Transforming genes and gene products of polyoma and SV40

Lessons in signaling and tumorigenesis from polyomavirus middle T antigen

The small DNA tumor viruses have provided a very long-lived source of insights into many aspects of the life cycle of eukaryotic cells. In recent years, the emphasis has been on cancer-related signaling. Here we review murine polyomavirus middle T antigen, its mechanisms, and its downstream pathways of transformation. We concentrate on the MMTV-PyMT transgenic mouse, one of the most studied models of breast cancer, which permits the examination of in situ tumor progression from hyperplasia to metastasis.

Noninvasive imaging of protein-protein interactions in living animals

Protein-protein interactions control transcription, cell division, and cell proliferation as well as mediate signal transduction, oncogenic transformation, and regulation of cell death. Although a variety of methods have been used to investigate protein interactions in vitro and in cultured cells, none can analyze these interactions in intact, living animals. To enable noninvasive molecular imaging of protein-protein interactions in vivo by positron-emission tomography and fluorescence imaging, we engineered a fusion reporter gene comprising a mutant herpes simplex virus 1 thymidine kinase and green fluorescent protein for readout of a tetracycline-inducible, two-hybrid system in vivo. By using micro-positron-emission tomography, interactions between p53 tumor suppressor and the large T antigen of simian virus 40 were visualized in tumor xenografts of HeLa cells stably transfected with the imaging constructs. Imaging protein-binding partners in vivo will enable functional proteomics in whole animals and provide a tool for screening compounds targeted to specific protein-protein interactions in living animals.

Transformation by ras suppresses expression of the neurotrophic growth factor pleiotrophin

An 18-kDa protein (p18) was detected in lysates and conditioned medium from contact-arrested NIH 3T3 fibroblasts, but was not detected when the cells were transformed by the oncogene ras. Analysis of transformation-defective cell clones generated after mutagenesis of the ras-retroviral vector used to transduce the ras gene showed an inverse correlation between p18 expression and the degree of transformation. p18 expression was high in non-transformed clones, intermediate in a partially transformed clone, undetectable in fully transformed clones, and detectable only at the non-permissive temperature in a clone which was cold-sensitive for ras transformation. In non-transformed cells, p18 expression varied with the degree of confluence. It was almost undetectable in medium from sparse, proliferating cells, but increased as the cells approached confluence and peaked 2-4 days after confluence. Microsequencing of partially purified p18 identified it as the developmentally regulated neurotrophic factor pleiotrophin. In further experiments, pleiotrophin was undetectable or almost undetectable in medium from fully transformed cells expressing the oncogenes v-src, truncated c-raf, activated c-fms, or polyomavirus middle tumor antigen; it was low but easily detectable in medium from SV40 large tumor antigen-expressing cells, which form soft agar colonies but not foci. Thus, pleiotrophin expression in NIH 3T3 cells is associated with quiescence, and suppression of pleiotrophin is related to oncogenic transformation.

Cell-type specific activation of the polyomavirus F9-1 regulatory region in transgenic mice

To characterize the activity of the polyomavirus regulatory region, two hybrid marker genes were constructed. In the first construct, the early promoter regulates expression of the CAT gene and the late promoter regulates expression of the lacZ gene. In the second construct, the lacZ gene was placed under the control of the early promoter. The fusion constructs were introduced into the mouse germline. Gene expression was analysed in the generated transgenic mice. A pronounced cell-type specific activation of the transcriptional control region was found in different tissues of the developing embryo and in the adult animal. The control region is recognized and activated in early preimplantation embryos. Around the time of implantation, sequential activation of the Py regulatory region was first observed in differentiating cells. Stage- and tissue-specific expression were noted later in embryonic development. Comparing reporter gene expression on the single-cell level, the different viral promoters display identical expression patterns throughout ontogenesis. Quantitative analysis revealed that marker gene expression from the late promoter was significantly higher than from the early promoter. Furthermore, the cell-type specificity of the control region is not altered in the presence of its regulatory protein, the LT.

Genetic analysis of polyomavirus large T nuclear localization: nuclear localization is required for productive association with pRb family members

Polyomavirus large T antigen (LT) is a multifunctional nuclear protein. LT has two nuclear localization signals (NLS2), one spanning residues 189 to 195 (NLS1) and another spanning residues 280 to 286 (NLS2). Site-directed mutagenesis showed that each signal contains at least two critical residues. The possibility of connections between NLSs and adjacent phosphorylations has attracted much attention. Cytoplasmic LT (CyT) mutants were underphosphorylated, particularly at sites adjacent to NLS2. However, since a nuclear LT bearing an inactivated NLS2 was phosphorylated normally at adjacent sites, the signal was not directly required for phosphorylation. Conversely, LT could be translocated to the nucleus via NLS2 even when the adjacent phosphorylation sites were deleted. CyT was examined to probe the importance of LT localization. CyT was unable to perform LT functions related to interactions with retinoblastoma susceptibility gene (pRb) family members. Hence, CyT was unable to immortalize primary cells or to transactivate an E2F-responsive promoter. Consistent with these findings, CyT, though capable of binding pRb in vitro, did not cause relocalization of pRb in cells. Assays of transactivation of the simian virus 40 late promoter and of the human c-fos promoter showed that defects of CyT were not limited to functions dependent on pRb interactions.

The third subunit of protein phosphatase 2A (PP2A), a 55-kilodalton protein which is apparently substituted for by T antigens in complexes with the 36- and 63-kilodalton PP2A subunits, bears little resemblance to T antigens

The small and middle T (tumor) antigens of polyomavirus have been shown previously to associate with the 36-kDa catalytic subunit and the 63-kDa regulatory subunit of protein phosphatase type 2A, apparently substituting for a normal third 55-kDa regulatory subunit (D.C. Pallas, L.K. Shahrik, B.L. Martin, S. Jaspers, T.B. Miller, D.L. Brautigan, and T.M. Roberts, Cell 60:167-176, 1990). To facilitate a comparison of the normal regulatory subunit and T antigens, we isolated a 2.14-kb cDNA clone encoding this 55-kDa subunit from a rat liver library. Using a probe from the coding region of this gene, we detected a major 2.4-kb mRNA transcript in liver and muscle RNAs. The 55-kDa protein phosphatase 2A subunit purified from rat skeletal muscle generates multiple species when analyzed on two-dimensional gels. Transcription and translation of the clone in vitro produced a full-length protein that comigrated precisely on two-dimensional gels with three of these species, indicating that the 55-kDa protein is apparently modified similarly in vivo and in reticulocyte lysates. Additional species in the purified preparation were not found in the translate, suggesting that there are probably two or more isoforms of this protein in rat muscle. Somewhat surprisingly, there was no clear homology with T-antigen amino acid sequences.

