SEARCH

SEARCH BY CITATION

Keywords:

  • actin;
  • endosome;
  • GTP-binding proteins;
  • melanosome;
  • myosin V;
  • Rab27;
  • secretory granule

Abstract

  1. Top of page
  2. Abstract
  3. Myosin Motors in Organelle Trafficking
  4. Rab27a recruits myosin Va to melanosomes via an effector protein
  5. Rab27a Recruits Two Distinct Myosins to Melanosomes via Distinct Effectors
  6. Two Distinct Rabs Recruit One Class V Myosin to Distinct Organelles in Yeast
  7. Rab11a Recruits Myosin Vb to Endosomes
  8. Why Rabs as Organelle Receptors for Myosins?
  9. Do Rabs Coordinate Myosin-dependent Transport and Membrane Fusion?
  10. Do Rab Proteins Regulate Cooperation between Myosin and Kinesin?
  11. Acknowledgments
  12. References

The actin cytoskeleton is essential to ensure the proper location of, and communication between, intracellular organelles. Some actin-based myosin motors have been implicated in this process, particularly members of the class V myosins. We discuss here the emerging role of the Ras-like GTPases of the Rab family as regulators of myosin function in organelle transport. Evidence from yeast secretory vesicles and mitochondria, and mammalian melanosomes and endosomes suggests that Rab GTPases are crucial components of the myosin organelle receptor machinery. Better understood is the case of the melanosome where Rab27a recruits a specific effector called melanophilin, which in turn binds myosin Va. The presence of a linker protein between a Rab and a myosin may represent a general mechanism. We argue that Rabs are ideally suited to perform this role as they are exquisite organelle markers. Furthermore, the molecular switch property of Rabs may enable them to regulate the timing of the myosin association with the target organelle.

The actin-based movement of organelles within cells is a general process that is essential to the proper distribution and communication between cellular compartments. One type of actin-based motility is based on actin polymerization propelling organelles within the cytoplasm, such as some movements observed for endosomes and lysosomes (1,2). This type of actin-based movement was reviewed recently in Traffic and will not be discussed much further here (3). The second type of movement depends on acto-myosin forces to move organelles or vesicles on actin tracks. We will focus our discussion on this type of motility, and on the emerging role of Rab GTPases as regulators of this process.

Myosin Motors in Organelle Trafficking

  1. Top of page
  2. Abstract
  3. Myosin Motors in Organelle Trafficking
  4. Rab27a recruits myosin Va to melanosomes via an effector protein
  5. Rab27a Recruits Two Distinct Myosins to Melanosomes via Distinct Effectors
  6. Two Distinct Rabs Recruit One Class V Myosin to Distinct Organelles in Yeast
  7. Rab11a Recruits Myosin Vb to Endosomes
  8. Why Rabs as Organelle Receptors for Myosins?
  9. Do Rabs Coordinate Myosin-dependent Transport and Membrane Fusion?
  10. Do Rab Proteins Regulate Cooperation between Myosin and Kinesin?
  11. Acknowledgments
  12. References

Advances in video-enhanced contrast microscopy in the early 1980s enabled the visualization of cytoplasmic movement of organelles. Pioneering experiments in algae and the squid giant axon suggested that organelles moved on actin filaments, in addition to long-distance displacements along microtubules (4–6). These early studies suggested that actin-based movements carried organelles or vesicles directionally along actin tracks and involved myosins, the molecular motors associated with actin filaments. The first myosin proposed to move an organelle on actin filaments was a myosin I from the amoeba Acanthamoeba castellanii on the basis of an in vitro motility assay (5).

Myosins share a common structure and are composed of three modules: the head domain comprising the motor domain, which hydrolyzes adenosine 5′ triphosphate (ATP) to power movement along actin filaments; the neck domain, which acts as a lever arm and binds regulatory light-chains such as calmodulin; and the tail domain, which is highly divergent among the different myosin classes and serves to bind specific cargo. Despite their common motor domain structures, myosins present fundamental differences in the kinetics of their ATP cycle and oligomeric structure. Some myosins are processive, i.e. can perform many consecutive steps along actin without dissociating, whereas other myosins can exert tensions between actin filaments, or actin filaments and membrane domains.

