{"uniprot":"Q96C57","gene":"CUSTOS","name":"Protein CUSTOS","organism":"Homo sapiens (Human)","status":"dark","length":262,"sequence":"MAAPSGTVSDSESSNSSSDAEELERCREAAMPAWGLEQRPHVAGKPRAGAANSQLSTSQPSLRHKVNEHEQDGNELQTTPEFRAHVAKKLGALLDSFITISEAAKEPAKAKVQKVALEDDGFRLFFTSVPGGREKEESPQPRRKRQPSSSSEDSDEEWRRCREAAVSASDILQESAIHSPGTVEKEAKKKRKLKKKAKKVASVDSAVAATTPTSMATVQKQKSGELNGDQVSLGTKKKKKAKKASETSPFPPAKSATAIPAN","function_summary":"A human regulatory scaffold that suppresses canonical Wnt signaling by assembling protein complexes at the nuclear envelope. By using multivalent binding rather than catalysis, it positions and retains pathway effectors at the nuclear periphery, promoting β-catenin turnover or preventing its productive assembly with transcriptional coactivators. This perinuclear checkpoint dampens Wnt-driven transcriptional outputs by spatially restricting signaling flux at the nuclear membrane.","plain_language":"CUSTOS has barely been studied. The model's best guess: A human regulatory scaffold that suppresses canonical Wnt signaling by assembling protein complexes at the nuclear envelope. By using multivalent binding rather than catalysis, it positions and retains pathway effectors at the nuclear periphery, promoting β-catenin turnover or preventing its productive assembly with transcriptional coactivators.","uniprot_summary":"Involved in the inhibition of Wnt signaling.","go_terms":{"molecular_function":[{"id":"GO:0005515","label":"protein binding"}],"biological_process":[{"id":"GO:0090090","label":"negative regulation of canonical Wnt signaling pathway"}],"cellular_component":[{"id":"GO:0031965","label":"nuclear membrane"}]},"interpro":[{"id":"IPR026694","name":"Protein CUSTOS","type":"family","start":1,"end":262}],"reasoning":"I begin with the InterPro architecture: IPR026694 (Protein CUSTOS family) spans residues 1–262 and covers the entire polypeptide. A single full-length family signature without embedded catalytic domains indicates a non-enzymatic regulator. CUSTOS-family proteins are characterized by membrane-associated scaffolding that modulates signaling pathways through selective protein–protein interactions rather than enzymatic chemistry. This architecture therefore drives a binding-centric mechanism, consistent with GO:0005515 protein binding as the primary molecular function.\n\nA binding-driven scaffold at a membrane interface can impose pathway specificity by sequestering or relocalizing effectors. The nuclear envelope placement inferred below provides the spatial logic: anchoring a signaling regulator at the nuclear membrane allows it to intercept cytoplasmic effectors that traffic to the nucleus. Such spatial control naturally produces negative regulation of the canonical Wnt pathway (GO:0090090). In canonical Wnt signaling, β-catenin accumulates and translocates to the nucleus to coactivate transcription with TCF/LEF factors. A nuclear-envelope-tethered CUSTOS-family scaffold can suppress this outcome by retaining β-catenin at the periphery, promoting its turnover, or preventing assembly of nuclear β-catenin–TCF complexes. Thus, the domain architecture and localization together cause inhibition of Wnt-driven transcriptional outputs.\n\nCellular placement follows from both the family’s membrane-tethering behavior and the experimentally grounded label of a nuclear-envelope resident. A scaffold that restrains nuclear entry of β-catenin is most effective when stationed at the nuclear membrane (GO:0031965), where it can couple to nuclear pore and lamina-associated complexes. This location explains how the protein exerts pathway control: it creates a perinuclear sink that biases signaling flux away from transcriptional activation.\n\nMechanistically, I hypothesize that the CUSTOS-family fold forms a perinuclear platform that binds β-catenin and associated adaptors, recruiting ubiquitin–proteasome machinery and nuclear transport modulators. By concentrating β-catenin near the nuclear envelope, the scaffold can enhance its degradation or prevent productive complex formation, thereby dampening canonical Wnt signaling. Likely partners include β-catenin itself and its regulators (e.g., APC/Axin/GSK3 complexes), nuclear pore components that gate cargo flux, and lamina-associated factors that stabilize perinuclear assemblies. Together, these interactions implement a membrane-proximal checkpoint that curtails Wnt-responsive transcription.","category":"dark","control_evaluation":null,"validation_plan":null,"structure":{"source":"AlphaFold DB","pdb":"/structures/AF-Q96C57.pdb","alphafold":"https://alphafold.ebi.ac.uk/entry/Q96C57"},"links":{"uniprot":"https://www.uniprot.org/uniprotkb/Q96C57/entry","atlas":"/p/Q96C57"},"provenance":{"model":"BioReason-Pro (ESM3 esm3-sm-open-v1)","source_list":"Pharos Tdark","note":"AI prediction, not experimentally verified"}}