1

2

3Q1

4

5

6

7891011121314Q2

1 5

1617181920

21Q422232425

39

4041

42

43

44Q5

45

46

Neurobiology of Disease xxx (2014) xxx–xxx

YNBDI-03151; No. of pages: 9; 4C: 4, 5, 6

Contents lists available at ScienceDirect

Neurobiology of Disease

j ourna l homepage: www.e lsev ie r .com/ locate /ynbd i

Exonic microdeletions of the gephyrin gene impair GABAergic synapticinhibition in patients with idiopathic generalized epilepsy

PRO

OFBorislav Dejanovic a,1, Dennis Lal b,c,d,1, Claudia B. Catarino e,1, Sita Arjune a, Ali A. Belaidi a, Holger Trucks b,h,

Christian Vollmar e, Rainer Surges f,h, Wolfram S. Kunz f,h, Susanne Motameny b, Janine Altmüller b,Anna Köhler a, Bernd A. Neubauer d, EPICURE Consortium h,2, Peter Nürnberg b,c,g,h, Soheyl Noachtar e,Günter Schwarz a,c,g,⁎,3, Thomas Sander b,h,⁎⁎,3

a Department of Chemistry, Institute of Biochemistry, University of Cologne, 50674 Cologne, Germanyb Cologne Center for Genomics (CCG), University of Cologne, 50931 Cologne, Germanyc Cologne Excellence Cluster on Cellular Stress Responses in Aging-Associated Diseases (CECAD), University of Cologne, 50674 Cologne, Germanyd Department of Neuropediatrics, University Medical Center Giessen and Marburg, 35392 Giessen, Germanye Epilepsy Center, Department of Neurology, University of Munich, 81377 Munich, Germanyf Department of Epileptology, University Clinics Bonn, 53105 Bonn, Germanyg Center for Molecular Medicine (CMMC), University of Cologne, 50931 Cologne, Germanyh EPICURE Consortium, Germany

UNAbbreviations: GABAARs, γ-aminobutyric acid type-A

variations; IGE, idiopathic generalized epilepsy; TMS, tranDTI, diffusion tensor imaging; HEK293, human embryolobe epilepsy; qPCR, quantitative polymerase chain reacti

⁎ Correspondence to: G. Schwarz, Institute of BiochZülpicher Str. 47, 50674 Cologne, Germany. Fax: +49 221

⁎⁎ Correspondence to: T. Sander, Cologne Center forGWeyertal 115b, 50931 Cologne, Germany. Fax: +49 221 4

E-mail addresses: gschwarz@uni-koeln.de (G. Schwthomas.sander@uni-koeln.de (T. Sander).

Available online on ScienceDirect (www.sciencedir1 Authors contributed equally.2 EPICURE Consortium participants are listed in Append3 Equal senior authorship.

http://dx.doi.org/10.1016/j.nbd.2014.02.0010969-9961/© 2014 Published by Elsevier Inc.

Please cite this article as: Dejanovic, B., et al.idiopathic generalized epilepsy, Neurobiol. D

Da b s t r a c t

a r t i c l e i n f o

26

27

28

29

30

31

32

33

34

35

Article history:Received 30 September 2013Revised 9 January 2014Accepted 10 February 2014Available online xxxx

Keywords:Idiopathic generalized epilepsyMicrodeletionGPHNGephyrin

36

37

38

RECTEGephyrin is a postsynaptic scaffolding protein, essential for the clustering of glycine and γ-aminobutyric acid

type-A receptors (GABAARs) at inhibitory synapses. An impairment of GABAergic synaptic inhibition representsa key pathway of epileptogenesis. Recently, exonic microdeletions in the gephyrin (GPHN) gene have been asso-ciated with neurodevelopmental disorders including autism spectrum disorder, schizophrenia and epilepticseizures. Here we report the identification of novel exonic GPHN microdeletions in two patients with idiopathicgeneralized epilepsy (IGE), representing themost common group of genetically determined epilepsies. The iden-tified GPHN microdeletions involve exons 5–9 (Δ5–9) and 2–3 (Δ2–3), both affecting the gephyrin G-domain.Molecular characterization of the GPHN Δ5–9 variant demonstrated that it perturbs the clustering of regulargephyrin at inhibitory synapses in cultured mouse hippocampal neurons in a dominant-negative manner,resulting in a significant loss of γ2-subunit containing GABAARs. GPHN Δ2–3 causes a frameshift resulting in apremature stop codon (p.V22Gfs*7) leading to haplo-insufficiency of the gene. Our results demonstrate thatstructural exonic microdeletions affecting the GPHN gene constitute a rare genetic risk factor for IGE and otherneuropsychiatric disorders by an impairment of the GABAergic inhibitory synaptic transmission.

© 2014 Published by Elsevier Inc.

R

47

48

49

50

COIntroduction

Epilepsy is characterized by recurrent spontaneous seizures due to aneuronal hyperexcitability and an abnormal cortical synchronization.

51

52

53

54

55

56

57

58

59

60

61

62

63

receptors; CNVs, copy numberscranial magnetic stimulation;nic kidney 293; TLE, temporalon.emistry, University of Cologne,470 5092.enomics, University of Cologne,78 96866.arz),

ect.com).

ix A.

, Exonic microdeletions of theis. (2014), http://dx.doi.org/1

Approximately 3% of the general population is affected by epilepsyuntil the age of 40 years (Hauser et al., 1993). Idiopathic generalizedepilepsies (IGEs) represent the most common group of geneticallydetermined epilepsies, accounting for 20–30% of all epilepsies (Jallonet al., 2001). Their clinical features are characterized by age-relatedrecurrent unprovoked generalized seizures, in the absence of detectablebrain lesions or metabolic abnormalities (Anon., 1989; Nordli, 2005).Genetic factors play a predominant role in the etiology of IGE withheritability estimates of 80%. However, the vast majority of IGE syn-dromes have an oligo-/polygenic predisposition and their genetic basisremains elusive (Dibbens et al., 2007).

Structural genomic copy number variations (CNVs) account for asubstantial fraction of the genetic variance in about 3% of patientswith idiopathic epilepsies (de Kovel et al., 2010; Heinzen et al., 2010;Helbig et al., 2009; Mefford et al., 2010). Recurrent microdeletions at15q11.2, 15q13.3, and 16p13.11 (de Kovel et al., 2010; Helbig et al.,2009) and exonic deletions in NRXN1 (Moller et al., 2013) and RBFOX1

gephyrin gene impair GABAergic synaptic inhibition in patients with0.1016/j.nbd.2014.02.001

T

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141Q6

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160Q7

161

162

163

164

165

166Q8

167

168

169

170

171

172

173

174

175

176

177

178

179Q9

180

2 B. Dejanovic et al. / Neurobiology of Disease xxx (2014) xxx–xxx

UNCO

RREC

(Lal et al., 2013) increase the risk of IGE and a wide range ofneurodevelopmental disorders (Coe et al., 2012a). Recently, exonicmicrodeletions in the gene encoding the synaptic scaffolding proteingephyrin (GPHN) have been found to represent a rare cause of neuro-developmental disorders, including autism spectrum disorder (ASD),schizophrenia and epileptic seizures (Lionel et al., 2013). Two out of sixpatients carrying exonic GPHN deletions exhibited seizures (Lionelet al., 2013).