Among the 18 classes that constitute the myosin superfamily (7), four classes of myosins (I, II, V and VI) have been implicated in the dynamics of a large variety of organelles such as secretory vesicles, vacuoles, Golgi, endoplasmic reticulum and mitochondria in yeast (8–12), and endoplasmic reticulum, recycling endosomes, lysosomes, secretory granules and melanosomes in mammalian cells (13–18).

Myosin V may be ideally suited for organelle transport as it is an efficient, processive motor that works in 37-nm steps which correspond to the helical periodicity of the actin filament (19). Indeed, several class V myosins have been involved in organelle trafficking. In yeast, one of two class V myosins, Myo2p, delivers various organelles to the bud tip during cell division in Saccharomyces cerevisiae8–12). There are three myosin V-related genes in mammals, myosin Va, myosin Vb and myosin Vc (7). In addition, myosin genes may be alternatively spliced, thus generating tissue-specific isoforms (20). These different myosin V isoforms are also involved in the transport of multiple organelles such as the endoplasmic reticulum, melanosomes, secretory granules and recycling endosomes (Table 1).

Table 1. Summary of reported interactions between Rab GTPases and myosin motors
RabsMotorsMode of interactionOrganelleReferences
Rab 11Myosin VbDirect?Recycling(17,54)
  FIP2endosomes 
Sec4pMyo2 (class?Secretory(8)
 V myosin) vesicles 
Ypt11pMyo2Direct?Mitochondria(11)
Rab27a (GriscelliMyosin VaMelanophilinSkin melanocyte(38,41,66)
syndrome, ashen mice)  melanosomes 
Rab27aMyosin VIIaMyRIPRPE melanosomes(43,48)
Rab8?Myosin Vc?Endosomes(55)

The divergent tail domain of myosins contains the information necessary for the targeting of myosins to specific intracellular compartments. However, until recently, little was known concerning the link between myosin tails and their cognate organelle. Genetic, morphologic and biochemical analysis of three mouse coat color mutants, dilute, ashen and leaden, gave some clues as to the mechanisms regulating the binding of one class V myosin to melanosomes, the pigment-containing granules of skin melanocytes (21–23).

Rab27a recruits myosin Va to melanosomes via an effector protein

  1. Top of page
  2. Abstract
  3. Myosin Motors in Organelle Trafficking
  4. Rab27a recruits myosin Va to melanosomes via an effector protein
  5. Rab27a Recruits Two Distinct Myosins to Melanosomes via Distinct Effectors
  6. Two Distinct Rabs Recruit One Class V Myosin to Distinct Organelles in Yeast
  7. Rab11a Recruits Myosin Vb to Endosomes
  8. Why Rabs as Organelle Receptors for Myosins?
  9. Do Rabs Coordinate Myosin-dependent Transport and Membrane Fusion?
  10. Do Rab Proteins Regulate Cooperation between Myosin and Kinesin?
  11. Acknowledgments
  12. References

Mercer et al. showed that the dilute locus in mice encodes the murine ortholog of myosin Va (24). This coat color pigment dilution phenotype, associated with neurologic disease leading to early neonatal death in dilute-lethal mice, is due to loss-of-function myosin Va mutations. Less severe alleles such as dilute-viral that exhibit only pigment dilution primarily affect the synthesis of the melanocyte-specific alternatively spliced form (containing exon F) of myosin Va (20).

Visible pigmentation in mammals requires the transfer of pigment granules from the melanocytes to adjacent keratinocytes in the basal layers of the epidermis. Once mature and fully pigmented, melanosomes must be transported to the distal end of the melanocyte dendrite extensions in order to be efficiently transferred to keratinocytes (25,26). The pigment dilution in dilute mice results from defects in melanosome transport within skin melanocytes, such that dilute melanosomes appear clustered around the perinuclear area and only a few melanosomes are observed at the periphery of the cell, in contrast to what is observed in a wild type melanocyte (Figure 1). Association of myosin Va with melanosomes has been revealed by immuno-electron microscopy and subcellular fractionation (27–29). Furthermore, dominant-negative protein expression and in vitro motility assay reconstituting the movements of melanosomes on actin bundles further confirmed the role of myosin Va as the actin-based motor involved in the transport of melanosomes (18,29).

image

Figure 1. Melanosome transport defects in mouse models of Griscelli Syndrome. Phase contrast photographs of primary cultures skin melanocytes derived from wild type, ashen (Rab27aash), leaden (Mlphln) and dilute (Myo5ad) are shown.

