PTK2 (human, Q05397) review notes
Automated deep research was unavailable for this gene (no deep-research provider
keys in this environment). These notes were compiled manually from the UniProt
record, the cached publications in publications/ (abstract-only for many; see
full_text_available: in each file), the Reactome cache, QuickGO and
gocams/index.tsv. No *-deep-research-*.md file was created.
Identity and architecture
- Focal adhesion kinase (FAK1), 1052 aa. FERM - kinase - proline-rich regions - FAT
(focal adhesion targeting) domain. Isoform 6 (FRNK) is the C-terminal non-catalytic
fragment that inhibits FAK signalling (UniProt).
- Catalytic activity: EC 2.7.10.2, L-tyrosyl-[protein] + ATP (UniProt; RHEA:10596).
Core molecular function: tyrosine kinase at integrin adhesions
Adhesion targeting
Processes
- Canonical FAK is holozoan, not animal-specific: PMID:20479219;
PMID:20479219.
- Choanoflagellates lost it (with integrins and the IPP complex), retaining only a FAK-related kinase domain:
PMID:20479219;
PMID:20479219.
- Apusozoan Amastigomonas has most of the adhesome but not FAK or c-Src
PMID:20479219.
The authors' preferred scenario is a holozoan origin of the canonical machinery, with the
alternative that FAK and c-Src were lost in apusozoans.
- Capsaspora integrin adhesion is functional, but FAK was not examined (abstract only):
PMID:32857975;
PMID:32857975.
- Interpretation:
- Ancestral, inferred from domain conservation only: the FAK domain set (FERM, kinase, FAT) and so,
plausibly, kinase signalling at integrin adhesions. There is no functional FAK data from any
non-animal.
- Animal-specific on current evidence: all developmental roles (placenta, heart, vasculature,
neurons), growth-factor/GPCR/netrin/Eph receptor crosstalk, immune contexts, nuclear p53 regulation,
and specialised locations such as ciliary basal bodies.
- This matches the TLN1/VCL reviews, where the integrin-talin-vinculin linkage is treated as ancestral
(sponge biochemistry; Capsaspora filopodia), and the CSK review on Src-family regulation.
GO-CAM
gocams/index.tsv has PTK2 in two models, both with GO:0004715 non-membrane spanning protein
tyrosine kinase activity:
- 682fbcd000002551 (FAISL): part of GO:0051894 positive regulation of focal adhesion assembly.
- 689e7a5d00003515 (Ephrin-A1/EPHA2 negative regulation of cell adhesion mediated by integrin):
part of GO:0033634 positive regulation of cell-cell adhesion mediated by integrin, in focal adhesion.
The model gives FAK a positive role that EphA2/SHP2 suppresses. That is why the GOA row
"negative regulation of cell adhesion mediated by integrin" on FAK was marked over-annotated.
The cell-cell restriction does not fit the matrix-adhesion assays, so the positive row was
modified to GO:0033630.
Annotation decisions (summary)
- 75 generic protein-binding rows: 66 removed under the repository policy. 9 modified where the
binding mode is defined: SH2 domain binding (SRC pY397, GRB2 pY925, Src SH3-SH2 peptide), SH3
domain binding (Src SH3 NMR), p53 binding (x2), integrin binding (ITGB4), PH domain binding
(Etk/BMX), talin binding.
- Protein tyrosine phosphatase activity (IMP, PMID:9360983) was modified to tyrosine kinase activity.
The paper shows FAK phosphorylating Cas, and FAK has no phosphatase domain.
- UNDECIDED: actin binding (PMID:16803572, abstract silent on FAK); JUN kinase binding (PMID:10925297,
abstract describes Jak2, possible JAK/JNK mix-up); two IL-34/macrophage IGI rows (PMID:26754294,
abstract silent on FAK); molecular function activator activity (PMID:12467573, structure paper).
- NEW: peptidyl-tyrosine autophosphorylation (GO:0038083). FAK itself catalyses the Tyr397 step
(Reactome R-HSA-354073).