YgiQ, also named RlmX, is a 739-residue cobalamin-dependent radical S-adenosyl-L-methionine (SAM) methyltransferase of the UPF0313 family and the enzyme that installs a stereoselective methyl group on carbon 5'(S) of the ribose of two 23S rRNA residues in the peptidyl-transferase centre of the 50S subunit: dihydrouridine 2449, giving D-5S-m2449, and cytidine 2498, giving Cm-5S-m2498 once RlmM has added the 2'-O methyl. This is a methylation of the sugar-phosphate backbone at C5' rather than the familiar 2'-O-methylation, and is the first RNA ribose modification described outside the 2' position. The protein combines an N-terminal cobalamin-binding domain, a radical SAM domain carrying one [4Fe-4S] cluster ligated by Cys386, Cys390 and Cys393, and a C-terminal DUF3362 domain, plus an enzyme-specific surface cysteine triad (Cys453, Cys458, Cys464) that is essential for activity and has been attributed to RNA recognition through zinc binding. Catalysis consumes two SAM per methyl group transferred: one is reductively cleaved by the cluster to the 5'-deoxyadenosyl radical that abstracts the 5'(S) hydrogen of the target ribose, and the second remethylates cobalamin, which is the immediate methyl donor. The enzyme acts on naked 23S rRNA early in 50S assembly, not on the mature subunit, and recognises domain V as its minimum efficient substrate. Both modifications, together with the RlhA-dependent 5-hydroxycytidine at position 2501, rise sharply during anaerobic growth, when RlmX protein levels and activity increase; they stabilise the P site and the peptidyl-transferase centre and raise the rate of peptide bond formation, so that cells lacking both RlmX and RlhA grow slowly without oxygen. Activity requires ferric iron and cobalamin supplied from the medium. RlmX homologues are widespread in bacteria and present in some archaea.
Definition: Catalysis of the stereoselective transfer of a methyl group to carbon 5'(S) of the ribose moiety of dihydrouridine at position 2449 and of cytidine at position 2498 of bacterial 23S ribosomal RNA, using methylcobalamin as the immediate methyl donor and S-adenosyl-L-methionine both as the source of the 5'-deoxyadenosyl radical that abstracts the 5'(S) hydrogen and as the methyl donor that regenerates methylcobalamin.
Justification: Ishiguro et al. show that E. coli RlmX (YgiQ, Q46861) methylates the ribose C5' position of 23S rRNA D2449 and C2498, the first RNA ribose modification described outside the 2' position. GO has no molecular-function term for ribose C5' methylation of RNA: every ribose methyltransferase term in GO is a 2'-O-ribose term (GO:0008650, GO:0070039, GO:0070677, GO:0030743, GO:0090486, GO:0180021 and the tRNA series), and the children of GO:0008649 rRNA methyltransferase activity are base-specific terms (GO:0016433, GO:0016434, GO:0016435, GO:0016436, GO:0070037). Annotating this activity with a 2'-O-methyltransferase term would assert the wrong chemistry, and GO:0008649 alone loses both the position and the stereochemistry. GO:0008649 is the right parent because its definition already admits the ribose group as the methyl acceptor. No EC number or RHEA reaction has been assigned to this activity.
