Studies on the human retinoblastoma susceptibility gene

The retinoblastoma susceptibility (RB) gene is unique among other cloned cancer genes because its causal role in a human cancer, retinoblastoma, was established by classical genetic methods before its isolation. Earlier hypotheses and experimental data suggested that inactivation of a gene in chromosome band 13q14 resulted in retinoblastoma formation. A gene in this region was identified as the RB gene on the basis of mutations found specifically in retinoblastoma tumors; however, its proposed biological activity in suppressing neoplasia has yet to be demonstrated. The RB gene product was identified as a nuclear phosphoprotein of 110 kD associated with DNA binding activity, suggesting that the RB protein may regulate other genes. Probes for the RB gene and gene product will be useful for genetic diagnosis of retinoblastoma susceptibility in affected families; for direct detection of mutant RB alleles; and, potentially, for genetic diagnosis of susceptibility to osteosarcoma and other tumors tentatively linked to RB-gene dysfunction. Continued study of the RB gene should yield further insight into mechanisms of oncogenesis, development, and gene regulation.     

Exclusion of the retinoblastoma gene and chromosome 13q as the site of a primary lesion for human breast cancer.

Chromosome 13q has been suggested as the site of a gene predisposing to human breast cancer, because loss of heterozygosity of alleles on this chromosome has been observed in some ductal breast tumors and because two breast cancer lines are altered at the retinoblastoma gene (RB1) at 13q14. To test this possibility, linkage of breast cancer susceptibility to 14 loci on chromosome 13q loci was assessed in extended families in which breast cancer is apparently inherited as an autosomal dominant trait. RB1 was excluded as the site of a breast cancer gene by a lod score of Z = -7.60 at close linkage for 13 families. Multipoint analysis yielded negative lod scores throughout the region between 13q12 and 13q34; over most of this distance, Z less than -2.0. Therefore, chromosome 13q appears to be excluded as the site of primary lesion for breast cancer in these families. In addition, comparison of tumor versus normal tissues of nonfamilial breast cancer patients revealed an alteration at the 5' end of RB1 in a mucoid carcinoma but no alterations of RB1 in five informative ductal adenocarcinomas. Linkage data and comparisons of tumor and normal tissues suggest that changes in the RBI locus either are secondary alterations associated with progression of some tumors or occur by chance.     

A splicing mutation in RB1 in low penetrance retinoblastoma

The pediatric eye-tumor retinoblastoma is widely held as a paradigm of human cancer genetics and has been a model system for both the two-hit hypothesis of dominantly inherited cancer as well as for the concept of tumor-specific loss of constitutional heterozygosity to achieve expression of the tumorigenic phenotype. Familial retinoblastoma is usually inherited as an autosomal dominant disease with high penetrance and expressivity. In a small but significant number of families, however, retinoblastoma is inherited with greatly reduced penetrance and expressivity. In these families, retinoblastoma tumors occur relatively late, are often unilateral, and unaffected carriers may exist. We have identified a mutation in such a family that exhibited extremely low penetrance and expressivity. This mutation appeared to affect splicing of the mutant allele such that both a normal length RB1 mRNA and a truncated RB1 mRNA were expressed from the same allele. Received: 7 March 1997 / Accepted: 29 April 1997     

Prediction of homologous binding sites on RB and p107 common for viral oncoproteins and cellular ligands.

Hydropathic anticomplementarity of amino acids specifies that peptides translated from complementary DNA strands may acquire amphiphilic conformations and bind to each other. This concept has been coined 'Molecular Recognition Theory' (MRT) or 'complementary peptide theory'. Inactivation of retinoblastoma protein (RB), a tumor suppressor gene product, has been shown to be involved in the pathogenesis of many tumors and to be due to either mutation of the RB gene, hyperphosphorylation or complex formation with viral oncoproteins. The viral oncoproteins share a common RB binding motif with cellular ligands. The exact site on RB associating with this common RB binding motif of viral oncoproteins and cellular ligands has not been identified yet. This study is the first to predict putative binding sites on RB and p107, a cellular protein with RB sequence homology, respectively, by using the hydropathic complementarity approach. These sites are residues 649-654 of RB and 657-662 of p107. Moreover, this paper proposes a structure for a potential antineoplastic agent based on the amino acid sequence of the predicted RB binding site. The data presented herein should have important implications both for the understanding of cancer pathophysiology and for the drug design of antineoplastic compounds.     

Human retinoblastoma gene: long-range mapping and analysis of its deletion in a breast cancer cell line.

