URM1

URM1 is a dual-function ubiquitin-like protein that serves both as a sulfur carrier in tRNA modification and as a covalent protein modifier under oxidative stress (PMID:19145231, PMID:21209336). Its NMR solution structure adopts a ubiquitin fold most closely resembling the prokaryotic MoaD/ThiS sulfur-carrier proteins, positioning Urm1 as an evolutionary link between bacterial sulfur-donor systems and eukaryotic ubiquitin-like modifiers (PMID:16864801, PMID:16046629). Activation proceeds through the E1-like enzyme Uba4 (MOCS3 in humans): Uba4 adenylates the Urm1 C-terminus and, via a thioester linkage at Cys225, shuttles Urm1 between its adenylation and rhodanese domains to generate a C-terminal thiocarboxylate (Urm1-COSH), the chemistry that drives both pathway branches (PMID:19145231, PMID:29718331, PMID:32901956, PMID:39673271). The thiocarboxylated species donates sulfur for 2-thiolation of wobble uridines in cytoplasmic tRNAs through downstream thiolases including Ncs6 (PMID:19145231, PMID:39673271). The same thiocarboxylate intermediate enables urmylation: under oxidative stress Urm1 conjugates to lysine residues of target proteins—including the peroxiredoxin Ahp1—via an E2/E3-independent mechanism accompanied by cysteine persulfidation, with substrates spanning yeast Ahp1 and mammalian MOCS3, CTU2, and CAS (PMID:21209336, PMID:36102610). Sulfur supply mechanistically couples the tRNA-thiolation and protein-conjugation branches, both depending on the Uba4 rhodanese domain (PMID:28357324). Beyond modification chemistry, urmylation promotes stress-induced phase separation, driving deposition of stress-sensitive proteins into stress granules and nuclear condensates and conferring stress resilience (PMID:38942013). In human cells the canonical neddylation enzymes NAE1/UBA3 and UBE2M act as E1 and E2 for the urmylation cascade, and pharmacological NAE1 inhibition by pevonedistat blocks urmylation and synergizes with cisplatin in liver cancer cells (PMID:42056084).

1. 2005 Medium

   Establishing the Urm1 fold answered whether this ubiquitin-like protein is built like a conjugation modifier or a sulfur carrier, revealing its closest structural kinship to bacterial MoaD.

   Evidence NMR solution structures of mouse and yeast Urm1 with structural/phylogenetic comparison

   PMID:16046629 PMID:16864801

   Open questions at the time

   • Structure alone did not demonstrate which biochemical activity (sulfur transfer vs conjugation) dominates in vivo
   • No partner enzyme captured in the structural studies
2. 2008 High

   Detection of a thiocarboxylate intermediate defined Urm1's first biochemical activity as a sulfur donor distinct from canonical UBL conjugation, mechanistically aligning it with prokaryotic sulfur carriers.

   Evidence Functional proteomics and enzymatic intermediate detection

   PMID:19017811

   Open questions at the time

   • Did not resolve the full enzymatic route of thiocarboxylate formation
   • Downstream tRNA targets not yet mapped at this stage
3. 2009 High

   Reconstitution placed Urm1 in a defined tRNA-thiolation pathway, showing Uba4 adenylates then transfers sulfur to Urm1 and that Ncs6 binds tRNA, establishing the sulfur-relay architecture.

   Evidence In vitro sulfur-transfer and tRNA thiolation assays with genetic screens in S. cerevisiae

   PMID:19145231

   Open questions at the time

   • Atomic mechanism of intramolecular sulfur transfer within Uba4 not yet defined
   • Did not address a protein-conjugation role
4. 2011 High

   Discovery of urmylation answered whether Urm1 also acts as a protein modifier, showing oxidative-stress-driven isopeptide conjugation to lysines using the C-terminal thiocarboxylate rather than canonical UBL chemistry.

   Evidence In vivo conjugation assays, mutagenesis, MS substrate identification and Co-IP in yeast and mammalian cells

   PMID:21209336

   Open questions at the time

   • E2/E3 requirements and enzymatic route of conjugation unresolved
   • Functional consequence of modifying each substrate not defined
5. 2015 Medium

   Cross-species complementation established that human URM1/MOCS3 and a Drosophila pathway are functional orthologs, demonstrating conservation of both branches and substrate overlap.

