Isolation and structural analysis of a 1.2-megabase N-myc amplicon from a human neuroblastoma

Extrachromosomal DNA (ecDNA) in cancer pathogenesis

In cancer, oncogenes and surrounding regulatory regions can untether themselves from chromosomes, forming extrachromosomal DNA particles (ecDNAs). Because of their non-chromosomal inheritance, ecDNA drives high oncogene copy number and intratumoral genetic heterogeneity, endowing tumors with the ability to rapidly change their genomes, accelerating tumor evolution, and contributing to therapeutic resistance. Further, the circular topology of ecDNA leads to enhanced chromatin accessibility, altered gene regulation, and massive oncogene transcription, driving tumor growth and progression, and placing ecDNA at the interface of cancer genomics and epigenetics. Recent studies show that ecDNA is a common event in many of the most aggressive forms of cancer, potentially challenging our current precision oncology approaches. In this review, we discuss what is known about ecDNA and its biological and clinical impact, highlighting new research and suggesting the promise, and some of the challenges ahead for the field.

The biogenesis and roles of extrachromosomal oncogene involved in carcinogenesis and evolution

More and more extrachromosomal DNA (ecDNA) was found in human tumor cells in recent years, which has a high copy number in tumors and changes the expression of oncogenes, thus different from normal chromosomal DNA. These circular structures were identified to originate from chromosomes, and play critical roles in rapid carcinogenesis, tumor evolution and multidrug resistance. Therefore, this review mostly focuses on the biogenesis and regulation of extrachromosomal oncogene in ecDNA as well as its function and mechanism in tumors, which are of great significance for our comprehensive understanding of the role of ecDNA in tumor carcinogenic mechanism and are expected to provide ecDNA with the potential to be a new molecular target for the diagnosis and treatment of tumors.

Small extrachromosomal circular DNAs, microDNA, produce short regulatory RNAs that suppress gene expression independent of canonical promoters

Interest in extrachromosomal circular DNA (eccDNA) molecules has increased recently because of their widespread presence in normal cells across every species ranging from yeast to humans, their increased levels in cancer cells and their overlap with oncogenic and drug-resistant genes. However, the majority of eccDNA (microDNA) in mammalian tissues and cell lines are too small to carry protein coding genes. We have tested functional capabilities of microDNA by creating artificial microDNA molecules mimicking known microDNA sequences and have discovered that they express functional small regulatory RNA including microRNA and novel si-like RNA. MicroDNA are transcribed in vitro and in vivo independent of a canonical promoter sequence. MicroDNA that carry miRNA genes form transcripts that are processed by the endogenous RNA-interference pathway into mature miRNA molecules, which repress a luciferase reporter gene as well as endogenous mRNA targets of the miRNA. Further, microDNA that contain sequences of exons repress the endogenous gene from which the microDNA were derived through the formation of novel si-like RNA. We also show that endogenous microDNA associate with RNA polymerases subunits, POLR2H and POLR3F. Together, these results suggest that microDNA may modulate gene expression through the production of both known and novel regulatory small RNA.

A chemical screen identifies the chemotherapeutic drug topotecan as a specific inhibitor of the B-MYB/MYCN axis in neuroblastoma

The transcription factor MycN is the prototypical neuroblastoma oncogene and a potential therapeutic target. However, its strong expression caused by gene amplification in about 30% of neuroblastoma patients is a considerable obstacle to the development of therapeutic approaches aiming at eliminating its tumourigenic activity. We have previously reported that B-Myb is essentially required for transcription of the MYCN amplicon and have also shown that B-MYB and MYCN are engaged in a feed forward loop promoting the survival/proliferation of neuroblastoma cells. We postulated that pharmacological strategies breaking the B-MYB/MYCN axis should result in clinically desirable effects. Thus, we implemented a high throughput chemical screen, using a curated library of ~1500 compounds from the National Cancer Institute, whose endpoint was the identification of small molecules that inhibited B-Myb. At the end of the screening, we found that the compounds pinafide, ellipticine and camptothecin inhibited B-Myb transcriptional activity in luciferase assays. One of the compounds, the topoisomerase-1 inhibitor camptothecin, is of considerable clinical interest since its derivatives topotecan and irinotecan are currently used as first and second line treatment agents for various types of cancer, including neuroblastoma. We found that neuroblastoma cells with amplification of MYCN are more sensitive than MYCN negative cells to camptothecin and topotecan killing. Campothecin and topotecan caused selective down-regulation of B-Myb and MycN expression in neuroblastoma cells. Notably, forced overexpression of B-Myb could antagonize the killing effect of topotecan and camptothecin, demonstrating that the transcription factor is a key target of the drugs. These results suggest that camptothecin and its analogues should be more effective in patients whose tumours feature amplification of MYCN and/or overexpression of B-MYB.