Mutational inactivation of the retinoblastoma (RB) gene is considered a fundamental event in the formation of several types of human cancer. A substantial proportion of RB gene mutations are partial or complete deletions that extend an unknown distance beyond one or both ends of the gene. To provide a framework for measuring the extent of these deletions, we have constructed a long-range restriction map of SfiI sites spanning 850 kilobases around the RB gene. This map was applied in a molecular analysis of RB gene deletion in breast cancer cell line MB468. A previous study of this cell line demonstrated deletion of the entire RB gene except for exons 1 and 2 (E. Y.-H. P. Lee, H. To, J.-Y. Shew, R. Bookstein, P. Scully, and W.-H. Lee, Science 241:218-221, 1988). Genomic clones containing the deletion junction were isolated from a library made from MB468 DNA. A probe obtained from the far side of the deletion junction was used to localize and clone the unknown 3' endpoint, demonstrating that the chromosomal mutation in this case was a simple deletion spanning 200 kilobases. Sequence analysis of the deletion junction indicated a conservative deletion with no loss or gain of nucleotides. The deletion endpoints had no sequence homology to each other or to any repetitive sequence family, such as Alu, so the recombination event was illegitimate. Structural analysis of this and other RB gene deletions is important for understanding molecular mechanisms of recessive oncogenesis.     

Detection of mutations of the RB1 gene in retinoblastoma patients by using exon-by-exon PCR-SSCP analysis.

Most sporadic cases of retinoblastoma, malignant eye tumor of children, may require the identification of a mutation of the retinoblastoma gene (RB1 gene) for precise genetic counseling. We established a mutation detection system of and screened for the RB1 gene mutation in 24 patients with retinoblastoma--12 bilateral patients and 12 unilateral patients. Mutation analysis was performed by PCR-mediated SSCP analysis in the entire coding region and promoter region, as an initial screening method, followed by direct genomic sequencing. Possible oncogenic mutations were identified in 14 (58%) of 24 tumors, of which 6 were single base substitutions, 4 were small deletions, 3 were small insertions, and 1 was a complex alteration due to deletion-insertion. A constitutional somatic mosaicism was suggested in one bilateral patient. A majority (57%) of mutations were found in E1A binding domains, and all were presumed to truncate the normal gene products. The mutation analysis presented here may provide a basis for the screening system of RB1 gene mutations in retinoblastoma patients.     

Mutation Spectrum of RB1 Gene in Unilateral Retinoblastoma Cases from Tunisia and Correlations with Clinical Features

Retinoblastoma, an embryonic neoplasm of retinal origin, is the most common primary intraocular malignancy in children. Somatic inactivation of both alleles of the RB1 tumor suppressor gene in a retinal progenitor cell through diverse mechanisms including genetic and epigenetic modifications, is the crucial event in initiation of tumorigenesis in most cases of isolated unilateral retinoblastoma. We analyzed DNA from tumor tissue and from peripheral blood to determine the RB1 mutation status and seek correlations with clinical features of 37 unrelated cases of Tunisian origin with sporadic retinoblastoma. All cases were unilateral except one who presented with bilateral disease, in whom no germline coding sequence alteration was identified. A multi-step mutation scanning protocol identified bi-allelic inactivation of RB1 gene in 30 (81%) of the samples tested. A total of 7 novel mutations were identified. There were three tumors without any detectable mutation while a subset contained multiple mutations in RB1 gene. The latter group included tumors collected after treatment with chemotherapy. There were seven individuals with germline mutations and all presented with advanced stage of tumor. There was no difference in age of onset of RB based on the germline mutation status. Thus 20% of the individuals with sporadic unilateral RB in this series carried germline mutations and indicate the importance of genetic testing all children with sporadic retinoblastoma. These findings help to characterize the spectrum of mutations present in the Tunisian population and can improve genetic diagnosis of retinoblastoma.     

Insights from animal models on the origins and progression of retinoblastoma

The RB gene was discovered 20 years ago because of its role in the childhood eye cancer retinoblastoma. However, surprisingly little progress was made in defining the role of RB protein in the retina. In the last two years, new models exploiting conditional deletion of the mouse Rb gene have altered this picture radically. These models provide insight into the first Rb function, the cell of origin of retinoblastoma, the window during which Rb acts, distinct cell-specific defenses against Rb loss, the number and type of post-Rb lesions required for transformation, why pediatric tumors exist, the controversial role of the p53 pathway in retinoblastoma, and the reason why the disease is virtually unique to humans. Two years have dramatically improved our understanding of Rb function in the tissue that gave us this important tumor suppressor.     