Gephyrin is a postsynaptic scaffolding protein, essential for the clus-tering and localization of glycine and a subset of GABAA receptors atinhibitory synapses (Fritschy et al., 2008; Kirsch et al., 1993;Kowalczyk et al., 2013; Mukherjee et al., 2011; Tretter et al., 2008,2011). Gephyrin is composed of three domains — the C-terminalE-domain binds directly to the glycine and GABAA receptor sub-units, whereas the N-terminal G-domain is crucial for gephyrin'soligomerization. Both domains are connected by the central (C) domain(Fritschy et al., 2008). Aberrant function of gephyrin impairs GABAergicsynaptic inhibition, which thereby may promote neuronal hyperexcit-ability and seizure susceptibility. Previously, we have reported agephyrin-specific effect in temporal lobe epilepsy (TLE), where stress-induced irregular splicing ofGPHN resulted in the expression of truncat-ed gephyrin variants that impaired the function of regular gephyrin bydominant-negative interaction (Forstera et al., 2010). Additionally,reduced gephyrin expression has been detected in temporal lobe surgi-cal specimens of patients with medically refractory TLE, as well as in ratmodels (Fang et al., 2011). These accumulating lines of evidencesupport the hypothesis that gephyrin dysfunction impairs GABAergicsynaptic inhibition and thereby contributes to epileptogenesis (Fanget al., 2011; Gonzalez et al., 2013). In the present study, we screenedthe GPHN gene formicrodeletions in 1469 European patients with com-mon IGE syndromes and 2256 population controls (Lal et al., 2013). Weidentified exonic GPHN deletions affecting the gephyrin N-terminalG-domain in two IGE patients and characterized the underlyingmolecular mechanism promoting epileptogenesis.

Materials and methods

Study cohort and CNV screening

This study has been approved by the local Research Ethics Commit-tees. All study participants including the family members of the GPHNdeletion-carriers provided an informed written consent. CNV screeningof the genomic GPHN sequence (14q23.3:66,974,124–67,648,524;human genome build 37/hg19) was carried out in 1469 unrelated pa-tients with IGE of self-identified North-Western European ancestry and2256 German population controls (Lal et al., 2013). DNA samples wereinvestigated by the Affymetrix Genome-Wide Human SNP Array 6.0(Affymetrix, Santa Clara, CA, USA). CNV analysis was performed usingthe BirdSuit algorithm implemented in the Affymetrix Genotyping Con-sole version 4.1.1. Regional log2 ratios of the signal intensities and theSNP heterozygosity state were visualized in the Chromosome AnalysisSuite v1.2.2 (Affymetrix, Santa Clara, CA, USA). The copy number stateof the GPHN microdeletions identified by the array-based CNV analysiswas examined by real-time quantitative PCR (qPCR), using a TaqMan®CNV assays located in GPHN exon 7 (Assay ID: Hs01711518_cn) andintron 2 (Assay ID: Hs07084813_cn; Life Technologies, Carlsbad,CA, USA). Genome-wide CNV screening beyond the GPHN locus was re-stricted to CNVswith a segment size N500 kb (Nicoletti et al., 2012) anda minimum of 50 markers to achieve a high accuracy and reliability.

Transcript analysis

RNA was isolated from whole blood samples using the PAXgeneBlood RNA Kit (Qiagen, Hilden, Germany). For amplification of theΔ5–9 deletion flanking primers were designed: Forward (within exon4) AGGAACAGGATTTGCACCAC and reverse (within exon 13) GCGATG

Please cite this article as: Dejanovic, B., et al., Exonic microdeletions of theidiopathic generalized epilepsy, Neurobiol. Dis. (2014), http://dx.doi.org/1

ED P

RO

OF

TCTTCTAGCCACCT. Amplification of the Δ2–3 deletion was performedwith primers: Forward (within exon 1) CCGAGGGAATGATCCTTACTAAand reverse (within exon 7) TCAAGTTCATCATGCACCTCC. The sizes ofthe amplified DNA fragmentswere assessed using 1.5% agarose gel elec-trophoresis and the Agilent 2200 TapeStation Nucleic Acid System(Agilent Technologies, Santa Clara, California, USA). Bidirectional Sangersequencing of the identified GPHN cDNA amplicons was carried outfollowing standard protocols.

Paired-pulse transcranial magnetic stimulation

For Family 1, the patient, his father and six healthy controls with nopersonal history of seizures or other neurological disorders and nomedication (three females/three males, median age 38 years, range24–44 years) had transcranial magnetic stimulation (TMS). After evalu-ating the resting motor threshold (MT), paired-pulse TMS was used tostudy intra-cortical inhibition (ICI; interstimulus interval [ISI] of 3 ms)and intra-cortical facilitation (ICF; ISI of 13 ms). The protocol used wasadapted from Werhahn et al. (2000). IBM SPSS Statistics 21.0 softwarewas applied for statistical analysis.

Expression construct

Regular enhanced green fluorescent protein (EGFP)-taggedgephyrin (Lardi-Studler et al., 2007) served as the basis for Δ5–9gephand Δ2–3geph expression constructs. Both variants were introducedinto the pEGFP-C2 vector (Clontech) using XhoI and HindIII restrictionsites. To createmyc-tagged gephyrin, EGFP-tagged gephyrinwas ampli-fied with primers that introduce the sequence for the myc-tag(EQKLISEEDL) at the N-terminus of gephyrin and a stop-codon afterthe last amino acid of gephyrin. Fragment was introduced into thepEGFP-N2 vector (Clontech, Mountain View, CA, USA) using XhoI andHindIII restriction sites.

Hippocampal neuron cultures and transfection

Primary neuron cultureswere prepared fromhippocampi of C57BL/6mice andplated on poly-L-lysine coated coverslips at a density of 75,000/24-well dish. Neurons were cultured in Neurobasal medium, supple-mented with B-27, N-2 and glutamine. Neurons were transfected after11 days in vitro (DIV) unless otherwise stated. Transfection was carriedout with Lipofectamin 2000 (Invitrogen) according to manufacturer'smanual. Constructs were expressed for 48 h unless otherwise stated.

Human embryonic kidney cell culture and transfection

Human embryonic kidney (HEK293) cells were grown in Dulbecco'smodified Eagle'smedium supplementedwith 10% fetal calf serum and 2mM L-glutamine at 37 °C and 5% CO2. For confocal laser-scanningmicroscopy, 1 × 105HEK293 cells were seeded on collagenized coverslipsin 12-well plates and immediately transfected with polyethylenimine(1 mg/ml, diluted in H2O, pH 7.0) using standard protocols. Cells weregrown for 24–48 h depending on the assay.

Immunostaining of cultured cells

Cells were fixed with 4% PFA for 10 min at room temperature.Unspecific binding sites were blocked with blocking solution (2% BSA,10% goat serum, 0.2% Triton X-100) for 1 h and primary antibodieswere applied for 1 h in 10% goat serum. After three washing steps inPBS, cells were incubated with secondary antibodies in 10% goatserum for 1 h at room temperature. After the final three washing stepsin PBS, slides were mounted on coverslips with Fluoro gel II containingDAPI (Science Service, Munich, Germany) to stain the nuclei. Antibodiesused for the staining are as follows: mouse anti-gephyrin (1:50 cell cul-ture supernatant, clone 3B11); rabbit anti-gephyrin (1:1000, rGeph-C,

gephyrin gene impair GABAergic synaptic inhibition in patients with0.1016/j.nbd.2014.02.001

T

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195Q10

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253Q11

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294Q12

295

296

297

298

299

300

3B. Dejanovic et al. / Neurobiology of Disease xxx (2014) xxx–xxx

UNCO

RREC

epitope EDLPSPPPPLSPPP, Eurogentec, Belgium); guinea-pig anti-GABAA-gamma2 receptor (Jean-Marc Fritschy, ETH, Zurich); rabbitanti-VGAT (1:500, Synaptic Systems, Göttingen, Germany); and mouseanti-myc-tag (1:10 cell culture supernatant, clone 9E10). Secondary an-tibodies were all goat raised Alexa Fluor antibodies (1:500, Invitrogen,Carlsbad, CA, USA).

Image analysis and quantification

Images were acquired with a confocal laser scanning microscope(Nikon A1) at 60× using a resolution of 1024 × 1024. Images wereprocessed using the software NIS Elements (Nikon). For analysis of neu-ronal cultures at least two independent preparations per conditionwereused. Imageswere acquired as a z-stackwith three optical sectionswith0.5 μm steps. Maximum intensity projections were created and ana-lyzed using NIS Elements software. Usually two 20 × 5 μm regions ofinterest (ROI) were placed on dendrites and clusters were countedusing the analyze particles option in NIS Elements. For statistical analy-sis, gephyrin clusters were compared pairwise between EGFP and Δ5–9geph expressing neurons. Mean valueswere compared for significanceusing Student's t-test with the software SigmaPlot (Systat Software,Chicago, IL, USA). Errors are presented as standard error of the mean(SEM). Significance levels are indicated as *P b 0.01, **P b 0.05, and***P b 0.001. Images were processed with the software ImageJ (NIH).

