Vol. XXIII No. 1
Editor: Heather J. Stalker, M.Sc.
Director: Roberto T. Zori, M.D.
By: Charles A. Williams1, Agatino Battaglia2
Introduction
Epilepsy is one of the most common brain disorders with one in 10 individuals having at least one seizure during his/her lifespan, and about 30% of these developing epilepsy. Epilepsy occurs for a variety of reasons, and can occur due to an environmental event alone without any genetic involvement. Other seizures occur without apparent cause and have a chronic course with recurrent unprovoked episodes, and these seizures are classified in the group of common epilepsies. The common epilepsies can be distinguished from epilepsy disorders that have single gene causation or have a syndromic component associated with brain or other clinical anomalies. The epilepsies have a population prevalence of 0.5 to 1% (Hauser et al., 1996).
The epilepsies can be classified in many different ways, but the most common classification comes from the International League Against Epilepsy (ILAE). The most recent report classifies epilepsies into three broad groups: 1. Genetic (includes single gene causes without structural brain anomalies or syndromic associations). 2. Structural-metabolic (applied to syndromic disorders such as Tuberous Sclerosis or Rett syndrome in which there are syndromic features or pervasive brain anomalies. 3. Unknown cause (Berg et al., 2010).
The biological underpinnings of epilepsy are extremely heterogeneous. The common thread is the presumption that each of these mechanisms disturbs the balance between excitation and inhibition of neuronal circuits.
Historically, epilepsy was believed to be due to dysfunction of ion channels, but over the past decade new genes have emerged which indicate many different mechanisms for epilepsy (Poduri and Lowenstein, 2011). In the first category is the “major” epilepsy gene SCN1A, the most well described channelopathy gene. In the second category, we discuss the ARX disorders which include seizure phenotypes as well as intellectual impairment and/or severe developmental brain malformations, and the chromosome abnormalities associated with seizures such as the classical Wolf-Hirschhorn syndrome (Battaglia et al., 2009), the 1p36 deletion syndrome (Bahi-Buisson et al., 2008b), variants within 15q11.2-q13 (Battaglia, 2008; Battaglia and Guerrini, 2005) and the more subtle or more newly discovered copy number variants which confer varying degrees of epilepsy risk.
Although we will discuss discrete genes associated with epilepsies, the most compelling issue in understanding the genetic etiology for the common epilepsies involves multifactorial inheritance. This concept is crucial in thinking about the polygenic origin of the common epilepsies and also for our understanding of how single gene defects operate in our genomic constitution and in the midst of environmental exposures.
Multifactorial Inheritance
For many decades it was known that the common epilepsies demonstrated multifactorial inheritance, meaning the involvement of both genetic and environmental factors. This inheritance pattern results in higher recurrence risk among siblings ( 3-5%), compared to the <1% risk for the normal population (Anderson et al., 1991; Hemminki et al., 2006), and a higher degree of concordance of seizures in monozygotic compared to dizygotic twins (Berkovic et al., 1998; Kjeldsen et al., 2001). There is a strong genetic influence, and heritability is estimated to be approximately 70% (Hemminki et al., 2006). A crucial part of multifactorial inheritance model is the threshold effect (Fig 1). In this model, a population has a distribution of risk genes that in combination with environmental factors lead to a threshold effect whereby the disease phenotype becomes expressed (Lobo, 2008; Turnpenny and Ellard, 2004).

Polygenic Effects
New molecular tools are now helping to better appreciate how multiple genes contribute to the epilepsy disorders. These tools include genome wide association studies (GWASs) which detects subtle gene variations or polymorphisms that may associate with the disease under study.
GWAS studies looking for gene risks with >1.3 odds ratio have not identified
any gene with this degree of risk effect (Kasperaviticiute et al., 2010). One interpretation of these data is that the contribution of risk genes may be more subtle, conferring risks <1.3. The study of small deletions/duplications known as copy number variants (CNVs) has been more productive in identifying genomic changes that are clearly associated with epilepsy and other developmental conditions. These studies have identified deletion or duplication regions particularly involving chromosome regions 16p13.11, 15q13.3 and 15q11.2 (de Kovel et al., 2010; Dibbens et al., 2009; Heinzen et al., 2010; Masurel-Paulet et al., 2010; Mefford et al., 2010; van Bon et al., 2009). Evaluation of family histories indicate that often a parent inherits one of the same CNVs but does not demonstrate any seizure phenotype, indicating that these CNV’s serve as only risk factors but the disease expression is modulated presumably by other aspects of one’s genetic constitution and/or yet to be identified environmental effects. It is also significant that most CNV’s that are enriched in epilepsy cohorts have failed to find any significant risk genes. It could be that the epilepsy-linked CNVs confer risk by haploinsufficiency of single or multiple genes or by relative large increases in gene dosage through the deletion or duplication event, each mechanism leading to large differences in the amount of expressed protein.