   Evidence Yeast gene-shuffle complementation, urmylation and tRNA thiolation assays; Drosophila null mutants and JNK epistasis

   PMID:25747390 PMID:26715182

   Open questions at the time

   • Reduced efficiency of human enzymes in yeast leaves native human kinetics unaddressed
   • Connection between urmylation loss and JNK induction not mechanistically dissected
6. 2018 High

   Identifying the Uba4 Cys225 thioester resolved how Urm1 is handed between Uba4's domains, defining the indispensable step linking activation to thiocarboxylation and tRNA thiolation.

   Evidence In vitro thiocarboxylation assays, Cys225 mutagenesis and in vivo validation

   PMID:29718331

   Open questions at the time

   • Did not provide a full structural view of the heterodimer
   • Release of Urm1-SH to downstream factors not defined
7. 2018 Medium

   A virally relevant urmylation event was characterized: modification of HTLV-1 Tax redirects it to the cytoplasm and tunes NF-kB target transcription, extending urmylation to host-pathogen signaling.

   Evidence Co-IP, subcellular localization, NF-kB reporter assays and a tax-transgenic Drosophila model

   PMID:29665857

   Open questions at the time

   • Single-lab study without reciprocal validation of the conjugation site
   • Direct biochemical demonstration of Tax urmylation chemistry limited
8. 2020 High

   Crystal structures of full-length Uba4 and the Uba4-Urm1 complex revealed how the adenylation and rhodanese domains recognize and thiocarboxylate Urm1 and how Uba4 avoids self-conjugation.

   Evidence X-ray crystallography of full-length enzyme and heterodimer with mechanistic biochemistry

   PMID:32901956

   Open questions at the time

   • Dynamic positioning of the rhodanese domain during catalysis not captured
   • Did not address mammalian enzyme architecture
9. 2022 High

   In vitro reconstitution proved urmylation is E2/E3-independent and oxidative-stress-dependent, and showed it transfers sulfur to target cysteines (persulfidation), unifying the sulfur-carrier and modifier roles at the chemical level.

   Evidence In vitro urmylation reconstitution, crystal structures of Ahp1 +/- Urm1, persulfidation assays

   PMID:36102610

   Open questions at the time

   • Whether mammalian urmylation requires dedicated enzymes left open
   • Substrate-selection rules for which lysines are modified not defined
10. 2024 High

    Demonstrating Urm1-driven phase separation answered what cellular outcome urmylation produces under stress, showing pH-triggered Urm1 self-association deposits target proteins into protective condensates.

    Evidence Live-cell condensate imaging, Urm1 deletion phenotyping, biochemical self-association and stress urmylation assays

    PMID:38942013

    Open questions at the time

    • Structural basis of Urm1 self-association not resolved
    • Generality of condensate role beyond yeast not established
11. 2024 High

    A cryo-EM structure plus live-cell delivery of synthetic URM1 refined the activation mechanism and directly visualized stress-regulated URM1 redistribution and conjugation in cells.

    Evidence Cryo-EM of Uba4/Urm1, interface mutagenesis with in vitro/in vivo assays; suspension bead loading with live-cell imaging under redox stress

    PMID:39246272 PMID:39673271

    Open questions at the time

    • Bead-loading study is single-lab with a novel delivery method
    • Precise compartments of stress-induced redistribution not fully mapped
12. 2026 High

    Identification of NAE1/UBA3 and UBE2M as human E1/E2 for urmylation answered the long-standing question of mammalian conjugation enzymes and opened a therapeutic angle via pevonedistat.

    Evidence Activity-based URM1 probe, proteomics, cell-based validation and pharmacological NAE1 inhibition in liver cancer cells

    PMID:42056084

    Open questions at the time

    • Reconciliation with the E2/E3-independent yeast mechanism not addressed
    • Role of DCN1 module remains tentative

• How urmylation substrate specificity is encoded, and how the apparently E2/E3-independent yeast chemistry reconciles with a dedicated human E1/E2 cascade, remains unresolved.
• No unified model of lysine-site selection across substrates
• Mechanistic reconciliation of E2/E3-independent vs NAE1/UBE2M-dependent conjugation lacking
• Disease relevance of mammalian urmylation beyond cancer cell models unexplored