Polyomavirus tumor induction in mice: effects of polymorphisms of VP1 and large T antigen

By testing recombinants between "high tumor" (inducing a high incidence of tumors) and "low tumor" (inducing a low incidence of tumors) strains of polyomavirus, we have previously shown that the key determinant(s) for induction of a high tumor profile resides in coding regions of the high tumor strain (R. Freund, G. Mandel, G. G. Carmichael, J. P. Barncastle, C. J. Dawe, and T. L. Benjamin, J. Virol. 61:2232-2239, 1987). Three single-amino-acid differences between the PTA (high tumor) and RA (low tumor) virus strains have now been identified by DNA sequencing, one each in the large T antigen, in the region common to the middle and small T antigens, and in the major capsid protein VP1. Further tests of appropriate recombinants and oligonucleotide-induced mutants show that VP1 of PTA is the major determinant for induction of a high tumor profile, including all tumors of epithelial origin. The differential effect of the VP1s of PTA and RA on the tumor profile is discussed in terms of a likely contribution of the polymorphic region of VP1 to binding of receptors and infection of different cell types in the animal. The polymorphism in the large T antigen has a more restricted action, which is seen only when tested in virus carrying the VP1 type of PTA; the PTA large T antigen then promotes more rapid growth of tumors of salivary gland and thymus than the RA large T antigen.

A viable mouse polyomavirus mutant without immortalizing or transforming activities

The polyomavirus mutant, dl1041, has a 375-base pair deletion. It removes most of the sequences that are unique to rodent polyomaviruses and encodes part of the large and middle T-antigens. The mutant was conditionally viable, although both the immortalizing and transforming functions of the T-antigens produced by this mutant were found to be defective. However, the dl1041 mutant was found to be capable of DNA replication in rapidly growing mouse 3T6 cells. In contrast, dl1041 DNA synthesis could not be detected in serum-deprived mouse 3T3 cells. In these cells, the low efficiency of dl1041 DNA replication could be attributed to deficiencies in both large and middle T-antigen, suggesting a link between the mitogenic and oncogenic activities of these proteins. Transfection of growing mouse 3T6 cells with dl1041 DNA resulted in the formation of infectious virus, demonstrating that the dl1041 mutant is able to complete an infection cycle. The ability to activate the viral late promoter in trans was retained by the dl1041 mutant large T-antigen, suggesting that immortalization and trans-activation of the late promoter represent two distinct activities of the protein. An essential element of the immortalizing activity in the large T-antigen polypeptide chain appeared to be in a segment consisting of amino acid residues 136-184, since the dl1041 deletion abolished the activity and the 184 amino acid residue N-terminal dl1354 fragment of large T-antigen retained the activity.

The amino terminus of polyomavirus middle T antigen is required for transformation

In polyomavirus-transformed cells, pp60c-src is activated by association with polyomavirus middle T antigen. These complexes have a higher tyrosine kinase activity compared with that of unassociated pp60c-src. Genetic analyses have revealed that the carboxy-terminal 15 amino acids of pp60c-src and the amino-terminal half of middle T antigen are required for this association and consequent activation of the tyrosine kinase. To define in greater detail the borders of the domain in middle T antigen required for activation of pp60c-src, we constructed a set of unidirectional amino-terminal deletion mutants of middle T antigen. Analysis of these mutants revealed that the first six amino acids of middle T antigen are required for it to activate the kinase activity of pp60c-src and to transform Rat-1 fibroblasts. Analysis of a series of insertion and substitution mutants confirmed these observations and further revealed that mutations affecting the first four amino acids of middle T antigen reduced or abolished its capacity to activate the kinase activity of pp60c-src and to transform Rat-1 cells in culture. Our results suggest that the first four amino acids of middle T antigen constitute part of a domain required for activation of the pp60c-src tyrosyl kinase activity and for consequent cellular transformation.

Differentiation of simian virus 40 transformed human mammary epithelial stem cell lines to myoepithelial-like cells is associated with increased expression of viral large T antigen

Cloned simian virus 40 (SV40)-transformed human breast epithelial cell lines can differentiate to myoepithelial-like cells, and these can be isolated as clonal cell lines. Immunofluorescent and immunocytochemical analysis of such cell lines growing on plastic surfaces, collagen gels, and as tumor-nodules in nude mice indicate that all the cell lines produce SV40 large T antigen, but that the production of this antigen is qualitatively increased in the myoepithelial-like cells and cell lines. The myoepithelial-like cell lines produce 4-6 times more immunoprecipitable large T antigen than the parental epithelial cells. The amount of mRNA for large T antigen is also increased by 3.5-5-fold in the myoepithelial-like cell lines when analysed by dot-blot or by Northern hybridisations. Thus, differentiation along the myoepithelial-like cell pathway is associated in these SV40-transformed cells with increased expression of the viral large T antigen. It is suggested that immortalization of primary breast epithelial cell cultures may be, in part, due to the expression of large T antigen preventing processes of terminal keratinization.