Overexpression of SERTAD3, a putative oncogene located within the 19q13 amplicon, induces E2F activity and promotes tumor growth

The amplified region of chromosome 19q13.1-13.2 has been associated with several cancers. The well-characterized oncogene AKT2 is located in this amplicon. Two members of the same gene family (SERTAD1 and SERTAD3) are also located within this region. We report herein the genomic structure and potential functions of SERTAD3. SERTAD3 has two transcript variants with short mRNA half-lives, and one of the variants is tightly regulated throughout G1 and S phases of the cell cycle. Overexpression of SERTAD3 induces cell transformation in vitro and tumor formation in mice, whereas inhibition of SERTAD3 by small interfering RNA (siRNA) results in a reduction in cell growth rate. Furthermore, luciferase assays based on E2F-1 binding indicate that SERTAD3 increases the activity of E2F, which is reduced by inhibition of SERTAD3 by siRNA. Together, our data support that SERTAD3 contributes to oncogenesis, at least in part, via an E2F-dependent mechanism.

Negative implication of C-MYC as an amplification target in esophageal cancer

Chromosomal aberrations (amplifications and deletions) underlie the genesis or development of cancer. Amplification of 8q24 is one of the most frequent events in esophageal cancer. To define whether C-MYC is the target gene for 8q24 amplification, we performed fluorescence in situ hybridization using a MYC (8q24.12 approximately q24.13) probe in esophageal cancer from southern China. Furthermore, we detected the expression status of several genes including C-MYC, TRIB1 (alias C8FW), and FAM84B (alias NSE2) in the regions of 8q24 via reverse transcriptase-polymerase chain reaction or immunohistochemical analysis (or both). Distinct amplification of 8q24 was found in esophageal carcinomas. Only 4 of 46 cases showed obvious protein expression in part of the esophageal cancerous nest. In particular, increased protein expression of C-MYC was shown only in a small part of a cancerous nest in the four cases. Positive C-MYC staining was detected mainly in the cytoplasm of esophageal cancer cells. No expression of TRIB1 was detected in esophageal squamous cell carcinomas. Of 59 cases, 39 (66%) cases showed increased expression of FAM84B in esophageal carcinomas. The results suggest that C-MYC and TRIB1 may not be the amplification target of 8q24 in esophageal cancer. FAM84B might be involved in the genesis or development of esophageal cancer in southern China. Whether FAM84B is the amplification target of esophageal cancer awaits further investigation.

Manifestation, mechanisms and mysteries of gene amplifications

Gene amplifications are essential features of advanced cancers and have prognostic as well as therapeutic significance in clinical cancer treatment. Models explaining the amplification process, such as breakage-fusion-bridge cycle and excision and unequal segregation of extrachromosomal DNA fragments, predict that independent DNA double-stranded breaks must occur to induce amplification formation. Many cellular, tissue and environmental factors induce DNA damage and amplifications. Also labile DNA sequence features like fragile sites facilitate amplifications. Although, databases and data mining tools of various genomic attributes are already available, extra-large scale systems biology endeavors to decipher dynamics, interactions and dependencies between different factors contributing to amplification process fail, because current databases of DNA copy number aberrations and fragile sites comprise conventional cytogenetics results obtained at far too coarse chromosome band resolution. Array comparative genomic hybridization (aCGH) enables genome-wide gene copy number measurements and amplification detection at molecular genetic resolution. Similarly, cloning and sequencing of fragile sites produce mapping information of vastly improved resolution. In conclusion, databases of aCGH and sequenced fragile sites are needed to resolve the mechanisms of gene amplifications in systems biology configuration.

