a Department of Medical Chemistry and Biochemitry, Sofia Medical University, 2 Zdrave str., Sofia, Bulgariab Genetic Medico-Diagnostic Laboratory Genica, 90 Tzar Asen str., Sofia, Bulgariac Clinic of Child Neurology, St. Naum University Hospital of Neurology and Psychiatry, 1 Lyuben Rusev str., Sofia, Bulgariad Department of Pediatrics, Division of Genetic Medicine, University of Washington, Health Sciences Building, K-253 Seattle,WA, USAe Department of Neuropediatrics, University Medical Center Schleswig-Holstein (Kiel Campus), Schwanenweg 20, Kiel, Germany
Received 16 July 2012; received in revised form 11 October 2012; accepted 30 October 2012Available online 23 January 2013
SummaryPurpose: The chromosome 15q13.3 region is a genomic rearrangement hotspot linked to idio-pathic generalized epilepsies (IGEs) and such rearrangements remain the strongest risk factorfor IGE known to date. Increasing evidence suggests that genetic variations can be highlypopulation-specific. Therefore, we aimed to assess the frequency of 15q13.3 microdeletions
15q13.3;Deletion
in IGE patients from Bulgaria.Methods: A cohort of 100 patients with various IGE syndromes was screened for large dele-tions/duplications by MLPA. All deletions and duplications were confirmed by array CGH analysis
as previously described.Results: In 100 prospectively recruited Bulgarian patients with IGE, we found one case with amicrodeletion, which amounted to 1% frequency for this copy number variant.
Conclusion: We confirm the frequency of 1% for the 15q13.3 microdeletion in a prospectivelyrecruited cohort of Bulgarian epilepsy patients, demonstrating that this variation representsa significant risk factor for IGE for various populations and that it is retrospectively detected
tion
s res
I
TiBedcg(asdMsi
tdsmaed2l(ahhacimfi(me
rneippar(
map
M
C
Dsspwot
tBtdIwsisottMg
r
tACoo
A
ItpacdWa(yment of Manova-Tomova (in younger children, for examplebelow the age of 3 years); specialized tests (e.g. test of Ben-ton, etc.). The psychological testing of the patients with IQ
win studies and family studies suggest a strong geneticmpact in a broad range of epilepsies (Lennox, 1947;erkovic et al., 1998; Helbig et al., 2008). Although sev-ral genes have been already identified in rare autosomalominant and severe sporadic forms of epilepsy, the exactause remains unknown in most of the cases with presumedenetic background. The idiopathic generalized epilepsiesIGEs) constitute about 30% of all epileptic disorders andre the most common forms of presumably genetic epilep-ies. IGEs have a complex inheritance pattern most probablyue to the interaction of various genetic and other factors.ultiple studies have studied common genetic variants and
everal candidate genes in IGE, but failed to identify signif-cant genetic risk factors.
Copy number variants (CNVs) are known to play an impor-ant role in the genetic etiology of many neuropsychiatricisorders, including intellectual disability (ID), autism, andchizophrenia. Recently, recurrent microdeletions on chro-osomes 15q13.3, 16p13.11 and 15q11.2 were identified
s important genetic factors predisposing to IGE (Dibbenst al., 2009; Mulley and Dibbens, 2009; Helbig et al., 2009;e Kovel et al., 2010; Mefford et al., 2010; Muhle et al.,011). These loci are genomic rearrangement hotspotsinked with different neurological disorders, including IGEMefford and Mulley, 2010). The presumed genetic mech-nism leading to these recurrent aberration is non-allelicomologous recombination (NAHR) between large, highlyomologous flanking segmental duplications (Stankiewicznd Lupski, 2002). The prevalence of these rearrangementsan be significant and 15q13.3 microdeletions are detectedn about 1% of patients with IGE. This makes 15q13.3icrodeletions one of the most common genetic causes
or IGE (Mefford et al., 2010). The CHRNA7 gene includedn this microdeletion represents the major candidate geneHelbig et al., 2009). The estimated frequency of 15q13.3icrodeletion in the general population is ∼0.02% (Dibbens
t al., 2009; Helbig et al., 2009).The IGE cases associated with this deletion could be
egarded as a part of a complex neurodevelopmental phe-otype including intellectual disability and seizures (Muhlet al., 2011). Reciprocal duplications in 15q13.3, eithernvolving the entire region between the common break-oints or the CHRNA7 gene alone, are detected in both IGEatients and healthy controls. Therefore, these duplicationsre not considered a major genetic defect in this unstableegion of the genome, involved in the development of IGEStein, 2009).
