be genetic in etiology.

View Answer

21.2: b. Genetic epilepsies result from gene abnormalities that may be inherited, but many arise as de novo mutations without a family history.

EPILEPSY POPULATION GENETICS

Epilepsy affects about 1% of the population worldwide by age 20 and about 3% by age 75.¹⁰ Twin studies suggest that the heritability of epilepsy ranges from about 25 to 75%,⁵ and the probandwise concordance rate in monozygotic twins is 37%.¹¹ Similar rates of heritability were found in an analysis of 4 million common single nucleotide polymorphisms (SNPs) in the genomes of 1258 UK patients with epilepsy (958 with focal epilepsy) and 5129 population control subjects.¹² Total heritability was 32% for all epilepsy, 23% for focal epilepsy, and 36% for nonfocal epilepsy. The relative risk of developing epilepsy is increased two- to fourfold in first-degree relatives of people with epilepsy of unknown cause, including the idiopathic generalized epilepsies and nonlesional focal epilepsies.¹³

Febrile Seizures

Febrile seizures (FS) are a prime example of the complex interplay between genes and environment. FS are the most common seizure type in childhood, occurring in neurologically normal infants and children between the ages of 3 months and 5 years.¹⁴ Approximately 4% of children will have at least one FS by the age of 7 years.¹⁵ In the United States, the prevalence of febrile seizures in African American children is 4.2% vs 3.5% in Caucasian children.¹⁵ The rate is even higher in the Japanese (6%-9%)¹⁶ and Pacific Islander (14%-15%)¹⁷ populations. Febrile seizures are slightly more common in boys than girls.

The pathophysiology of febrile seizures remains incompletely understood. Febrile seizures arise from a complex interaction of immature brain development, fever, and genetic predisposition. Genetic factors play a role in febrile seizure susceptibility, but the mode of inheritance in most cases is unknown. Possible inheritance
mechanisms include polygenic, autosomal recessive, and autosomal dominant with reduced penetrance. A positive family history for febrile seizures can be found in 25%-40% of patients with febrile seizures, and the reported frequency in siblings of children with febrile seizures has ranged from 9% to 22%.¹⁸

FS are usually benign and resolve spontaneously without antiseizure drug (ASD) therapy, with no lasting neurologic impact. However, febrile seizures recur in approximately one-third of children who experience a first febrile seizure. About 2%-7% of children with FS may have subsequent or spontaneous epileptic seizures later in life, which is 2-10 times more than the general population.¹⁹ Despite the fact that the vast majority of patients with FS do not go on to develop epilepsy, up to 10%-15% of people with epilepsy report having had FS.²⁰ Age, family history, duration of illness, and temperature at the time of the seizure serve as the primary predictors of recurrence.²¹ These risk factors can be combined to create a useful prediction scheme. Patients with none of the four risk factors (age <18 months, family history of febrile seizures, low temperature at the time of the seizure, and short duration of illness) have a 4% risk of recurrence, with one factor 23%, with two 32%, with three 62%, and with all four 76%.²¹

Some patients with FS had additional seizure types and a stronger family history of FS or seizures. Linkage on chromosomes 2q and 19q associated with the phenotype of febrile seizures with generalized epilepsy (tonic-clonic, absence, and myoclonic) suggested evidence of sodium channel involvement. By 2004, at least 5 different genes had been linked to epilepsy syndromes, which include febrile seizures,²² and by 2018 that number was up to 25.¹⁴ Many of these gene defects are associated not only with febrile seizures but also seizures induced by fever past the age of 6, as well as spontaneous generalized seizures, with or without associated encephalopathy, producing a syndrome known as generalized epilepsy with febrile seizures plus (GEFS+). GEFS+ most commonly arises via autosomal dominant defects in voltage-gated sodium channel subunits (SCN1B, SCN1A, and SCN2A) or a defect in the gamma2-subunit of the GABA_A receptor.²³ These channel-related syndromes will be discussed in more detail below.

REVIEW

21.3: Which of the following is TRUE about febrile seizures?

a. A positive family history of febrile seizures is present in >50% of cases.

b. Most children with febrile seizures go on to have epilepsy later in life.

c. About 10%-15% of adult epilepsy patients report having had febrile seizures.

d. Children with febrile convulsions should be treated with antiseizure medications.

e. Predictors of febrile seizure recurrence include age, family history, and high temperature at the time of seizure.

