SV40 large T-antigen: dual oncogene

Simian virus 40 (SV40) is a small, DNA-containing tumour virus. One of its gene products, the large tumour antigen (T-ag), is essential for both viral replication and cell transformation. SV40 T-ag can be considered a dual oncogene protein; it is a composite transforming protein that provides distinct functions at different subcellular locations. In addition to its roles in virus replication, T-ag exerts numerous effects on host cells. Those cellular effects reflect viral stimulation of host cell entry into S phase. Numerous chemical modifications have been ascribed to T-ag. They might be involved in defining subpopulations of T-ag that are, in turn, responsible for mediating various T-ag biochemical functions. The T-ag polypeptide, 90,000-100,000 in molecular weight, appears to contain multiple, discrete functional domains; several biological activities have been assigned to relatively small defined regions of the molecule. The cellular progenitors of the T-ag biochemical activities are not obvious. A cellular protein, p53, thought to be involved in regulation of cell proliferation, becomes complexed with T-ag in transformed cells and is stabilized. The interaction of T-ag with this cellular substrate may play an important part in SV40 transformation. T-ag and T-ag/p53 complexes are localized in both the nucleus and plasma membrane of transformed cells. T-ag is transported to the nucleus because of a 7-residue nuclear transport signal contained within its primary sequence. Its migration to the membrane is by an unknown pathway. Only a minor fraction of the total cellular T-ag is present at the cell surface. Both amino and carboxy termini of the T-ag polypeptide are exposed on the extracellular face of the cell. Nuclear and membrane T-ag are structurally very similar, although a portion of membrane T-ag is acylated and nuclear T-ag is not. The nuclear and membrane forms of T-ag apparently provide separate and complementary functions necessary for cell transformation. Nuclear T-ag is important in immortalizing primary cells and membrane T-ag may mediate more pronounced morphological changes. A model is presented, postulating how the two forms of T-ag might cooperate to mediate phenotypic transformation.     

Establishment and characterization of SV40 T-antigen immortalized human esophageal epithelial cells.

Normal human esophageal autopsy tissue was explanted in serum-free medium. The epithelial outgrowths were subcultured and then transfected by strontium phosphate coprecipitation with plasmid pRSV-T consisting of the RSV-LTR promoter and the sequence encoding the simian virus 40 large T-antigen. The transfected cells, but not the sham-transfected controls, formed multilayered colonies within 3-4 weeks, after which the colonies were transferred and cell strains (HE-451 and HE-457) developed. Both cell strains grew exponentially for 8-10 weeks and then senesced. After a "crisis" of 6-8 months, growth resumed in isolated colonies. One line, HET-1A from HE-457, was developed and has now undergone more than 250 population doublings. This line has retained epithelial morphology, stains positively for cytokeratins and the simian virus 40 T-antigen gene by immunofluorescence, and has remained nontumorigenic in athymic, nude mice for more than 12 months. Karyotypic analysis by Giemsa banding has shown that HET-1A is hypodiploid (34-40 chromosomes). Growth factor studies have shown that HET-1A is stimulated by Ca2+, and inhibited by fetal bovine serum, transforming growth factor-beta 1, and transforming growth factor-beta 2. This serum-free immortalized esophageal cell system will be useful for investigating the action of putative esophageal carcinogens.     

Detection and quantification of SV40 large T-antigen DNA in mesothelioma tissues and cell lines

SV40 viral DNA sequences, particularly large T-antigen (T-ag) DNA, have been reported in malignant pleural mesothelioma (MPM) and suggested to play a role in the tumorigenesis of this cancer. These results remain somewhat controversial owing to variability among a number of laboratories in reported SV40 DNA and protein detection in MPM tissues. This could be explained in part if SV40 DNA is present in relatively low abundance in many MPM tissues determined to contain viral sequences and is therefore difficult to detect. To this end, we investigated the efficacy of real time quantitative PCR in detecting low copy number SV40 DNA sequences, then we quantified SV40 copy number in MPM tissues and cell lines at our institution. Quantitative PCR demonstrated consistent precision and accuracy in detecting SV40 DNA sequences over a large logarithmic range of viral copy number. In addition, SV40 sequences were found in 2/35 tissues and 3/7 cell lines in relatively low abundance. We conclude that SV40 is not a contributing factor in the pathobiology of the majority of MPM tumors from patients at our institution.     

Dependence of SV40 large T-antigen cell cycle regulation on T-antigen expression levels.