With the aim to assess the frequency of 15q13.3icrodeletions in IGE patients from Bulgaria we performed
prospective screening for 15q13.3 rearrangements in 100atients.
di
bias.erved.
ethods
linical data
uring a one-year period 100 IGE patients were con-ecutively recruited. The patients were with various IGEyndromes. All patients were followed up as in- and out-atients at the Clinic of Child Neurology in Sofia, Bulgaria,ith about 2/3 of them representing patients with new-nset epilepsy; the other being followed up regularly forheir epilepsy.
This study was approved by the local ethics commit-ee for scientific research at Medical University Sofia,ulgaria. Patient consent forms were signed from all par-icipants in the study. Classification of the epilepsy wasone according to the ILAE classification (EngelJr., 2001).nclusion criteria were: idiopathic generalized epilepsy,hich was diagnosed based on the medical history with
eizure description; on the clinical examination includingn some cases provocation and observation of the reportedeizures; on the EEG and especially on video-EEG (as inur cohort) registering the habitual seizures in most ofhe patients. Exclusion criteria were: focal epilepsy; symp-omatic epilepsy proven by neuroimaging (including brainRI), or another underlying cause for epilepsy (e.g. provenenetic syndrome).
We included patients with a broad range of common andare IGE syndromes.
The aim of the study was to screen exactly 100 consecu-ively recruited IGE patients, covering the inclusion criteria.ll patients were followed up as in- and out-patients at thelinic of Child Neurology in Sofia, Bulgaria, with about 2/3f them representing patients with new-onset epilepsy; thether being followed up regularly for their epilepsy.
ssessment of intellectual disability
ntellectual disability was established in 13 individuals ofhe whole cohort (13%), including one parent. Among theatients with MR, three of the children with CAE have had
relative with epilepsy; familial EMA was present in onehild with EMA and ID. In all patients with MR the IQ wasetermined by means of following tests: HAWIK-R (Hamburg-echsler Intelligenztest für Kinder Revision 1983) — for the
ge range 6—15 years, and Stanford—Binet Intelligence Testknown also as Binet—Terman test) — for the age range 3—14ears; the Bulgarian scales for neuropsychological develop-
eficit (n = 13) revealed mild degree of intellectual disabil-ty, i.e. IQ < 70 (70—55).
243
Figure 1 (A) Pedigree of the family. (B) EEG at age 11 years8 months. While awake and during IPS a dominating alpha-background is seen. Slow waves with frontal maximum and someleft-side predominance, lasting about 3 s are not accompaniedb
t3((
3Eltm
15q13.3 microdeletions in IGE patients from Bulgaria
Neurophysiological (EEG) verification of theepilepsy diagnosis
EEG recordings were performed in all patients. In 87 patients(0.91), a standard (routine) EEG recording was performed,i.e. EEG in awake state, for a time period of 10—20 mindepending on patient compliance, and performing thestandard activation procedures (eyes opening and closing;intermittent photic stimulation [IPS] at frequencies 2, 4, 6,8, 10, 12, 15, 18, 20, 22, 25 and 30 Hz; and hyperventilation[HV]). In all patients the EEG included a video-monitoring.The video-EEG enabled the registration of seizures in 54out of 96 children (56%) allowing a precise classification ofthe underlying epilepsy syndrome. Most commonly, we reg-istered absences (n = 27) and myoclonic seizures (n = 14). In13/96 patients (13.5%), all with JME, a sleep-deprived EEGrecording was performed for the registration of myoclonicseizures correlating to the generalized paroxysmal activityof polyspikes-slow wave complexes (PSW).
Neuroimaging
Due to financial restrictions in all individuals a brain com-puted tomography (CT) was done. In four children, a brainmagnetic resonance imaging (MRI) was performed due tofocal features of the reported seizures or focal featuresin the EEG, or because of drug-resistance of the epilepticseizures. Potentially causal CT/MRI findings could not beidentified in any of the patients. However, some incidental,unrelated findings (including in case 1) were identified.
The control group consists of middle aged individuals(>40—50 years), in whom epilepsies and other early tomiddle-onset neurodevelopmental disorders were excluded.They belong to an internal database and they were notrecruited for the study.
Screening and confirmation of 15q13.3microdeletions
Genomic DNA was isolated using commercial kits (Qiagen,
Valencia, CA) and screening for 15q13.3 microdeletionswas performed using SALSA MLPA Kit P297-B1 MicrodeletionSyndromes-2 (MRC-Holland, Amsterdam, the Netherlands)following manufacture instructions.
cn01
Table 1 Clinical data of 100 IGE patients.
Number of
Childhood absence epilepsy 36
Juvenile myoclonic epilepsy 20
Juvenile absence epilepsy 2
Early-onset absence epilepsy 5
Myoclonic-astatic epilepsy 7
Epilepsy with myoclonic absences 2
Epilepsy with generalized tonic—clonic seizures 11
Idiopathic generalized epilepsy — unclassified 12
Jeavons syndrome (eyelid myoclonia and absences) 5
Total 100
y any clinical manifestation.
