Fibrillin-2 has three retained core molecular activities: extracellular matrix structure, calcium binding in its EGF-like domain architecture, and hormone activity of the released C-terminal product placensin. Matrix-chain and hormone functions are distinct. The two original splice records remain unchanged; the shorter 1,473-aa product lacks the mapped placensin interval.
Human FBN2 microfibril and coassembly evidence supports a broad structural role, including the scaffold for early elastic-fiber assembly and nonelastic matrix organization. UniProt annotates early assembly with ortholog evidence, while PMID8120105 provides target sequence and preferential elastic-matrix localization. The calcium-binding core now cites the target primary-sequence analysis in PMID8120105 alongside the mapping reference. The exact UniProt calcium-binding EGF-like domain features and their evidence are located in the direct source guide; no calcium signaling or measured affinity at every repeat is asserted. In mouse osteoblasts, Fbn2-dependent matrix sequestration of latent TGF-β complexes limits signaling. This provides a scaffold mechanism for the retained regulatory process without assigning receptor catalysis or making fibrillin-2 an obligatory mineralization substrate. Bone and eye outcomes remain contextual non-core biology.
The generic basement-membrane location is supported by actual human renal GBM and Descemet/anterior lens-capsule evidence. The separate Bruch membrane study places FBN2 particularly in its central elastic layer; that measurement is not relabeled as RPE or choriocapillaris basal-lamina incorporation. Original source and tissue limits remain visible in each complete reference assessment.
The two applied classes use reviewed UniProt chain mappings: matrix fibrillin-2 at 78–2779 and placensin at 2780–2912. These curated coordinates are not newly measured endogenous termini. The placensin paper's Methods use 2780–2912, whereas its Figure 2 caption says 2779–2912; both statements are preserved. Human trophoblast secretion and serum IP-MS support the released product, while recombinant human ligand was tested in mouse metabolic systems and human trophoblast cells. Neither an endogenous human glucose intervention nor hormone output from the truncated splice product is inferred.
Placensin positively regulates gluconeogenesis in the tested mouse hepatocyte systems; those cells execute the glucose-conversion reactions. Figure 3 and the hepatocyte Methods in PMID:32329225 report glucose output in glucose-free lactate/pyruvate medium with gluconeogenic transcript induction. This supports GO:0045722 without claiming tracer-resolved flux or a human metabolic intervention. FBN1/asprosin's raw glucose-metabolism annotation is recognized, and its reviewed record likewise uses a regulatory-role correction. A shared cAMP/PKA response does not establish a placensin GPCR class. UniProt explicitly assigns linked secretion and subsequent furin separation with ECO:0000269/PMID32329225. The processing model and positive product-release evidence are retained. Its original site prediction, colocalization and internal peptides after proteolytic digestion leave the precise intact endogenous terminus and quantitative furin dependence unresolved. The proposed experiment addresses this bounded biological gap with intact-product mapping and controlled furin perturbation. No processing activity is assigned to the FBN2 substrate.
The unresolved PMID28327460 HDA target identification retains UNDECIDED because the exact supplementary FBN2 accession join was not recovered. Independent microfibril evidence supports its generic structural classification, but cannot reconstruct that particular screen hit. In contrast, actual main-table/workbook records support the other HDA matrix detections. The inverted fibrillin/fibulin names in the source table are not treated as proof of curator error. The PAINT ECM inference and verified HDA detections are retained as positive non-core matrix claims. Their evidence scopes remain distinct: the phylogenetic statement concerns matrix fibrillin-2, whereas the HDA extracts do not resolve intact molecular assembly or processing state. The unresolved exact HDA source remains UNDECIDED rather than being verified by inference from another study.
Broad extracellular location remains informative for the gene and is ACCEPTed consistently across all eleven source contexts, including the seven mature-matrix Reactome routes. The independently verified product scope stays in each source assessment and the functional-product model; no imported-row isoform identifier is invented. TOO_GENERAL judges the usefulness of the term, not the strength or breadth of one evidence row. Broad structural molecular function is retained because it covers nonelastic matrix organization as well as the contextual elasticity role.
All 32 complete reference objects retain their identities and inherited content; manual assessments include the direct and database feedback. Six shared-paper assessments contain both independently read scopes. Publication caches remain unchanged; independently recovered originals and their limits are recorded in the assessments. No new annotation or supporting quotation was introduced.