RASA1/p120RasGAP: GAP-Independent Scaffolding Mechanism

Research question: What is the evidence for RASA1 functions independent of its catalytic Ras-GAP activity — especially p190RhoGAP recruitment, directed cell movement, and vascular development — and how should GO curation express a non-catalytic molecular function?

Iteration 1. Literature-based synthesis. Citations are PubMed-cached abstracts (PMIDs verified); DOIs are provided where confidently known and flagged as uncertain/uncached otherwise.

1. Summary (answer)

RASA1 has a genuine, mechanistically defined GAP-independent scaffolding function: through its tandem SH2 domains it directly, phosphotyrosine-dependently recruits p190RhoGAP (and assembles a FAK–RASA1–p190RhoGAP complex) to control RhoA-dependent cytoskeletal reorientation and directed cell migration, a role explicitly shown to be separable from — and not requiring — Ras-GAP catalysis. Recruitment is direct and high-affinity (Kd ≈ 0.3 µM), and partner binding leaves catalytic activity unchanged, so the two functions are cleanly separable at the molecular level. However, the recruitment/scaffold role is not sufficient to explain the CM-AVM vascular phenotype: developmental blood/lymphatic vascular disease from RASA1 loss is driven by the catalytic (Ras-MAPK) arm via collagen IV export, and the EPHB4–RASA1 physical interaction is dispensable for angiogenesis. For GO curation, the non-catalytic role should be expressed as a specific, evidence-backed binding/scaffold molecular function, scoped to migration/polarity — not recorded merely as a residual mechanism gap.

2. Catalytic Ras-GAP evidence (kept separate)

• Vascular development / CM-AVM is Ras-dependent. In endothelial cells (EC), RASA1 loss dysregulates Ras-MAPK, impairing collagen IV folding/export from the ER → EC apoptosis and defective angiogenesis. Chen et al. 2019, PMID 31185000 (JCI Insight; DOI 10.1172/jci.insight.125940 — likely, uncertain).
• EPHB4–RASA1 act in the same catalytic pathway; physical interaction dispensable. EPHB4-mutant mice expressing EPHB4 that cannot engage RASA1 (but retains kinase activity) had normal angiogenesis; RASA1/EPHB4 protect against CM-AVM "independent of physical interaction." Chen et al. 2022, PMID 35015735 (JCI; DOI uncertain/uncached).
• Review of EPHB4–RASA1 negative regulation of Ras-MAPK. Chen et al. 2023, PMID 37259315 (ATVB; DOI uncertain).
• Second-hit somatic RASA1 inactivation in CM-AVM EC confirms loss-of-function (catalytic) genetics. Lapinski et al. 2018, PMID 29024832.

These establish the Ras-dependent (catalytic) disease mechanism and are deliberately partitioned from the scaffold evidence below.

3. GAP-independent scaffold evidence

3.1 Direct, phosphotyrosine-dependent p190RhoGAP recruitment

• N-SH2 of RASA1 binds p190RhoGAP pTyr-1105 via canonical FLVR arginine R207; Kd = 0.3 ± 0.1 µM (ITC). Jaber Chehayeb et al. 2019, PMID 31891593 (J Struct Biol).
• The SH2-SH3-SH2 module simultaneously engages two proximal p190RhoGAP phosphotyrosines with synergistic (avidity) binding and induced compaction — a selectivity mechanism. Stiegler, Vish & Boggon 2022, PMID 36417908 (J Biol Chem).
• RASA1 makes distinct SH2 interactions with doubly-phosphorylated partners p190RhoGAP, Dok1, EphB4; these interactions do not impact catalytic activity. Vish, Stiegler & Boggon 2023, PMID 37507023 (J Biol Chem). → clean biochemical separation of scaffolding from catalysis.
• FLVR-unique C-SH2 / methods: PMIDs 32540970, 37668970.

