Engineering active siRNA therapeutics

Abstract

The discovery of RNA interference (RNAi) has led to a rush to develop short interfering RNAs (siRNAs) for a variety of applications, especially for therapeutics. A particular advantage of RNAi-based therapeutics is that they utilize the existing regulatory machinery of gene expression in eukaryotic cells. For this reason, siRNAs have the potential to modulate cell function with exquisite specificity and limited side effects. Nonetheless, the development of siRNAs as therapeutics will require continued research focused on the critical bottlenecks in the process. To initiate RNAi in vivo, siRNAs are first encapsulated into a delivery vehicle that is typically cationic, to balance charge, and either lipid-like or polymeric in nature. As with other drugs, the vehicle–siRNA complexes are either administered directly to the site of interest (e.g. by injection) to minimize trafficking to nontargeted tissues and maximize local concentration of the drug, or delivered through the bloodstream (intravenous) or gastrointestinal system (oral) to simplify administration. Once delivered to the tissue of interest, the vehicle–siRNA complexes must reach the target cells and cross the plasma membrane to enter the cytoplasm. At this point, the siRNA must be released from the vehicle in order to be recognized by the RNAi machinery. In humans, the primary proteins of the RNAi machinery are Dicer, TAR RNA-binding protein (TRBP) and Argonaute 2 (Ago2). Together with the siRNA, these three proteins form the RNA-induced silencing complex (RISC) and its precursor, the RISC loading complex (RLC). The three proteins bind the siRNA and select which strand of the double-stranded siRNA will become the single-stranded guide strand in RISC. Activated RISC, possessing the single-stranded guide strand, then specifically cleaves an mRNA that contains a sequence complementary to the guide strand. The reduction in mRNA concentration leads to a consequential reduction in protein concentration and the corresponding therapeutic effect. However, many other proteins also recognize double-stranded RNAs (dsRNAs), and many of these proteins are involved in innate immunity and the cellular response to viral infections. Although shorter dsRNAs, like siRNAs, are less immunostimulatory than longer dsRNAs, siRNA therapeutic design must focus on avoiding nonspecific recognition by either cell-surface or cytoplasmic proteins. Understanding how to control potential interactions with nonpathway proteins will greatly expand the therapeutic window for siRNA-based therapeutics. The focus of this minireview series is on the means by which siRNA therapeutics can be generated with the greatest activity and specificity. The three articles focus on different steps in this process: (a) selection of the siRNA sequence most likely to have high activity, (b) in vivo delivery of that siRNA to the tissue and cells of interest, and (c) avoiding side effects associated with the administration of foreign nucleic acids. Current approaches and future challenges are discussed.