High-Throughput Quantification of In Vivo Adeno-Associated Virus Transduction with Barcoded Non-Coding RNAs.

Abstract

Recombinant adeno-associated viruses (rAAVs) have become favorable gene delivery vehicles for expressing therapeutic transgenes. Capsid engineering efforts to produce novel AAVs with improved transduction efficiencies, unique tissue specificities, and reduced host immunities are a direct response to the high demand for treatment needs that preexisting rAAVs cannot currently fulfill. New AAV capsids discovered by directed evolution methods, in silico design, or from natural proviral sequences ultimately require extensive characterization in relevant in vivo models. Consequently, quantitative screening of candidate capsid libraries now requires reliable high-throughput sequencing approaches. In this study, we have developed a vector/transgene tracking system that employs the indexing of a non-coding RNA. Specifically, a barcoded Tough Decoy (bcTuD) that express highly stable RNA transcripts that can be used as readouts for transduction efficiency. The pseudo-hairpin structure of the bcTuD contains a variable region that is amenable to barcode insertion, which can be detected by target amplicon sequencing. The described approach, named AAV-bcTuD screening, offers a new alternative for in vivo assessment of rAAV that can accurately quantify vector genomes and transcript abundances in tissues, as exampled by the demonstration in liver and brain infections. Proof-of-concept is provided to show that vector genome and transcript detection in tissues with this method is accurate and consistent for a vector dose range of upwards to four logs in a mixed vector injection, showing that this technique is robust, sensitive, and applicable for multiplexed screening of capsid performance in vivo.