# ASCC3 review notes ## Summary Automated deep research status: `just deep-research-falcon human ASCC3 --fallback perplexity-lite` was run for this review. Falcon timed out after 600 seconds, and the configured `perplexity-lite` fallback failed with an API quota 401. No provider deep-research artifact was produced, so this review uses the cached UniProt, GOA, Reactome, and primary publication files plus the manual notes below. ASCC3 is a large ATP-dependent SF2 helicase with two well-supported functional contexts. In the nucleus, it is the catalytic helicase subunit of the ASCC-ALKBH3 DNA dealkylation repair complex. Dango et al. show that ASCC3 unwinds DNA to provide the single-stranded substrate used by ALKBH3 for repair [PMID:22055184 "ASCC3 unwinds DNA to generate the single-stranded substrate needed for ALKBH3-mediated DNA repair"]. Brickner et al. further show alkylation-specific recruitment of ASCC3 to nuclear repair foci [PMID:29144457 "Endogenous ASCC3 formed nuclear foci upon treatment of U2OS cells with the alkylating agent"]. In the cytosol, ASCC3 is the hRQT/RQT2 ATPase subunit of the ribosome quality control trigger complex. Hashimoto et al. define hRQT as ASCC3, ASCC2, and TRIP4 [PMID:32099016 "The hRQT complex is composed of ASCC3, ASCC2, and TRIP4"], and Juszkiewicz et al. show that ASCC acts on ubiquitinated collided ribosomes to disassemble the lead ribosome [PMID:32579943 "ASCC acts on ubiquitinated collided ribosomes to selectively disassemble the lead ribosome"]. Narita et al. refine this as K63-uS10-dependent hRQT-mediated subunit dissociation [PMID:36302773 "the trimeric hRQT complex-mediated subunit dissociation"]. ## PN projection The PN workbook places ASCC3 under `Translation > Cytosolic translation > Ribosome-associated QC > Ribosomal rescue` with UniProt ID Q8N3C0 and synonym RQT2. The local projection report maps this path to two GO targets: `GO:0072344 rescue of stalled cytosolic ribosome`, which is already in GOA, and the broader `GO:0006515 protein quality control for misfolded or incompletely synthesized proteins`, listed as new to GOA. Conservative decision: retain the exact ribosome-rescue/RQC annotations and do not add a new ASCC3 annotation to `GO:0006515`. ASCC3 has direct evidence for splitting K63-ubiquitinated collided ribosomes and thereby initiating RQC, but the existing GOA already captures this more specifically with `GO:0072344`, `GO:0032790`, and `GO:1990116`. The broad PN projection is biologically compatible as an ancestor/context term but would be redundant and less informative as a new gene-level assertion. ## Annotation decisions - Accept the DNA repair core: `GO:0043138 3'-5' DNA helicase activity`, `GO:0006307 DNA alkylation repair`, nuclear/nucleoplasm/nuclear-speckle locations, and `GO:1990391 DNA repair complex`. - Accept the RQC core: `GO:0180022 RQC-trigger complex`, `GO:0016887 ATP hydrolysis activity`, `GO:0022626 cytosolic ribosome`, `GO:0032790 ribosome disassembly`, `GO:0072344 rescue of stalled cytosolic ribosome`, and `GO:1990116 ribosome-associated ubiquitin-dependent protein catabolic process`. - Treat generic binding annotations conservatively. `protein binding` rows are interaction evidence but not informative molecular-function curation. Broad nucleic-acid/helicase terms should defer to the specific DNA helicase/ATPase/RQT annotations. - Remove the `DNA replication` annotation: the cited ASCC1/ASCC damage-response paper supports alkylation repair-complex regulation, not a general DNA replication process for ASCC3. - Remove the high-throughput `membrane` localization: ASCC3 is a soluble nuclear/cytosolic protein with no transmembrane-domain basis, and stronger curated evidence places it in ASCC/hRQT contexts. ## Falcon deep research findings (2026-06-07) A Falcon (Edison Scientific) deep-research