Abstract A34: Artificial multi-target microRNAs: A new RNA interference approach to enable simultaneous suppression of multiple genes

Abstract

Abstract RNA interference (RNAi) has become a powerful tool for the suppression of gene expression and the identification of therapeutic targets. RNAi screens have revealed synthetic lethal cancer vulnerabilities, and progress has been made toward enabling therapeutic delivery of short interfering RNAs (siRNAs) in vivo. However, current RNAi libraries are designed to detect synthetic lethal interactions between individual genes and an oncogenic lesion. They are less effective for detecting more subtle interactions in which a synthetic lethal phenotype requires the combined inhibition of several genes in addition to the sensitizing mutation. Identifying such interactions, where multiple genes cooperate with an oncogenic lesion, may open up new avenues for therapeutic intervention, but their discovery remains difficult with current approaches. We therefore sought to develop a new RNAi system in which a single short RNA sequence can suppress multiple genes simultaneously. As the basis for our design, we mimicked the activity of endogenous microRNAs (miRNAs), and refer to our system as “artificial miRNAs.” In contrast to traditional siRNAs and short hairpin RNAs (shRNAs) that are perfectly complementary to their target, miRNAs utilize partial complementarity to direct the repression of multiple genes at the same time. By screening a library of artificial miRNAs, we expect to reveal a novel class of synthetic lethal interactions that can be exploited for cancer therapy. Moreover, in vivo delivery of artificial miRNAs may enable combined inhibition of multiple therapeutic targets that cannot currently be targeted by small molecules. We developed an algorithm to generate artificial miRNA sequences targeting a given set of genes of interest. Based on the mechanisms of endogenous miRNA targeting, our algorithm designs artificial miRNAs that feature perfect complementarity between the miRNA 5'-end “seed region” and each of the target sequences as well as partial complementarity between the miRNA 3' end and each target. To test the hypothesis that miRNAs can be rationally designed for the simultaneous knockdown of multiple genes, we designed and synthesized 234 artificial miRNAs to target two metabolic genes implicated in glioblastoma: pyruvate carboxylase (PC) and glutaminase (GLS). In addition, we generated 74 non-targeting control miRNAs. Using luciferase reporter assays, we observed that PC/GLS-targeted artificial miRNAs were significantly more effective at suppressing both genes than non-targeting controls. We then confirmed by immunoblotting that 13 artificial miRNAs knockdown endogenous PC and GLS simultaneously. Thus, we demonstrated that artificial miRNAs are effective for multi-gene RNAi. We also found that artificial miRNA activity was predicted by the miRNA:target binding energy and the degree of complementarity at the 3' end of the miRNA. Therefore, we are refining the design algorithm to generate artificial miRNAs with enhanced activity. Taken together, our results validate the artificial miRNA system as an effective RNAi-based approach for multiple gene suppression. We are currently developing a library of ~16,000 artificial miRNAs that will permit functional screens to identify combinatorial synthetic lethal interactions. We expect that multi-target artificial miRNAs will provide an important tool for the identification of new therapeutic targets. Citation Format: Jason D. Arroyo, Emily N. Gallichotte, Muneesh Tewari. Artificial multi-target microRNAs: A new RNA interference approach to enable simultaneous suppression of multiple genes. [abstract]. In: Proceedings of the AACR Precision Medicine Series: Synthetic Lethal Approaches to Cancer Vulnerabilities; May 17-20, 2013; Bellevue, WA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(5 Suppl):Abstract nr A34.