Automated deep research was attempted with just deep-research-falcon human FERMT2 --fallback perplexity-lite, but the run timed out before producing a deep-research file. This review therefore uses cached GOA publications, the UniProt record, Reactome records, and the PANTHER family fetch.
FERMT2 encodes kindlin-2, a FERM/PH domain-containing cytoplasmic adaptor and scaffold. UniProt summarizes its core role as a phosphoinositide-binding focal-adhesion protein that cooperates with talin to activate integrins, links cell-ECM adhesion sites to the actin cytoskeleton, and supports focal adhesion assembly, cell adhesion, spreading, and shape modulation. The local PANTHER fetch places FERMT2 in PTHR16160, a FERMITIN 2-related family.
The strongest core evidence is the podocyte/integrin paper. Although only the abstract is cached, it directly states that "Kindlin-2 is a FERM and PH domain-containing integrin-binding protein" and that kindlin-2 interacts with phosphoinositides, preferentially PIP3 [PMID:21325030 "Kindlin-2 is a FERM and PH domain-containing integrin-binding protein"; PMID:21325030 "kindlin-2 interacts with multiple phosphoinositides, preferentially with phosphatidylinositol 3,4,5-trisphosphate"]. It also supports the biological outputs: "depletion of kindlin-2 reduced integrin activation, matrix adhesion and fibronectin matrix deposition, whereas overexpression of kindlin-2 promoted these processes" PMID:21325030.
The structural PIP3 paper supports phosphoinositide binding and membrane engagement. The abstract states that the kindlin PH domain recognizes PIP2 and PIP3, that PIP3 binds kindlin-2 PH more strongly than PIP2, and that phosphoinositide-binding-defective kindlin-2 is impaired for talin-cooperative integrin activation [PMID:22030399 "the kindlin FERM domain contains an inserted pleckstrin homology (PH) domain that recognizes membrane phosphoinositides"; PMID:22030399 "PIP3 site-specifically binds to kindlin-2 PH with substantial chemical shift changes that are much larger than PIP2"; PMID:22030399 "this ability is significantly impaired for a phosphoinositide binding-defective kindlin-2 mutant"].
The ILK-kindlin-2 papers support an integrin outside-in/focal-adhesion axis rather than generic protein binding. PMID:25160619 reports that "human kindlin-2 binds to human ILK with high affinity" and that kindlin-2 binding to ILK is crucial for focal-adhesion localization and cell spreading [PMID:25160619 "human kindlin-2 binds to human ILK with high affinity"; PMID:25160619 "kindlin-2 binding to ILK is crucial for the kindlin-2 localization to focal adhesions and cell spreading"]. PMID:30254023 independently reports that "the interaction between ILK and kindlin-2 is critical for cell spreading and focal adhesion localization" PMID:30254023.
Several papers support important but context-specific branches. The Wnt paper states that kindlin-2 forms a tripartite complex with beta-catenin and TCF4 and enhances Wnt target gene expression [PMID:22699938 "Kindlin 2, as a focal adhesion protein, controls integrin activation"; PMID:22699938 "Kindlin 2 forms a tripartite complex with β-catenin and TCF4"]. The TGF-beta/Smad fibrosis paper states that kindlin-2 physically mediates interaction of TGF-beta type I receptor with Smad3 PMID:23723426. The renal EMT paper supports Ras/ERK/AKT-dependent EMT in tubular cells PMID:24226523.
The vascular-barrier and mesenchymal stem-cell papers are also relevant but non-core. Endothelial work shows kindlin-2 at adherens junctions, linking cadherin-catenin complexes to cortical actin and limiting vascular leak [PMID:28799653 "Kindlin-2 co-localized and co-immunoprecipitated with vascular endothelial cadherin-based complexes"; PMID:28799653 "maintaining the vascular barrier by linking Vascuar endothelial cadherin-based complexes to cortical actin"]. The MSC paper reports that kindlin-2 controls YAP1/TAZ, RhoA, MLC phosphorylation, stress-fiber formation, focal-adhesion assembly, and adipogenic/osteogenic fate decisions [PMID:29496737 "Loss of kindlin-2 inhibits RhoA activation and reduces myosin light-chain phosphorylation, stress fiber formation, and focal adhesion assembly"; PMID:29496737 "Depletion of kindlin-2 in MSCs is sufficient to induce adipogenesis and inhibit osteogenesis in vitro and in vivo"].
For curation, accept integrin binding, PIP3/phosphoinositide membrane engagement, focal adhesion localization/assembly, cell-matrix adhesion, integrin activation/signaling, membrane/cortical/focal-adhesion localization, actin binding, and ILK/actin-linked spreading/morphology terms as core or directly supporting core. Keep Wnt, TGF-beta/Smad, EMT, ERK/AKT, adherens-junction/barrier, MSC differentiation, cell migration, and tissue-development terms as non-core because they are context-specific outputs. Modify generic protein binding annotations to informative terms where the cached reference makes the partner clear: beta-catenin/DNA-binding transcription factor binding for the Wnt paper, and protein kinase binding for the ILK paper.
The second-pass audit confirmed the existing FERMT2 review and manual reference metadata. No annotation action changes were needed: FERMT2 remains curated as a kindlin-2 FERM/PH-domain adaptor linking phosphoinositide-rich membranes, integrin tails, ILK/focal adhesions, and the actin cytoskeleton, with Wnt, TGF-beta/Smad, EMT, barrier, MSC-differentiation, and tissue-development outputs retained as context-specific non-core biology.