The database assertions were assessed against the target protein and each source's role. The 625- and 571-residue UniProt products remain distinct. The latter deletes residues 109–162; neither this deletion nor a platelet-associated transcript proves equivalent secretion, activation, receptor binding or catalytic activity. Signal-peptide removal (1–18) and activated heavy/light chains (19–387 and 388–625) are processing of the long precursor, not additional splice products. Historical mature-protein numbering is retained where reported; it is not globally shifted onto every construct.
PMID:22961984(https://pmc.ncbi.nlm.nih.gov/articles/PMC3488089/) provides the directly read human internal-cleavage evidence. Its Experimental Procedures, Results and relevant Discussion describe human FXI cDNA expressed in HEK293 cells, conditioned-medium purification, FXIIa activation and controls against activating-enzyme carryover. Activated FXI cleaves factor IX sequentially at two internal bonds; engineered apple-domain/catalytic-domain constructs dissect binding and catalysis. This is direct proteolytic work on a zymogen substrate, not an inference from the necessity of F11 activation. Figure captions and text were read; no new figure pixels or full supplement were inspected.
The PMID:10823824 author manuscript was accessible independently of the abstract-only normal cache. Selected Methods cover human FXI cDNA/pJVCMV expression in 293 cells, purification, washed human platelets, radiolabel binding and competition. The opening Results show saturable cofactor-dependent binding; the platelet-versus-phospholipid comparison links A3-domain variants to platelet-supported coagulation. This supports a peripheral extracellular-facing pool. It does not establish integral membrane topology or identify the short natural splice product. Exact private line locators are retained; this was not a complete new paper or pixel review.
PMID:89876(https://doi.org/10.1182/blood.V54.4.850.850) was recovered only as actual publisher-indexed original pages 851, 853, 855, 857 and 859. They contain human plasma purification, Hageman-fragment activation, plasminogen-dependent fibrin assays, antibody-adsorption and kallikrein-contamination controls, and discussion of discrepant earlier work. The comparison on page 857 used about 30 times plasma concentrations. Positive plasminogen activation and increased fibrin digestion are retained as contextual functions; physiological dominance is not inferred. The complete article/PDF, Table I, remaining pages and figure pixels were not recovered. Failed PDF responses were HTML, and an Elsevier response was bibliographic metadata only. Search-engine visual descriptions are not figure inspection.
All 11 source events were traced through the captured raw participant graph, including catalyst and regulatedBy branches: 160 retrieved objects, with no missing reachable IDs in the declared participant/reference/modification edge set. Literature and pathway-ancestry links are outside that graph scope. Four target entities resolve to P03951: extracellular or receptor-bound residues 19–625, and activated heavy/light chains 19–387/388–625. The modeled GPIb–IX–V association and the independent human platelet experiment support the more precise external side of plasma membrane. The source's broad compartment is preserved in every imported object.
| Event | Target role and source boundary |
|---|---|
| 158145 | Extracellular FXI becomes GPIb–IX–V-associated FXI. |
| 9936440 | Extracellular FXI binds kininogen; F11 is a partner, not a kininogen-cleaving enzyme in this event. |
| 9936451 | FXI–kininogen binds the receptor complex. |
| 158300, 9936430 | FXIIa activates receptor-bound FXI; F11 is the substrate of this step. |
| 158419 | Thrombin activates the same receptor-bound target; drug-regulation branches were included. |
| 158333 | Native FXIa is the catalyst that activates factor IX. |
| 9653838, 9655851, 9657706 | Native receptor-associated FXIa is retained in SERPING1-, SERPINE1- or APP-associated products. These models do not imply that every inhibited or internalized FXIa molecule occupies the membrane. The SERPINE1 narrative's endothelial uptake experiment is distinguished from its GPIb-associated source model. |
| 9673223 | The disease variant is factor IX substrate; the F11 catalyst is native. No mutant-F11 projection is asserted. |
Event narratives and relevant target objects were read; the graph was mechanically replayed. This does not mean all primary papers cited by those events were inspected.