Simian virus 40 (SV40) large T antigen (Tag) expression results in reduced percentages of G1-phase cells and increased percentages of S- and G2+M-phase cells in exponentially growing fibroblast populations as compared with identical cell populations not expressing Tag. This effect is the result of reduced G1 and increased G2+M cell cycle phase durations caused by Tag [Sladek, T.L. & Jacobberger, J.W. (1992). J. Virol., 66, 1059-1065]. Using recombinant retroviruses to manipulate Tag expression over a 25-fold range, it is shown here that the magnitude of this cell cycle phenotype increases as a function of increasing intracellular Tag concentration. This effect of Tag on the cell cycle is not independent of negative regulation by cellular mechanisms since exponentially growing cell populations producing high and increasing levels of Tag, increase the fraction of cells residing in G1 and decrease the fraction in S and G2+M as a function of cell density. Therefore, the data in this paper show, first, that Tag is a concentration-dependent, positive cell cycle regulator in exponentially proliferating cells and, second, that endogenous cellular mechanisms negatively regulating the cell cycle in response to cell density override the effect of Tag.     

Regulation of SV40 large T-antigen stability by reversible acetylation

Reversible acetylation on protein lysine residues has been shown to regulate the function of both nuclear proteins such as histones and p53 and cytoplasmic proteins such as alpha-tubulin. To identify novel acetylated proteins, we purified several proteins by the affinity to an anti-acetylated-lysine antibody from cells treated with trichostatin A (TSA). Among the proteins identified, here we report acetylation of the SV40 large T antigen (T-Ag). The acetylation site was determined to be lysine-697, which is located adjacent to the C-terminal Cdc4 phospho-degron (CPD). Overexpression of the CBP acetyltransferase acetylated T-Ag, whereas HDAC1, HDAC3 and SIRT1 bound and deacetylated T-Ag. The acetylation and deacetylation occurred independently of p53, a binding partner of T-Ag, but the acetylation was enhanced in the presence of p53. T-Ag in the cells treated with TSA and NA or the acetylation mimic mutant (K697Q) became unstable in COS-7 cells, suggesting that acetylation regulates stability of T-Ag. Indeed, NIH3T3 cells stably expressing K697Q showed decreased anchorage-independent growth compared with those expressing wild type or the K697R mutant. These results demonstrate that acetylation destabilizes T-Ag and regulates the transforming activity of T-Ag in NIH3T3 cells.     

beta-catenin mutations are absent in hepatocellular carcinomas of SV40 T-antigen transgenic mice

beta-catenin mutations have been found not only in melanoma and prostatic carcinoma but also in hepatocellular carcinomas in human, c-myc, H-ras genes transgenic mice and chemically-induced models. We investigated beta-catenin mutations in human hepatocellular carcinomas (HCCs), Hep G2 cell line and HCCs in SV40 T-antigen transgenic mice, in order to examine whether beta-catenin mutations are frequently observed in HCC in general. We found a point mutation of beta-catenin in one of nine HCCs in human and a deletion of it in Hep G2 cell line. However, we found no mutation in HCC in SV40 TG mice liver.     

Decreased methylation rates of DNA in SV40-transformed human fibroblasts

The rates of methylation of total cellular DNA and newly synthesized DNA were measured in four unrelated SV40-transformed human fibroblast lines and in four control parent fibroblast lines. Rates of methylation of total cellular DNA were decreased by a factor of 1.8-2.3 in the transformed cells relative to control cells. Methylation was largely (75%-87%) restricted to newly synthesized DNA in control and transformed fibroblasts, and methylation rates of newly synthesized DNA were diminished in transformed cells by 12- to 19-fold relative to control cells. Growth rates were similar in the normal and transformed cells. The cellular uptake of methionine and conversion to S-adenosylmethionine were similar in the normal and transformed cells, suggesting no major differences between the normal and transformed cells in the cellular transport of methionine, methionine S-adenosyltransferase activity, or the intracellular concentrations of methionine and S-adenosylmethionine. The diminished rates of DNA methylation that we have observed suggest a possible mechanism for altered gene expression and growth control in transformed cells.     

Mitochondrial DNA sequence variation in human leukemic cells

The Long PCR followed by the RFLP technique has been used to search for abnormally structured mitochondrial DNA (mtDNA) and specific sequence differences implicated in the pathogenesis of acute lymphoblastic leukaemia (ALL). We have studied 54 specific sites whose combinations define groups of mtDNA types, in 30 leukemic patients of French Caucasian origin. Results were compared with those in 100 French heathy individuals. Nucleotide substitutions have been defined in 11 patients. This polymorphism is expressed by single base substitution at 6 sites which corresponds to 5 morphs, 2 of which were not found in the reference group. Combining the 11 observed morphs, we have identified 7 different mtDNA types, defined in 30 patients with ALL. Two of the morphs (MspI-2 and AvaII-3) and 3 of the types (17-2, 55-2, New^Fr150) were not found in the group of healthy individuals. We have observed significant statistical changes in type 28-2 in ALL patients compared with the controls. Int. J. Cancer 76:495–498, 1998.© 1998 Wiley-Liss, Inc.     

