How to Fine-Tune a Riboswitch
by Janie
It has the inklings of a lame biology joke, but in actuality has profound implications among microbial communities: What's the similarity between male moths and RNA? The answer is the ability to detect extremely specific compounds. Like male moths that can detect a single pheromone molecule released by a female from over 10 kilometers away, riboswitches detect select molecules and ions, some at concentrations even down to the picomolar range. This is incredibly sensitive, considering the rule of thumb that in E. coli a concentration of 1 nanomolar roughly corresponds to one molecule per cell! The riboswitch aptamer binds its ligand, thereby inducing a shift in structure in the expression platform to trigger a genetic response. Riboswitch and ligand pairs run the whole chemical gamut: amino acids, metal cations, coenzymes like TPP and SAM, the alarmone ppGpp, and vitamin cofactors like thiamin, riboflavin, vitamin B12 …
Fig. 1. (From Figure 1 from the paper) (A) The structure of cobalamin/B12. (B) The different corrinoids used in this study. Frontispiece: X‑ray crystal structure of a cobalamin riboswitch, depicted with base-on cobalamin alignments. Source
That last one is particularly interesting. Vitamin B12 (aka cobalamin) is a member of the corrinoid enzyme cofactors, which consist of a cobalt ion nestled within a four-pyrrole ring. Computationally identified cobalamin riboswitches are a dime a dozen in bacterial genomes, manifesting as structural variants that regulate a whole spectrum of functions, from cobalamin metabolism and transport to less obvious roles like antibiotic synthesis. But as shown in Fig. 1, the "lower ligand" varies widely amongst natural corrinoids. These variants are also used by cobalamin-dependent enzymes. So, how are riboswitches – with a reputation for being so persnickety about ligand specificity – to contend with all the variety among cobalamin's corrinoid kith and kin?
A preprint from Michi Taga's lab at UC Berkeley explores this ribonucleic terra incognita, in work led by Kristopher Kennedy. First, they hooked up 38 different cobalamin riboswitches from 12 different bacterial species to a GFP reporter system in Bacillus subtilis. The riboswitches covered a roomy sample space derived from a range of Gram-positive and Gram-negative bacteria across Firmicutes and Gammaproteobacteria (now also known as Bacillota and Pseudomonadota) including corrinoid-producers. They then exposed these strains to varying concentrations of four different corrinoids purified from bacterial cultures. The cobalamin riboswitches were all responsive to multiple corrinoids – not so picky, unlike most other riboswitches (Fig. 2).
Fig. 2. (From Figure 3 from the paper) (A,B) Corrinoid dose-responses of two riboswitch reporter strains with distinct corrinoid specificities, representing the riboswitches that responded to some but not all corrinoids (A) and those that responded to all four corrinoids (B). (C‑E) Pairwise comparisons of GFP fold repression induced by 100 nM of different corrinoids. Source
To parse how this relaxed specificity might be playing out at a more fine-grained molecular level, the authors then produced chimeric riboswitches by mixing and matching pieces of their riboswitches. Perhaps corrinoid specificity can be traced to certain lengths of RNA? Not so. Swapping in part of a riboswitch that responded to all four corrinoids sometimes gave the new Franken-switch the same broad corrinoid selectivity, but not in other cases. Trying to assign responsibility to any one riboswitch part didn't lead to broadly generalizable conclusions. Corrinoid selectivity here isn't so simple.
Shifting focus from the riboswitch towards the corrinoids themselves proved a more fruitful angle. Many corrinoids flip-flop between two isomeric forms: "base-on," in which the corrinoid structure is stiffly constrained, since a nitrogen in the lower ligand is coordinated to the cobalt (as in Fig. 1), or "base-off," in which the lower ligand tail can swing around because the coordination is absent. Exposing the riboswitches to a panel of corrinoids with different base-on/base-off equilibrium ratios led to the discovery that this is what the riboswitches are attuned to (Fig. 3). It appears cobalamin riboswitches are sensitive to the shape of their ligands; X‑ray crystal structures show that base-on cobalamin fits neatly within the riboswitch binding site, whereas base-off cobalamin with its floppy tail does not.
Fig. 3. (From Figure 4 from the paper) (A) The sixteen corrinoids used to test the relationship between corrinoid base-on tendency and riboswitch response. (B) The responses of different riboswitches to the tested corrinoids, which have differing base-on tendencies. Base-off corrinoids are represented by the empty circles. Source
This broad specificity for corrinoids looks like a smart strategy for microbes: induce a genetic response in the presence of only the types of corrinoids that the cell likes to use. Perhaps this is why the corrinoid selectivity of the tested cobalamin riboswitches group by taxonomic class rather than by genomic context – raising interesting implications in interactions between diverse microbes. The role of cobalamin in microbial community development and dynamics was a topic covered on the blog a few years back, here⛓️💥. Perhaps this new study's findings hint at riboswitches having a hand in such processes, as a molecule-level phenomenon that ripples out to the level of the cell and then the level of microbial ecology, a biological bullwhip effect. After all, over 70% of sequenced bacterial species use corrinoids.
This study depicts one tactic by which riboswitches distinguish between members of a suite of chemically similar molecules. Meanwhile, other riboswitches do in fact possess particular regions that are responsible for ligand-detecting, and yet others rely on electrostatics between RNA and ligand. Perhaps another possibility might be RNA editing. RNA modifications, like little chemical diacritics, alter biological meaning for an RNA molecule by tweaking the secondary structure and the RNA's interaction with other macromolecules. Such epitranscriptomics have to date been a very eukaryote-centric topic of study, but increasing evidence finds that these RNA-calibrations pan out in bacteria as well. Since changes in RNA allow quick responses to environmental shifts like nutrient availability, riboswitches might very well fall within the purview of modifying enzymes. (One relevant example: In eukaryotes, the presence of the m6A modification induces a structural remodeling of RNA loops that promotes the binding of proteins. These loops are even termed "m6A switches" – maybe such switches are also at play within riboswitches.)
Riboswitch fine-tuning is a complex joint enterprise between sequence and structural factors. Never one to be simple, that RNA.
