pH-dependent distribution of chlorin e6 derivatives across phospholipid bilayers probed by NMR spectroscopy

The pH-dependent membrane adsorption and distribution of three chlorin derivatives, chlorin e6 (CE), rhodin G7 (RG), and monoaspartyl-chlorin e6 (MACE), in the physiological pH range (pH 6-8) were probed by NMR spectroscopy. Unilamellar vesicles consisting of dioleoyl-phosphatidyl-choline (DOPC) were used as membrane models. The chlorin derivatives were characterized with respect to their aggregation behavior, the pK(a) values of individual carboxylate groups, the extent of membrane adsorption, and their flip-flop rates across the bilayer membrane for pH 6-8. External membrane adsorption was found to be lower for RG than for CE and MACE. Both electrostatic interactions and the extent of aggregation seemed to be the main determinants of membrane adsorption. Rate constants for chlorin transfer across the membrane were found to correlate strongly with the pH of the surrounding medium, in particular, for CE and RG. In acidic solution, CE and RG transfer across the membrane was strongly accelerated, and in basic solution, all compounds were retained, mostly in the outer monolayer. In contrast, MACE flip-flop across the membrane remained very low even at pH 6. The protonation of ionizable groups is suggested to be a major determinant of chlorin transfer rates across the bilayer. pK(a) values of CE and RG were found to be between 6 and 8, and two of the carboxylate groups in MACE had pK(a) values below 6. For CE and RG, the kinetic profiles at acidic pH indicated that the initial fast membrane distribution was followed by secondary steps that are discussed in this article.

Interactions between selected photosensitizers and model membranes: an NMR classification

Membrane interactions of porphyrinic photosensitizers (PSs) are known to play a crucial role for PS efficiency in photodynamic therapy (PDT). In the current paper, the interactions between 15 different porphyrinic PSs with various hydrophilic/lipophilic properties and phospholipid bilayers were probed by NMR spectroscopy. Unilamellar vesicles consisting of dioleoyl-phosphatidyl-choline (DOPC) were used as membrane models. PS-membrane interactions were deduced from analysis of the main DOPC (1)H-NMR resonances (choline and lipid chain signals). Initial membrane adsorption of the PSs was indicated by induced changes to the DOPC choline signal, i.e. a split into inner and outer choline peaks. Based on this parameter, the PSs could be classified into two groups, Type-A PSs causing a split and the Type-B PSs causing no split. A further classification into two subgroups each, A1, A2 and B1, B2 was based on the observed time-dependent changes of the main DOPC NMR signals following initial PS adsorption. Four different time-correlated patterns were found indicating different levels and rates of PS penetration into the hydrophobic membrane interior. The type of interaction was mainly affected by the amphiphilicity and the overall lipophilicity of the applied PS structures. In conclusion, the NMR data provided valuable structural and dynamic insights into the PS-membrane interactions which allow deriving the structural constraints for high membrane affinity and high membrane penetration of a given PS. (C) 2011 Elsevier B.V. All rights reserved.

Distribution of intracellular and secreted surfactant during postnatal rat lung development

Pulmonary surfactant prevents alveolar collapse via reduction of surface tension. In contrast to human neonates, rats are born with saccular lungs. Therefore, rat lungs serve as a model for investigation of the surfactant system during postnatal alveolar formation. We hypothesized that this process is associated with characteristic structural and biochemical surfactant alterations. We aimed to discriminate changes related to alveolarization from those being either invariable or follow continuous patterns of postnatal changes. Secreted active (mainly tubular myelin (tm)) and inactive (unilamellar vesicles (ulv)) surfactant subtypes as well as intracellular surfactant (lamellar bodies (lb)) in type II pneumocytes (PNII) were quantified before (day (d) 1), during (d 7), at the end of alveolarization (d 14), and after completion of lung maturation (d 42) using electron microscopic methods supplemented by biochemical analyses (phospholipid quantification, immunoblotting for SP-A). Immunoelectron microscopy determined the localization of surfactant protein A (SP-A). (1) At d 1 secreted surfactant was increased relative to d 7-42 and then decreased significantly. (2) Air spaces of neonatal lungs comprised lower fractions of tm and increased ulv, which correlated with low SP-A concentrations in lung lavage fluid (LLF) and increased respiratory rates, respectively. (3) Alveolarization (d 7-14) was associated with decreasing PNII size although volume and sizes of Lb continuously increased. (4) The volume fractions of Lb correlated well with the pool sizes of phospholipids in lavaged lungs. Our study emphasizes differential patterns of developmental changes of the surfactant system relative to postnatal alveolarization.