Alternative mechanisms of gene amplification in human cancers

Gene amplification is a common phenomenon in cancer. Cytogenetic analyses have indicated that breakage-fusion-bridge (BFB) cycles drive intrachromosomal amplification of some oncogenes in a head-to-head manner in human cancers. However, the complex structures of an amplified sequence found in cancers are not always explained by the BFB model. At the 17q21 locus, which is not linked to common fragile sites, we discovered a recombination hot spot harboring amplicon repeats in tandem in a head-to-tail orientation, with the interamplicon junctions in each cancer cell being homogeneous. These findings clearly show the presence of alternative mechanisms other than BFB cycles in oncogene amplification.

Neuroblastoma: biological insights into a clinical enigma

Neuroblastoma is a tumour derived from primitive cells of the sympathetic nervous system and is the most common solid tumour in childhood. Interestingly, most infants experience complete regression of their disease with minimal therapy, even with metastatic disease. However, older patients frequently have metastatic disease that grows relentlessly, despite even the most intensive multimodality therapy. Recent advances in understanding the biology and genetics of neuroblastomas have allowed classification into low-, intermediate- and high-risk groups. This allows the most appropriate intensity of therapy to be selected - from observation alone to aggressive, multimodality therapy. Future therapies will focus increasingly on the genes and biological pathways that contribute to malignant transformation or progression.

MYCN gene amplification. Identification of cell populations containing double minutes and homogeneously staining regions in neuroblastoma tumors

Neuroblastoma is the second most common solid tumor occurring in children. Amplification of the MYCN oncogene is associated with poor prognosis. To identify neuroblastoma tumors with MYCN amplification, we studied the number of copies of MYCN in interphase cells by fluorescence in situ hybridization in 20 neuroblastoma patients. MYCN amplification appeared in 7 tumor specimens. Interphase and metaphase studies showed a tumor cell population with both forms of amplification, double minutes and homogeneously staining regions, in two patients. These patients showed a smaller tumor cell subpopulation with the presence of more than one homogeneously staining region, suggesting that gene amplification was undergoing karyotype evolution.

MYCN is the only highly expressed gene from the core amplified domain in human neuroblastomas

MYCN amplification in neuroblastomas is strongly associated with advanced stages of disease and a poor prognosis. We have recently defined a 130 kb core region of the MYCN amplicon that is consistently amplified in neuroblastomas. However, it has been argued that other expressed sequences were coamplified with MYCN and, as a result, might contribute to the aggressive phenotype of MYCN-amplified neuroblastomas. Therefore, we have screened cosmids representing the core MYCN-amplified domain and surrounding DNA by using a differential hybridization approach to detect other amplified, highly expressed genes from this region. Our results suggest that MYCN is the only highly expressed gene consistently amplified in human neuroblastomas, and that the MYCN gene is likely to be the only selective marker for genomic amplification in these tumors.

Molecular genetics in the diagnosis and prognosis of solid pediatric tumors

The field of molecular genetics continues to see an ever increasing number of applications to pediatric tumor analysis. Studies in pediatric tumors have identified novel genes and other genetic changes, a large number of which reflect one of the following mechanisms: (1) activation of proto-oncogenes; (2) loss of tumor suppressor genes; or (3) creation of novel fusion proteins. At least one of these mechanisms is operational in each of the following pediatric tumors: neuroblastoma, Ewing sarcoma and peripheral primitive neuroectodermal tumor (pPNET), intra-abdominal desmoplastic small-cell tumor, rhabdomyosarcoma, synovial sarcoma, and Wilms tumor. Out of this research has come not only an increased understanding of oncogenesis but also, for each of the tumors listed above, diagnostic and/or prognostic markers that can be used by the pathologist and oncologist to improve overall patient management.