Environmental effects
Little progress has been made in identifying specific environmental factors that modulate or interact with these risk genes, so this area remains largely a “black box” problem (Berg et al., 2010). Environmental effects are well known to induce seizures and include the effects of infectious diseases, tumors, autoimmune disorders, toxic exposures and traumas. Although many of these events are associated with abnormalities in certain genetic or molecular pathways, it has not been possible to implicate these related environmental factors as contributors to the general heritable problem of idiopathic epilepsy in the population.
Taken together the molecular studies illustrate how multiplex the problem of epilepsy is likely to be. These studies stand in contrast to the “single gene” epilepsy’s in which established mutations lead predictably to disease or lead to an established high risk for the seizure phenotype.
Single Gene Disorders
SCN1A
The SCN1A gene is located at chromosome 2q24 and encodes a component of a sodium channel gene which is widely expressed in the brain. This component consists of a voltage sensor and a sodium conducting pore. The 2q24 region contains a cluster of sodium channel genes within a 1.4 Mb area and includes SCN3A, SCN2A, SCN 9A and SCN7A (Meisler et al., 2010).
The intron and exon structure of SCN1A is identical to SCN8A on 12q13.13.
Mutations that disrupt function of SCN1A play a major role in single gene related epilepsy and it is the most clinically relevant epilepsy gene (Mulley et al., 2005). SCN1A mutations were initially identified in individuals having genetic (generalized) epilepsy with febrile seizures plus (GEFS+) (Escayg et al., 2000; Scheffer et al., 2009). GEFS+ is a familial epilepsy syndrome that expresses various different seizure phenotypes from classical febrile seizures, febrile seizures plus, followed by a chronic course of generalized and sometimes focal seizures. Shortly after this discovery, SCN1A mutations were identified in individuals with Dravet Syndrome or severe myoclonic epilepsy of infancy (SMEI) (Claes et al., 2001). Dravet Syndrome is a severe condition having seizure onset before the first year of life in an otherwise normal infant. Between one and four years, myoclonic, atypical absences, focal and other types of seizures appear. There is psychomotor delay and often an ataxic or clumsy-like movement problem. Dravet Syndrome can be associated with frequent status epilepticus and resistance to anticonvulsant treatment (Dravet, 1978; Dravet, 2011). As experience increased with mutation analysis of SCN1A, additional SCN1A-associated epilepsy phenotypes were identified including borderline cases of SMEI (SMEB) and intractable childhood epilepsy with generalized tonic clonic seizures (ICEGTC). SCN1A mutations have not been described as
associated with infantile spasms.
ARX
The ARX (Aristaless-related homeobox) gene is one of the more frequently mutated genes associated with brain malformations, seizures and/or intellectual deficiency. ARX is relevant to epilepsy since its mutations can cause the EEG findings of hypsarrhythmia and burst suppression seen in the West (Kato et al., 2003; Stromme et al., 2002) and Ohtahara syndromes(Giordano et al., 2010; Kato et al., 2010) .
ARX maps to region Xp22.13, and it is known to be a regulator of multiple genes. Since ARX is on the X chromosome, those identified with mutations are typically males although females may be less affected. Familial seizure disorders with X-linked inheritance and onset in infancy are thus prime candidates for consideration for ARX mutation testing. The clinical spectrum of ARX mutations is immense, ranging from X-linked lissencephaly with ambiguous genitalia (XLAG) to nonspecific intellectual deficiency; ten different clinical disorders have been associated with ARX mutations (Kato et al., 2004; Shoubridge et al., 2010). Although mutations have been described in various regions of ARX, seven recurrent mutations account for about 70% of the approximate 100 families reported.