Co-immunoprecipitation

Protein extracts were prepared in an IP-buffer (PBS, 0.2% Triton X-100, protease and phosphatase inhibitor cocktail (Roche, Mannheim,Germany)). The post-nuclear fractionwas incubated with primary anti-bodies for 1 h at room-temperature or overnight at 4 °C. 20 μl Protein-GSepharose beads were incubated with the extracts for 2 h at roomtemperature. After a brief centrifugation (500 g for 3 min), theimmune-beads were washed three times with the IP buffer. Theadsorbed proteins were eluted from the immune-beads by boiling in50 μl SDS-loading buffer. Immunoprecipitated samples were subjectedto SDS-PAGE followed by immunoblotting.

Western Blot analysis and antibodies

For immunoblotting, a standard protocol was followed and detec-tion was carried out using chemiluminescence and an ECL systemwith a cooled CCD camera (Decon Science Tec, Germany). The followingprimary antibodies were used and diluted in TBS-Tween containing 1%dry-milk: mouse anti-gephyrin (1:50, clone 3B11 cell culture superna-tant), rabbit anti-β-tubulin (1:100, Santa Cruz, Santa Cruz, CA, USA),rabbit anti-GFP (1:5000, Abcam, Cambridge, UK), and mouse anti-myc-tag (1:10 cell culture supernatant, clone 9E10). As secondary anti-bodies anti-mouse or anti-rabbit conjugates were used in a 1:10,000dilutions in TBS-Tween containing 1% dry milk.

Results

Detection of exonic GPHNmicrodeletions and mRNA transcription analysis

We screened the genomic GPHN sequence (14q23.3:66,974,124–67,648,524, reference coding sequence: NM_020806, CCDS9777.1;hg19) for microdeletions (size N 40 kb, number of probe sets N 20)using high coverage microarrays from 1469 unrelated European pa-tients with common IGE syndromes and 2256 German population con-trols. We identified two hemizygous exonic GPHNmicrodeletions in theIGE cohort: i) a 129 kb microdeletion (chr14:67,314,917–67,443,729)affecting GPHN exons 5–9 (Δ5–9) in a German male patient with juve-nile myoclonic epilepsy (JME, Family 1), and ii) a 158 kb microdeletion(chr14:67,136,658–67,295,196) encompassingGPHN exons 2–3 (Δ2–3)in a German male patient with myoclonic astatic epilepsy (Doose

Please cite this article as: Dejanovic, B., et al., Exonic microdeletions of theidiopathic generalized epilepsy, Neurobiol. Dis. (2014), http://dx.doi.org/1

ED P

RO

OF

syndrome, Family 2) (Fig. 1A). Genome-wide screening for large CNVs(size N 500 kb), revealed no additional known pathogenic CNVs previ-ously associated with epilepsy in both index patients (de Kovel et al.,2010; Heinzen et al., 2010; Mefford et al., 2007, 2010, 2011). None ofthe 2256 German controls carried a microdeletion at the GPHN locus.The hemizygous copy number state of theGPHN deletionswas validatedby TaqMan quantitative PCR (qPCR).

Exome sequencing of the GPHN coding regions and exon/intronboundaries did not reveal any additional indels and rare missense mu-tations in the parent–offspring trio of Family 1. Direct blood cDNA am-plification of GPHN exons 4–13 for Family 1 and exons 1–7 for Family2 followed by Sanger sequencing of the major amplicons confirmedthe transcription of the truncatedGPHN variants and revealed a paternaltransmission of both microdeletions (Figs. 1B–D and SupplementaryMaterial S1). On the protein level, the GPHN Δ5–9 microdeletion leadsto the truncation of the gephyrin G-domain and C-domain (hereaftertermed “Δ5–9geph” — Fig. 1E) but retains an intact E-domain. TheGPHN Δ2–3 microdeletion results in a frameshift and a premature stopcodon (p.V22Gfs*7, Fig. 1F) given that residue 22 is encoded by a splitcodon composed of exons 1 and 2. Consequently, the short N-terminalpeptide of gephyrin (p.V22Gfs*7; full length: 736 amino acids) lacksall structural domains that are required for the physiological oligomer-ization and clustering processes at inhibitory synapses.

Clinical features of the GPHN deletion carriers and segregation analysis

Family 1 (GPHN Δ5–9 deletion)The pedigree of Family 1 is presented in Fig. 1A. The non-

consanguineous German parents (57-year-old father, 55-year-oldmother) and the patient's 30-year-old sister had no unequivocal historyof seizures. The GPHN Δ5–9 deletion was paternally inherited. Themother showed a diploid copy number status at the GPHN locus. DNAfrom the clinically unaffected sister was not available. The 33-year-oldmale index IGE patient was born at term after an uneventful pregnancyanddelivery. Hehad recurrent simple febrile seizures from the age of sixmonths to six years, unprovoked generalized tonic–clonic seizures sincethe age of five years and myoclonic seizures recorded by video-EEGstarting at the age of 16 years. Complete remission of epileptic seizuresby monotherapy with levetiracetam was reached since the age of24 years. Neurological examinations and brainMRIs (3-T)were normal.The epilepsy diagnosis was classified as JME. Neuropsychiatric anddevelopmental comorbidities were mild deficits in motor coordination,hyperactivity, and learning difficulties during childhood. Neuropsycho-logical testing at the age of 19 years showed normal cognitive abilitieswith average performance scores. Recurrent episodes of major depres-sion with suicide attempts started at the age of 16 years, accompaniedby generalized anxiety, panic attacks and phobic vertigo. Non-epilepticpsychogenic attacks became evident since the age of 32 years. Thepatient's father experienced three unprovoked events of loss of con-sciousness between 4 and 6 years of age. There is a strong family historyof migraine affecting the maternal branch of the family. Any history ofpsychiatric disorders, dysmorphism, developmental delay, intellectualdisability or miscarriages was absent.

Neurophysiological examinationsTo evaluate cortical excitability, we performed paired-pulse

transcranial magnetic stimulation (TMS) in Family 1 (Reutens andBerkovic, 1992). For both, the JME index-patient and his father,intra-cortical inhibition was significantly reduced as compared tothat in six healthy controls (P b 0.001), while intra-cortical facilitationwas more pronounced than that in controls (P b 0.001 — Fig. 2). Like-wise, diffusion tensor imaging (DTI) showed an enhanced structuralconnectivity between the mesial frontal region and the descendingmotor pathways in the index patient and his father that was beyondthe range seen in healthy controls (Supplementary Material Fig. S2).For the patient the peak voxels reached statistical significance for the

gephyrin gene impair GABAergic synaptic inhibition in patients with0.1016/j.nbd.2014.02.001

RECTED P

RO

OF

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

Fig. 1.Genetic analyses ofGPHNmicrodeletions. (A) Genomic localization (hg19) ofGPHNmicrodeletions identified in the IGE index patients. The deletion segregating in Family 1 is span-ning approximately 129 kb involving exons 5–9. The microdeletion of Family 2 is approximately 158 kb in size and deletes exons 2–3. (B) Familial segregation of exonic GPHNmicrodeletions. Deletion carriers are indicated by copy number state (CN) = 1, whereas CN = 2 represents an analyzed individual with a diploid copy number state;black symbols = affected by IGE; TMS path. = increased cortical excitability without seizures; arrows highlight index patients. (C) Electropherogram of the cDNA Δ5–9GPHN transcript sequence of the index patient F1II.1 and his father F1I.1, confirming the transcription of a truncated Δ5–9 GPHN transcript. Arrow separates the sequenceof exon 4 from the fused adjacent sequence of exon 10. The exon annotation refers to the genomic organization of GPHN (NM_020806.4). (D) Electropherogram of thecDNA Δ2–3 GPHN transcript sequence of the index patient F2II.2, confirming the transcription of a truncated Δ2–3 GPHN transcript (NM_020806.4). Arrow separates the se-quence of exon 1 from the fused adjacent sequence of exon 4. (E) Schematic structure of regular gephyrin and Δ5–9geph with numbering of deleted residues in Δ5–9geph.(F) Sequence of the truncated Δ2–3 GPHN transcript expressed by the father F2I.1 and his son F2II.2. Residue Val22 is encoded by a splitted codon composed of exons 1 and 2;the GPHN Δ2–3 variant results in a frameshift and a premature stop codon (p.V22Gfs*7, NP_001019389.1).