Molecular detection of chromosomal translocations that disrupt the putative retinoblastoma susceptibility locus.

A candidate DNA sequence with many of the properties predicted for the retinoblastoma susceptibility (RB1) locus has been cloned (S. H. Friend, R. Bernards, S. Rogelj, R. A. Weinberg, J. M. Rapaport, D. M. Albert, and T. P. Dryja, Nature [London] 323:643-645, 1986). The large size of this gene (ca. 200 kilobases [kb]) and its multiple dispersed exons (Wiggs et al., N. Engl. J. Med. 318:151-157, 1988) complicate molecular screening strategies important in prenatal and presymptomatic diagnosis and in carrier detection. Here we used field inversion gel electrophoresis (FIGE) to construct a restriction map of approximately 1,000 kb of DNA surrounding the RB1 locus and to detect the translocation breakpoints in three retinoblastoma patients. DNA probes from either the 5' or 3' end of the gene were used to detect a 250-kb EagI restriction fragment in DNA from unaffected individuals. Both probes identified an additional hybridizing fragment in the DNA from each patient, permitting the breakpoints in all three to be mapped within the cloned RB1 gene. Analysis of the breakpoint in one translocation cell line allowed the RB1 gene to be oriented with its 5' end toward the centromere. The 5' end of the gene also appeared to be associated with a clustering of sites for several infrequently cleaving restriction enzymes, indicating the presence of an HpaII tiny fragment island. The detection and mapping of the translocation breakpoints of all three retinoblastoma patients to within the putative RB1 gene substantiated the authenticity of this candidate sequence and demonstrated the utility of FIGE in detecting chromosomal rearrangements affecting this locus.     

Stability of retinoblastoma gene expression determines the tumorigenicity of reconstituted retinoblastoma cells.

Mutational inactivation of the retinoblastoma gene (RB) is an invariant feature of the childhood eye cancer retinoblastoma and of tumor cells derived therefrom. In a previous study, retrovirus-mediated transfer of wild-type RB into cultured retinoblastoma cells resulted in a marked enlargement and reduced growth rate of these cells, as well as loss of their tumorigenic properties in nude mice. It was therefore difficult to separate the proposed growth-suppressing and tumor-suppressing activities of RB protein. Here, we show that clones of RB-reconstituted retinoblastoma cells can be isolated that stably express apparently normal RB protein for at least 20 months of continuous culture. These clones were indistinguishable from nonreconstituted cells by multiple parameters including morphology, growth rate, and cell cycle distribution. Despite similar phenotypes in culture, clones with stable RB expression were uniformly nontumorigenic in nude mice, whereas those that lost such expression regained their tumorigenic properties. These results indicate that the tumorigenicity of these cells is entirely determined by the presence or absence of exogenous RB protein expression and that suppression of tumorigenicity is distinct from inhibition of cellular growth in culture.     

No evidence for sequences structurally related to the RB1 gene in the human genome

Summary The retinoblastoma (RB1) gene is a ubiquitously expressed gene encoding a cell-cycle control protein. Inactivation of this gene plays a crucial role in the development of retinoblastoma, osteosarcoma, and other tumors. In a search for structurally related gene sequences we identified a 5.5-kb BamHI fragment strongly cross-hybridizing with the 5 end of the RB1 cDNA. Molecular cloning, in situ hybridization, restriction mapping, and sequence analysis identified this DNA segment as the 28S rRNA gene. The absence of other cross-hybridizing sequences suggests that the RB1 gene is not part of a structurally related gene family.     

The quest for a tumor suppressor gene phenotype

Our current definitions of the tumor suppressor gene (TSG) have been guided by the identification of the prototypical gene, RB1, a TSG that is implicated in the development of both the inherited and sporadic forms of retinoblastoma. The hallmark feature of this TSG is loss of function in tumoral cells, which can be restored by reintroduction of a normally functioning protein with concomitant reversion of tumorigenicity. Key to this discovery was that loss of function is often achieved by deletion of a normal copy of the TSG and retention of a mutated allele, which was either inherited or acquired. Suppression of tumorigenicity and the loss-of-function concept of TSGs was also demonstrated in early studies where normal cellular growth was achieved when tumorigenic cells were fused with normal cells. Thus loss of genetic content and restoration of gene function has guided studies aimed at the discovery of novel TSGs. Here we review the successes of TSG discovery using three approaches that are based on the genetic analysis of inherited predisposition to cancer, tumors that display chromosome loss, and tumorigenic cells that display a suppression of tumorigenicity as a result of transfer of normal chromosomes. Based on a review of the literature we conclude that the discovery of TSGs has been highly successful in the genetic analysis of inherited predisposition to cancer with a dominant mode of inheritance. In contrast, the latter two approaches have yielded a paucity of TSGs that exhibit features similar to the prototypical RB1 in that they are rarely inactivated by somatic mutations in tumors displaying LOH, although decreased gene expression is observed. Nevertheless, some of these genes have been shown to suppress tumorigenicity when normal function is restored in tumorigenic cells consistent with the loss-of-function concept. These observations continue to challenge our current definition of TSG.     