report was generated (`ASCC3-deep-research-falcon.md`) and reviewed against the existing COMPLETE review. PMIDs below were resolved from the report's DOIs via the PubMed ID-converter and verified for exact titles; none of these five papers are cached in `publications/`, so they are added to `references:` as statement-only entries (`full_text_unavailable: true`, no `supporting_text`) and are NOT used to alter any annotation action. - CONFIRMS (core MF/architecture): ASCC3 is a dual-cassette ("dual-engine") Ski2-like helicase whose two tandem NTPase/helicase cassettes are both catalytically active and can thread one substrate through both cassettes; TRIP4 docks via a Zn-finger and positions an ASCH domain next to the C-terminal cassette to stimulate the motor [PMID:37019967 Jia 2023 "ASCC3 can thread substrates through both helicase cassettes... TRIP4 docks on ASCC3 via a zinc finger domain"]. This reinforces the existing `GO:0043138`/`GO:0016887` MF calls; no annotation change. - NEW (partner-driven routing): TRIP4 and ALKBH3 bind ASCC3 mutually exclusively, providing a mechanism that routes the same helicase between the ALKBH3 alkylation-repair arm and TRIP4-associated functions [PMID:37019967 Jia 2023 "TRIP4 binds ASCC3 mutually exclusively with the DNA/RNA dealkylase, ALKBH3"]. Useful context for the two-arm model already captured; does not require a new GO term. - NEW (translation initiation arm): The ASC-1 complex/ASCC3 also associates with scanning ribosomes and promotes translation initiation; ASCC3 localizes predominantly to 5' UTRs and its knockdown impairs 43S preinitiation-complex loading and scanning for a subset of transcripts [PMID:37092320 Kito 2023 "ASCC3... localizes predominantly to the 5' untranslated region... knockdown impairs 43S preinitiation complex loading and scanning dynamics"]. This is a distinct, non-RQC role; recorded as context and a suggested question rather than a new gene-level annotation (single study, mechanism still emerging). - CONFIRMS/refines (RQC mechanism): In vitro reconstitution shows ribosomal collision is not strictly required for ZNF598 ubiquitination or ASCC-mediated disassembly; ASCC can split polysomes, monosomes, 80S elongation complexes, and even 48S initiation complexes once uS10 is K63-polyubiquitinated, provided >=30-35 nt of 3' mRNA downstream of the P site [PMID:38366554 Miscicka 2024 "ribosomal collision is not a strict prerequisite... as long as such complexes contained >= 30-35 3'-terminal mRNA nt downstream from the P site"]. Consistent with existing `GO:0032790`/`GO:0072344` ribosome-rescue calls; refines mechanism, no action change. - NEW (disease link): Biallelic loss-of-function ASCC3 variants cause an autosomal-recessive neurodevelopmental/neuromuscular syndrome (MIM:620700) with developmental delay, intellectual disability, hypotonia and motor/language impairment; three new Chinese families with compound-heterozygous variants reported [PMID:39286456 Li 2024 "neuromuscular syndrome caused by biallelic variants in ASCC3... compound heterozygous variants"]. Supports the biological importance of ASCC3 helicase function; added as reference + suggested question, no GO annotation change. - PROVISIONAL (cancer/immune, single study, distinct pathway): In NSCLC, ASCC3 is reported upregulated and to stabilize STAT3 by recruiting CAND1 (inhibiting STAT3 ubiquitination), dampening the type I interferon response and fostering an immunosuppressive microenvironment [PMID:38148115 Ao 2023 "ASCC3 stabilized STAT3... by recruiting CAND1... impairing the type I interferon response"]. This is a single mechanism-of-disease study describing an indirect/regulatory role not clearly tied to the helicase core; recorded as reference and a suggested question only — deliberately NOT projected to any new MF/BP annotation.