CD44 and β3 Integrin Organize Two Functionally Distinct Actin-based Domains in Osteoclasts

Abstract

The actin cytoskeleton of mature osteoclasts (OCs) adhering to nonmineralized substrates is organized in a belt of podosomes reminiscent of the sealing zone (SZ) found in bone resorbing OCs. In this study, we demonstrate that the belt is composed of two functionally different actin-based domains: podosome cores linked with CD44, which are involved in cell adhesion, and a diffuse cloud associated with beta3 integrin, which is involved in cell adhesion and contraction. Wiskott Aldrich Syndrome Protein (WASp) Interacting Protein (WIP)-/- OCs were devoid of podosomes, but they still exhibited actin clouds. Indeed, WIP-/- OCs show diminished expression of WASp, which is required for podosome formation. CD44 is a novel marker of OC podosome cores and the first nonintegrin receptor detected in these structures. The importance of CD44 is revealed by showing that its clustering restores podosome cores and WASp expression in WIP-/- OCs. However, although CD44 signals are sufficient to form a SZ, the presence of WIP is indispensable for the formation of a fully functional SZ.

Rho GTPases in osteoclasts: Orchestrators of podosome arrangement

“…For example, the ultrastructural organization of the sealing zone, which is encountered only on bone-like substrate, suggests that the compaction of podosome cores could be the result of a RhoA-dependent increase in contractility mediated by the actomyosin system that may connect podosome cores together. Spreading of osteoclasts following myosin inhibition by blebbistatin and high levels of activated RhoA in osteoclast maintained on bone-like substrate supports that hypothesis (Chabadel et al, 2007). This implies that RhoA has to be correctly localized and able to interact with its downstream effector ROCK.…”

“…In macrophages from WAS patients in which WASp is truncated and not functional, podosome formation and macrophage chemotaxis are dramatically impaired (Zicha et al, 1998;Linder et al, 1999). Critical function of WASp is supported by analysis of dendritic cells or osteoclasts coming from WASp (Burns et al, 2001;Calle et al, 2004) and WIP (WASp interacting protein) null mice (Chou et al, 2006;Chabadel et al, 2007). As WIP binds and protects WASp from calpain-dependent degradation (Chou et al, 2006;Chabadel et al, 2007).…”

“…Critical function of WASp is supported by analysis of dendritic cells or osteoclasts coming from WASp (Burns et al, 2001;Calle et al, 2004) and WIP (WASp interacting protein) null mice (Chou et al, 2006;Chabadel et al, 2007). As WIP binds and protects WASp from calpain-dependent degradation (Chou et al, 2006;Chabadel et al, 2007). Interestingly, deletion of WIP only affects podosome core formation while the F-actin cloud is still present in osteoclasts (Chabadel et al, 2007).…”

“…As WIP binds and protects WASp from calpain-dependent degradation (Chou et al, 2006;Chabadel et al, 2007). Interestingly, deletion of WIP only affects podosome core formation while the F-actin cloud is still present in osteoclasts (Chabadel et al, 2007). It thus defines two independent domains inside the podosome, the podosome core which coincides with the localization of actin polymerization regulating proteins (WASp, WIP, Arp2/3, gelsolin) and the F-actin cloud to which scaffolding proteins are recruited (paxillin, vinculin, talin).…”

“…Interestingly, OPN but not vitronectin is sufficient to restore normal podosome organization in WIP-/-osteoclasts suggesting that WASp function can be regulated by both the  v  3 integrins and CD44 (Chellaih, 2005;Chellaih et al, 2007;Chabadel et al, 2007).…”