FBXO34 Substrate Scope and SCF Adaptor Function: Evidence Review

Summary

FBXO34 (F-box only protein 34; UniProt Q9NWN3) is a confirmed SCF (SKP1-CUL1-F-box) E3 ubiquitin ligase substrate-recognition adaptor with one well-validated degradative substrate: HNRNPU/hnRNP U. Among 38 unique interaction partners catalogued in IntAct, only HNRNPU has direct evidence of FBXO34-dependent ubiquitination leading to proteasomal degradation with functional consequences (HIV-1 latency reactivation). The remaining ~35 interactors derive exclusively from high-throughput screens (Y2H, AP-MS, XL-MS, LUMIER) and lack any mechanistic validation as substrates. Current GO-style curation correctly assigns FBXO34 broad "ubiquitin-like protein ligase substrate adaptor activity" but should also support a specific HNRNPU substrate annotation. High-throughput interactors should remain classified as non-core interaction leads until independent substrate-level evidence is obtained.

1. Evidence for SCF-FBXO34 Complex Formation

F-box Domain

FBXO34 contains a canonical F-box domain at residues 572-624 (UniProt annotation), which mediates binding to SKP1, the adaptor protein linking F-box proteins to the CUL1 scaffold.

SKP1 Interaction

• Method: Yeast two-hybrid (Y2H) pooling approach
• Source: Stelzl et al. 2005 (PMID: 16169070) — a large-scale human interactome screen of 4,456 baits × 5,632 preys
• Evidence level: High-throughput detection; consistent with canonical F-box–SKP1 binding but not independently validated by targeted biochemistry

CUL1 Interaction

• Method: Anti-tag co-immunoprecipitation (AP-MS)
• Source: BioPlex 2.0, Huttlin et al. 2017 (PMID: 28514442)
• Evidence level: Co-purification in a systematic AP-MS campaign; supports SCF complex assembly

UniProt Curation

UniProt annotates the SUBUNIT field as "Directly interacts with SKP1 and CUL1" based on ECO:0000250 (sequence similarity evidence). The GO annotation GO:0031146 ("SCF-dependent proteasomal ubiquitin-dependent protein catabolic process") is assigned by NAS (Non-traceable Author Statement) via ComplexPortal, referencing Thompson et al. 2021 (PMID: 34445249), a review of SCF complex roles.

Assessment

The combination of a canonical F-box domain, Y2H-detected SKP1 binding, and AP-MS-detected CUL1 co-purification provides strong but indirect support for a functional SCF-FBXO34 complex. No crystal structure, targeted co-IP, or in vitro reconstitution of the trimeric SCF-FBXO34 complex has been published. The current evidence is sufficient for annotation of SCF complex membership but would benefit from targeted validation.

2. HNRNPU as the Best-Supported Degradative Substrate

Key Study: Yang et al. 2022 (PMID: 36285453)

"FBXO34 promotes latent HIV-1 activation by post-transcriptional modulation" — Emerging Microbes & Infections 11(1):2785-2799.

Evidence Lines

Substrate Evidence Grade: Strong (Tier 1)

HNRNPU meets all standard criteria for a bona fide E3 ligase substrate:
- Direct physical interaction (AP-MS)
- Direct ubiquitination (ubiquitination assay)
- Consequent degradation
- Functional phenocopy (knockout = overexpression of E3)
- Biological pathway validation

Important Mechanistic Context: HNRNPU's Dual Regulation by SCF Complexes

A remarkable finding from the literature is that HNRNPU has a contrasting relationship with a different F-box protein, β-TrCP (FBXW1):

• SCF-FBXO34 → HNRNPU is a true degradative substrate (ubiquitinated and degraded; PMID: 36285453)
• SCF-β-TrCP → HNRNPU is a stable pseudosubstrate (binds β-TrCP substrate pocket but is NOT degraded; PMID: 11850407, Davis et al. 2002)

In the β-TrCP context, HNRNPU occupies the substrate-binding WD domain of β-TrCP in a stoichiometric manner, stabilizes the E3, and determines its nuclear localization. This binding is competed by phospho-IκBα (a true substrate). The pseudosubstrate relationship serves as a regulatory mechanism controlling β-TrCP localization, stability, and substrate-binding threshold.