Table 1: Major Epilepsy Genes
| Syndrome/ Condition | Location | Gene | Protein Function | Inherit |
|---|---|---|---|---|
| BFNS1, BFNS2 | 20q13., 38q24 | KCNQ2, KCNQ3 | K channel, K channel | AD |
| BFNIS3 | 2q23-q24.3 | SCN2A | Na α-subunit | –– |
| GEFS+ | 2q23-q24.3, 2q23-q24.3, 19q13.1, 5q31.1- q33.1 | SCN1A, SCN2A, SCN1B, GABRG2 | Na α-subunit, Na α-subunit, Na β-subunit, GABA A-type subunit | AD |
| Dravet/SMEI | 2q23-q24.3 | SCN1A | Na α-subunit | AD |
| Ohtahara/EIEE syndrome; | 9q34.11, Xp22.13 | STXBP1, ARX | Involved in synaptic vesicle release Homeobox gene | AD |
| EFMR | Xq22 | PCDH19 | Member of protocadherin family,cell-cell adhesion | XLD |
| ADNFLE | 20q13.3, 1q21, 8p21 | CHRNA4, CHRNB2, CHRNA2 | Acetylcholine receptor subunit a4- subunitβ2-subunit, a2-subunit | AD |
| EJM1, EJM5, EJM6, EJM7, EJM8 | 6p12-6 p11, 5q33, 2q22-23, 1p36.3, 3q26 | EFHC1, GABRA1, CACNB4, GABRD, CLCN2 | Calcium homeostasis GABA A-type subunit, Calcium channel subunitGABA subunit Chloride channel | AD |
| ADLTEAF | 10q24 | LGI1 | Secreted protein with a G- protein componentinteraction domain | AD |
Table 2: Seizure Disorders Associated with Clinical Syndromes
| Syndrome | Location | Gene | Protein Function |
|---|---|---|---|
| Rett | xq28 | MeCP2 | Methylation binding protein |
| Angelman | 15q11.2-q13 | UBE3A | Ubiquitin ligase enzyme |
| Fragile X | Xq27.3 | FMR1 | Translation repressor |
| ARX disorder (Infantile spasm; lissencephaly; variable phenotype) | Xp22.13 | ARX | Homeobox |
| CDKL5 disorder (Rett-like phenotype) | Xp22 | STK9 (CDKL5) | Serine-Threonine Protein Kinase |
| Cortical dysplasia- focal epilepsy syndrome (CDFE) | 7q35-36 | CNTNAP2 (CASPR2) | adhesion |
| EAST syndrome: epilepsy, ataxia, sensorineural hearing loss, tubulopathy | 1q23.2 | KCNJ10 | Potassium channel protein |
Table 3: Chromosome Disorders Associated with Epilepsy
| Chromosome region | Syndrome name | Gene | Percent with Seizures |
|---|---|---|---|
| Del 16p13.11 | CNV | –– | ?<50% |
| Del 15q13.3 | CNV | ?CHRN7 | 50% |
| Del 15q11.2 | CNV | –– | ?20% |
| Del 1p36 | 1p36 microdeletion syndrome | –– | >50 |
| Del 4p16.3 | Wolf-Hirschhorn syndrome | –– | >50 |
| Trisomy 12p | Trisomy 12p syndrome | –– | ? |
| Ring 14 | Ring 14 syndrome | –– | ? |
| Del 15q11.2-q13 | Angelman syndrome | UBE3A | >50 |
| Dup 15q11.2-q13Idic(15) | Autism/PDDSevere DD/ID, epilepsy, autism | –– | 10-30>50 |
| Del 17p13.3 | Lissencephaly, Type 1 | LIS1 | >50 |
| Ring 20(p13;q13.3) | Ring 20 syndrome | –– | >50 |
| Trisomy 21 | Down syndrome | –– | <10 |
| Del 22q11.2 | Velo-cardio-facial, DiGeorge | –– | <10 |
| XXY | Klinefelter syndrome | –– | 10-20% |
Other genes of varied function
Another X-linked gene, PCDH19, cause a disorder of epilepsy and mental retardation limited to females (EFMR) (Dibbens et al., 2008) with a clinical presentation that mimics Dravet syndrome (Depienne et al., 2011).
Disruption of UBE3A, is responsible for Angelman syndrome. 90% of AS patients have epilepsy (Zori et al., 1992), with atypical absences, myoclonic seizures, GTC, and unilateral seizures. Mutations in STXBP1 have been linked to Ohtahara syndrome or early infantile epileptic encephalopathy with suppression burst (EIEE) disorder, and the encoded protein appears to be involved in synaptic vesicle release (Saitsu et al., 2008). Disruption of MECP2 causes Rett syndrome. MECP2, like UBE3A appears crucial to the process of neuronal plasticity. STK9 (CDKL5) mutations impair a serine/threonine protein kinase, an enzyme involved in intracellular signaling (Tao et al., 2004; Weaving et al., 2004). Such mutations have been reported in patients with severe neurodevelopmental disorders characterized by early- onset seizures, infantile spasms, some Rett syndrome features (deceleration of head growth, stereotypies, and hand apraxia), and refractory epilepsy (Bahi-Buisson et al., 2008c). Mutations in CNTNAP2 (Neurexin IV) disrupt a protein that is a member of the Neurexin super family and cause the cortical dysplasia focal epilepsy syndrome (CDFE) (Strauss et al., 2006). Neurexin IV resides in the synaptic cleft in the nodes of Ranvier. Identification of these and other genes illustrate the enormous cell complexity and varied pathophysiology that leads to epilepsy, including defects in cell membrane function, synaptic development and maintenance, intracellular signaling and transcriptional regulation.