Parent term: rRNA methyltransferase activity
Supporting Evidence:
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0003824 catalytic activity | IEA GO_REF:0000002 | MODIFY | Summary: Root-level catalytic activity from the radical SAM InterPro signatures. The reaction is now known: YgiQ is RlmX, which methylates carbon 5'(S) of the ribose of 23S rRNA D2449 and C2498 using methylcobalamin as methyl donor and a radical SAM [4Fe-4S] cluster to generate the 5'-deoxyadenosyl radical. Reason: Per the RADICAL_SAM treatment rule for generic rows, GO:0003824 is replaced by the specific reaction once catalysis is established. It now is: the ygiQ deletion loses both methylations and plasmid-borne ygiQ restores them, the authors rename ygiQ as rlmX, and reconstituted recombinant protein methylates rRNA in vitro. GO has no term for ribose C5' methylation of RNA (all GO ribose methyltransferase terms are 2'-O-ribose terms, and the children of GO:0008649 are base terms), so the replacement is the proposed new term in this review rather than a force-fitted 2'-O term. Proposed replacements: 23S rRNA (dihydrouridine(2449)/cytidine(2498)-ribose-5'-C)-methyltransferase activity Supporting Evidence: PMID:41380683 of a plasmid-encoded ygiQ gene, indicating that ygiQ is essential PMID:41380683 Hereafter, we refer to ygiQ as rlmX. PMID:41380683 methylation outside the 2′ position, representing an unprecedented example of RNA backbone modification. We identified |
| GO:0005506 iron ion binding | IEA GO_REF:0000104 | MODIFY | Summary: Iron ion binding from a UniRule. The iron of RlmX is held as a single [4Fe-4S] cluster, not as a mononuclear iron ion: iron quantification of reconstituted recombinant protein gave 3.69 Fe per protein, consistent with one cluster at about 92% occupancy, and no separate iron site is known. Reason: Lateral correction, applying the RADICAL_SAM treatment rule for GO:0046872 metal ion binding to this sibling term: the iron is bound as a cluster, so the supported term is GO:0051539 4 iron, 4 sulfur cluster binding, which the gene already carries. Proposed replacements: 4 iron, 4 sulfur cluster binding Supporting Evidence: PMID:41380683 revealed 3.69 iron atoms bound to one recombinant protein file:ECOLI/ygiQ/ygiQ-uniprot.txt Binds 1 [4Fe-4S] cluster. The cluster is coordinated with 3 |
| GO:0051536 iron-sulfur cluster binding | IEA GO_REF:0000002 | MODIFY | Summary: Generic iron-sulfur cluster binding from the radical SAM signatures. The cluster at the CxxxCxxC motif (Cys386, Cys390, Cys393) is a [4Fe-4S] cluster, demonstrated by iron quantification and by the complete loss of activity on substituting any one of the three cysteines. Reason: Correct but general. Per the RADICAL_SAM treatment rule, replace with the descendant GO:0051539, which the gene already carries and which the measured cluster content supports. Proposed replacements: 4 iron, 4 sulfur cluster binding Supporting Evidence: PMID:41380683 C390, and C393) in the [4Fe-4S] cluster, which resulted in PMID:41380683 revealed 3.69 iron atoms bound to one recombinant protein |
| GO:0051539 4 iron, 4 sulfur cluster binding | IEA GO_REF:0000002 | ACCEPT | Summary: [4Fe-4S] cluster binding, triggered here by an InterPro methylthiotransferase conserved-site signature. YgiQ is not a methylthiotransferase, but the term itself is correct and now experimentally grounded: reconstituted RlmX carries one [4Fe-4S] cluster at Cys386/Cys390/Cys393, the cluster is indispensable for methylation in vitro, and each single cysteine substitution abolishes activity in vivo. Reason: Per the RADICAL_SAM treatment rule, GO:0051539 is core for every member whose catalysis is established, because the radical SAM cluster is the catalytic cofactor that generates the 5'-deoxyadenosyl radical. Catalysis is now established for Q46861, so this row is accepted as a core function. Supporting Evidence: PMID:41380683 revealed 3.69 iron atoms bound to one recombinant protein PMID:41380683 either the [4Fe-4S] cluster or Me-Cbl was indispensable for PMID:41380683 C390, and C393) in the [4Fe-4S] cluster, which resulted in file:ECOLI/ygiQ/ygiQ-uniprot.txt Binds 1 [4Fe-4S] cluster. The cluster is coordinated with 3 |
