Genetic analysis of tumorigenesis. VIII. Suppression of SV40 transformation in cell hybrids and cytoplasmic transferants

Intraspecies somatic cell hybrids of BALB/c mouse 3T3 and SV40-transformed embryonic fibroblast (SVT2) cells were analyzed for transformation-associated properties and their tumorigenic potential in nude mice. In confirmation of our earlier findings, hybrids expressing the viral T-antigen were not suppressed for the ability to clone in medium with 1% serum. In contrast, division rate in medium with 1% or 10% serum, anchorage independence, cytochalasin-sensitive growth control, and tumorigenicity were suppressed noncoordinately, and the extent of suppression varied from one hybrid to another. Suppression was not simply determined by the increased chromosome content of the hybrid cells, nor was suppression correlated with rearrangements of the integrated viral sequence (SAGER et al., 1981a, b). Similar results were found in cytoplasmic transferants expressing T-antigen. Four independent transferants and subclones derived from them varied in the extent of suppression of anchorage independence and tumorigenicity. In both hybrids and transferants, a low serum requirement for clonal growth apparently was determined solely by expression of SV40 T-antigen, but other transformation properties, as well as tumorigenicity, appeared to require multiple changes in the cellular genome for their expression. These changes must occur during or after viral integration, since they are not expressed in uninfected 3T3 cells.     

Contact-Inhibited Revertant Cell Lines Isolated from Simian Virus 40-Transformed Cells III. Concanavalin A-Selected Revertant Cells

Contact-inhibited variants have been isolated by treatment of simian virus 40 (SV40)-transformed Balb/c 3T3 cells (SVT2) with the plant lectin concanavalin A. These con A revertant cells exhibit the following properties: (i) they resemble 3T3 cells morphologically and grow to saturation densities which are similar to that of 3T3 cells; (ii) they synthesize the SV40-specific T antigen and yield infectious virus after fusion with permissive monkey cells; (iii) they contain a high sialic acid content similar to that of 3T3 cells and not to that of SVT2 cells; sialic acid composition was found to be independent of serum concentration; (iv) they contain more chromosomes with the average number in the tetraploid range than the SVT2 cells from which they were derived; and (v) SVT2 and revertant cells, confluent or subconfluent, produce more collagen than Balb/3T3 cells. The relationship of surface membrane properties to contact inhibition of growth and the mechanisms for generating revertant cells are discussed.     

Binding of p53 and p105-RB is not sufficient for oncogenic transformation by a hybrid polyomavirus-simian virus 40 large T antigen.

To identify regions on the large T antigens of simian virus 40 (SV40) and polyomavirus which are involved in oncogenic transformation, we constructed plasmids encoding hybrid polyomavirus-SV40 large T antigens. The hybrid T antigens were expressed in G418 sulfate-resistant pools of rat F2408 cells, and extracts of such pools were immunoprecipitated with an antibody against p53. Two hybrid T antigens containing SV40 amino acids 337 to 708 bound to p53, whereas another hybrid T antigen containing SV40 amino acids 412 to 708 did not. This suggests that a binding domain on SV40 large T antigen for p53 is contained within amino acids 337 to 708, with amino acids 337 to 411 playing an important role. One of the two hybrids that bound to p53 was chosen for further study. This T antigen contained SV40 large T antigen amino acids 336 to 708 joined to polyomavirus large T antigen amino acids 1 to 521 (PyT1-521-SVT336-708). Immunoprecipitation with antibodies directed against the product of the retinoblastoma susceptibility gene, p105-RB, showed that this hybrid bound p105-RB as well as p53. Pools expressing the hybrid PyT1-521-SVT336-708 did not grow in soft agar, nor did they form foci on confluent monolayers of nontransformed F2408 cells. The hybrid T antigen was expressed at levels comparable to those seen in retrovirus-infected F2408 cells expressing only SV40 large T antigen, which do show a transformed phenotype. Thus, this level of expression was sufficient for transformation by SV40 large T antigen but not for the hybrid large T antigen. These data, combined with genetic studies from other laboratories, suggest that complex formation with p53 and p105-RB is necessary but not sufficient for the oncogenic potential of papovavirus large T antigens.     

