Changes in gene expression and protein phosphorylation in murine cells, transformed or abortively infected with wild type and mutant simian virus 40

Karyotype alteration generates the neoplastic phenotypes of SV40-infected human and rodent cells

BACKGROUND

Despite over 50 years of research, it remains unclear how the DNA tumor viruses SV40 and Polyoma cause cancers. Prevailing theories hold that virus-coded Tumor (T)-antigens cause cancer by inactivating cellular tumor suppressor genes. But these theories don't explain four characteristics of viral carcinogenesis: (1) less than one in 10,000 infected cells become cancer cells, (2) cancers have complex individual phenotypes and transcriptomes, (3) recurrent tumors without viral DNA and proteins, (4) preneoplastic aneuploidies and immortal neoplastic clones with individual karyotypes.

RESULTS

As an alternative theory we propose that viral carcinogenesis is a form of speciation, initiated by virus-induced aneuploidy. Since aneuploidy destabilizes the karyotype by unbalancing thousands of genes it catalyzes chain reactions of karyotypic and transcriptomic evolutions. Eventually rare karyotypes evolve that encode cancer-specific autonomy of growth. The low probability of forming new autonomous cancer-species by random karyotypic and transcriptomic variations predicts individual and clonal cancers. Although cancer karyotypes are congenitally aneuploid and thus variable, they are stabilized or immortalized by selections for variants with cancer-specific autonomy. Owing to these inherent variations cancer karyotypes are heterogeneous within clonal margins. To test this theory we analyzed karyotypes and phenotypes of SV40-infected human, rat and mouse cells developing into neoplastic clones. In all three systems we found (1) preneoplastic aneuploidies, (2) neoplastic clones with individual clonal but flexible karyotypes and phenotypes, which arose from less than one in 10,000 infected cells, survived over 200 generations, but were either T-antigen positive or negative, (3) spontaneous and drug-induced variations of neoplastic phenotypes correlating 1-to-1 with karyotypic variations.

CONCLUSIONS

Since all 14 virus-induced neoplastic clones tested contained individual clonal karyotypes and phenotypes, we conclude that these karyotypes have generated and since maintained these neoplastic clones. Thus SV40 causes cancer indirectly, like carcinogens, by inducing aneuploidy from which new cancer-specific karyotypes evolve automatically at low rates. This theory explains the (1) low probability of carcinogenesis per virus-infected cell, (2) the individuality and clonal flexibility of cancer karyotypes, (3) recurrence of neoplasias without viral T-antigens, and (4) the individual clonal karyotypes, transcriptomes and immortality of virus-induced neoplasias - all unexplained by current viral theories.

A casein-kinase-2-related protein kinase is tightly associated with the large T antigen of simian virus 40

The simian virus 40 (SV40) large T antigen is a multifunctional protein involved in SV40 cell transformation and lytic virus infection. Some of its activities are regulated by interaction with cellular proteins and/or by phosphorylation of T antigen by various protein kinases. In this study, we show that immuno-purified T antigen from SV40-transformed cells and from baculovirus-infected insect cells is tightly associated with a protein kinase that phosphorylates T antigen in vitro. In the presence of heparin or a peptide resembling a protein kinase CK2 recognition site, the phosphorylation of T antigen by the associated kinase is reduced whereas a p34cdc2-kinase-specific peptide has no influence. In addition, the T-antigen-associated protein kinase can use GTP and ATP as phosphate donors. These properties together with the observation that immunopurified T antigen can be phosphorylated by the addition of protein kinase CK2 suggest that at least one of the T-antigen-associated protein kinases is CK2 or a protein-kinase-CK2-related enzyme. The association of recombinant CK2 with T antigen was strongly confirmed by in vitro binding studies. Experiments with temperature-sensitive SV40-transformed cells provide evidence for a close correlation between cell transformation and phosphorylation of T antigen by the associated protein kinase.

Enhanced protein phosphorylation in SV40-transformed and -infected cells

We have studied the phosphorylation of cellular phosphoproteins and, in more detail, of SV40 T antigen and the cellular protein p53 in SV40 tsA-transformed cells. As detected by radiolabeling cold-sensitive tsA1499- or heat-sensitive tsA58-transformed rat fibroblasts with [32P]orthophosphate or by in vitro labeling extracts with [gamma-32P]ATP the hyperphosphorylation of certain cellular phosphoproteins including p53 and also of free SV40 large T antigen and T antigen complexed with p53 is strictly correlated with the expression of the transformed phenotype. This hyperphosphorylation can be observed as early as 30 min after shifting to the temperature where the cells expressed the transformed phenotype and, furthermore, it is dependent on protein synthesis. To evaluate the influence of a functional T antigen and to exclude properties of individual transformants we 32P labeled in vitro cellular proteins from rat F111, mouse NIH 3T3, and monkey TC-7 cells infected with tsA58 or tsA1499. In tsA58-infected cells we found a heat-sensitive enhancement of protein phosphorylation just as in tsA58 transformants. In tsA1499-infected monkey cells we observed a heat-sensitive and in abortively infected rat or mouse cells a cold-sensitive hyperphosphorylation of proteins. Thus in tsA-transformants and in various tsA-infected cells we found a strong correlation among the transformed phenotype, functions of T antigen, and the phosphorylation of various cellular proteins and in particular T antigen and p53.

Identity of the proliferating cell nuclear antigen and cyclin

Studies of growth regulation and cellular transformation will be assisted by the identification of proteins that are preferentially synthesized in dividing cells. The 'proliferating cell nuclear antigen' ( PCNA ), distinguished by its apparent association with cell division, is defined by reaction with an antibody found in the autoimmune disease systemic lupus erythematosus (SLE). This antibody reacts with proliferating cells including tumour cells but gives weak or undetectable immunofluorescence with resting cells of normal tissues. Peripheral blood lymphocytes are devoid of PCNA until activated by mitogen in vitro. In synchronized cultures its level and distribution fluctuate through the cell cycle, with a striking accumulation in the nucleolus late in the G1 phase and early in the S phase. Many of these properties are shared by ' cyclin '. This nuclear protein, identified by its position in a two-dimensional separation of cell proteins, is also transformation-sensitive and is preferentially synthesized in the S phase. We establish here that PCNA and cyclin are identical, and show that PCNA is an acidic nuclear protein of apparent molecular weight 35,000.

Decreased adenylate cyclase responsiveness of transformed cells correlates with the presence of a viral transforming protein

The adenylate cyclase responsiveness of transformed fibroblastic and epithelial cell lines to forskolin, fluoride, guanine nucleotides and cholera toxin was reduced compared to their parental counterparts. This phenomenon was observed in lines transformed by either RNA or DNA tumor viruses, and in the case of polyoma virus, coincided with the expression of middle T antigen. The data suggest that decreased responsiveness of adenylate cyclase to non-hormone activators is a general consequence of viral transformation and may be related to viral regulation of protein kinase activity.