| GO:1904047 S-adenosyl-L-methionine binding | IDA PMID:41380683 Hypoxia-induced ribosomal RNA modifications in the peptidyl-... | NEW | Summary: RlmX binds SAM twice over in its catalytic cycle: one molecule is reductively cleaved at the [4Fe-4S] cluster to the 5'-deoxyadenosyl radical, and a second remethylates cobalamin. Reconstituted RlmX methylates rRNA only in the presence of SAM, transfers a CD3 group from CD3-SAM, and the side chain of Glu525 contacts the SAM ribose, with E525A impairing activity. Reason: GOA has no SAM-binding row for Q46861, although the RADICAL_SAM project treats GO:1904047 as a core cosubstrate-binding function of every member whose radical SAM catalysis is established. This is a molecular function, so the participation test is met directly: RlmX itself binds and cleaves SAM, shown with purified protein. Comparator check, against GOA rather than against this project's own proposals: GOA does annotate GO:1904047 to radical SAM enzymes in exactly this role, by UniRule -- ECOLI/queE (UR000100349), ECOLI/moaA and PSEPK/moaA (UR000101305), METEA/pqqE and PSEPK/pqqE (UR000098303). So the absence here is patchy UniRule coverage, which rules happen to carry the term, not a GO convention against annotating a radical SAM enzyme's own cosubstrate. That is also why those five genes have an existing row to ACCEPT or MODIFY while this one needs a NEW row. Supporting Evidence: PMID:41380683 in a SAM-dependent manner (Figure S7B). Moreover, deuterium-labeled SAM (CD3-SAM) was used as a substrate for PMID:41380683 Regarding SAM recognition, the side chain of E525 directly interacts with the ribose of SAM (Figure S5E). E525A mutation PMID:41380683 a second SAM molecule to regenerate Me-Cbl (III). In our |
| GO:0031419 cobalamin binding | IDA PMID:41380683 Hypoxia-induced ribosomal RNA modifications in the peptidyl-... | NEW | Summary: RlmX is a cobalamin-dependent radical SAM methyltransferase with an N-terminal cobalamin-binding domain. Methylcobalamin bound to reconstituted recombinant protein, detected by its 320 nm absorbance; Me-Cbl is indispensable for methylation in vitro and is the immediate methyl donor; and G160N in the GG motif that contacts the dimethylbenzimidazole moiety of cobalamin abolishes methylation in vivo. In vivo the modifications require cobalamin in the medium. Reason: GOA has no cobalamin-binding row for Q46861, and GO:0031419 is verified in QuickGO as the molecular function for binding cobalamin. The participation test is met directly: RlmX itself binds the cofactor, demonstrated spectroscopically on purified protein and by loss-of-function of a cobalamin-contact residue. Comparator check: cobalamin binding is the standard annotation for the other cobalamin-dependent radical SAM methyltransferases of this class (TsrM, OxsB, TokK, Mmp10) and for cobalamin-dependent methionine synthase. Supporting Evidence: PMID:41380683 as a Cbl-dependent radical SAM enzyme, which is composed PMID:41380683 92% occupancy. We also added methylcobalamin (Me-Cbl) during the reconstitution of the holoenzyme. Correct binding of Cbl PMID:41380683 completely lost methylation ability (Figure 2C), highlighting the PMID:41380683 of an active holoenzyme. Me-Cbl also functions as a methyl |
| GO:0019843 rRNA binding | IDA PMID:41380683 Hypoxia-induced ribosomal RNA modifications in the peptidyl-... | NEW | Summary: RlmX recognises structured naked 23S rRNA. Holoenzyme methylates deproteinized rRNA and in vitro transcribed 23S rRNA but not the assembled 50S subunit, and among sub-fragments only the domain V transcript (G2018 to C2628) is an efficient substrate, while shorter H89-H93 and H89 transcripts are not. The enzyme-specific surface cysteine triad is proposed to contribute to this RNA recognition. Reason: GOA has no rRNA-binding row for Q46861. This is a molecular function the protein itself performs: productive methylation of defined rRNA transcripts by purified protein requires direct binding, and the mapping of substrate competence to domain V rather than to the local helix shows recognition of the rRNA itself. Comparator check: rRNA binding (GO:0019843) is routinely annotated to E. coli 23S rRNA methyltransferases that recognise their substrate directly, such as RlmM and RlmE. Supporting Evidence: PMID:41380683 introduced into the domain V transcript (G2018-C2628), whereas PMID:41380683 suggest that domain V is a minimum substrate for RlmX-mediated PMID:41380683 any methylations in the 50S subunit (Figure 3C), but clearly |