4 B. Dejanovic et al. / Neurobiology of Disease xxx (2014) xxx–xxx

R

connectivity to the motor pathways (P b 0.001), whereas the father'svalues showed only a trend but did not reach statistical significance.

O 323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

UNCFamily 2 (GPHN Δ2–3 deletion)

The pedigree of Family 2 is presented in Fig. 1A. The non-consanguineous parents (50-year-old father, 48-year-old mother) areof German ancestry. The GPHN Δ2–3 deletion was paternally inherited.The mother and the brother had diploid copy numbers at the GPHNlocus. The 16-year-old male index IGE patient was affected by febrileseizures at the age of 15 months and by myoclonic astatic seizuresstarting at the age of 1.5 years. Since antiepileptic treatment withvalproate, which was started shortly after epilepsy-onset, he isseizure-free. The EEG exhibited irregular sharp–slow wave dis-charges at 3-Hz. The patient showed a normal physical but delayedcognitive development with persistent learning disability. His epilepsywas classified as idiopathic myoclonic-astatic epilepsy (Doose syn-drome). The father had no history of seizures and a normal psychomo-tor development. Themother experienced three seizures during infancywith spontaneous remission and normal psychomotor development.Her EEGs displayed interictal epileptiform discharges. The 18-year-oldbrother of the index patient had two febrile seizures during childhood

Please cite this article as: Dejanovic, B., et al., Exonic microdeletions of theidiopathic generalized epilepsy, Neurobiol. Dis. (2014), http://dx.doi.org/1

while no subsequent history of afebrile seizures, normal psychomotordevelopment and EEG were reported. None of the family membershad a history of psychiatric disorders. The members of Family 2 werenot available for TMS and exome/GPHN mutation screening.

Molecular and biochemical characterization of the GPHN Δ5–9 andΔ2–3 variants

To elucidate the molecular effects of the GPHN microdeletions onoligomerization and synaptic clustering, we examined the functionalalterations of the truncated gephyrin variants and their interactiveinfluences on the functions of regular gephyrin. Upon overexpressionin non-neuronal HEK293 cells, regular gephyrin oligomerizes andforms cytosolic protein-aggregates (“blobs”, Fig. 3A). In contrast,overexpressed GFP-tagged Δ5–9geph (Δ5–9geph-GFP) and Δ2–3geph(Δ2–3geph-GFP) were homogenously distributed in the cytosol ofHEK293 cells (Fig. 3A), suggesting that both truncated gephyrin variantsare unable to oligomerize due to the loss of the G-domain trimerizationinterface encoded mainly by GPHN exon 6 (Schwarz et al., 2001). Next,we co-expressed myc-tagged regular gephyrin with Δ5–9geph-GFP orΔ2–3geph-GFP in HEK293 cells. We found that in the presence of Δ5–

gephyrin gene impair GABAergic synaptic inhibition in patients with0.1016/j.nbd.2014.02.001

T340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

Fig. 2.Paired-pulse transcranialmagnetic stimulation. The relative variation of the averagepeak-to-peak amplitude of the motor evoked potential in relation to the baseline (set at100%), is plotted as a percentage, for the conditions “intracortical inhibition” (ISI 3 ms),white bars; and “intracortical facilitation” (ISI 13 ms), black bars. In Family 1, both theindex patient and his father exhibited an increased cortical excitability, with decreasedICI and increased ICF, statistically significant compared to healthy controls (***P b 0.001,ANOVA).

5B. Dejanovic et al. / Neurobiology of Disease xxx (2014) xxx–xxx

9geph-GFP, gephyrin blobs were dissolved resulting in either diffuseregular gephyrin or the formation of microclusters co-localizing withΔ5–9geph-GFP (Figs. 3B, D). Therefore, the observed dominant-negative effect of Δ5–9geph on regular gephyrin oligomerization ismediated by direct interaction, as shown by co-immunoprecipitation(Fig. 3E). By contrast, the short N-terminal Δ2–3geph (p.V22Gfs*7)peptide did not co-localize with regular gephyrin and did not influenceits ability to form blobs (Figs. 3C, D). Likewise, Δ2–3geph did not inter-act with regular gephyrin (Fig. 3E) and did not perturb the functionalproperties of regular gephyrin.

UNCO

RREC

Fig. 3.Δ5–9geph, but notΔ2–3, interactswith regular gephyrin and affects oligomerization in ain HEK293 cells, while Δ5–9geph-GFP and Δ2–3geph-GFP are diffusively distributed. (B) Upoleading to either completely diffuse regular gephyrin (B1) or the formation of microclustas in (B), Δ2–3geph-GFP does not co-localize with gephyrin-myc. (D) Quantification of morpGephyrin alone or co-expressed with Δ2–3geph forms mainly blobs, whereas in the presencderived from three independent transfectionswere quantified. (E)Δ5–9geph-GFP, but notΔ2–co-expression of both proteins. Co-immunoprecipitation of gephyrin-myc by GFP-tagged regu

Please cite this article as: Dejanovic, B., et al., Exonic microdeletions of theidiopathic generalized epilepsy, Neurobiol. Dis. (2014), http://dx.doi.org/1

ED P

RO

OF

Next, we investigated the consequences of dominant-negative Δ5–9geph action on neuronal gephyrin clustering. When expressed in cul-tured neurons, gephyrin forms postsynaptic clusters along the neuritesas well as the soma. By contrast, Δ5–9geph-GFP was diffusively distrib-uted and filled all morphological structures, i.e. spines (Fig. 4A). Wethen expressed Δ5–9geph-GFP or GFP alone in cultured hippocampalneurons and immunostained endogenous gephyrin. To distinguish be-tween endogenous gephyrin andΔ5–9geph-GFP,we used amonoclonalantibody against a peptide in the C-domain that is deleted in Δ5–9geph(rGeph-C, Fig. 4B). The expression of Δ5–9geph-GFP significantlyreduced the number of endogenous gephyrin clusters confirming thepreviously stated dominant-negative effect (Figs. 4C, D). Moreover,the remaining gephyrin clusters were significantly reduced in size indi-cating that Δ5–9geph disturbs the clustering of regular gephyrin(Fig. 4E). To determine how the Δ5–9geph-GFP expression influencesthe clustering of GABAARs, we visualized the γ2-subunit (Fig. 4F),which plays a central role in gephyrin-dependent postsynaptic cluster-ing of GABAARs (Alldred et al., 2005; Essrich et al., 1998). The numberand cluster size of γ2-containing GABAARs were significantly decreasedin the presence of Δ5–9geph-GFP (Figs. 4G, H). Together with theobserved reduced quantity of gephyrin clusters, our data imply thatreceptor homeostasis responds to decreased amounts of clusteredgephyrin.

To investigate whether the Δ5–9geph-mediated reduction inGABAergic innervation influences synaptic spines, typical morphologi-cal structures of the excitatory synapse (Kasai et al., 2003), we visual-ized spines in control and Δ5–9geph-GFP-expressing neurons bythresholding the EGFP-signal to an identical value. Spine headmorphol-ogy and amount were similar in EGFP and Δ5–9geph-expressingneurons, suggesting that diminished GABAergic synapses did not inter-fere with synaptic spines (Figs. 4I–K).

