Genome wide analysis of novel copy number variations duplications/deletions of different epileptic patients in Saudi Arabia

Detection of 6p12.1 microduplications

Whole genome 2x 400K oligonucleotide based microarray analysis showed 99.9kb duplication at cytoband 6p12.1. Figure 2 presented 5 horizontal red lines of 5 patients with duplicated regions; each red line is marked with each patients ID along with the duplicated region of that specific patient. Out of these 5 patients, 2 were members of the same family whereas the rest of 3 were sporadic cases. No deletion or duplication was observed in father: 05, normal daughter: 1192 and normal son: 1191 (Figure 1). In affected mother: 1190, a microduplication of 51.6kb (55722611-55774293) and in affected daughter: 1193, a microduplication of 55.8kb (55718462-55774293) was observed. Rest of 3 sporadic cases also showed duplications of different sizes such as 241 showed a duplication of 95.6kb (55678636-55774293); 494 showed a duplication of 82.5kb (55691753-55774293) and 440 showed a duplication of 51.7kb (55726917-55778628). The gene that was observed in the duplicated region of all the patients is BMP5 (Table 1 and Figure 2); determined by UCSC Genome Browser (http://genome.ucsc.edu/) and Database of Genomic Variants (http://dgv.tcag.ca/dgv/app/home). BMP5 is a member of BMP family which belongs to transforming growth factor-β (TGFβ) superfamily. It is expressed in hippocampus, cerebellum and striatum [14, 15]. Studies also showed its expression in developing rat superior cervical ganglion corresponding to initial extension of primary dendrites and in early postnatal period during maximal dendritic growth. Immunofluorescence analysis of cultured superior cervical ganglion cells detected Bmp5 all over cell body of glial [16]. It was also observed that cell proliferation was reduced by BMP5 through cell death into the progenitor cells specifically in ventral forebrain. Insertion of BMP5 in developing chicken prosencephalon causes loss of ventral forebrain secondary to immense cell death localized to that region. Experimental embryos showed a loss of basal telencephalon due to implantation of beads soaked in recombinant BMP5 or BMP4 into neural tube of chicken forebrain that resulted in holoprosencephaly (a single cerebral hemisphere), cyclopia (a single midline eye) and loss of ventral midline structures [17]. Noradrenergic (NA) neurons are critical modulators of brain functions and have been implicated in common CNS disorders [18]. Locus coeruleus (LC), the major NA nucleus providing the main source of noradrenaline in the brain is formed in rostral hindbrain during embryogenesis. LC development requires either Bmp5 or Bmp7 and one is able to compensate the loss of the other. Also, the position of mid-hindbrain organizer determines the size of LC and propose that Bmp5/7 play an important role in mediating this organizer function [19]. Studies also uncovered novel roles of BMP signaling during development of heart, allantois, branchial arches, somites and forebrain development. Bmp5 do not appear to be involved in establishing pattern in these tissues but necessary for proliferation and maintenance of specific cell populations [20]. Observed duplications in all patients with reference to controls were confirmed by qPCR which showed a significant fold increase in all patients compared to controls Figure 3).

Detection of 7q32.3 microdeletions

Whole genome 2x 400K oligonucleotide based microarray analysis showed 63.9kb microdeletion at cytoband 7q32.3. Figure 4 showed 3 horizontal green lines of 3 patients with deleted regions; each green line is marked with each patients ID along with the deleted region of that specific patient. In patient 1090, a microdeletion of 63.9kb (130860344-130924315); in 1186, a microdeletion of 59.5kb (130864738-130924315) and in 495, a deletion of 56.6kb (130864738-130921416) was observed. The gene that was observed in the deleted regions of all the patients was PODXL (Table 1 and Figure 4); determined by UCSC Genome Browser (http://genome.ucsc.edu/) and Database of Genomic Variants (http://dgv.tcag.ca/dgv/app/home). Podocalyxin (PODXL) is type-1 membrane mucin protein belongs to CD34 family expressed abundantly in kidney epithelial cells (podocytes) [21]; its expression has also reported in the brain with maximum level in cortical plate, hippocampus, cerebellum and basal forebrain nuclei [22, 23]. In humans, it is also linked with malignant progression of brain astrocytic tumors [24]. Studies also showed that ventricles of Podxl ^−/− mice were enlarged in coronal and horizontal brain MR images which were also confirmed through histological analysis of paraffin embedded brain [25]. Reason of this could be either an enhanced formation or reduced disposal of cerebrospinal fluid (CSF) or it might be due to a combination of both mechanisms. Enhanced volume of CSF might be the result of a new steady state equilibrium of choroid plexuses ion membrane transporters; Na⁺-K⁺ATPase, K⁺ channels and Na⁺-2Cl ⁻ -K⁺ co-transporters expressed in apical membrane and Cl ⁻ -HCO3 exchangers, a variety of Na⁺ coupled HCO3 ⁻ transporters and K⁺-Cl ⁻ cotransporters expressed in the basolateral membrane [26]. Further analysis is required to check the involvement of these ion transporters toward ventricles enlargement in the absence of PODXL. Numerous human pathological disorders are linked with enlarged ventricles such as trauma, autism, bardet-biedl syndrome and alzheimer disease [27–29]. However, an earlier and possibly best human pathological condition studied was schizophrenia [30]. Finding of enlarged ventricles in the patients having first schizophrenia episode has raised the question of whether this neurodegenerative disease exhibited at the time of symptom onset or it is considered as a neurodevelopmental process which produces abnormal brain volumes at an early age [31]. Observed deletions in all the patients with reference to controls were confirmed by qPCR which showed a significant fold decrease in all the patients compared to controls (Figure 5).

