Cynthia Winter

Copy number variation at 1q21.1 associated with neuroblastoma

Common copy number variations (CNVs) represent a significant source of genetic diversity, yet their influence on phenotypic variability, including disease susceptibility, remains poorly understood. To address this problem in human cancer, we performed a genome-wide association study of CNVs in the childhood cancer neuroblastoma, a disease in which single nucleotide polymorphism variations are known to influence susceptibility. We first genotyped 846 Caucasian neuroblastoma patients and 803 healthy Caucasian controls at ∼550,000 single nucleotide polymorphisms, and performed a CNV-based test for association. We then replicated significant observations in two independent sample sets comprised of a total of 595 cases and 3,357 controls. Here we describe the identification of a common CNV at chromosome 1q21.1 associated with neuroblastoma in the discovery set, which was confirmed in both replication sets. This CNV was validated by quantitative polymerase chain reaction, fluorescent in situ hybridization and analysis of matched tumour specimens, and was shown to be heritable in an independent set of 713 cancer-free parent–offspring trios. We identified a previously unknown transcript within the CNV that showed high sequence similarity to several neuroblastoma breakpoint family (NBPF) genes and represents a new member of this gene family (NBPF23). This transcript was preferentially expressed in fetal brain and fetal sympathetic nervous tissues, and the expression level was strictly correlated with CNV state in neuroblastoma cells. These data demonstrate that inherited copy number variation at 1q21.1 is associated with neuroblastoma and implicate a previously unknown neuroblastoma breakpoint family gene in early tumorigenesis of this childhood cancer.

Chromosome 6p22 Locus Associated with Clinically Aggressive Neuroblastoma

BACKGROUND Neuroblastoma is a malignant condition of the developing sympathetic nervous system that most commonly affects young children and is often lethal. Its cause is not known. METHODS We performed a genomewide association study by first genotyping blood DNA samples from 1032 patients with neuroblastoma and 2043 control subjects of European descent using the Illumina HumanHap550 BeadChip. Samples from three independent groups of patients with neuroblastoma (a total of 720 patients) and 2128 control subjects were then genotyped to replicate significant associations. RESULTS We observed a significant association between neuroblastoma and the common minor alleles of three consecutive single-nucleotide polymorphisms (SNPs) at chromosome band 6p22 and containing the predicted genes FLJ22536 and FLJ44180 (P=1.71x10(-9) to 7.01x10(-10); allelic odds ratio, 1.39 to 1.40). Homozygosity for the at-risk G allele of the most significantly associated SNP, rs6939340, resulted in an increased likelihood of the development of neuroblastoma (odds ratio, 1.97; 95% confidence interval, 1.58 to 2.45). Subsequent genotyping of the three 6p22 SNPs in three independent case series confirmed our observation of an association (P=9.33x10(-15) at rs6939340 for joint analysis). Patients with neuroblastoma who were homozygous for the risk alleles at 6p22 were more likely to have metastatic (stage 4) disease (P=0.02), amplification of the MYCN oncogene in the tumor cells (P=0.006), and disease relapse (P=0.01). CONCLUSIONS A common genetic variation at chromosome band 6p22 is associated with susceptibility to neuroblastoma.

Common variations in BARD1 influence susceptibility to high-risk neuroblastoma

We conducted a SNP-based genome-wide association study (GWAS) focused on the high-risk subset of neuroblastoma. As our previous unbiased GWAS showed strong association of common 6p22 SNP alleles with aggressive neuroblastoma, we restricted our analysis here to 397 high-risk cases compared to 2,043 controls. We detected new significant association of six SNPs at 2q35 within the BARD1 locus (Pallelic = 2.35 × 10−9–2.25 × 10−8). We confirmed each SNP association in a second series of 189 high-risk cases and 1,178 controls (Pallelic = 7.90 × 10−7–2.77 × 10−4). We also tested the two most significant SNPs (rs6435862, rs3768716) in two additional independent high-risk neuroblastoma case series, yielding combined allelic odds ratios of 1.68 each (P = 8.65 × 10−18 and 2.74 × 10−16, respectively). We also found significant association with known BARD1 nonsynonymous SNPs. These data show that common variation in BARD1 contributes to the etiology of the aggressive and most clinically relevant subset of human neuroblastoma.

