Regulation of invadosomes by microtubules: Not only a matter of railways

Abstract

Invadosomes, which encompass podosomes and invadopodia, are actin rich adhesive and protrusive structures facilitating invasion and migration in various cell types. Podosomes are mostly found in normal cells, while invadopodia are hallmarks of invasive transformed cells. Despite evident structural differences, both structures mostly rely on the same pathways for their formation and their activity. While the role of actin cytoskeleton is undeniable, the involvement of microtubules (MTs) in invadosome formation/activity has recently been demonstrated but also somehow underestimated. MTs are components of the eukaryotic cytoskeleton well known for their essential roles for cell division, the maintenance of cell shape, intracellular transport and cell motility. Until now, MTs were mostly seen as railways for the delivery of various cargos required for invadosome functions but recent data suggest a more complex role. In this review, we address the specific functions of MTs on invadosome dynamics, activity, maturation and organization in light with recent data, which extended far beyond simple track delivery. Indeed, MT dynamic instability, which in turn modulates Rho GTPase signalling and likely MT post-translational modifications are playing major roles in invadosome functions.

Introduction

Podosomes and invadopodia, collectively called invadosomes, are adhesive actin-enriched membrane structures protruding at the ventral side of cells when grown in 2D, which can have extracellular matrix (ECM) degradation properties. The term podosome is used to define the structure found in normal cells (e.g. monocytic cells, megakaryocyte, endothelial cells, smooth muscle cells) while in invasive cancer cells and in Src-transformed fibroblasts organised into rosettes it is called invadopodium (Fig. 1). Invadosomes are essential for processes involving cells to cross tissue barriers such as cell transmigration for immune cells in physiological conditions but also for pathological processes such as dissemination of cancer cells during metastasis. Despite sharing several similarities, invadosomes present major differences. Indeed, podosomes have a diameter of around 1 μm and a height of 0.5 μm whereas invadopodia can reach 8 μm in diameter and 5 μm in depth suggesting a more aggressive matrix degradation in this latter. The dynamics of both structures as well as their abundance in cells are also significantly different. While invadopodia can last hours, podosome lifetime is usually in the range of minutes (Linder et al., 2011; Paterson and Courtneidge, 2018). Furthermore, podosomes and invadopodia rely on different signalling proteins and inputs (Hoshino et al., 2013). For example, the scaffold protein Tks5 is characteristic and essential for signalling pathway leading to invadopodia formation (Eddy et al., 2017). Finally, a ring containing adhesive molecules, such as integrins or vinculin, surrounds the F-actin core of podosomes (Fig. 1A). The presence of such a ring structure in invadopodia is apparently conserved (Branch et al., 2012; Pignatelli et al., 2012) (Fig. 1B).

Among the similarities, podosomes and invadopodia are both F-actin rich dot-like structures involving actin polymerisation, mostly mediated by the Arp2/3 complex and formins, and adhesive molecules (Linder et al., 2011; van den Dries et al., 2019) (Fig. 1). The role of actin cytoskeleton in invadosome activity has been addressed in various excellent reviews to which readers can refer (Kedziora et al., 2016; Linder et al., 2011; van den Dries et al., 2019). Briefly, the core of podosomes is composed by branched actin filaments contacting adhesion molecules such as CD44 in osteoclasts (Chabadel et al., 2007). This branched network is surrounded by unbranched actin filaments bundled by myosin II linked to adhesion molecules such as integrins through ring proteins containing, among others, talin and vinculin (Chabadel et al., 2007; Linder et al., 2011). Finally, these actin networks are covered by cap proteins comprising for example the formin FMNL-1 (Mersich et al., 2010; van den Dries et al., 2019) (Fig. 1A). Branched and unbranched actin filaments are also found in invadopodia. Indeed, the base contains a branched network potentially linked to ring proteins and adhesion molecules (Branch et al., 2012; Pignatelli et al., 2012). A similar interaction is also possibly present at the side of invadopodia (Beaty et al., 2014; Proszynski and Sanes, 2013) (Fig. 1B). Finally, bundled actin filaments are found at the tip of the invadopodium (Schoumacher et al., 2010) (Fig. 1B). Besides the crucial role played by actin cytoskeleton, invadosomes are also influenced by microtubules (MTs) (Linder et al., 2011). For a long time according to most of the publications in the field, the function of MT on invadosome activity was restricted to tracks for the delivery of cargos. However, this implies the targeting or capture of MTs to invadosomes, a process which starts to be well documented. Furthermore, recent data demonstrated that MTs also have signalling functions regulating the formation of podosomes and potentially invadopodia.

In this review, we would like to focus on the specific functions of MTs on invadosome dynamics, activity and organisation in light with recent data.