View Answer

21.3: c. About 10%-15% of adult epilepsy patients report having had febrile seizures in childhood. A positive family history of febrile seizures is present in 25%-40% of cases. Only 2%-7% of children with febrile seizures go on to have epilepsy later in life. Children with febrile convulsions do not benefit from antiseizure medications, which may cause cognitive adverse effects and may not prevent febrile seizures effectively. Predictors of febrile seizure recurrence include age, family history, and low temperature at the time of seizure.

SYSTEMATIC APPROACHES TO EPILEPSY GENE IDENTIFICATION

The Epilepsy Phenome/Genome Project (EPGP) was an NIH-funded study to identify genes that influence the development of epilepsy and responses to treatment, involving 25 major epilepsy centers in the United States, Australia, and Argentina. The EPGP study enrolled 4199 patients and their family members with the aim to identify genetic variants of common forms of epilepsy and determine genetic influences in rare severe epilepsies. A subsequent study known as Epi4K²⁴ analyzed the genomes of 4000 subjects with well-characterized epilepsies to identify their genetic associations, mostly involving mutations in ion channel and neurotransmitter receptor genes, as well as de novo mutations and copy number variants in epileptic encephalopathies. The approach used whole exome or whole genome “trio sequencing” (patient/proband and parents), many of whom also participated in EPGP, particularly families with more than one affected individual. In a whole exome screen for de novo mutations among patients/parents with infantile spasms (n = 149) and Lennox-Gastaut syndrome (n = 115), there were 329 confirmed de novo mutations,²⁵ including 4 patients with mutations in the GABA_A receptor beta-3 subunit (GABRB3) and 2 patients with the same de novo mutation in ALG1, a mannosyltransferase involved in protein glycosylation. Other genes with de novo mutations in this cohort included CACNA1A, CHD2, FLNA, GABRA1, GRIN1, GRIN2B, HNRNPU, IQSEC2, MTOR, and NEDD4L. The de novo mutations were frequently among genes regulated by the fragile X protein. In another Epi4K study, genome sequencing revealed an excess of ultrarare mutations among patients with epilepsy.²⁶

Even those who are very familiar with genetics will probably not recognize all of these genes. The next section will help you decode this alphabet soup by introducing some of the major epilepsy genes.

GENES ASSOCIATED WITH EPILEPSY, AND WHERE TO FIND THEM: A GENETIC BESTIARY

Many of the monogenic epilepsy syndromes result from mutations in ion channels.¹³ This is not surprising, since epilepsy is a disease of neuronal excitability, and ion channels are the molecular devices that allow neurons to become excitable and pass electric signals from one to another. The ion channel genes currently known to be associated with epilepsy are listed in Table 21.1, along with their genetic loci and epilepsy syndromes. As discussed in Chapter 1, there are two main types: voltage-gated ion channels that open in response to a change in transmembrane voltage and ligand-gated ion channels that open in response to binding a neurotransmitter or other chemical signal. The voltage-gated channels can be further divided into those that conduct

sodium, potassium, or calcium ions. Mutations in all three of these voltage-gated ion channel families are responsible for specific types of epilepsy or developmental syndromes in which seizures are a prominent feature. Among the ligand-gated channels, epilepsy-related mutations have been found in excitatory N-methyl-D-aspartate (NMDA) glutamate receptor subunits, inhibitory GABA_A receptor subunits, and presynaptic nicotinic acetylcholine receptor (nAChR) subunits. Mutations at different positions on the same channel subunit gene can cause very different clinical syndromes. The effects of the mutations depend on what specific channel is affected and where they are in the channel structure, since channel subtypes are highly specialized in their cellular and subcellular localization and biophysical actions.