Rab5 Induces Rac-independent Lamellipodia Formation and Cell Migration

Rab5 is a regulatory GTPase of vesicle docking and fusion that is involved in receptor-mediated endocytosis and pinocytosis. Introduction of active Rab5 in cells stimulates the rate of endocytosis and vesicle fusion, resulting in the formation of large endocytic vesicles, whereas dominant negative Rab5 inhibits vesicle fusion. Here we show that introduction of active Rab5 in fibroblasts also induced reorganization of the actin cytoskeleton but not of microtubule filaments, resulting in prominent lamellipodia formation. The Rab5-induced lamellipodia formation did not require activation of PI3-K or the GTPases Ras, Rac, Cdc42, or Rho, which are all strongly implicated in cytoskeletal reorganization. Furthermore, lamellipodia formation by insulin, Ras, or Rac was not affected by expression of dominant negative Rab5. In addition, cells expressing active Rab5 displayed a dramatic stimulation of cell migration, with the lamellipodia serving as the leading edge. Both lamellipodia formation and cell migration were dependent on actin polymerization but not on microtubules. These results demonstrate that Rab5 induces lamellipodia formation and cell migration and that the Rab5-induced lamellipodia formation occurs by a novel mechanism independent of, and distinct from, PI3-K, Ras, or Rho-family GTPases. Thus, Rab5 can control not only endocytosis but also actin cytoskeleton reorganization and cell migration, which provides strong support for an intricate relationship between these processes.

RAC1 Regulates Adherens Junctions through Endocytosis of E-Cadherin

The establishment of cadherin-dependent cell–cell contacts in human epidermal keratinocytes are known to be regulated by the Rac1 small GTP-binding protein, although the mechanisms by which Rac1 participates in the assembly or disruption of cell–cell adhesion are not well understood. In this study we utilized green fluorescent protein (GFP)-tagged Rac1 expression vectors to examine the subcellular distribution of Rac1 and its effects on E-cadherin–mediated cell–cell adhesion. Microinjection of keratinocytes with constitutively active Rac1 resulted in cell spreading and disruption of cell–cell contacts. The ability of Rac1 to disrupt cell–cell adhesion was dependent on colony size, with large established colonies being resistant to the effects of active Rac1. Disruption of cell–cell contacts in small preconfluent colonies was achieved through the selective recruitment of E-cadherin–catenin complexes to the perimeter of multiple large intracellular vesicles, which were bounded by GFP-tagged L61Rac1. Similar vesicles were observed in noninjected keratinocytes when cell–cell adhesion was disrupted by removal of extracellular calcium or with the use of an E-cadherin blocking antibody. Moreover, formation of these structures in noninjected keratinocytes was dependent on endogenous Rac1 activity. Expression of GFP-tagged effector mutants of Rac1 in keratinocytes demonstrated that reorganization of the actin cytoskeleton was important for vesicle formation. Characterization of these Rac1-induced vesicles revealed that they were endosomal in nature and tightly colocalized with the transferrin receptor, a marker for recycling endosomes. Expression of GFP-L61Rac1 inhibited uptake of transferrin-biotin, suggesting that the endocytosis of E-cadherin was a clathrin-independent mechanism. This was supported by the observation that caveolin, but not clathrin, localized around these structures. Furthermore, an inhibitory form of dynamin, known to inhibit internalization of caveolae, inhibited formation of cadherin vesicles. Our data suggest that Rac1 regulates adherens junctions via clathrin independent endocytosis of E-cadherin.

Toxin-induced pore formation is hindered by intermolecular hydrogen bonding in sphingomyelin bilayers