Complete transformation of embryonal rat fibroblasts by polyomavirus occurs during passage in vitro

The tumorigenicity of secondary rat embryo fibroblasts transfected with a plasmid harboring a replication origin-defective polyomavirus was found to increase during in vitro propagation. Thus, polyomavirus-transfected cells were found to be more than 10,000-fold more tumorigenic when injected into syngenic rats at 3 months after transfection compared to those injected at an earlier time point. Furthermore, most clones of polyomavirus-transfected cells did not grow in semisolid medium at 52 days after transfection but did grow at 95 days. Addition of glucocorticoid hormones, but not of 25% fetal calf serum, to the growth medium of the early passage cells resulted in limited anchorage-independent growth. An altered level of expression of a number of proteins was found in cells analyzed at different times after transfection. Notably, the expression of a component of the actin filament system, tropomyosin 2, was shown to decrease during growth in vitro. The development of a more fully transformed phenotype at late passages correlated with loss of the requirement for large T-antigen for growth. Thus, cells transfected with a polyomavirus mutant encoding a thermolabile large T-antigen did not grow at the restrictive temperature at 6 weeks after transfection, but grew well at 5 months after transfection. We suggest that these phenomena may be explained by assuming that establishment of rodent fibroblasts, and thereby sensitivity to transformation by middle T-antigen, is not an immediate consequence of expression of large T-antigen but occurs after a period of growth in vitro.

Mechanisms of transformation by polyoma virus middle T antigen

This review addresses a fundamental question of polyoma virus biology: What is the molecular mechanism by which the polyoma virus middle T antigen (MTAg) transforms cells in culture? Since MTAg has no known intrinsic biochemical activity, it is believed to act by modulating the properties of the host cell's proteins (see review by Courtneidge [26]). Experiments to date have largely focused on the interaction of MTAg with the cellular tyrosine kinase, pp60c-src. However, recent data from a number of laboratories have demonstrated the importance of other MTAg-associating cellular proteins in MTAg-mediated transformation, including pp62c-yes and a phosphatidylinositol kinase. In this review, we will summarize what is presently known about the proteins interacting with MTAg. The extent to which the currently known details of the biochemistry of MTAg and its associated proteins can explain the transforming properties of the various mutant alleles of MTAg will be assessed.

Cellular proteins that associate with the middle and small T antigens of polyomavirus

We have used two-dimensional gel electrophoresis to analyze in more detail the cellular proteins which associate with the middle and small tumor antigens (MT and ST, respectively) of polyomavirus. Proteins with molecular masses of 27, 29, 36, 51, 61, 63, and 85 kilodaltons (kDa) that specifically coimmunoprecipitated with MT were identified on these gels. The 36-, 51-, 61-, 63-, and 85-kDa proteins are probably the same as the proteins of similar sizes previously reported by a number of groups, whereas the 27- and 29-kDa proteins represent proteins that are heretofore undescribed. The 27- and 29-kDa proteins were abundant cellular proteins, whereas the others were minor cellular constituents. The association of each of these proteins with MT was sensitive to one or more mutations in MT that rendered it transformation defective. The association of the 85-kDa protein was the most sensitive indicator of the transformation competence of MT mutants. In addition, the 85-kDa protein was the only associated protein whose association with MT changed consistently in parallel with MT-associated phosphatidylinositol kinase activity. Furthermore, the fraction of the 85-kDa protein which was found associated with the MT complex contained 15 to 20% of its phosphate content on tyrosine. The 36- and 63-kDa proteins complexed with both polyomavirus MT and ST and comigrated on two-dimensional gels with two simian virus 40 ST-associated proteins originally described by Rundell and coworkers (K. Rundell, E. O. Major, and M. Lampert, J. Virol. 37:1090-1093, 1981). None of the other MT-associated proteins associated significantly with ST.

Interactions between polyomavirus medium T antigen and three cellular proteins of 88, 61, and 37 kilodaltons

Affinity-purified medium T antigen of wild-type polyomavirus and dl8, a transforming mutant with a deletion in the medium T gene, is associated with three cellular proteins with apparent molecular weights of 88,000 (88K protein), 61,000 (61K protein), and 37,000 (37K protein). Medium T antigen encoded by the nontransforming hrt mutants fails to associate with these proteins, whereas medium T antigen of the nontransforming mutant dl1015 is able to do so. Medium T antigen of the nontransforming mutant dl23 binds to the 61K and 37K proteins; however, binding to the 88K protein is uncertain. The pattern of complex formation between these proteins and medium T antigen resembles that of pp60c-src and medium T antigen. The binding of medium T antigen to the 88K, 61K, and 37K proteins, as well as to pp60c-src, might represent a necessary but insufficient step in transformation. By mixing extracts from infected and uninfected cells, complex formation between medium T antigen and the 88K, 61K, and 37K proteins can be demonstrated in vitro. Pulse-chase experiments indicated that in vivo the association between medium T antigen and the 61K and 37K proteins is a slow process. The latter two proteins are probably bound to each other in uninfected cells. On two-dimensional gels of whole-cell extract, the 61K protein comigrated with a minor protein with an isoelectric point of 5.2. The 61K protein was neither phosphorylated nor glycosylated. Polyomavirus tumor serum precipitated the 61K and 37K proteins independently of medium T antigen. Therefore, the 61K protein or the 37K protein or both have the properties of a cellular tumor antigen.

Association of viral and plasmid DNA with the nuclear matrix during productive infection

The association of simian virus 40 (SV40) DNA or plasmid DNA in subcellular fractions from either infected or transfected cells was examined. In lytically infected cells, approx. 25% of viral specific DNA during the infection cycle was retained in nuclei after washing with low ionic strength buffer and 1% Triton X-100. Viral replicating DNA found in the nuclear matrix was capable of performing limited DNA synthesis by the endogenous DNA polymerase in vitro. Viral DNA synthesized in vitro hybridized preferentially to SV40 Hind-III B and C fragments which are in proximity to the origin of replication. In plasmid-transfected COS-7 cells (SV40-transformed cells), the amount of plasmid DNA found in the nuclear matrix was related to its replication efficiency in cells. More than 80% of the plasmid DNA was tightly associated with subnuclear structures. Little or no plasmid DNA was found in the cytoplasmic fraction. The results suggest that, in extrachromosomal model systems, the association of DNA with nuclear matrix is important for the regulation of DNA replication.