Tumorspheres but Not Adherent Cells Derived from Retinoblastoma Tumors Are of Malignant Origin

Verification that cell lines used for cancer research are derived from malignant cells in primary tumors is imperative to avoid invalidation of study results. Retinoblastoma is a childhood ocular tumor that develops from loss of functional retinoblastoma protein (pRb) as a result of genetic or epigenetic changes that affect both alleles of the RB1 gene. These patients contain unique identifiable genetic signatures specifically present in malignant cells. Primary cultures derived from retinoblastoma tumors can be established as non-adherent tumorspheres when grown in defined media or as attached monolayers when grown in serum-containing media. While the RB1 genotypes of tumorspheres match those of the primary tumor, adherent cultures have the germline RB1 genotype. Tumorspheres derived from pRb-negative tumors do not express pRb and express the neuroendocrine tumor markers synaptophysin and microtubule-associated protein 2 (MAP2). Adherent cells are synaptophysin-negative and express pRb, the epithelial cell marker cytokeratin that is expressed in the retinal pigmented epithelium and the vascular endothelial cell marker CD34. While tumorspheres are of malignant origin, our results cast doubt on the assumption that adherent tumor-derived cultures are always valid in vitro models of malignant cells and emphasize the need for validation of primary tumor cultures.     

Loss of the retinoblastoma tumor suppressor protein in murine calvaria facilitates immortalization of osteoblast-adipocyte bipotent progenitor cells characterized by low expression of N-cadherin

The retinoblastoma gene, RB1, is frequently inactivated in a subset of tumors, including retinoblastoma and osteosarcoma (OS). One characteristic of OS, as well as other tumors in which RB1 is frequently inactivated, is the lack of N-cadherin-mediated cell-cell adhesions. The frequent inactivation of RB1 and parallel loss of N-cadherin expression in OS prompted us to ask whether these observations are directly related to each other. In this study, we observed reduced N-cadherin expression in RB1(-/-) calvarial osteoblasts. In addition, RB1(-/-) cell lines had increased migration potential compared to their RB1(+/+) counterparts. These properties of RB1(-/-) cell lines correlated with an adipogenic potential lacking in RB1(+/+) cell lines, suggesting that each property is present in an immature progenitor cell. The isolation of a cell population with low surface expression of N-cadherin and enhanced adipogenic ability supports this view. Interestingly, the acute loss of pRb does not affect N-cadherin expression or migration or confer adipogenic potential to immortalized RB1(+/+) calvarial cells, suggesting that these traits are not a direct consequence of pRb loss; rather, pRb loss leads to the expansion and immortalization of an immature progenitor pool characterized by these properties.     

Detection of retinoblastoma gene copy number in metaphase chromosomes and interphase nuclei by fluorescence in situ hybridization.

Fluorescence in situ hybridization (FISH) was applied to detect the copy number of the retinoblastoma (RB1) tumor suppressor gene in metaphase chromosomes and interphase nuclei. We used 14 lambda phage clones spanning the whole RB1 gene region as a probe and obtained a specific hybridization signal in normal metaphase chromosomes at 13q14. Normal interphase nuclei showed two RB1 signals in about 90% of cases, whereas two cell lines with cytogenetically defined deletions involving the RB1 gene showed only one hybridization signal in about 80% of the nuclei. Analogous changes were detected in metaphase chromosomes. Multicolor FISH with subsets of the phage clones allowed visualization of subregions within the 200-kb gene in interphase nuclei. Analysis of clinical breast cancer samples showed that most of the cells contained two copies of the RB1 gene, even when restriction fragment length polymorphism analysis showed loss of heterozygosity (LOH) at the RB1 locus. This indicates that LOH at the RB1 locus in breast cancer cells probably involves mechanisms other than physical deletion.