BAIAP2

IRSp53 is fundamentally an adaptor that bridges the plasma membrane and the Rho-family GTPases to the actin machinery. The central partial-CRIB and proline-rich region receives activated Cdc42 and Rac1; the C-terminal SH3 domain recruits the actin effectors that do the polymerizing and bundling -- WAVE2, N-WASP, Eps8 and Mena/VASP. IRSp53 itself neither nucleates nor elongates filaments; its contribution is to hold the GTPase signal, the membrane surface and the nucleation-promoting factor in the same place, which is what determines whether a protrusion forms and what shape it takes. The same adaptor logic operates in neurons, where binding to PSD-95/PSD-93 and the Shank family positions IRSp53 in the postsynaptic density and drives dendritic spine morphogenesis.

• PMID:11696321
  Cdc42 induces filopodia by promoting the formation of an IRSp53:Mena complex.
• PMID:15673667
  Here, we report that IRSp53 interacts with two specific

The N-terminal I-BAR/IMD domain reads negative membrane curvature, and this is what makes IRSp53 a shape sensor rather than a constitutive scaffold. Quiroga et al. compressed fibroblasts to generate bud-shaped plasma membrane evaginations on the ~100 nm scale and found IRSp53 selectively enriched at them, with maximal sorting near 0.05 nm^-1 and roughly twofold enrichment over control. Enrichment creates a local node that couples membrane topography to a transient Rac1/Arp2/3 actin burst, which re-flattens the evagination -- a mechanochemical feedback loop maintaining plasma membrane shape homeostasis.

• PMID:37747150
  We show that upon cell compression, bud-shaped PM deformations of negative curvature (evaginations) on the 100 nm scale are formed and enriched by IRSp53, a negative curvature-sensing protein.

Direct binding of the IMD/I-BAR domain to PI(4,5)P2 is the membrane-targeting step on which the adaptor and curvature-sensing activities both depend, and it is the feature that distinguishes the I-BAR subfamily mechanically. Mattila et al. mapped the PI(4,5)P2-binding site and showed that its geometry is inverted relative to classical BAR domains, so the domain binds the interior of the tubule and induces curvature of the opposite sign; this is the molecular reason I-BAR proteins generate plasma membrane protrusions where BAR proteins generate invaginations. The same study found the filopodia-forming activity of the IMD to depend on this membrane-deforming activity rather than on the previously proposed F-actin-bundling or GTPase-binding activities -- a point worth carrying, since it means the actin bundling terms this gene carries by IBA are a secondary property of the domain and not the mechanism by which it builds protrusions.

• PMID:17371834
  Here, we provide evidence that IMDs deform PI(4,5)P2-rich membranes into tubular structures.
• PMID:17371834
  that the membrane-deforming activity of IMDs, instead of the previously proposed

The IMD binds F-actin directly and cross-links filaments into tightly packed parallel bundles, an activity mapped to conserved basic residues at the two extreme ends of the zeppelin-shaped dimer. This is the molecular function underlying the actin filament bundle assembly and actin crosslink formation processes the gene is annotated to, and it had no molecular-function representation in GOA. It is listed last deliberately: Mattila et al. showed the filopodia-forming activity of the domain survives loss of F-actin bundling and tracks membrane deformation instead, so actin binding is a genuine property of the domain rather than the mechanism by which protrusions are built. In cells it operates alongside Eps8, with which IRSp53 synergizes to bundle actin.

• PMID:14752106
  The IMD alone, derived from either IRSp53 or MIM, induced filopodia in HeLa cells and the formation of tightly packed parallel F-actin bundles in vitro
• PMID:15635447
  Mutagenesis of conserved basic residues at the extreme ends of the dimer abrogated actin bundling in vitro and filopodia formation in vivo