Sticholysin I and II (StnI and StnII) are pore-forming toxins that use sphingomyelin (SM) for membrane binding. We examined how hydrogen bonding among membrane SMs affected the StnI- and StnII-induced pore formation process, resulting in bilayer permeabilization. We compared toxin-induced permeabilization in bilayers containing either SM or dihydro-SM (lacking the trans 4 double bond of the long-chain base), since their hydrogen-bonding properties are known to differ greatly. We observed that whereas both StnI and StnII formed pores in unilamellar vesicles containing palmitoyl-SM or oleoyl-SM, the toxins failed to similarly form pores in vesicles prepared from dihydro-PSM or dihydro-OSM. In supported bilayers containing OSM, StnII bound efficiently, as determined by surface plasmon resonance. However, StnII binding to supported bilayers prepared from dihydro-OSM was very low under similar experimental conditions. The association of the positively charged StnII (at pH 7.0) with unilamellar vesicles prepared from OSM led to a concentration-dependent increase in vesicle charge, as determined from zeta-potential measurements. With dihydro-OSM vesicles, a similar response was not observed. Benzyl alcohol, which is a small hydrogen-bonding compound with affinity to lipid bilayer interfaces, strongly facilitated StnII-induced pore formation in dihydro-OSM bilayers, suggesting that hydrogen bonding in the interfacial region originally prevented StnII from membrane binding and pore formation. We conclude that interfacial hydrogen bonding was able to affect the membrane association of StnI- and StnII, and hence their pore forming capacity. Our results suggest that other types of protein interactions in bilayers may also be affected by hydrogen-bonding origination from SMs.

Microscopy imaging of liposomes:from coverslips to environmental SEM

A plethora of techniques for the imaging of liposomes and other bilayer vesicles are available. However, sample preparation and the technique chosen should be carefully considered in conjunction with the information required. For example, larger vesicles such as multilamellar and giant unilamellar vesicles can be viewed using light microscopy and whilst vesicle confirmation and size prior to additional physical characterisations or more detailed microscopy can be undertaken, the technique is limited in terms of resolution. To consider the options available for visualising liposome-based systems, a wide range of microscopy techniques are described and discussed here: these include light, fluorescence and confocal microscopy and various electron microscopy techniques such as transmission, cryo, freeze fracture and environmental scanning electron microscopy. Their application, advantages and disadvantages are reviewed with regard to their use in analysis of lipid vesicles.

Liposome-mediated DNA immunisation via the subcutaneous route

Compared to naked DNA immunisation, entrapment of plasmid-based DNA vaccines into liposomes by the dehydration-rehydration method has shown to enhance both humoural and cell-mediated immune responses to encoded antigens administered by a variety of routes. In this paper, we have investigated the application of liposome-entrapped DNA and their cationic lipid composition on such potency after subcutaneous immunisation. Plasmid pI.18Sfi/NP containing the nucleoprotein (NP) gene of A/Sichuan/2/87 (H3N2) influenza virus in the pI.18 expression vector was incorporated by the dehydration-rehydration method into liposomes composed of 16 μmol egg phosphatidylcholine (PC), 8 μmoles dioleoyl phosphatidylethanolamine (DOPE) or cholesterol (Chol) and either the cationic lipid 1,2-diodeoyl-3-(trimethylammonium) propane (DOTAP) or cholesteryl 3-N-(dimethyl amino ethyl) carbamate (DC-Chol). This method, entailing mixing of small unilamellar vesicles (SUV) with DNA, followed by dehydration and rehydration, yielded incorporation values of 90-94% of the DNA used. Mixing or rehydration of preformed cationic liposomes with 100 μg plasmid DNA also led to similarly high complexation values (92-94%). In an attempt to establish differences in the nature of DNA association with these various liposome preparations their physico-chemical characteristics were investigated. Studies on vesicle size, zeta potential and gel electrophoresis in the presence of the anion sodium dodecyl sulphate (SDS) indicate that, under the conditions employed, formulation of liposomal DNA by the dehydration-rehydration generated submicron size liposomes incorporating most of the DNA in a manner that prevents DNA displacement through anion competition. The bilayer composition of these dehydration-rehydration vesicles (DRV(DNA)) can also further influence these physicochemical characteristics with the presence of DOPE within the liposome bilayer resulting in a reduced vesicle zeta potential. Subcutaneous liposome-mediated DNA immunisation employing two DRV(DNA) formulations as well as naked DNA revealed that humoural responses (immunoglobulin total IgG, and subclasses IgG1 and 1gG2a) engendered by the plasmid encoded NP were substantially higher after dosing twice, 28 days apart with 10 μg liposome-entrapped DNA compared to naked DNA. At all time points measured, mice immunised with naked DNA showed no greater immune response compared to the control, non-immunised group. In contrast, as early as day 49, responses were significantly higher in mice injected with DNA entrapped in DRV liposomes containing DOTAP compared to the control group and mice immunised with naked DNA. By day 56, all total IgG responses from mice immunised with both DRV formulations were significantly higher. Comparison between the DRV formulations revealed no significant difference in immune responses elicited except at day 114, where the humoural responses of the group injected with liposomal formulation containing DC-Chol dropped to significantly lower levels that those measured in mice which received the DOTAP formulation. Similar results were found when the IgG1 and IgG2a subclass responses were determined. These results suggest that, not only can DNA be effectively entrapped within liposomes using the DRV method but that such DRV liposomes containing DNA may be a useful system for subcutaneous delivery of DNA vaccines. © 2003 Taylor & Francis Ltd.

