Use of the siRNA-triggered RNAi pathway to study gene function has become common in molecular biology and holds promise for the development of new therapeutics. However, specific issues exist that limit the application of this promising technology. These issues include off-target effects that arise from the ability of siRNAs to interact with proteins of the innate immune system and the ability of an siRNA guide strand to function as a miRNA. In addition, antisense oligonucleotides targeting miRNAs (anti-miRs) have significant therapeutic potential. In this competitive renewal of an R01 project, we propose to use tools from nucleic acid chemistry, molecular and structural biology to advance our understanding of siRNA off-target effects and provide new modifications to siRNAs and anti-miRs to improve potency and selectivity. Because the guide strand of an siRNA can function as a miRNA, siRNAs often repress the expression of gene targets that have mismatches in the guide-target duplex outside the seed region (nucleotides 2-8 of the guide). A challenge in the field of siRNA research is to develop guide strand modifications that will lead to more specific targeting (i.e. reduced miRNA-like off target effects). Here we will use a combination of x-ray crystallography (in collaboration with Prof. Ian MacRae, Scripps, La Jolla), computational screening, RNA modification, hAgo2 binding studies and RNAi experiments to establish the basis for selective targeting by hAgo2-guide complexes and discover new nucleoside analogs that reduce miRNA-like effects. Regulation of innate immune responses by binding Toll-like receptors (TLRs) is another important siRNA off target effect to be addressed in this project. Sequence motifs found in the component strands of certain siRNAs are known to activate TLRs and induce an immune response in human cells. Modification of riboses in the backbone of siRNA (e.g. 2’-OMe, 2’-F, LNA), can reduce TLR-mediated immune stimulation. However, the modified oligonucleotides typically function as TLR antagonists indicating they maintain TLR-binding activity. Here we propose to develop guanosine analogs that are truly “immune silent” i.e. lacking TLR-binding activity such that the modified oligonucleotides are neither receptor agonists nor antagonists. Two different approaches are proposed to achieve this goal. Inhibition of miRNA function is also a promising therapeutic strategy since increased expression of specific miRNAs has been linked to human disease. Antisense oligonucleotides that bind miRNAs by Watson-Crick pairing (anti-miRs) are candidates as miRNA inhibitors. Since the functional fraction of a miRNA is loaded into an Argonaute (Ago) protein, modifications of the anti-miR that enhance binding to Ago are likely to enhance anti-miR activity and selectivity. Crystal structures of hAgo2 bound to miRNA-target complexes will be used to inform the design of new anti-miR modifications. We will modify two different nucleotide positio...