Ten years after the initial description of myosin Va as the dilute gene, several independent lines of investigation focused on Rab27a as a component of the organelle trafficking machinery acting in concert with myosin Va. Firstly, Rab27a mutations were found in patients suffering from Griscelli syndrome (30). Griscelli syndrome is a disease that associates partial albinism with characteristic gray hair and immunodeficiency due to loss of cytotoxic T-lymphocyte killing. Concomitantly, the ashen gene was identified as Rab27a, leading to the identification of ashen as a model of Griscelli syndrome (31). Secondly, both Rab27a and myosin Va colocalize on wild type melanosomes in skin melanocytes, and expression of Rab27a in melanocytes from patients suffering from Griscelli syndrome restores the distribution of melanosomes to the tips of the dendrites (32–34). Thirdly, the association of Rab27a and myosin Va was confirmed by coimmunoprecipitation (33). However, Rab27a and myosin Va do not interact directly. Instead, they require the product of the leaden gene, which was called melanophilin (Mlph, also called Slac-2a) (35). The ashen, leaden and dilute gene products are thus believed to be responsible for the recruitment of melanosomes to actin filaments at the cell periphery. The absence of any one of these gene products leads to striking clustering of melanosomes around the nucleus as in dilute melanocytes (Figure 1).

Rab27a localizes to melanosomes irrespective of the presence of Mlph or myosin Va, suggesting that it represents the organelle-recognition element in the system (36–38). In wild type melanocytes, Rab27a associates with mature melanosomes, and upon activation via GTP loading, Rab27a-GTP recruits melanophilin. Melanophilin stabilized at the melanosome surface then binds the tail of the melanocyte-specific spliced form of myosin Va. The motor domain of this myosin then allows association of melanosomes with the actin network and leads to their accumulation in dendrite extensions (Figure 2A).

image

Figure 2. Molecular mechanisms of Rab-dependent, myosin-driven organelle motility. See text for details.

Rab27a Recruits Two Distinct Myosins to Melanosomes via Distinct Effectors

  1. Top of page
  2. Abstract
  3. Myosin Motors in Organelle Trafficking
  4. Rab27a recruits myosin Va to melanosomes via an effector protein
  5. Rab27a Recruits Two Distinct Myosins to Melanosomes via Distinct Effectors
  6. Two Distinct Rabs Recruit One Class V Myosin to Distinct Organelles in Yeast
  7. Rab11a Recruits Myosin Vb to Endosomes
  8. Why Rabs as Organelle Receptors for Myosins?
  9. Do Rabs Coordinate Myosin-dependent Transport and Membrane Fusion?
  10. Do Rab Proteins Regulate Cooperation between Myosin and Kinesin?
  11. Acknowledgments
  12. References

Rab27a and its close homolog Rab27b interact with at least 10 different effectors. Most of the effectors interact with Rab27 via a conserved Rab27-binding domain, which is present in Mlph and the Synaptotagmin-like proteins (Slp) (39–42). MyRIP (for Myosin and Rab Interacting Protein, also called Slac-2c) was originally identified as a myosin VIIa-binding protein and the presence of a canonical N-terminal Rab27-binding domain as well as its close homology with Mlph suggested a role as a linker between Rabs and myosins (43,44).

Myosin VIIa plays a critical role in the inner ear and the retina, since mutations in the MYO7A gene lead to the blindness–deafness Usher syndrome type 1B (45). In the retina, myosin VIIa is required for the normal localization of melanosomes in the apical area of retinal pigment epithelial (RPE) cells, for phagocytosis of photoreceptor disks by RPE cells and for the transport of opsins through the connecting cilium in photoreceptors (43,46,47). Recent observations indicate that Rab27a also contributes to the distribution of melanosomes in RPE cells (48). Altogether, the available data suggest that a Rab27a/MyRIP/myosin VIIa complex regulates melanosome distribution in the RPE (Figure 2B). Thus, Rab27 appears to be able to recruit distinct myosins (Va or VIIa) via different effectors (Mlph or MyRIP) to regulate the dynamics of the same type of organelle (melanosome) in different cell types (melanocytes or RPE cells).

