Molecular motors are differentially distributed on Golgi membranes from polarized epithelial cells

Microtubules (MT) are required for the efficient transport of membranes from the trans-Golgi and for transcytosis of vesicles from the basolateral membrane to the apical cytoplasm in polarized epithelia. MTs in these cells are primarily oriented with their plus ends basally near the Golgi and their minus-ends in the apical cytoplasm. Here we report that isolated Golgi and Golgi-enriched membranes from intestinal epithelial cells possess the actin based motor myosin-I, the MT minus- end-directed motor cytoplasmic dynein and its in vitro motility activator dynactin (p150/Glued). The Golgi can be separated into stacks, possessing features of the Golgi cisternae, and small membranes enriched in the trans-Golgi network marker TGN 38/41. Whereas myosin-I is present on all membranes in the Golgi fraction, dynein is present only on the small membrane fraction. Dynein, like myosin-I, is associated with membranes as a cytoplasmic peripheral membrane protein. Dynein and myosin-I coassociate with membranes that bind to MTs and cross-link actin filaments and MTs in a nucleotide-dependent manner. We propose that cytoplasmic dynein moves Golgi membranes along MTs to the cell cortex where myosin-I provides local delivery through the actin- rich cytoskeleton to the apical membrane.     

Myosin-I moves actin filaments on a phospholipid substrate: implications for membrane targeting

Acanthamoeba myosin-I bound to substrates of nitrocellulose or planar lipid membranes on glass moved actin filaments at an average velocity of 0.2 micron/s. This movement required ATP and phosphorylation of the myosin-I heavy chain. We prepared planar lipid membranes on a glass support by passive fusion of lipid vesicles (Brian, A. A., and H. M. McConnell. 1984. Proc. Natl. Acad. Sci. USA. 81:6159-6163) composed of phosphatidylcholine and containing 0-40% phosphatidylserine. The mass of lipid that bound to the glass was the same for membranes of 2 and 20% phosphatidylserine in phosphatidylcholine and was sufficient to form a single bilayer. Myosin-I moved actin filaments on planar membranes of 5-40% but not 0-2% phosphatidylserine. At the low concentrations of phosphatidylserine, actin filaments tended to detach suggesting that less myosin-I was bound. We used the cooperative activation of Acanthamoeba myosin-I ATPase by low concentrations of actin to assess the association of phospholipids with myosin-I. Under conditions where activity depends on the binding of actin to the tail of myosin-I (Albanesi, J. P., H. Fujisaki, and E. D. Korn. 1985. J. Biol. Chem. 260:11174-11179), phospholipid vesicles with 5-40% phosphatidylserine inhibited ATPase activity. The motility and ATPase results demonstrate a specific interaction of the tail of myosin-I with physiological concentrations of phosphatidylserine. This interaction is sufficient to support motility and may provide a mechanism to target myosin-I to biological membranes.     

The enterocyte microvillus is a vesicle-generating organelle

For decades, enterocyte brush border microvilli have been viewed as passive cytoskeletal scaffolds that serve to increase apical membrane surface area. However, recent studies revealed that in the in vitro context of isolated brush borders, myosin-1a (myo1a) powers the sliding of microvillar membrane along core actin bundles. This activity also leads to the shedding of small vesicles from microvillar tips, suggesting that microvilli may function as vesicle-generating organelles in vivo. In this study, we present data in support of this hypothesis, showing that enterocyte microvilli release unilamellar vesicles into the intestinal lumen; these vesicles retain the right side out orientation of microvillar membrane, contain catalytically active brush border enzymes, and are specifically enriched in intestinal alkaline phosphatase. Moreover, myo1a knockout mice demonstrate striking perturbations in vesicle production, clearly implicating this motor in the in vivo regulation of this novel activity. In combination, these data show that microvilli function as vesicle-generating organelles, which enable enterocytes to deploy catalytic activities into the intestinal lumen.     

Measuring orientation of actin filaments within a cell: orientation of actin in intestinal microvilli.