| GO:0031167 rRNA methylation | IMP PMID:41380683 Hypoxia-induced ribosomal RNA modifications in the peptidyl-... | NEW | Summary: RlmX carries out the post-transcriptional methylation of 23S rRNA at the ribose C5'(S) of D2449 and C2498. Deletion of ygiQ converts D-5S-m2449 and Cm-5S-m2498 to the unmethylated D2449 and Cm2498, and plasmid-encoded ygiQ restores both; reconstituted protein reproduces the methylation on isolated rRNA. Reason: GOA has no biological-process row for Q46861. The participation test is met in the strongest form available: RlmX itself catalyses the methyl transfer step of this process, shown both by knockout and complementation in vivo and by in vitro reconstitution, so the entity that performs the step is this gene product. GO:0031167 is the most specific applicable term, because its ribose-specific children are all 2'-O-methylation terms and its base-methylation sibling GO:0070475 does not apply; the ancestor GO:0000154 rRNA modification is deliberately not proposed as well. Comparator check: the other E. coli 23S rRNA methyltransferases (RlmM, RlmE, RlmN and the rest of the Rlm series) all carry rRNA methylation or one of its children. No anaerobic-growth or translation process term is proposed: the growth and peptidyl-transfer phenotypes of the rlmX rlhA double knockout are consequences of the altered ribosome, not steps that RlmX performs. Supporting Evidence: PMID:41380683 of a plasmid-encoded ygiQ gene, indicating that ygiQ is essential PMID:41380683 for D5Sm2449 and Cm5Sm2498 formation in E. coli (Figures 2A PMID:41380683 any methylations in the 50S subunit (Figure 3C), but clearly |
Loading supporting content…
Download this section (compressed HTML)Q: UniProt Q46861 and NCBI Gene 948963 still carry no rlmX gene name and no function, although Ishiguro et al. rename ygiQ as rlmX and establish its activity. Can the gene name, protein name and function be synchronised, and should the UPF0313 family description be updated to a cobalamin-dependent radical SAM rRNA ribose C5'-methyltransferase family? Until then the gene symbol used in this repository stays ygiQ.
Suggested experts: UniProt curation team, Tsutomu Suzuki
Q: Should GO add a molecular-function term for ribose C5' methylation of RNA, as proposed here under GO:0008649, and a matching process term distinguishing ribose backbone methylation from 2'-O-methylation and base methylation under GO:0031167?
Suggested experts: GO ontology editors
Q: Does the RlmX-specific cysteine triad Cys453, Cys458 and Cys464 bind zinc, and is the zinc attribution transferable to other UPF0313 members?
Suggested experts: Squire J Booker, Kensuke Ishiguro
Q: Do RlmX homologues in other bacteria and in archaea modify the equivalent peptidyl-transferase-centre riboses, and is the hypoxia induction conserved?
Suggested experts: Tsutomu Suzuki, Jaanus Remme
Experiment: Quantify zinc, iron and other metals in anaerobically reconstituted wild-type RlmX and in the C453A, C458A and C464A variants by ICP-MS, and measure domain V rRNA binding (filter binding or MST) and methylation rate separately for each variant, so that a binding defect can be distinguished from a catalytic one.
Hypothesis: The RlmX-specific cysteine triad is a zinc site that mediates rRNA binding rather than catalysis.
Type: metal analysis with RNA binding and kinetic assays
Experiment: Replace methyl viologen in the reconstituted methylation assay with purified E. coli flavodoxin, flavodoxin-NADP+ reductase and NADPH, and measure D-5S-m2449 and C-5S-m2498 formation by RNA-MS; in parallel, quantify the modifications in strains depleted of fldA or fpr grown anaerobically.
