Possible association between childhood absence epilepsy and the gene encoding GABRB3

Possible Association between Childhood Absence Epilepsy and the Gene Encoding GABRB3 Martha Feucht, Karoline Fuchs, Ernest Pichlbauer, Kurt Hornik, Joachim Scharfetter, Ralph Goessler, Thomas Fu¨reder, Nevenka Cvetkovic, Werner Sieghart, Siegfried Kasper, and Harald Aschauer Background: Childhood Absence Epilepsy (CAE) is considered to have a predominantly, perhaps exclusively, genetic background. To date, genes responsible for susceptibility to CAE have not been identified. The object of the present study was to test association between CAE and the genes encoding the ␥-aminobutyric acid (GABA) type-A receptor subunits ␣5 (GABRA5) and ␤3 (GABRB3) located on the long arm of chromosome 15 (15q11– q13). Methods: A family-based candidate gene approach was applied: 50 Austrian nuclear families ascertained for the presence of an affected child were investigated. GABRA5 and GABRB3 subunit genes were genotyped using DNA gained from peripheral blood samples by Polymerase Chain Reactions (PCR). Genetic association was tested using a Monte Carlo Version of the multi-allele Transmission-Disequilibrium Test (TDT). Results: The TDT displayed significant overall association with GABRB3 (p ⫽ .0118). Conclusions: The present data suggest that the tested polymorphism may be either directly involved in the etiology of CAE or in linkage disequilibrium with diseasepredisposing sites. Biol Psychiatry 1999;46:997–1002 © 1999 Society of Biological Psychiatry Key Words: Childhood absence epilepsy, family-based association study, ␥-aminobutyric acid type-A receptor subunits ␤3 and ␣5 (GABRB3, GABRA5), haplotype relative risk statistic, transmission disequilibrium test (TDT)

Introduction

E

pileptic seizure disorders are among the most frequently occurring neuropsychiatric diseases in children. Accounting for 13–24%, childhood absence epilepsy From the University Hospital for Child and Adolescent Neuropsychiatry, Vienna, Austria (MF, RG); University Hospital for Psychiatry, Division for Biochemical Psychiatry, Vienna, Austria (KF, EP, NC, WS); University Hospital for Psychiatry, Clinical Department of General Psychiatry, Vienna, Austria (JS, SK, HA); and University of Technology, Department of Statistics and Probability Theory, Vienna, Austria (KH, TF). Address reprint requests to Martha Feucht, MD, Universita¨tsklinik fu¨r Neuropsychiatrie des Kindes und Jugendalters, Wa¨hringer Gu¨rtel 18 –20; A1090 Vienna, Austria Received July 2, 1998; revised February 2, 1999; accepted February 10, 1999.

© 1999 Society of Biological Psychiatry

(CAE) is one of the most frequently recognized syndromes among the idiopathic generalized epilepsies (IGEs). According to the International Classification of Epilepsies and Epileptic Syndromes (Commission 1989), the clinical and electroencephalographic (EEG) diagnostic criteria are: female preponderance; age-dependent onset between 3 and 12 years; very frequent (“pyknoleptic”) typical absence seizures (AS) of any kind— except myoclonic absences—as the initial seizure type; normal psychomotor development before seizure onset, normal neurological and neuroradiological findings; generalized, high-amplitude, bilaterally synchronous 3 Hz spike and wave discharges (SWD) on a normal background activity in the EEG; AS, as well as SWDs provoked by hyperventilation; increase of SWDs during slow wave sleep, but normal organization of sleep; photosensitivity in 20 – 40% of cases; usually complete response to specific antiepileptic drug treatment with sodium valproate or ethosuximide; favorable outcome with spontaneous and durable remission of AS before puberty in 60% or development of (rare) generalized tonic clonic seizures during adolescence in about 40%, respectively. CAE is not associated with an identifiable cause and, due to the consistent results of epidemiological, family and twin studies, is considered to have a predominantly, perhaps exclusively genetic background (Malafosse et al 1994). The precise genetic characterization of the syndrome, however, has remained elusive and several modes of inheritance have been proposed. The perhaps most likely explanation is that CAE is polygenic or, even less probable, multifactorial in origin (Andermann 1985). Genetic complexity and the consequent uncertainties surrounding the mode of transmission at any given locus may impede traditional lod score linkage approaches (Godfrey 1993). In support of this assumption is the fact, that, although it has been the subject of intense investigation, evidence to prove the existence of a major susceptibility gene by means of linkage analysis has not been obtained so far (Sander et al 1995a, 1995b, 1996a, 1996b, 1997; Whitehouse et al 1990). A complementary approach to identify multiple genes at 0006-3223/99/$20.00 PII S0006-3223(99)00039-6

