Microchromosomal deletions involving SCN1A and adjacent genes in severe myoclonic epilepsy in infancy

Summary

Purpose: Genetic abnormalities of the gene encoding α1 subunit of the sodium channel (SCN1A), which can be detected by direct sequencing, are present in more than 60% of patients with severe myoclonic epilepsy in infancy (SMEI) or its borderline phenotype (SMEB). Microchromosomal deletions have been recently reported as additional causes of SMEI. This study examines whether such microdeletions are associated with SMEI as well as with SMEB.

Methods: We recruited patients with SMEI (n = 35) and SMEB (n = 34), who were confirmed previously to have no mutations of SCN1A by direct sequencing. Microdeletions were sought by multiplex ligation-dependent probe amplification (MLPA), and then confirmed and characterized by fluorescence in situ hybridization (FISH) and array-based comparative genomic hybridization (aCGH), respectively.

Results: Heterozygous multiple exonic deletions were identified in 7/35 SMEI patients (20%) and 0/34 SMEB patients (0%), with a net frequency of 10.1% (7/69 patients). Deletions were confirmed by FISH and aCGH analysis. The concomitant deletions of adjacent genes were revealed by aCGH. None of the parents who agreed to undergo the analysis had such deletions suggesting that the deletions were de novo. The phenotypes of patients with the deletions were indistinguishable from those of SMEI resulting from point mutations.

Discussion: Our findings indicate that microchromosomal deletion, often involving not only SCN1A but also several adjacent genes, is associated with core SMEI. As microchromosomal deletion cannot be anticipated by the phenotypes or detected by conventional methods, genetic abnormalities in SMEI should be carefully sought by techniques that can detect microdeletions.

Severe myoclonic epilepsy in infancy (SMEI), or Dravet syndrome, is a rare and distinct malignant epilepsy syndrome (ILAE, 1989; Dravet et al., 1992; Engel 2001). Children with SMEI show normal development before the onset and often have febrile seizures that easily precipitate status epilepticus as the first episode of seizures in infancy. As these children grow, the clinical features evolve into a variety of afebrile or fever-induced seizure types, including myoclonic, partial, and atypical absence. Other features that develop during the course of SMEI include psychomotor slowing and regression, and ataxia. In late life, seizures remain refractory and intellectual outcome is usually poor. It is difficult to make the correct diagnosis of SMEI at an early stage of the disease with incomplete symptoms.

Some patients are labeled borderline SMEI (SMEB), whose clinical features are almost identical to those of core SMEI but do not exhibit distinct myoclonic seizures or atypical absences (Dravet et al., 1992; Oguni et al., 2001). The intellectual outcome in SMEB may be somehow better than that of SMEI, but it is again difficult to distinguish SMEI and SMEB at an early stage of the disorders.

Mutations of the gene encoding the neuronal sodium channel α1 subunit, SCN1A, were discovered in SMEI (Claes et al., 2001; Ohmori et al., 2002; Sugawara et al., 2002). Subsequent studies have so far identified more than 250 different SMEI-associated mutations in SCN1A; about 95% of which are considered de novo (Nabbout et al., 2003; Ceulemans et al., 2004; Fukuma et al., 2004; Mulley et al., 2005; Harkin et al., 2007; Marini et al., 2007). A few mutations are familial, and familial cases probably due to germline mosaic mutations in mildly or unaffected parents have been reported recently (Depienne et al., 2006; Gennaro et al., 2006; Marini et al., 2006; Morimoto et al., 2006). Mutations in SCN1A have been also found in SMEB. The prevalences of mutations in SMEI and SMEB are more than 60%. Identification of mutation in SCN1A may help establish early diagnosis of SMEI and SMEB.