Common elements in growth factor stimulation and oncogenic transformation: 85 kd phosphoprotein and phosphatidylinositol kinase activity

The phosphorylation of proteins on tyrosine in vivo and in vitro was examined in 3T3 cells stimulated by platelet-derived growth factor (PDGF) and transformed by polyoma middle T antigen (MTAg) by using an antibody directed against phosphotyrosine (P-tyr). Two common events were observed upon PDGF stimulation or MTAg transformation of cells: the appearance in the immunoprecipitates of an 85 kd phosphoprotein, and increased phosphatidylinositol (PI) kinase activity. In PDGF-stimulated cells, the 85 kd phosphoprotein and PI kinase activity appeared rapidly, within 1 min of growth factor addition. The PI kinase activity and 85 kd phosphorylation were also increased in anti-P-tyr immunoprecipitates from cells transformed by v-fms and v-sis, but not by SV40 T antigen. The presence of the tyrosine-phosphorylated 85 kd protein correlated with PI kinase activity during several purification steps. These results suggest that the 85 kd phosphoprotein, a putative PI kinase, is a substrate for both the PDGF receptor and MTAg/pp60c-src tyrosine kinase activities.

Overproduction of polyomavirus middle T antigen in mammalian cells through the use of an adenovirus vector

To overproduce biologically active polyomavirus middle T antigen, we used an adenovirus vector and human 293 cells as hosts. Two helper-independent recombinant adenoviruses were isolated that contain a hybrid transcription unit, in differing orientations, at a site in the adenovirus genome from which the E1a and most of the E1b transcription units have been deleted. The hybrid transcription unit consists of the adenovirus type 2 major late promoter and tripartite leader and a cDNA segment capable of encoding polyomavirus middle T antigen and accompanying 3' RNA-processing signals. Both recombinant viruses were stable and replicated to high titers in human 293 cells. The polyomavirus sequences were expressed, predominantly at late times after infection of 293 cells, to yield mRNAs that encoded middle T antigen. One of the recombinant viruses also expressed a middle T antigen-related protein in 293 cells. The latter was translated from one of several novel mRNA species that resulted from aberrant splicing and incomplete RNA processing of precursor RNA transcripts. Comparison of the amount of middle T antigen produced in 3T6 cells infected with polyomavirus with that in 293 cells infected with either of the recombinant adenoviruses, under optimal conditions for each system, revealed at least a 10-fold greater yield of the protein on a per-cell basis in the latter system than in the former. The recombinant-virus-encoded middle T antigen was biologically active, as evidenced by its ability to associate with and serve as a substrate for human pp60c-src. The functionality of the middle T antigen was further confirmed by demonstrating that both recombinant viruses efficiently transformed Rat-1 cells. These recombinant viruses will be useful to overproduce middle T antigen and to introduce the polyomavirus oncogene into a wide variety of mammalian cells.

Characterization of middle T antigen expressed by using an adenovirus expression system

The adenovirus Ad5(pymT) has been used to express middle T antigen at very high levels in 293 cells. The middle T antigen produced was localized to membranes and was modified in the same way as that expressed in polyoma virus-infected mouse cells. It was phosphorylated in vivo on serine residues and in vitro on tyrosine residues. The in vivo phosphorylations occurred between residues 223 and 275. The middle T antigen encoded by A d5(pymT) was phosphorylated in vitro in a complex with human pp60c-src. Interestingly, the extreme overexpression of middle T antigen did not cause a parallel increase in the amount of complex; most of the pp60c-src remained unassociated. Immunoaffinity purification resulted in approximately 100 micrograms of middle T antigen from a 100-mm tissue culture dish. Several cell proteins copurified with the Ad5(pymT)-derived middle T antigen. Two of these, the 74- and 63-kilodalton species, are of particular interest because they were also purified from mouse tumors expressing middle T antigen.

Phosphorylation of polyomavirus large T antigen: effects of viral mutations and cell growth state

Phosphorylation is responsible for the shift in electrophoretic mobility of polyomavirus large T antigen observed in pulse-chase or continuous-labeling experiments. Phosphorylated forms migrated more slowly than newly synthesized [35S]methionine large T antigen, and alkaline phosphatase treatment reversed the mobility shift. Analysis of phosphopeptides with Staphylococcus aureus V8 protease showed that large T antigen forms of intermediate mobility were enriched in peptides 1 to 4, 8, and 9, while the slower migrating species had all nine phosphopeptides, including peptides 5 and 7. The phosphorylations represented by phosphopeptides 5 and 7 were of particular interest. These phosphopeptides were entirely lacking in large T antigen from tsa mutants such as ts616 labeled at the nonpermissive temperature. Also, the phosphorylation of peptides 5 and 7 depends on the growth state of the cell. Early in infection of quiescent cells intermediate mobility forms of large T antigen with little or no phosphorylation, particularly of peptides 5 and 7, were seen, whereas peptides 5 and 7 were well represented at the same time in patterns from growing cells. Later in infection of growth-arrested cells, these phosphorylations were observed, suggesting that infection stimulates the relevant kinase. Because large T antigen of hrt mutants, which lack middle and small T antigens, showed phosphorylation of peptides 5 and 7, large T antigen was apparently responsible for the stimulation. Because some differences in the distribution of phosphopeptides were noted between hrt mutants and the wild type, middle T antigen, small T antigen, or both may play a modulating role in large T antigen phosphorylation.

Regulation of pp60c-src synthesis by inducible RNA complementary to c-src mRNA in polyomavirus-transformed rat cells

To determine the potential role of pp60c-src in polyomavirus-transformed cells, we constructed a recombinant plasmid with the mouse metallothionein-I promoter upstream of a src gene in an anti-sense orientation. We cotransfected this plasmid into middle tumor antigen-transformed FR3T3 cells with a plasmid containing the neomycin resistance gene, and G418 resistant colonies were selected. Analysis of these cells for pp60c-src expression revealed that 50 of the 200 cellular clones screened were found to have decreased levels of c-src expression when compared with the parental middle tumor antigen-transformed cells. Three independent clones which transcribed the expected 3.6-kilobase src complementary RNA and had levels of pp60c-src kinase activity comparable to that of normal FR3T3 cells were further analyzed. In the presence of Cd2+, these clones grew significantly slower in monolayer cultures than either the parental transformed cells (FR18-1) or FR18-1 cells transfected with the neomycin resistance gene alone. The morphology of these clones in the presence of Cd2+ was distinct from that of either the parental FR18-1 cells or normal FR3T3 cells. The clones expressing the complementary src RNA were found to form fewer colonies in soft agar, form fewer foci on monolayers of normal rat cells, and form tumors more slowly following injection into syngenic rats when compared with parental FR18-1 cells. The results of these studies suggest that the level of pp60c-src kinase activity affects the growth characteristics and transformation properties of polyoma virus-transformed rat cells.