998

M. Feucht et al

BIOL PSYCHIATRY 1999;46:997–1002

different loci, each of small effect on susceptibility, may be offered by disease-marker (candidate) association studies (Hodge 1993). Recently, in vitro studies have defined numerous molecular candidates for reversible epileptiform bursting in neurons (Noebels 1996). The major type of receptors for the inhibitory neurotransmitter ␥-aminobutyric acid (GABA), the postsynaptic type A receptors (GABRs) are hetero-oligomeric ion-channels that seem to be responsible for mediating rapid central nervous system (CNS) neural inhibition as well as they are the site of action of several antiepileptic drugs (Olsen and Avoli 1997; Sieghart 1989, 1992; Snead 1995). Molecular cloning of complementary DNA (cDNA) encoding GABRs in humans has resulted in the identification of at least 15 different protein subunits of five different sequence classes (␣1– 6, ␤1–3, ␥1–3, ␦ and ε). GABRs with different functional properties are produced by different combinations of these subunits, with each subunit being produced by a different gene. Chromosomal mapping indicates that several of these genes seem to be organized as clusters (MacDonald and Olsen 1994). There is also growing evidence for selective regional and temporal expression of these genes in the brain (Brooks-Kayal and Pritchett 1993; Laurie et al 1992; Poulter et al 1992). Consequently, the effects of gene mutations will be found in limited brain areas within a limited time frame, and the existence of genes of regional effect seems likely, that are switched on as a function of age, controlling the predisposition, expression, modulation, and temporal sequence of childhood-onset IGEs. The genes encoding GABR subunits therefore represent leading candidates to play a role in the genetic basis of these epilepsy syndromes. For the following reasons, the gene cluster ␣5 (GABRA5), ␤3 (GABRB3), and ␥3 (GABRG3) located on chromosome 15 (Sinnet et al 1996) is of special interest in this context: Defects in GABRA5 and GABRB3 function have been suggested to play a role in producing the electroclinical pattern in Angelman syndrome (AGS). The chromosomal abnormalities observed are deletions on the maternally inherited chromosome 15 or uniparental paternal disomy (Williams et al 1995; Greger et al 1997). Interstitial duplication-inversion of the GABRA5 and GABRB3 genes on the maternally derived chromosome 15q was reported by Bundey et al (1994) in a patient with epilepsy, autism and ataxia. An abnormal karyotype 47 XY ⫹ inv. dup.15q11– q13 (parental origin) was found by Aguglia et al. (1995) in a case of Prader–Willi syndrome associated with seizures. In a recently published paper, Elia et al (1998) report about the relationship between myoclonic absence seizures and underlying chromosome disorders. Among 14 patients observed in 3 centres, 7 presented with chromo-

somopathies. In 4 cases abnormal expression of genes codifying for GABRB3 was demonstrated. Alterations of GABRs have been described in several genetic animal models of epilepsy (Malafosse et al 1994; Olsen and Avoli 1997). The enhancement of [3H]flunitrazepam binding and the abundance of ␤2–␤3 subunits were both reported to be decreased in the sensory motor cortex and the anterior thalamus of genetically epilepsy-prone rats (GAERS) vs. control animals (Spreafico et al 1993). In the mouse pink-eyed cleft palate (pcp) mutation, a mouse model for AGS, the genes encoding the ␣5, ␤3 and ␥3 subunits are disrupted by deletion (Nakatsu et al 1993). In the stargazer mouse, a mutant inbred mouse strain with generalized cortical SWDs and absence-like attacks, the seizure predisposition is inherited in an autosomal recessive fashion and the gene has been localized to chromosome 15 (Noebels et al 1990). Finally, ␤3-deficient mice display hyperactive behavior, frequent myoclonus and occasional epileptic seizures (Homanics et al 1997). The thalamocortical GABRA5 and GABRB3 gene expression undergoes a prominent peak in early brain and subsequently substantially diminishes to be superseded in the adult brain by ␣1, ␣4, ␤2 and ␦ subunit mRNAs (Brooks-Kayal and Pritchett 1993; Laurie et al 1992). Although there is little known about the function of GABRB3 in vivo, pharmacological studies revealed that in GABRs containing the ␤3 and ␤2 subunits some AEDs are far more effective than in those containing the ␤1 subunit (Wafford et al 1994). Based on these observations, the objective of the present study was to test allelic associations between CAE and the DNA polymorphism of GABR subunits ␤3 (155 CA2) and ␣5 (85 CA) on chromosome 15q11–13. The hypothesis under investigation was that these candidate genes carry mutations that increase the susceptibility to CAE and that the alleles studied are in linkage disequilibrium with these mutations.

