Riboswitches & Regulation: An Introduction

Sarrah Rose
8 min readJul 3, 2021


Photo by Myriam Zilles on Unsplash

All living organisms possess the ability to take in environment stimuli, thereby converting these signals into a cellular response. Oftentimes, these responses are mediated by transcription factors that bind to DNA & control the activity of RNA polymerase, or of proteins which elicit allosteric effects of their regulatory targets.

In the 1970s however, researchers began to realise the role of regions of mRNA transcripts in regulating the production of downstream products. One key example of this were Riboswitches — mRNA elements that bind & target small molecules, undergoing conformational changes in the process, in order to regulate mRNA expression.

Basic Structure & Function

What kinds of molecules? Well, scientists have found that riboswitches have the capacity to sense a variety of ligands from magnesium ions, to acid precursors and amino acid residues! Crucially, this makes it an extremely flexible & adaptive tool in gene regulation.

It’s also worth noting that you’ll usually find these riboswitches located in the 5’ untranslated region of many bacteria mRNAs — a segment of the transcript that precedes the translation initiation site, enabling it to regulate signals for transcription attenuation (i.e. the premature termination of transcription) or translation initiation.

Now that we’ve gotten the basics down, let’s consider the structure of the riboswitch, made up of:

  1. The aptamer domain
  2. The expression platform

The aptamer domain functions as a typical receptor, recognising the ligand and binding to it with high specificity & affinity. In fact, aptamers have a level of structural complexity that approaches that of proteins. For instance, the receptor of a guanine-binding riboswitch from Bacillus subtilis forms a three-dimensional (3D) structure in which the ligand is almost completely enveloped.

Meanwhile, the expression platform directly regulated gene expression (transcription termination or translation initiation) by interpreting the binding state of the aptamer. Fundamentally, this means that in response to the presence (or absence) of ligand binding, the expression platform will toggle between 2 different 3D-conformations.

Wait a moment…how does the expression platform actually tell that the aptamer bound to a ligand? Simple — a switching sequence, that functions intuitively, like a switch. For example, if the ligand binding to the aptamer, the switching sequence is incorporated into the aptamer domain, causing the expression platform to fold into a specific secondary structure. So you can basically think of this as turning the switch “on”, and the opposite situation as turning the switch “off”.


To achieve this great specificity of aptamers we were talking about, we go through an iterative discovery process called SELEX: Systematic Evolution of ligands by exponential enrichment.

That was a mouthful. But basically, it’s a producer used to produce single-stranded RNA (aptamers) that can specifically bind to a target ligand or ligands.

  1. The process beings with the initial library — a RNA pool consisting of a huge library of different RNA species — produced through the random generation of oligonucleotide sequences of a fixed length
  2. These sequences are then exposed to the target ligand, with “selective pressures” applied. Mimicking the process of natural selection, sequences which fail to bind to the target are removed, typically through affinity chromatography or target capture on paramagnetic beads
  3. Successful sequences are then eluted (essentially getting them to release the target ligand) & amplified by PCR to prepare for subsequent rounds of selection
  4. The process is repeated with increasingly stringent selection conditions to identify the binding sequences of the highest affinity to the target ligand

Conformational Changes & Function

As you’ve probably grasped by now, the conformational changes undertaken by these RNAs are crucial to the structure it plays in regulating gene expression. Let’s look at a couple examples.

A. Regulating Translation

In this scenario, researchers have integrated an aptamer into the 5’ UTR of a eukaryotic mRNA. When a ligand is introduced to the system, the aptamer shifts from its slightly-structured state into the defined tertiary structure. In this conformation, the aptamer stops the scanning ribosome, preventing it from recognising the START codon, rapidly decreasing protein expression levels.

An experiment conducted by researchers actually showed that when a protein — NOP14, involved in ribosome genesis, is brought under the control of a tetracycline aptamer; in the presence of tetracycline ligands they successfully inhibited the growth of tetracycline.

In fact, if we place 3 riboswitches adjacent to each other, researchers found that you completely inhibit the production of these enzymes altogether! This has really cool applications in therapeutics, which we’ll cover shortly.

B. Controlling splicing of pre-mRNA molecule

Splicing is an important mechanism to process mRNA strands, by removing introns (non-coding regions) and joining exons (coding regions) together. This is crucial because without its removal by the spliceosome (a ribonucleoprotein that removes introns), an mRNA with extra “junk” will be made, producing the wrong protein in translation.