Besides its synaptic function, gephyrin catalyzes the last step of themolybdenum cofactor (Moco) biosynthesis being essential for the activ-ity of four molybdenum-dependent enzymes (Belaidi and Schwarz,2013; Reiss et al., 2001; Stallmeyer et al., 1999). We measured typicalurinary biomarkers of twoMoco-dependent enzymes, xanthine oxidase

dominant-negativemanner. (A)Myc-tagged gephyrin forms large “blobs” upon expressionn co-expression, Δ5–9geph-GFP dominant negatively affected gephyrin oligomerizationers harboring regular and truncated gephyrin-variants (B2). (C) Upon co-expressionhological appearance of regular gephyrin upon expression in HEK293 cells as indicated.e of Δ5–9geph gephyrin blob formation is greatly decreased. 100 cells of each condition3geph-GFP, coimmunoprecipitatedmyc-tagged gephyrin in HEK293 cell-lysates followinglar gephyrin served as the positive control.

gephyrin gene impair GABAergic synaptic inhibition in patients with0.1016/j.nbd.2014.02.001

ORRECTED P

RO

OF

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

Fig. 4. GABAergic innervation is decreased upon expression of Δ5–9geph-GFP. (A) Morphology and localization of GFP-tagged regular gephyrin and Δ5–9geph in cultured hippocampalneurons. Insets showhigh-magnification of the dendrites. Arrowheads point to spine heads,morphological structures of excitatory synapses. Scale bar=20 μm. (B) Schematic structure ofgephyrin and the epitope of the rGeph-C antibody, whichwas used to distinguish between endogenous gephyrin andΔ5–9geph. Representative dendrites (20 μm) of EGFP- orΔ5–9geph-expressing cultured hippocampal neurons (11 + 2 DIV) immunostained for endogenous gephyrin (C) or GABAAR γ2 subunit (F). (C–H)Δ5–9geph-GFP significantly reduced the numberand size of endogenous gephyrin clusters and GABAA γ2-containing receptors. (I–K) Morphology and numbers of spine heads were not affected by Δ5–9geph-GFP. Arrowheads point toexemplary spines that have been quantified. Scale bar = 10 μm. All data are expressed as mean± SEM (**P b 0.05; ***P b 0.001, Student's t-test; n = 14–18 neurons per condition fromtwo independent preparations).

6 B. Dejanovic et al. / Neurobiology of Disease xxx (2014) xxx–xxx

UNC

(uric acid, xanthine) and sulfite oxidase (S-sulfocysteine) as well as theMoco degradation product urothion (Bamforth et al., 1990; Veldmanet al., 2010) in the patient and his parents of Family 1. Concentrationsof all analyzed metabolites were inconspicuous, suggesting that thehemizygous GPHN Δ5–9 variant did not cause any subclinical Mocodeficiency phenotype in the affected patient (Supplementary MaterialFig. S3).

Discussion

We present a comprehensive clinical, genetic, functional and neuro-physiological characterization of two IGE patients with novel hemizy-gous microdeletions encompassing GPHN exons that code for theN-terminal gephyrin G-domain. Our finding strengthens a previousreport of six unrelated subjectswith awide range of neurodevelopmentaldisorders (ASD, schizophrenia, seizures) who carried hemizygous GPHNmicrodeletions affecting the gephyrin G-domain (Lionel et al., 2013),

Please cite this article as: Dejanovic, B., et al., Exonic microdeletions of theidiopathic generalized epilepsy, Neurobiol. Dis. (2014), http://dx.doi.org/1

and extends the phenotypic spectrum related to N-terminal GPHNmicrodeletions by IGE syndromes (Supplementary Material Fig. S4).The study reported by Lionel et al. deduced evidence of pathogenicityof GPHN microdeletions from the significant association with neuro-developmental disorders (P = 0.009; 6/8775 cases versus 3/27,019controls) and found that three out of five GPHN microdeletions testedfor inheritance arose as de novo events (Lionel et al., 2013). In thisstudy, we present additional functional data elucidating the molecularmechanisms by which the N-terminal GPHN deletions cause dysfunc-tional GABAergic synaptic inhibition and thereby increase susceptibilityof IGE.

Molecular characterization demonstrates that the truncated GPHNΔ5–9 variant (Family 1) exerts a dominant-negative effect on the struc-tural synaptic organization of gephyrin and GABAAR clusters. Gephyrinoligomerization is thought to be a concerted interaction mediated byG-domain trimerization (Schwarz et al., 2001), G- and E-domain inter-action (Fritschy et al., 2008), E-domain dimerization (Lardi-Studler

gephyrin gene impair GABAergic synaptic inhibition in patients with0.1016/j.nbd.2014.02.001

T

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

507

508

509

510Q13

511

512

513

514

515

516

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

541

542

543

544

545

546

547

548

549

7B. Dejanovic et al. / Neurobiology of Disease xxx (2014) xxx–xxx

UNCO

RREC

et al., 2007) and C-domain binding to other proteins controlled by post-translational modifications (Herweg and Schwarz, 2012). The truncatedΔ5–9geph protein itself is not able to form clusters and impairs the olig-omerization process of regular gephyrin (Fig. 4),which can be explainedby the loss of the G-domain trimerization interface encoded mainly byexon 6 (Schwarz et al., 2001). Instead, Δ5–9geph probably forms irreg-ular dimers by interaction of the intact E-domain with regular gephyrinE-domains (Lardi-Studler et al., 2007). Thereby, it stochastically affectsoligomerization of regular gephyrin in a dominant-negative manner.Importantly, the size of the remaining gephyrin and GABAAR clusterswas significantly reduced, which will probably impair the recentlydescribed homeostatic regulation of gephyrin scaffolds and normal de-velopment of the neuronal inhibitory GABAergic circuits (Vlachoset al., 2012). Consistentlywith ourfindings, it was speculated that cellu-lar stressors such as elevated temperature or alkalosis, which can arisefrom seizure activity, might induce irregular exon skipping in GPHNmRNA leading to deletions within the G- and C-domain thus causing adominant-negative effect on the oligomerization of synaptic gephyrinclusters (Forstera et al., 2010). Moreover, given the constitutionalexpression of the truncated Δ5–9geph protein, it is also likely thatduring development the Δ5–9geph expression negatively affects theestablishment of certain networks which cumulatively contribute tothe genetic variance of the IGE phenotype.

In Family 2, the GPHNΔ2–3 variant causes a frameshift with prema-ture stop codon, resulting in a truncatedGPHNprotein,which consists ofthe first 21 residues encoded by exon 1 followed by seven residuesencoded by out of frame codons before the premature stop codon(p.V22Gfs*7 — Fig. 1F). Such a severely truncated gephyrin lacks allrelevant structural domains for normal functioning. In contrast to theGPHN Δ5–9 variant from Family 1, it does not confer a dominant-negative effect on regular gephyrin oligomerization and thus does notinfluence gephyrin's functions. Most likely, the truncated GPHN Δ2–3transcript will be degraded in vivo by nonsense-mediated mRNAdecay (NMD), a translation-dependent posttranscriptional processthat selectively recognizes and degrades mRNAs whose open readingframe is truncated by a premature translation termination codon(Chang et al., 2007), resulting in negligible amounts of the peptide.Given that we were able to amplify the GPHN Δ2–3 transcript fromblood cell cDNA (Supplementary Material Fig. S2), at least a partialexpression of the GPHN Δ2–3 transcript is possible. Irrespective ofthese two pathways, the resulting loss-of-function of the GPHN Δ2–3variant suggests that the epileptogenic effect most likely results fromhaplo-insufficiency of the gene.