Methods and Materials The study was performed according to a protocol approved by the local Ethical Committee.

Study Sample Nuclear families ascertained for the presence of an affected child were recruited from the outpatient department for seizure disorders at the Department of Child and Adolescent Neuropsychiatry. Patients were eligible if they had an established diagnosis of CAE (Commission 1989), if they had been under continuous surveillance for a follow-up period of at least five years and if both parents were willing to participate in the study. Required pre-entry confirmatory examinations included detailed diagnostic re-evaluation based on the review of medical

Association between CAE and a GABAA-Receptor Subunit Gene

BIOL PSYCHIATRY 1999;46:997–1002

999

Table 1. GABAA Receptor Subunit Gene Polymorphisms n alleles

Allele size (bp)

Heterozygosity (%)

85 CA

11

106 –126

85

155 CA2

12

95–123

91

GABRA Subunit gene

Polymorphic repeat

GABRA5 GABRB3

records and re-analysis of EEG/Video recordings. The possibility of progressive neurologic diseases, significant medical illness and other exogenous factors known to be strongly associated with risk for epilepsy had to be excluded. Neuroimaging studies (MRI) had to be performed within the last six months. Patients or their parents provided informed written consent.

Genotyping High molecular weight DNA was extracted from peripheral blood leukocytes using the Nuncleon II kit (Scotlab, UK). GABR subunit genes were genotyped by polymerase chain reaction (PCR). The primers used revealed simple sequence repeat polymorphisms. The chromosome 15 DNA markers used (Glatt et al 1994) are presented in Table 1. Long range restriction mapping suggests that the two genes are arranged in head-to-head transcriptional orientation (GABRB3 100 kb proximal of GABRA5). The polymorphic (CA)n repeats used in this study are located in the intron between the exons 3 and 4 (155CA2) and in the promoter region (85CA), respectively (Glatt et al 1997). PCR amplification was performed in a reaction volume of 40 ␮l using 500 ng of genomic DNA, 10 mM Tris-HCl, pH 8.4, 50 mM KCl, 200 ␮M of each dNTP, 10 ␮M unlabeled primer A and 10 ␮M 33P-labeled primer B, and 2 units of DyNAzyme DNA Polymerase, Native Enzyme (Finnzymes OY/Finland). DyNAzyme is a thermostabile DNA polymerase, isolated and purified from Thermus brockianus (Finnzymes Inc. strain F 500). Denaturation was for 1 min at 95°C, annealing was for 1 min at 55°C (␣5), resp. 62°C (␤3), and extension was for 1 min at 72°C. After 30 cycles, a final extension was performed at 72°C for 5 min. PCR samples were resolved by electrophoresis on formamide-containing sequencing gels and polymorphic bands were detected by autoradiography for 12–24 hr. Genotypes were scored blindly to diagnosis and pedigree position by two independent raters.

Statistical Analysis A nuclear family-based candidate gene approach was applied. In families ascertained for the presence of an affected child, the affected offspring’s genotype at a marker locus (made up of the transmitted parental marker alleles) was used as the case sample, and an artificial genotype derived from the not transmitted parental alleles (haplotypes) formed the affected family-based control sample (AFBAC). This matched design for “patient” (parental transmitted) and “control” (parental nontransmitted) marker alleles avoids ethnic confounding in the case of a stratified population (Thomson 1995).