Overexpression of a DEAD box protein (DDX1) in neuroblastoma and retinoblastoma cell lines

The DEAD box gene, DDX1, is a putative RNA helicase that is co-amplified with MYCN in a subset of retinoblastoma (RB) and neuroblastoma (NB) tumors and cell lines. Although gene amplification usually involves hundreds to thousands of kilobase pairs of DNA, a number of studies suggest that co-amplified genes are only overexpressed if they provide a selective advantage to the cells in which they are amplified. Here, we further characterize DDX1 by identifying its putative transcription and translation initiation sites. We analyze DDX1 protein levels in MYCN/DDX1-amplified NB and RB cell lines using polyclonal antibodies specific to DDX1 and show that there is a good correlation with DDX1 gene copy number, DDX1 transcript levels, and DDX1 protein levels in all cell lines studied. DDX1 protein is found in both the nucleus and cytoplasm of DDX1-amplified lines but is localized primarily to the nucleus of nonamplified cells. Our results indicate that DDX1 may be involved in either the formation or progression of a subset of NB and RB tumors and suggest that DDX1 normally plays a role in the metabolism of RNAs located in the nucleus of the cell.

Chromosomal fragile sites and DNA amplification in drug-resistant cells

It has been well established that DNA amplification is one of the important mechanisms by which cultured cells acquire resistance to many cytotoxic compounds. Amplification of important genes including those encoding oncoproteins, growth factors, their receptors and cell-cycle regulators has been reported in human neoplasms. Yet, despite intensive research since the first description of DNA amplification in cultured cells about 20 years ago, the mechanisms of DNA amplification remain largely unknown. Many models have been proposed to account for the diverse manifestations of amplified DNA in many different cell sources. It is not the intention of this commentary to review these many different models. Rather, we wil focus on the recent advances in this area of research, made mainly via the fluorescence in situ hybridization technique, that have revealed a fairly common chromosomal manifestation of amplified DNA in the drug-resistant hamster cell lines and have demonstrated the association of chromosomal fragile site breakage with early events in DNA amplification. These new developments underscore the importance of future research toward understanding the molecular bases of chromosomal fragile sites, including mechanisms involved in DNA strand breakage and repair, chromosomal translocations, and deletions, which may, in turn, provide important new insights into genomic plasticity and neoplastic transformation.

Relational mapping of MYCN and DDXI in band 2p24 and analysis of amplicon arrays in double minute chromosomes and homogeneously staining regions by use of free chromatin FISH

MYCN amplification has been observed in diverse neuronal tumors including neuroblastoma, retinoblastoma, and small cell carcinoma of the lung, and has been correlated with a poor prognosis in advanced-stage neuroblastomas. Recent studies have shown a co-amplification of DDXI, a DEAD box gene, and MYCN in retinoblastoma and neuroblastoma. DDXI has been mapped to within a megabase of the MYCN gene in band 2p24. In the present study, the relational map of DDXI and MYCN by fluorescence in situ hybridization (FISH) mapping to metaphase cells and extended free chromatin fibers indicated that DDXI is telomeric to MYCN. Dual-color FISH analysis of amplicons within arrays of extended chromatin fibers was performed to examine the physical relationship of MYCN and DDXI within double minute chromosomes (dmins) and homogeneously staining regions (hsrs). No regular reiterated amplicon repeat unit was present in the hsrs, but detailed analysis of the configurations of DDXI and MYCN within each array indicated that multiple rearrangements generated a complex hsr amplicon structure. Similarly, analysis of a cell line bearing dmins showed that a composite amplicon structure involving deletions and/or duplications of MYCN and DDXI is a feature of dmin formation. These data are consistent with a molecular mechanism involving many rearrangements during the evolution of gene amplification, resulting in complex amplicon structures with distinct changes in relative gene copy number and considerable variation in intragenic distances between coamplified genes.