Notably, five out of six GPHN multi-exon deletions (Δ2–5 (n = 2),Δ3–8, Δ3–11 and Δ3–12) reported by Lionel et al. (2013) in patientswith neurodevelopmental disorders result in frameshifts and prema-ture stop codons, causing truncated GPHN proteins including the first21 or 47 amino acids encoded by exon 1 or exons 1–2, respectively.Although truncated gephyrins comprising the first 47 amino acidshave the capacity to impair the oligomerization of regular gephyrin(Forstera et al., 2010), the expression of these frameshift-causingGPHN multi-exon deletions will probably be negligible due to NMD.Taken together, patients with GPHN microdeletions encompassingexons 1–3 most likely exhibit their pathogenic effect by haplo-insufficiency due to a loss-of-function of the essential functionaldomains of the truncated gephyrin protein.

Given the intriguing impact of an impaired GABAergic synaptictransmission in epileptogenesis (Noebels, 2003), the reduced amountof synaptic GABAARs in the presence of truncated Δ5–9 gephyrin exertsa functionally convergent effect similar to that observed by mutations ofgenes encodingGABAAR subunits (GABRA1,GABRB3,GABRG2 andGABRD)in rare families with dominantly inherited IGE syndromes (for reviewssee Cossette et al., 2012; Macdonald et al., 2010). On the molecularlevel, GABR gene mutations generally lead to a reduced surface expres-sion of GABAARs, which results in the reduction of the amplitude ofGABA-evoked currents. Although GPHN deletions should also affect

Please cite this article as: Dejanovic, B., et al., Exonic microdeletions of theidiopathic generalized epilepsy, Neurobiol. Dis. (2014), http://dx.doi.org/1

ED P

RO

OF

glycine receptor clustering, this seems to have no significant pathogenicconsequences considering that typical symptoms of glycinergic deficits(Lynch, 2004), such as hyperekplexia, are missing.

Interestingly, in Family 1, the clinically unaffected father as well asthe JME-affected index patient carrying the GPHN Δ5–9 deletion exhib-ited an impaired cortical inhibition leading to neuronal hyperexcitabili-ty, as demonstrated by TMS, which mainly reflects the functional stateof GABAergic interneuronal circuits (Ziemann et al., 1996). This imbal-ance of the inhibitory/excitatory ratio corresponds well with the exper-imentally validated dominant-negative effect of the GPHN Δ5–9deletion on the reduced size and number of GABAAR clusters at inhibito-ry synapses. Moreover, diffusion tensor imaging showed a statisticallysignificant increase in the structural connectivity of the mesial frontalregion and the descending motor circuits in the JME-affected indexpatient and a slightly enhanced connectivity in the father of Family 1.An altered structural connectivity of the mesial frontal region hasrecently been found in JME (Vollmar et al., 2012) and may explain thepredominant manifestation of myoclonic seizures in the JME-indexpatient of Family 1. Altogether, our present findings add to convergentlines of evidence that dysfunctional GABAergic synaptic inhibitionrepresents a key pathogenic process shared by a wide spectrum ofneurodevelopmental disorders, such as epilepsy, ASD, schizophre-nia, and intellectual disability (Coghlan et al., 2012). Consistently, ahigh prevalence of epilepsy has been found in autism spectrum dis-order (ASD), ranging from 21% in ASD subjects with an intellectualdisability to 8% in those without intellectual disability (Amiet et al.,2008). Notably, the index patient of Family 1 not only had a typicalJME but also exhibited a broad range of psychiatric and behavioraldisorders.

Carriers of GPHN microdeletions encompassing the N-terminalexons display a high phenotypic variability and incomplete penetrance,strongly depending on the genetic background. This pleiotropic expres-sion has been observed for all microdeletions (at 1q21.1, 15q11.2,15q13.3, 16p11.2, 16p13.11, 22q11.2, A2PB1,NRXN1) that have been as-sociated with IGE or other neurodevelopmental disorders (Coe et al.,2012b). Most likely, these microdeletions impair neurodevelopmentalprocesses in a rather unspecific manner and contribute to the geneticvariance of awide spectrumof neurodevelopmental disorders. The indi-vidual disease phenotype is probably further specified by the interplaywith genetic background effects and environmental influences follow-ing an oligo-/polygenic inheritance model with substantial geneticheterogeneity (Dibbens et al., 2007). In particular, bi-parental geneticrisk factors may contribute to the epilepsy phenotype of the index sub-ject. Accordingly, segregation analysis in Family 2 suggests a complexbi-parental inheritance pattern composed of a paternal inheritance oftheGPHNΔ2–3 deletion and a predominant epileptogenic susceptibilityfactor(s) transmitted by the mother, who exhibited afebrile seizuresduring infancy. Likewise, the brother showing a diploid GPHN copystatus had a history of febrile seizures. Similar inheritance patternswere also reported earlier for exonic GPHN microdeletions (Lionelet al., 2013) and other recurrent microdeletions increasing the risk forIGE (Cossette et al., 2012;Macdonald et al., 2010). Aswe did not observemetabolic Moco deficiency symptoms (Veldman et al., 2010), the hemi-zygous microdeletions of N-terminal GPHN exons presumably affectonly the synaptic functions of gephyrin. However, it cannot be excludedthat minor alterations in the Moco-enzyme activity during brain devel-opment might contribute to disease manifestation.

In summary, our study strengthens a previous statistical associationof GPHN microdeletions affecting the gephyrin G-domain with a widerange of neuropsychiatric disorders and expands this spectrum to IGEsyndromes (Lionel et al., 2013). GPHN microdeletions affecting theN-terminal gephyrin domain confer a susceptibility effect with highphenotypic variability and incomplete penetrance depending onthe type of the exonic GPHN deletion and genetic background. We pro-vide genetic, functional and neurophysiological evidence that structuralexonic microdeletions affecting the gephyrin G-domain can increase

gephyrin gene impair GABAergic synaptic inhibition in patients with0.1016/j.nbd.2014.02.001

T

550

551

552

553Q14

554

555

556

557

558

559

560

561

562

563

564

565

566

567

568

569

570

571

572

573

574

575

576

577

578

579

580

581

582

583

584

585

586

587

588

589

590

591

592

593

594

595

596

597

598

599

600

601

602

603

604

605

606

607

608

609

610

611

612

613

614

615

616

617

618

619

620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684

8 B. Dejanovic et al. / Neurobiology of Disease xxx (2014) xxx–xxx

UNCO

RREC

neuronal excitability by an impairment of GABAergic synaptic inhibi-tion, and thereby confer susceptibility for IGE.

Funding

This work was supported by grants from the European Community[FP6 Integrated Project EPICURE, grant LSHM-CT-2006-037315 to T.S.],the German Research Foundation (DFG) within the EUROCORES Pro-gramme EuroEPINOMICS [grants NE416/5-1 to B.N., SA434/5-1 to T.S.,NU50/8-1 to P.N.], the SFB635 [grant A11 to G.S.]; the German FederalMinistry of Education and Research, National Genome Research Net-work [NGFNplus: EMINet, grants 01GS08120 to T.S., and 01GS08121to P.N.; IntenC, grant 01DL12011 to T.S.]; the PopGen biobank [grantto A.F.], the Center for Molecular Medicine Cologne [to G.S.] and theFonds der Chemischen Industrie [to G.S.]. The PopGen project receivedinfrastructure support through the German Research Foundation excel-lence cluster “Inflammation at Interfaces”. The KORA research platform(KORA, Cooperative Research in the Region of Augsburg) was initiatedand financed by the Helmholtz Zentrum München — German ResearchCenter for Environmental Health, which is funded by theGerman Feder-al Ministry of Education and Research and by the State of Bavaria; thisresearch was supported within the Munich Center of Health Sciences(MC Health) as part of LMUinnovativ.

Conflict of interest statement

R.S. has received support for congress participation and fees forlectures from EISAI and had a consultancy agreement with UCB.The other authors declare no conflict of interest.

Acknowledgments

We thank Prof. Jean-Marc Fritschy (University of Zurich, Switzerland)for providing the GABAAR γ2 antibody, Joana Stegemann for technicalsupport and Ruth Ruscheweyh for help with the transcranial magneticstimulation.

