Genome-wide identification of natural RNA aptamers in prokaryotes and eukaryotes

Sidika Tapsin,Lian Hui Zhang,Teodorus Theo Susanto,Miao Sun,Yue Wan,Alexander Lezhava,Xin Ni Lim,Gui Sheng Zeng,Niranjan Nagarajan,Huibin Zhang,Yue Wang,Huimin Zhao,Yang Shen,Siwy Ling Yang,Jasmine Lee,Ee Lui Ang

doi:10.1038/s41467-018-03675-1

Abstract

RNAs are well-suited to act as cellular sensors that detect and respond to metabolite changes in the environment, due to their ability to fold into complex structures. Here, we introduce a genome-wide strategy called PARCEL that experimentally identifies RNA aptamers in vitro, in a high-throughput manner. By applying PARCEL to a collection of prokaryotic and eukaryotic organisms, we have revealed 58 new RNA aptamers to three key metabolites, greatly expanding the list of natural RNA aptamers. The newly identified RNA aptamers exhibit significant sequence conservation, are highly structured and show an unexpected prevalence in coding regions. We identified a prokaryotic precursor tmRNA that binds vitamin B2 (FMN) to facilitate its maturation, as well as eukaryotic mRNAs that bind and respond to FMN, suggesting FMN as the second RNA-binding ligand to affect eukaryotic expression. PARCEL results show that RNA-based sensing and gene regulation is more widespread than previously appreciated in different organisms.

Highlights

RNAs are well-suited to act as cellular sensors that detect and respond to metabolite changes in the environment, due to their ability to fold into complex structures
The known thiamine pyrophosphate (TPP) and S-adenosylmethionine (SAM) riboswitches were used as positive controls and other RNA sequences not known to bind TPP or SAM were used as negative controls in this experiment[9,10]
We identified 17 out of 20 known riboswitches that interact with key metabolites (TPP, FMN, and SAM) in B. subtilis and P. aeruginosa, including 4/5 known TPP riboswitches, 2/2 FMN riboswitches, 9/11 SAM riboswitches in B. subtilis, as well as 1/1 TPP and 1/1 FMN riboswitches in P. aeruginosa[11], highlighting the high sensitivity of the method (85%; Fig. 1c, Supplementary Fig. 5a-d)

Summary

Introduction

RNAs are well-suited to act as cellular sensors that detect and respond to metabolite changes in the environment, due to their ability to fold into complex structures. One class of cellular RNA sensors, known as riboswitches, can recognize and respond to specific metabolites by altering gene expression[2] Upon binding to their ligands, riboswitches undergo conformational changes that result in the regulation of gene expression through diverse means, such as changes in transcription, translation and decay[2], informing the host of its environmental conditions. Subtle differences between these two libraries point to the few true ligand-specific structure changing RNA elements in the genome and we developed a sensitive and robust computational analysis pipeline to identify these (Methods) This experimental and computational approach, termed Parallel Analysis of RNA Conformations Exposed to Ligand binding (PARCEL), revealed the breadth of RNA-ligand interactions in prokaryotic and eukaryotic transcriptomes, identifying many new natural RNA aptamers in the process

Methods

Results

Conclusion