At present, screening for mutations in SCN1A in patients with SMEI is mainly carried out by direct DNA sequencing or direct DNA sequencing combined with either denaturing high performance liquid chromatography (DHPLC) (Claes et al., 2001; Ohmori et al., 2002; Sugawara et al., 2002; Claes et al., 2003; Nabbout et al., 2003; Ceulemans et al., 2004; Fukuma et al., 2004). While these methods can detect point mutations, they are not suited to identify chromosomal microdeletions and duplications of one or more entire exons (Gille et al., 2002; Schouten et al., 2002). Therefore, microdeletions or duplications of SCN1A might be missed in SMEI. Recently, microdeletions of SCN1A were reported in SMEI by multiplex ligation-dependent probe amplification (MLPA) (Mulley et al., 2006), multiple amplicon quantification (MAQ) (Suls et al., 2006), and a technique similar to MAQ (Madia et al., 2006). However, microdeletions of SCN1A in SMEB have not been reported. Here, we report that microdeletions of SCN1A are found only in SMEI but not in SMEB and that such deletions involve not only SCN1A but also the adjacent genes.

Materials and Methods

Patients

We recruited 35 individuals with core SMEI and 34 with SMEB, in whom no mutation of SCN1A had been identified previously by direct DNA sequencing. They comprised 32 previously reported patients (Fukuma et al., 2004) and 37 newly diagnosed Japanese patients recruited by the method described previously by our group (Fukuma et al., 2004). Preparation of genomic DNA and the direct sequencing method used in this study was similar to that reported previously (Fukuma et al., 2004). The diagnostic criteria and classification of SMEI and SMEB used in this study were described previously (Fukuma et al., 2004). In short, the diagnosis of SMEI was based on fulfillment of all the accepted diagnostic criteria for SMEI (ILAE, 1989; Dravet et al., 1992), whereas the diagnosis of SMEB was based on the presence of almost identical clinical features of SMEI but excluding myoclonic and atypical absence seizures.

We also recruited 20 healthy volunteers as the control group. Each participant or parent/guardian signed an informed consent form approved by the Ethics Review Committee of Fukuoka University or equivalent committees of the participating institutions.

MLPA

MLPA was conducted using a commercially available kit for SCN1A (SALSA MLPA KIT P137 SCN1A, Lot 0107 or Lot 0805, MRC-Holland, Amsterdam, the Netherlands). Details of the probe sequences can be found on http://www.mrc-holland.com. MLPA tests were conducted according to the protocol supplied by the manufacturer. In brief, genomic DNA of 300 ng was denatured and hybridized with MLPA probes. Following probe hybridization for an average of 16 h, ligation was performed. The ligation products were then amplified by 30 cycles of PCR according to the protocol recommended by the manufacturer, using two primers; one unlabelled and one labeled with 6-FAM. Fragment analysis of the PCR product was carried out on ABI model 310 capillary sequencer (Applied Biosystems, Foster City, CA, U.S.A.) using GeneScan TM-500LIZ as size standards (Applied Biosystems) and deionized formamide (HiDi Formamide, Applied Biosystems). Data were analyzed using the Genescan software (Applied Biosystems). Specific peaks corresponding to each exon were identified according to their migration pattern relative to the size standards. For sample analysis, the peak areas were imported into a Microsoft Excel template provided by FALCO Biosystems (Tokyo, Japan). The peak fractions were calculated by dividing the peak area of a given exon by the sum of peak areas of all 14 references in various parts of the entire chromosomes (outside SCN1A) in the individual sample. Subsequently, this relative peak area of each exon was divided by the average relative peak area of the corresponding exon of the control samples. Under this condition, 1.0 should indicate two copies of the target sequence in the sample. We set thresholds of <0.65 for deletions and >1.35 for duplications. The peak pattern of a patient's sample superimposed over the peak pattern of the control was visually inspected.

Fluorescent in situ hybridization (FISH)

The probe for FISH for SCN1A was prepared by LA PCR with a primer pair (ACC ACC TCA CAA GGG TTA GCG TTA TCA G and TAG CCT TGG CTA ACA TTG TCA GAC ACA G) designed for the part of SCN1A between exons 22 to 24 (5,673 bp), and labeled with digoxigenin-11-dUTP by the nick translational method. As a reference probe for another part of chromosome 2, RP11-384O8 BAC DNA containing PAX3 was chosen and labeled with biotin-16-dUTP.

