Identification of the domains required for direct interaction of the helicase-like and polymerase-like RNA replication proteins of brome mosaic virus

Abstract

Brome mosaic virus is a positive-strand RNA virus whose RNA replication requires viral protein 1a, which has putative helicase and capping functions, and 2a, which has putative polymerase function. Since domains of related sequence are conserved in a wide range of plus-strand RNA viruses, analysis of 1a and 2a function should have applicability to many other viruses. We have recently demonstrated that 1a and 2a form a complex in vivo and in vitro. Using immune coprecipitation and mutant polypeptides made in reticulocyte lysates, we have now mapped both the 1a and 2a domains necessary for complex formation. The sequences needed to bind 2a map to the carboxy-terminal helicase-like domain of 1a. Truncated polypeptides containing this domain were able to bind to 2a, while several small insertions in the helicase-like domain disrupted binding. The sequence required for binding 1a lies within a 115-residue subset of the 2a N-terminal segment preceding the polymerase-like domain. Truncations or fusion polypeptides containing this segment can bind 1a. We also determined that highly purified 2a protein made in insect cells can form a complex with highly purified 1a helicase-like domain made in Escherichia coli, suggesting that no other factor is required to mediate 1a-2a interaction. Previous genetic analyses of 1a and 2a are consistent with this mapping and show that the newly defined 1a and 2a binding regions are required for RNA synthesis. The locations of these interacting regions are discussed with regard to models of viral replication and the evolution of positive-strand RNA virus genomes.