There are significant differences between RPE and melanocyte melanosomes both in terms of cellular distribution and ability to undergo cell transfer (25,48). This discrepancy might reflect the different properties of the two myosins with which they associate. Myosin Va is clearly a processive motor and has been observed to move melanosomes on actin filaments, while myosin VIIa has been proposed to bridge membrane domains to actin filaments rather than to act as a processive motor (19,49).

Rab27a and MyRIP have also been implicated in the regulation of exocytosis in pancreatic β cells and neuro-endocrine PC12 cells (50,51). However, the Rab27/MyRIP protein complex does not appear to require recruitment of a myosin on the secretory granules for function (50,51). The association of myosin Va with Rab27/MyRIP has not been demonstrated in vivo, although it contributes to the cytoplasmic distribution of secretory granules in PC12 cells (13). Myosin VIIa is apparently not expressed in pancreatic β cells (50).

Interestingly, the two known Rab27 and myosin linker proteins, Mlph and MyRIP, also contain a putative actin-binding domain near the C-terminus (44,52) (Figure 2A and B). Overexpression of an actin-binding domain mutant form of Mlph leads to a dominant-negative effect in cultured melanocytes, i.e. induces perinuclear clustering of melanosomes, as observed in ashen, leaden or dilute (52) (Figure 1). This domain might dock melanosomes on actin filaments and facilitate the interaction between myosin Va and Rab27a (Figure 2A and B). Alternatively, it may contribute to the capture and immobilization of melanosomes on actin filaments in the vicinity of the plasma membrane, prior to transfer to adjacent keratinocytes (52).

Two Distinct Rabs Recruit One Class V Myosin to Distinct Organelles in Yeast

  1. Top of page
  2. Abstract
  3. Myosin Motors in Organelle Trafficking
  4. Rab27a recruits myosin Va to melanosomes via an effector protein
  5. Rab27a Recruits Two Distinct Myosins to Melanosomes via Distinct Effectors
  6. Two Distinct Rabs Recruit One Class V Myosin to Distinct Organelles in Yeast
  7. Rab11a Recruits Myosin Vb to Endosomes
  8. Why Rabs as Organelle Receptors for Myosins?
  9. Do Rabs Coordinate Myosin-dependent Transport and Membrane Fusion?
  10. Do Rab Proteins Regulate Cooperation between Myosin and Kinesin?
  11. Acknowledgments
  12. References

The S. cerevisiae Myo2p is another class V myosin involved in organelle motility. Two Rab proteins, Sec4p and Ypt11, have been implicated in the docking of Myo2p onto secretory vesicles and mitochondria, respectively. Genetic evidence suggests that Sec4p interacts with the tail domain of Myo2p, but a direct interaction has not been demonstrated (53). Furthermore, mutations in Sec2p, a guanine nucleotide exchange factor (GEF) and thus an activator of Sec4p, uncouple Myo2p from secretory vesicles (53). These observations support a similar mode of action of a Rab GTPase and a myosin motor in yeast, whereby a class V myosin is recruited to the surface of an organelle upon activation of a Rab protein (Sec4p) (Figure 2D).

Ypt11p has recently been implicated in the inheritance of mitochondria (11). Deletion of Ypt11 induces a partial delay in transmission of mitochondria to the bud, whereas its overexpression results in the accumulation of mitochondria in the bud (11). The Ypt11 GTPase binds to the tail domain of Myo2p, although it is unclear whether the interaction is direct or indirect. Furthermore, residues in Myo2p, whose substitution induces defects in mitochondrial transport, do not affect transport of secretory vesicles and vice versa, suggesting that Sec4p and Ypt11 bind different motifs on the tail of Myo2p (11,53) (Figure 2D).

