Genome-Wide Screening of Copy Number Variation In Childhood Neuroblastoma

Structural genetic variation, including copy-number variation (CNV), constitutes a substantial fraction of total genetic variability and the importance of structural variants in modulating human disease is increasingly being recognized. Recent studies showed that chromosomal aberrations are detectable in childhood cancers and can be associated with susceptibility to childhood cancers. However its relationship with neuroblosatoma in particular is not fully understood. To gain insight into the incidence of the chromosomal aberrations in neuroblastoma in children, we examined Korean childhood neuroblastoma genomes using high-resolution single-nucleotide polymorphism (SNP) array-based analysis.

13 cases analyzed had been diagnosed with neuroblastoma (5 male, 8 female). 620,901 SNP markers were considered on these samples using Human610Quad v1.0 DNA analysis BeadChip (Illumina). Fragmented DNAs were hybridized on bead chips. Data analysis was carried out with GenomeStudio v2009.1, Genotyping 1.1.9, cnvPartition_v2.4.4 softwares. Overall call rate were more than 99.8%. Genome-wide CNV, genotyping of markers including 7,577 non-synonymous SNPs, loss of heterozygosity (LOH) analyses were performed using the GenomeStudio v2010.1. Linkage disequilibrium was analyzed by Haploview 4.2. The gene set enrichment analysis was performed using GO software, Panther.

The average call rates were 99.8 %. In total 343 CNVs were identified across the whole genome. Average number of CNVs per genome in this study (17.15) is higher than that of CNVs called in the recent studies using lower-resolution SNP- or CNV arrays. The median size of CNVs was 30,056 (range 569 ~ 1,260,297 bp). The largest portion of CNVs (235 calls) were found to be 10 kb~500 kb in length. Gain/loss of CNV was 2.05/4.90 having 2.4 fold higher frequencies in loss calls. We defined CNV regions (CNVRs) by merging overlapping CNVs (30% of overlap threshold) detected in two or more genomes. In total 155 CNVRs identified. The median size of CNVR was 27,482 (range 806 ~ 1,270,815 bp). Like CNVs, CNVRs-losses were more frequent than CNVR-gains. Defined CNVRs encompassing 13.4Mb accounted for ~0.5% of the human genome. Total of 1029 NM numbered transcripts were located near or within the 155 CNVRs. Through gene ontology (GO) analysis, putative target genes within the commonly gained or deleted region were categorized. Gene functions significantly enriched in the identified CNVRs include receptors for signal transduction pathways, transcription factors with nucleic acid binding proteins, transporters and regulatory molecule related functions involved in developmental processes. Genotype distributions for 7,577 non-synonymous SNPs in neuroblastoma were also examined and compared to two lab-specific as well as 90 Korean HapMap samples as control reference.

High-resolution single-nucleotide polymorphism (SNP) array-based analysis allowed us a high incidence of gains and losses in childhood neuroblastoma. Many of those detectable legions were found to be previously unidentified cryptic chromosomal aberrations. Those CNVRs could be potentially Korean-specific novel CNVRs indicating that previous CNV coverage of the human genome is incomplete and there is human genome diversity among different ethnic populations. Although results reveals high degrees of heterogeneity in the genomic alterations detectable in neuroblastoma, genes of the signal transduction pathway and transcriptional regulatory members were the most frequently altered targets whose deregulation may play a role in the pathogenesis of neuroblastoma in children. CNVs/CNVRs identified in the study will be solid resources for investigating chromosomal aberrations in childhood cancer and its potential association with childhood neuroblastoma. Further studies on larger sample size, as well as functional analyses will define their role in the pathogenesis of neuroblastoma in children.

No relevant conflicts of interest to declare.