TABLE 21.1 Ion channel genes associated with epilepsy^a

Gene	Protein	Locus	Phenotype	OMIM
Voltage-Gated Channels
Sodium Channels
SCN1A	Na_V1.1	2q24	Dravet syndrome; GEFS+	182389
SCN1B	Na_Vb1	19q13	GEFS+, TLE, EIEE	600235
SCN2A	Na_V1.2	2q24	BFNIE, EEE, NDD, BFIS3	182390
SCN3A	Na_V1.3	2q24.3	Familial focal epilepsy, EIEE	617935, 38
SCN8A	Na_V1.6	12q13.13	BFIS5, EE	600702
Potassium Channels
KCNA1	K_V1.1	12p13	Partial epilepsy and episodic ataxia	176260
KCNA2	K_V1.2	1p13.3	Early infantile epileptic encephalopathy	176262
KCNB1	K_V2.1	20q13.13	Early infantile epileptic encephalopathy	600397
KCNC1	K_V3.1	11p15.1	Progressive myoclonus epilepsy, EPM7	616187, 176258
KCNJ10	K_IR 4.1	1q23.2	SeSAME syndrome	612780
KCNMA1	KCal.1	10q22	Epilepsy, paroxysmal dyskinesia	600150
KCNMB3	K_VCaβC	3q26.32	Juvenile absence epilepsy
KCNQ2	K_V7.2	20q13.3	BFNE, epileptic encephalopathy	602235
KCNQ3	K_V7.3	8q24	BFNE	602232
KCNT1	KNal.1	9q34.3	ADNFLE, EIMFS	608167
KCTD7	KCTD7	7q11.21	Progressive myoclonus epilepsy, EPM3	611725
HCN1	HCN1	5p12	IGE, GEFS+, EIEE	602780
Calcium Channels
CACNA1A	CaV2.1	19p13	Epilepsy, episodic ataxia, EE	601011
CACNA1H	CaV3.2	16p13.3	GGE, CAE	607904
Ligand-Gated Receptor Channels
NMDA Receptor Subunits
GRIN1	GluN1	9q34.3	Epileptic encephalopathy	138249
GRIN2A	GluN2A	16p13.2	Epileptic encephalopathy	138253
GRIN2B	GluN2B	12p13.1	Epileptic encephalopathy	138252
GRIN2D	GluN2D	19q13.33	Epileptic encephalopathy	602717
GABA_A Receptor Subunits
GABRA1	GABA_AR α1	5q34	GGE, epileptic encephalopathy	137160
GABRB3	GABA_AR β3	15q12	CAE, epileptic encephalopathy	137192
GABRG2	GABA_AR γ2	5q34	FS/GEFS+, epileptic encephalopathy	137164
Nicotinic Acetylcholine Receptor Subunits
CHRNA2	NAChR α2	8p21	ADNFLE, BFIS6	118502
CHRNA4	NAChR α4	20q13.33	ADNFLE	118504
CHRNB2	NAChR β2	1q21	ADNFLE	605375
^aModified from Oyrer J, Maljevic S, Scheffer IE, Berkovic SF, Petrou S, Reid CA. Ion channels in genetic epilepsy: from genes and mechanisms to disease-targeted therapies. Pharmacol Rev. 2018;70:142-173. doi: 10.1124/pr.117.014456.
ADNFLE, autosomal dominant nocturnal frontal lobe epilepsy; BFIE, benign familial infantile epilepsy; BFNIE, benign familial neonatal-infantile epilepsy; BFIS3/5/6, benign familial infantile seizures type 3, type 5 or type 6; EE, epileptic encephalopathy; EEE, early epileptic encephalopathy; EIEE, early infantile epileptic encephalopathy; EIMFS, epilepsy of infancy with migrating focal seizures; FS, febrile seizures; GEFS+, generalized epilepsy with febrile seizures plus; GGE, genetic generalized epilepsy; GOF, gain of function; LOF, loss of function; NDD, neurodevelopmental disorder; OMIM, Online Mendelian Inheritance in Man; SeSAME, epilepsy, ataxia, sensorineural deafness, and tubulopathy (renal electrolyte imbalance) syndrome.

The gene descriptions that follow come primarily from online resources, particularly the Online Mendelian Inheritance in Man (OMIM) website, omim.org. This is the latest implementation of a database originated by Dr. Victor McKusick at Johns Hopkins University in the early 1960s and is maintained by the McKusick-Nathans Institute of Genetic Medicine at JHU. Since the pace of discovery is swift and the field ever changing, these listings are illustrative only, and the reader is strongly encouraged to consult the updated online resources to address clinical questions.

Voltage-Gated Sodium Channels

The voltage-gated sodium channels are essential for the propagation of action potentials. They are composed of a large alpha subunit with six transmembrane segments that form the ion channel pore, the voltage sensor that enables the channel to open in response to depolarization, and an inactivation gate that rapidly closes the channel after opening. Resetting the inactivation gate so the channel can open again requires repolarization. There are nine alpha subunits (Na_V1.1 to Na_V1.9, encoded by genes SCN1A to SCN11A, with the 6A and 7A genes belonging to a different family), each with slightly different biophysical properties and expression patterns.¹³ While the alpha subunit alone is sufficient for sodium channel function, it is usually associated with two beta subunits (from four beta subunit genes, SCN1B-SCN4B) that modulate sodium channel behavior. More than 800 sodium channel mutations have been described in patients with epilepsy.¹³