Expression of malignancy traits in the interspecific somatic hybrids of tumor and normal cells

Tumorigenicity and anchorage independence in two types of the interspecies hybrids of the tumor and normal mammalian cells were studied. One hybrid type was derived from fusion of spontaneously transformed Chinese hamster and normal mouse cells; the second type was obtained by fusion of SV40-transformed Djungarian hamster and the same mouse cells. The tumorigenicity in the athymic nude mice was suppressed in the first type of hybrids. The hybrid clones derived from fusion of SV40-transformed and normal cells could form tumor in nude mice. Testing of hybrid clones for their ability to form colonies in soft agar showed that all hybrids grew well in the medium, similar to tumor parental cells. These data suggest that malignancy and anchorage independence are under separate genetic control. The influence of the origin of the tumor parental cells (spontaneous or SV40-virus transformation) on the expression of the malignancy in hybrids of the tumor and normal cells is discussed.     

Transformation of Simian Virus 40-resistant Hamster Cells with an Adenovirus 7-Simian Virus 40 Hybrid

When the hamster cell lines BHK21 and Nil-2 were infected at a multiplicity of 100 with the adenovirus 7-simian virus 40 (SV40) hybrid (strain LLE46), SV40 T antigen was induced in 0.1 to 6% of the cells during the first 96 hr postinfection, morphological changes occurred 3 to 7 weeks later, and eventually all the cells contained SV40 T antigen, but no adeno 7 T antigen. Results were similar when primary and secondary monolayer cultures of hamster embryo (HE) cells were infected with the adeno 7-SV40 hybrid, and when primary HE cells were infected with SV40. However, infection of BHK21, Nil-2, and secondary HE cells with the same multiplicity of SV40 did not induce SV40 T antigen or morphological transformation. This suggests that the target cells required for infection with SV40 virions, but not those required for infection with the hybrid, are lost or altered in secondary HE cultures and in the two cell lines. In most of the virus-host cell systems in which SV40 T antigen and transformation were induced, there was a decrease in the number of T antigen-positive cells after the initial infection. This was followed by a lag period of up to 2 months before the onset of a progressive increase in the number of positive cells. The beginning of the rise in T antigen production coincided with the first morphological changes.     

Studies of Nondefective Adenovirus 2-Simian Virus 40 Hybrid Viruses VII. Characterization of the Simian Virus 40 RNA Species Induced by Five Nondefective Hybrid Viruses

Five nondefective adenovirus 2 (Ad2)-simian virus 40 (SV40) hybrid viruses have been isolated and found to contain segments of SV40 DNA covalently linked to Ad2 DNA. The quantity of SV40 DNA present is a stable characteristic of each hybrid virus, and varies from less than 5% (in Ad2(+)ND(3)) to more than 30% (in Ad2(+)ND(4)) of the SV40 genome. We have characterized the SV40 portions of these hybrids by relating the SV40-specific RNA sequences transcribed in cells infected with each hybrid virus to those transcribed in cells infected with each of the other hybrid viruses and with SV40 itself. RNA-DNA hybridization-competition experiments indicate that the number of unique SV40 RNA sequences transcribed in infected cells is proportional to the size of the SV40 DNA segment contained within each hybrid and, in the case of the three hybrids which induce detectable SV40-specific antigens, to the number of SV40 antigens induced. Furthermore, the SV40-specific RNA sequences transcribed from any one of the hybrids are completely represented in the RNA transcribed from all other hybrids with longer SV40 segments. Thus, the SV40 DNA regions in the five hybrid viruses appear to contain some nucleotide sequences in common. The SV40-specific RNA transcribed from Ad2(+)ND(4), the hybrid containing the largest SV40 segment, is qualitatively similar to the SV40-specific RNA transcribed early (i.e., prior to viral DNA replication) in SV40 lytic infection. Thus, it appears that no significant amount of late SV40 DNA is transcribed during infection by any of the five nondefective Ad2-SV40 hybrid viruses.     

The level of expression of adenovirus type 2 transforming genes governs sensitivity to nonspecific immune cytolysis and other phenotypic properties of adenovirus 2-simian virus 40-transformed cell hybrids.