Integrative genomics identifies LMO1 as a neuroblastoma oncogene

Neuroblastoma is a childhood cancer of the sympathetic nervous system that accounts for approximately 10% of all paediatric oncology deaths. To identify genetic risk factors for neuroblastoma, we performed a genome-wide association study (GWAS) on 2,251 patients and 6,097 control subjects of European ancestry from four case series. Here we report a significant association within LIM domain only 1 (LMO1) at 11p15.4 (rs110419, combined P = 5.2 × 10−16, odds ratio of risk allele = 1.34 (95% confidence interval 1.25–1.44)). The signal was enriched in the subset of patients with the most aggressive form of the disease. LMO1 encodes a cysteine-rich transcriptional regulator, and its paralogues (LMO2, LMO3 and LMO4) have each been previously implicated in cancer. In parallel, we analysed genome-wide DNA copy number alterations in 701 primary tumours. We found that the LMO1 locus was aberrant in 12.4% through a duplication event, and that this event was associated with more advanced disease (P < 0.0001) and survival (P = 0.041). The germline single nucleotide polymorphism (SNP) risk alleles and somatic copy number gains were associated with increased LMO1 expression in neuroblastoma cell lines and primary tumours, consistent with a gain-of-function role in tumorigenesis. Short hairpin RNA (shRNA)-mediated depletion of LMO1 inhibited growth of neuroblastoma cells with high LMO1 expression, whereas forced expression of LMO1 in neuroblastoma cells with low LMO1 expression enhanced proliferation. These data show that common polymorphisms at the LMO1 locus are strongly associated with susceptibility to developing neuroblastoma, but also may influence the likelihood of further somatic alterations at this locus, leading to malignant progression.

