Factor V is a secreted procofactor. Its processed Va chains assist factor Xa in prothrombin activation on phospholipid surfaces; Xa performs the proteolysis. Cleavage of FV by thrombin, Xa or activated protein C makes FV the substrate of those particular reactions. Neither a catalyst complex containing FV nor a strong coagulation phenotype makes FV a protease.
The original human reconstitution in PMID:9252393(https://doi.org/10.1074/jbc.272.33.20721), printed pages 20722–20723, uses normal plasma FV and homozygous-patient Leiden FV with human coagulation factors, PC/PS membranes, thrombin-generation readout and prothrombin immunoblots. The inspected Methods/Results support a cofactor step in fibrin clot formation and a membrane-associated context. They do not establish an F5 catalytic active site. The complete PDF is available, but this database assessment read those operative pages without inspecting figure pixels.
PMID:9556620(https://www.researchgate.net/publication/13722970_Protein_C_Inhibitor_Secreted_from_Activated_Platelets_Efficiently_Inhibits_Activated_Protein_C_on_Phosphatidylethanolamine_of_Platelet_Membrane_and_Microvesicles), JBC 273:11281–11287, contains a useful F5 control within a PCI study. Its immunoelectron-microscopy Methods use anti-factor-V antibody as a control. Results report 267 of 330 F5 gold particles inside alpha granules; activated-platelet membrane/canalicular F5 is reported as data not shown. These positive locations are retained. The measured inhibitory complex is APC–PCI, not an experimentally demonstrated F5–PCI heterodimer in the inspected passages. Only selected original text was recovered here; no figure pixels were read. The direct-evidence assessment separately addresses the unresolved imported F5–SERPINA5 interaction.
PMID:6490642(https://www.researchgate.net/publication/16704949_Coagulation_Factor_V_contains_copper_ion), JBC 259:12949–12951, directly measures copper in three bovine and one human FV preparation by atomic absorption and emission spectroscopy after extensive dialysis, with buffer and prothrombin controls. Approximately one copper ion per FV supports copper binding even though the generic InterPro cupredoxin description calls FV/FVIII non-copper-binding. The paper does not establish a copper-dependent assembly or cofactor mechanism. Its negative oxidase assay is bovine FV/Va with a ceruloplasmin control. Complete author-transcribed main text was available; purification, measurement, controls and the functional-limit discussion were read, without pixels.
The actual PANTHER v19 gene mapping places human P12259 at PTN002554984 in PTHR24543. The ungapped 2224-residue MSA sequence is exactly the displayed human sequence. Extracellular-region IBD PTN009099846 and coagulation, circulation and platelet-alpha-granule IBDs at PTN002815661 all lie on that target path, with no target-path NOT. Current UniProt identifies PTHR46806; that current family and the captured v19 family are kept as distinct versions. The earlier missing target in the other v19 family is not evidence of functional loss. A broad extracellular ancestor can include retinoschisin O15537 without transferring retinoschisin activity to F5. Donor count and a target in its own WITH list are not objections to an IBA.
The mouse donor is F5/O88783, including ENSMUSP00000083204. In PMID:12816857(https://doi.org/10.1182/blood-2003-04-1224), fetal-liver-cell transplantation, gel-filtered platelet/plasma clotting and FV immunoblots establish the mouse platelet pool. The actual Methods distinguish a one-stage assay from a two-stage assay using mouse thrombin and human FV-deficient plasma; the authors correct substantial one-stage platelet-activity overestimation. These mouse synthesis results do not establish the same biosynthetic route for human platelet FV. The complete author transcription was accessible; Methods, pertinent Results, captions and interpretation limits were read without pixels. Mouse knockout/low-plasma-FV rescue sources 8900278, 10669158 and 12855561 were available only as original abstracts. Their positive developmental/bleeding and rescue observations are retained with that limit.