Orientational distribution of actin filaments within a cell is an important determinant of cellular shape and motility. To map this distribution we developed a method of measuring local orientation of actin filaments. In this method actin filaments within cells are labeled with fluorescent phalloidin and are viewed at high magnification in a fluorescent microscope. Emitted fluorescence is split by a birefringent crystal giving rise to two images created by light rays polarized orthogonally with respect to each other. The two images are recorded by a high-sensitivity video camera, and polarization of fluorescence at any point is calculated from the relative intensity of both images at this point. From the value of polarization, the orientation of the absorption dipole of the dye, and thus orientation of F-actin, can be calculated. To illustrate the utility of the method, we measured orientation of actin cores in microvilli of chicken intestinal epithelial cells. F-actin in microvillar cores was labeled with rhodamine-phalloidin; measurements showed that the orientation was the same when microvillus formed a part of a brush border and when it was separated from it suggesting that "shaving" of brush borders did not distort microvillar structure. In the absence of nucleotide, polarization of fluorescence of actin cores in isolated microvilli was best fitted by assuming that a majority of fluorophores were arranged with a perfect helical symmetry along the axis of microvillus and that the absorption dipoles of fluorophores were inclined at 52 degrees with respect to the axis. When ATP was added, the shape of isolated microvilli did not change but polarization of fluorescence decreased, indicating statistically significant increase in disorder and a change of average angle to 54 degrees. We argue that these changes were due to mechanochemical interactions between actin and myosin-I.     

Cytoskeleton and vesicle mobility in astrocytes

Exocytotic vesicles in astrocytes are increasingly viewed as essential in astrocyte-to-neuron communication in the brain. In neurons and excitable secretory cells, delivery of vesicles to the plasma membrane for exocytosis involves an interaction with the cytoskeleton, in particular microtubules and actin filaments. Whether cytoskeletal elements affect vesicle mobility in astrocytes is unknown. We labeled single vesicles with fluorescent atrial natriuretic peptide and monitored their mobility in rat astrocytes with depolymerized microtubules, actin, and intermediate filaments and in mouse astrocytes deficient in the intermediate filament proteins glial fibrillary acidic protein and vimentin. In astrocytes, as in neurons, microtubules participated in directional vesicle mobility, and actin filaments played an important role in this process. Depolymerization of intermediate filaments strongly affected vesicle trafficking and in their absence the fraction of vesicles with directional mobility was reduced.     

Endocytosis of junctional cadherins in bovine kidney epithelial (MDBK) cells cultured in low Ca2+ ion medium

The release of intercellular contacts in MDBK cells, initiated by the depletion of Ca2+ ions from the culture medium, results in the endocytotic uptake of membrane vesicles containing specific membrane constituents of the zonula adhaerens (ZA). During this process the junction-derived, endocytosed vesicles remain associated with the ZA plaque components, while the plaque and its attached actin filaments retract as a whole in a ring-like fashion from the plasma membrane, often accumulating, usually in fragments, in the juxtanuclear cytoplasm. Double-label immunofluorescence microscopy with antiplakoglobin and antivinculin has indicated that both plaque proteins colocalize with the hallmark membrane glycoprotein of this junction type, E-cadherin (uvomorulin). When HRP used as a fluid phase marker is applied to the culture medium, simultaneously with the Ca2+ ion-chelator EGTA, numerous HRP-positive vesicles are found in close association with the dislocated plaque material, suggesting that the HRP is contained in the vesicles formed upon EGTA-induced junction splitting. Immunoelectron microscopy with various cadherin-specific antibodies revealed vesicle-associated labeling, confirming the derivation of these plaque-associated vesicles from the ZA. As the desmosome-specific cadherin, desmoglein, is recovered in another type of junction-derived vesicle, which is characterized by its association with a desmoplakin-plaque, we conclude that the membrane domains of both kinds of junction are endocytosed during Ca2+ depletion but stay in different vesicle populations, emphasizing the selective interaction of the specific cadherins with their respective plaque and filament partners.     