Small cationic DDA:TDB liposomes as protein vaccine adjuvants obviate the need for TLR agonists in inducing cellular and humoral responses

Most subunit vaccines require adjuvants in order to induce protective immune responses to the targeted pathogen. However, many of the potent immunogenic adjuvants display unacceptable local or systemic reactogenicity. Liposomes are spherical vesicles consisting of single (unilamellar) or multiple (multilamellar) phospholipid bi-layers. The lipid membranes are interleaved with an aqueous buffer, which can be utilised to deliver hydrophilic vaccine components, such as protein antigens or ligands for immune receptors. Liposomes, in particular cationic DDA:TDB vesicles, have been shown in animal models to induce strong humoral responses to the associated antigen without increased reactogenicity, and are currently being tested in Phase I human clinical trials. We explored several modifications of DDA:TDB liposomes--including size, antigen association and addition of TLR agonists--to assess their immunogenic capacity as vaccine adjuvants, using Ovalbumin (OVA) protein as a model protein vaccine. Following triple homologous immunisation, small unilamellar vesicles (SUVs) with no TLR agonists showed a significantly higher capacity for inducing spleen CD8 IFN? responses against OVA in comparison with the larger multilamellar vesicles (MLVs). Antigen-specific antibody reponses were also higher with SUVs. Addition of the TLR3 and TLR9 agonists significantly increased the adjuvanting capacity of MLVs and OVA-encapsulating dehydration-rehydration vesicles (DRVs), but not of SUVs. Our findings lend further support to the use of liposomes as protein vaccine adjuvants. Importantly, the ability of DDA:TDB SUVs to induce potent CD8 T cell responses without the need for adding immunostimulators would avoid the potential safety risks associated with the clinical use of TLR agonists in vaccines adjuvanted with liposomes.

Manipulation of the surface pegylation in combination with reduced vesicle size of cationic liposomal adjuvants modifies their clearance kinetics from the injection site, and the rate and type of T cell response

The mechanism behind the immunostimulatory effect of the cationic liposomal vaccine adjuvant dimethyldioctadecylammonium and trehalose 6,6′- dibehenate (DDA:TDB) has been linked to the ability of these cationic vesicles to promote a depot after administration, with the liposomal adjuvant and the antigen both being retained at the injection site. This can be attributed to their cationic nature, since reduction in vesicle size does not influence their distribution profile yet neutral or anionic liposomes have more rapid clearance rates. Therefore the aim of this study was to investigate the impact of a combination of reduced vesicle size and surface pegylation on the biodistribution and adjuvanticity of the formulations, in a bid to further manipulate the pharmacokinetic profiles of these adjuvants. From the biodistribution studies, it was found that with small unilamellar vesicles (SUVs), 10% PEGylation of the formulation could influence liposome retention at the injection site after 4 days, whilst higher levels (25 mol%) of PEG blocked the formation of a depot and promote clearance to the draining lymph nodes. Interestingly, whilst the use of 10% PEG in the small unilamellar vesicles did not block the formation of a depot at the site of injection, it did result in earlier antibody response rates and switch the type of T cell responses from a Th1 to a Th2 bias suggesting that the presence of PEG in the formulation not only control the biodistribution of the vaccine, but also results in different types of interactions with innate immune cells. © 2012 Elsevier B.V.