Regulation of pp60c-src synthesis by inducible RNA complementary to c-src mRNA in polyomavirus-transformed rat cells

To determine the potential role of pp60c-src in polyomavirus-transformed cells, we constructed a recombinant plasmid with the mouse metallothionein-I promoter upstream of a src gene in an anti-sense orientation. We cotransfected this plasmid into middle tumor antigen-transformed FR3T3 cells with a plasmid containing the neomycin resistance gene, and G418 resistant colonies were selected. Analysis of these cells for pp60c-src expression revealed that 50 of the 200 cellular clones screened were found to have decreased levels of c-src expression when compared with the parental middle tumor antigen-transformed cells. Three independent clones which transcribed the expected 3.6-kilobase src complementary RNA and had levels of pp60c-src kinase activity comparable to that of normal FR3T3 cells were further analyzed. In the presence of Cd2+, these clones grew significantly slower in monolayer cultures than either the parental transformed cells (FR18-1) or FR18-1 cells transfected with the neomycin resistance gene alone. The morphology of these clones in the presence of Cd2+ was distinct from that of either the parental FR18-1 cells or normal FR3T3 cells. The clones expressing the complementary src RNA were found to form fewer colonies in soft agar, form fewer foci on monolayers of normal rat cells, and form tumors more slowly following injection into syngenic rats when compared with parental FR18-1 cells. The results of these studies suggest that the level of pp60c-src kinase activity affects the growth characteristics and transformation properties of polyoma virus-transformed rat cells.

Phosphatidylinositol metabolism and polyoma-mediated transformation

The effect of polyoma middle-sized tumor antigen (MTAg) on phosphatidylinositol metabolism has been characterized in vivo and in vitro using polyoma-transformed and polyoma-infected cells. Cells infected with transformation-competent polyoma virus exhibit increased levels of inositol phospholipids and the second messenger inositol trisphosphate. MTAg or pp60c-src immunoprecipitates from MTAg-transformed cells contain an activity that phosphorylates phosphatidylinositol and phosphatidylinositol 4-phosphate. This activity is induced in parallel with MTAg when the MTAg synthesis is regulated by hormonal or heavy metal inducers. Immunoprecipitates from one class of polyoma mutants defective in transformation have a reduced level of associated phosphatidylinositol kinase activity in vitro yet are capable of tyrosine phosphorylation on exogenous protein substrates at rates comparable to wild-type virus. Thus, for these mutants, phosphatidylinositol kinase activity is more tightly correlated with transformation than is protein kinase activity. These results suggest that alterations in phosphatidylinositol metabolism by MTAg play a role in transformation by polyoma virus.

A subclass of polyomavirus middle tumor antigen binds to DNA cellulose

We examined the binding of polyomavirus large (L-T)-, middle (M-T)-, and small-tumor antigens to DNA cellulose. At pH 6.0, the majority of L-T bound to calf thymus DNA cellulose, while little or no small tumor antigen was retained under these conditions. Unexpectedly, a small but reproducible proportion of M-T bound to both native and denatured DNA cellulose. M-T encoded by polyomavirus mutant dl 8, which expressed shortened L-T and M-T, bound to DNA, indicating that the deleted sequences are not required for DNA binding. Also, M-T from transformed BMT-1 rat cells, which synthesize exclusively this polyomavirus tumor antigen, bound to DNA, indicating that its binding is not due to association with other polyomavirus-encoded proteins. Using the DNA fragment immunoassay, we found that, under conditions in which L-T bound specifically to DNA fragments containing viral regulatory sequences, no viral DNA fragments were bound by M-T. The existence of distinct subpopulations of M-T that differ in their DNA-binding properties was indicated by rebinding experiments in which M-T that had bound to DNA cellulose rebound very efficiently, while that which had not been originally retained by DNA cellulose rebound poorly. Furthermore, the M-T-pp60 c-src complex did not bind to DNA cellulose. These data suggest that polyomavirus M-T is heterogeneous, consisting of populations of molecules that differ in their interactions with DNA cellulose.

Guanine nucleotide contacts within viral DNA sequences bound by polyomavirus large T antigen

Essential nucleotide contacts between the polyomavirus large T antigen and its multiple specific binding regions within the regulatory sequences of the polyomavirus genome were determined in vitro by methylation interference. Methylation of any of the guanine residues of the 5'-G(A/G)GGC-3' pentanucleotide repeats in large-T-antigen-binding regions A, B, C, and 3 (A. Cowie and R. Kamen, J. Virol. 52:750-760, 1984) interfered with T antigen binding. Within regions A, B, and C these pentanucleotides are spaced 5 or 6 base pairs apart. Therefore, the clusters of contacted nucleotides within each of these binding regions are localized along one face of the DNA helix. Methylation of guanines within the sequences between the pentanucleotide repeats did not interfere with binding. The ORI binding region contains four additional pentanucleotide sequences within a region of dyad symmetry. Methylation of only particular guanines of these pentanucleotides interfered with T antigen binding. The spatial arrangement of the pentanucleotides in the ORI is such that the clusters of contacted guanines are situated around the DNA helix, thereby forming a very different arrangement from that found in the other binding regions. A model is discussed in which cooperative interactions between T antigen protomers, recognizing individual pentanucleotides, determines the strength and the function of different T antigen-DNA interactions.

Analysis of pp60c-src protein kinase activity in hamster embryo cells transformed by simian virus 40, human adenoviruses, and bovine papillomavirus 1

We have examined the effect of DNA tumor virus transformation of primary hamster embryo cells on the tyrosyl kinase activity of pp60c-src. Our present study demonstrates that some clones of hamster embryo cells transformed by simian virus 40, adenovirus type 2, adenovirus type 12, or bovine papillomavirus 1 can possess elevated pp60c-src kinase activity when compared with normal hamster embryo cells. However, other clones of hamster embryo cells transformed by these same viruses were found to have normal levels of pp60c-src kinase activity. In those clones of transformed cells where pp60c-src kinase activity was elevated, the increased levels of kinase activity were the result of an apparent increase in the specific activity of the pp60c-src phosphotransferase rather than an increase in the amount of the src gene product. Additionally, pp60c-src was not found to be physically associated with tumor antigens known to be encoded by these viruses. These results indicate that elevated levels of pp60c-src kinase activity can be found in hamster embryo cells transformed by several different DNA tumor viruses and suggest that the molecular mechanism by which pp60c-src kinase activity is elevated may differ from that previously observed in polyomavirus-transformed cells. These results also imply that elevation of pp60c-src kinase activity is not required for the transformation of hamster cells by these viruses.

