The normal human Q9H211 intake supplies 60 unique annotation assertions and no alternative-product section. Its 61 GOA records include one Hec1-binding assertion twice, differing only in database attribution/date (IntAct and CAFA); the deterministic seed collapses that duplicate. All source identifiers, terms, evidence codes, partners and qualifiers are retained. No NEW assertion is added. The 60 decisions are 27 ACCEPT, 29 KEEP_AS_NON_CORE, one MODIFY, one MARK_AS_OVER_ANNOTATED and two UNDECIDED.
The source evidence extract preserves exact GOA joins and the bounded IntAct fields behind target-pair checks, including query URL/hash, record IDs, host organisms, tags and negative flags. It contains no interaction-comment prose. Current-query missing pairs are not treated as binding-negative results. Supported generic binding is retained under the ClinGen campaign instruction, pinned for this review at main 3eb71be. A partner’s kinase, ligase, acetyltransferase or deubiquitinase activity is not assigned to CDT1 from association alone.
A genuine Falcon research attempt timed out; the explicit Perplexity-lite fallback failed with an HTTP401 quota response. No provider report was fabricated. The following manual evidence record is the research basis. Existing shared publication/Reactome caches were selected from authenticated main bytes; genuine new normal outputs were used only where absent. An abstract-only cache does not mean external full text was unavailable: successful indexed primary reading is identified below, and no cache was rewritten to simulate it.
CDT1 cooperates with ORC and CDC6 to load MCM2–7 onto DNA. Purified human reconstitution separates recruitment from high-salt-resistant loading, and omission of CDT1 disrupts loading. Full-length and later truncated constructs in the 2024 study are distinct preparations. This directly performed co-loader role supports the pre-RC assembly core; it does not give CDT1 an intrinsic ATPase or helicase activity. PMID:39604729(https://pubmed.ncbi.nlm.nih.gov/39604729/), selected full Results/Fig.1 and expression/loading Methods.
The 2026 human reconstitution tests WT CDT1 and the MGS variants R462Q/E468K; R453W was not successfully purified for the assay. R210C is derived from a fly mutant and is not identified here as a human MGS allele. Recruitment/loading deficits are measured, whereas suggested interface/salt-bridge explanations are structural interpretations. [PMID:41448435 Molecular impacts of Meier-Gorlin syndrome mutations on human origin licensing, "MGS variants of CDT1 specifically reduce MCM recruitment"].
Direct DNA binding is grounded in mouse biochemical experiments and separately corroborated by human recombinant-protein DNA-cellulose assays. Mouse full-length 557-aa Cdt1 and mapped deletions were tested by DNA-cellulose and mobility shift; this does not establish sequence-specific origin recognition. Human 2026 experiments use tagged WT/mutant CDT1, with the DNA-binding control distinguished from cellular synthesis inhibition. PMID:12192004(https://pubmed.ncbi.nlm.nih.gov/12192004/), indexed full primary Methods/Figs.2,4–6; PMID:41004347(https://pubmed.ncbi.nlm.nih.gov/41004347/), full Materials/Methods and selected Results/Fig.4. The human dynamic-chromatin study’s authentic abstract independently supports G1 chromatin association; its original full body was not recovered. PMID:17318181(https://pubmed.ncbi.nlm.nih.gov/17318181/).
The InterPro-derived preinitiation-complex assembly assertion is refined to mitotic pre-replicative complex assembly (GO:1902985). Current GO definitions distinguish the licensing/loading step from downstream preinitiation assembly enabling origin opening and helicase activation. These are distinct branches, so this is a functional-stage refinement rather than a claimed direct-parent replacement. The original IEA/source is unchanged and the existing experimental pre-RC assertion is retained.