The official GO_REF:0000033 method and actual IBD assertions at PTN002543071 and PTN002548665 were inspected. Their donor lists and legitimate target self-WITH are preserved. The immutable UniProt/InterPro intake records place P03951 in PTHR24252; actual PANTHER v19 geneinfo instead reports PTHR24253:SF44. A recovered PTHR24252 alignment has 299 entries but no exact ungapped match to the canonical 625-residue target. Required tree retrievals and mapped-family alignment retrieval timed out. Therefore no target ancestral path, exact target alignment, loss or family reassignment is claimed. These access/version limits do not invalidate the independent human activity and plasma evidence.
Actual InterPro target intervals and mappings were recovered for IPR000177, IPR001254 and IPR018114. The trypsin domain and histidine signature, together with EC 3.4.21.27, agree with the internal factor IX-cleavage chemistry. SL-0243 explicitly maps secretion to extracellular region. Current ARBA00087422 and ARBA00089948 each match the captured target FunFam/Primates clause. Current ARBA00027128 has no matching clause among its complete captured condition sets; the historical match remains unresolved while independent native-human extracellular evidence remains positive.
The aminopeptidase IEA was traced to rat F11 A0A0G2K4I9/ENSRNOP00000073061 and two RGD IDA papers. PMID:15331420(https://pmc.ncbi.nlm.nih.gov/articles/PMC1618603/) was recovered in full by genuine normal acquisition after earlier access failures. The main Methods, Results and Discussion show a female Lewis-rat arthritis experiment with plasma FXI functional measurements. The FXI method is cited to a separate 1988 microplate-assay paper; the adjacent S-2302 method is for prekallikrein and was not reassigned to FXI. Figure 4 Results report changes in activity. No figure pixels or original 1988 assay paper were read. PMID:10899350(https://pubmed.ncbi.nlm.nih.gov/10899350/) remained abstract/selected Animals-passage accessible only. The inspected evidence does not resolve release of single N-terminal amino acids. The proposed human correction is independently supported endopeptidase chemistry, not a declaration that the unread rat experiment or orthology is false.
The campaign instructions prohibit new publication quotations and permit supported non-core binding where no specific replacement is justified. No new publication quotations or annotations are introduced. Reference availability flags describe whether a complete original is available; exact actual reading scope and immutable normal-cache metadata are recorded separately. In particular, selected pages of PMID:89876 remain incomplete, whereas the complete PMID:15331420 main text was genuinely recovered. No source cache was rewritten.
The broad proteolysis assertion is refined to zymogen activation because FXIa directly performs the proteolytic conversion of FIX to FIXa. The current term lies beneath protein processing/proteolysis. An exact-term comparator query of human F2, F10 and F12 recovered four F12 annotations, including three IDA, and no exact-term F2/F10 hit; this uneven usage was not treated as universal support. The positive human target assay is decisive. The separate plasminogen-activation descendant remains contextual and should not be repeated as redundant coverage in a core synthesis.
Factor XI is a plasma zymogen whose activated product, FXIa, supplies proteolytic activity in coagulation. The native, recombinant and transcript experiments below answer different questions. The current UniProt products are 625 and 571 residues; the shorter sequence deletes precursor residues 109–162. Signal removal and activation into the 19–387 and 388–625 chains are processing events, not additional splice products. Historical papers often number the mature sequence rather than the precursor.
Hsu et al., PMID:9593722 reports human platelet and megakaryocytic-cell RNA, RT-PCR, Northern analysis, in-situ hybridization and a cDNA lacking exon V. The author manuscript's Experimental Procedures and operative Results were read. Its references to platelet proteolysis and membrane-associated activity describe earlier work. This study does not purify the shorter product and establish its catalytic equivalence to plasma FXI. The paper's cDNA is AF045649; current product reconstruction and a historical transcript result are distinct evidence.