The verified zebrafish donor is f5/B7ZVT7, ZFIN:ZDB-GENE-030131-5531. PMID:31167819(https://pmc.ncbi.nlm.nih.gov/articles/PMC6560344/) supplied selected original Methods/Results for CRISPR deletion, vascular injury and rescue. The rescue cDNA is zebrafish f5, including constructs bearing human-associated variant positions; it is not a human-protein experiment. Complete main text was not recovered in the saved bounded captures. PMID:38874909(https://pmc.ncbi.nlm.nih.gov/articles/PMC11230853/) is available as complete XML-derived main text. The read Methods/Results specify an ENU f5 stop mutation, larval injury, adult bleeding and plasma kPTT/kPT fibrin-formation measurements. The larval FV comparison includes pooled heterozygote-incross progeny, whereas adult plasma measurements compare heterozygotes with wild types. No figure pixels or supplements were inspected. The exploratory accession Q502M6 resolved to an unrelated protein and is not used as an f5 donor.
The rat donor A0A0G2K3W2/ENSRNOP00000072800 carries response-to-vitamin-K IEP from PMID:658784(https://doi.org/10.1159/000214261). The actual publisher abstract reports modest FV reductions after warfarin or biliary fistula and restoration after vitamin K1. Full Methods/Results were unavailable. This is a positive response observation, not evidence that F5 requires vitamin-K-dependent gamma-carboxylation. Its conservation in human F5 remains unresolved rather than contradicted.
The target routes were followed through participant sets and complexes to human P12259 entities with their own compartments. Public event pages and entity numbers are listed in the corresponding reference assessments.
Captured target routes are complete; other cargo branches in the large trafficking assemblies were not exhaustively acquired. Reused and newly captured API objects retain their respective acquisition provenance rather than being represented as one synchronous database release.
The actual current ARBA00027486, ARBA00084788 and ARBA00092682 condition sets match the target's recorded FunFam memberships and taxonomy. FunFam names mentioning FVIII were not used to reject verified P12259 membership. The current ARBA00027128 conditions did not reproduce a match; the combined historical GOA row also has an independently valid Secreted/SL-0243 mapping and mouse/rat sources.
Current GO definitions support external-side plasma-membrane association, including peripheral proteins; alpha-granule lumen; fibrin clot formation; and the broad response-to-vitamin-K meaning. The latter is not a biosynthetic vitamin-K-dependence assertion. Non-core transport locations are retained as positive context, not removed because FV does not perform transport.
The displayed source is 2224 residues. Signal 1–28, mature chain 29–2224, heavy chain 29–737, released region 738–1573 and light chain 1574–2224 are processing features. The UniProt record has no alternative-products comment, and no splice-product objects are invented. Reactome intermediate chains, clinical variants, paper-specific mature numbering and separate external transcript models remain distinct.
Reference full_text_unavailable describes whether complete original main text was recovered, independently of the immutable cache's full_text_available field. A false flag does not claim that every paragraph, figure pixel or supplement was read. Each reference assessment names the actual read scope and public locator.
This checkpoint reviews nine publication contexts. It preserves the imported source assertions and the displayed P12259 record; processing into factor Va heavy/light chains is not a set of newly invented natural isoforms. The other 44 contexts belong to the database review. No publication quotations or new annotations were added at that original checkpoint. The feedback revision below separately evaluates two NEW annotations; the campaign instructions prohibit new publication quotations, not evidence-backed annotation proposals.
The complete 1984 article, PMID:6323392, pages 187–195, was read as extracted text, including Methods, Results, captions, Discussion and references. Human activated protein C (APC), protein C inhibitor (PCI/SERPINA5), and other proteases are the identified reagents. The experiments establish APC–PCI complexes and inhibition of several proteases. F5 appears in background as a cofactor/APC substrate; an executed F5–PCI binding assay was not identified. No figure pixels were inspected.
The 1998 platelet article, PMID:9556620, was available as selected author-upload text. Preparation of Proteins, Immunoelectron Microscopic Analysis, the inhibition/complex Methods, and corresponding Results were read. The human platelet F5 immunogold control is positive: the authors count 267 of 330 gold particles within alpha granules. They also report F5 on activated platelet membranes and canalicular structures, explicitly as data not shown. The adjacent complex experiment is APC–PCI. The displayed PCI localization images were not inspected and are not described as F5 images. These observations support F5 localization, not an F5–PCI heterodimer. Complete main-article access was not established.
