A genomic selection strategy to identify accessible and dimerization blocking targets in the 5′‐UTR of HIV‐1 RNA

Defining target sites for antisense oligonucleotides in highly structured RNA is a non‐trivial exercise that has received much attention. Here we describe a novel and simple method to generate a library composed of all 20mer oligoribonucleotides that are sense‐ and antisense to any given sequence or genome and apply the method to the highly structured HIV‐1 leader RNA. Oligoribonucleotides that interact strongly with folded HIV‐1 RNA and potentially inhibit its dimerization were identified through iterative rounds of affinity selection by native gel electrophoresis. We identified five distinct regions in the HIV‐1 RNA that were particularly prone to antisense annealing and a structural comparison between these sites suggested that the 3′‐end of the antisense RNA preferentially interacts with single‐stranded loops in the target RNA, whereas the 5′‐end binds within double‐stranded regions. The selected RNA species and corresponding DNA oligonucleotides were assayed for HIV‐1 RNA binding, ability to block reverse transcription and/or potential to interfere with dimerization. All the selected oligonucleotides bound rapidly and strongly to the HIV‐1 leader RNA in vitro and one oligonucleotide was capable of disrupting RNA dimers efficiently. The library selection methodology we describe here is rapid, inexpensive and generally applicable to any other RNA or RNP complex. The length of the oligonucleotide in the library is similar to antisense molecules generally applied in vivo and therefore likely to define targets relevant for HIV‐1 therapy.