Control of actin assembly in cells through regulation of Capping Protein

点击翻译按钮获取中文摘要

CARMILs (Capping protein Arp2/3 Myosin I Linker) are 1000 residue, multi-domain scaffold proteins expressed from protozoa to man that have been studied extensively with regard to their ability to bind Capping Protein (CP) and reduce its affinity for the actin filament barbed end. CARMIL proteins also appear to play important roles in signal transduction, as they exhibit genetic and physical interactions with the Rac GEF Trio (Liang et al MBoC 2009; Vanderzalm et al. Dev. 2009), and T cells lacking CARMIL-2 exhibit a profound block in signaling downstream of CD28, the major co-receptor for T cell signaling (Liang et al Nat. Immunol. 2013). In previous work (Jung et al JCB 2001), we showed that Dictyostelium CARMIL binds CP, the Arp2/3 complex and myosin I (through its SH3 domain), and it is required for actin-dependent processes such as chemotaxis and micropinocytosis to be robust. Here we describe initial studies of CARMIL-GAP, a second Dictyostelium CARMIL that contains, in addition to all the normal CARMIL domains (including the CP-binding CPI domain), a 130 residue insertion that, by homology, is a GTPase activating (GAP) domain for Rho-GTPases. This domain is probably functional given that full length CARMIL-GAP can only be over-expressed if its GAP domain contains a point mutation (R737A) that blocks GAP activity in all characterized GAP proteins. Moreover, mass spec analyses of GAP domain pull-downs indicate that CARMIL-GAP binds the Rac isoform Rac1A. Current efforts are directed at demonstrating in vitro that CARMIL-GAP exhibits GAP activity towards Rac1A. Like CARMIL, CARMIL-GAP localizes to actin-rich structures and is expressed in both vegetative and starved, developing cells. Consistently, CARMIL-GAP null cell lines created by homologous recombination exhibit pronounced defects in several actin-based processes occurring in vegetative and starved cells, including phagocytosis, motility and chemotactic aggregation. Importantly, these defects are specific, as expression of GFP-FL CARMIL-GAP in CARMIL-GAP null cells rescues them. Finally, rescue of CARMIL-GAP null cells with versions of CARMIL-GAP that lack either GAP activity or the ability to bind CP show that while both domains contribute significantly to CARMIL-GAP function, the GAP domain plays a bigger role.