Myo1c binding to submembrane actin mediates insulin-induced tethering of GLUT4 vesicles

GLUT4-containing vesicles cycle between the plasma membrane and intracellular compartments. Insulin promotes GLUT4 exocytosis by regulating GLUT4 vesicle arrival at the cell periphery and its subsequent tethering, docking, and fusion with the plasma membrane. The molecular machinery involved in GLUT4 vesicle tethering is unknown. We show here that Myo1c, an actin-based motor protein that associates with membranes and actin filaments, is required for insulin-induced vesicle tethering in muscle cells. Myo1c was found to associate with both mobile and tethered GLUT4 vesicles and to be required for vesicle capture in the total internal reflection fluorescence (TIRF) zone beneath the plasma membrane. Myo1c knockdown or overexpression of an actin binding–deficient Myo1c mutant abolished insulin-induced vesicle immobilization, increased GLUT4 vesicle velocity in the TIRF zone, and prevented their externalization. Conversely, Myo1c overexpression immobilized GLUT4 vesicles in the TIRF zone and promoted insulin-induced GLUT4 exposure to the extracellular milieu. Myo1c also contributed to insulin-dependent actin filament remodeling. Thus we propose that interaction of vesicular Myo1c with cortical actin filaments is required for insulin-mediated tethering of GLUT4 vesicles and for efficient GLUT4 surface delivery in muscle cells.     

Organization of an actin filament-membrane complex. Filament polarity and membrane attachment in the microvilli of intestinal epithelial cells

The association of actin filaments with membranes is now recognized as an important parameter in the motility of nonmuscle cells. We have investigated the organization of one of the most extensive and highly ordered actin filament-membrane complexes in nature, the brush border of intestinal epithelial cells. Through the analysis of isolated, demembranated brush borders decorated with the myosin subfragment, S1, we have determined that all the microvillar actin filaments have the same polarity. The S1 arrowhead complexes point away from the site of attachment of actin filaments at the apical tip of the microvillar membrane. In addition to the end-on attachment of actin filaments at the tip of the microvillus, these filaments are also connected to the plasma membrane all along their lengths by periodic (33 nm) cross bridges. These bridges were best observed in isolated brush borders incubated in high concentrations of Mg++. Their visibility is attributed to the induction of actin paracrystals in the filament bundles of the microvilli. Finally, we present evidence for the presence of myosinlike filaments in the terminal web region of the brush border. A model for the functional organization of actin and myosin in the brush border is presented.     

Evidence for the participation of actin microfilaments and bristle coats in the internalization of gap junction membrane

Thin sections of rabbit granulosa, human SW-13 adrenal cortical adenocarcinoma, and mouse B-16 melanoma cells revealed an apparent single-layered basket of 4- to 7-nm filaments surrounding cytoplasmic gap junction vesicles. This interpretation was based upon apparent longitudinal, cross, and en face sections of structures surrounding the vesicle profiles in tissue treated with tannic acid-glutaraldehyde. In granulosa cells incubated with the S-1 fragment of heavy meromyosin, arrowhead-decorated filaments were observed at the periphery of the gap junction vesicles, suggesting that these filaments contained actin. In addition, we found that small gap junctional blebs and vesicles at the cell surface were coated with short electron-dense bristles similar in appearance to the cloathrin-containing coat of coated vesicles of nonjunctional membrane. The role of these and other cytoskeletal elements in the possible endocytosis of gap junction membrane is discussed.     

Localization of brush border cytoskeletal proteins in gastric oxynticopeptic cells from the bullfrog Rana catesbeiana

The contribution of brush border cytoskeletal proteins (actin, villin, fimbrin, and brush border myosin-1) to organization of the cytoskeletal network underlying apical plications of oxynticopeptic cells was examined by immunohistochemical techniques in frozen sections of gastric mucosa from the bullfrog, Rana catesbeiana. Apical localization of F-actin with phalloidin in oxynticopeptic cells inhibited with cimetidine revealed small, punctate domains within the apical cytoplasm that were consistent with the presence of short microvilli revealed by electron microscopy. Localization of F-actin in cells stimulated with forskolin was limited to a wide continuous band of cytoplasm corresponding to the location of numerous long surface folds. Inhibition of protein synthesis with cycloheximide did not prevent acid secretion or formation of actin filaments within surface folds in stimulated oxynticopeptic cells, suggesting that the formation of filaments does not require actin synthesis. Staining of gastric mucosae with fluorescent DNase-1 demonstrated that oxynticopeptic cells possess an unusually large pool of non-filamentous actin. Taken together, these results suggest that actin-filament formation in stimulated cells occurs by polymerization of an existing pool of non-filamentous actin. Localization of antibodies specific for villin and fimbrin revealed that these proteins were present within intestinal absorptive cells and gastric surface and neck cells but were not present within inhibited or stimulated oxynticopeptic cells. Brush border myosin-1, present in intestinal absorptive cells, was not present in gastric epithelium. Thus, we propose that actin-containing projections in oxynticopeptic cells are not organized like intestinal microvilli and that filament formation occurs after stimulation by modulating intracellular pools of filamentous and non-filamentous actin.     