Primer 5⬘-CCATGATAGTGGAATAG-3⬘ 5⬘-AGTCGATTTATCAGCCTCC-3⬘ 5⬘-GAAGGAACATTTCTGGGTC-3⬘ 5⬘-CACAACCATTTTTGGTCTGC-3⬘

The null hypothesis of no marker association with the disease was tested using MCT_m, a Monte Carlo version of the multiallele Transmission Disequilibrium Test for nuclear family data (T_m test, described by Kaplan et al 1997), with 1000000 shuffels (code by E. R. Martin, obtained from ftp://statgen. ncsu.edu/pub/martin/Mctest). First the table of transmitted and non-transmitted alleles was computed that then was passed as input to the MCT_m program. The Transmission Disequilibrium Test (TDT) originally was proposed as a test of linkage in the presence of association. If the sample consists of only nuclear families with a single affected child, then the TDT also can be used to test for association in the presence of linkage. The use of Monte Carlo randomization techniques is recommended if the sample size is small, to guard against an overly conservative test. Regardless of the test statistic, this procedure always leads to a test with a significance level close to the nominal value (Kaplan et al 1997; Martin et al 1997).

Results Fifty CAE patients, 31 females (62%) and 19 males (38%), between 10.2 and 18.6 years of age (mean 12.7 years) and their parents were genotyped. Onset of AS had been between 4.9 and 8.2 years (mean 6.8 years) of age. Forty children (80%) suffered from absence seizures exclusively, 10 (20%) had absence seizures and rare generalized tonic clonic seizures during the course of the illness. 8 children (16%) had a history of febrile seizures. A positive family history of epilepsy was found in 20 cases (40%); paternal in 11 (55%), maternal in 9 (45%). Distributions of allele frequencies as well as results obtained by the TDT are displayed in Tables 2 and 3. For GABRB3 155 CA2, the multi-allele TDT revealed a significant result (value of Tm statistic 24.5758, p ⫽ .0118). Results for GABRA5 85 CA did not reach significance (value of Tm statistic 16.7893, p ⫽ .1060). According to visual analysis of allele distribution (Table 2), significant differences were seen for the 107 bp allele of GABRB3, that was over represented in the cases (25:10) and for the 95 bp allele of GABRB3, that was over represented in the control group (20:46); 95 bp was the most frequently observed allele of this subunit, that gives this result even more impact.

1000

M. Feucht et al

BIOL PSYCHIATRY 1999;46:997–1002

Table 2. Allele Frequencies and Statistics: GABRB3 (155 CA 2F/2R) Alleles

Patients (n ⫽ 48 a)

Internal controls (96 alleles)

20 8 8 2 25 9 12 3 5 3 1 0

46 6 2 1 10 11 12 3 2 2 0 1

95 97 103 105 107 109 111 113 115 117 119 123

MCT_m TDTb

p ⫽ .0118

a

Two genotypes missing for technical reasons. Monte Carlo Version with 1000000 shuffles.

b

Discussion The results of a significant overall association and linkage between GABR subunit ␤3 and CAE using the TDT, suggest that the gene for GABRB3 or loci in linkage disequilibrium with it, might contribute to CAE susceptibility and thus be involved in the etiopathogenesis of the syndrome. The negative association observed for allele 95 bp GABRB3 by visual analysis of the allele distribution (Table 2) suggests that there might perhaps exist some “modifier” or “protector (seizure resistance) gene,” that exerts an “antiepileptic influence” in healthy individuals and which is in connection with or at least located near these repeats. Up to date, there is a limited amount of literature on GABR subunit genes and susceptibility to the IGEs, especially CAE. Sander et al (1996b) investigated the hypothesis that allelic variants of GABRA1 and GABRG2 or tightly linked neurotransmitter receptor genes located on chromoTable 3. Allele Frequencies and Statistics: GABRA5 (85 CA) Alleles 106 108 110 112 114 116 118 120 122 124 126 a

Patients (n ⫽ 50)

Internal controls (100 alleles)

MCT_m TDTa

4 10 19 36 1 6 5 0 13 4 3

2 13 33 31 4 3 6 1 4 4 0

p ⫽ .1060

Monte Carlo Version with 1000000 shuffles.

