LOADING OF A DNA HELICASE ON THE DNA UNWINDING ELEMENT IN THE YEAST REPLICATION ORIGIN - MECHANISM OF DNA-REPLICATION IN A MODEL SYSTEM

We found that initiation of DNA replication occurs from the region containing the yeast autonomously replicating sequence 1 (ARS1), by incubating negatively supercoiled plasmid DNA with the proteins required for SV40 DNA replication in addition to DNA gyrase (Ishimi, Y., & Matsumoto, K. (1993) Proc. Natl. Acad. Sci. U.S.A. 90, 5399-5403). Here, the mechanism of DNA replication and the roles of the replication proteins in this model system were analyzed. Both SV40 T antigen as a DNA helicase and multisubunit human single-stranded DNA binding protein (HSSB) (also called RP-A) were required for the initial step of DNA synthesis. Furthermore, it has been shown that T antigen plays an essential role in the initiation of DNA replication from the ARS region in this system. The digestion of negatively supercoiled DNA with the single-strand-specific nuclease P1 revealed that regions containing A, B, and C domains of ARS1 can be unwound under the conditions used for DNA replication. Footprinting with KMnO4 indicated that T antigen interacted with the unwound B domain where initiation of DNA replication mainly occurred. When circular DNAs of different negative-superhelical densities were replicated in the absence of DNA gyrase, short fragments were synthesized from the ARS region in proportion to its density and they were elongated by addition of HeLa topoisomerase I, which inhibits the initiation of DNA replication in this system. These results suggested that T antigen loads on the B domain of ARS1, where the DNA duplex is destabilized by the torsional stress of negative supercoiling, to form a preinitiation complex with the assistance of HSSB, and that DNA gyrase plays a major role at the elongation step of DNA synthesis as a swivelase. The role of T antigen in this system is consistent with the model of ARS DNA replication, in which a DNA helicase enters the ARS region through the 3' flanking AT-rich region (B domain) of the consensus sequence.