Syrian hamster embryo cells transformed by adenovirus type 2 (Ad2) or simian virus 40 (SV40) differ markedly in morphology, tumorigenicity, and susceptibility to in vitro lysis by nonspecific cytotoxic cells. Hybrid cells formed by fusing Ad2- and SV40-transformed Syrian hamster embryo cells may express only SV40 T antigens or both SV40 and Ad2 T antigens. Hybrids that express only SV40 T antigens are indistinguishable from the nonhybrid SV40-transformed phenotype, whereas hybrid cells that express T antigens from both viruses closely resemble the nonhybrid parental Ad2-transformed phenotype. Because these hybrid cells have been useful in the study of neoplastic transformation, we determined the amount of viral antigens that they accumulate in an attempt to correlate the level of expression of the transforming viral genes with some of their phenotypic properties. Hybrid cells that expressed proteins from both viruses showed reduced levels of SV40 T antigens compared with those of hybrid cells that did not express Ad2 T antigens. We also found that the production of several cellular proteins that influence cytomorphology was inhibited in hybrid and nonhybrid cells that expressed Ad2 T antigens, and the repression of these cellular proteins correlated with a change in cytomorphology from fibroblastic to spherical. Finally, we showed that the susceptibility of our hybrid cells to in vitro lysis by natural killer cells and activated macrophages, two putative host-effector cells involved in defense against neoplasia, correlated closely with the level of expression of a 58,000-dalton Ad2 protein. The results reported here, together with the results of previous studies, indicate that the oncogenic potential of hybrid cells that express both Ad2 and SV40 antigens is extremely sensitive to Ad2 expression, whereas other phenotypic properties depend on Ad2 expression in a dose-dependent manner.     

Simian virus 40 DNA replication, transcription, and antigen induction during infection with two adenovirus 2-SV40 hybrids that contain the entire SV40 genome.

The Ad2++hey hybrid virus population produces simian virus 40 (SV40) efficiently during lytic infection, whereas Ad2++ley does not, although both hybrids contain a complete SV40 genome. In this report, we demonstrate the synthesis of nonhydrid SV40 DNA in Ad2++HEY-infected Vero cells, but only early SV40 RNA is transcribed efficiently in Ad2++LEY-infected cells. Ad2++HEY induces SV40 U, T, and V antigens during lytic infection of African green monkey kidney cells, whereas Ad2++LEY induces only SV40 U and T antigens. These variations in the behavior of Ad2++HEY and Ad2++LEY regarding expression of SV40 functions probably reflect differences in the rate of SV40 excision from the hybrid genomes.     

Simian virus 40 T- and U-antigens: immunological characterization and localization in different nuclear subfractions of simian virus 40-transformed cells.

Simian virus 40 (SV40)-transformed cells and cells infected by the nondefective adenovirus 2(Ad2)-SV40 hybrid viruses Ad2+ND1 and Ad2+ND2 were analyzed for SV40 T- and U-antigens, respectively, using individual hamster SV40 tumor sera or serum for which U-antibodies were removd by absorption. These studies showed that (i) T- and U-antigens can be defined by separate classes of antigenic determinants and (ii) the U-antigenic determinants in SV40-transformed cells and in hybrid virus-infected cells are similar. The apparent discrepancy in the subcellular location of U-antigen in SV40-transformed cells (nuclear location) and in hybrid virus-infected cells (perinuclear location) as determined by immunofluorescence staining of methanol/acetone-fixed cells could be resolved by treating hybrid virus-infected cells with a hypotonic KCl solution before fixation. Upon this treatment hybrid virus-infected cells also showed nuclear U-antigen staining. The possibility of an association of T- and U-antigens with different nuclear subfractions in SV40-transformed cells was investigated. Detergent-cleaned nuclei of SV40-transformed cells were fractionated into nuclear matrices and a DNase-treated, high-salt nuclear extract. Analysis of the nuclear matrices by immunofluorescence microscopy with T+U+ and T+U- hamster SV40 tumor serum revealed that U-antigen remained associated with the nuclear matrices, whereas T-antigen could not be detected in this nuclear subfraction. T-antigen, however, could be immunoprecipitated from nuclear extracts of the SV40-transformed cells.     