Hypothesis: Flavodoxin and flavodoxin-NADP+ reductase are the physiological reductants of the cobalamin cofactor of RlmX.
Type: anaerobic enzymology with RNA mass spectrometry
Experiment: Measure RlmX protein level, cluster occupancy (iron and UV-visible absorbance on affinity-purified protein) and D-5S-m2449 frequency in cells sampled through an aerobic-to-anaerobic shift, and test whether overexpression of iron-sulfur cluster assembly or repair systems raises the aerobic modification level.
Hypothesis: RlmX activity, rather than rlmX expression, is limited by oxygen through the lability of its [4Fe-4S] cluster.
Type: physiology with cofactor and modification quantification
What is not known — curated, literature-grounded statements of the open unknowns (the inverse of core functions).
Gap: Whether the RlmX-specific surface cysteine triad Cys453, Cys458 and Cys464 binds zinc, and what it actually does, is unresolved. All three residues are essential for methylation, and the authors attribute the triad to RNA recognition through zinc binding by analogy to earlier work, but no metal was measured on RlmX and the triad could instead contact rRNA directly or hold the protein fold.
OPEN BIOLOGY
What is known: Established: the triad is RlmX-specific, conserved across bacterial homologues, located on the protein surface in the AlphaFold2 model, and each single alanine substitution abolishes 5'(S)-methylation in vivo. Also established: iron quantification gives one [4Fe-4S] cluster at about 92% occupancy, so the triad is not a second iron-sulfur cluster. Not established: the metal content of the triad, and whether it contacts rRNA.
Significance: A zinc site would be the first metal-dependent RNA-recognition module described in a cobalamin-dependent radical SAM methyltransferase and would explain how this family selects a single ribose in a folded rRNA domain; if the triad instead ligates rRNA or supports the fold, the zinc attribution carried over from earlier work should not be propagated to UPF0313 family members.
What would resolve it: Quantify zinc and other metals in reconstituted RlmX, and test rRNA binding and methylation by the triad mutants separately so that loss of binding can be distinguished from loss of catalysis.
Provenance (the field's own admissions):
Gap: The physiological electron donor that re-reduces cobalamin(II) during each RlmX turnover in E. coli is unknown.
OPEN BIOLOGY
What is known: Established: in vitro, methyl viologen serves as the electron carrier for the cobalamin redox cycle, and the reaction fails without it. Flavodoxin, flavodoxin reductase and NADPH are suggested by analogy with other cobalamin-dependent radical SAM methyltransferases but were not tested here.
Significance: Identifying the donor would complete the pathway and would say whether the anaerobic induction of these modifications is partly a property of the reducing system rather than of RlmX itself.
What would resolve it: Reconstitute RlmX turnover with purified E. coli flavodoxin, flavodoxin-NADP+ reductase and NADPH in place of methyl viologen, and test whether the modifications are lost in strains depleted of those components under anaerobic growth.
Provenance (the field's own admissions):
Gap: How RlmX activity is raised under hypoxia is unknown: rlmX mRNA levels do not change between aerobic and anaerobic growth, yet the modifications and the protein increase, and at equal induced expression the modification level is still much higher anaerobically.
OPEN BIOLOGY
What is known: Established: RNA-seq shows no change in rlmX mRNA, proteomics shows more RlmX protein anaerobically, and arabinose-titrated expression gives more D-5S-m2449 anaerobically than aerobically at the same inducer concentration. Proposed but untested: that the oxygen sensitivity of the [4Fe-4S] cluster limits activity in air, with post-transcriptional regulation or protein stabilisation raising protein levels.
Significance: This is the control point that makes these modifications hypoxia-specific, and so the point at which ribosome tuning is coupled to oxygen availability.
What would resolve it: Measure RlmX protein stability, translation efficiency and cluster occupancy in cells shifted between aerobic and anaerobic growth, and test whether cluster-repair systems change the aerobic modification level.
Provenance (the field's own admissions):
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)