Mechanisms for the transactivation of the cyclin A gene by polyomavirus large and small tT antigens

DNA tumorviruses (Adenoviruses and the papovavirus group, papillomavirus, Simian virus 40 and polyoma virus) are DNA viruses with genome sizes between about 5000 and 35000 base pairs. These viruses preferentially infect differentiated, and hence growth arrested cells but require cells in S phase for their replication because they depend almost completely on host replication enzymes for the synthesis of their DNA and many of these enzymes are present in sufficient quantity only in this phase of the cell cycle. These viruses, therefore, encode a few proteins which interfere with the growth regulation of the host cell and allow to drive cells from the quiescent state into the cell cycle. While the evolutionary important role of these viral proteins is their capacity to prepare the permissive host cell for efficient replication of the viral genome, the same activities are also responsible for immortalizing and transforming activities of the tumor viruses in the event of abortive infection of non-perrnissive cells. In case of the mouse virus, polyonia, these activities rest in three early viral proteins, the large, middle and small tumor (T) antigens. We found previously that the large T antigen is capable of transactivating S phase specific enzymes involved in DNA replication and precursor productions but does not itself lead to efficient S phase induction of sertum-starved Swiss 3T3 mouse fibroblasts. lf present together with small T antigen, a significant fraction of cells are driven into S phase. This led us to think about possible candidates for proteins required for S phase induction which are not synthesized in sufficient quantity in response to large T alone. One such protein could be cyclin A which is known to be synthesized at the G1/S border of the cell cycle and to be required for induction and propagation of the S phase. Looking for the synthesis of cyclin A in cells harbouring large T, small T or both proteins, we found that both viral proteins individually are capable of induction of cyclin A in serum starved cells, albeit to a lower degree than addition of serum and also less efficiently than early proteins of other DNA tumorviruses viruses, for instance the E1A protein of adenovirus 5. Both T antigens together cause an effective induction of cyclin A. The promoter sequences of the murine and human cyclin A genes are known but the mechanism of the growth- and cell cycle regulation of this promoter is so far largely unknown. Clarification as to how polyomavirus large and small T antigens transactivate cyclin A will therefore, on the one hand, contribute to our knowledge as to the variety of ways by which products of DNA tumorviruses interfere with events regulating the growth and cell cycle status of cells. lt will, on the other hand, also allow us to investigate the mechanisms by which one decisive step in the control of S phase initiation, the transcription of the cyclin A gene, is regulated in mammalian cells. lt is, consequently, the aim of this proposal to elucidate the mechanisms by which these two viral proteins activate the synthesis of cyclin A, which transcription factors active at the cyclin A promoter are the targets for and which regulatory interactions are disrupted or changed by the viral protein. Towards this goal, the wild type and mutated forms of the murine cyclin A promoter-luciferase constructs will first be employed in cells carrying wild type or mutated versions of the viral proteins. This should allow to define those portions of the promoter involved in transactivation by the viral protein and to determine which amino acid sequences within the viral protein is involved. This should set the stage to define the cellular partners for the viral proteins which play a role in the regulated expression of cyclin A and should further allow us to learn about the mechanism of growth and cell cycle regulation of the cyclin A gene.

Nearly everybody makes at least once in his life the acquaintance with warts. These benigne swellings are caused by papillomaviruses and usually disappear by themselves. Papillomaviruses are members of a group of DNA viruses which in some cases can cause tumors. In women, papillomaviruses are responsible for the development of cervix carcinoma. Other members of this group of DNA viruses are the mousevirus polyoma, the monkey virus Simian virus 40 (SV40) and the adenoviruses. Common to these viruses is their infection mostly of differentiated cells which are growth arrested. This causes a problem for the viruses because they depend to a large extent on the replication machinery of the host cell for the replication of their own DNA, and this machinery is not epressed in arrested cells. To overcome this problem, these viruses code for proteins which interfere with the hosts regulatory system. This allows them to drive cells out of the arrested state into the S phase, the cell cycle phase in which DNA replication takes place. The very same viral proteins are responsible for the cancerogenic activity of the viruses. Our research dealt with two such proteins the large (LT) and the small (ST) tumorantigen of polyomavirus. We found earlier, that only if both proteins are expressed can a mouse fibroblast cell be driven from the quiescent state into S phase. LT is responsible for the transactivation of DNA synthesis enzymes. In doing so, it interacts with the cellular tumor suppressor proteins pRB, the retinoblastoma proteins, and its relatives. More recently we were studying the transactivation of one of the essential regulators of S phase induction and propagation, cyclin A. This protein, together with the protein kinase cdk2, phosphorylates members of the replication machinery. Work on the induction of cyclin A/cdk2 activity by the combined activities of LT and ST allowed us to explaine why we need both proteins for S phase induction. While LT transactivated cyclin A, ST leads to the destruction of an inhibitor of cdk2 and thereby is essential for the initiation of cyclinA/cdk2 activity. We then found that also ST can transactivate cyclin A, but not DNA synthesis enzymes, so that it alone can not induce an S phase. The target sites of the viral proteins on the region of the cyclin A gene responsible for the control of expression were uncovered as well as the functional interactions of the T antigens which are required for transactivation of cyclin A. In case of LT this is the aforementioned interaction with pRB and its relatives (p107 and p130) as well as a region in the C- terminal domain of LT which originally was thought to function only in the replication of viral DNA. In case of ST this is an interaction with a major cellular protein phosphatase, PP2A. This protein is known to be inhibited by the contact with ST protein. The results of our studies help to explain the mechanisms involved in the deregulation of growth and cell cycle by proteins coded by DNA tumorviruses.