Appendix A

EPICURE Integrated Project (contributing centers listed by country):Department of Clinical Neurology (Fritz Zimprich) and Department ofPediatrics and Neonatology (Martina Mörzinger, Martha Feucht),Medical University of Vienna, Währinger Gürtel 18-20, A-1090 Vienna,Austria. VIB Department of Molecular Genetics, University of Antwerp,Universiteitsplein 1, 2610 Antwerpen, Belgium (Arvid Suls, SarahWeckhuysen, Lieve Claes, Liesbet Deprez, Katrien Smets, Tine VanDyck, Tine Deconinck, Peter De Jonghe). Research Department, DanishEpilepsy Centre, Dianalund Kolonivej 7, 4293 Dianalund, Denmark(Rikke S Møller, Laura L. Klitten, Helle Hjalgrim). Wilhelm JohannsenCentre for Functional Genome Research, University of Copenhagen,Blegdamsvej 3, 2200 Copenhagen N, Denmark (Rikke SMøller). Depart-ment of Neuropediatrics, University Medical Center Schleswig-Holstein(Kiel Campus), Schwanenweg 20, 24105 Kiel, Germany (Ingo Helbig,Hiltrud Muhle, Philipp Ostertag, Sarah von Spiczak, Ulrich Stephani).Cologne Center for Genomics, University of Cologne, Weyertal 115b,50931 Cologne, Germany (Peter Nürnberg, Thomas Sander, HolgerTrucks). Department of Epileptology, University Clinics Bonn, SigmundFreud Strasse 1, 53105 Bonn, Germany (Christian E. Elger, Ailing A.Kleefuß-Lie, Wolfram S. Kunz, Rainer Surges). Department of Neurolo-gy, Charité University Medicine, Campus Virchow Clinic, HumboldtUniversity of Berlin, Augustenburger Platz 1, 13353 Berlin, Germany(Verena Gaus, Dieter Janz, Thomas Sander, Bettina Schmitz). Epilepsy-Center Hessen, Department of Neurology, Philipps-University Marburg,35043 Marburg, Germany (Karl Martin Klein, Philipp S. Reif, WolfgangH. Oertel, Hajo M. Hamer, Felix Rosenow). Abteilung Neurologie mitSchwerpunkt Epileptologie, Hertie Institut für klinische Hirnforschung,

Please cite this article as: Dejanovic, B., et al., Exonic microdeletions of theidiopathic generalized epilepsy, Neurobiol. Dis. (2014), http://dx.doi.org/1

ED P

RO

OF

Universität Tübingen, Hoppe-Seyler-Str. 3, 72076 Tübingen, Germany(Felicitas Becker, Yvonne Weber, Holger Lerche). NetherlandsSection Complex Genetics, Department of Medical Genetics, UniversityMedical Center Utrecht, Str. 2.112 Universiteitsweg 100, 3584 CGUtrecht, The Netherlands (Bobby P.C. Koeleman, Carolien de Kovel,Dick Lindhout). SEIN Epilepsy Institute in the Netherlands, P.O. Box540, 2130AM Hoofddorp, Netherlands (Dick Lindhout).

Appendix B. Supplementary data

Further data are provided as SupplementaryMaterial. Supplementarydata associated with this article can be found, in the online version, athttp://dx.doi.org/10.1016/j.nbd.2014.02.001.

References

Alldred, M.J., Mulder-Rosi, J., Lingenfelter, S.E., Chen, G., Luscher, B., 2005. Distinctgamma2 subunit domains mediate clustering and synaptic function of postsynapticGABAA receptors and gephyrin. J. Neurosci. 25, 594–603. http://dx.doi.org/10.1523/JNEUROSCI.4011-04.2005.

Amiet, C., et al., 2008. Epilepsy in autism is associated with intellectual disability andgender: evidence from a meta-analysis. Biol. Psychiatry 64, 577–582. http://dx.doi.org/10.1016/j.biopsych.2008.04.030.

Proposal for revised classification of epilepsies and epileptic syndromes. Commission onClassification and Terminology of the International League Against Epilepsy. Epilepsia30, 389–399 (1989).

Bamforth, F.J., et al., 1990. Biochemical investigation of a child with molybdenum cofactordeficiency. Clin. Biochem. 23, 537–542.

Belaidi, A.A., Schwarz, G., 2013. Metal insertion into the molybdenum cofactor: product–substrate channelling demonstrates the functional origin of domain fusion ingephyrin. Biochem. J. 450, 149–157. http://dx.doi.org/10.1042/BJ20121078.

Chang, Y.F., Imam, J.S., Wilkinson, M.F., 2007. The nonsense-mediated decay RNA surveil-lance pathway. Annu. Rev. Biochem. 76, 51–74. http://dx.doi.org/10.1146/annurev.biochem.76.050106.093909.

Coe, B.P., Girirajan, S., Eichler, E.E., 2012a. The genetic variability and commonality ofneurodevelopmental disease. Am. J. Med. Genet. C: Semin. Med. Genet. 160C,118–129. http://dx.doi.org/10.1002/ajmg.c.31327.

Coe, B.P., Girirajan, S., Eichler, E.E., 2012b. A genetic model for neurodevelopmentaldisease. Curr. Opin. Neurobiol. 22, 829–836. http://dx.doi.org/10.1016/j.conb.2012.04.007.

Coghlan, S., et al., 2012. GABA system dysfunction in autism and related disorders: fromsynapse to symptoms. Neurosci. Biobehav. Rev. 36, 2044–2055. http://dx.doi.org/10.1016/j.neubiorev.2012.07.005.

Cossette, P., Lachance-Touchette, P., Rouleau, G.A., 2012. In: Noebels, J.L., et al. (Eds.),Jasper's Basic Mechanisms of the Epilepsies.

de Kovel, C.G., et al., 2010. Recurrent microdeletions at 15q11.2 and 16p13.11 predisposeto idiopathic generalized epilepsies. Brain 133, 23–32. http://dx.doi.org/10.1093/brain/awp262.

Dibbens, L.M., Heron, S.E., Mulley, J.C., 2007. A polygenic heterogeneitymodel for commonepilepsies with complex genetics. Genes Brain Behav. 6, 593–597. http://dx.doi.org/10.1111/j.1601-183X.2007.00333.x.

Essrich, C., Lorez, M., Benson, J.A., Fritschy, J.M., Luscher, B., 1998. Postsynaptic clusteringof major GABAA receptor subtypes requires the gamma 2 subunit and gephyrin. Nat.Neurosci. 1, 563–571. http://dx.doi.org/10.1038/2798.

Fang, M., et al., 2011. Downregulation of gephyrin in temporal lobe epilepsy neurons inhumans and a rat model. Synapse 65, 1006–1014. http://dx.doi.org/10.1002/syn.20928.

Forstera, B., et al., 2010. Irregular RNA splicing curtails postsynaptic gephyrin in the cornuammonis of patients with epilepsy. Brain 133, 3778–3794. http://dx.doi.org/10.1093/brain/awq298.

Fritschy, J.M., Harvey, R.J., Schwarz, G., 2008. Gephyrin: where do we stand, where do wego? Trends Neurosci. 31, 257–264. http://dx.doi.org/10.1016/j.tins.2008.02.006.

Gonzalez, M.I., Cruz Del Angel, Y., Brooks-Kayal, A., 2013. Down-regulation of gephyrinandGABAA receptor subunits during epileptogenesis in theCA1 region of hippocampus.Epilepsia 54, 616–624. http://dx.doi.org/10.1111/epi.12063.

Hauser, W.A., Annegers, J.F., Kurland, L.T., 1993. Incidence of epilepsy and unprovokedseizures in Rochester, Minnesota: 1935–1984. Epilepsia 34, 453–468.

Heinzen, E.L., et al., 2010. Rare deletions at 16p13.11 predispose to a diverse spectrum ofsporadic epilepsy syndromes. Am. J. Hum. Genet. 86, 707–718. http://dx.doi.org/10.1016/j.ajhg.2010.03.018.

Helbig, I., et al., 2009. 15q13.3 microdeletions increase risk of idiopathic generalizedepilepsy. Nat. Genet. 41, 160–162. http://dx.doi.org/10.1038/ng.292.