For PMID:2844223(https://doi.org/10.1021/bi00412a005), only verified bibliographic/abstract material was recovered. The abstract discusses APC, kallikrein and XIa with PCI; it does not establish whether F5 was also tested in the unavailable full article. That omission is not used as a negative result.
The imported three P05154 interaction contexts remain UNDECIDED. A SUSPECTED/CHECK source query on the P12259 UniProt inhibition/heterodimer citation mapping is attached to the 6323392 context. It now explicitly includes UniProt-designated PMID:15140131, asks for clarification, proposes no replacement accession and asserts neither F5 nonbinding nor a global wrong-gene error. The limited MISCITED reference judgment concerns the particular F5–PCI claim, not the validity of the APC–PCI experiments. The current serpin-family binding term was considered, but no refinement was made without a verified F5–PCI assay.
The original BioPlex 2.0 paper, PMID:28514442, was available in complete BioC main text. Selected Online Experimental Methods cover human ORFeome clones, sequence validation, C-terminal HA-FLAG tagging, cell culture and anti-HA AP-MS. The author-labeled 2.0 293T directed-edge table has OLR1/GeneID 4973 as bait and F5/GeneID 2153 as prey. The BioPlex download site supplies the data provenance preserved in the exact table receipts.
The complete main of BioPlex 3.0, PMID:33961781, was also available. Selected clone, cell-culture, AP-MS, peptide/protein-identification and interaction Methods were read. The 293T network contains P78380/P12259 with pInt 0.99998471197731; the separate directed table identifies OLR1 bait and F5 prey. Network A/B ordering alone is not interpreted as direction. Absence from inspected HCT116 files is a file-level observation, not negative binding evidence or cell-line exclusivity. The two publications are related releases, not independent purified reconstitutions.
OLR1/P78380 is a scavenger receptor, but the AP-MS association alone does not establish F5 as its functional ligand. GO:0005124 has the signaling receptor binding parent. The issue is this role assignment, not a blanket requirement that every binding study measure downstream signaling. Both imported generic associations are retained as non-core under the campaign supported-binding exception. Neither a purified binary interface nor physiological plasma binding, exact historical insert sequence, F5 precursor/active-chain state or receptor response is claimed. The generic archive named BioPlex_interactionList_v2.tsv is not equated with the explicitly labeled 2.0 dataset; an additional positive v4a archive is retained as provenance only and not assigned to a paper from its filename.
Selected original Data Sources/Methods and Table II of PMID:14718574(https://doi.org/10.1074/mcp.M300127-MCP200) were read. Its explicit P12259 row is present in the literature, LCMS1 and LCMS2 columns and absent from 2DEMS. This supports circulating human F5 in extracellular region. The complete main and table pixels were not recovered; no unique peptide, current-product exclusivity or processing-state claim follows from this compilation.
The complete original PDF of PMID:9252393 was recovered. Materials and Methods on printed page 20722 and Results/Figure 1 on 20723 were read; page 20723/Figure 1 pixels were inspected. The preparations include F5 from normal human plasma and homozygous Leiden patients. Reconstituted human coagulation reactions on PC/PS vesicles measure thrombin generation and APC control. F5 supplies cofactor work to Xa-mediated prothrombin activation; Xa is the protease. This is positive participation, not merely F5 being cleaved or necessary for an outcome. The assay does not itself demonstrate an in-vivo clot endpoint.
GO:0072378, a current descendant of blood coagulation, covers the protein-activation cascade leading to fibrin clot formation and its regulation. It is therefore the more specific replacement for the broad coagulation context. It does not assert that F5 polymerizes fibrin. Normal and Leiden preparations remain separate. The database lane independently uses the same supported refinement; its phylogenetic/donor evidence is separate from this human assay.
