Dynein Clusters into Lipid Microdomains on Phagosomes to Drive Rapid Transport toward Lysosomes

摘要

class="head no_bottom_margin" id="sec1title">IntroductionMicrotubule motors of the kinesin and dynein families drive many cellular processes such as organelle transport, chromosome segregation, and beating of cilia/flagella. This diversity of function requires the cellular localization and activity of motors to be regulated in many ways (). Regulation of motors at the single-molecule level by motor-associated regulatory proteins has been studied extensively (, ). However, most cellular functions require large forces that can only be generated collectively by a team of many motors (). Little is known about how such motor teams are assembled at appropriate cellular locations before they can execute a specific task. The substrate on which motor teams must assemble inside cells is usually a lipid membrane, for example, the bilayer membrane covering vesicular cargoes that are transported by motors. We therefore wondered if motor recruitment to a lipid membrane can be controlled by the membrane itself, perhaps in coordination with other membrane-bound proteins that regulate vesicle trafficking (e.g., Rab GTPases).In this respect, the heterogeneity of biological membranes is of particular interest. Cholesterol and sphingolipids appear enriched within lipid microdomains (also known as lipid rafts), where they enhance membrane packing to promote microdomain formation (, , ). This process is likely facilitated by a combination of protein-lipid and protein-protein interactions, because microdomains are enriched in specific proteins (e.g., glycosylphosphatidylinositol [GPI]-anchored proteins) and may be maintained by active processes that drive the membrane away from thermodynamic equilibrium (). Motors could be localized to microdomains by direct binding to lipids () or via adaptor proteins (). Membranous regions of high motor density could potentially be created by clustering many copies of a motor within a microdomain. Such geometrical clustering may be of advantage if multiple motors are to work cooperatively as a team (, ). Geometrical arguments suggest that motor clustering is necessary for efficient transport of micron-sized cargoes (). Indeed, cooperative improvement in transport of artificial liposomes through clustering-induced dimerization of kinesin-3 motors has been reported (). A minus-end-directed kinesin is also shown to localize into membrane domains near the apical subplasma membrane of polarized epithelial cells ().However, the functional relevance of clustering of motors and its impact on specific cellular processes is unknown. In this context, the appearance of microdomains on phagosomes with maturation is particularly interesting (href="#bib8" rid="bib8" class=" bibr popnode">Dermine et al., 2001, href="#bib9" rid="bib9" class=" bibr popnode">Dermine et al., 2005, href="#bib15" rid="bib15" class=" bibr popnode">Goyette et al., 2012). Phagocytosis and subsequent encapsulation of microbes into a membranous vesicle result in the formation of a phagosome. Phagosome maturation is intimately connected to microtubule (MT) motor-driven motion. Early phagosomes (EPs) move in a bidirectional (back-and-forth) manner on MTs, when they physically interact with and exchange lipids and proteins with endosomes (href="#bib5" rid="bib5" class=" bibr popnode">Blocker et al., 1997, href="#bib44" rid="bib44" class=" bibr popnode">Vieira et al., 2002). Intriguingly, this motion changes as the phagosome matures, so that late phagosomes (LPs) exhibit rapid unidirectional dynein-driven transport toward the MT minus end. The mechanism of this change is important to understand because it facilitates fusion of phagosomes with perinuclear lysosomes and is essential for pathogen clearance. MT depolymerization blocks delivery of fluid phase markers from endosomes to phagosomes and also reduces phagosome-lysosome fusion (href="#bib4" rid="bib4" class=" bibr popnode">Blocker et al., 1996, href="#bib11" rid="bib11" class=" bibr popnode">Desjardins et al., 1994, href="#bib17" rid="bib17" class=" bibr popnode">Harrison et al., 2003). Importantly, pathogens such as Mycobacterium tuberculosis (href="#bib39" rid="bib39" class=" bibr popnode">Sun et al., 2007) and Salmonella (href="#bib18" rid="bib18" class=" bibr popnode">Harrison et al., 2004) specifically inhibit this switch to dynein-dependent transport as a survival strategy.We therefore wondered if microdomains on the phagosome membrane could upregulate dynein-driven transport of phagosomes. Cholesterol appears to be a major player in microdomain formation on cellular membranes (href="#bib28" rid="bib28" class=" bibr popnode">Mayor and Rao, 2004, href="#bib33" rid="bib33" class=" bibr popnode">Rao and Mayor, 2014, href="#bib35" rid="bib35" class=" bibr popnode">Simons and Ikonen, 1997). Dynein-driven transport of endosomes increases in cholesterol storage disorders like Niemann-Pick disease, where cholesterol-laden “paralyzed” endosomes cluster around the MT minus ends (href="#bib24" rid="bib24" class=" bibr popnode">Lebrand et al., 2002). Cholesterol accumulation into endolysosomes results in cholesterol-poor phagosomes that are unable to fuse with lysosomes (href="#bib20" rid="bib20" class=" bibr popnode">Huynh et al., 2008). Interestingly, the GTPase Rab7 that recruits dynein to phagosomes interacts with the cholesterol sensor ORP1L (href="#bib34" rid="bib34" class=" bibr popnode">Rocha et al., 2009) and is enriched in a cholesterol-rich detergent resistant fraction of phagosomal membranes (href="#bib15" rid="bib15" class=" bibr popnode">Goyette et al., 2012).Taken together, the above observations suggest a molecular connection between dynein, Rab7, and cholesterol within microdomains on the phagosome membrane. Here, we show using multiple experimental approaches that dynein clusters into microdomains on the membrane of a phagosome as it matures inside cells. This geometrical clustering allows many dyneins to simultaneously contact a single MT and generate large cooperative force. This force drives rapid retrograde transport of late phagosomes (LPs), likely enabling their fusion with degradative lysosomes. We also show that lipophosphoglycan, the main molecule used by pathogenic Leishmania donovani parasites to survive inside macrophages, specifically disrupts the clustering of dynein on LP membranes to block retrograde transport of LPs.