Alexander A. Kolykhalov

Safe, long-term hepatic expression of anti-HCV shRNA in a nonhuman primate model.

The hepatitis C virus (HCV) chronically infects 2% of the world population and effective treatment is limited by long duration and significant side-effects. Here, we describe a novel drug, intended as a “single-shot ” therapy, which expresses three short hairpin RNAs (shRNAs) that simultaneously target multiple conserved regions of the HCV genome as confirmed in vitro by knockdown of an HCV replicon system. Using a recombinant adeno-associated virus (AAV) serotype 8 vector for delivery, comprehensive transduction of hepatocytes was achieved in vivo in a nonhuman primate (NHP) model following a single intravenous injection. However, dose ranging studies performed in 13 NHP resulted in high-expression levels of shRNA from wild-type (wt) Pol III promoters and dose-dependent hepatocellular toxicity, the first demonstration of shRNA-related toxicity in primates, establishing that the hepatotoxicity arises from highly conserved features of the RNA interference (RNAi) pathway. In the second generation drug, each promoter was re-engineered to reduce shRNA transcription to levels that circumvent toxicity but still inhibit replicon activity. In vivo testing of this modified construct in 18 NHPs showed conservation of hepatocyte transduction but complete elimination of hepatotoxicity, even with sustained shRNA expression for 50 days. These data support progression to a clinical study for treatment of HCV infection.

997. Loop Composition as Well as Orientation of the Guide Strand in shRNA May Dramatically Modulate Activity of the shRNA: Effect on the Development of RNAi-Based Inhibitors

RNA interference (RNAi) is a new promising technology for therapeutic drug development. It can be initiated either by chemically synthesized double-stranded RNA delivered to a cell, or by transcription of a short hairpin RNA in the cell nucleus from a DNA template. It has been realized that many RNA sequences although |[ldquo]|active|[rdquo]| as siRNAs, are often inactive as shRNAs. The composition of the loop sequence in shRNA has been suggested to play a role in determining its activity by affecting shRNA transport from the cell nucleus to the cytoplasm (Kawasaki and Taira, Nucleic Acids Res. 2003, 31(2):700-7). In the present study we identified several siRNA sequences that efficiently inhibit HCV replication in the replicon system. In subsequent attempts to generate active shRNAs spanning these validated sequences, we tested the effect of two different loop sequences, as well as the guide strand orientation on the shRNA activity. Functional tests were performed in the HCV replicon systems, as well as in a model luciferase-HCV fusion systems. We demonstrate that the context of the loop within the shRNA has a dramatic effect on shRNA activity, and a loop that is |[ldquo]|inactive|[rdquo]| in one shRNA can be fully active in the context of another shRNA. We also demonstrate that the orientation of the guide strand in the shRNAs (sense vs antisense) may dramatically influence the shRNA activity. Changes in the extent and precision of the shRNA processing by Dicer will be presented. The effect of these findings on the development of ddRNAi-based inhibitors will be discussed.