Array-based comparative genomic hybridization (acgh) analysis

Genome-wide DNA screening on Affymetrix GeneChip Mapping 250K Nsp EA array was performed using the GeneChip Instrument system according to the standard protocols supplied by the manufacturer (Affymetrix, Santa Clara, CA, U.S.A.). Briefly, genomic DNA was digested with restriction endonuclease, ligated to an adaptor, and subjected to PCR amplification with adaptor-specific primers. The PCR products were digested with DNase I, and labeled with a biotinylated nucleotide analogue using terminal deoxynucleotidyl transferase. The labeled DNA fragments were hybridized to the microarray, the hybridized DNA probes were captured by streptavidin-phycoerythrin conjugates, and the array was scanned. The signal intensity ratio was calculated as described previously with some modifications (Kojima et al., 2006). In this analysis, we used publicly available data from the Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/) accession number GSE5013, as reference data. A moving average of the signal intensity ratio across five adjacent markers was calculated and mapped onto the NCBI Homo sapiens Genome Build 35.1 according to the physical position of each marker.

Results

Seven (20%) of the 35 patients with core SMEI had multiple exonic deletions; deletions of all 26 exons were found in four patients (Fig. 1A). Deletions of 8–26th exons, 20–26th exons, and 17–26th exons were found in one patient each (Fig. 1B). All deletions were confirmed and found to be heterozygous by repeated MLPA tests. No exonic deletion was found in patients with SMEB. In total, multiple exon deletions were found in 7/69 patients (SMEI + SMEB) corresponding to 10.1%.

Figure 1. Multiplex ligation-dependent probe amplification (MLPA). (A) Comparison of the amplitude of the peak of each exon of the normal control (blue) and patient #5 (orange) shows that the copy numbers of exons 1–26 are half of those of the control. Thus, patient #5 has deletions of the exons on one allele. (B) Comparison of the amplitude of the peak of each exon of the normal control (blue) and patient #33 (orange) shows that the copy numbers of exons 17–26 are half of those of the control. Thus, patient #33 has deletions of the exons on one allele. Numbers in (A) and (B) indicate the corresponding number of exons. Arrows and arrowheads indicate peaks of exons that are deleted and controls, respectively.

As MLPA indicated that all the deletions affected 22–24th exons, we confirmed the deletions by FISH with a probe covering the same part of SCN1A. FISH showed two pairs of spots with the probe for PAX3 and only one pair of spots with the probe for 22–24th exons of SCN1A (Fig. 2). These findings confirmed the corresponding chromosomal deletions and heterozygosity identified by MLPA.

Figure 2. Results of fluorescent in situ hybridization (FISH). A representative FISH image taken for patient #5 shows that the probes for PAX3 hybridize both chromosome 2 whereas the probe for SCN1A hybridizes one of chromosome 2, suggesting of lack of SCN1A in one of allele.

aCGH analyses with Affymetrix GeneChip extended the above findings, showing that all the deletions affected not only SCN1A but also several adjacent genes, including TAIP-2, GALNT3, LOC440923, FLJ11457, SCN9A, SCN7A, and CMYA3 (Fig. 3). The length of the deleted sequences ranged from 156 to 1,549 kb.

Figure 3. Map of microdeletions. Micro chromosomal deletions do not only affect SCN1A but also often involve the adjacent genes. Solid boxes and lines indicate exons and introns, respectively. Blank boxes and dotted lines indicate exons and introns that are deleted in an allele, respectively.

We carried out MLPA analysis for the parents of patients #54 and #83. None of them had microdeletions of SCN1A. The other parents refused to undergo the analysis.