Rab11a Recruits Myosin Vb to Endosomes

  1. Top of page
  2. Abstract
  3. Myosin Motors in Organelle Trafficking
  4. Rab27a recruits myosin Va to melanosomes via an effector protein
  5. Rab27a Recruits Two Distinct Myosins to Melanosomes via Distinct Effectors
  6. Two Distinct Rabs Recruit One Class V Myosin to Distinct Organelles in Yeast
  7. Rab11a Recruits Myosin Vb to Endosomes
  8. Why Rabs as Organelle Receptors for Myosins?
  9. Do Rabs Coordinate Myosin-dependent Transport and Membrane Fusion?
  10. Do Rab Proteins Regulate Cooperation between Myosin and Kinesin?
  11. Acknowledgments
  12. References

In mammals, another member of the class V myosin, myosin Vb, has been shown to regulate a membrane traffic step (plasma membrane recycling) through its association with a distinct Rab GTPase (Rab11a) (17) (Figure 2C). It is not yet clear whether the interaction is direct or whether it requires the Rab11a and myosin Vb-binding protein, Rab11 family interacting protein-2 (Rab11-FIP2) (54). The interaction of one Rab and one myosin via a linker protein is reminiscent of the Rab27a/melanophilin/myosin Va interaction.

Myosin Vc, the major class V isoform in epithelial cells is also involved in regulating the cellular distribution of endocytic compartments distinct from those exhibiting myosin Vb (55). Overexpression of myosin Vc perturbs the distribution of compartments labeled for Rab8 and the transferrin receptor. Thus Rab8 is another potential candidate Rab, in addition to Rab27, Rab11, Sec4 and Ypt11, which might selectively recruit a class V myosin to a defined organelle.

Why Rabs as Organelle Receptors for Myosins?

  1. Top of page
  2. Abstract
  3. Myosin Motors in Organelle Trafficking
  4. Rab27a recruits myosin Va to melanosomes via an effector protein
  5. Rab27a Recruits Two Distinct Myosins to Melanosomes via Distinct Effectors
  6. Two Distinct Rabs Recruit One Class V Myosin to Distinct Organelles in Yeast
  7. Rab11a Recruits Myosin Vb to Endosomes
  8. Why Rabs as Organelle Receptors for Myosins?
  9. Do Rabs Coordinate Myosin-dependent Transport and Membrane Fusion?
  10. Do Rab Proteins Regulate Cooperation between Myosin and Kinesin?
  11. Acknowledgments
  12. References

The available experimental evidence suggests that there is great complexity in myosin-driven organelle dynamics. The same class V myosin isoform transports multiple organelles with unique destinations. The generation of multiple splice isoforms creates diversity in the myosin tail domain and thus enables myosins to operate in multiple transport steps. Furthermore, myosin V-driven motility of an organelle is differentially controlled during the cell cycle, as shown for the endoplasmic reticulum in Xenopus egg extracts (56). Therefore, the interaction of myosin with organelles needs to be regulated in time and space. The recent evidence suggesting that the interaction between Rabs and myosins might be a general feature in organelle motility raises important questions concerning the suitability of Rab GTPases to perform the role required of a myosin receptor in membrane traffic. Several of the known properties of Rabs provide some clues to the answers.

Rab GTPases represent the largest family within the Ras superfamily and mammalian genomes contain more than 60 independent genes encoding Rab proteins (57,58). Rabs are localized within cellular organelles and increasing evidence suggests that Rabs may be important recognition elements in the organelle identification machinery (59). As such, Rabs are among the best known organelle markers and myosins could thus connect precisely to the desired organelle. In addition to their role as organelle markers, Rabs function as GTP-dependent molecular switches. The ability to control the activation and inactivation of any one Rab in time and space may provide the means to regulate the association and dissociation of myosins with the appropriate organelle.

Do Rabs Coordinate Myosin-dependent Transport and Membrane Fusion?