Ho et al., PMID:10823824 supplies the more direct surface evidence. Its author manuscript describes native human FXI and recombinant human FXI expressed in 293 cells, washed human platelets, radioligand binding with competition and nonspecific-binding controls, and clotting assays in FXI-deficient plasma. Results/Table III distinguish platelet-supported activity from phospholipid-supported activity for A3-domain variants. This supports peripheral association with the extracellular face of the platelet membrane, not integral membrane topology. HK/prothrombin cofactor experiments do not establish a specific natural splice product as the membrane-bound form.
Wojcik et al., PMID:9169594 investigates human FIX substrates. The abstract and original publisher-indexed Results on p. 633 were read: FXIa activates normal FIX more readily than FIX Boxtel under the compared conditions. Full Methods and FXIa reagent provenance remain unavailable after finite PMC/publisher PDF attempts. The paper's phospholipid-binding measurements concern FIX and are not evidence that F11 itself was measured in that assay. The separate human FXI clotting experiments above provide independent coagulation support.
Jin et al., PMID:15545266 tests the human FXIa catalytic domain, paper residues 370–607, produced in Pichia pastoris, with E. coli-produced ecotin variants. The original Experimental Procedures and operative structural and kinetic Results were read. Ecotin M84R is a potent inhibitor, whereas M84D/M84P also show substrate behavior under the reported conditions. Purified complexes and substrate-like contacts support binding. The 2:2 assembly involves the ecotin dimer; it is not evidence for native FXI homodimerization. No inhibitor activity is assigned to F11, and generic binding is retained as a non-core facet after checking more specific, but role-inverting, ontology alternatives.
Meijers et al., PMID:2844223 explicitly reports human FXIa inhibition and complex formation with protein C inhibitor in its abstract. The publisher and institutional record verify the citation, but the full original Methods and gel results were not recovered. This is a positive report with an access-limited judgment, not evidence against binding or a claim of curator error.
Papagrigoriou et al., PMID:16699514 has an accessible main abstract and an official three-page Supplementary Methods PDF. The latter was read in full and describes native plasma FXI purification, crystallization and diffraction. Complete main Results and figure pixels were not inspected. Samuel et al., PMID:17884987 was read directly in PMC: operative Results, captions and Methods resolve the recombinant A4-domain dimer interface and examine whole FXI/FXIa preparations by solution scattering. A4 residues 272–361 are paper coordinates. This supports the specific homodimerization refinement of the two imported P03951-1 self-binding contexts; the proposed compact activated conformation remains a model. It does not establish the shorter product's assembly or function. Structural database endpoint failures were retained and are not presented as an independent assembly verification.
Mandle and Kaplan, PMID:89876 was inspected through actual publisher-indexed original pages 851, 853, 855, 857 and 859. These include purification of human plasma FXI, activation with Hageman fragments, plasminogen-dependent fibrin-plate assays, anti-FXI/anti-prekallikrein immunoadsorption, and enzymatic comparisons addressing kallikrein contamination. They provide positive evidence for the measured capacity. The authors also discuss disagreement with earlier studies. Overnight activation and concentrated preparations—including a Figure 7 comparison at 30-fold plasma concentration—limit inference about physiological dominance. Plasminogen activation and positive regulation of fibrinolysis are therefore retained as non-core contexts. Complete original PDF pages and figure pixels were not available; automatically returned image descriptions were not counted as pixel inspection.
Gonzales et al., PMID:19056867 describes pooled urine from eight healthy people, differential ultracentrifugation, gel fractionation and LC-MS/MS. The actual main Methods, operative Results and linked NHLBI exosome table were read. The exact F11 record names NP_000119 and one peptide. The authors' ambiguity filter permits shared peptides among splice products of one gene, so this is not product-resolved detection. The urinary preparation does not establish a general exosome function, luminal versus surface topology, or universal extracellular-vesicle residence. Its anatomical source qualifier remains distinct from a broader cellular location.