This dual regulation demonstrates that F-box protein identity, not merely substrate binding, determines whether HNRNPU is degraded or stabilized, providing a clear example of substrate fate determination by E3 ligase specificity.

Citation-gating note: the initial OpenScientist draft cited PMID: 42348677 for an HNRNPU NEDDylation/RNA-binding-repertoire claim. PubMed verification on 2026-07-06 shows that PMID resolves to a 2026 Science article titled "Ubiquitin-like proteins NEDD8 and SUMO2 control epithelial homeostasis, regeneration, and inflammation," not the stated "Winge et al. 2026, Cell" citation. Treat the HNRNPU NEDDylation claim as an unverified/miscited lead unless full text or another verified source confirms direct HNRNPU evidence. It should not be propagated into FBXO34-ai-review.yaml as fact from this report.

3. Other Reported FBXO34 Interactors: Detailed Assessment

3A. CCNB1/Cyclin B1 — Functional Target, Not Direct Substrate (PMID: 33842473)

Study: Zhao et al. 2021, "FBXO34 Regulates the G2/M Transition and Anaphase Entry in Meiotic Oocytes" (mouse oocyte system)

Assessment: CCNB1 is a functional pathway target but not a validated direct substrate. The rescue experiment shows that FBXO34 depletion leads to low MPF activity (CDK1/Cyclin B1), and CCNB1 overexpression can compensate. However, this is consistent with FBXO34 targeting an upstream inhibitor of MPF (e.g., a CDK inhibitor or Cyclin B1 degradation regulator) rather than CCNB1 directly. This work was performed in mouse oocytes; human relevance is inferred by sequence similarity.

Substrate evidence grade: Insufficient. Should not be annotated as a substrate without ubiquitination or degradation evidence.

3B. High-Throughput Binary Interactors (Y2H-Only)

Source 1: HuRI (PMID: 32296183)
Luck et al. 2020, "A reference map of the human binary protein interactome" — Nature 580:402-408.

17 partners detected, including:
- 10 KRTAPs (KRTAP4-5, 4-11, 4-12, 5-3, 9-2, 9-3, 9-8, 10-5, 10-8, 10-11, 12-4)
- ALPP, COL8A1, DISC1, FGF14, OIT3, VWC2L

Methods: Y2H array, two-hybrid prey pooling, validated two-hybrid (3 independent Y2H methods per pair in most cases)

Source 2: Rolland et al. (PMID: 25416956)
Rolland et al. 2014, "A proteome-scale map of the human interactome network" — Cell 159:1212-1226.

3 partners: MTUS2, KRT40, MDFI

Assessment: These are legitimate binary protein-protein interactions detected in well-controlled Y2H screens. However:
- No ubiquitination, degradation, half-life, or functional evidence for any of these partners
- The KRTAP enrichment (10/17 HuRI partners) is suspicious — KRTAPs are small, cysteine-rich proteins known to be promiscuous Y2H interactors. Their biological relevance to FBXO34 (a ubiquitin ligase adaptor) is unclear
- DISC1, FGF14, ALPP, COL8A1, OIT3 are biologically diverse proteins with no obvious connection to FBXO34's known ubiquitin ligase function

Substrate evidence grade: None. These should remain as non-core interaction leads in annotation databases.