  1. Top of page
  2. Abstract
  3. Myosin Motors in Organelle Trafficking
  4. Rab27a recruits myosin Va to melanosomes via an effector protein
  5. Rab27a Recruits Two Distinct Myosins to Melanosomes via Distinct Effectors
  6. Two Distinct Rabs Recruit One Class V Myosin to Distinct Organelles in Yeast
  7. Rab11a Recruits Myosin Vb to Endosomes
  8. Why Rabs as Organelle Receptors for Myosins?
  9. Do Rabs Coordinate Myosin-dependent Transport and Membrane Fusion?
  10. Do Rab Proteins Regulate Cooperation between Myosin and Kinesin?
  11. Acknowledgments
  12. References

The coordination of vesicle motility is just one of the roles Rabs play. In fact, Rabs have been implicated in almost every step of vesicular transport (57,60). The key to the multiple roles of Rabs appears to reside in the ability to interact with multiple effectors. For example, recombinant activated Rab5-GTP interacts with more than 20 bovine brain cytosolic proteins (61). Rabs and their effectors may play multiple roles in transport pathways and may thus have sequential functions, such as the recruitment of myosins via one specific effector and recruitment of tethering factors via other effectors. For Rab27, effector proteins of the Slp family contain tandem Ca++ and phospholipid-binding C2 domains in addition to a Rab27-binding domain, and thus may play tethering roles. Another recently identified effector of Rab27 is Munc13–4, which is thought to regulate SNARE function in membrane fusion (62). This evidence raises the interesting possibility that Rabs recruit hierarchically a set of effectors and thereby coordinate organelle transport with tethering and fusion to their target compartment. However, the mechanistic details of how this hierarchy of effects could be achieved remain mysterious.

Do Rab Proteins Regulate Cooperation between Myosin and Kinesin?

  1. Top of page
  2. Abstract
  3. Myosin Motors in Organelle Trafficking
  4. Rab27a recruits myosin Va to melanosomes via an effector protein
  5. Rab27a Recruits Two Distinct Myosins to Melanosomes via Distinct Effectors
  6. Two Distinct Rabs Recruit One Class V Myosin to Distinct Organelles in Yeast
  7. Rab11a Recruits Myosin Vb to Endosomes
  8. Why Rabs as Organelle Receptors for Myosins?
  9. Do Rabs Coordinate Myosin-dependent Transport and Membrane Fusion?
  10. Do Rab Proteins Regulate Cooperation between Myosin and Kinesin?
  11. Acknowledgments
  12. References

Microtubule-based and actin-based movements cooperate in the transport of organelles from the perinuclear region to the cell periphery, and vice versa. In skin melanocytes, the bidirectional microtubule-dependent transport system provides a way of transporting the melanosomes to the cell periphery, where the actomyosin system acts to ensure that the mature melanosomes are captured and stay in position to be transferred to keratinocytes. This system has been a fruitful one to study cooperation between the cytoskeletal networks. Several models have been proposed to describe the cooperation between microtubule-based motors and myosins (21,63,64). Most recently, the ‘tug-of-war’ model proposes a competition between microtubule-based motors and myosins. In Xenopus, myosin V contributes to the dispersion of melanosomes by counteracting the action of dynein, particularly shortening the length of dynein-driven runs, and by ensuring the regular motion of melanosomes towards the microtubule plus-end (65).

The productive association of myosins with organelles might therefore be coordinated with the inactivation of microtubule motors. Genetic evidence in yeast suggests that the kinesin-related protein, Smy1, stabilizes the protein complex formed by Sec4p, and Myo2p. As molecular switches, Rabs are ideally suited to regulate this type of coordination of events but a direct role for Rabs in this process has yet to be uncovered.

In conclusion, the experimental evidence reviewed herein strongly suggests that the recruitment of different isoforms of the processive motor myosin V onto organelles is regulated by Rab proteins. Whether or not other myosins involved in membrane trafficking such as class I, II or VI myosins are also regulated by their specific recruitment by Rab proteins remains an interesting possibility.

References

  1. Top of page
  2. Abstract
  3. Myosin Motors in Organelle Trafficking
  4. Rab27a recruits myosin Va to melanosomes via an effector protein
  5. Rab27a Recruits Two Distinct Myosins to Melanosomes via Distinct Effectors
  6. Two Distinct Rabs Recruit One Class V Myosin to Distinct Organelles in Yeast
  7. Rab11a Recruits Myosin Vb to Endosomes
  8. Why Rabs as Organelle Receptors for Myosins?
  9. Do Rabs Coordinate Myosin-dependent Transport and Membrane Fusion?
  10. Do Rab Proteins Regulate Cooperation between Myosin and Kinesin?
  11. Acknowledgments
  12. References