# ABTB1 review notes

## Scope

ABTB1 is reviewed for the Human Proteostasis Network batch because the PN projection places it in the UPS Cul3 substrate receptor branch. The PN projection for ABTB1 is one candidate addition: `GO:1990756 ubiquitin-like ligase-substrate adaptor activity`, from `Ubiquitin Proteasome System|E3 ubiquitin and UBL ligases|Cul3 substrate receptor|BTB-BACK|ankyrin`.

I am treating that projection as context only for this review. The PN mapping audit marks the Cul3 substrate receptor mapping as requiring manual gene-level review before changing a gene review, and the UPS branch is domain-heavy. For ABTB1, local evidence supports BTB/POZ/ankyrin architecture, CUL3 interaction, and inferred ubiquitin ligase complex membership, but I did not find ABTB1-specific substrate identification or a direct assay showing that ABTB1 bridges CUL3 to a substrate. Therefore I do not add `GO:1990756` as a new annotation in this review.

## Evidence synthesis

ABTB1/BPOZ was cloned as a 478 aa protein with an N-terminal ankyrin repeat and two BTB/POZ domains [PMID:10891360 "A novel human gene containing an ankyrin repeat and BTB/POZ domains (BPOZ) was isolated from a human leukocyte cDNA library"; PMID:10891360 "Sequence pattern analysis shows that BPOZ contains an N-terminal ankyrin repeat, a bipartite nuclear localization signal and two BTB/POZ domains"]. The same paper found ubiquitous fetal expression and proposed involvement in protein-protein interaction/complex formation [PMID:10891360 "ubiquitously expressed in all fetal tissues examined"; PMID:10891360 "functionally involved in protein-protein interaction, perhaps in forming protein complexes"].

UniProt summarizes ABTB1 as cytoplasmic and as a possible mediator of PTEN growth-suppressive signaling [file:human/ABTB1/ABTB1-uniprot.txt "May act as a mediator of the PTEN growth-suppressive"; file:human/ABTB1/ABTB1-uniprot.txt "SUBCELLULAR LOCATION: Cytoplasm"]. The PTEN-response paper reports that BPOZ overexpression suppresses cancer cell growth and G1/S progression, while antisense knockdown accelerates growth [PMID:11494141 "BPOZ were able to suppress growth of cancer cells"; PMID:11494141 "over-expression of BPOZ inhibited progression of the cell cycle at the G(1)/S transition"; PMID:11494141 "Anti-sense oligonucleotides for BPOZ or EGR2 effectively inhibited their expression, and cell growth was accelerated"].

The strongest proteostasis-relevant evidence is CUL3 association plus domain architecture. UniProt lists an ABTB1-CUL3 interaction with multiple experiments [file:human/ABTB1/ABTB1-uniprot.txt "Q969K4; Q13618: CUL3; NbExp=5"]. The Bennett et al. CRL-network paper provides the broader cullin-ligase proteomics context, reporting that cullins are assembled with adaptor modules and that adaptor-module abundance drives CRL network organization [PMID:21145461 "a large fraction of cullins are assembled with adaptor modules"; PMID:21145461 "the abundance of adaptor modules, rather than cycles of neddylation and CAND1 binding, drives CRL network organization"]. This supports a specific replacement of generic CUL3-linked `protein binding` with `cullin family protein binding` and supports refining `ubiquitin ligase complex` toward `Cul3-RING ubiquitin ligase complex`. It does not, by itself, establish a known ABTB1 substrate.

The remaining IntAct protein-binding rows are mostly large-scale interactome evidence for EEF1A2, EEF1D, ZBTB21, ATXN1, TARDBP, and additional CUL3 interactions. These data support physical-interaction context but the generic `protein binding` term is not informative as a molecular function. I mark these rows as over-annotated rather than using them to infer translation elongation factor activity, neurodegeneration-specific function, or broad proteostasis function.

UniProt lists three GO rows that were not present in the QuickGO GOA file used to seed this review: nucleolus, plasma membrane, and translation elongation factor activity [file:human/ABTB1/ABTB1-uniprot.txt "GO; GO:0005730; C:nucleolus; IDA:HPA"; file:human/ABTB1/ABTB1-uniprot.txt "GO; GO:0005886; C:plasma membrane; IDA:HPA"; file:human/ABTB1/ABTB1-uniprot.txt "GO; GO:0003746; F:translation elongation factor activity"]. The two HPA localization rows may reflect broader or condition-specific localization, but they are not in the seeded GOA file. The translation elongation factor activity IEA looks especially suspect: ABTB1 binds EEF1A2 reproducibly [file:human/ABTB1/ABTB1-uniprot.txt "Q969K4; Q05639: EEF1A2; NbExp=12"], but local evidence does not show that ABTB1 itself has elongation factor activity.