This P297-B1 Microdeletion syndromes-2 probemix con-ains MLPA probes the following regions: 1q21.1; 1q21.1;q29; 7q36.1 (CNTNAP2 gene); 12p11.23; 15q13; 15q24.1PML gene); 16p11; 17q12; 18q21.2 (TCF4 gene) and 20p12.2PAK7 gene).
The obtained PCR products were analyzed on ABI Prism10 Genetic Analyser (Applied Biosystems, Foster city, CA).ach patient sample was analyzed simultaneously with ateast two normal controls. The relative peak ratios in ordero assess deletion/duplication were calculated using the for-ula: r = (peak areapatient/mean peak area both neighboring
ontrol peakspatient)/(peak areacontrol/mean peak area botheighboring control peakscontrol). The relative ratio less than
.6, corresponds to a deletion. The relative ratio more than.5, corresponds to a duplication.
patients Positive family history Intellectual disability
The suspected deletion and duplication were confirmedy array CGH analysis as previously described (Helbig et al.,009).
esults
linical data
linical data of 100 consequtively recruited patients arehown in Table 1. Average age of the whole cohort was 8.2ears. The age of epilepsy onset was between 10 months and6 years (mean 7.2 years). All patients had EEG examinationsnd neuroimaging. All patients were without dysmorphism.
enetic analysis
he results of the MLPA analysis revealed one microdeletionf the 15q13.3 region in 1/100 patients. The clinical data of
wawT
igure 2 Genetic analysis for the 15q13.3 microdeletion. (A) TheRPM1, KLF13, CHRNA7 — 15q13.4]. The deleted peaks are shown b5q13.3 rearrangements the patient and the family. The X-axis correumber differences in comparison to normal control sample (from −lack color, the regions with copy loss were colored in green and thef the references to color in figure legend, the reader is referred to
A. Kirov et al.
he patient with the 15q13.3 microdeletion are presentedelow. In addition, we identified one patient with CHRNA7icroduplication (∼500 kb). No reciprocal duplication of the
5q13.3 microdeletion spanning the BP4—BP5 region coulde identified. Additional copy number variations along theegions screened by MLPA were not detected.
ase 1: microdeletion 15q13.3
he patient was an 11 years and 8 month old boy diagnosedith CAE. At the time point of recruitment, the patient hadlready been followed-up for a period of 4 years becausef absence seizures and learning disability. The child wasorn after normal pregnancy, without signs of asphyxia. He
alked independently at age of 18 months and presented
speech delay (first words at the age of 3). Seizure onsetas at 8 years of age with multiple typical absences per day.he seizures were confirmed on EEG (Fig. 1B) and valproate
MLPA profile of the case with deletion [15q13.2 — MTMR15,y arrows. (B) High-resolution oligonucleotide array mapping ofsponds to chromosome region, the Y-axis corresponds to copy
4 to +4). All regions with normal copy number are presented in regions with copy gain were colored in red. (For interpretation
the web version of the article.)
Iot
1ovps
A
TM
ir
R
B
C
d
D
H
H
H
L
M
M
M
M
M
S
S
15q13.3 microdeletions in IGE patients from Bulgaria
(VPA) treatment was recommended. Brain CT demonstratedan arachnoid cyst in the left temporal polar region and wasinterpreted as an incidental finding. The patient was notcompliant with the antiepileptic medication and absenceseizures, as well as the typical generalized SW activity inthe EEG persisted for the next year. Behavioral disorderwas reported including aggressive and auto-aggressive out-bursts, even when seizures were controlled and the EEGwas unremarkable later in the course of the epilepsy. Fromthe age of 9 years onwards, seizures were fully controlledon VPA monotherapy with no further seizures until the lastvisit at inclusion in the study. At that time, EEG demon-strated slightly disorganized background of alpha rhythm,with frequent bilaterally synchronous and often rhythmicdelta-waves about 2.5—3.5/s, without clinical manifesta-tion, including on HV (Fig. 1B).
Neurological exam was normal. With the exception of asyndactyly of the 2nd—3rd digit of the left foot, no otherdysmorphic features were noticed. Psychological testingestablished an IQ of 64.
Parents were unaffected by epilepsy or intellectual dis-ability. Two older daughters from a previous marriage ofthe mother had been treated for epilepsy in childhood, butmedical records were not available. One of the half-sistersagreed to be tested for the 15q13.3 microdeletion.
The MLPA analysis showed a deletion [15q13.2 — MTMR15,TRPM1, KLF13, CHRNA7 — 15q13.4] in the patient and theunaffected mother, which was confirmed by array CGH. The15q13.3 microdeletion could not be confirmed in the half-sister affected by unclassified epilepsy. Both patient and themother carried a BP4—BP5 deletion (Fig. 2).
