by Janie
Pseudouridine, an isomer of uridine, is perhaps now most widely known for its critical role in the success of the COVID-19 mRNA vaccines. Three years ago on STC, Roberto wrote an appreciative note about this key feature of the vaccines, the N1-methylpseudouridine that acts a bit like an invisibility cloak to prevent our immune systems from lashing out against foreign RNA.
Figure 1. Pseudouridine is an isomer of uridine. Pseudouridine synthases isomerize uridine into pseudouridine, so that the nucleobase is connected to the sugar moiety via a carbon-carbon bond instead of a carbon-nitrogen bond. Frontispiece: Structure of the ring-open pseudouridine monophosphate (ΨMP). Source
Prior to this COVID-catapulted stardom, pseudouridine was better known in RNA biology circles as a post-transcriptional regulator. It is a common RNA modification found in all RNA types, responsible for calibrating a panoply of essential processes. With its altered hydrogen bonding capacity, pseudouridine base-pairs with all four canonical nucleobases and stabilizes RNA molecules by coordinating a water molecule between the phosphate backbone and the newly exposed nitrogen, thereby decreasing the flexibility of an RNA molecule. From stabilizing proper RNA structure to enabling stop-codon readthrough to increasing antibiotic resistance, the many cellular roles of pseudouridine in RNA are perhaps befitting of a separate future post...
This RNA-biologist-centric characterization of pseudouridine is all well and good, but I recently learned that the nucleobase also finds its way into a few bacterial natural products. (This was particularly intriguing to me, as someone who spent a few years in natural-products-oriented labs before hopping over into labs that are more RNA-biology-oriented.)
Figure 2. Examples of C-nucleoside natural products, all of which contain nucleosides or nucleoside analogs connected via carbon-carbon bonds to a sugar moiety. Also shown is indigoidine, a blue pigment whose biosynthetic pathway overlaps with those of some C-nucleoside antibiotics. Source
A few nucleoside analog antibiotics, including pseudouridimycin, malayamycin A, and minimycin, contain pseudouridine or similar residues (Figure 2). Pseudouridimycin, for example, competes with UTP for RNA polymerase's NTP addition site. Once it binds there, the polymerase is inhibited from further elongating the nascent RNA chain.
Why incorporate pseudouridine into these nucleoside analogs instead of uridine (or any other canonical base)? Perhaps the ability of pseudouridine to form hydrogen-bonding patterns different from those of uridine confer it extra specificity when slotting into target enzymes, making resistance more difficult. Stability of the resulting molecule is also a consideration: the C-C bond is resistant to common enzymes like hydrolases that break C-N glycosidic bonds, which equates to longer-lived molecules. In the context of bacterial chemical warfare, that means less easily dismantled weapons.
Figure 3. Pathway for pseudouridine catabolism in E. coli. The pseudouridine kinase YeiC first phosphorylates the pseudouridine into pseudouridine 5’-monophosphate, then the glycosidase YeiN cleaves the C-C glycosidic bond via a ring-opening mechanism not depicted here. Source
Whatever the cause for that C-C bond, at the end of the catabolic day it presents an interesting challenge. What happens to pseudouridine left lying around in a cell after RNA degradation? When it comes to bonds, the carbon-carbon variety are rather tough cookies. Enzymes tasked with breaking them typically go about it via oxidation and reshuffling of bonds to create either unstable intermediates or easily hydrolysable products, or take the Gordian Knot approach and brute-force it with radical chemistry. In the case of pseudouridine, in mammals it is simply excreted intact via urine. In bacteria (and plants), however, catabolism involves cleavage of the C-C bond! The process (Figure 3) begins with the pseudouridine kinase YeiC, which phosphorylates the pseudouridine, followed by the glycosidase YeiN, which cleaves the nucleobase off the sugar via a novel ribose ring-opening mechanism. The resulting uracil and ribose 5'-phosphate are funneled into other metabolic pathways, such as the pentose phosphate pathway for the latter, and for the former, nucleotide salvage…
… and, perhaps, C-nucleoside antibiotic production? Or so one might think. Wouldn't it make sense for a cell to upcycle spent metabolites into newly useful products?
For reasons that are beyond me (any ideas?), this appears not to be how cells go about it. In the case of pseudouridimycin, the cell forgoes the eco-friendly route and instead rebuilds from scratch. The first step of the biosynthetic workflow is thought to be performed by PumJ, an oddball of a pseudouridine synthase found within the pseudouridimycin gene cluster. Unlike other pseudouridine synthases that recognize uridines already incorporated within RNA transcripts and isomerize them into pseudouridine (Figure 2), PumJ produces free pseudouridine from free uridine – it's the eccentric cousin of the TruD family of pseudouridine synthases.
For the pseudouridine-containing malayamycin (Figure 2) as well, biosynthesis goes the way of reinventing the wheel. Here, too, the biosynthetic gene cluster features a divergent pseudouridine synthase that shares significant sequence similarity to TruD, reported as MalD. This synthase produces the free pseudouridine that becomes the basis of the final antibiotic.
Figure 4. Biosynthesis of both minimycin and indigoidine begin with glutamate. Source
Finally, there is minimycin, which is structurally related to pseudouridine with the exception of an oxygen in the place of a nitrogen. Biosynthesis begins not with a pseudouridine or even uridine precursor, but instead with a glutamate and an NRPS module called MinA. The subsequent player in this pathway, MinB, shares ~50% sequence identity with YeiN. (I wonder if these cytoplasmic pseudouridine-like bases shown in Figure 4 can end up incorporated into RNA, or if they are somehow sequestered to the purview of antibiotic biosynthetic pathways?)
There is a fuzzy link here to bacterial blue pigments (the breadth of which we previously discussed here!). YeiN, which cleaves pseudouridine's C-C bond, is a figure of intrigue in the story of indigoidine, a blue pigment with antimicrobial properties produced by various bacteria, from the natural-product-powerhouse Streptomyces to the plant pathogen Dickeya dadantii (formerly Erwinia chrysanthemi). E. coli's YeiN is homologous to Thermotoga maritima's indigoidine synthase A IndA, which also has pseudouridine-cleaving activity: both enzymes share conserved sequences and protein folds. Even in the case of the more structurally divergent minimycin, there is a connection to blue: the NRPS MinA, key to minimycin biosynthesis, also appears to be the key to indigoidine's biosynthesis, since E. coli equipped with MinA produces the blue pigment.
Even C-C bond formation can be added to the resume of YeiN-like proteins. The homologous AlnA is a C-glycosynthase that does exactly so during the biosynthesis of alnumycin A in Streptomyces, and even E. coli YeiN itself has some C-C bond forming activity. From pseudouridine catabolism to antibiotic biosynthesis to blue pigment biosynthesis, what else might these enzymes be dabbling in? Similar enzymes are certainly found in a wide spread of organisms beyond bacteria: YeiN/IndA N-terminal domain and a YeiC C-terminal domains are found (via BLAST search) on bifunctional proteins in various eukaryotes like zebrafish, chickens, Drosophila, C. elegans, and some fungi (Aspergillus and Schizosaccharomyces pombe, but not Saccharomyces cerevisiae) but not in any mammals. Perhaps evolution has improvised other functionalities.
Do you want to comment on this post? We would be happy about it! Please comment on Mastodon or Bluesky.
Comments