## Falcon

Falcon deep research was started for ABTB1 with `perplexity-lite` fallback on 2026-06-03. Falcon timed out after 600 seconds and did not produce `ABTB1-deep-research-falcon.md`. The configured fallback then failed with a Perplexity API 401 quota error and did not produce `ABTB1-deep-research-perplexity-lite.md`.

Because both providers failed, this review relies on the local UniProt record, GOA file, cached publications, fetched PANTHER family data, and PN projection/mapping-audit files. If a later Falcon/fallback report identifies direct ABTB1 substrate-adaptor evidence, the main YAML should be updated to reconsider `GO:1990756`; otherwise this review remains intentionally conservative about the PN projection.

## Falcon deep research findings (2026-06-07)

A Falcon (Edison Scientific) deep research report was generated on 2026-06-07 (`ABTB1-deep-research-falcon.md`), superseding the earlier failed run noted above. Key findings, emphasizing what is NEW relative to the existing review:

- NEW (regulation): ABTB1 is a direct, experimentally validated target of miR-125b in myeloid cell models — ABTB1 mRNA down ~3.9-fold and protein down ~60% on miR-125b overexpression, up ~40% with miR-125b inhibitor, with a 3'UTR luciferase reporter confirming direct targeting; ABTB1 is described as an "anti-proliferative factor" [PMID:22689670 "we demonstrated that ABTB1, an anti-proliferative factor, is a new direct target of miR-125b"]. This is post-transcriptional regulation context not previously captured; it does not change any GO molecular-function/localization annotation.
- NEW (disease/clinical context): In prostate cancer, ABTB1 is an androgen-receptor (AR)-associated gene retained under androgen-deprivation adaptation, and TCGA PRAD multivariable Cox modeling links ABTB1 expression to progression-free survival (low vs high HR 0.46, 95% CI 0.28-0.74, p=0.0013), i.e. higher ABTB1 associated with worse PFS in that cohort [PMID:36809527 "Expressions of 47 AR-associated or association gaining genes were statistically significant markers for progression-free survival"]. This correlative oncology context is new; it contrasts with the tumor-suppressive/anti-proliferative narrative elsewhere and the report attributes the discrepancy to context/isoform complexity. Correlative only — does not justify a GO process annotation.
- CONFIRMS existing review (molecular function / complex): the BTB/POZ-as-CUL3-substrate-adaptor model (Geyer et al. 2003, Mol Cell, doi:10.1016/S1097-2765(03)00341-1) is cited as the family-level mechanistic framing, consistent with the existing CRL3-complex / cullin-binding refinements. This remains family-level inference; the report explicitly notes no ABTB1-specific CUL3 reconstitution or substrate was found. Geyer 2003 PMID could not be reliably resolved with available tools (the citation-lookup tool was denied; a DOI->PMID conversion did not return a PMID), so it is recorded here in notes only and NOT added to the YAML references to avoid an unverified identifier.
- CONFIRMS existing review (PTEN signaling): ABTB1 is reiterated as a mediator of PTEN growth-suppressive signaling with overexpression inhibiting growth and G1/S progression — already captured via PMID:11494141.
- PROVISIONAL (do not annotate): Open Targets genetics aggregation links ABTB1 to refractive error, asthma, and neurodegenerative disease (GWAS credible sets plus a glutamatergic-neuron CRISPRi peroxidized-lipid phenotype). These are platform prioritization scores, not effect sizes or mechanism; treated as leads only.
- GAP (consistent with existing review): the report found no robust subcellular localization beyond the cytoplasmic curation already annotated, and no direct ABTB1 ubiquitination substrate — reinforcing the conservative stance against adding GO:1990756.

PMID provenance verified via PubMed metadata for the two new primary references added to the YAML: PMID:22689670 (Bousquet 2012, Haematologica, "MicroRNA-125b transforms myeloid cell lines by repressing multiple mRNA.") and PMID:36809527 (Nätkin 2023, PLoS One, "Adaptive and non-adaptive gene expression responses in prostate cancer during androgen deprivation.").