CDT1 itself can restrain DNA synthesis. Human-cell experiments and purified human replication systems in PMID:36608667 distinguish this effect from checkpoint signaling and rereplication. CDT1 inhibits established forks and a minimal CMG/Pol-epsilon/RPA system, while Pol-epsilon-only primer extension is unaffected; MCM-contacting domains are required. The purified CDT1 preparation is 2×FLAG human protein from insect cells. ATR/WEE1 inhibition and CDC6-degron experiments provide separate cellular controls. These support an enzyme-inhibitor core and retention of the existing negative replication assertion. They do not provide a CMG-alone unwinding assay or a solved inhibitory structure; the paper explicitly leaves altered helicase–polymerase coupling as an alternative. PMID:36608667(https://pubmed.ncbi.nlm.nih.gov/36608667/), selected full Results and STAR protein/replication Methods.
The additional human MCF-7 study uses engineered nondegradable, licensing-impaired and geminin-nonbinding CDT1 constructs, with MCM-contacting mutants as controls. EdU/flow-cytometry and binding experiments corroborate synthesis restraint. This engineered overexpression context is not substituted for a measurement of its unperturbed physiological magnitude. PMID:41004347(https://pubmed.ncbi.nlm.nih.gov/41004347/).
The original 2009 source combines a human truncated CDT1–geminin structure, Xenopus licensing assays and separately described human U2OS overexpression-induced G2 arrest. Its supplementary Fig.S9 pixels and detailed SI Methods were not independently recovered; the explicit main Results description was read. The cell-cycle annotation remains contextual. New human licensing reconstitution shows inhibition with a CDT1–geminin heterotrimer, so an obligatory inhibitory hexamer is not asserted. This qualifies the older mechanism without denying binding or CDT1’s independently demonstrated fork-inhibitory function. PMID:19906994(https://pubmed.ncbi.nlm.nih.gov/19906994/), indexed full main text; PMID:41365879(https://pubmed.ncbi.nlm.nih.gov/41365879/), full selected Results/Figs.1–3.
The older checkpoint citation measures binding/loading: [PMID:14672932 The regulated association of Cdt1 with minichromosome maintenance proteins and Cdc6 in mammalian cells, "Cdc6 and Cdt1 are mutually dependent on one another"]. Actual primary Methods/Results/Figs.1–6 were read through the indexed author-uploaded article. Those assays support licensing rather than the signal-transduction step required by GO:0000076; that source-specific IDA assertion is over-annotated. The PAINT checkpoint node remains unresolved because the complete ancestral reconstruction and fly experimental grounding were not re-adjudicated. The combined electronic chain is also unresolved despite inspection of the mouse binding source. A short donor list or human self-donor is not a flaw, and licensing perturbation is not proof that CDT1 can never influence checkpoints.
G2-specific CDT1 perturbation after normal S phase, rescue, Hec1-loop mutants and cold-stable attachment measurements separate the mitotic function from prior replication defects. The source states [PMID:22581055 Recruitment of the human Cdt1 replication licensing protein by the loop domain of Hec1 is required for stable kinetochore-microtubule attachment, "human Cdt1 has a separable essential mitotic function"]. Human HeLa experiments are distinguished from PTK2/LLCPK1 localization. Hec1 interaction is supported by human two-hybrid, GST pull-down and endogenous co-IP; NUF2 recovery is co-complex evidence, not a purified binary interface.
Independent purified-protein work demonstrates direct microtubule binding by tag-cleaved full-length human CDT1, using porcine-brain microtubules and no-MT/GST/GFP controls. GFP-truncated TIRF experiments and cellular phosphomutant rescue are separate assays. This supports GO:0008017 in the core. Because this MF is already a descendant of the existing generic protein-binding class, no redundant NEW row is added, and Hec1-binding assertions are preserved as interaction evidence. PMID:30154187(https://pubmed.ncbi.nlm.nih.gov/30154187/), full Fig.1E/F and selected Results/Methods. Kinetochore is used as an anatomical location, not misfiled as a protein-containing complex.