The SCN1A sodium channel encoding Na_V1.1 is expressed primarily at the axon initial segment of inhibitory GABAergic interneurons and is one of the most commonly mutated channels associated with epilepsy with several hundred known mutations.¹³ Most of these produce a syndrome of febrile generalized tonic-clonic seizures “plus” additional seizure types including absence, myoclonic, or focal seizures, which as noted above is known as GEFS+. The more severe variants produce Dravet syndrome, which begins with febrile convulsions at about age 6 months followed by multiple seizure types, developmental delay or regression, and gait disturbance. A normally developing child develops either focal-onset or generalized clonic seizures. The initial seizures may be considered complex febrile seizures because they usually occur in association with fever, immunization, or infection. Over time, seizures progress and become more frequent and prolonged and occur both with and without fever. Myoclonic jerks appear by 2-5 years of age and eventually worsen, becoming refractory to antiepileptic treatment. Global developmental regression appears, along with ataxia. The disorder is difficult to treat but does respond to cannabidiol.

Dravet syndrome is usually sporadic, arising from de novo mutations. Some mutations result in channel truncation and loss of function, but gain-of-function mutants have also been reported. The EEG in GEFS+ typically shows generalized spike-and-wave discharges; those with Dravet will initially have a normal EEG but eventually develop an abnormal background with generalized spike-wave complexes. Mutations in the SCN1B beta subunit are also associated with GEFS+, and homozygous mutant alleles can cause early infantile epileptic encephalopathy (EIEE) consistent with Dravet syndrome.

SCN2A is expressed mainly at the axon initial segment of excitatory neurons, where it may play a role in backpropagation of depolarizing signals into the soma and dendrites.¹³ Mutations in SCN2A have been associated with benign familial neonatal-infantile seizures, a self-limited epilepsy syndrome of the first year of life. However, other mutations in this channel are associated with early-onset epileptic encephalopathies with multiple seizure types, including Ohtahara syndrome, epileptic spasms of West syndrome, Lennox-Gastaut syndrome, and a Dravet-like syndrome, all with prominent neurodevelopmental symptoms including autism, intellectual disability, and schizophrenia. Both gain- and loss-of-function mutations occur. Mutations in SCN3A are less common and can cause a familial focal epilepsy or EIEE.

SCN8A encodes Na_V1.6, found throughout the brain in the distal axon initial segment associated with action potential initiation, as well as at nodes of Ranvier in myelinated axons where they are critical for saltatory conduction. Heterozygous mutations result in epileptic encephalopathy with developmental delay and intractable seizures by age 18 months, including epileptic spasms, GTC seizures, and absence and focal seizures. Most of the mutations are missense resulting in gain of function and increased excitability.

Voltage-Gated Potassium Channels

Voltage-gated potassium channels are composed of 4 subunits, each with 6 transmembrane domains, selected from about 40 different genes.¹³ Two of the transmembrane segments (S5 and S6) form the ion channel pore, and the other four contribute to voltage sensing and gating. Since the potassium concentration is higher inside the cell than outside, K⁺ ions generally flow out when the channel opens, repolarizing the cell and helping to terminate action potentials.

The KCNA1 gene encodes K_V1.1, which is broadly expressed in the CNS, particularly in the hippocampus at axon initial segments and nodes of Ranvier.¹³ Mutations in KCNA1, mostly causing loss of function and broadening of the action potential, cause episodic ataxia type 1 (EA-1) with attacks of limb ataxia lasting 1-2 minutes and myokymia (involuntary twitching) of the face and limbs. Seizures are 10-fold more likely in patients with EA-1 than the general population. Treatment includes acetazolamide in addition to conventional ASDs.

K_V1.2, encoded by KCNA2, is a “delayed rectifier” channel (like KCNA1 and a few others) that remains open for a while even after the membrane potential repolarizes, allowing K⁺ ions to pass back into the cell. This helps restore the resting membrane potential after efflux of K⁺ with action potentials. KCN2A mutations cause EIEE, with seizure onset between 5 and 17 months, both with and without fever, including myoclonic, myoclonic-atonic, absence, focal or hemiclonic, and generalized convulsive seizures, as well as intellectual disability with delayed speech and severe ataxia.¹³ Other mutations cause milder familial epilepsies, hereditary spastic paraplegia, and ataxia. Both loss- and gain-of-function mutations occur. Mutations in KCNB1, which encodes K_V2.1, a similar delayed rectifier channel, also cause EIEE.

KCNC1 encodes K_V3.1, which is highly expressed in fast-spiking inhibitory GABAergic interneurons as well as cerebellar granule neurons.¹³ De novo mutations are a major cause of progressive myoclonic epilepsy, which presents with myoclonus, GTC seizures, and progressive neurologic deterioration.

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