Characterization of a 54K dalton cellular SV40 tumor antigen present in SV40-transformed cells and uninfected embryonal carcinoma cells.

SV40 infection or transformation of murine cells stimulated the production of a 54K dalton protein that was specifically immunoprecipitated, along with SV40 large T and small t antigens, with sera from mice or hamsters bearing SV40-induced tumors. The same SV40 anti-T sera immunoprecipitated a 54K dalton protein from two different, uninfected murine embryonal carcinoma cell lines. These 54K proteins from SV40-transformed mouse cells and the uninfected embryonal carcinomas cells had identical partial peptide maps which were completely different from the partial peptide map of SV40 large T antigen. An Ad2+ND4-transformed hamster cell line also expressed a 54K protein that was specifically immunoprecipitated by SV40 T sera. The partial peptide maps of the mouse and hamster 54K protein were different, showing the host cell species specificity of these proteins. The 54K hamster protein was also unrelated to the Ad2+ND4 SV40 T antigen. Analogous proteins immunoprecipitated by SV40 T sera, ranging in molecular weight from 44K to 60K, were detected in human and monkey SV40-infected or -transformed cells. A wide variety of sera from hamsters and mice bearing SV40-induced tumors immunoprecipitated the 54K protein of SV40-transformed cells and murine embryonal carcinoma cells. Antibody produced by somatic cell hybrids between a B cell and a myeloma cell (hybridoma) against SV40 large T antigen also immunoprecipitated the 54K protein in virus-infected and -transformed cells, but did not do so in the embryonal carcinoma cell lines. We conclude that SV40 infection or transformation of mouse cells stimulates the synthesis or enhances the stability of a 54K protein. This protein appears to be associated with SV40 T antigen in SV40-infected and -transformed cells, and is co-immunoprecipitated by hybridomas sera to SV40 large T antigen. The 54K protein either shares antigenic determinants with SV40 T antigen or is itself immunogenic when in association with SV40 large T antigen. The protein varies with host cell species, and analogous proteins were observed in hamster, monkey and human cells. The role of this protein in transformation is unclear at present.     

Properties of Somatic Cell Hybrids Between Mouse Cells and Simian Virus 40-Transformed Rat Cells

Hybrids between mouse cells and simian virus 40 (SV40)-transformed rat cells were made, and their properties and chromosome constitution were investigated over many generations. Their hybrid nature was confirmed by enzyme studies. During a period of 1 year a loss of 10 to 20% of the total number of chromosomes was observed. The SV40 tumor antigen was present and remained present in the hybrids. The parental and hybrid cells were studied for agglutination with concanavalin A, for growth in soft agar, and for serum requirement. These growth and surface characteristics of the transformed cells appeared in the hybrids.     

Simian virus 40 large T antigen oligomers: analysis of electrophoresis in the absence of detergent.

Large T antigen of simian virus 40 is found as monomeric and oligomeric species in transformed cells. These can be demonstrated in cell extracts by velocity centrifugation in sucrose gradients. We analyzed them further in a transformed human line cell (SV80) and a transformed mouse line cell (SVT2). Individual fractions from sucrose gradients were subjected to polyacrylamide gel electrophoresis in the absence of detergent. T-antigen species were then detected by protein blotting and antibody overlay with polyclonal anti-D2 T antibody or monoclonal Pab419, Pab101, or Pb1700 antibody. The rapidly sedimenting species (14S and larger) of large T antigen from both cell lines reproducibly showed two major bands with estimated molecular weights of 670,000 and 850,000. A third band of 1,200,000 was more prominent in SVT2 cells than in SV80 cells. In SV80 cells the slowly sedimenting species of large T antigen (5S to 11S) contained two reproducible bands. A band with a molecular weight of 95,000 was the predominant one in all fractions between 5S and 11S. A relatively minor band with a molecular weight of 230,000 was found in fractions between 9S and 11S. The low-molecular-weight forms were seen in SVT2 cells only when a prominent peak at 5S to 7S was present, that is, when extracts were stored before analysis. In fresh extracts, the low-molecular-weight bands and slowly sedimenting forms were absent.     