Liposome (Lipodine™)-mediated DNA vaccination by the oral route

Plasmid DNA pRc/CMV HBS encoding the S (small) region of hepatitis B surface antigen (HBsAg) was incorporated by the dehydration-rehydration method into Lipodine™ liposomes composed of 16 μmoles phosphatidylcholine (PC) or distearoyl phosphatidylcholine (DSPC), 8 μmoles of (dioleoyl phosphatidylethanolamine (DOPE) or cholesterol and 4 μmoles of the cationic lipid 1,2-dioleoyl-3-(trimethylammonium propane (DOTAP) (molar ratios 1:0.5:0.25). Incorporation efficiency was high (89-93% of the amount of DNA used) in all four formulations tested and incorporated DNA was shown to be resistant to displacement in the presence of the competing anionic sodium dodecyl sulphate molecules. This is consistent with the notion that most of the DNA is incorporated within the multilamellar vesicles structure rather than being vesicle surface-complexed. Stability studies performed in simulated intestinal media also demonstrated that dehydration-rehydration vesicles (DRV) incorporating DNA (DRV(DNA)) were able to retain significantly more of their DNA content compared to DNA complexed with preformed small unilamellar vesicles (SUV-DNA) of the same composition. Moreover, after 4h incubation in the media, DNA loss for DSPC DRV(DNA) was only minimal, suggesting this to be the most stable formulation. Oral (intragastric) liposome-mediated DNA immunisation studies employing a variety of DRV(DNA) formulations as well as naked DNA revealed that secreted IgA responses against the encoded HBsAg were (as early as three weeks after the first dose) substantially higher after dosing with 100 μg liposome-entrapped DNA compared to naked DNA. Throughout the fourteen week investigation, IgA responses in mice were consistently higher with the DSPC DRV(DNA) liposomes compared to naked DNA and correlated well with their improved DNA retention when exposed to model intestinal fluids. To investigate gene expression after oral (intragastric) administration, mice were given 100 μg of naked or DSPC DRV liposome-entrapped plasmid DNA expressing the enhanced green fluorescent protein (pCMV.EGFP). Expression of the gene, in terms of fluorescence intensity in the draining mesenteric lymph nodes, was much greater in mice dosed with liposomal DNA than in animals dosed with the naked DNA. These results suggest that DSPC DRV liposomes containing DNA (Lipodine™) may be a useful system for the oral delivery of DNA vaccines.

Desenvolvimento e estudo em microcanais de fluidos inovadores análogos ao sangue baseados em vesículas unilamelares gigantes

O escoamento sanguíneo é um dos temas de grande interesse para a comunidade científica. Assim, a busca de fluidos que sejam análogos ao sangue bem como o estudo do seu escoamento em microcanais, tal como acontece com o sangue nos capilares, continua a ser alvo de investigação. Numa primeira fase deste trabalho, procedeu-se ao desenvolvimento de um modelo inovador para produzir glóbulos vermelhos artificiais, constituído por Vesículas Unilamelares Gigantes, vulgarmente designadas Giant Unilamellar Vesicles (GUVs), com três concentrações diferentes. Pretende-se que estas vesículas tenham um comportamento reológico idêntico ao escoamento dos glóbulos vermelhos (GVs) em microcanais, permitindo assim proceder a vários estudos hemodinâmicos. No desenvolvimento destas vesículas, foi verificado que as mais adequadas são constituídas por uma mistura natural de lípidos e lecitina de soja. Foi realizado um estudo relativamente à sua concentração, onde se verificou que, com o aumento da quantidade da lecitina de soja nas soluções, a concentração de GUVs tende a aumentar. Foi também realizado um estudo relativo aos diâmetros dos GUVs para verificar se estes se aproximavam em termos de tamanho dos GVs, onde foi verificado que a maioria dos GUVs possuem diâmetros com dimensões entre os 5 e 7 μm, tal como os GVs. Foi ainda verificado que a solução com a menor concentração de lecitina de soja possui uma maior quantidade de GUVs com diâmetros entre os 5 e 7 μm. Na segunda fase, foi estudado experimentalmente o escoamento das três soluções de GUVs em microcanais hiperbólicos, com três caudais diferentes, com o objetivo de visualizar a Camada Livre de Células (CLC), determinar a deformação e estudar as velocidades destes. Foi verificado que existe a formação de CLC em todas as concentrações e que aumenta com o aumento do caudal. Relativamente à deformação, esta é bastante mais evidente na contração do microcanal onde a taxa deformação é máxima. Para o caso da velocidade, foi observado um aumento bastante significativo e linear da velocidade na região da contração do microcanal hiperbólico e uma velocidade baixa e aproximadamente constante a montante e jusante da contração. vi Por fim, foi também realizado o estudo reológico dos GUVs, de forma a investigar se estes têm uma viscosidade próxima do sangue. Foi verificado que os GUVs apresentam uma viscosidade inferior à do sangue total e que existe um ligeiro aumento da viscosidade dos GUVs com o aumento da sua concentração. Por último, também foi efetuada uma comparação da viscosidade da solução de GUVs com uma solução de 5% de Hematócrito (Hct) em soro fisiológico, onde foi verificado que ambas as viscosidades são muito próximas.