Sequences from polyomavirus and simian virus 40 large T genes capable of immortalizing primary rat embryo fibroblasts

We developed a procedure to evaluate quantitatively the capacity of subgenomic fragments from polyomavirus and simian virus 40 (SV40) to promote the establishment of primary cells in culture. The large T antigen from both of these viruses can immortalize primary rat embryo fibroblasts. Both antigens have amino-terminal domains that retain biological activity after deletion of other parts of the polypeptide chain. However, this activity varies considerably among various mutants, presumably because of alterations in the stability or conformation of the truncated polypeptides. The polyomavirus middle T gene alone immortalizes at a low efficiency, which indicates that this oncogene can have both immortalization and transformation potentials depending on the assay system chosen. We generated deletions in the polyomavirus and SV40 large T genes to localize more precisely the functional domains of the proteins involved in the immortalization process. Our results show that the region of the SV40 large T antigen involved in immortalization is localized within the first 137 amino acid residues. This region is encoded by the first large T exon and a small portion from the second exon which includes the SV40 large T nuclear location signal. The polyomavirus sequence involved in immortalization comprises a region from the second large T exon, mapping between nucleotides 1016 and 1213, which shares no homology with SV40 and is thought to be of cellular origin. We suggest that this region of the polyomavirus large T gene functions either as a nuclear location signal or as part of the large T protein sequence involved in DNA binding.

Large-scale production of polyoma middle T antigen by using genetically engineered tumors

A recombinant plasmid containing a metallothionein promoter-polyoma middle T cDNA fusion was constructed and used to transfect NIH 3T3 cells. Transformed cells expressing middle T were injected into nude mice. Within 3 weeks, each mouse produced tumors containing middle T equivalent to that in 250 to 1,000 100-mm dishes of polyomavirus-infected cells. This middle T, partially purified by immunoaffinity chromatography, retained activity as measured by its ability to be phosphorylated in vitro. The combined approach of fusing strong promoters to genes of interest and utilizing nude mice to grow large quantities of cells expressing the gene provides a quick, inexpensive alternative to other expression systems.

Association of phosphorylated simian virus 40 T-antigen with subnuclear fractions of infected and transformed cells

To define the roles of subnuclear structure in SV40 infection, the relative distribution of T-antigen (T-ag) in various subnuclear fractions obtained from both lytically infected and transformed African green monkey kidney cells was determined. Depending on the differential sensitivity of nuclear T-ag to extraction by salt and detergent, nuclear T-ag could be separated into nucleoplasmic T-ag, salt-sensitive T-ag and matrix-bound T-ag subclasses. At least fivefold less matrix-bound T-ag was found in transformed cells than in lytically infected cells. While a cAMP-independent protein kinase was detected in the nuclear matrix, the matrix-bound T-ag (94K) could not be phosphorylated in vitro. The removal of cellular chromosomes by DNase caused changes in the interaction of T-ag with nuclear components. The results suggest that the compartmentalization of nuclear T-ag may be determined by its interaction with host chromosomes.

Large-scale production of polyoma middle T antigen by using genetically engineered tumors

A recombinant plasmid containing a metallothionein promoter-polyoma middle T cDNA fusion was constructed and used to transfect NIH 3T3 cells. Transformed cells expressing middle T were injected into nude mice. Within 3 weeks, each mouse produced tumors containing middle T equivalent to that in 250 to 1,000 100-mm dishes of polyomavirus-infected cells. This middle T, partially purified by immunoaffinity chromatography, retained activity as measured by its ability to be phosphorylated in vitro. The combined approach of fusing strong promoters to genes of interest and utilizing nude mice to grow large quantities of cells expressing the gene provides a quick, inexpensive alternative to other expression systems.

Monomers through trimers of large tumor antigen bind in region I and monomers through tetramers bind in region II of simian virus 40 origin of replication DNA as stable structures in solution

Large tumor (T) antigen and its bound multimeric states are positioned by scanning transmission electron microscopy (STEM) within a few base pairs at control sequences of the simian virus 40 DNA origin of replication region. Proximal and distal edge positions for each multimer group match the end positions of previously mapped fragments protected from DNase cleavage. Since chance correspondence is shown to be extremely unlikely, STEM mass measurements, obtained concurrently with STEM map positions, indicate that the DNase fragments arise from bound monomers, dimers, trimers, and tetramers in binding region II and monomers, dimers, and trimers in binding region I. Simultaneous binding of seven monomer-equivalent masses is observed, three in region I and four in region II, with an ordered and interpretable mass distribution in the plane of the foil. Although this observation does not prove that the six G-A-G-G-C and one T-A-G-G-C sequences, similarly distributed, function as recognition sequences for T-antigen monomer, it provides strong support for such a model. The stable existence in solution of low-and intermediate-mass structures, observed at lower T-antigen concentrations, suggests a role as assembly intermediates.

Association of phosphatidylinositol kinase activity with polyoma middle-T competent for transformation

Polyoma middle-T antigen is required for viral transformation of cultured cells and for tumorigenesis in animals. Like many other transforming gene products, middle-T is bound to the membrane and has an associated tyrosine kinase activity in vitro. This activity seems to result from the interaction of middle-T with pp60c-src, the cellular homologue of the transforming gene product of the Rous sarcoma virus, pp60v-src (refs 3-5). Both pp60v-src (ref. 6) and another retrovirus transforming gene product, pp68v-ros (ref. 7) were shown recently to have an associated phosphatidylinositol (PI) kinase activity in vitro and to increase PI turnover in vivo. These results suggest that viral transformation may be directly connected to a complex network of second messengers generated from PI turnover. Here, we assayed for PI kinase activity in immunoprecipitates made with middle-T- or pp60c-src-specific antisera of cells infected with polyoma virus. A PI kinase activity was detected in those immunoprecipitates which contained middle-T. Studies of mutants of middle-T defective in transformation indicate a close correlation between PI kinase activity and transformation.