some 5q32–35 contribute a frequent major gene effect to the expression of the most common IGEs. Their linkage results in the entire group of 63 families with Juvenile Myoclonic Epilepsy (JME: n ⫽ 30), Juvenile Absence Epilepsy (JAE: n ⫽ 30) and CAE (n ⫽ 18) provided evidence against this assumption, although the operational exclusion region was not reached for the families ascertained through CAE patients. The authors concluded that their results might be in part due to the methodological difficulties inherent in parametric linkage analysis of genetically complex traits. Sander et al (1997) tested linkage between the GABAA receptor ␣5, ␤3, and ␥3 subunit gene cluster on chromosome 15 and idiopathic generalized epilepsies (JME: n ⫽ 61, JAE: n ⫽ 38 and CAE: n ⫽ 29). Their results provide evidence against an IGE susceptibility locus within this gene cluster both in the entire family set and the subsets of families. Slightly positive lod scores were obtained, however, in families of CAE patients under diagnostic Scheme 2: idiopathic generalized seizures before the age of 26 years (assuming an autosomal dominant mode of inheritance with 50% penetrance) and for the JME subgroup; and under diagnostic Scheme 3: either idiopathic generalized SW trait or generalized SW EEG (assuming genetic heterogeneity and an autosomal recessive mode of inheritance). In the same paper, the authors report, that they were not able to find an allelic association between JME and GABRB3/GABRA5 microsatellites, using the familybased HRR method. They conclude that the high mutation rate of the tested polymorphisms might reduce the likelihood that a nearby trait causing mutation remains in linkage disequilibrium with an originally linked allele. Guipponi et al (1997) reported lack of association between JME and GABRA5 and GABRB3 genes. According to the authors, the negative result may be due to population stratification and the possible heterogeneity of the IGEs. In addition, the overlap of the various subgroups makes the attempt to clearly define a homogeneous phenotype a difficult if not impossible task. Advantages of the present study are that all patients included belonged to the same subsyndrome. In addition, the statistical approach chosen is reasonably robust to stratification effects. It is therefore better suited to testing for the effects of genetic factors in a population when A) a clear definition of the phenotype is unknown; B) the hypothetical susceptibility to the given disease is produced by many genes, none of which has a major effect; and C) each of them is potentially responsible for a limited amount of the total genetic variance. Association mapping of candidate genes by use of nuclear-based sample collections is therefore relevant to genetic studies of complex diseases like the IGEs in which standard linkage approaches may be less effective. Recent

Association between CAE and a GABAA-Receptor Subunit Gene

theoretical analyses demonstrated that, apart from Type 1 error, an association uncovered by use of nuclear familybased data must implicate a disease predisposing locus linked to the marker locus (Thomson 1995). One major problem is the large number of marker alleles when using polymorphic microsatellites. To our knowledge there is no published and widely accepted statistical procedure for this problem. The problems of multi-allele testing when using polymorphic markers were also extensively discussed by Steinlein et al (1997). The result obtained for allele bp 95 has therefore to be checked in an independent study in a new sample.

Conclusion Genes responsible for CAE have not yet been identified. Numerous genetic, pharmacological and semiological data suggest that CAE does not constitute a homogeneous entity neither in humans nor in animals, but seems to be a polygenic disorder with the phenotype probably being expressed by a specific combination of genes, perhaps not located on the same chromosome. Some of these genes must be very common, and might be shared with other IGEs as well as with the idiopathic localization-related epilepsies (Neubauer et al 1997), others not. It is tempting to speculate that the common link in genetic epilepsies is an alteration of some seizure protecting mechanisms and that remission occurs when these mechanisms mature. The result of this association study, for the first time, indicates linkage between a certain GABR subunit gene and the susceptibility to CAE. Taking into consideration the role of GABA in epileptogenesis in general and that of GABRB3 in detail, the result seems plausible. Yet, results have to be interpreted with caution and it remains to be determined what role, if any, the genes investigated in this study play in the etiology of CAE. Further subsequent studies have to be undertaken to replicate the present results in larger cohorts. Finally, a consequence should be the effort to sequence the first three exons and the 5⬘-UTR (untranslated region) of the GABRB3 subunit in cases and controls, to identify and isolate the suspected mutations. This work was supported by the Austrian Fund of Scientific Research grant P10460-MED.