Analysis of candidate gene co-amplification with MYCN in neuroblastoma

Previous studies have revealed that the MYCN gene spans approximately 7kb, while the amplicon has been estimated to be 100 kb to 1 Mb long [1-3]. This implies that several other genes may be present on the MYCN amplicon. Such co-amplified genes could contribute to the malignant phenotype and might provide an explanation for why not all patients with MYCN amplification have a poor outcome. We investigated 7 neuroblastoma cell lines and 167 primary tumours for the co-amplification of candidate genes known to be present near the MYCN locus: ornithine decarboxylase, ribonucleotide reductase, syndecan-1 and a DEAD box protein gene, DDX1. We also investigated further the pG21 expressed sequence previously reported to be co-amplified with MYCN. No co-amplification with the first 3 genes was found in any of the cell lines or tumour samples. DDX1 was found to be amplified along with MYCN in 4/6 (67%) cell lines and 18/38 (47%) of the MYCN amplified tumours. No amplification of DDX1, ODC1, RRM2 or syndecan-1 was found in the absence of MYCN amplification. DDX1 co-amplification was observed to occur mainly in stage 4 or 4S patients. With the exclusion of those with 4S disease, patients with DDX1 co-amplification had a significantly shorter mean (+/- SE) disease-free interval (4.1 +/- 1.4, n = 8 months) compared with patients with MYCN amplification alone (19.6 +/- 4.5, n = 17) (P = 0.04, Welch's unpaired t-test). The pG21 sequence was identical to part of the DDX1 gene. These observations indicate that there is at least 1 other gene co-amplified with MYCN in a proportion of cases and that those patients with DDX1 co-amplification tend to relapse more quickly. It also implies that the MYCN amplicon is of varied size and/or position relative to the MYCN gene.

Amplification of 19q13.1-q13.2 sequences in ovarian cancer. G-band, FISH, and molecular studies

In this study of ovarian carcinoma, we extended previous findings by performing FISH using chromosome 19 paint and microFISH probes and patient samples with and without abnormalities of chromosome 19 identified by G-banding. Karyotype interpretations of der(19) were confirmed, while additional 19 translocations were also detected by FISH with 19WCP in some cases. Similar FISH studies of ovarian carcinoma cell lines found chromosome 19 abnormalities even after extensive in vitro culture. MicroFISH probes were generated by chromosome microdissection from two cases with hsr(19) and mapped to 19q13.2 and 19q13.1-.2, respectively. FISH with these microFISH probes alone or in combination with a 19WCP probe to four patient samples and seven cell lines showed that 65% of chromosome 19 structural abnormalities contained 19q13.1-q13.2 sequences, sometimes as large hsrs. Ovarian cancer cell lines showed amplification and overexpression of the AKT2 putative oncogene, but not the ERCC-2 DNA repair gene in this chromosomal region. In addition to AKT2, amplification and overexpression of other yet-unidentified genes in the 19q13.1-q13.2 region may contribute to ovarian carcinoma pathogenesis or progression.

The DDX1 gene maps within 400 kbp 5' to MYCN and is frequently coamplified in human neuroblastoma

Human neuroblastoma cells frequently show amplification of the oncogene MYCN, which maps to 2p24. Previous studies have localized the DEAD box motif gene DDX1 to the same chromosome band and demonstrated coamplification of DDX1 and MYCN in two retinoblastoma cell lines. Recently, a high frequency of coamplification of DDX1 and MYCN has been shown in human neuroblastoma cells. We have determined the physical distance between the two genes by pulsed field gel electrophoresis in normal tissue and have found that DDX1 maps to a position at a maximum distance of 400 kbp 5' to MYCN. Two neuroblastoma cell lines with coamplification of DDX1/MYCN showed a similar topographic relationship of the two genes. In contrast, in two cell lines with high copy number, the DDX1 gene was not present in all amplified units recognized by MYCN and had changed its position in the amplified DNA relative to MYCN from 5' to 3', presumably by rearrangement during the amplification process. Our data show that the high frequency of DDX1 coamplification is due to its close physical distance to MYCN. Although amplification has resulted in an elevated expression of DDX1 the significance of overexpression for neuroblastoma remains unclear.