p53 in SV40-transformed DNA-damaged human cells binds to its cognate sequence but fails to transactivate target genes

Human SV40-transformed cells contain high levels of stabilized p53 of which only a fraction is complexed with the SV40 large tumor antigen (T-antigen). This raises the question whether the p53 which is not complexed with T-antigen retains some biological activity. Two human SV40-transformed cell lines, BEAS and SV80, were investigated. A significant level of constitutive cognate-sequence-specific DNA-binding of p53 was detected by electrophoretic mobility shift assay (EMSA) of cell extracts. Upon DNA damage by treatment with mitomycin C the DNA-binding activity was increased, as known for cells with wild-type p53. However, in both cell lines, before and after DNA damage, p53 was not able to transactivate a target gene as shown by reporter gene assay. Hence, the capability of p53 to bind its cognate sequence is a prerequisite but no proof of p53 transactivating activity. Nuclear p53 levels were not further increased after mitomycin C treatment, occasionally rather slightly decreased, often accompanied by an even larger decrease in amount of T-antigen. In conclusion, SV40-transformation of human cells has caused a loss of essential features of wild-type p53 activity, even in that fraction of p53 not in physical complex with SV40 T-antigen.     

Carcinogen-mediated induction of SV40 DNA synthesis in SV40 transformed Chinese hamster embryo cells

Exposure of SV40-transformed Chinese hamster embryo cells (lineCO50) to a series of physical and chemical carcinogens (includingactivation-dependent and activation-independent varieties) resultedin the induction of viral DNA synthesis. The carcinogen mediatedamplification of SV40 DNA was demonstrated by a highly sensitivein situ hybridization procedure for the detection of cells synthesizingSV40 DNA. Treatment of CO50 cells with an inhibitor of polycyclichydrocarbon metabolism (7,8-benzoflavone) prior to the applicationof benzo[a]pyrene or 7,12-dimethyl-benz[a]anthracene preventedthe induction of SV40 DNA synthesis, indicating that the inductiondepends upon the metabolic activation of these compounds. Non-carcinogenichydrocarbons were inactive under this assay. Two different protocolsfor determining the inducing potential of a compound are presented.The properties of this test and its possible use as a short-termassay for potential carcinogens is discussed. The possibilitythat the induction of SV40 DNA synthesis is a reflection ofa general gene amplification phenomenon mediated by carcinogensis discussed.     

SV40 large T-antigen disturbs the formation of nuclear DNA-repair foci containing MRE11

The accumulation of DNA repair proteins at the sites of DNA damage can be visualized in mutagenized cells at the single cell level as discrete nuclear foci by immunofluorescent staining. Formation of nuclear foci in irradiated human fibroblasts, as detected by antibodies directed against the DNA repair protein MRE11, is significantly disturbed by the presence of the viral oncogene, SV40 large T-antigen. The attenuation of foci formation was found in both T-antigen immortalized cells and in cells transiently expressing T-antigen, indicating that it is not attributable to secondary mutations but to T-antigen expression itself. ATM-mediated nibrin phosphorylation was not altered, thus the disturbance of MRE11 foci formation by T-antigen is independent of this event. The decrease in MRE11 foci was particularly pronounced in T-antigen immortalized cells from the Fanconi anaemia complementation group FA-D2. FA-D2 cells produce essentially no MRE11 DNA repair foci after ionizing irradiation and have a significantly increased cellular radiosensitivity at low radiation doses. The gene mutated in FA-D2 cells, FANCD2, codes for a protein which also locates to nuclear foci and may, therefore, be involved in MRE11 foci formation, at least in T-antigen immortalized cells. This finding possibly links Fanconi anaemia proteins to the frequently reported increased sensitivity of Fanconi anaemia cells to transformation by SV40. From a practical stand point these findings are particularly relevant to the many studies on DNA repair which exploit the advantages of SV40 immortalized cell lines. The interference of T-antigen with DNA repair processes, as demonstrated here, should be borne in mind when interpreting such studies.     

SV40 infection in human cancers.