References Aguglia U, Gambarella D, Concolino D, Porta G, Sirchia S, Quattrone A (1995): Severe epilepsy in a patient with Prader–Willi syndrome due to a inversion-duplication of chromosome 15q11–13. A clinical, neurophysiological and molecular genetic study (abstract). Epilepsia 36(Suppl 3):S1. Andermann E (1985): Genetic aspects of the epilepsies. In: Sakai

BIOL PSYCHIATRY 1999;46:997–1002

1001

T, Tsuboi T, editors. Genetic Aspects of Human Behavior. Tokyo: Igaku-Shoin, pp 129 –145. Brooks-Kayal AR, Pritchett DB (1993): Developmental changes in human GABAA receptor subunit composition. Ann Neurol 34:687– 693. Bundey S, Hardy C, Vickers V, Kilpatrick MW, Corbett JA (1994): Duplication of 15q11–13 region in a patient with autism, epilepsy and ataxia. Dev Med Child Neurol 36:736 – 742. Commission on Classification and Terminology of the International League against Epilepsy (1989): Proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia 30:389 –399. Elia M, Guerrini R, Musumeci SA, Bonanni P, Gambardella A, Aguglia U (1998): Myoclonic absence-like seizures and chromosome abnormality syndromes. Epilepsia 39:660 – 663. Glatt K, Sinnett D, Lalande M (1994): The human ␥-aminobutyric acid receptor subunit ␤3 and ␣5 in chromosome 15q11–13 is rich in highly polymorphic (CA)n repeats. Genomics 19:157–160. Glatt K, Glatt H, Lalande M (1997): Structure and organization of GABRB3 and GABRA5. Genomics 41:63– 69. Godfrey M (1993): Molecular heterogeneity: a clinical dilemma. Clinical heterogeneity: a molecular dilemma. Am J Hum Genet 53:22–25. Greger V, Knoll JHM, Wagstaff J, Woolf E, Lieske P, Glatt H, et al (1997): Angelman Syndrome associated with an inversion of chromosome 15q11.2q24.3. Am J Hum Genet 60: 574 –580. Guipponi M., Pierre Th, Girard-Reydet C, Feingold J, BaldyMoulinier M, Malafosse A (1997): Lack of association between juvenile myoclonic epilepsy and GABRA5 and GABRB3 genes. Am J Med Genet 74:150 –153. Hodge SE (1993): Linkage analysis vs. association analysis: distinguishing between two models that explain diseasemarker associations. Am J Hum Genet 53:367–384. Homanics GE, DeLorey T, Firestone LL, Quinlan JJ, Handforth A, Harrison NL, et al (1997): Mice devoid of ␥-aminobutyric type A receptor ␤3 subunit have epilepsy, cleft palate, and hypersensitive behavior. Proc Natl Acad Sci USA 94:4143– 4148. Kaplan NL, Martin ER, Weir BS (1997): Power studies for the transmission/disequilibrium tests with multiple alleles. Am J Hum Genet 60:691–702. Laurie DJ, Wisden W, Seeburg PH (1992): The distribution of 13 GABAA receptor subunit mRNAs in the rat brain. III. Embryonic and postnatal development. J Neurosci 12:4251– 4172. Mac Donald RL, Olsen RW (1994): GABAA receptor channels. Ann Rev Neurosci 17:569 – 602. Malafosse A, Genton P, Hirsch E, Marescaux C, Broglin D, Bernasconi R (1994): Idiopathic Generalized Epilepsies. London, GB: John Libbey & Company Ltd. Martin ER, Kaplan NL, Weir BS (1997): Tests for linkage and association in nuclear families. Am J Hum Genet 61:439 – 448. Nakatsu Y, Tyndale RF, DeLorey TM, Durham-Pierre D, Gardner JM, McDanel H, et al (1993): A cluster of three GABAA receptor subunit genes is deleted in a neurological mutant of the mouse p locus. Nature 364:448 – 450.