For PMID:19946888(https://doi.org/10.1002/jms.1696), the verified abstract describes membrane fractions from human NK-like YTS cells. The original F5 identification and peptide table remain unavailable after bounded publisher and repository attempts. Independent platelet localization cannot verify this particular screen. The membrane HDA assertion remains UNDECIDED, without an integral-membrane, contamination or negative-localization claim.
The official PRIDE PXD000608 project explicitly identifies PMID:24769233(https://doi.org/10.1016/j.jprot.2014.04.028), human CSF and the two pooled samples. The author-deposited MaxQuant archive was actually downloaded. Its proteinGroups table has P12259 as the majority accession of the F5 group, with 52/44 peptides in exo_A/exo_G and 9/15 in sn_B/sn_H. The group also contains CON__Q28107 and B4DU26. These are author protein-group counts, not independently recalculated counts of peptides unique to a current human product. A separate exact extraction finds 40 peptide-table rows listing P12259; that projection is not asserted to reproduce the proteinGroups aggregate counts. The design labels, nine fractions per experiment and search parameters were inspected; no normalized enrichment calculation was performed.
The archive's group identity and positive vesicle-enriched fractions support retaining the specialized extracellular-vesicle location as non-core. The complete main was not recovered and PRIDE's sample/data-processing protocol fields say Not available. Vesicle lumen versus surface, neuron-specific origin, exclusive vesicle localization, physiological F5 processing and an F5-specific vesicle function remain unestablished. The raw GOA CSF anatomy extension is preserved at intake and is not invented as a seeded field.
Reference full_text_unavailable describes complete-original availability, not whether the unchanged normal cache contains a full main. It is false for 28514442, 33961781, 6323392 and 9252393, whose originals are accessible, but only the specified portions of the first, second and fourth were read. It is true for 2844223, 9556620, 14718574, 19946888 and 24769233. Positive partial-original or deposited-data verification is retained independently of that flag.
The private ledger pins exact original captures, source-table rows, selected text ranges, and the one viewed PDF page. It also retains failed access attempts. Existing normal bytes were not replaced by the newly recovered originals. No new normal publication dependency was required in this direct lane; its nine source PMIDs already existed in the accepted intake.
The principal procoagulant activity is represented by GO:0061133 endopeptidase activator activity. Its current definition requires binding to and increasing an endopeptidase's activity, matching Va enhancement of Xa. The accepted human reconstitution study (Materials and Methods, pp. 20722–20723; Results/Figure 1 and Discussion pp. 20727–20728) distinguishes the cofactor from the protease and its APC-mediated inactivation. The experiment measures thrombin generation and protein cleavage; fibrin-clot participation follows the established cofactor step, not an assertion that the experiment directly measured an in vivo clot. Reactome 140664 and 140686 supply the explicit human Xa/Va assembly and external membrane context. Plasma and granule residence remain valid accepted locations; the active prothrombinase unit uses the external membrane face.
A separately read original, PMID:30046712, Factor V-short and protein S as synergistic TFPIalpha cofactors, provides direct evidence for an anticoagulant contribution. Materials and Methods describes human plasma-derived FV and coagulation reagents, BHK-expressed recombinant full-length FV/FV-short, and the author-numbered R709Q/R1545Q FV-short QQ mutant. Purified-component S2765 assays and Figures 1–3 compare TFPIalpha with FV-short and protein S; Figures 4–9 and the corresponding Results use deficient/normal plasma, antibody blocking and thrombin-cleavage controls. FV-short enhances TFPIalpha inhibition of Xa in synergy with protein S; intact FV is less efficient and the synergistic FV-short activity is lost after thrombin cleavage. Full XML and its complete normal cache are available. Methods, Results and captions were read, but figure pixels and the cited earlier construct-generation articles were not. The paper's introductory schematic chain-domain wording is not used to define the Va domain architecture.
