Model of invadopodium maturation. (stage 1) The invadopodium precursor assembles in response to growth factor stimulation (core structure consists of actin, cortactin [Cttn], cofilin, N-WASp, Arp2/3, and Tks5). During this time, kinases (Src and Arg) are also recruited. Between stage 1 and 2, Tks5 anchors the core structure to PI(3,4)P₂ at the invadopodium core (surrounding membrane not shown; Sharma et al., 2013a). (stage 2) After growth factor stimulation, Arg binding to β1 integrin disrupts the Arg autoinhibited conformation and growth factor receptors induce Src-dependent Arg phosphorylation on tyrosine 439 (Y439), resulting in full Arg kinase activation and phosphorylation of Y421 on cortactin (Beaty et al., 2013). Nck1 localizes to invadopodia by binding to phosphorylated cortactin and N-WASp (Oser et al., 2010). Talin is recruited to invadopodia in an integrin- and moesin (msn)-independent manner by binding to actin via its C-terminal I/LWEQ binding site. (stage 3) Talin binds to β1 integrin, which enhances the interaction between the talin rod domain (R11-DD) and the moesin FERM domain, to recruit a complex of moesin–NHE-1 to invadopodia. NHE-1 increases the intracellular pH to disrupt the inhibitory pH-dependent interaction between cortactin and cofilin, allowing active cofilin to be released. Cofilin severs F-actin to generate free actin barbed ends and new filament elongation. (stage 4) The new cofilin-generated filaments support dendritic nucleation by the Nck1–N-WASP–Arp2/3 complex. This actin polymerization, in turn, induces invadopodial membrane protrusion. MT1-MMP is locally delivered to invadopodia and activated to degrade the ECM either directly or indirectly through activation of pro-MMP-2 and -9 zymogens (Sakurai-Yageta et al., 2008).