The nine original direct-evidence publication caches remain abstract-only and their metadata is unchanged. The authored reference flag describes original full-text availability: it is false for the complete accessible author-manuscript pages (9593722, 10823824 and 15545266) and complete PMC originals (17884987 and 19056867), while the exact portions actually read are listed above. It remains true for selected-page, supplement-only, partial-original or abstract-only access (89876, 16699514, 9169594 and 2844223). Availability does not imply that every passage or figure was inspected. No figure-pixel inspection is claimed in this review.
The core couples serine-type endopeptidase activity to two compatible process facets: zymogen activation and blood coagulation. The actor is activated FXIa and its substrate is FIX. These current process terms are not ancestors or descendants of one another. The separate plasminogen-activation descendant is retained only as its accepted non-core context, not added again to the core.
PMID:22961984(https://pmc.ncbi.nlm.nih.gov/articles/PMC3488089/), Experimental Procedures and Results Figures 1–5, directly measures human FIX conversion after activation of recombinant human FXI. FIX was resolved from its intermediate and active product on gels, rather than inferred only from a downstream chromogenic endpoint. A3-domain replacement, blocking antibody and isolated-domain experiments support substrate recognition outside the catalytic site. The paper's FIX Arg145/Arg180 numbering is retained as reported; the core states the reaction without imposing those mature-protein coordinates on precursor records. Its Discussion distinguishes a release/rebind model from possible partial repositioning and does not require the two FXIa subunits to execute separate cuts.
PMID:17884987(https://doi.org/10.1073/pnas.0703080104), A4-domain NMR Results, whole-protein SAXS/EM Results and main Materials and Methods, supports homodimer organization. The compact activated conformation and simultaneous-cleavage arrangement were modeled. No new figure pixels or supplementary structural restraints were inspected in this synthesis. The platelet association and assay limits are described above; soluble plasma activity and surface organization are both retained without an integral-membrane or universal lipid-binding claim.
The natural shorter product remains a specific experimental question. The proposed comparisons separate transcript presence, secreted protein, native processing, dimer formation and active-site-normalized FIX cleavage. A second question asks whether the specialized fibrinolytic capacity measured in PMID:89876 contributes appreciably under physiological concentration/time conditions; it does not presume that contextual capacity is absent or dominant.
GO:0007597 explicitly describes FXIa conversion of FIX to FIXa within the intrinsic cascade. The current typed path to blood coagulation is part_of through fibrin clot formation (GO:0072378), not an is_a relation. The three broad coagulation contexts and the single catalytic core now use the intrinsic-pathway term. The human assays support this refinement without excluding other routes of FXI activation.
In PMID:22961984(https://pmc.ncbi.nlm.nih.gov/articles/PMC3488089/), Experimental Procedures describe human recombinant FXI expression, activation and controls, and direct gel-resolved FIX cleavage; Results and Figure 2 analyze conversion to active FIXa. In PMID:10823824(https://doi.org/10.1074/jbc.M001890200), Coagulation Assays specify kaolin activation, FXI-deficient plasma and platelets or phospholipids, with the functional comparison in Table III. These are independent positive human sources; they do not repair the unread complete Methods of PMID:9169594 or validate catalytic equivalence of the shorter transcript described by PMID:9593722.
GO:0031639 is an is_a child of zymogen activation, GO:0031638. That hierarchy does not make the substrate-specific plasminogen reaction a description of FIX activation. The central FIX reaction therefore retains GO:0031638, while the earlier positive plasminogen assay stays non-core with its high-concentration and physiological-scope limits. This is an existing-source refinement, not a new assertion justified by missing graph coverage.
The seven extracellular-region annotations describe soluble or complex-associated extracellular pools. GO:0005615 is currently obsolete as equivalent in concept to GO:0005576. GO:0072562 specifies a blood microvesicle and cannot be inferred simply from plasma purification; it is not added. Platelet external-surface association remains supported independently. The ecotin NC and SERPINA5 U dispositions also remain: positive engineered binding and an incompletely accessible endogenous-inhibitor source have different verification limits, not reversed biological roles. No new publication quotations are introduced under the explicit campaign instruction.