Germline PHOX2B Mutation in Hereditary Neuroblastoma

To the Editor: We read with interest the study by Trochet and colleagues (2004), published in the April 2004 issue of The American Journal of Human Genetics, that described germline mutations of the paired-like homeobox 2B gene (PHOX2B [MIM 603851]) in neuroblastoma (MIM 256700). We have also considered PHOX2B as a candidate gene for predisposition to neuroblastoma, and we now report on a germline PHOX2B mutation in a pedigree with neuroblastoma. However, we also show that there is no evidence for mutation of this gene in eight other pedigrees with neuroblastoma screened to date. We think these data establish PHOX2B as the first bona fide gene that can predispose to neuroblastoma when mutated in the germline, and the findings further emphasize the complex genetics of this important pediatric malignancy. We previously demonstrated linkage of hereditary neuroblastoma to 16p12-13 by use of a genomewide screening strategy (Maris et al. 2002). Positional cloning of a putative 16p12-13 hereditary neuroblastoma-predisposition gene (HNB1) is ongoing, but the critical genomic region for this gene remains large. We had previously considered and excluded other genes known to be mutated in Hirschsprung disease (MIM 142623) and/or in congenital central hypoventilation syndrome (CCHS [MIM 209880]) as candidates for HNB1, because these disorders can occur coincident with both sporadic and hereditary neuroblastoma (Maris et al. 2002). Because of the recent reports that the vast majority of patients with CCHS harbor PHOX2B mutations, including two patients also affected with neuroblastoma (Amiel et al. 2003; Weese-Mayer et al. 2003), we initiated a screen for germline mutations in this gene in our series of pedigrees with neuroblastoma. Oligonucleotide primer pairs flanking the coding regions of exons 1, 2, and 3 of PHOX2B were designed by use of the program Primer 3.0; these primer pairs were used for PCR amplification and bidirectional sequencing of purified PCR products (primer sequences available on request). We screened germline DNA from the proband and an unaffected family member for each of the seven families that showed cosegregation of a 16p haplotype with disease, as well as for two pedigrees that consisted of cousins with neuroblastoma with no cosegregation of 16p marker haplotypes (see Maris et al. [2002] for details of pedigrees). We also sequenced 109 control DNA samples from the Coriell SNP500 Cancer Panel (Coriell Cell Repositories). All sequence aberrations were confirmed by repeat sequencing after cloning of purified PCR products (TOPO TA Cloning Kit [Invitrogen]), and DNA samples from the remaining available members of the pedigree were also screened for the variant. The Children’s Hospital of Philadelphia institutional review board approved this work. A heterozygous single-base deletion (676delG) was discovered in a complex pedigree with neuroblastoma (fig. 1) (see dbSNP Home Page). This family has seven members in three generations affected with neuroblastoma, and two of these individuals were also shown to have Hirschsprung disease. The proband was affected with neuroblastoma, Hirschsprung disease, and neurofibromatosis type 1 (MIM 162200). The putative nonsense mutation 676delG segregated with neuroblastoma through all three generations, and the frameshift was predicted to produce a slightly truncated protein that would no longer code for the second polyalanine tract. This family had previously been shown to cosegregate a 16p12-13 haplotype with neuroblastoma, and the proband was also shown to have an inactivating mutation in NF1 (3775delT) that was not present in either of her parents (Maris et al. 2002). Tumor material was available only for patient 1-001, and the tumor exon 3 sequence remained heterozygous for the 676delG mutation. In addition, loss-of-heterozygosity studies using microsatellite markers (D4S2912, D4S1587, D4S405, D4S2971, and D4S428) that are closely linked to the PHOX2B locus showed no evidence for allelic deletion. The only other sequence variant discovered in the remaining eight pedigrees was a putative SNP (C552T) in pedigree 12 that is not predicted to affect the resultant protein sequence (S184S) (see dbSNP Home Page). This sequence variant was present in the proband but was not detected in the patient’s affected father (no maternal DNA sample was available). It is important to note that neither sequence variant was identified in the bidirectional sequencing of 218 alleles from the control sample set. This strongly suggests that the 676delG sequence variant that segregates with the disease phenotype is a true mutation. The C552T sequence variant, which does not segregate with the disease, is more likely a very rare polymorphism, but we cannot formally exclude the possibility that there might be a functional effect of this presumably neutral polymorphism. Figure 1 Germline PHOX2B mutation in a pedigree segregating neuroblastoma and Hirschsprung disease. A, Family 1 pedigree structure. DNA samples from this family with neuroblastoma were available only for patients with a PHOX2B result shown (wt = wild type; 676delG ... Accumulated data strongly implicate PHOX2B as an essential regulator of normal autonomic nervous system development (Pattyn et al. 1999; Brunet and Pattyn 2002). The discovery of polyalanine-expansion mutations in the majority of patients with CCHS clearly defines a role for this gene in human disease (Amiel et al. 2003; Weese-Mayer et al. 2003), and there appears to be a correlation between the severity of the respiratory symptoms and the length of polyalanine expansion (Weese-Mayer et al. 2003; Matera et al. 2004). Neuroblastoma represents perhaps the most aberrant phenotype that results from abnormal adrenergic tissue development. The rare but well-described synchronous appearance of neuroblastoma with other disorders of the autonomic nervous system has suggested a common genetic etiology often referred to as a “neurocristopathy” (Gaisie et al. 1979; Nemecek et al. 2003). Although other genes implicated in Hirschsprung disease and/or CCHS have not been excluded as hereditary neuroblastoma-predisposition genes (Maris et al. 2002; Perri et al. 2002), our data further establish PHOX2B as an important gene involved in the initiation of neuroblastoma tumorigenesis. However, the fact that the majority of pedigrees studied here do not show PHOX2B mutations clearly implicates locus heterogeneity for hereditary predisposition to neuroblastoma. Assuming that our inferences of linkage to 16p are correct, and in light of the observation of two germline mutations in the proband of the family presented here, we suggest that an oligogenic mechanism for neuroblastoma initiation should be considered, as has been shown for other diseases of neural crest–derived tissues (Gabriel et al. 2002). It is not yet clear if the PHOX2B mutations discovered in patients with hereditary or sporadic neuroblastoma result in gain or loss of protein function. The hypothesis that PHOX2B functions as a tumor suppressor is supported by the potential predicted consequence of the five mutations described, to date, in patients with neuroblastoma. Weese-Mayer and colleagues discovered a nonsense mutation that predicts a significantly truncated protein that would miss most of exon 3, including all of the 20-alanine repeat motif (Weese-Mayer et al. 2003). The frameshift mutation described here is similar to that reported by Amiel and colleagues (2003) in a patient who also had CCHS, Hirschsprung disease, and neuroblastoma, and, in both cases, the changes in reading frame are predicted to abolish the polyalanine tract. Trochet and colleagues (2004) discovered two missense mutations, both of which map to a conserved portion of the homeodomain and thus may interfere with DNA binding. On the other hand, 4p12 is not a known site of frequent allelic deletion in neuroblastoma (Maris and Matthay 1999), and, to date, biallelic inactivation of PHOX2B has not been demonstrated, although far too few cases have been examined to assert this with confidence. Taken together, these data suggest that PHOX2B mutations are involved in the initiation of neuroblastoma tumorigenesis, especially in patients with associated disorders of the autonomic nervous system. Our data also indicate that germline mutational events in this gene are not involved in the majority of hereditary neuroblastoma cases and that alternative genetic events may predispose to tumorigenesis. Examination of additional patients will facilitate the definition of PHOX2B mutation frequency in the genetic and (apparently) sporadic forms of neuroblastoma and will help to clarify the role of PHOX2B mutations in tumor initiation and progression.