Location of a protein of the fodrin-spectrin-TW260/240 family in the mouse intestinal brush border

We have determined that a protein of the fodrin-spectrin-TW260/240 (FST) family is a component of the thin fibrils (～5 nm wide, 100–200 nm long) that cross-link bundles of actin filaments to adjacent actin bundles and to the plasma membrane in the terminal web of the brush border of the intestinal epithelium. When isolated brush borders were incubated with anti-fodrin antibodies and prepared for electron microscopy by the quick-freeze, deepetch technique, these ～5 nm fibrils were specifically decorated with the antibody. In addition, these cross-linking fibrils disappeared when the anti-fodrin-reactive proteins were extracted from the brush border. We conclude that FST is a component of a cross-linking system composed of ～5 nm fibrils that are morphologically distinct from the ～8 nm myosin-containing fibrils which were identified by antimyosin decoration. In addition to linking actin bundles to adjacent actin bundles and to the plasma membrane, these FST fibrils may mediate actinvesicle, actin-intermediate filament and vesicle-plasma membrane linkages.     

Distinct cytoskeletal tracks direct individual vesicle populations to the apical membrane of epithelial cells

A key aspect in the structure of epithelial and neuronal cells is the maintenance of a polarized organization based on highly specific sorting machinery at the exit site of the trans Golgi network (TGN). Epithelial cells sort protein and lipid components into different sets of carriers for the apical or basolateral plasma membrane. The two intestinal proteins lactase-phlorizin hydrolase (LPH) and sucrase-isomaltase (SI) are delivered to the apical plasma membrane of epithelial cells with high fidelity but differ in their affinity to detergent-insoluble, glycolipid-enriched complexes (DIGs). Using a two-color labeling technique, we have recently characterized two post-Golgi vesicle populations that direct LPH and SI separately to the apical cell surface. Here, we investigated the structure and identification of protein components in these vesicle populations and assessed the role of cytoskeletal post-Golgi transport routes for apical cargo. Apart from the central role of microtubules in vesicle transport, we demonstrate that the transport of SI-carrying apical vesicles (SAVs) occurs along actin tracks in the cellular periphery, whereas LPH-carrying apical vesicles (LAVs) are transferred in an actin-independent fashion to the apical membrane. Our data further indicate that myosin 1A is the actin-associated motor protein that drives SAVs along actin filaments to the apical cell surface.     

Endogenous plasma membrane t-SNARE syntaxin 4 is present in rab11 positive endosomal membranes and associates with cortical actin cytoskeleton

Membrane fusion requires the formation of a complex between a vesicle protein (v-SNARE) and the target membrane proteins (t-SNAREs). Syntaxin 4 is a t-SNARE that, according to previous overexpression studies, is predominantly localized at the plasma membrane. In the present study endogenous syntaxin 4 was found in intracellular vesicular structures in addition to regions of the plasma membrane. In these vesicular structures syntaxin 4 colocalized with rab11, a marker of recycling endosomes. Furthermore, syntaxin 4 colocalized with actin at the dynamic regions of the plasma membrane. Treatment with N-ethylmaleimide, the membrane transport inhibitor, caused an increased accumulation of syntaxin 4/rab11 positive vesicles in actin filament-like structures. Finally, purified recombinant syntaxin 4 but not syntaxin 2 or 3 cosedimented with actin filaments in vitro, suggesting direct interaction between these two proteins. Taken together, these data suggest that syntaxin 4 regulates secretion at the actin-rich areas of the plasma membrane and may be recycled through rab11 positive intracellular membranes.     

