Background

RNA is the central molecule in gene expression in all extant life, serving as the messenger. It is also central to biocatalysis, playing a key role in ribosome assembly and function. Ultimately, RNA function is regulated by its structure and there is a seemingly limitless variety of structures that allows for diverse functions. RNA molecules can form both inter- and intramolecular Watson-Crick and Hoogsteen base pairs, which lead to the formation of secondary structures such as hairpins, apical and internal loops, 3-way-junctions or pseudoknots (Cruz and Westhof, 2009). Such secondary structures can impart catalytic activity, serve as molecular scaffolds as well as provide or occlude binding sites for RNA-binding proteins (RBPs) and other RNAs. Secondary structures on messenger RNA are known to affect a plethora of biological processes such as RNA stability and processing (Soemedi et al., 2017), alternative splicing (Bartys et al., 2019), subcellular localisation (Lazzaretti and Bono, 2017) and translation regulation (Jacobs et al., 2012).

Whilst RNA molecules have the capacity to adopt a multitude of distinct conformations, the most thermodynamically and kinetically favourable structures will be populated under certain conditions. As a consequence, RNA secondary structures can be highly dynamic and interconvert between different structural arrangements as the environment changes (Ganser et al., 2019; Mustoe et al., 2014). For example, binding of RBPs can induce or facilitate structural rearrangement, while riboswitches change their conformation and exert regulatory function upon binding of small ligands. Finally, physical parameters such as temperature and osmolarity strongly affect the distribution and the stability of RNA secondary structures and thereby promote structural transitions. Given the biological relevance of RNA secondary structure formation, several biochemical and biophysical strategies have been developed to monitor RNA structural changes in real-time, including Circular Dichroism, NMR, DMS-footprinting, and Förster Resonance Energy Transfer (FRET).

As RNA secondary structures are dependent on their thermodynamic free energies, temperature is one of the most direct parameters to control such structures for translation regulation. Thermosensitive RNA secondary structures – referred to as RNA thermometers – have been harnessed by multiple organisms to fine-tune translation in a temperature-dependent manner (Kortmann and Narberhaus, 2012; Krajewski and Narberhaus, 2014). A classic example of an RNA thermometer is found in Listeria, where a secondary structure in the prfA transcript controls virulence in response to temperature: At temperatures up to 30 °C, a hairpin structure forms and prevents access to the ribosome binding site; this structure partially melts when temperature rises to 37 °C (the body temperature of the human host), exposing the ribosome binding site and allowing translation to occur (Johansson et al., 2002). Many prokaryotes employ such zipper-like RNA thermometers as a switch to control translation. However, increased temperature may not necessarily lead to a complete loss of secondary structure, but favour the adoption of alternative conformations instead. We recently identified several eukaryotic RNA thermoswitches in the model flowering plant Arabidopsis thaliana, which are activated at warm temperatures (27-32 °C) by adopting a more relaxed hairpin conformation to enhance translation (Chung et al., 2020).

These examples emphasise the importance of investigating RNA structural dynamics in order to rationalize their biological functions. Chemicals or enzymes that react selectively with single or double stranded RNA can be employed to probe secondary structure both in vitro and in vivo (Mailler et al., 2019). However, these approaches are irreversible, leaving the RNA molecules permanently modified and preventing real-time observation of the same molecules’ dynamics over a range of conditions. Several biophysical methods can be used to observe RNA secondary structure as well, but come with their own set of limitations: besides significant requirement of RNA input, nuclear magnetic resonance (NMR) spectroscopy and X-ray crystallography preclude observation under physiological conditions, while cryo electron microscopy still suffers from limited resolution and is thus currently restricted to larger structures (Mailler et al., 2019). In our characterisation of plant RNA thermoswitches, we employed a combination of Förster Resonance Energy Transfer (FRET) measurements and circular dichroism (CD) spectroscopy to observe RNA hairpin dynamics over a range of temperatures in real-time. A detailed protocol of these methods is outlined below.