Flaviviruses, including dengue, West Nile and Zika viruses, pose significant threats as emerging diseases and potential bioterror agents. Despite the considerable impact of flavivirus infection on world-wide health, no antiviral therapies are available and existing flavivirus vaccines are of limited utility. Our long-term goal is to obtain detailed structural and biochemical information regarding the flavivirus replication process and to use this information for the development of antiviral therapeutics and vaccines. The flavivirus replication complex, consisting of virally-encoded non-structural proteins (NS), unidentified cellular proteins, and the viral RNA genome, is responsible for copying the viral genome. Viral replication begins with negative-strand synthesis from the positive-strand RNA genome, leading to dsRNA formation, which in turn is transcribed into positive- strand RNA. Because the positive-sense RNA genome is used for both viral translation and replication, flavivirus must be able to regulate whether the genome is used either as a transcript for viral protein synthesis or as a template for RNA synthesis. This decision tree depends on a reorganization of the genome from a linear to a circular form, which promotes RNA synthesis. Viral polymerase NS5 then recognizes an RNA promoter at the 5’-end of the circularized genome, called stem-loop A (SLA), and translocates to the 3’ terminus to initiate RNA synthesis. However, the mechanism by which NS5 selectively recognizes the circular genome for negative- strand RNA synthesis is not well understood. In particular, the viral genome has differently predicted 3’ terminal stem-loop (3’SL) structures in the linear and circular genome, yet how these differences relate to negative strand synthesis is not known. The goal of this project is thus to understand the molecular mechanism of negative- strand RNA synthesis. In aim 1, we will determine how the structural changes of 3’SL in the linear and circular forms of the viral genome modulate NS5 interaction and polymerase activity, and whether NS5 recognizes both 5’ SLA and 3’SL structures simultaneously. In aim 2, we will determine the structures of 3’SL in the linear and circular genomes using a tRNA-scaffold approach, and also characterize the NS5 initiation complex assembled on dengue virus mini-genome. The change from linear to circular genome is conserved in all flaviviruses. Thus, the combined structural, biochemical, and virological studies will help elucidate the mechanism for negative- strand RNA synthesis in flavivirus.