FBXO4

FBXO4 (FBX4) is the substrate-recognition F-box subunit of an SCF (SKP1-CUL1-RBX1-FBXO4) E3 ubiquitin ligase that functions as a tumor suppressor by targeting growth-promoting proteins for proteasomal degradation (PMID:17081987, PMID:21911473). Its best-defined substrate is Thr286-phosphorylated cyclin D1, which FBXO4 binds in cooperation with the co-adaptor alphaB-crystallin and polyubiquitinates at Lys269 to drive proteasomal turnover (PMID:17081987, PMID:19767775); loss of this activity stabilizes nuclear cyclin D1 and, in knockout mice, produces lymphomas, sarcomas, and Ras- or Braf-driven tumors (PMID:21911473, PMID:24019069). Catalytic activity requires cell-cycle-dependent FBXO4 dimerization, which is triggered by GSK3β phosphorylation downstream of the Ras-Akt axis and is structurally organized as an antiparallel domain-swapped homodimer in which the linker of one protomer engages the C-terminal substrate-binding domain of the other within the CRL1 complex (PMID:18598945, PMID:20181953, PMID:39406020). Beyond cyclin D1, FBXO4 ubiquitinates a range of additional substrates—Pin2/TRF1 to control telomere length, Fxr1, Mcl-1, PPARγ, and β-catenin—and separation-of-function mutations such as I377M genetically uncouple cyclin D1 from TRF1 targeting, indicating substrate-specific recognition surfaces (PMID:16275645, PMID:24019069, PMID:29142209, PMID:28776569, PMID:29925002, PMID:40599691). FBXO4-mediated cyclin D1 degradation is also engaged as a genotoxic-stress response in an ATM-dependent, p53-independent manner (PMID:33784509).

1. 2002 High

   Established that FBXO4 does not act alone but partners with alphaB-crystallin, whose phosphorylation state controls FBXO4 localization and ligase output—defining a co-adaptor mechanism.

   Evidence Co-IP with phosphomimetic mutagenesis and cellular ubiquitination/fractionation assays

   PMID:12468532

   Open questions at the time

   • Did not identify the physiological substrate of the complex
   • Mechanism by which phosphorylated alphaB-crystallin redirects FBXO4 not resolved
2. 2004 Medium

   Mapped the alphaB-crystallin-driven recruitment of FBXO4 to nuclear SC35 speckles, localizing the complex to a defined subnuclear compartment.

   Evidence Immunofluorescence co-transfection with detergent and nuclease resistance assays

   PMID:15511225

   Open questions at the time

   • Functional consequence of speckle recruitment not established
   • Whether substrate degradation occurs at speckles unknown
3. 2005 High

   Identified the first FBXO4 substrate, Pin2/TRF1, linking the ligase to telomere length homeostasis.

   Evidence Two-hybrid, Co-IP, in vitro/in vivo ubiquitination, and RNAi with telomere length measurement

   PMID:16275645

   Open questions at the time

   • Did not address dimerization or regulatory inputs
   • Relationship to other substrates unknown at the time
4. 2006 High

   Reconstituted SCF(FBX4-alphaB crystallin) as a bona fide E3 ligase for Thr286-phosphorylated cyclin D1, establishing the central tumor-suppressive substrate.

   Evidence In vitro ubiquitination reconstitution, Co-IP, and overexpression/knockdown with cyclin D1 turnover

   PMID:17081987

   Open questions at the time

   • How ligase activity is temporally regulated not addressed
   • Acceptor lysine on cyclin D1 not yet defined
5. 2008 High

   Revealed that GSK3β phosphorylation drives cell-cycle-dependent FBXO4 dimerization required for ligase activity, and that cancer-associated mutations abolishing dimerization cause oncogenic cyclin D1 accumulation.

   Evidence Phosphorylation/dimerization assays, mutagenesis, transformation assays, and human cancer sequencing

   PMID:18598945

   Open questions at the time

   • Structural basis of dimerization not yet visualized
   • Direct GSK3β phosphosite occupancy not fully mapped
6. 2009 High

   Identified Lys269 of cyclin D1 as the residue required for degradation, dissecting ubiquitination from proteolytic targeting.