Structure of two adenovirus-simian virus 40 hybrids which contain the entire SV40 genome

Two defective adenovirus-simian virus 40 hybrids which contain the entire SV40 genome (Ad2⁺⁺HEY and Ad2⁺⁺LEY)² have been isolated. Upon infection of cells permissive for SV40 both hybrids give rise to infectious SV40 virions, but with markedly different efficiencies. In the case of Ad2⁺⁺HEY nearly all cells infected with a hybrid particle yield SV40 progeny, whereas in the case of Ad2⁺⁺LEY infectious SV40 is produced in only about one in 10⁴ cells infected with hybrid particles. The structures of the DNA molecules in the Ad2⁺⁺HEY and Ad2⁺⁺LEY populations were examined using electron microscope heteroduplex methods. Both populations were found to be heterogeneous. Ad2⁺⁺HEY contained three hybrids (HEY-I, HEY-II, and HEY-III) whose genomes differed only in their content of SV40 DNA (0.45 ± 0.02, 1.43 ± 0.04, and 2.39 ± 0.09 SV40 genomes, respectively). Ad2⁺⁺LEY contained two hybrids (LEY-I and LEY-II), which also differed only in their content of SV40 DNA (0.03 ± 0.01 and 1.05 ± 0.01 SV40 genomes, respectively). In those hybrids which contained more than one complete SV40 genome (HEY-II, HEY-III, LEY-II) the excess SV40 DNA was shown to be organized as a tandem repetition. These data suggest that the various hybrid genomes within each population are interconvertible by recombination events, which insert or excise an SV40 genome. It is proposed that HEY-II and HEY-III yield infectious SV40 with higher efficiency than LEY-II because their SV40 DNA segments contain longer tandem repetitions; thus, the probability of an intramolecular recombination event which results in excision of an SV40 genome is greater.     

Transformation of a continuous rat embryo fibroblast cell line requires three separate domains of simian virus 40 large T antigen.

Mouse C3H 10T1/2 cells and the established rat embryo fibroblast cell line REF-52 are two cell lines widely used in studies of viral transformation. Studies have shown that transformation of 10T1/2 cells requires only the amino-terminal 121 amino acids of simian virus 40 (SV40) large T antigen, while transformation of REF-52 cells requires considerably more of large T antigen, extending from near the N terminus to beyond residue 600. The ability of a large set of linker insertion, small deletion, and point mutants of SV40 T antigen to transform these two cell lines and to bind p105Rb was determined. Transformation of 10T1/2 cells was greatly reduced by mutations within the first exon of the gene for large T antigen but was only modestly affected by mutations affecting the p105Rb binding site or the p53 binding region. All mutants defective for transformation of 10T1/2 cells were also defective for transformation of REF-52 cells. In addition, mutants whose T antigens had alterations in the Rb binding site showed a substantial reduction in transformation of REF-52 cells, and the degree of this reduction could be correlated with the ability of the mutant T antigens to bind p105Rb. There was a tight correlation between the ability of mutants to transform REF-52 cells and the ability of their T antigens to bind p53. These results demonstrate that multiple regions of large T antigen are required for full transformation by SV40.     

Use of Nondefective Adenovirus-Simian Virus 40 Hybrids for Mapping the Simian Virus 40 Genome

A series of viable recombinants between adenovirus 2 (Ad2) and simian virus 40 (SV40) (nondefective Ad2-SV40 hybrids) have been isolated. The members of this series (designated Ad2(+)ND(1) through Ad2(+)ND(5)) differ from one another in the early SV40-specific antigens and the SV40-specific RNA species which they induce in infected cells. They also contain different amounts of SV40 DNA as shown by RNA-DNA hybridization techniques. We have examined the structure of the DNA molecules from these hybrids, using electron microscope heteroduplex mapping techniques. Each hybrid was found to contain a single segment of SV40 DNA of characteristic size covalently inserted at a unique location in the adenovirus 2 DNA molecule. The SV40 segments of the various hybrids formed an overlapping series with a common end point. When the results of the electron microscopic study were combined with data on antigen induction, it was found that a self-consistent map could be constructed which related specific regions of the SV40 genome to the induction of specific antigens. The order of these early SV40 antigen inducing regions in the SV40 DNA segments contained in the nondefective hybrids is: U antigen, tumor specific transplantation antigen, and T antigen with the U antigen region being nearest the common end point.     