A large-tumor-antigen-specific monoclonal antibody inhibits DNA replication of simian virus 40 minichromosomes in an in vitro elongation system

In productively infected cells, a fraction of large-tumor antigen (T antigen) is tightly bound to replicating simian virus 40 (SV40) minichromosomes and does not dissociate at salt concentrations of greater than 1 M NaCl. We present electronmicrograms demonstrating the presence of T antigen on the replicated sections of replicating SV40 minichromosomes. We also show that the fraction of tightly bound T antigen is recognized by antibodies from mouse tumor serum and, more specifically, by a particular T-antigen-specific monoclonal antibody, PAb 1630. A second T-antigen-specific monoclonal antibody, PAb 101, does not react with the T-antigen fraction remaining on replicating SV40 chromatin at high salt concentrations. We used an in vitro replication system which allows, via semiconservative DNA replication, the completion of in vivo-initiated replicative intermediate DNA molecules. We show that monoclonal antibody PAb 1630, but not monoclonal antibody PAb 101, inhibits viral DNA replication. We discuss the possibility that SV40 T antigen may play a role in chain elongation during SV40 chromatin replication.

A polyoma mutant that encodes small T antigen but not middle T antigen demonstrates uncoupling of cell surface and cytoskeletal changes associated with cell transformation

The hr-t gene of polyoma virus encodes both the small and middle T (tumor) antigens and exerts pleiotropic effects on cells. By mutating the 3' splice site for middle T mRNA, we have constructed a virus mutant, Py808A, which fails to express middle T but encodes normal small and large T proteins. The mutant failed to induce morphological transformation or growth in soft agar, but did stimulate postconfluent growth of normal cells. Cells infected by Py808A became fully agglutinable by lectins while retaining normal actin cable architecture and normal levels of extracellular fibronectin. These properties of Py808A demonstrated the separability of structural changes at the cell surface from those in the cytoskeleton and extracellular matrix, parameters which have heretofore been linked in the action of the hr-t and other viral oncogenes.

An adenovirus vector system used to express polyoma virus tumor antigens

We have used a generalized adenovirus vector system to express the three polyoma tumor (T) antigen proteins under the control of the adenovirus major late promoter. One hybrid virus, Ad-PySVR498, expresses high levels of polyoma middle and small T antigens. A second hybrid virus, Ad-LTSVR545, which contains a cDNA copy of the polyoma A gene, overproduces large T antigen. The T antigens produced are indistinguishable from their authentic polyoma counterparts as determined by immunoprecipitation and partial cleavage by V8 protease. Analysis of polyoma mRNAs encoded by the recombinant viruses showed that they initiate from the adenovirus major late promoter and contain the tripartite leader at their 5' ends. Large T antigen isolated from Ad-LTSVR545-infected cells by immunoaffinity was shown to bind selectively to polyoma DNA sequences that contain the origin of viral DNA replication as well as the sites for transcription initiation.

Multiple binding sites for polyomavirus large T antigen within regulatory sequences of polyomavirus DNA

Polyomavirus large T antigen binds specifically to multiple sites within the regulatory region of the viral genome. Experiments done with crude extracts from wild-type virus-infected mouse cells and immunoprecipitation of protein-DNA complexes localized two high-affinity binding sites on the early region side of the DNA replication origin. Purification of the large T antigen by immunoaffinity chromatography made it possible to refine the analysis through application of DNase I footprinting. The high-affinity interactions were resolved into three closely spaced, but distinct, binding regions. These begin at a site only slightly overlapping the early boundary of the core replication origin, a location highly homologous to that of simian virus 40 large T antigen-binding site I, but then extend away from the origin toward the early coding sequence and thus span the early region transcriptional initiation sites. Each tight-binding region contains from two to four copies of the sequence 5'-(A = T)G(A greater than G)GGC-3' repeated at 9- to 11-base-pair spacing. At high protein concentrations and at low ionic strength, additional sites within the core replication origin and in the enhancer region were protected from DNase I digestion. These minor binding sites also included repeats of sequences related to the consensus, but at different spacings. Our results suggest that, unlike simian virus 40 DNA, the polyomavirus genome may have distinct regions of interaction with its large T antigen which separately are involved in initiation of DNA replication and the regulation of viral transcription.

A polyoma mutant that encodes small T antigen but not middle T antigen demonstrates uncoupling of cell surface and cytoskeletal changes associated with cell transformation

The hr-t gene of polyoma virus encodes both the small and middle T (tumor) antigens and exerts pleiotropic effects on cells. By mutating the 3' splice site for middle T mRNA, we have constructed a virus mutant, Py808A, which fails to express middle T but encodes normal small and large T proteins. The mutant failed to induce morphological transformation or growth in soft agar, but did stimulate postconfluent growth of normal cells. Cells infected by Py808A became fully agglutinable by lectins while retaining normal actin cable architecture and normal levels of extracellular fibronectin. These properties of Py808A demonstrated the separability of structural changes at the cell surface from those in the cytoskeleton and extracellular matrix, parameters which have heretofore been linked in the action of the hr-t and other viral oncogenes.

Inhibition of polyoma DNA synthesis by base pair substitutions at the replication origin

The effect of base pair substitutions on the function of the polyoma virus origin of DNA replication was studied. The mutations were all C-G to T-A transitions, induced by bisulfite treatment of recombinant DNA molecules. The mutagenesis was directed to short single-stranded gaps in duplex DNA, or to loops in heteroduplex molecules. Modification of a 34 base pair sequence of dyad symmetry led to cis-acting inhibition of viral DNA synthesis, ranging from slight defects to total inactivation. One of the mutants was temperature sensitive. Mutants with base changes in an adjacent DNA segment, including an 18 base pair long purine-pyrimidine tract, had similar, but less severe, deficiences. In contrast to the effect of mutations in the homologous region of the simian virus 40 genome, there was no strict relationship between mutation of the putative large T-antigen-binding base sequence GPuGGC and defective viral DNA synthesis.