We collected further clinical information from patients with microdeletions, including other clinical features that were deduced to result from the adjacent genes. The phenotypes of the patients were uniform, irrespective of the size of deletion, and consistent with those of typical SMEI, although deletion involving multiple genes in a part of the chromosome was reminiscent of contiguous gene syndromes. No obvious dysmorphologic features were identified in any of the patients (Table 1).

Table 1.  Phenotypes of seven patients with microdeletions

Patient	#5	#9	#24	#33	#54	#83	#108
The phenotypes of patients were uniform and consistent with those of typical SMEI irrespective of the size of deletions. NI, not investigated.
Sex	F	M	M	F	M	M	F
Age at genetic analysis (years)	10	9	11	15	7	9	31
Range of deleted exons	1–26	1–26	1–26	17–26	8–26	1–26	20–26
Size of deletion (kbps)	1,139	750	1,549	164	262	1,277	156
Seizure type
Generalized tonic and/or clonic seizures	–	+	+	+	+	+	+
Myoclonic seizures	+	+	+	+	+	+	+
Atypical absences	–	+	–	+	–	–	–
Focal seizures	+	+	+	+	–	+	+
Status epilepticus	+	+	+	+	+	+	+
Seizure in fever	+	+	+	+	+	+	+
Onset of seizures (months)	5	2	4	5	4	6	6
Neurological abnormalities
Mental retardation (age at diagnosis, months)	+(24)	+(18)	+(24)	+(48)	+(12)	+(24)	+(12)
Delayed motor development (age at diagnosis, years)	+(4)	+(1.5)	–	+ (unknown)	–	+(2)	+(3)
Ataxia	+	–	–	+	–	+	+
Photosensitivity	–	–	–	–	–	–	–
Neuroimaging	Normal	Normal	Normal	Normal	NI	Normal	Normal
Symptoms resulting from mutation in adjacent genes
Erythermalgia (SCN9A)	–	–	–	–	–	–	–
Insensitivity to pain (SCN9A)	–	–	–	–	–	–	–
Hyperostosis (GALNT3)	–	–	–	–	–	–	–
Family history
Febrile seizures	–	–	+	–	–	+	+
Epilepsy	–	–	–	–	–	–	–
Other neurological disorders	–	–	–	–	–	–	–

Discussion

In this study, we identified genomic deletions of SCN1A and adjacent genes in patients with core SMEI. The results provide compelling evidence that such genetic abnormalities undetectable by conventional methods underlie SMEI and extend the findings of recent reports (Madia et al., 2006; Mulley et al., 2006; Suls et al., 2006).

Identification of this class of mutations is important because changes in copy numbers of such specific chromosomal sequences are frequently implicated as a cause of, or disposition to, human diseases and syndromes (Gille et al., 2002; Schouten et al., 2002). The deletion frequency of 20% in 35 patients with core SMEI, who were confirmed to be mutation-negative by DNA sequencing analysis, is close to the 15% in 13 SMEI patients reported by Mulley et al. (2006). We demonstrated that the deletions involved one copy of almost the entire SCN1A and several contiguous exons. No single exonic deletion, however, was detected. The part of the chromosome in the vicinity of SCN1A may be vulnerable to chromosomal rearrangements, which may result, for yet unknown reason, in copy number variation.

It is noteworthy that exonic deletions of SCN1A were found exclusively in individuals with core SMEI but not with SMEB. Our previous study (Fukuma et al., 2004; Kanai et al., 2004) suggested that genetic abnormalities of SCN1A such as truncation mutation, which may have a great impact, tend to be more common in core SMEI than in SMEB. Furthermore, only missense mutations of SCN1A have been identified so far as causes of autosomal dominant epilepsy with febrile seizure plus or generalized epilepsy with febrile seizure plus (GEFS+), which is considered a milder phenotype of a disease spectrum encompassing SMEI. In turn, microdeletions affecting hundreds to thousands kilo base pairs of the gene should have a great impact and thus may result in the most severe phenotype; core SMEI.