The inhibitor unit uses contributes_to GO:0004867 serine-type endopeptidase inhibitor activity: TFPIalpha supplies the direct protease-inhibitory contact, while F5 contributes to the inhibitory assembly. This is supported by functional component-omission, titration and cleavage experiments, not only association or loss-of-function necessity. The synthetic phospholipid assays do not independently establish a native endothelial or platelet location, so none is assigned to this unit. GO:0030195 negative regulation of blood coagulation and the procoagulant GO:0072378 fibrin clot formation are non-ancestral process facets of distinct activities. No new annotation rows were proposed at the original core-synthesis checkpoint; the focused feedback review below now proposes two supported rows.
The study's recombinant splice-derived FV-short, engineered QQ variant, full-length FV and thrombin-cleaved preparations remain distinct. No UniProt isoform identifier or functional equivalence across current transcript models was inferred. The accepted Reactome 9931624 and 9931628 instead model a proteolytic FV intermediate: their actual P12259 chains are 29–737 and 1047–2224. Their discussion of FV-short does not turn that modeled intermediate into a splice-product record. The second core is grounded in the newly read experiment; the earlier database assessment established modeled assembly/localization only.
Hayashi et al., PMID:15140131 is the fourth paper cited in UniProt's PCI regulation statement and the one expressly designated for heterodimer formation. Recovered publisher-indexed Materials and methods identify human APC and human PCI transgenic-mouse plasma, APC–PCI immunoassays and residual APC activity. Results on exogenous human APC and endogenous mouse APC/Protac, with Figures 3–5 captions, support those positive assays. Full main/PDF access remains unavailable; no figure pixels or supplement were read. These passages do not settle an F5–PCI interaction, and their omission of a target assay is not a negative experiment. The three imported binding rows remain unresolved, with a reduced-confidence citation-mapping query.
The two NEW rows describe performed work already represented in the core: Va enhancement of Xa-dependent prothrombin activation, and FV-short-dependent enhancement of TFPIalpha/protein-S anticoagulation. The former uses the human reconstitution and the explicit human Xa/Va assembly; the latter uses the purified-component assays, plasma experiments, blocking antibodies and cleavage controls in PMID:30046712. FVIII, tissue factor and thrombomodulin were checked as comparable protease cofactors. They lack the exact activator term in the captured complete human records, so no consensus or systematic oversight is claimed. Thrombomodulin carries negative regulation of coagulation, while protein S/protein Z have different or broader annotations. Target-specific performed cofactor work, rather than those annotation absences, justifies the proposals. Typed is_a/part_of closure shows no redundancy with the retained coagulation/clot-formation terms.
Four committed production GO-CAM models—665912ed00003542, 665912ed00003594, 666b894f00000010 and 666b894f00000232—contain F5 as an input of APC proteolysis in a protein-C pathway. They do not assign that proteolysis to F5. That substrate role is kept separate from the experimentally performed TFPIalpha cofactor role; it is not used to justify the new process assertion.
Native processed-F5 Reactome entities have their own plasma-membrane compartment and a defined extracellular reaction context. The reported native human platelet-surface F5 observation in PMID:9556620 is explicitly data not shown. The PC/PS reconstitution in PMID:9252393 corroborates functional membrane assembly, not intact-cell imaging. Together these support the peripheral external face for the native/processed prothrombinase pool. The separate FV-short/QQ experiments in PMID:30046712 use engineered reagents and synthetic lipids and do not identify a native platelet or endothelial pool for that anticoagulant unit. The NK-cell HDA membrane assertion remains unresolved because its F5 target table is unavailable; the valid platelet location does not verify that distinct source context.
The preserved UniProt P12259 record also lists Cu-cation sites at residues 1843 and 1845. Both are annotated by homology (ECO:0000250); the direct whole-protein human copper measurement remains PMID:6490642. Neither evidence supplies a copper-dependent oxidase claim or a measured requirement for copper in cofactor function.
The TFPIalpha core retains a contributes_to inhibitor relationship: TFPIalpha supplies the protease-inhibitory contact and F5 supplies measured cofactor work in the assembly. A named GO complex identifier is not required by the schema, and none is invented. The experiments establish a contribution despite incomplete structural detail; a missing term identifier is not recast as an unknown biological function. No additional autonomous F5 inhibitor annotation is proposed merely to mirror the core.