Co-amplification and concomitant high levels of expression of a DEAD box gene with MYCN in human neuroblastoma

MYCN gene amplification is strongly correlated with poor prognosis in neuroblastoma (NB), the second most common solid pediatric tumor. However, increased MYCN expression seen in tumors that lack MYCN amplification does not correlate with aggressive clinical behavior. Whereas the MYCN gene spans only 7 kb, the MYCN amplicon has been shown to range in size from 350 kb to more than 1 Mb. Given the large size of the amplicon, it is possible that additional genes are co-amplified in NBs whose expression may contribute to the aggressive phenotype associated with MYCN-amplified tumors. We isolated a cDNA clone from a human NB library that is identical to DDXI, a gene recently reported to be preferentially expressed in two retinoblastoma cell lines that also express high levels of MYCN. DDXI belongs to a family of genes that encode DEAD (Asp-Glu-Ala-Asp) box proteins, putative ATP-dependent RNA helicases implicated in a number of cellular processes involving alterations of RNA secondary structure. We examined the frequency of DDXI amplification in 15 NB cell lines, 1 neuroepithelioma cell line, and 122 NB tumors by Southern blot analyses, and we found that 7 of 10 MYCN-amplified cell lines and 27 of 40 (68%) MYCN-amplified tumors also harbored multiple copies of the DDXI gene. Amplification of DDXI was associated with high levels of DDXI mRNA expression in the NB cell lines and tumors as examined by Northern analysis. Neither DDXI gene amplification nor enhanced expression was observed in tumors or cell lines that lacked MYCN amplification. Because RNA helicases play important roles in both post-transcriptional and translational gene regulation, high levels of DDXI expression consequent to genomic amplification may contribute to the malignant phenotype of a subset of NBs.

Gene amplification in human gliomas

Gliomas represent the largest group of primary brain tumors in adults. The astrocytic variants are the most common and the adult forms are histologically stratified into three malignancy grades. Of these glioblastoma is the most common and the most malignant; it has also been best studied by molecular genetics and cytogenetics. Double-minute chromosomes, known to represent amplified genes, are found in 50% of glioblastomas. Amplified genes are not detected in the most benign of the astrocytomas. Many genes have been shown to be amplified in more than single cases of gliomas and these include EGFR, CDK4, SAS, MDM2, GLI, PDGFAR, MYC, N MYC, MYCL1, MET, GADD153, and KIT. The most commonly amplified genes in glioblastomas are EGFR (in approximately 40%), CDK4, and SAS (in approximately 15%). The remainder of the genes are amplified at lower frequency. The best mapped amplicon in gliomas involves the 12q13-14 region. The amplicon is of undetermined size, encompasses a number of genes, and may be rearranged. It occurs in 15% of glioblastomas and almost always includes the CDK4 and SAS genes, in about 10% of tumors the MDM2 gene, and at lower frequency GLI, GADD153, and A2MR. All but A2MR are overexpressed if amplified. The amplified EGFR gene is frequently rearranged, resulting in changes in the regions of the transcript that codes for the extracellular domain. The resultant receptor is constitutively activated. These findings provide examples of the impact the use of modern molecular biological techniques has had on our understanding of oncogenic mechanisms in gliomas.

Mitochondrial ATP synthase alpha-subunit gene amplified in a retinoblastoma cell line maps to chromosome 18

The human retinoblastoma cell line Y79 has multiple copies of the MYCN gene and the DEAD box gene DDXI. Both genes have been mapped to chromosome band 2p24. A third gene, encoding the alpha-subunit of mitochondrial ATP synthase (ATPSA), is also amplified in Y79. Here we report that there are at least four human mitochondrial ATPSA-related genes located on four different chromosomes. The ATPSA gene that is amplified in Y79 originates from chromosome 18. In Y79, the amplified copies of both the ATPSA and the MYCN genes are located on a homogeneously staining region (HSR) at chromosome band Ip34.

The evolution of small DNA viruses of eukaryotes: past and present considerations