   Evidence Site-directed mutagenesis with in vivo ubiquitination and degradation/transformation assays

   PMID:19767775

   Open questions at the time

   • K269R still permits ubiquitination, so the degradation-specific signal is unexplained
   • How K269-linked chains differ from non-degradative chains unknown
7. 2010 High

   Provided the crystal structure of FBXO4-SKP1 showing an antiparallel dimer and defined the domains essential for dimerization-dependent activity.

   Evidence X-ray crystallography with in vitro ubiquitination and mutagenesis

   PMID:20181953

   Open questions at the time

   • Structure lacked CUL1/RBX1 and substrate
   • Did not capture the GSK3β-phosphorylated state
8. 2011 High

   Demonstrated in vivo that Fbxo4 loss stabilizes nuclear cyclin D1, drives proliferation and transformation, and produces tumors, validating FBXO4 as a tumor suppressor.

   Evidence Fbxo4 knockout mouse, MEF characterization, Ras transformation, and stability assays

   PMID:21911473

   Open questions at the time

   • Contribution of non-cyclin-D1 substrates to tumorigenesis not isolated
   • Tissue-specific requirements not dissected
9. 2013 High

   Showed substrate targeting is genetically separable—I377M disrupts cyclin D1 but not TRF1 degradation—and that cyclin D1 accumulation mediates FBXO4-deficient melanoma.

   Evidence Braf/Fbxo4 compound mouse model and separation-of-function mutagenesis with dual-substrate assays

   PMID:24019069

   Open questions at the time

   • Structural determinants distinguishing substrate surfaces not mapped
   • Generalizability of separation-of-function to other substrates unknown
10. 2017 High

    Expanded the FBXO4 substrate repertoire to Fxr1 and Mcl-1, linking the ligase to a translational feedback loop and to apoptotic/chemoresistance control.

    Evidence SCFFbxo4 complex purification/MS, Co-IP, in vitro ubiquitination, KO cells, and translation/apoptosis readouts

    PMID:28776569 PMID:29142209

    Open questions at the time

    • Mcl-1 targeting rests on a single lab without structural validation
    • Whether dimerization/GSK3β regulation applies to these substrates not tested
11. 2018 Medium

    Connected FBXO4 to metabolic control by showing it degrades PPARγ under Hsp20 negative regulation, extending the ligase beyond proliferation.

    Evidence Co-IP, ubiquitination assay, and Hsp20 knockout mice with adipocyte phenotype

    PMID:29925002

    Open questions at the time

    • Single-lab evidence
    • Direct PPARγ acceptor lysine and chain topology not defined
12. 2021 Medium

    Placed FBXO4-mediated cyclin D1 degradation within the DNA damage response as an ATM-dependent, p53-independent event.

    Evidence siRNA knockdown, proteasome/ATM inhibition, cycloheximide chase, and γ-H2AX foci assay

    PMID:33784509

    Open questions at the time

    • How ATM signals to FBXO4 mechanistically unknown
    • Single-lab, single cell system
13. 2024 High

    Resolved the full CRL1FBXO4 architecture by cryo-EM, confirming a domain-swapped homodimer engaging SKP1 and contacting CUL1, and added HPV-18 E6 as an E6AP-independent target.

    Evidence Cryo-EM structural determination; and siRNA screen with E6 stability and p53-dependent death assays

    PMID:39406020 PMID:39688415

    Open questions at the time

    • Cryo-EM lacked bound substrate
    • HPV E6 targeting lacked a direct ubiquitination assay
14. 2025 Medium

    Identified FBXO4 as a STAT1-induced regulator that degrades β-catenin to suppress muscle differentiation, embedding the ligase in IFNα/JAK/STAT-Wnt signaling.

    Evidence Co-IP, ubiquitination assay, KO muscle cells with RNA-seq, and STAT1-FBXO4 promoter luciferase reporter

    PMID:40599691

    Open questions at the time

    • Single-lab evidence
    • Whether β-catenin targeting requires alphaB-crystallin or dimerization not tested

• How distinct co-adaptors and substrate surfaces select among FBXO4's many substrates, and whether GSK3β-dependent dimerization regulates all of them, remains unresolved.
• No structure of FBXO4 with any substrate bound
• Co-adaptor requirement for non-cyclin-D1 substrates unknown
• Whether phosphorylation-dependent dimerization gates all targets untested