3C. High-Throughput Co-complex Interactors (AP-MS)

Source: BioPlex 2.0/3.0 (PMID: 28514442, 33961781)
- Huttlin et al. 2017 and 2021

Partners: FBXO30, MYO6, YEATS4, NEXN, GSN, TMOD3, GAP43, ABRA, ERLEC1

Assessment: AP-MS co-purification indicates co-complex membership or proximity but does not distinguish direct binding from indirect association. CUL1 co-purification from the same studies is meaningful because it matches the expected SCF scaffold interaction. The other partners (cytoskeletal proteins MYO6/NEXN/GSN/TMOD3/GAP43/ABRA; ER lectin ERLEC1; chromatin factor YEATS4; another F-box protein FBXO30) lack any substrate-level evidence.

Substrate evidence grade: None.

3D. Other Detection Methods

Substrate evidence grade: None.

4. Summary Classification Table

5. Curation Recommendations

Current GO Annotations (Adequate)

• GO:0031146 — SCF-dependent proteasomal ubiquitin-dependent protein catabolic process (NAS): Correctly reflects FBXO34's role as an SCF adaptor

Recommended Additional Annotations

1. Substrate-specific annotation: HNRNPU should be annotated as a specific substrate of FBXO34 in GO/UniProt with experimental evidence (ECO:0000269) from PMID: 36285453. This supports a pathway-specific annotation beyond the broad "ubiquitin-like ligase substrate adaptor activity."

2. HIV-1 latency regulation: FBXO34's role in HIV-1 latency through HNRNPU degradation should be captured as a microbial infection-specific function (already present in UniProt FUNCTION field).

3. HTP interactors should NOT be converted to substrate annotations without additional evidence. The threshold for substrate annotation should require, at minimum:

4. Direct ubiquitination evidence (in vivo or in vitro ubiquitination assay)
5. Evidence of FBXO34-dependent degradation or protein level change

6. Ideally: functional rescue or reconstitution

7. Oocyte maturation function: The CCNB1 functional rescue (PMID: 33842473) supports a GO annotation for meiotic cell cycle regulation but should be qualified as "by similarity" (mouse data) and should NOT annotate CCNB1 as a substrate.

What Would Strengthen the FBXO34 Literature

• In vitro reconstitution of SCF-FBXO34 ubiquitination of HNRNPU
• Identification of the HNRNPU degron recognized by FBXO34
• Crystal or cryo-EM structure of FBXO34 alone or in complex with SKP1
• Cycloheximide chase experiments quantifying HNRNPU half-life ± FBXO34
• Identification of additional substrates beyond HNRNPU with full validation
• Confirmation of CCNB1 as direct or indirect target with mechanistic detail

6. Limitations

1. PMID: 36285453 (Yang et al. 2022) — The abstract was available but the full text was not systematically reviewed in this analysis. Specific experimental details (e.g., whether in vitro ubiquitination was performed, whether a degron motif was mapped) require full-text verification.

2. PMID: 33842473 (Zhao et al. 2021) — Mouse oocyte study; species extrapolation to human is annotated "by similarity" (ECO:0000250) in UniProt. The direct substrate in oocyte meiosis remains unidentified.

3. PMID: 42348677 was PubMed-verified as a real 2026 Science paper, but the report's original "HNRNPU NEDDylation regulation" wording is not verified by the PubMed title/metadata and may be miscited. Keep this as an unverified lead unless a curator checks the full text.

4. PMID: 40497153 was PubMed-verified as a 2025 Journal of Virus Eradication HIV-1 latency review. Treat it as background only unless full text is checked for a specific FBXO34 claim.

5. The HuRI/BioPlex interactors were assessed at the annotation level. Individual follow-up studies on any of these partners (beyond the original HTP screens) were not found in PubMed, suggesting they remain uncharacterized.

6. FBXO34 remains a relatively understudied F-box protein with only 5 primary PubMed entries as of this analysis, compared to dozens for well-characterized F-box proteins like β-TrCP, FBXW7, or SKP2.

7. Key Citations

Report generated: Iteration 1 of 1. Analysis based on PubMed literature search, UniProt (Q9NWN3), IntAct (89 interaction records, 38 unique partners), and BioGRID (58 interactions) as of July 2026.