However, the numbers of deleted genes along with the deletion of SCN1A did not seem to influence the phenotypes of SMEI. It is known that abnormalities of one of the adjacent genes, SCN9A, which were found to be heterozygously deleted in four of the patients, may lead to human disorders. SCN9A encodes α subunits of Na_v1.7, a voltage sodium channel preferentially expressed in the dorsal root ganglia or spinal sensory neurons, and its missense mutations may cause either inherited erythermalgia (Cummins et al., 2004; Nassar et al., 2004; Yang et al., 2004) or paroxysmal extreme pain disorder (Fertleman et al., 2006, 2007), while its nonsense mutations can be a cause of channelopathy-associated insensitivity to pain (Cox et al., 2006). Another adjacent gene, SCN7A, encodes a sodium-level-sensitive sodium channel, which plays an important role in the central sensing of body-fluid sodium level and regulation of salt intake behavior (Hiyama et al., 2002). Furthermore, hyperostosis and hyperphosphatemia have been associated with GALNT3 mutation (Frishberg et al., 2005). Nonetheless, our findings suggest that SCN1A deletion involving the adjacent genes cannot be anticipated by phenotypes of patients with SMEI.

Based on the present findings, we recommend the inclusion of a comprehensive mutation screening for small to mid-size deletions or insertions and large genomic rearrangements in any genetic testing of SCN1A in patients with SMEI-related epileptic syndromes. In particular, it is crucial to detect mid-size deletions or duplications ranging from 400 kbp to 3 Mbp because conventional techniques for genetic tests may overlook such genetic derangements. Even a high resolution of G-band pattern of human chromosomes can show chromosomal deletions only when they are larger than several Mbp, whereas the sequence methods may allow the detection of only small deletions that are less than approximately 400 kbp. At present, many techniques, such as MLPA, FISH, multiplex amplifiable probe hybridization (MAPH), Southern blots, and real-time PCR, can be used to identify mid-size and large deletions or duplications of genes (Schouten et al., 2002). Several studies (Gille et al., 2002; Taylor et al., 2003; Slater et al., 2004; Rooms et al., 2005, 2006) compared MLPA with one or more of the above methods for detection of cryptic subtelomeric rearrangements and genes of MSH2, MLH1, BRCA1, and PMP22. Their results indicated that MLPA seems to be easier to perform and is a rapid, reliable, sensitive, cost-effective and robust gene dosage method that can be readily adopted by diagnostic services.

Our findings suggest that microchromosomal deletion does not only affect SCN1A but also often involves the adjacent genes. As the phenotypes of individuals with the deletions were consistent with those of typical SMEI, they were indistinguishable from those resulting from point mutations. As microchromosomal deletions are not rare, cannot be anticipated by the phenotypes, and are not detected by the conventional methods, genetic abnormalities in patients with SMEI should be carefully sought using MLPA or other techniques that can detect microdeletions in order to establish the correct genetic diagnosis and thereby to provide better genetic counseling.

Acknowledgments

We are indebted to our patients and their families for their helpful cooperation and encouragement in our research. We thank Ms. Akiyo Hamachi and Ms. Minako Yonetani for their technical assistance and Ms. Takako Umemoto for typing and formatting the manuscript. Our research was conducted as part of a comprehensive project organized by The Epilepsy Genetic Study Group, Japan (Chairperson, S. K.) and supported in part by Grants-in-Aid for Scientific Research (S) 16109006 and (A)18209035 and for Exploratory Research 1659272, and “High-Tech Research Center” Project for Private Universities: matching fund subsidy from the Ministry of Education, Culture, Sports, Science and Technology, 2006–2010: “The Research Center for the Molecular Pathomechanisms of Epilepsy, Fukuoka University” and a Research Grant (19A-6) for Nervous and Mental Disorders from the Ministry of Health, Labor and Welfare.

Conflict of interest: We confirm that we have read the Journal's position on issues involved in ethical publication and confirm that this report is consistent with those guidelines. None of the authors has any conflict of interest to disclose.