SHAPE Selection (SHAPES) enrich for RNA structure signal in SHAPE sequencing-based probing data.

Line Dahl Poulsen,Sofie R Salama,Anders Krogh,Jeppe Vinther,Lukasz Jan Kielpinski

doi:10.1261/rna.047068.114

Abstract

Selective 2′ Hydroxyl Acylation analyzed by Primer Extension (SHAPE) is an accurate method for probing of RNA secondary structure. In existing SHAPE methods, the SHAPE probing signal is normalized to a no-reagent control to correct for the background caused by premature termination of the reverse transcriptase. Here, we introduce a SHAPE Selection (SHAPES) reagent, N-propanone isatoic anhydride (NPIA), which retains the ability of SHAPE reagents to accurately probe RNA structure, but also allows covalent coupling between the SHAPES reagent and a biotin molecule. We demonstrate that SHAPES-based selection of cDNA–RNA hybrids on streptavidin beads effectively removes the large majority of background signal present in SHAPE probing data and that sequencing-based SHAPES data contain the same amount of RNA structure data as regular sequencing-based SHAPE data obtained through normalization to a no-reagent control. Moreover, the selection efficiently enriches for probed RNAs, suggesting that the SHAPES strategy will be useful for applications with high-background and low-probing signal such as in vivo RNA structure probing.

Highlights

Under physiological conditions, RNA has the ability to form structures through internal base-pairing
We demonstrate that RNase I treatment of cDNA/RNA– N-propanone isatoic anhydride (NPIA)–biotin hybrids followed by selection on streptavidin beads enriched for probed RNAs and effectively removes the large majority of background signal present in Selective 2′-Hydroxyl Acylation analyzed by Primer Extension (SHAPE) probing data
We obtained N-propanone isatoic anhydride (NPIA), which is identical to the canonical SHAPE reagent N-methyl isatoic anhydride (NMIA), except that the N-methyl group has been exchanged with an N-propanone group (Fig. 1A)

Summary

Introduction

RNA has the ability to form structures through internal base-pairing. This additional layer of information encoded in the RNA sequence will in many cases be key to understanding the function of RNA molecules. This is true for the abundant noncoding RNAs, and for protein-coding mRNAs, which often contain functional regulatory RNA structures. A successful approach to improve RNA secondary structure prediction has been to use experimental probing data to guide the computational predictions. It has been shown that data from Selective 2′-Hydroxyl Acylation analyzed by Primer Extension (SHAPE) experiments significantly improve secondary RNA structure prediction (Deigan et al 2009; Weeks and Mauger 2011)

Methods

Results

Conclusion