The visualization of actin filament polarity in thin sections. Evidence for the uniform polarity of membrane-associated filaments

We have developed an improved method for visualizing actin filament polarity in thin sections. Myosin subfragment-1 (S-1)-decorated actin filaments display a dramatically enhanced arrowhead configuration when fixed in a medium which contains 0.2 % tannic acid. With the exception of brush borders from intestinal epithelial cells, the arrowhead periodicity of decorated filaments in a variety of nonmuscle cells is similar to that in isolated myofibrils. The periodicity of decorated filaments in brush borders is significantly smaller. Actin filaments which attach to membranes display a clear, uniform polarity, with the S-1 arrowheads pointing away from the plasma membrane, while those which comprise the stress fibers of myoblasts and CHO cells have antiparallel polarities. These observations are consistent with a sliding filament mechanism of cell motility.     

Actin filament-membrane attachment: are membrane particles involved?

The association of actin filaments with membranes is an important feature in the motility of nonmuscle cells. We investigated the role of membrane particles in the attachment of actin filaments to membranes in those systems in which the attachment site can be identified. Freeze fractures through the end-on attachment site of the acrosomal filament bundles in Mytilus (mussel) and Limulus (horseshoe crab) sperm and the attachment site of the microvillar filament bundles in the brush border of intestinal epithelial cells were examined. There are no particles on the P face of the membrane at these sites in the sperm systems and generally none at these sites in microvilli. In microvilli, the actin filaments are also attached along their lengths to the membrane by bridges. When the isolated brush border is incubated in high concentrations of Mg++ (15 mM), the actin filaments form paracrystals and, as a result, the bridges are in register (330 A period). Under these conditions, alignment of the particles on the P face of the membrane into circumferential bands also occurs. However, these bands are generally separated by 800-900 A, indicating that all the bridges cannot be directly attached to membrane particles. Thus membrane particles are not directly involved in the attachment of actin filaments to membranes.     

Synthetic lipid (DOPG) vesicles accumulate in the cell plate region but do not fuse

The cell plate is the new cell wall, with bordering plasma membrane, that is formed between two daughter cells in plants, and it is formed by fusion of vesicles (approximately 60 nm). To start to determine physical properties of cell plate forming vesicles for their transport through the phragmoplast, and fusion with each other, we microinjected fluorescent synthetic lipid vesicles that were made of 1,2-dioleoyl-sn-glycero-3-[phospho-rac-(1-glycerol)] (DOPG) into Tradescantia virginiana stamen hair cells. During interphase, the 60-nm wide DOPG vesicles moved inside the cytoplasm comparably to organelles. During cytokinesis, they were transported through the phragmoplast and accumulated in the cell plate region together with the endogenous vesicles, even inside the central cell plate region. Because at this stage microtubules are virtually absent from that region, while actin filaments are present, actin filaments may have a role in the transport of vesicles toward the cell plate. Unlike the endogenous vesicles, the synthetic DOPG vesicles did not fuse with the developing cell plate. Instead, they redistributed into the cytoplasm of the daughter cells upon completion of cytokinesis. Because the redistribution of the vesicles occurs when actin filaments disappear from the phragmoplast, actin filaments may be involved in keeping the vesicles inside the developing cell plate region.     

Lipid rafts exist as stable cholesterol-independent microdomains in the brush border membrane of enterocytes

Glycosphingolipid/cholesterol-rich membranes ("rafts")can be isolated from many types of cells, but their existence as stable microdomains in the cell membrane has been elusive. Addressing this problem, we studied the distribution of galectin-4, a raft marker, and lactase, a protein excluded from rafts, on microvillar vesicles from the enterocyte brush border membrane. Magnetic beads coated with either anti-galectin-4 or anti-lactase antibodies were used for immunoisolation of vesicles followed by double immunogold labeling of the two proteins. A morphometric analysis revealed subpopulations of raft-rich and raft-poor vesicles by the following criteria: 1) the lactase/galectin-4 labeling ratio/vesicle captured by the anti-lactase beads was significantly higher (p < or = 0.01) than that of vesicles captured by anti-galectin-4 beads, 2) subpopulations of vesicles labeled by only one of the two antibodies were preferentially captured by beads coated with the respective antibody (p < or = 0.01), 3) the average diameter of "galectin-4 positive only" vesicles was smaller than that of vesicles labeled for lactase. Surprisingly, pretreatment with methyl-beta-cyclodextrin, which removed >70% of microvillar cholesterol, did not affect the microdomain localization of galectin-4. We conclude that stable, cholesterol-independent raft microdomains exist in the enterocyte brush border.     