Discussion
In 100 consecutively recruited IGE patients from Bulgaria weconfirmed the previously reported frequency of about 1% for15q13.3 microdeletions. As emphasized in earlier studies,the microduplication, both the smaller CHRNA7 duplicationand the larger BP4—BP5 reciprocal duplication probably rep-resents a benign variant, while the 15q13.3 microdeletionremains the strongest risk factor for IGE known to date.We confirm that this variant has incomplete penetrance andcan be identified at almost identical frequency in variousIGE cohorts, including in this prospectively selected epilepsycohort of IGE patients from a single medical center, thusallowing minimal selection bias.
It is worth mentioning that our patient had epilepsy butalso intellectual disability and behavioral problems, evenwhen seizures were controlled. As emphasized in earlierstudies, the epilepsy accompanied by intellectual disabil-ity may represent a common phenotype of the 15q13.3microdeletion in epileptic children (Muhle et al., 2011;Mefford et al., 2011).
Although BP4—BP5 microdeletion was well-establishedrisk factor for epilepsy, the role of the reciprocal microdu-plication is not clear. In recent publications, the typical
microduplication of CHRNA7 gene was not detected incohorts of IGE (Cooper et al., 2011). In the same time, thisvariant was detected in control cohorts at a relatively highfrequency of 0.5% (Helbig et al., 2009; Cooper et al., 2011).
E
245
n other words, this variant is more frequent in people with-ut epilepsy than in IGE patients and it may be speculatedo have a protective effect (Helbig, 2011).
In summary, we could confirm the frequency of 1% for the5q13.3 microdeletion in a prospectively recruited cohortf epilepsy patients from Bulgaria, demonstrating that thisariation represents a significant risk factor for IGE in variousopulations and that the observed frequency is not due toelection bias.
cknowledgments
he study was supported by the grant no. 29/2010, Sofiaedical University, Bulgaria.
We confirm that we have read the Journal’s position onssues involved in ethical publication and affirm that thiseport is consistent with those guidelines.
eferences
erkovic, S.F., Howell, R.A., et al., 1998. Epilepsies in twins: genet-ics of the major epilepsy syndromes. Ann. Neurol. 43, 435—445.
ooper, G.M., Coe, B.P., et al., 2011. A copy number variation mor-bidity map of developmental delay. Nat. Genet. 43, 838—846.
e Kovel, C.G., Trucks, H., et al., 2010. Recurrent microdeletionsat 15q11.2 and 16p13.11 predispose to idiopathic generalizedepilepsies. Brain 133, 23—32.
ibbens, L.M., Mullen, S., et al., 2009. Familial and sporadic15q13.3 microdeletions in idiopathic generalized epilepsy:precedent for disorders with complex inheritance. Hum. Mol.Genet. 18, 3626—3631.
elbig, I., 2011. Does the 15q13.3 microduplication protectagainst epilepsy?, http://www.euroepinomics.wordpress.com(retrieved 02.04.12).
elbig, I., Scheffer, I.E., et al., 2008. Navigating the channels andbeyond: unravelling the genetics of the epilepsies. Lancet Neu-rol. 7, 231—245.
elbig, I., Mefford, H.C., et al., 2009. 15q13.3 microdeletionsincrease risk of idiopathic generalized epilepsy. Nat. Genet. 41,160—162.
ennox, W.G., 1947. The genetics of epilepsy. Am. J. Psychiatry 103,457—462.
efford, H.C., Mulley, J.C., 2010. Genetically complex epilepsies,copy number variants and syndrome constellations. GenomeMed. 2, 71.
efford, H.C., Muhle, H., et al., 2010. Genome-wide copy num-ber variation in epilepsy: novel susceptibility loci in idiopathicgeneralized and focal epilepsies. PLoS Genet. 6, e1000962.
efford, H.C., Yendle, S.C., et al., 2011. Rare copy number vari-ants are an important cause of epileptic encephalopathies. Ann.Neurol. 70, 974—985.
uhle, H., Mefford, H.C., et al., 2011. Absence seizures with intel-lectual disability as a phenotype of the 15q13.3 microdeletionsyndrome. Epilepsia 52, e194—e198.
ulley, J.C., Dibbens, L.M., 2009. Chipping away at the commonepilepsies with complex genetics: the 15q13.3 microdeletionshows the way. Genome Med. 1, 33 (review).
tein, R.A., 2009. An important risk factor in idiopathic generalized
epilepsies. Clin. Genet. 76, 21—23.
ngel Jr., 2001. A proposed diagnostic scheme for people withepileptic seizures and with epilepsy: report of the ILAE TaskForce on Classification and Terminology. Epilepsia 42, 796—803.