Herweg, J., Schwarz, G., 2012. Splice-specific glycine receptor binding, folding, and phos-phorylation of the scaffolding protein gephyrin. J. Biol. Chem. 287, 12645–12656.http://dx.doi.org/10.1074/jbc.M112.341826.

Jallon, P., Loiseau, P., Loiseau, J., 2001. Newly diagnosed unprovoked epileptic seizures: pre-sentation at diagnosis in CAROLE study. Coordination Active du Reseau ObservatoireLongitudinal de l' Epilepsie. Epilepsia 42, 464–475.

Kasai, H., Matsuzaki, M., Noguchi, J., Yasumatsu, N., Nakahara, H., 2003. Structure–stability–function relationships of dendritic spines. Trends Neurosci. 26, 360–368.http://dx.doi.org/10.1016/S0166-2236(03)00162-0.

gephyrin gene impair GABAergic synaptic inhibition in patients with0.1016/j.nbd.2014.02.001

685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720

721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753

9B. Dejanovic et al. / Neurobiology of Disease xxx (2014) xxx–xxx

Kirsch, J., Wolters, I., Triller, A., Betz, H., 1993. Gephyrin antisense oligonucleotidesprevent glycine receptor clustering in spinal neurons. Nature 366, 745–748.http://dx.doi.org/10.1038/366745a0.

Kowalczyk, S., et al., 2013. Direct binding of GABAA receptor beta2 and beta3 subunits togephyrin. Eur. J. Neurosci. 37, 544–554. http://dx.doi.org/10.1111/ejn.12078.

Lal, D., et al., 2013. Rare exonic deletions of the RBFOX1 gene increase risk of idiopathicgeneralized epilepsy. Epilepsia 54, 265–271. http://dx.doi.org/10.1111/epi.12084.

Lardi-Studler, B., et al., 2007. Vertebrate-specific sequences in the gephyrin E-domainregulate cytosolic aggregation and postsynaptic clustering. J. Cell Sci. 120, 1371–1382.http://dx.doi.org/10.1242/jcs.003905.

Lionel, A.C., et al., 2013. Rare exonic deletions implicate the synaptic organizer gephyrin(GPHN) in risk for autism, schizophrenia and seizures. Hum. Mol. Genet. 22,2055–2066. http://dx.doi.org/10.1093/hmg/ddt056.

Lynch, J.W., 2004.Molecular structure and function of the glycine receptor chloride channel.Physiol. Rev. 84, 1051–1095. http://dx.doi.org/10.1152/physrev.00042.2003.

Macdonald, R.L., Kang, J.Q., Gallagher, M.J., 2010. Mutations in GABAA receptor subunitsassociated with genetic epilepsies. J. Physiol. 588, 1861–1869. http://dx.doi.org/10.1113/jphysiol.2010.186999.

Mefford, H.C., et al., 2007. Recurrent reciprocal genomic rearrangements of 17q12 areassociated with renal disease, diabetes, and epilepsy. Am. J. Hum. Genet. 81,1057–1069. http://dx.doi.org/10.1086/522591.

Mefford, H.C., et al., 2010. Genome-wide copy number variation in epilepsy: novelsusceptibility loci in idiopathic generalized and focal epilepsies. PLoS Genet. 6,e1000962. http://dx.doi.org/10.1371/journal.pgen.1000962.

Mefford, H.C., et al., 2011. Rare copy number variants are an important cause of epilepticencephalopathies. Ann. Neurol. 70, 974–985. http://dx.doi.org/10.1002/ana.22645.

Moller, R.S., et al., 2013. Exon-disrupting deletions of NRXN1 in idiopathic generalizedepilepsy. Epilepsia 54, 256–264. http://dx.doi.org/10.1111/epi.12078.

Mukherjee, J., et al., 2011. The residence time of GABA(A)Rs at inhibitory synapses isdetermined by direct binding of the receptor alpha1 subunit to gephyrin. J. Neurosci.31, 14677–14687. http://dx.doi.org/10.1523/JNEUROSCI.2001-11.2011.

Nicoletti, P., et al., 2012. Genomewide pharmacogenetics of bisphosphonate-inducedosteonecrosis of the jaw: the role of RBMS3. Oncologist 17, 279–287.http://dx.doi.org/10.1634/theoncologist.2011-0202.

Noebels, J.L., 2003. The biology of epilepsy genes. Annu. Rev. Neurosci. 26, 599–625.http://dx.doi.org/10.1146/annurev.neuro.26.010302.081210.

UNCO

RRECT

Please cite this article as: Dejanovic, B., et al., Exonic microdeletions of theidiopathic generalized epilepsy, Neurobiol. Dis. (2014), http://dx.doi.org/1

PRO

OF

Nordli Jr., D.R., 2005. Idiopathic generalized epilepsies recognized by the InternationalLeague Against Epilepsy. Epilepsia 46 (Suppl. 9), 48–56. http://dx.doi.org/10.1111/j.1528-1167.2005.00313.x.

Reiss, J., et al., 2001. A mutation in the gene for the neurotransmitter receptor-clusteringprotein gephyrin causes a novel form of molybdenum cofactor deficiency. Am.J. Hum. Genet. 68, 208–213. http://dx.doi.org/10.1086/316941.

Reutens, D.C., Berkovic, S.F., 1992. Increased cortical excitability in generalised epi-lepsy demonstrated with transcranial magnetic stimulation. Lancet 339,362–363.

Schwarz, G., Schrader, N., Mendel, R.R., Hecht, H.J., Schindelin, H., 2001. Crystal struc-tures of human gephyrin and plant Cnx1 G domains: comparative analysis andfunctional implications. J. Mol. Biol. 312, 405–418. http://dx.doi.org/10.1006/jmbi.2001.4952.

Stallmeyer, B., et al., 1999. The neurotransmitter receptor-anchoring protein gephyrinreconstitutes molybdenum cofactor biosynthesis in bacteria, plants, and mammaliancells. Proc. Natl. Acad. Sci. U. S. A. 96, 1333–1338.

Tretter, V., et al., 2008. The clustering ofGABA(A) receptor subtypes at inhibitory synapses isfacilitated via thedirect bindingof receptor alpha 2 subunits to gephyrin. J. Neurosci. 28,1356–1365. http://dx.doi.org/10.1523/JNEUROSCI.5050-07.2008.

Tretter, V., et al., 2011. Molecular basis of the gamma-aminobutyric acid A receptoralpha3 subunit interaction with the clustering protein gephyrin. J. Biol. Chem. 286,37702–37711. http://dx.doi.org/10.1074/jbc.M111.291336.

Veldman, A., et al., 2010. Successful treatment of molybdenum cofactor deficiencytype A with cPMP. Pediatrics 125, e1249–e1254. http://dx.doi.org/10.1542/peds.2009-2192.

Vlachos, A., Reddy-Alla, S., Papadopoulos, T., Deller, T., Betz, H., 2012. Homeostatic regula-tion of gephyrin scaffolds and synaptic strength at mature hippocampal GABAergicpostsynapses. Cereb. Cortex. http://dx.doi.org/10.1093/cercor/bhs260.

Vollmar, C., et al., 2012. Altered microstructural connectivity in juvenile myoclonicepilepsy: the missing link. Neurology 78, 1555–1559. http://dx.doi.org/10.1212/WNL.0b013e3182563b44.

Werhahn, K.J., Lieber, J., Classen, J., Noachtar, S., 2000. Motor cortex excitability in patientswith focal epilepsy. Epilepsy Res. 41, 179–189.

Ziemann, U., Lonnecker, S., Steinhoff, B.J., Paulus, W., 1996. Effects of antiepileptic drugson motor cortex excitability in humans: a transcranial magnetic stimulation study.Ann. Neurol. 40, 367–378. http://dx.doi.org/10.1002/ana.410400306.

D

E

gephyrin gene impair GABAergic synaptic inhibition in patients with0.1016/j.nbd.2014.02.001