CRB2 is a membrane-associated Crumbs protein whose cytoplasmic tail interacts with polarity proteins. Its main supported role is organizing and maintaining the apical epithelial junction region. Human fetal RPE depletion impairs junction ultrastructure and recovery after a calcium switch, even when baseline transepithelial resistance is unchanged. This establishes maintenance and reestablishment in that assay; it does not establish a universal requirement for initial junction formation. PMID:34262913
The terminal CRB2 peptide has direct, measured PDZ-domain binding. The fragmentomics experiment tests the biotin-tagged final ten residues, 1276–1285, against isolated PDZ constructs. Its 21 source-specific partner annotations map to 66 positive curator records, including the actual domains, tags and construct features. These are valid domain-binding observations; they do not establish that every partner contacts intact CRB2 in its native compartment. The separate EPB41L5 interaction concerns a FERM domain, rather than a PDZ domain. Detailed motif-deletion controls in that earlier paper predominantly concern CRB1 and are not reassigned to CRB2. PMID:36115835 PMID:17920587(https://doi.org/10.1016/j.yexcr.2007.08.025)
The fragmentomics author correction PMID:36477203(https://pmc.ncbi.nlm.nih.gov/articles/PMC9729566/) restores missing figure labels and a column, and corrects Figure 5 panel ordering. Its operative text was read independently of the metadata-only normal cache. It does not state that the CRB2 interaction records were withdrawn. Supplementary Data 1 was not recovered; the literal interaction records are retained, and a malformed affinity exponent is not converted into an invented numerical value.
Mouse RPE CRB2 is apicolateral, including the tight-junction region. Human RPE experiments independently corroborate peripheral junction localization; the mouse antibody and HEK construct controls in the earlier paper are not human retinal microscopy. The anatomical apical junction complex must also be distinguished from a molecular complex of fixed stoichiometry. The GO subapical-complex term explicitly describes an invertebrate epithelial compartment, so the mammalian location is represented by the broader anatomical apical junction unit. PMID:26404741
Mouse Crb2 deletion disrupts gastrulation, mesoderm production and associated axial and cardiac morphogenesis. Later live imaging directly connects Crb2 loss with cells that fail to complete apical detachment and ingression after initiating apical constriction. These are positive source-specific developmental roles, retained outside the distilled epithelial polarity core. The later study separates these events from a universal primary loss of all classical polarity markers and leaves the biochemical connection to Myosin IIB incompletely resolved. PMID:22072575 PMID:27870829
The independent Poglut1 study directly compares Crb2-null embryos and shows a wider primitive streak, retained E-cadherin-positive cells and deficient mesoderm. It also shows that Crb2 is a glycosylation substrate whose surface trafficking depends on Poglut1. That does not make CRB2 a glycosyltransferase, and the separate Notch defect of Poglut1-null embryos is not assigned to CRB2. PMID:26496195
The earlier Crb2 knockout study reports reduced Bmp4 expression during E7–E7.75, with expression retained in amnion/allantois and Brachyury/Fgf8 unchanged. Gross abnormalities become apparent later in that interval. This stage- and region-selective response supports the existing indirect positive BMP-regulation inference as non-core. A downstream BMP-output assay, direct molecular mechanism and the contribution of altered cell composition remain unresolved; the observation is not described as a pSMAD assay or universal CRB2-to-BMP4 pathway.
The retinal conditional-knockout paper genuinely assays Crb2, despite its title discussing human CRB1 disease. Its original Methods and Results support junction disorganization, photoreceptor orientation defects, progressive degeneration and shorter developing cone outer segments. These findings support retinal homeostasis, photoreceptor maintenance and cone-development annotations at mouse source scope. The human RPE maintenance assay does not negate separate developmental polarity evidence. PMID:23001562
Human renal disease variants and rescue with human CRB2 in zebrafish support a conserved renal epithelial role, with species and construct scope kept explicit. Recent podocyte experiments connect CRB2 depletion with contractility and increased YAP-associated readouts. Culture Methods and Results identify human immortalized cells, while the immunoblot Methods say murine; that internal source ambiguity is unresolved. Twenty-four-hour YAP or rapamycin inhibition did not reverse the force phenotype, so the data do not establish a sole CRB2–Hippo–YAP–contractility chain. No additional YAP-signaling activity is inferred. PMID:25557779(https://pubmed.ncbi.nlm.nih.gov/25557779/) PMID:40062402(https://pubmed.ncbi.nlm.nih.gov/40062402/)
Human CRB2 expression reduces APP cleavage products, whereas endogenous SH-SY5Y knockdown increases them. CHAPSO-solubilized membrane reactions with recombinant substrate provide cell-free corroboration; they are not purified binary CRB2–protease experiments. The association depends on the CRB2 transmembrane region, while disrupting the PDZ- and FERM-binding motifs does not remove the measured inhibition. PMID:20299451
The original also co-immunoprecipitates native CRB2 with the four gamma-secretase components from cynomolgus monkey brain and observes corresponding complexes in human cell membrane preparations. Reciprocal immunoprecipitation was limited by background. The current gamma-secretase-complex definition includes regulatory subunits, making a CRB2-associated regulatory pool a defensible source-specific refinement. This does not make CRB2 a universal catalytic-core subunit, establish fixed stoichiometry or identify its directly contacted partner. The inhibitory/complex-binding activities remain positive contextual functions outside the primary polarity core; physiological human-brain relevance is unresolved.
