KCTD14: CUL3 Substrate-Adaptor or Alternative Function?

Research question: What is the evidence for human KCTD14 molecular function—specifically, does it act as a CUL3‑RING ubiquitin ligase (CRL3) substrate adaptor like several KCTD paralogs, or does it have another activity? What GO‑style curation posture is justified?

Date: 2026‑07‑07 · Iteration 1 (literature synthesis; no primary data files provided)

1. Summary (answer)

There is no direct experimental evidence that human KCTD14 binds CUL3, oligomerizes (beyond prediction), recognizes a substrate, or participates in a CRL3 complex. The entire KCTD14 primary literature is transcriptomic/biomarker work with no biochemical characterization. Its BTB/T1 fold and its close relationship to the validated Cul3 adaptor KCTD7 make a CRL3 substrate‑adaptor role a plausible, high‑priority hypothesis, but family membership is not sufficient to assert it, because Cul3 binding is demonstrably not universal among KCTDs (KCTD1 and KCTD16 do not bind Cul3). The only functionally defensible molecular‑function annotation today is protein homo‑oligomerization via the BTB/T1 domain, and even that rests on AlphaFold prediction plus family‑level inference rather than KCTD14‑specific experiment. KCTD14 should remain largely MF‑dark: keep non‑core "protein binding," optionally add homo‑oligomerization at an inferred (IBA/computational) evidence level, and do not annotate a specific adaptor/substrate‑recognition function.

2. Key findings and evidence

2.1 The KCTD14 primary literature is functionally empty

A systematic PubMed search returns only ~9 records mentioning KCTD14. All are transcriptomic signature / prognostic‑biomarker studies:
- Pancreatic‑cancer DC signature, "KCTD14–TNF axis" (correlative, not mechanistic) — PMID 41080575 (2025).
- Ovarian‑cancer methylation prognostic panel — PMID 36978087 (2023).
- Prostate‑cancer CAF signature — PMID 36195908 (2022).
- Dengue transcriptome biomarkers — PMIDs 39397194 (2025), 35220956 (2022).
- Diabetic rat ileal/colon transcriptome — PMID 34806320 (2021).
- Mouse oocyte/follicle expression — PMID 40044995 (2025).
- Incidental co‑regulation in a Znf230‑KO transcriptome — PMID 25505846 (2014).

None report protein interactions, enzymatic activity, oligomerization, or a CUL3 relationship. → KCTD14 is MF‑dark.

2.2 No KCTD14–CUL3 binding has ever been tested

The definitive family binding survey — Ji, Chu, Nielsen, Benlekbir, Rubinstein & Privé, "Structural Insights into KCTD Protein Assembly and Cullin3 Recognition" (PMID 26334369, 2016) — measured Cul3 binding for KCTD1, 5, 6, 9, 16, 17 but did not include KCTD14. No pulldown, ITC, cryo‑EM, or proteomic study places KCTD14 in a CRL3 complex.

2.3 Cul3 binding is member‑specific → family transfer is unreliable

From the same study (PMID 26334369): "the KCTD proteins 5, 6, 9 and 17 bind to Cul3 with high affinity, while the KCTD proteins 1 and 16 do not have detectable binding," and KCTDs "do not share many of the previously identified determinants for Cul3 binding." Because two tested KCTDs (KCTD1, KCTD16) fail to bind Cul3 despite the shared BTB fold, homology‑based inference of adaptor function to the untested KCTD14 is not safe.

2.4 KCTD14 oligomerization is predicted, not observed

Esposito, Balasco & Vitagliano, "AlphaFold Predictions Provide Insights into the Structural Features of the Functional Oligomers of All Members of the KCTD Family" (PMID 36362127, 2022) predict a reliable pentameric KCTD14 assembly, but with an atypical CTD: "the structure of the related proteins KCTD7 and KCTD14, although pentameric, appears to be characterized by a different organization of the CTD region, with the five chains forming a circle‑like structure with a large cavity." Family‑wide negative‑stain EM (Smaldone et al., PMID 27152988, 2016) shows KCTD BTB domains across five clades "prevalently assume pentameric states." Together these support homo‑oligomerization as an inherited fold property, but there is no KCTD14‑specific experimental oligomerization data, and the atypical CTD hints KCTD14/KCTD7 may engage partners non‑canonically.

2.5 The closest paralog KCTD7 is a validated Cul3 partner (hypothesis, not proof)

• Staropoli et al. (PMID 22748208, 2012): an NCL/PME disease mutation "abrogated interaction with cullin‑3, a ubiquitin‑ligase component and known KCTD7 interactor."
• Jiang, Wang, Kong & Zheng (PMID 37450587, 2023): cryo‑EM structures of the "KCTD5–Gβγ fusion complex and the KCTD7–Cul3 complex."

Because AlphaFold groups KCTD14 with KCTD7 by CTD architecture, KCTD7's proven Cul3 partnership makes a CRL3 role for KCTD14 a strong, testable prediction — but strictly a prediction, especially given §2.3.