The exact positive screens comprise human proteins tested in HeLa or 293T affinity-purification contexts, or in yeast-host two-hybrid assays. Each source-specific reason names its record and assay context. Direct small-scale evidence separately supports PCNA, MCM6 and GRWD1 interactions. The MCM6 NMR/pull-down experiment uses human CDT1 410–445 and MCM6 708–821; the GRWD1 paper includes purified binding but did not detect the endogenous CDT1 co-IP in its stated experiment. PMID:16482215(https://pubmed.ncbi.nlm.nih.gov/16482215/), full Fig.7; PMID:20202939(https://pubmed.ncbi.nlm.nih.gov/20202939/), authentic abstract plus indexed primary Methods; PMID:25990725(https://pubmed.ncbi.nlm.nih.gov/25990725/), full Fig.2.
CDT1–FAF1 association is inspected in the human nuclear-lysate/p97-trap experiment, separately from the paper’s worm work. The USP37 paper detects endogenous CDT1 with tagged USP37 and studies its substrate turnover. PMID:26842564(https://pubmed.ncbi.nlm.nih.gov/26842564/), full selected Fig.6/Methods; PMID:27296872(https://pubmed.ncbi.nlm.nih.gov/27296872/), full Results3.2/Methods. The sorbitol study’s genuine full-text variant supports stress-dependent loss of HBO1 association and retained geminin association; its main text describes PCNA/SKP2 supplementary results, but the Fig.S4 pixels were not inspected. PMID:21856198(https://pubmed.ncbi.nlm.nih.gov/21856198/).
The original 22581055 geminin-binding assay linkage is unresolved in the inspected main body, although the same binding pair is independently established by human structural and pull-down studies. Lack of geminin at mitotic kinetochores is not interpreted as lack of binding. Similarly, the Idas paper’s seeded partner is geminin, not Idas. These supported pairs remain non-core with source limits visible.
Nuclear activity is supported by performed licensing and inhibition, not localization alone. HPA nucleoplasm and nuclear-body observations are retained; nucleoplasm is central, whereas nuclear-body localization is contextual. All six Reactome nucleoplasm sources retain that common core location, with independent nuclear evidence. Those pathway events do not assign other participants’ ubiquitination, phosphorylation or transcriptional activities to CDT1. The LSD1 source uses a CDT1-derived FUCCI fragment; independent full-protein evidence supports nucleus, while the fragment limitation is retained. PMID:24217620(https://pubmed.ncbi.nlm.nih.gov/24217620/), full target-relevant Results/Fig.2; HPA CDT1.
Finally, Arabidopsis CDT1a/b TAP recovers polymerase-epsilon subunits POL2A and DPB2, and two-hybrid directly tests DPB2 association. Exact donor-locus joins support the inherited polymerase-binding annotation; no human direct assay or loss of the ancestral activity is inferred. PMID:22773747(https://pubmed.ncbi.nlm.nih.gov/22773747/), indexed full Table3/Fig.8 and TAP/Y2H Methods. Resolving the human interface and the checkpoint propagation questions remains useful; these bounded uncertainties do not prevent adjudication of the other functions.
The FAF1 interaction reason now names the ATPase-defective p97 substrate trap without an unverified residue assignment. The primary source itself is inconsistent: its Results and Figure 6j legend use E587Q, whereas its Protein interaction studies Methods use E578Q. Both point to the same construct reference. The original rationale followed the Results spelling; this audit does not silently convert a discrepancy in the paper into a confidently resolved sequence identity. PMID:26842564(https://pubmed.ncbi.nlm.nih.gov/26842564/), full Results, Figure 6 legend and Methods read. The cited construct paper describes an ATPase-inactive p97 EQ substrate trap and inducible tagged cell lines, but the retrieved full text does not give either residue number. The measured HA-FAF1/CDT1 association, human nuclear-lysate context and non-core decision remain unchanged.
All six nonblocking suggestions in the approved review were considered:
All 60 original source objects, qualifiers, partners and source references remain intact, as do all 60 actions and the three cores. This entry appends to the exact prior notes. No new annotation, product, literature quotation or source-cache edit is introduced.