The three official sequences independently reconstruct to 1285, 1176 and 953 residues. The displayed product retains its membrane anchor and cytoplasmic binding motifs. Product 2 lacks the canonical anchor and tail and is predicted secreted; full-length ectodomain-release negatives do not assay that splice product. Product 3 retains the tail but has no established equivalent activity solely from sequence. No publication construct is silently assigned an official isoform merely because a reference-sequence checksum matches.
The actual PAINT ancestral node is on the human CRB2 path, with four source IBDs and no target-path loss assertion. The current resolving PTHR24049 tree differs from the preserved PTHR12916 cache/official-cross-reference boundary. Both are exposed in the evidence index; the older cache is not presented as the current resolving topology. Human CRB2 among descendant evidence is legitimate, rather than circular. The adhesion donor experiments use fish proteins, including recombinant extracellular-domain fusion coexpression/lysate pulldown and extracellular-domain-dependent HEK aggregation, and are not relabeled as direct human assays.
CRB2/NP_775960 is listed with one peptide in the original-linked NHLBI urinary-exosome table, under the reference key for PMID:19056867. This is retained as a non-core fraction detection, not evidence that CRB2 performs vesicle biogenesis or cargo sorting. The second EPS-urine screen remains undecided because its target-bearing supplementary Table 2 was not recovered. Its apparent workbook responses were challenge/error content; absence from an unavailable table is not claimed. PMID:19056867 PMID:23533145(https://pubmed.ncbi.nlm.nih.gov/23533145/)
The source and evidence index supplies portable original locators, cache/access distinctions and exact source-record joins. It is authored provenance, not an independent biological experiment.
The developmental retinal deletion in PMID:23001562 affects the orientation and segment formation of newly born photoreceptors. The two existing establishment assertions therefore refine to establishment of epithelial apical/basal polarity (GO:0045198). This is a developmental formation inference; it is not evidence that CRB2 initiates polarity in every epithelium. The human RPE maintenance/recovery experiments remain a separate source scope. The broader polarity core integrates both contexts. PMID:23001562(https://pubmed.ncbi.nlm.nih.gov/23001562/) PMID:34262913(https://pubmed.ncbi.nlm.nih.gov/34262913/)
The predicted secreted splice product is now represented by a curated functional class mapped to Q5IJ48-2 and extracellular region. Official sequence changes remove the canonical membrane anchor and cytoplasmic motifs. This is a prediction with a defined product scope, not a newly measured secretion result; the original GOA row has no isoform identifier and retains its exact source metadata.
The PDZ-binding core is supported by complementary evidence: the exact CRB2 terminal-peptide/PDZ-domain assay records in PMID:36115835 establish domain specificity, while the target-specific CRB2/MPP5 association in PMID:38570189 provides retinal complex context. The latter uses human CRB2 with porcine retinal lysate and does not alone isolate the contacted domain. The previous screen-scale core excerpt has been replaced with a CRB2-specific association excerpt; the direct assay constructs remain explicit in the evidence index. PMID:36115835(https://pubmed.ncbi.nlm.nih.gov/36115835/) PMID:38570189(https://pubmed.ncbi.nlm.nih.gov/38570189/)
The evidence index is an authored provenance sidecar, not an independent experimental source or a schema-validated gene review. Its literal records preserve all 76 GOA rows behind the 74 seeded source identities, including the distinct UniProt and CAFA assignments.