For example, activation of the epidermal growth factor (EGF) receptor by EGF has been shown to trigger and/or enhance invadopodia formation [1] as have platelet-derived growth factor (PDGF) and reactive oxygen species (ROS) (as reviewed in [2]). Following stimulation of the EGF receptor, the tyrosine kinase Src is activated, followed by the Abelson-related nonreceptor tyrosine kinase, Arg [3]. Src has traditionally been known as a vital component in the signaling cascade that governs the initiation of invadopodia formation [4, 5, 6, 7] however more recent studies have also highlighted the importance of other Src family kinases [8] and protein kinase C [9] in this process.

Activation of these kinases are reported to increase the number of free actin barbed ends and lead to the phosphorylation of cortactin [10] and the scaffolding protein Tks5 (tyrosine kinase substrate with 5 SH3 domains) [7]. Together these events promote actin assembly polymerization and maturation of the invadopodium [10]. Tks5 is also associated with the production of invadopodia in response to ROS [11].

In addition, the GTPase Cdc42 has been implicated in invadopodia formation via the regulation of neuronal Wiskott-Aldrich Syndrome protein (N-WASP), which acts in concert with the Arp2/3 complex to promote actin polymerization [9]. Similarly, cortactin is able to promote actin polymerization through binding and activating the Arp2/3 complex [12]. Live-cell imaging experiments have shown the accumulation of cortactin in invadopodia precedes matrix metalloproteinases (MMPs) accumulation and matrix degradation [5].

Ultimately the outcome of initiating invadopodia formation has been speculated to result in the production of a podosome-like precursor. This precursor structure comprises a small branched actin network, which is suggested to continue to polymerize and extend deeper into the extracellular matrix (ECM) [13] as MMPs are released from the tip of the elongating structure.

 
Two distinct types of filamentous actin networks are suggested to cooperatively form invadopodia - a branched actin network that forms the base of the structure and a parallel array of bundled actin filaments that form along the length of the invadopodial shaft. The latter is responsible for its elongation [15] and works together with microtubules and intermediate filaments to extend the invadopodia [15].

The production of these two actin networks are driven by different molecular components. Formation of the branched array of actin filaments at the base of invadopodia is driven by Arp2/3 and its activator N-WASP [15, 16, 1]. The activation of N-WASP is facilitated by the Rho GTPase Cdc42, WIP (WASP interacting protein) and Nck1 (non-catalytic region of tyrosine kinase adaptor protein 1) - all of which have been found to localize to invadopodia and are necessary for their formation [1]. The resulting structure resembles the actin core of podosomes.

Elongation of invadopodia is suggested to be driven by the extension of parallel arrays of bundled actin filaments that are present along the length of invadopodia [15]. Extension of these filaments is driven by formins (Functional Module: Formin and Profilin in Actin Nucleation), whilst their bundling is coordinated by the actin cross-linker fascin. Studies have shown both formins [15, 17] and fascin [18] localize to invadopodia and are necessary for their formation and stability. Additional proteins required for maintaining the stability of these structures include members of the Ena/VASP family. The invasive isoform of the Ena/VASP protein MenaINV has been shown to stabilize invadopodia and is hypothesized to do so either via stimulating actin polymerization from free barbed ends or through the bundling of actin filaments [19].

Invadopodia extension is driven by actin polymerization and hence the concentration of G-actin-ATP monomers at the invadopodia tip contributes to the rate of extension. The actin severing protein cofilin has been shown to generate free barbed ends onto which G-actin monomers can be added [20]. The activity of cofilin is focused within the core of the invadopodium by the action of the RhoGTPase RhoC [21]. RhoC activity is regulated by its activator, the guanine nucleotide exchange factor (GEF) p190RhoGEF and its deactivator, the GTPase activating protein (GAP) p190RhoGAP. The phosphorylation of cofilin by RhoC decreases its severing ability, therefore in order to promote extension of the invadopodium, RhoC activity is limited to regions beyond the invadopodium core. This is achieved by restricting the localization of the RhoC activator p190RhoGEF to regions surrounding invadopodia, whilst the RhoC deactivator p190RhoGAP localizes within invadopodia [21]. The spatially regulated activity of RhoC, and subsequently cofilin, leads to barbed end production within invadopodia, which facilitates actin polymerization and filament extension.

The rate of extension is also dependent upon the ability of the growing filaments to overcome membrane resistance and the concomitant incorporation of new membrane. This can be achieved through the addition of membrane from vesicles delivered to sites of invadopodia formation. The Golgi apparatus is found in close proximity to invadopodia and is speculated to be positioned in such a manner to meet the demands of new membrane incorporation [22]. New membrane must be remodeled to allow for the growth of invadopodia into the extracellular matrix (ECM). F-BAR proteins have long been known to modulate membrane curvature and recently the F-BAR protein, CIP4 (Cdc42 interacting protein 4), has been implicated in invadopodia as both a scaffold and a means to promote membrane curvature [23]. The combination of actin polymerization and membrane incorporation ultimately results in the extension of invadopodia into the extracellular matrix (ECM).

ECM degradation itself is carried out by a variety of secreted matrix metalloproteinases (MMPs) and serine proteases. Currently over 25 different MMPs have been identified that together have the potential to degrade the entire ECM [26]. Of all the MMPs, MMP14 (also known as MT1-MMP), is considered to be the major regulator of invadopodia-mediated ECM degradation across several cell models [27]. MMP docking has been shown to be essential for ECM degradation in melanoma cells, with knock-down [5] or overexpression [26] showing a decrease or increase in invadopodial activity, respectively. Cortactin, commonly known as a weak activator of the actin nucleator Arp2/3, also has a role to play in MMP secretion and ECM degradation [27].

Cortactin accumulation has been shown to precede MMP accumulation at the tips of invadopodia and has been suggested to regulate their secretion [5]. This cortactin-dependent maturation process has been shown to be dependent on the activity of LIM kinases [28].