Construction and functional characterization of polyomavirus genomes that separately encode the three early proteins

Modified polyomavirus genomes that individually encode the large and small T proteins were constructed by exchanging restriction endonuclease fragments between cDNA copies of the respective mRNAs and cloned genomic DNA. The efficacies of the new constructs, and that of the middle T protein gene described previously (R. Treisman , U. Novak, J. Favaloro , and R. Kamen , Nature [London] 292:595-600, 1981), were demonstrated with simian virus 40 (SV40)-polyomavirus recombinants in which part or all of the SV40 late region was replaced with the modified polyomavirus early genes. Each of the three recombinant viruses induced the synthesis of only the expected polyomavirus early protein in infected CV-1 cells. The rates of synthesis of large, middle, and small T proteins were ca. 1.5, 4.0, and 9.0 times the rate of synthesis of SV40 large T protein, respectively. The deletion of introns had no detrimental effect on mRNA biogenesis. Indeed, a further polyomavirus-SV40 recombinant, containing wild-type polyomavirus early region DNA, expressed an aberrant 58,000-dalton form of the middle T protein which we believe to result from utilization of a cryptic splice site. Immunofluorescence studied with monkey cells infected by the recombinant viruses allowed us to determine the cellular locations of the polyomavirus early proteins. Overproduction of the middle T protein did not result in a corresponding overproduction of the middle T protein-associated tyrosine phosphokinase activity.

DNA binding activity of polyoma virus large tumor antigen

Polyoma virus large tumor antigen from productively infected mouse cells has been purified to greater than 50% homogeneity by a simple immunoaffinity procedure using monoclonal antibodies. A radioimmunoreaction was devised for assaying purity. The purified large tumor antigen retained its antigenicity and its ability to bind DNA specifically. The regions on the polyoma virus genome recognized by the protein were characterized. Three binding regions were localized within the portion of the genome between the viral origin of DNA replication and the protein coding sequence, overlapping the early promoter and the sites of initiation of mRNAs that specify the viral tumor antigens. The binding regions each contain direct repeats of the pentanucleotide sequence G-R-G-G-C.

Isolation and characterization of polyoma nucleoprotein complexes

A method for the isolation of polyoma nucleoprotein complexes has been developed using neuraminidase treatment of infected cell lysates. At least three distinct forms of polyoma virion intermediates were identified by their [3H]thymidine labeling kinetics and sedimentation coefficients: a rapidly labeled 95 S "replicating complex" which chases to a 75 S minichromosome and then to a 240 S virion structure. The general properties of these distinct intermediates were similar to those found for SV40. In contrast to SV40, however, a continuum of labeled polyoma viral DNA sedimented between 240 S and 95 S. These complexes were characterized by their release from cell debris with neuraminidase, precipitation with antivirion antibody, complete disruption in 1 M NaCl, and association with hemagglutinating (HA) activity. These intermediates may represent incremental capsid protein additions to the 75 S minichromosome, hypothesized in the current models for SV40 assembly. The ability to isolate a complete complement of polyoma subviral complexes provides a basis for studying the growth defect of polyoma host-range mutants, and the properties of neuraminidase release, hemagglutination, and specific immunoprecipitation suggest purification steps for further characterization of these virion assembly intermediates.

Analysis of simian virus 40 chromosome-T-antigen complexes: T-antigen is preferentially associated with early replicating DNA intermediates

The fraction and DNA composition of simian virus 40 chromosomes that were complexed with large T-antigens (T-Ag) were determined at the peak of viral DNA replication. Simian virus 40 chromatin containing radiolabeled DNA was extracted by the hypotonic method of Su and DePamphilis (Proc. Natl. Acad. Sci. U.S.A. 73:3466-3470, 1976) and then fractionated by sucrose gradient sedimentation into replicating (90S) and mature (70S) chromosomes. Viral chromosomes containing T-Ag were isolated by immunoprecipitation with saturating amounts of either an anti-T-Ag monoclonal antibody or an anti-T-Ag hamster serum under conditions that specifically precipitated T-Ag protein from cytosol extracts. An average of 10% of the uniformly labeled DNA in the 90S pool and 7.5% in the 70S pool was specifically precipitated, demonstrating that under these conditions immunologically reactive T-Ag was tightly bound to only 8% of the total viral chromosomes. In contrast, simian virus 40 replicating intermediates (RI) represented only 1.2% of the viral DNA, but most of these molecules were associated with T-Ag. At the shortest pulse-labeling periods, an average of 72 +/- 18% of the radiolabeled DNA in 90S chromosomes could be immunoprecipitated, and this value rapidly decreased as the labeling period was increased. Electron microscopic analysis of the DNA before and after precipitation revealed that about 55% of the 90S chromosomal RI and 72% of the total RI from both pools were specifically bound to T-Ag. Comparison of the extent of replication with the fraction of RI precipitated revealed a strong selection for early replicating DNA intermediates. Essentially all of the RI in the 70S chromosomes were less than 30% replicated and were precipitated with anti-T-Ag monoclonal antibody or hamster antiserum. An average of 88% of the 90S chromosomal RI which were from 5 to 75% replicated were immunoprecipitated, but the proportion of RI associated with T-Ag rapidly decreased as replication proceeded beyond 70% completion. By the time sibling chromosomes had separated, only 3% of the newly replicated catenated dimers in the 90S pool (<1% of the dimers in both pools) were associated with T-Ag. Measurements of the fraction of radiolabeled DNA in each quarter of the genome confirmed that T-Ag was preferentially associated with newly initiated molecules in which the nascent DNA was nearest the origin of replication. These results are consistent with a specific requirement for the binding of T-Ag to viral chromosomes to initiate DNA replication, and they also demonstrate that T-Ag does not immediately dissociate from chromosomes once replication begins. The biphasic relationship between the fraction of T-Ag-containing RI and the extent of DNA replication suggests either that 1 or 2 molecules of T-Ag remain stably bound until replication is about 70% completed or that 4 to 6 molecules of T-Ag are randomly released from each RI at a uniform rate throughout replication.