Defining the Rules for Designing Fully Chemically Modified siRNAs to Treat Genetically Linked Central Nervous System Disorders

点击翻译按钮获取中文摘要

PROJECT SUMMARY Small interfering RNA (siRNA) therapeutics specifically and potently block the expression of disease-related genes. siRNA clinical utility is currently limited to disease targets in the liver, but the Khvorova lab has developed a novel, fully chemically modified siRNA platform that enables delivery to the central nervous system (CNS) and results in potent modulation of gene expression in mouse and monkey brain for 6 months. This technology provides an opportunity to treat genetically-defined neurological disorders, including Huntington’s disease, amyotrophic lateral sclerosis, and Alzheimer’s disease (AD). Extensive chemical modification protects siRNAs from degradation and is essential for in vivo delivery, but lowers the gene silencing efficacy of many siRNA sequences. Bioinformatics algorithms have been developed to predict the activity of non-modified siRNAs, but these algorithms cannot predict whether a chemically modified siRNA will be functional. Identifying a hyperfunctional siRNA chemical modification pattern and developing a predictive algorithm for modified siRNA sequences will be critical for the widespread application of this platform in vivo to treat AD and other genetically-linked neurological disorders. The goal of this proposal is to identify parameters that impact the silencing efficacy of chemically modified siRNAs. Aim 1 will identify the step(s) that limit the activity of chemically modified siRNAs. Using a validated AGO2- immunoprecipitation technique and a small RNA high-throughput sequencing protocol, loading of 32 chemically modified siRNA sequences, each with 3 different chemical modification patterns, into RNA-induced silencing complex (RISC) will be quantified, and these results will be compared to their in vitro silencing activity. These efforts will provide a data set to determine the impacts of chemistry and sequence on the different steps of RISC function. Aim 2 will design and screen 192 siRNAs in 6 different chemical modification patterns (i.e., for a total of 1152 siRNAs) to systematically assess if and how changes to siRNA sequence and chemical modification patterns impact siRNA efficacy. These 1152 siRNAs will target 4 different mRNAs identified as therapeutic targets for AD. Completing this aim will identify siRNAs that effectively reduce the expression of AD targets. Using multiple bioinformatics analysis methods, including multi-parameter linear regression, support vector machine, and random forest, Aim 3 will model algorithms that specifically predict functional, chemically modified siRNAs. The best performing algorithm will be determined by independent and cross-validations. Completion of this project will decrease the extent of in vitro screening needed to identify functional chemically modified siRNAs capable of targeting disease genes in the CNS and streamline the design of siRNA therapies to treat genetically-defined neurological disorders.