Integrative Genomics Identifies Distinct Molecular Classes of Neuroblastoma and Shows That Multiple Genes Are Targeted by Regional Alterations in DNA Copy Number

Neuroblastoma is remarkable for its clinical heterogeneity and is characterized by genomic alterations that are strongly correlated with tumor behavior. The specific genes that influence neuroblastoma biology and are targeted by genomic alterations remain largely unknown. We quantified mRNA expression in a highly annotated series of 101 prospectively collected diagnostic neuroblastoma primary tumors using an oligonucleotide-based microarray. Genomic copy number status at the prognostically relevant loci 1p36, 2p24 (MYCN), 11q23, and 17q23 was determined by PCR and was aberrant in 26, 20, 40, and 38 cases, respectively. In addition, 72 diagnostic neuroblastoma primary tumors assayed in a different laboratory were used as an independent validation set. Unsupervised hierarchical clustering showed that gene expression was highly correlated with genomic alterations and clinical markers of tumor behavior. The vast majority of samples with MYCN amplification and 1p36 loss of heterozygosity (LOH) clustered together on a terminal node of the sample dendrogram, whereas the majority of samples with 11q deletion clustered separately and both of these were largely distinct from the copy number neutral group of tumors. Genes involved in neurodevelopment were broadly overrepresented in the more benign tumors, whereas genes involved in RNA processing and cellular proliferation were highly represented in the most malignant cases. By combining transcriptomic and genomic data, we showed that LOH at 1p and 11q was associated with significantly decreased expression of 122 (61%) and 88 (27%) of the genes mapping to 1p35-36 and all of 11q, respectively, suggesting that multiple genes may be targeted by LOH events. A total of 71 of the 1p35-36 genes were also differentially expressed in the independent validation data set, providing a prioritized list of candidate neuroblastoma suppressor genes. Taken together, these data are consistent with the hypotheses that the neuroblastoma transcriptome is a sensitive marker of underlying tumor biology and that chromosomal deletion events in this cancer likely target multiple genes through alteration in mRNA dosage. Lead positional candidates for neuroblastoma suppressor genes can be inferred from these data, but the potential multiplicity of transcripts involved has significant implications for ongoing gene discovery strategies.

RNAi screen of the protein kinome identifies checkpoint kinase 1 (CHK1) as a therapeutic target in neuroblastoma

Neuroblastoma is a childhood cancer that is often fatal despite intense multimodality therapy. In an effort to identify therapeutic targets for this disease, we performed a comprehensive loss-of-function screen of the protein kinome. Thirty kinases showed significant cellular cytotoxicity when depleted, with loss of the cell cycle checkpoint kinase 1 (CHK1/CHEK1) being the most potent. CHK1 mRNA expression was higher in MYC–Neuroblastoma-related (MYCN)–amplified (P < 0.0001) and high-risk (P = 0.03) tumors. Western blotting revealed that CHK1 was constitutively phosphorylated at the ataxia telangiectasia response kinase target site Ser345 and the autophosphorylation site Ser296 in neuroblastoma cell lines. This pattern was also seen in six of eight high-risk primary tumors but not in control nonneuroblastoma cell lines or in seven of eight low-risk primary tumors. Neuroblastoma cells were sensitive to the two CHK1 inhibitors SB21807 and TCS2312, with median IC50 values of 564 nM and 548 nM, respectively. In contrast, the control lines had high micromolar IC50 values, indicating a strong correlation between CHK1 phosphorylation and CHK1 inhibitor sensitivity (P = 0.0004). Furthermore, cell cycle analysis revealed that CHK1 inhibition in neuroblastoma cells caused apoptosis during S-phase, consistent with its role in replication fork progression. CHK1 inhibitor sensitivity correlated with total MYC(N) protein levels, and inducing MYCN in retinal pigmented epithelial cells resulted in CHK1 phosphorylation, which caused growth inhibition when inhibited. These data show the power of a functional RNAi screen to identify tractable therapeutical targets in neuroblastoma and support CHK1 inhibition strategies in this disease.

