cact (Cactus, Q03017) review notes

Identity

• Drosophila melanogaster IkappaB-family protein, FBgn0000250; 500 aa; N-terminal signal-responsive region, five ankyrin repeats, C-terminal PEST region (UniProt Q03017).
• UniProt: "Belongs to the NF-kappa-B inhibitor family"; PANTHER PTHR46680:SF3 "NF-KAPPA-B INHIBITOR CACTUS".
• Isoforms A (maternal/zygotic) and C (zygotic).

Core biology (with provenance)

• Cactus binds Dorsal and holds it in the cytoplasm: PMID:7705656
• Mechanism in yeast reconstitution: PMID:7479760
• Cactus is the inhibitor of both Rel proteins downstream of Toll: PMID:24086459
• Signal-dependent degradation is the switch: PMID:7705656; gradient of Cactus degradation shapes the Dorsal nuclear gradient PMID:9025065.
• Kinase: Pelle (IRAK homolog), not IKK: PMID:24086459; PMID:24086459. Slimb (betaTrCP) is required for degradation in S2 cells.
• Loss of function ventralizes embryos; cactus acts via dorsal: PMID:1794309.
• Immunity: spz/Toll/cact cassette controls drosomycin PMID:8808632; fat-body mosaic PMID:10369678. cactus is itself a Toll target (autoregulation) PMID:9553105; feedback via Bre1/Rad6 PMID:36516751.
• Hematopoiesis: Cact knockdown in prohemocytes drives differentiation PMID:27163255.
• NMJ: Dorsal and Cactus concentrated in postsynaptic side of larval NMJ PMID:10192771; levels change with synaptic activity, possibly Toll-independent PMID:12532402.
• Partners (IPI): Dorsal, Dif, cactin PMID:10842059, Kurtz beta-arrestin PMID:20802461, IKKbeta/DLAK in LPS-stimulated cells PMID:10636911.

Curation decisions (summary)

• MF core: GO:0140311 protein sequestering activity + GO:0051059 NF-kappaB binding (ACCEPT IBA and experimental rows).
• BP core: GO:0045751 negative regulation of Toll signaling pathway (ACCEPT all). Dorsoventral patterning ACCEPT.
• GO:0043124 negative regulation of canonical NF-kappaB signal transduction (IBA, IMP): the GO definition is IKK-dependent; the fly Toll pathway uses Pelle as the Cactus kinase and IKK is dispensable (PMID:24086459). MODIFY to GO:0045751.
• protein binding rows: Dorsal partner rows -> MODIFY to NF-kappaB binding; cactin, Kurtz, IKKbeta rows -> REMOVE as uninformative.
• NMJ / subsynaptic reticulum / hemocyte differentiation: KEEP_AS_NON_CORE.
• HMP from the Schupbach & Wieschaus female-sterile screen (oogenesis, dorsal appendage formation): abstract does not mention cactus, full text not cached -> UNDECIDED.

Deep research

The review was drafted from the UniProt record and cached publications before Falcon deep research (cact-deep-research-falcon.md) was finished. The report was read afterwards and agrees with the curation decisions: cytoplasmic IkappaB that sequesters Dorsal/Dif through its ankyrin repeats, Toll-induced N-terminal phosphorylation and proteasomal degradation, and Toll-independent turnover through the C-terminal PEST region (CKII, Calpain A). New points it raises that do not change any annotation:
- Modelling (Barros et al. 2021) and facilitated-diffusion work (Carrell 2016; Schloop 2020) suggest the Dorsal-Cactus complex also acts as a mobile carrier that concentrates Dorsal ventrally. In this view Cactus promotes Toll-dependent Dorsal import on the ventral side while inhibiting basal import elsewhere.
- In Yorkie/polarity-deficient tumour models (a bioRxiv preprint), Cactus acts upstream of JNK.
Neither is backed by a cached primary paper here, so neither was used as annotation evidence. The report calls the direct Cactus kinase "not definitively established". That conflicts with PMID:24086459, which shows in vitro that Pelle phosphorylates the signal-responsive serines. The review follows the primary paper.