3.2 Directed cell movement requires the scaffold, not catalysis

• Cornerstone separation-of-function: In RasGAP-deficient cells, growth factor rescues spreading/Golgi/transport Ras-dependently, but stress-fiber/focal-adhesion turnover producing the elongate migratory morphology depends on the constitutive RasGAP–p190 association, independent of Ras regulation; disrupting the pTyr-mediated complex with p190 peptides suppresses migration. Kulkarni et al. 2000, PMID 10769036 (J Cell Biol; DOI 10.1083/jcb.149.2.457 — likely, uncertain).
• FAK–RASA1–p190RhoGAP polarity complex: FAK (pY397) drives SH2-mediated RASA1 recruitment and p190A phosphorylation, targeting the complex to leading-edge focal adhesions to transiently inhibit RhoA; loss abolishes polarity in fibroblasts, endothelial and carcinoma cells. Tomar et al. 2009, PMID 19435801 (Nat Cell Biol; DOI uncertain).
• Corroborating pathway/cross-talk context: β3-integrin/EGFR→Src→p190/p120 complex (21937717); Brk phosphorylation of p190 Y1105 promoting p190/p120 complex, migration, invasion (18829532); E-cadherin→membrane p190RhoGAP/p120RasGAP complex lowering RhoA (20228844).

3.3 Other non-catalytic RASA1 activities (context)

• Caspase-3 cleavage generates an N-terminal "fragment N" that drives an Akt-dependent anti-apoptotic response (independent of GAP catalysis). Peltzer et al. 2013, PMID 23826368.

4. Supported vs. refuted hypotheses

5. GO-curation guidance

• Express a specific molecular function, distinct from GAP catalysis (GO:0005096 Ras GTPase activator):
• phosphotyrosine residue binding (GO:0001784) / SH2 adaptor-scaffold activity, evidence PMIDs 31891593, 36417908, 37507023 (IPI/IDA + structural).
• Biological process: regulation of cell migration / establishment of cell polarity via RasGAP–p190RhoGAP, evidence 10769036, 19435801.
• Retain as non-core binding/scaffold evidence, and scope BP annotations to migration/polarity — do not annotate the scaffold to CM-AVM/angiogenesis pathogenesis, which the genetic separation-of-function assigns to the catalytic Ras-MAPK arm (35015735, 31185000).
• Residual mechanism gap remains for whether p190 recruitment is sufficient in vivo for vascular/migration phenotypes (existing in-vivo data point to catalysis for vasculature); flag this as an open annotation caveat.

6. Coordination / separation of Ras-dependent vs Ras-independent roles

Evidence supports context-partitioned roles rather than a single integrated switch: (i) catalytic Ras suppression governs EC survival and collagen IV homeostasis in developmental vasculature (CM-AVM); (ii) non-catalytic SH2 scaffolding of p190RhoGAP governs RhoA-dependent cytoskeletal reorientation for directed motility/polarity. Upstream tyrosine kinases (FAK/Src, EphB4, Brk, EGFR) generate the phosphotyrosines that toggle RASA1 between reading partners and localizing signaling — a spatial-temporal, not purely catalytic, layer of control (37507023).

7. Limitations & future directions

• Synthesis is abstract-level; DOIs are flagged as uncertain/uncached and several journal/DOI attributions were inferred, not verified.
• No direct in-vivo "SH2-scaffold-dead but GAP-intact" RASA1 knock-in exists in the retrieved literature to prove sufficiency of p190 recruitment for vascular/migration phenotypes — the key missing separation-of-function reagent.
• Most migration evidence is in fibroblast/carcinoma/EC culture; in-vivo relevance of the scaffold to human vascular disease remains unresolved (residual mechanism gap).
• Future: knock-in RASA1 SH2-binding mutants (R207/R231) vs GAP-dead (R789-type) alleles in EC-specific models to formally dissociate scaffold vs catalytic contributions to angiogenesis and migration.