Expression of the SV40-transformed phenotype in hybrids between conditional and wild-type transformants

The nature of two different SV40-transformed cell lines which are temperature sensitive (ts) for the expression of the transformed phenotype has been analysed by somatic cell hybridization. Ts 23A cells are temperature sensitive in their ability to grow in medium containing low serum; in hybrids between these cells and a standard SV40-transformed BALB/c 3T3 cell line, the mutation seems to be dominant, at least for some parameters. One hundred percent of the hybrid clones have a very low efficiency of plating in medium containing 1% serum, but only 50% of them show growth inhibition in mass culture under the same conditions. Ts H615 cells revert to a normal phenotype at 39 °C for most parameters of transformation. The mutation appears to be recessive, as the hybrids generally behave like transformed cells at 39 °C. However, the expression of this mutation is not completely suppressed, since DNA synthesis after confluence shows some degree of inhibition in the hybrids at 39 °C. A considerable variation in the behavior of individual hybrid clones and in the degree of inhibition of some parameters has been observed and is discussed in terms of possible mechanism(s) of transformation.     

Contact-inhibited revertant cell lines isolated from SV40-transformed cells. V. Contact inhibition of sugar transport

The transport of 2-deoxyglucose in BALB/c 3T3 cells, Simian virus 40-transformed BALB/c 3T3 (SVT2) cells, and concanavalin A-selected revertant cells of SVT2 has been measured. Sparsely-seeded BALB/c 3T3 cells transport the sugar at about one-fourth, and sparsely-seeded revertant cells at three-fourths, the rate of SVT2 cells. BALB/c 3T3 cells undergo a dramatic drop in sugar uptake at confluency, transporting sugar at about one-tenth the rate of subconfluent cells. Revertant cells (contact-inhibited variants of transformed cells) are similar in this respect, but the drop is only 5-fold. SVT2 cells show no such change in uptake over wide cell densities. Subconfluent BALB/c 3T3, SVT2, and revertant cells have similar K_m and V_max values for 2-deoxyglucose transport; however, confluent 3T3 and confluent revertant cells show a large increase in K_m and a 5-fold decrease in V_max as compared to their subconfluent counterparts or SVT2 cells—indications of a decreased number of transport sites and a decreased affinity of these sites for sugar when these cells make intimate contacts with each other. These data indicate that extensive changes in the architecture of the cell surface occur when contactinhibited cells are in close apposition with each other, regardless of the persistence of partially expressed SV40 genetic information, and are discussed with regard to the membrane compositions of these cell lines.     

Pyridine nucleotide levels as a function of growth in normal and transformed 3T3 cells.

The levels of NAD (NAD⁺ + NADH) and NADP (NADP⁺ + NADPH) and their redox states were measured as a function of growth in 3T3 mouse fibroblasts which exhibit density-dependent inhibition of growth and SV40 (simian virus #40)-transformed 3T3 cells (SVT2) which have lost this property. The levels were related to cell numbers, protein content, and rates of DNA synthesis. At corresponding cell densities, 3T3 cells contain approximately twice as much total protein as SVT2 cells. The levels of NAD relative to total cellular protein are density dependent in both 3T3 and SVT2, increasing with increasing cell density. Over a 30-fold range of cell densities, the NAD levels in 3T3 increase 2.4-fold, while the levels in SVT2 increase 1.6-fold. The levels of NAD are very similar in dividing 3T3 and SVT2 cells at corresponding cell densities; however, a marked increase in the levels of NAD is observed in 3T3 cells, but not in SVT2 cells, at cell densities just prior to where 3T3 cells enter density-dependent inhibition of growth. This increase in NAD levels is correlated with the cessation of DNA synthesis. The NAD pools are 15–25% NADH for 3T3 and 5–15% NADH for SVT2. NADP levels relative to protein in 3T3 and SVT2 are less density dependent, with overall increases of 1.3- and 1.2-fold, respectively, observed over the range of cell densities examined. NADP levels relative to protein are nearly twice as high in SVT2 cells as in 3T3 cells of corresponding cell densities. The NADP pools are approximately 70–80% NADPH in both cell types.