Characterization of Amoeba proteus myosin VI immunoanalog

Amoeba proteus, the highly motile free-living unicellular organism, has been widely used as a model to study cell motility. However, molecular mechanisms underlying its unique locomotion and intracellular actin-based-only trafficking remain poorly understood. A search for myosin motors responsible for vesicular transport in these giant cells resulted in detection of 130-kDa protein interacting with several polyclonal antibodies against different tail regions of human and chicken myosin VI. This protein was binding to actin in the ATP-dependent manner, and immunoprecipitated with anti-myosin VI antibodies. In order to characterize its possible functions in vivo, its cellular distribution and colocalization with actin filaments and dynamin II during migration and pinocytosis were examined. In migrating amoebae, myosin VI immunoanalog localized to vesicular structures, particularly within the perinuclear and sub-plasma membrane areas, and colocalized with dynamin II immunoanalog and actin filaments. The colocalization was even more evident in pinocytotic cells as proteins concentrated within pinocytotic pseudopodia. Moreover, dynamin II and myosin VI immunoanalogs cosedimented with actin filaments, and were found on the same isolated vesicles. Blocking endogenous myosin VI immunoanalog with anti-myosin VI antibodies inhibited the rate of pseudopodia protrusion (about 19% decrease) and uroidal retraction (about 28% decrease) but did not affect cell morphology and the manner of cell migration. Treatment with anti-human dynamin II antibodies led to changes in directionality of amebae migration and affected the rate of only uroidal translocation (about 30% inhibition). These results indicate that myosin VI immunoanalog is expressed in protist Amoeba proteus and may be involved in vesicle translocation and cell locomotion.     

Energy filtered electron tomography of ice-embedded actin and vesicles.

Semiautomatic single-axis tilt electron tomography has been used to visualize the three-dimensional organization of actin filaments in "phantom cells," i.e. lipid vesicles. The instrumentation consisted of a 120-kV electron microscope equipped with a postcolumn energy filter, which was used in the zero-loss imaging mode. Apart from changing the tilt angle, all steps required for automated tomography, such as recentering the image area, refocusing, and centering the energy-selecting slit, were performed by external computer control. This setup permitted imaging of ice-embedded samples up to a thickness of 800 nm with improved image contrast compared with that produced by tomography with a conventional electron microscope. In spite of the missing-wedge effect that is especially obvious in the study of membrane-filament interaction, single-axis tilt tomography was found to be an appropriate (in fact the only available) method for this kind of investigation. In contrast to random actin networks found in actin gels, actin filaments in and on vesicles with a bending radius of less than approximately 2 microns tend to be arranged in single layers of parallel filaments and often induce an elongated shape of the vesicles. Actin filaments located on the outside usually associate with the vesicle membrane.     

Cooperative regulation of endocytic vesicle transport by yeast Eps15-like protein Pan1p and epsins

Dynamic actin filaments are required for the formation and internalization of endocytic vesicles. Yeast actin cables serve as a track for the translocation of endocytic vesicles to early endosomes, but the molecular mechanisms regulating the interaction between vesicles and the actin cables remain ambiguous. Previous studies have demonstrated that the yeast Eps15-like protein Pan1p plays an important role in this interaction, and that interaction is not completely lost even after deletion of the Pan1p actin-binding domain, suggesting that additional proteins mediate association of the vesicle with the actin cable. Other candidates for mediating the interaction are endocytic coat proteins Sla2p (yeast Hip1R) and Ent1p/2p (yeast epsins), as these proteins can bind to both the plasma membrane and the actin filament. Here, we investigated the degree of redundancy in the actin-binding activities of Pan1p, Sla2p, and Ent1p/2p involved in the internalization and transport of endocytic vesicles. Expression of the nonphosphorylatable form of Pan1p, Pan1-18TA, caused abnormal accumulation of both actin cables and endocytic vesicles, and this accumulation was additively suppressed by deletion of the actin-binding domains of both Pan1p and Ent1p. Interestingly, deletion of the actin-binding domains of Pan1p and Ent1p in cells lacking the ENT2 gene resulted in severely defective internalization of endocytic vesicles and recruitment of actin cables to the site of endocytosis. These results suggest that Pan1p and Ent1p/2p cooperatively regulate the interaction between the endocytic vesicle and the actin cable.