Common variation at 6q16 within HACE1 and LIN28B influences susceptibility to neuroblastoma

Neuroblastoma is a cancer of the sympathetic nervous system that accounts for approximately 10% of all pediatric oncology deaths. Here, we report a genome-wide association study of 2,817 neuroblastoma cases and 7,473 controls. We identified two new associations at 6q16, the first within HACE1 (rs4336470; combined P = 2.7 × 10−11; odds ratio 1.26, 95% confidence interval (CI) 1.18–1.35) and the second within LIN28B (rs17065417; combined P = 1.2 × 10−8; odds ratio 1.38, 95% CI 1.23–1.54). Expression of LIN28B and let-7 miRNA correlated with rs17065417 genotype in neuroblastoma cell lines, and we observed significant growth inhibition upon depletion of LIN28B, specifically in neuroblastoma cells that were homozygous for the risk allele. Low HACE1 and high LIN28B expression in diagnostic primary neuroblastomas were associated with worse overall survival (P = 0.008 and 0.014, respectively). Taken together, these data show that common variants in HACE1 and LIN28B influence neuroblastoma susceptibility and indicate that both genes likely have a role in disease progression.

Definition and characterization of a region of 1p36.3 consistently deleted in neuroblastoma

Substantial genomic and functional evidence from primary tumors and cell lines indicates that a consistent region of distal chromosome 1p is deleted in a sizable proportion of human neuroblastomas, suggesting that this region contains one or more tumor suppressor genes. To determine systematically and precisely the location and extent of 1p deletion in neuroblastomas, we performed allelic loss studies of 737 primary neuroblastomas and genotype analysis of 46 neuroblastoma cell lines. Together, the results defined a single region within 1p36.3 that was consistently deleted in 25% of tumors and 87% of cell lines. Two neuroblastoma patients had constitutional deletions of distal 1p36 that overlapped the tumor-defined region. The tumor- and constitutionally-derived deletions together defined a smallest region of consistent deletion (SRD) between D1S2795 and D1S253. The 1p36.3 SRD was deleted in all but one of the 184 tumors with 1p deletion. Physical mapping and DNA sequencing determined that the SRD minimally spans an estimated 729 kb. Genomic content and sequence analysis of the SRD identified 15 characterized, nine uncharacterized, and six predicted genes in the region. The RNA expression profiles of 21 of the genes were investigated in a variety of normal tissues. The SHREW1 and KCNAB2 genes both had tissue-restricted expression patterns, including expression in the nervous system. In addition, a novel gene (CHD5) with strong homology to proteins involved in chromatin remodeling was expressed mainly in neural tissues. Together, these results suggest that one or more genes involved in neuroblastoma tumorigenesis or tumor progression are likely contained within this region.

Prevalence and functional consequence of PHOX2B mutations in neuroblastoma

PHOX2B is a homeodomain-containing protein that is involved in the development of the peripheral nervous system and is the major disease gene for the rare congenital breathing disorder congenital central hypoventilation syndrome (CCHS). Germline PHOX2B alterations were also recently discovered in neuroblastoma cases with CCHS and/or Hirschsprung disease, but a comprehensive survey for mutational frequency and functional consequence has not been performed. We therefore studied a large panel of hereditary neuroblastomas to understand the frequency and functional effects of PHOX2B mutations. Three of 47 individuals with presumed genetic predisposition to neuroblastoma showed a germline PHOX2B mutation (6.4%). Mutations were also discovered in 2 of 30 human neuroblastoma-derived cell lines, but none of 86 primary tumors from patients with sporadically occurring neuroblastoma. The vast majority of primary tumors showed abundant PHOX2B mRNA expression relative to the remainder of the transcriptome. Consistent with its role as an important neurodevelopmental gene, forced overexpression of wild-type PHOX2B in neuroblastoma cell lines suppressed cell proliferation and synergized with all-trans retinoic acid to promote differentiation. Patient-derived mutant PHOX2B constructs retained the ability to suppress cellular proliferation, but were not able to promote differentiation or activate expression of a known PHOX2B target gene in vitro. These findings show that PHOX2B alterations are a rare cause of hereditary neuroblastoma, but disruption of this neurodevelopmental pathway can interfere with transcription-dependent terminal differentiation. These data also suggest that the genetics of neuroblastoma initiation are complex, and highlight genes involved in normal noradrenergic development as candidate predisposition genes.