Researchers therefore exploited this key mechanism by integrating the aptamer into the 5’ splice site. If a ligand is present, it will bind to the aptamer, causing a tightening of the overall structure of the riboswitch. As a result, the 5’ splice site is no longer accessible to the spliceosome, resulting in the expression of a defective protein, inhibiting gene expression.

C. Degrading mRNA with ribozymes

Put simply, ribozymes are RNA molecules that have the ability to catalyse specific biochemical reactions, including self-cleaving mRNA strands. Following the cleavage, the mRNA is devoid of a phosphate group at the 5’ end, making it susceptible to degradation by cellular RNAse.

Critically, by adding ligands to the system which bind to the aptamers, researchers were able to induce a conformational change in the structure, turning the stem of the aptamer “rigid”. With this change, the specific cleavage sequence was no longer accessible to the ribozyme, preventing mRNA degradation.

Applications of Riboswitches as Antimicrobial Drug Targets

Researches have long explored the potential of alternative “antibiotics”, particularly with the rapid rise of antibiotic-resistance by most bacteria. Essentially, researchers design ligands that mimic natural metabolites, directing them to bind to specific aptamers in riboswitches. This works because the expression of a number of genes crucial to their survival & virulence (e.g. metabolite biosynthesis / transport) in most bacteria are regulated by these riboswitches. Additionally, given the highly specific nature of ligand-aptamer interactions, we could potentially design riboswitch-targeting compounds that are highly selective and avoid binding to other cellular targets, minimising unintended cellular responses.

This also makes sense because of what we already see happening in nature. Several known natural antibacterial compounds currently function by targeting riboswitches. For example, Pyrithiamine is phosphorylated to form pyrithiamine pyrophosphate, where it’s able to bind to several TPP riboswitches in bacteria, repressing the expression of a reporter gene and ultimately inhibiting bacterial/fungal growth!

How exactly do we begin designing these antibacterial compounds?

One answer is through rational design — creating new antibacterial compound, based on the ability to predict how it’s structure will affect function through physical models. And the cool thing is, it’s an incredibly virtuous cycle! As more 3-D structures of riboswitches are reported, additional avenues for rational design become available for the discovery of more antibacterial compounds.

One major challenge of this method though is how hard it is to couple structure to function. Physical properties of riboswitches, such as conformational flexibility, near total ligand encapsulation & chemical differences in RNA chemistry make it challenging for us to rationally design these compounds.

Another potential answer is high-throughput screening — the automated testing of massive numbers of biological compounds for a specific biological target. For example, researchers have developed a high-throughput assay to screen for other compounds that could activate the self-cleavage of this riboswitch. They do this by measuring properties such as fluorescence polarisation and fluorescence resonance energy transfer (FRET), in order to validate the cleavage of by these riboswitches.


Even still, the process of drug discovery is never that easy. Some considerations:

  1. A principal question that remains is the degree to which a riboswitch-targeting compound can repress gene expression.
  2. Additionally, even in cases where the synthesis of certain metabolites can be completely repressed by riboswitch targeting, the bacteria might still grow if the metabolite can be imported from the host tissue
  3. Therein also lies the issue of toxicity, because it’s possible that compounds which target riboswitches in bacteria may result in similar effects in their human host (which is bad, because it could cause the under-expression of important metabolites!)
  4. Researchers are also wary of the potential for bacterial resistance to these compounds. A common mechanism by which bacteria becomes resistant to antibiotics is by expressing a protein that modifies / removes the drug. Certain riboswitch-targeting compounds could be susceptible to this mechanism, given that many bacteria already express proteins that act on natural metabolites.
  5. There’s also a concern that bacteria could invoke a mutation that disrupts ligand-binding to a receptor; although it’s still really unclear how this mutation will affect other cellular responses in the cell

Ultimately, these are all important concerns to take into account — but ones that are not impossible to overcome (e.g. designing compounds that are chemically dissimilar to the natural metabolite, minimising susceptibility to bacterial resistance)

I think that (in the near term), it’s unlikely that we’re going to chance upon a mechanism that solves all our pathogenic problems; but the potential success of different mechanisms lends some assurance that someday we’ll possess the suite of options that truly allows for us to do so.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Hey, I’m Sarrah Rose! A 17 year old deeply passionate in utilising Synthetic Biology & Artificial Intelligence to solve major problems in the world today. If you enjoyed this article or would just like to chat, I’d love to hear from you:

email: sarrahrose04@gmail.com || twitter || Linkedin