Conclusion
The common epilepsies arise from a complex mixture of multiple genes and environmental effects and conform to the theoretical model of multifactorial inheritance. Although many epilepsy syndromes are attributed to mutations in single genes, the common epilepsies have remained elusive regarding identifying individual risk genes that have even a moderate effect. It is thus likely that a large repertoire of genes confer only mild risk effects for epilepsy. Availability of more refined genetic and biomolecular analyses will further delineate the genetic structure of the human genome and proteonome. Hopefully this refinement will help identify epilepsy risk genes and help clarify how certain CNVs contribute to epilepsy risk. As more knowledge about the polygenic aspects of epilepsy emerges, multiplex gene sequencing testing might then be used for diagnosis and early identification of many genetic-based epilepsy disorders.
References
- Anderson VE, et al. 1991. Epilepsy Res Suppl 4:89-103. Bahi-Buisson N, et al. 2008. Brain Dev 30(2):126-30.
- Battaglia A. 2008. Orphanet J Rare Dis 3:30.
- Battaglia A, et al. 2009. Dev Med Child Neurol 51(5):373-80. Battaglia A, Guerrini R. 2005. Epileptic Disord 7(3):181-92. Berg AT, et al. 2010. Epilepsia 51(4):676-85.
- Berkovic SF, et al. 1998. Ann Neurol 43(4):435-45. Claes L, et al. 2001. Am J Hum Genet 68(6):1327-32. de Kovel CG, et al. 2010. Brain 133(Pt 1):23-32.
- Depienne C, et al. 2011. Hum Mutat 32(1):E1959-75. Dibbens LM, et al. 2009. Hum Mol Genet 18(19):3626-31. Dibbens LM, et al. 2008. Nat Genet 40(6):776-81.
- Dravet C. 1978. Vie Med 8:543-8.
- Dravet C. 2011. Epilepsia 52 Suppl 2:3-9.
- Escayg A, Goldin AL. 2010. Epilepsia 51(9):1650-8.
- Giordano L, et al. 2010. Am J Med Genet A 152A(12):3133-7. Hauser WA, et al. 1996. Mayo Clin Proc 71(6):576-86.
- Heinzen EL, et al. 2010. Am J Hum Genet 86(5):707-18. Hemminki K, et al. 2010. Nat Neurosci 13(9):1120-7. Kasperaviciute D, et al. 2010. Brain 133(Pt 7):2136-47. Kato M, et al. 2004. Hum Mutat 23(2):147-59.
- Kato M, et al. 2003. Neurology 61(2):267-76.
- Kato M, Dobyns WB. 2005. J Child Neurol 20(4):392-7. Kjeldsen MJ, et al. 2001. Epilepsy Res 44(2-3):167-78. Lobo I. 2008. Nat Educ 1(1).
- Masurel-Paulet A, et al. 2010. Clin Genet 78(2):149-61. Mefford HC, et al. 2010. PLoS Genet 6(5):e1000962. Meisler MH, et al. 2010. J Physiol 588(Pt 11):1841-8. Mulley JC, et al. 2005. Hum Mutat 25(6):535-42.
- Saitsu H, et al. 2008. Nat Genet 40(6):782-8. Scheffer IE, et al. 2009. Brain Dev 31(5):394-400. Shoubridge C, et al. 2010. Hum Mutat 31(8):889-900.
- Strauss KA, et al. 2006. N Engl J Med 354(13):1370-7. Stromme P, et al. 2002. Nat Genet 30(4):441-5.
- Tao J, et al. 2004. Am J Hum Genet 75(6):1149-54.
- Turnpenny P, Ellard S. 2004. Elements of Medical Genetics. 12th ed: Elsevier. 436 p. van Bon BW, et al. 2009. J Med Genet 46(8):511-23.
- Weaving LS, et al. 2004. Am J Hum Genet 75(6):1079-93. Zori RT, et al. 1992. J Child Neurol 7(3):270-80.
- Raymond C. Philips Unit, Division of Genetics and Metabolism, Department of Pediatrics, University of Florida, Gainesville, FL 32605; Email: willicx@peds.ufl.edu
- Stella Maris Clinical Research Institute for Child and Adolescent Neuropsychiatry, Via dei Giacinti, 256128 Calambrone, Pisa, Italy; E-mail: agatino.battaglia@inpe.unipi.itCalambrone