1002

BIOL PSYCHIATRY 1999;46:997–1002

Neubauer BA, Ka¨mpfer F, Fiedler B, Schwabe G, Doose H, Stephani U (1997): Toward the mapping of a gene responsible for rolandic epilepsy (Abstract). Epilepsia 38(suppl 3):201. Noebels JL, Qiao X, Bronson RT, Spencer C, Davisson MT (1990): Stargazer: a new neurological mutant on Chromosome 15 in the mouse with prolonged cortical seizures. Epilepsy Res 7:129 –135. Noebels JL (1996): Targeting epilepsy genes. Neuron 16:24 –244. Olsen RW, Avoli M (1997): GABA and epileptogenesis. Epilepsia 38:399 – 407. Poulter MO, Barker JL, O⬘Carroll AM, Lolait SJ, Mahan LC (1992): Differential and transient expression of GABAA receptor ␣-Subunit mRNAs in the developing rat CNS. J Neurosci 12:2888 –2900. Sander T, Hildmann BC, Janz D, Wienker TF, Neitzel H, Bianchi A, et al (1995a): The phenotypic spectrum related to the human epilepsy susceptibility gene “EJM1.” Ann Neurol 38:210 –217. Sander T, Janz D, Ramel C, Ross CA, Paschen W, Hildmann T, et al (1995b): Refinement of map position of the human GluR6 kainate receptor gene (GRIK2) and lack of association and linkage with idiopathic generalized epilepsies. Neurology 45:1713–1720. Sander T, Hildmann BC, Wienker TF Ramel C, Beck-Managetta G, Bianchi A, et al (1996a): Common subtypes of idiopathic generalized epilepsies: lack of linkage to D20S19 close to candidate loci (EBN1, EEGV1) on chromosome 20. Am J Med Genet 67:31–39. Sander T, Hildmann T, Janz D, Wienker FT, Bianchi A, Bauer G, et al (1996b): Exclusion of linkage between the idiopathic generalized epilepsies and the GABAA receptor ␣1 and ␥2 subunit cluster on chromosome 5. Epilepsy Res 23:235–244. Sander T, Kretz R, Williamson MP, Elmslie FV, Rees M, Hildmann T, et al (1997): Linkage analysis between idiopathic generalized epilepsies and the GABAA receptor ␣5, ␤3 and ␥3 subunit gene cluster on chromosome 15. Acta Neurol Scand 96:1–7. Serratosa J, Delgado-Escueta A, Liu A, Pascual-Castroviejo I, Wang G, Sparkes R (1993): Exclusion of linkage between

M. Feucht et al

DNA markers in the juvenile myoclonic epilepsy locus of chromosome 6p and childhood absence epilepsy. Epilepsia 34(suppl 2):S149. Sieghart W (1989): Multiplicity of GABAA-benzodiazepine receptors. Trends Pharmacol Sci 10:407– 411. Sieghart W (1992): Ligand-gated Cl⫺ ion channels modulated by multiple drug binding sites. Trends Pharmacol Sci 13:446 – 450. Sinnet D, Woolf E, Xie W, Glatt K, Kirkness EF, Nielsen TO, et al (1996): Identification of a putative DNA replication origin in the ␥-aminobutyric acid receptor ␤3 and ␣5 gene cluster on human chromosome 15q11–13, a region associated with parental imprinting and allele-specific replication timing. Gene 173:171–177. Snead OC III (1995): Basic mechanisms of generalized absence seizures. Ann Neurol 7:77– 88. Spreafico R, Mennini T, Danober L, Cagnotto A, Regondi MC, Miari A, et al (1993): GABAA receptor impairment in the genetic absence epilepsy rats from Strasbourg (GAERS): an immunocytochemical and receptor binding autoradiographic study. Epilepsy Res 15:229 –238. Steinlein O, Sander T, Stoodt J, Kretz K, Janz D, Propping P (1997): Possible association of a silent polymorphism in the neuronal nicotinic acetylcholine receptor subunit ␣4 with common idiopathic generalized epilepsies. Am J Med Genet 74:445– 449. Thomson G (1995): Mapping disease genes: family-based association studies. Am J Hum Genet 57:487– 498. Wafford KA, Bain CJ, Quirk K, McKernan RM, Wingrove PB, Whiting PJ, et al (1994): A novel allosteric modulatory site on GABAA receptor ␤ subunit. Neuron 12:775–782. Whitehouse WP, Curtis D, Gardiner RM (1990): Linkage analysis with HLA-DQ A1 and A2 in juvenile myoclonic epilepsy and childhood absence epilepsy. Epilepsia 31:819. Williams CA, Angelman H, Clayton-Smith J, Driscoll DJ, Hendrickson JE, Knoll JHM, et al (1995): Angelman Syndrome: consensus for diagnostic criteria. Am J Med Genet 56:237–238.