The human reference supplies mechanistic evidence for the corresponding selected horse protein; the human conclusion alone is not validation of the horse sequence. The exact horse comparison is in genes/HORSE/MAP2K2/MAP2K2-bioinformatics/RESULTS.md. Research reports are source leads; annotation decisions cite the underlying publication or experimentally supported UniProt passages. Unresolved source-specific results retain UNDECIDED.
Identity. P36507 is the reviewed, canonical human MEK2 (400 aa, sequence version 1, MANE ENST00000262948). It is not a fragment, isoform or alternative gene model: it carries the full protein kinase domain (72-369), active-site Asp194, ATP-binding loop 78-86 and Lys101, RAF sites Ser222/Ser226, and the C-terminal class I PDZ-binding motif RTAV that MEK1 lacks.
Evolutionary placement. UniProt cross-references PTHR47448:SF3 (MITOGEN-ACTIVATED PROTEIN KINASE KINASE 2); the PANTHER 19.0 per-organism file instead places P36507 in PTHR48013:SF8 (also named MAP2K2), and MAP2K1 in PTHR48013:SF5. Both sources keep MEK2 in its own subfamily separate from MEK1, so the disagreement is a family-numbering/release conflict, not a biological one (recorded in interpro/panther/PTHR47448/PTHR47448-review.yaml and panther-members-overrides.tsv). The IBA rows come from PAINT node PTN000684494 (MAP2K clade, MAPKK activity + MAPK cascade), with P36507 itself among the descendant evidence PMID:8388392.
Changes vs previous review.
- Catalytic rows: generic kinase, Ser/Thr, Ser, Tyr, Ser/Thr/Tyr kinase and Ser/Thr kinase activator rows all MODIFY to GO:0004708 MAP kinase kinase activity (the activator-activity row describes ERK activation by phosphorylation, which is the catalytic MF).
- MAPK cascade IBA -> MODIFY to ERK1 and ERK2 cascade; positive regulation of Ser/Thr kinase activity -> MODIFY to positive regulation of MAP kinase activity.
- Extracellular region (NAS, PMID:8388392) REMOVE: name-derived from "extracellular signal-regulated kinase"; MEK2 is cytosolic.
- Cytoplasm -> MODIFY to cytosol; membrane kept non-core (KSR fractionation PMID:10409742).
- MAP kinase scaffold activity (IMP PMID:29433126, full text read) kept non-core: MEK binding to KSR1 drives KSR1-BRAF dimerization PMID:29433126; the authors frame it as KSR being a MEK-dependent allosteric BRAF activator, so the scaffold term is approximate (suggested question).
- Gene expression / DNA-templated transcription IMPs (PMID:24375836, MEK1/2 inhibitor AZD6244, miR-92a via ERK-AP1) MARK_AS_OVER_ANNOTATED: downstream outputs, not MEK2 work. Regulation of stress-activated MAPK cascade MARK_AS_OVER_ANNOTATED (MEK2 is ERK-specific). Peroxisomal membrane HDA (PEX14 co-purification) MARK_AS_OVER_ANNOTATED.
- Review-derived (PMID:19565474, abstract only) localizations and processes (mitochondrion, early/late endosome, Golgi, focal adhesion, Golgi inheritance, endosome transport) stay UNDECIDED: the best-known endosomal partner LAMTOR3/MP1 was reported as MEK1-selective and MEK2-specific data cannot be checked. DLG1-paper ER/microtubule/junction/Golgi IDAs UNDECIDED (abstract attributes ER/microtubules to hDlg); plasma-membrane inner face and perinuclear IDAs kept non-core (stated for both proteins).
- Protein binding: Q96II5 (an ARAF entry) row changed from REMOVE to MODIFY -> MAPKKK binding, consistent with other ARAF rows; boilerplate UniProt FUNCTION quotes removed from protein-binding rows.
- New description and two core functions (MAPKK activity in the ERK1/2 cascade, cytosol; scaffold protein binding to KSR1/DLG1).
Relative. Closest experimentally characterised relative by placement: mouse Map2k2 (Q63932), same MEK2 subfamily (PTHR47448:SF3 in the PANTHER entries list; source of UniProt's "By similarity" ERK1/2 statement). No mouse Map2k2 review exists in the repo. Paralog MAP2K1 (Q02750, different subfamily) has a complete review at genes/human/MAP2K1/. Module: modules/erk_cascade.yaml (mek_kinase annoton lists P36507).
Rows: GO:0005739 IEA (GO_REF:0000117, ARBA:ARBA00026962) and TAS (PMID:19565474), both located_in, both
UNDECIDED before. No sub-compartment rows.
Checks:
- ARBA00026962 (UniProt REST, 2026-11-03) pools condition sets from unrelated families; MEK2 matches the
IPR000719 + IPR050915 (MAP kinase kinase) set in Euarchontoglires. Unvetted, not used as support.
- QuickGO: rodent orthologs carry only the same ARBA and TAS rows; no experimental MEK row to project from.
- TAS source PMID:19565474 (Yao and Seger 2009 review) is abstract-only and never says "mitochondri".
- Primary studies fetched: PMID:18545666 (Galli 2008, full text) PMID:18545666, in mouse cells PMID:18545666; fraction purity by complex I/IV, LDH and PI only (no ER marker).
PMID:15030409 (Alonso 2004, rat brain, abstract-only) PMID:15030409.
- Both use MEK1/2 antibodies, so MEK2 cannot be separated from its paralog.
Outcome: both rows UNDECIDED -> KEEP_AS_NON_CORE (minor, transient signalling pool with the ERK module;
TRAF3/BRAF precedent). Stays at root: no transit peptide or anchor, sub-mitochondrial data split between OMM
and IMS. Added PMID:18545666 and PMID:15030409 to references.