2.6 Reliability of high‑throughput "protein binding" interactors

Reported KCTD14 GO "protein binding" annotations trace to high‑throughput/curated PPI compilations (e.g., Skoblov et al., PMID 23592240, 2013, which explicitly notes "The value of manual curation of PPIs and analysis of existing high‑throughput data should not be underestimated" but offers only hypotheses). Such Y2H/AP‑MS hits for a low‑expression, uncharacterized protein are prone to false positives, are non‑reciprocated, and none has been validated as a physiological KCTD14 substrate or partner. They justify only a non‑core, low‑confidence "protein binding" annotation, not a substrate‑recognition function.

3. Hypotheses: supported vs. refuted

4. Recommended GO‑style curation posture

1. Keep KCTD14 essentially MF‑dark. Retain only non‑core, low‑confidence protein binding (GO:0005515) from HT data, clearly flagged as unvalidated.
2. Best‑justified positive annotation: protein homo‑oligomerization (GO:0051260) / identical protein binding, at an inferred/computational evidence level (AlphaFold prediction + family IBA), not IDA.
3. Do NOT annotate Cul3‑RING ubiquitin ligase binding, ubiquitin‑protein transferase/ligase adaptor activity, or any specific substrate‑recognition MF. These are unsupported and would over‑annotate.
4. Record the CRL3 substrate‑adaptor role as an open hypothesis (paralogy to KCTD7), pending direct experiment.

5. Limitations and future directions

Limitations. This is a literature‑only synthesis (no primary data). PMIDs below were confirmed to resolve via the PubMed search tool; DOIs are supplied from model knowledge and should be independently verified before publication‑grade use (flagged). Absence of evidence for KCTD14 reflects the field's neglect of this gene, not a demonstrated negative.

Experiments that would resolve the question.
- Co‑IP / ITC / AP‑MS of tagged KCTD14 vs. CUL3 (± the BTB determinants mutated) — the exact assay Ji 2016 applied to other KCTDs.
- Reconstituted in‑vitro ubiquitination with CUL3–RBX1–E2 to test ligase‑adaptor activity.
- SEC‑MALS/native‑MS/cryo‑EM to confirm the predicted pentamer and the atypical CTD cavity.
- Unbiased BioID/TurboID interactome in a physiologically relevant tissue to identify candidate substrates and test whether any HT "protein binding" hits reciprocate.

6. Citations (PMIDs confirmed via PubMed; DOIs to be verified)

• PMID 26334369 — Ji et al. 2016, J Biol Chem — KCTD assembly & Cul3 recognition; member‑specific binding. DOI (verify): 10.1074/jbc.M115.669580.
• PMID 36362127 — Esposito, Balasco, Vitagliano 2022, Int J Mol Sci — AlphaFold oligomers of all KCTDs incl. KCTD14. DOI (verify): 10.3390/ijms232113346.
• PMID 27152988 — Smaldone et al. 2016, Biochim Biophys Acta Mol Cell Res — KCTD BTB domains are pentameric; KCTD6–Cul3 pinwheel. DOI (verify): 10.1016/j.bbamcr.2016.04.023.
• PMID 22748208 — Staropoli et al. 2012, Am J Hum Genet — KCTD7–cullin‑3 interaction; NCL/PME. DOI (verify): 10.1016/j.ajhg.2012.05.023.
• PMID 37450587 — Jiang, Wang, Kong, Zheng 2023, Sci Adv — cryo‑EM KCTD5–Gβγ and KCTD7–Cul3. DOI (verify): 10.1126/sciadv.adg8369.
• PMID 24747150 — Balasco et al. 2014, PLoS One — KCTD5–Cul3 5:5 heterodecamer, Kd ≈ 59 nM. DOI (verify): 10.1371/journal.pone.0091093.
• PMID 23592240 — Skoblov et al. 2013, BioEssays — KCTD PPI curation (hypothesis‑generating). DOI (verify): 10.1002/bies.201200169.
• PMID 24268103 — Liu, Xiang, Sun 2013 — KCTD family review (structure/function/disease). DOI (verify): 10.1007/s12035‑013‑8555‑y.
• PMID 34001146 — Angrisani et al. 2021, Semin Cancer Biol — KCTDs in cancer review. DOI (verify): 10.1016/j.semcancer.2021.05.019.
• PMID 40925965 — Jiang & Zheng 2026 — KCTDs as regulators of GPCR biased signaling (Gβγ degradation). DOI: not confirmed (flag as uncached).
• KCTD14 transcriptomic references: PMIDs 41080575, 40044995, 39397194, 36978087, 36195908, 35220956, 34806320, 25505846.

Suspect/uncached flags: All DOIs above are reconstructed from model knowledge and must be verified against CrossRef/PubMed. PMID 40925965 (2026) and PMID 41080575 (2025) are recent; treat their DOIs as unconfirmed. No previously reported KCTD14 citation was accepted without re‑confirmation in this PubMed session; any claim of a direct KCTD14–CUL3 experiment in secondary sources should be treated as suspect because none was found here.