The vesicular monoamine transporter 2 is present in small synaptic vesicles and preferentially localizes to large dense core vesicles in rat solitary tract nuclei.

In central neurons, monamine neurotransmitters are taken up and stored within two distinct classes of regulated secretory vesicles: small synaptic vesicles and large dense core vesicles (DCVs). Biochemical and pharmacological evidence has shown that this uptake is mediated by specific vesicular monamine transporters (VMATs). Recent molecular cloning techniques have identified the vesicular monoamine transporter (VMAT2) that is expressed in brain. This transporter determines the sites of intracellular storage of monoamines and has been implicated in both the modulation of normal monoaminergic neurotransmission and the pathogenesis of related neuropsychiatric disease. We used an antiserum against VMAT2 to examine its ultrastructural distribution in rat solitary tract nuclei, a region that contains a dense and heterogeneous population of monoaminergic neurons. We find that both immunoperoxidase and immunogold labeling for VMAT2 localize to DCVs and small synaptic vesicles in axon terminals, the trans-Golgi network of neuronal perikarya, tubulovesicles of smooth endoplasmic reticulum, and potential sites of vesicular membrane recycling. In axon terminals, immunogold labeling for VMAT2 was preferentially associated with DCVs at sites distant from typical synaptic junctions. The results provide direct evidence that a single VMAT is expressed in two morphologically distinct types of regulated secretory vesicles in central monoaminergic neurons.

Cross-linking studies and membrane localization and assembly of radiolabelled large mechanosensitive ion channel (MscL) of Escherichia coli

The gene encoding the large conductance mechanosensitive ion channel (MscL) of Escherichia coli and several deletion mutants of mscL were cloned under the control of the T7 RNA polymerase promoter. Transformation of these constructs into an E. coli strain carrying an inducible T7 RNA polymerase gene allowed the specific production and labelling of MscL with [S-35]methionine. Preparation of membrane fractions of E. coli cells by sucrose gradient centrifugation indicated that the radiolabelled MscL was present in the inner cytoplasmic membrane in agreement with results of several studies. However, treatment of the labelled cells and cell membrane vesicles with various cross-linkers resulted in the majority of labelled protein migrating as a monomer with a small proportion of molecules (approximate to 25%) migrating as dimers and higher order multimers. This result is in contrast with a finding of a study suggesting that the channel exclusively forms hexamers in the cell membrane off. coli (1) and therefore may have profound implication for the activation and/or ''multimerization'' of the channel by mechanical stress exerted to the membrane. In addition, from the specific activity of the radiolabelled protein and the amount of protein in the cytoplasmic membrane fraction we estimated the number of MscL ion channels expressed under these conditions to be approximately 50 channels per single bacterium. (C) 1997 Academic Press.

Intracellular targeting of GLUT4 in transfected insulinoma cells: evidence for association with constitutively recycling vesicles distinct from synaptophysin and insulin vesicles.

In adipocytes and muscle cells, the GLUT4 glucose transporter isoform is present in intracellular vesicles which continuously recycle between an intracytoplasmic location and the plasma membrane. It is not clear whether the GLUT4-vesicles represent a specific kind of vesicle or resemble typical secretory granules or synaptic-like microvesicles. To approach this question, we expressed GLUT4 in the beta cell line RINm5F and determined its intracellular localization by subcellular fractionation and by immunofluorescence and immunoelectron microscopy. GLUT4 was not found in insulin granules but was associated with a subpopulation of smooth-surface vesicles present in the trans-Golgi region and in vesicular structures adjacent to the plasma membrane. In the trans-Golgi region, GLUT4 did not colocalize with synaptophysin or TGN38. Incubation of the cells with horseradish peroxidase (HRP) led to colocalization of HRP and GLUT4 in some endosomal structures adjacent to the plasma membrane and in occasional trans-Golgi region vesicles. When cells were incubated in the presence of Bafilomycin A, analysis by confocal microscopy revealed GLUT4 in numerous large spots present throughout the cytoplasm, many of which costained for TGN38 and synaptophysin. By immunoelectron microscopy, numerous endosomes were observed which stained strongly for GLUT4. Together our data demonstrate that ectopic expression of GLUT4 in insulinoma cells reveals the presence of a subset of vesicular structures distinct from synaptic-like vesicles and insulin secretory granules. Furthermore, they indicate that GLUT4 constitutively recycles between the plasma membrane and its intracellular location by an endocytic route also taken by TGN38 and synaptophysin.