Patients and families

22 patients were selected with primary diagnosis of different types of epilepsies including epilepsy (Epi), epilepsy plus mental retardation (Epi+MR) and juvenile myoclonic epilepsy (JME) shown in the Table 1. Details of all patients are as follows; Case 1190 and 1193: 1193 is a 4 years old female proband; her seizures were started at the age of 3 years. Her 33 years old mother 1190 had a consanguineous marriage and also suffering with seizures that were started at the age of 17 years. 05: father; 1192: sister and 1191: brother of the proband were normal and no CNV was found in any of these normal members of the family. One patient of the same family was died at the age of 5 years due to seizures (Figure 1). Case 440: She is 39 years old female patient of JME and her seizures were started the age of 16 years. Case 241: He is 14 years old male patient of JME and his seizures were started the age of 15 years. Case 1090: she is 15 years old female patient of epilepsy, her parents had a consanguineous marriage and both are normal. Blood samples of her parents, one normal sister and one normal brother was not available. Her seizures were started when she was 8 years old. Case 1186: He is 27 years old male patient of JME and his seizures were started the age of 16 years. Case 495: She is 13 years old female patient of epilepsy plus mental retarded. Her seizures were started at the age of 6 years. Case 494: He is 7 years old male patient of epilepsy plus mental retarded. His seizures were started the age of 3 years.

DNA preparation

Blood samples from all the patients were taken at King Abdulaziz University Hospital with their informed consent after the approval of ethical committee. DNA was extracted by using QIAamp DNA blood mini kit (http://www.Qiagen.com) and was quantified by NanoDrop ND-10000 Spectrophotometer.

Array-CGH analysis

Array-CGH analysis was done by Agilent sure print G3 Hmn CGH 2x 400K arrays (Agilent Technologies, Santa Clara, USA) following manufacturer protocol. Briefly, 500ng patient’s DNA and also reference DNA of the same sex (Promega, Madison, USA) were digested with RsaI (Promega, Madison, USA) and AluI (Promega, Madison, USA) for 2hrs at 37ºC. Furthermore, these samples were labeled by random primers through Agilent labeling kit (Agilent Technologies, Santa Clara, USA) following manufacturer guidelines. Patient’s DNA was labeled with Cy5-dUTP whereas reference DNA was labelled with Cy3-dUTP. Labeled products were purified by Microcon YM-30 filter units (Millipore, Billerica, USA). Patient and reference DNAs were mixed with Cot-1 DNA (Invitrogen, USA) blocking agent and hybridization buffer following by Agilent instructions. After denaturation at 95ºC and pre-annealing at 37ºC, hybridization was done at 65ºC for 40hrs. After the two washing steps, array was analyzed through Agilent Scanner (G2505C) and Feature Extraction Software (V.1.5.1.0). Data was analyzed by using Partek Genomics Suite Software, Cytogenomics Software (V.2.0.6.0; Agilent Technologies, USA) and publicly available DGV.

Validation of CNVs by real-time quantitative PCR

The microdeletions and microduplications were confirmed by real-time quantitative PCR (qPCR). Targeted sequences were chosen for qPCR by using Primer-3 Software (V.0.4.0). Two pair of primers were selected for each of genomic region of two target genes; bone morphogenetic protein-5 (BMP5) and podocalyxin-like protein (PODXL); an endogenous gene β₂-microglobulin (B2M) and control samples. PCR was done in 10μl reaction volume comprising of 05μl SYBR-Green qPCR master mix (KAPA Biosystems, USA), 10 pmol forward and reverse primers and 20ng genomic DNA. Reaction cycling conditions were as follows; denaturation at 95°C for 10minutes, annealing at 60°C for 1minute, holding stage at 60°C for 30seconds followed by 45 cycles at 60°C for 15seconds. Each run of 96 well plate included control samples, target genes, reference gene and non-template control for each gene. Plate was analyzed with StepOnePlus™ Real-Time PCR Systems (StepOnePlus™ Real-Time PCR, Applied Biosystems, Canada) raw data was acquired using Light Cycler 480 Software. Ct is the threshold cycle defined as the mean cycle at which the fluorescence curve reaches an arbitrary threshold; ΔCt is the difference between the Ct of the target gene and that of reference gene; ΔΔCt is ΔCt value of patients obtained by dividing the patient ΔCt value with ΔCt value of control sample. T-test with significant P-value <0.05 was used to find out statistical significance of predicted copy number alterations.