Simian virus 40 (SV40) is a monkey virus recently detected ina variety of human cancers such as mesothelioma, central nervoussystem (CNS) tumours and non-Hodgkin's lymphoma [1]. Despitea large body of data suggesting that SV40 is implicated in differenttumours in humans, there are still controversies as to whetherSV40 has     

Analysis of proviral integration in human mammary epithelial cell lines immortalized by retroviral infection with a temperature-sensitive SV40 T-antigen construct

A panel of eight conditionally immortal lines derived by infection of human breast epithelial cells with an amphotropic retrovirus transducing a ts mutant of SV40 large T-antigen was analyzed with respect to individual retroviral integration patterns. Each line contained multiple integration sites which were clonal and stable over extended passage. Similar integration patterns were observed between individual lines arising separately from the same stock of pre-immortal cells, suggesting a common progenitor. Retroviral integration analysis of preimmortal cells at different stages of pre-crisis growth showed changes indicative of a progressive transition from polyclonality to clonality as the cells approached crisis. Each of the immortal lines contained a sub-set of the integration sites of their pre-immortal progenitors, with individual combinations and copy numbers of sites. Since all the cell lines appeared to originate from single foci in separate flasks, it is likely that each set arose from a common clone of pre-immortal cells as the result of separate genetic events. There was no evidence from this analysis to suggest that specific integration sites played any part either in the selection of pre-crisis clones or in the subsequent establishment of immortal lines. © 1994 Wiley-Liss, Inc.     

Integration of SV40 at 12q23 in SV40-immortalized human bronchial epithelial cells

In four SV40-immortalized human bronchial epithelial cell linesestablished in our laboratory, we identified the SV40 integrationsites by fluorescence in situ hybridization. We found that inthe late passage of all the four cell lines, SV40 integratedat 12q23. It is possible that only SV40 integration at 12q23is necessary for the immortalization of human bronchial epithelialcells. Some DNA sequences or genes in the region, such as IGF-1,may be involved with a proliferative advantage of the cellswith 12q23 SV40 integration.     

Association of SV40 with human tumors

In 1994, PCR and protein studies suggested that SV40 DNA sequences and proteins were present in 29/48 (60%) USA human mesothelioma samples. Sequence analysis confirmed that the sequences were homologous to SV40. One year later, SV40 was also found in 5/9 human mesotheliomas, and in 1996 SV40 was also reported to be present in 1/3 of the tumor specimens examined. These reports, in combination with an earlier study in 1992 which had detected SV40 in human brain tumors, raised concerns that SV40 was associated with certain types of human tumors, specifically mesothelioma, bone, and brain tumors. These findings raised concerns, because these tumor types are the same malignancies that had been observed in animals injected with SV40. However, a study in 1996 and a presentation made at the International Mesothelioma Interest Group, IMIG in 1997 failed to detect SV40 in mesotheliomas, suggesting the possibility that laboratory artifacts, such as PCR contamination, had caused the previous positive findings. In 1997, the FDA, the NIH, and the CDC organized an international conference in Bethesda to review the literature and to address the possibility that SV40 was present in, and was possibly the cause of, some human tumors. The results of that conference were reported the same year in a meeting review in Oncogene by Carbone and colleagues. Briefly, the consensus was that before accepting the possibility that SV40 was present in human tumors, a multi-laboratory study needed to be conducted. It was recommended that a blinded multi-laboratory study be directed by an independent scientist not previously associated with the controversial reports of SV40 in human specimens. It was also recommended that this study include laboratories that had reported positive findings as well as laboratories that had failed to detect SV40 in human specimens. Since 1997, about 30 independent reports have been published on this topic, including the multi-laboratory study. Evidence in favor and against a possible association of SV40 with human cancer was reviewed at an international consensus meeting at the University of Chicago on 20, 21 April 2001, entitled "Malignant Mesothelioma: Therapeutic Options and the Role of SV40, 2001". The main focus was the association of SV40 with mesothelioma and other human tumors. At the end of the meeting, a panel discussion, which included independent experts who had not published on this topic, critically reviewed the evidence presented at the meeting. The results of the meeting and of the final panel discussion are outlined below.     

Association of SV40 with human tumours

SV40 was discovered as a contaminant of poliovirus vaccines that were inadvertently administered to millions of people in Europe and the United States between 1955 and 1963. Shortly afterwards, SV40 was proven to be oncogenic in rodents and capable of transforming human and animal cells in vitro. The possibility that SV40 might cause tumours in humans thus became a subject of scientific and public interest and scrutiny. However, largely due to a lack of significant epidemiological evidence, interest in assessing SV40's potential carcinogenic role in humans diminished. Recently, many laboratories have reported the presence of SV40-like DNA in a high proportion of human mesotheliomas, ependymomas and osteosarcoma (the three main types of tumours caused by virus in hamsters), renewing the question whether SV40 might be a human tumour virus. Molecular data from these studies are reviewed to re-evaluate the potential role of SV40 as a human carcinogen.