Faculty Bibliography

One of the most important prognostic factors in neuroblastoma is amplification of the MYCN gene, which is strongly associated with advanced stages of disease and a poor prognosis. Although the MYCN amplicon sometimes spans more than 1 Mb, no other consistently expressed sequences from the MYCN amplicon have been reported. However, DDX1, a gene encoding a DEAD box protein, was recently mapped to chromosome 2p24 and is frequently co-amplified with MYCN. Therefore, we performed genomic mapping with YACs to determine the physical relationship between DDX1 and MYCN, and whether DDX1 was contained within the core region of amplification. Based on YAC restriction mapping and content analysis, DDX1 maps 340 kb 5' of MYCN, outside the core domain of consistent amplification. Interestingly, we also determined by sequence analysis and detailed restriction mapping that G21, previously isolated as a 'neuroblastoma-specific' cDNA clone from an MYCN amplicon, is a partial cDNA of DDX1. Our data confirm that DDX1 is amplified in some but not all MYCN-amplified tumors, and that it is rearranged in other cases. This suggests that the co-amplification of DDX1 is due to its proximity to MYCN.

Familial predisposition to neuroblastoma, a common embryonal cancer of childhood, segregates as an autosomal dominant trait with high penetrance. It is therefore likely that neuroblastoma susceptibility is due to germ line mutations in a tumor suppressor gene. Cytogenetic, functional, and molecular studies have implicated chromosome band 1p36 as the most likely region to contain a suppressor gene involved in sporadic neuroblastoma tumorigenesis. We now demonstrate that neuroblastoma predisposition does not map to any of eight polymorphic markers spanning 1p36 by linkage analysis in three families. In addition, there is no loss of heterozygosity at any of these markers in tumors from affected members of these kindreds. Furthermore, there is strong evidence against linkage to two Hirschsprung disease (a condition that can cosegregate with neuroblastoma) susceptibility genes, RET and EDNRB. We conclude that the neuroblastoma susceptibility gene is distinct from the 1p36 tumor suppressor and the currently identified Hirschsprung disease susceptibility genes.

The Krüppel-type zinc finger proteins are members of a conserved family of transcription factors that are important in developmental regulation. Altered expression of several of these proteins has been implicated in human diseases, including cancer. We report the cloning, mapping, and characterization of the zinc finger gene Human Krüppel-Related 3 (HKR3). Genomic clones of HKR3 were isolated from a P1 library and localized to human chromosome subband 1p36.3 by human-rodent somatic cell hybrid mapping and fluorescence in situ hybridization. The gene was physically mapped to within 40 kb of D1S214 by YAC content and long-range restriction mapping. HKR3 spans 9.5 kb of genomic DNA and is contained in 11 exons. Sequencing defined each of the exon/intron splice site junctions and identified a CpG island in the 5' region of the gene. HKR3 is ubiquitously expressed in human tissues as at least two major transcripts, the shorter of which excludes a conserved finger-associated box and a putative acidic activation domain contained in the full-length transcript. HKR3 is a novel zinc finger gene that maps to a region of the genome commonly rearranged or deleted in human cancers.

The tumor necrosis factor receptor 2 (TNFR2) gene localizes to 1p36. 2, a genomic region characteristically deleted in neuroblastomas and other malignancies. In addition, TNFR2 is the principal mediator of the effects of TNF on cellular immunity, and it may cooperate with TNFR1 in the killing of nonlymphoid cells. Therefore, we undertook an analysis of the genomic structure and precise physical mapping of this gene. The TNFR2 gene is contained on 10 exons that span 26 kb. Most of the functional domains of TNFR2 are encoded by separate exons, and each of the repeats of the extracellular cysteine-rich domain is interrupted by an intron. The genomic structure reveals a close relationship to TNFR1, another member of the TNFR superfamily. Based on electrophoretic analysis of yeast artificial chromosomes, TNFR2 maps within 400 kb of the genetic marker D1S434. In addition, we have identified a new polymorphic dinucleotide repeat within intron 4 of TNFR2. The genetic sequence information and exon-intron boundaries we have determined will facilitate mutational analysis of this gene to determine its potential role in neuroblastoma, as well as in other cancers with characteristic deletions or rearrangements of 1p36.

By screening a human fetal brain cDNA expression library using a monoclonal antiphosphotyrosine antibody and by 5' RACE procedures, we have isolated overlapping cDNAs encoding a receptor-type tyrosine kinase belonging to the EPH family, DRT (Developmentally Regulated EPH-related Tyrosine kinase gene). The DRT gene is expressed in three different size transcripts (i.e. 4, 5 and 11 kb). DRT transcripts are expressed in human brain and several other tissues, including heart, lung, kidney, placenta, pancreas, liver and skeletal muscle, but the 11 kb DRT transcript is preferentially expressed in fetal brain. Steady-state levels of DRT mRNA in several tissues, including brain, heart, lung and kidney, are greater in the midterm fetus than those in the adult. DRT transcripts are detectable at low levels in a human teratocarcinoma cell line (NTera-2), but its expression is greatly increased after the NTera-2 cells are induced to become postmitotic neurons (NTera-2N) by retinoic acid treatment. These data suggest that DRT plays a part in human neurogenesis. A large number of tumor cell lines derived from neuroectoderm express DRT transcripts, including 12 neuroblastomas, two medulloblastomas, one primitive neuroectodermal tumor and six small cell lung carcinomas (SCLC). Interestingly, several neuroblastoma cell lines with 1p deletion and one SCLC cell line express DRT transcripts of aberrant size (i.e. 3, 6 and 8 kb) in addition to those found in normal tissues. We mapped the DRT gene to human chromosome 1p35-1p36.1 by PCR screening of human-rodent somatic cell hybrid panels and by fluorescence in situ hybridization. As the distal end of chromosome 1p is often deleted in neuroblastomas and altered in some cases in SCLCs, these chromosomal abnormalities may have resulted in the generation of aberrant size transcripts. Thus, the DRT gene may play a part in neuroblastoma and SCLC tumorigenesis.

We analyzed 156 primary neuroblastoma tumor samples for loss of heterozygosity at the distal short arm of chromosome 1 (1p LOH). We also compared 1p LOH with known clinical and genetic prognostic variables as well as patient outcome. 1p LOH was detected in 30 of 156 tumors (19%) and was strongly associated with adverse clinical and biological features. 1p LOH was also strongly predictive of a poor outcome in univariate analyses (estimated 4-year survival, 32 +/- 10% SE versus 76 +/- 5% SE; P < 0.001). However, the prognostic value of 1p LOH was equivocal when stratified for amplification of the MYCN oncogene (P = 0.16). We conclude that 1p LOH is an important component of a pattern of genetic abnormalities in neuroblastoma associated with an aggressive clinical course.

Mutations of CDKN2 have been found recently in melanoma and many other tumor types. Neuroblastoma shares with melanoma a neuroectodermal origin and a high incidence of deletions of the short arm of chromosome 1. Therefore, we analyzed 18 primary neuroblastomas and 9 tumor-derived cell lines for mutations in CDKN2. We used PCR-single-strand conformation polymorphism to examine exons 1 and 2 of the CDKN2 gene for mutations, but none were detected. Furthermore, no homozygous deletions were detected and there was no loss of heterozygosity at the closely linked IFNA locus. We conclude that disruption of the CDKN2 gene is not required for malignant transformation of human neuroblastomas.