Historically, viral evolution has often been considered from the perspective of the ability of the virus to maintain viral pathogenic fitness by causing disease. A predator-prey model has been successfully applied to explain genetically variable quasi-species of viruses, such as influenza virus and human immunodeficiency virus (HIV), which evolve much faster rates than the host. In contrast, small DNA viruses (polyomaviruses, papillomaviruses, and parvoviruses) are species specific but are stable genetically, and appear to have co-evolved with their host species. Genetic stability is attributable primarily to the ability to establish and maintain a benign persistent state in vivo and not to the host DNA proofreading mechanisms. The persistent state often involves a cell cycle-regulated episomal state and a tight linkage of DNA amplification mechanisms to cellular differentiation. This linkage requires conserved features among viral regulatory proteins, with characteristic host-interactive domains needed to recruit and utilize host machinery, thus imposing mechanistic constrains on possible evolutionary options. Sequence similarities within these domains are seen amongst all small mammalian DNA viruses and most of the parvo-like viruses, including those that span the entire spectrum of evolution of organisms from E. coli to humans that replicate via a rolling circle-like mechanism among the entire spectrum of organisms throughout evolution from E. coli to humans. To achieve benign inapparent viral persistence, small DNA viruses are proposed to circumvent the host acute phase reaction (characterized by minimal inflammation) by mechanisms that are evolutionarily adapted to the immune system and the related cytokine communication networks. A striking example of this is the relationship of hymenoptera to polydnaviruses, in which the crucial to the recognition of self, development, and maintenance of genetic identity of both the host and virus. These observations in aggregate suggest that viral replicons are not recent "escapies" of host replication, but rather provide relentless pressure in driving the evolution of the host through cospeciation.

Molecular basis for heterogeneity in human neuroblastomas

Neuroblastomas demonstrate both clinical and biological heterogeneity. We have proposed that neuroblastomas may be classified in three genetically distinct subtypes, based on cytogenetic and molecular analysis. The first comprises those with hyperdiploid or triploid modal karyotypes (or compatible DNA content by flow cytometry), 1p LOH and MYCN amplification are absent, and TRKA expression is high. These patients are likely to be infants with low stages of disease (stages 1, 2, or 4S by the International Neuroblastoma Staging System), and they have a very favourable outcome (> 90% cure). The second group consists of tumours that generally have a near diploid or tetraploid modal chromosome number or DNA content but lack MYCN amplification. They usually have 1p allelic loss, 14q allelic loss or other structural changes, and TRKA expression is usually low. These patients are generally older with advanced stages of disease (stages 3 or 4), and they have a slowly progressive course, with a cure rate of 25-50%. The third group is characterised by tumours with MYCN amplification. These tumours are generally near diploid or tetraploid, with 1p allelic loss, and low or absent TRKA expression. The patients are usually between 1 and 5 years of age with advanced stages of disease, and they have a very poor prognosis (< 5%). It remains to be determined if tumours in one group ever evolve into a less unfavourable group, but current evidence suggests that they are distinct genetically. The identification of the oncogenes, suppressor genes and growth factor receptor pathways involved in neuroblastomas has provided great insight into the mechanisms of malignant transformation and progression, and ultimately they may provide the targets for future therapy.

Molecular biology of double-minute chromosomes

Double-minute chromosomes play a critical role in tumor cell genetics where they are frequently associated with the overexpression of oncogene products. They have been observed for many years in light microscopic examinations of metaphase chromosomes from tumor cells, but their origin remains unknown and is the subject of considerable speculation. However, molecular details of their structure and organization can now be described in conjunction with the microscopic examinations, to allow an evaluation of the various models that have been developed to explain the genesis of double-minutes. The evidence now favors simple models that invoke chromosome breakage and circularization of very large acentric chromosome fragments, permitting unequal segregation of the genes on the fragment during cell division. If there is selection for overexpression of one of the genes on the fragment, daughter cells with more fragments will grow faster than daughter cells with fewer fragments, and over time the population of cells will come to contain many double-minutes per cell.

Preferential amplification of the paternal allele of the N-myc gene in human neuroblastomas

Genomic imprinting plays a role in influencing the parental origin of genes involved in cancer-specific rearrangements. We have analysed 22 neuroblastomas with N-myc amplification to determine the parental origin of the amplified N-myc allele and the allele that is deleted from chromosome 1p. We analysed DNA from neuroblastoma patients and their parents, using four polymorphisms for 1p and three for the N-myc amplicon. We determined that the paternal allele of N-myc was preferentially amplified (12 out of 13 cases; P = 0.002). However, the paternal allele was lost from 1p in six out of ten cases, consistent with a random distribution (P > 0.2). These results suggest that parental imprinting influences which N-myc allele is amplified in neuroblastomas, but it does not appear to affect the 1p allele that is deleted in the cases that we have examined.