A 108-residue RING-H2 protein, the catalytic subunit shared by most human cullin-RING
ubiquitin ligases (CRLs). Two structural elements do all the work: an N-terminal strand
that pairs with the cullin C-terminal domain through an intermolecular beta-sheet, and a
cross-braced RING domain that coordinates two zinc ions (a third is seen in SCF
structures) and binds a modifier-charged E2.
Ubiquitin ligase. The RING activates E2~ubiquitin for direct transfer onto a lysine
of a receptor-bound substrate, with mostly K48 chains
PMID:33234069.
Independent confirmation that the RING is the catalytic site: Glomulin binds it and
inhibits ligase activity
PMID:22405651,
and arsenite binding to the RING blocks CUL3-KEAP1-dependent NRF2 turnover
PMID:29658272.
NEDD8 ligase. With UBE2M/UBC12 the same RING neddylates the conserved cullin lysine
[file:human/RBX1/RBX1-uniprot.txt "Promotes the neddylation of CUL1, CUL2, CUL4 and CUL4 via its interaction with UBE2M."];
the E2 split is UBE2M/RBX1 versus UBE2F/RBX2
PMID:19250909,
and the mechanism is RING rotation, caught in a trapped intermediate
PMID:24949976,
PMID:21765416.
Neddylation then reorients the cullin-RING arm and removes the CAND1 site
PMID:18805092.
DCN-type co-E3s (DCUN1D1-5) deliver UBE2M~NEDD8 to the cullin-RBX1 module
PMID:26906416.
Substrate choice belongs to the receptor, not to RBX1
[file:human/RBX1/RBX1-uniprot.txt "The functional specificity of the E3 ubiquitin-protein ligase complexes depends on the variable substrate recognition components."],
[file:human/RBX1/RBX1-deep-research-falcon.md "RBX1 supplies catalytic E2 recruitment but ordinarily does not select protein substrates by itself."].
This drives most of the grading below: pathway-level processes that follow from one
F-box/BTB/DCAF receptor are kept but marked non-core, and vague downstream phenotypes
(proliferation, apoptosis regulation, epigenetic regulation, MAPK cascade) are marked
over-annotated. The processes RBX1 itself executes — ubiquitination of all linkage
types, neddylation, and the CRL/SCF-dependent proteolysis that follows — are accepted.
RBX1 sits in CRL1/SCF PMID:15520277,
CRL2 (elongin BC plus VHL-box receptors; CUL2-Rbx1 versus CUL5-Rbx2 selectivity)
PMID:15601820,
CRL3 PMID:14528312,
CRL4A/B PMID:22118460,
CUL7-FBXW8 and the hexameric CUL9 assembly
PMID:24793696.
Two structural caveats worth recording: in CRL7(FBXW8) the RBX1 RING is held
non-productively and the ubiquitination is done by a separate neddylated CUL1-RBX1 module
PMID:35982156,
and in the tetrameric CRL4(DCAF1) resting state the RING is occluded until neddylation
PMID:34595758.
CUL5-RBX1 is retained but treated as non-core, since SOCS-box receptors select CUL5-RBX2.
Nucleus and cytoplasm, with the effective site set by the assembled ligase
[file:human/RBX1/RBX1-deep-research-falcon.md "RBX1 is reported in both the nucleus and cytoplasm, consistent with degradation of substrates in both compartments."].
ROC1 binding itself promotes nuclear accumulation of CUL1
PMID:11027288.
"Site of DNA damage" is accepted: the CRL4(CSA) ligase ubiquitinates stalled Pol II at
lesions PMID:34526721.
Golgi (from a CRL4-DCAF12 study) and centrosome (SCF(cyclin F)) are kept as non-core
receptor-driven pools.
Graded alongside genes/human/CUL1 (the scaffold: GO:0160072 scaffold activity, receptor
outputs non-core) and genes/human/ANAPC11 (the paralogous APC/C RING). ANAPC11 shares
the PTN000129805 IBA node with RBX1, and the cullin-binding split between them —
ROC1/ROC2 bind all cullins while APC11 is restricted to APC2 — comes from the same 1999
paper.