Viral membranes: model systems for studying biological membranes

Glucosylceramide in bunyavirus particles is essential for virus binding to host cells

Hexosylceramides (HexCer) are implicated in the infection process of various pathogens. However, the molecular and cellular functions of HexCer in infectious cycles are poorly understood. Investigating the enveloped virus Uukuniemi (UUKV), a bunyavirus of the Phenuiviridae family, we performed a lipidomic analysis with mass spectrometry and determined the lipidome of both infected cells and derived virions. We found that UUKV alters the processing of HexCer to glycosphingolipids (GSL) in infected cells. The infection resulted in the overexpression of glucosylceramide (GlcCer) synthase (UGCG) and the specific accumulation of GlcCer and its subsequent incorporation into viral progeny. UUKV and several pathogenic bunyaviruses relied on GlcCer in the viral envelope for binding to various host cell types. Overall, our results indicate that GlcCer is a structural determinant of virions crucial for bunyavirus infectivity. This study also highlights the importance of glycolipids on virions in facilitating interactions with host cell receptors and infectious entry of enveloped viruses.

The lipidomes of vesicular stomatitis virus, semliki forest virus, and the host plasma membrane analyzed by quantitative shotgun mass spectrometry

Although enveloped virus assembly in the host cell is a crucial step in the virus life cycle, it remains poorly understood. One issue is how viruses include lipids in their membranes during budding from infected host cells. To analyze this issue, we took advantage of the fact that baby hamster kidney cells can be infected by two different viruses, namely, vesicular stomatitis virus and Semliki Forest virus, from the Rhabdoviridae and Togaviridae families, respectively. We purified the host plasma membrane and the two different viruses after exit from the host cells and analyzed the lipid compositions of the membranes by quantitative shotgun mass spectrometry. We observed that the lipid compositions of these otherwise structurally different viruses are virtually indistinguishable, and only slight differences were detected between the viral lipid composition and that of the plasma membrane. Taken together, the facts that the lipid compositions of the two viruses are so similar and that they strongly resemble the composition of the plasma membrane suggest that these viruses exert little selection in including lipids in their envelopes.

A study of the glycoproteins of Autographa californica nuclear polyhedrosis virus (AcNPV)

Pulse labeling with tritiated mannose was used to follow the time course of Autographa californica nuclear polyhedrosis virus (AcNPV) glycoprotein synthesis in Spodoptera frugiperda IPLB-21 cells. Nine viral-induced intracellular glycoproteins were first detected from as early as 2 hr postinoculation (67K, early phase) to as late as 14 hr (36K and 19K glycoproteins, intermediate phase). Glycosylation of these proteins was observed to continue to the end of the experiment (28 hr postinoculation). Seven of these intracellular glycoproteins could also be detected in infected Trichoplusia ni TN-368 cells 24 hr postinoculation. When the glycosylation inhibitor tunicamycin was present (from 0 hr postinoculation) there was no detectable glycosylation of any of these viral-induced glycoproteins. Metabolic labeling of the nonoccluded virus budded from IPLB-21 and TN-368 with tritiated mannose or N-acetylglucosamine identified 11 structural glycoproteins, 8 of which were identical in both virus preparations. All of these structural glycoproteins were sensitive to the inhibitory action of tunicamycin. A single 42K structural glycoprotein was detected (with acetylglucosamine only) in the occluded form of AcNPV. Glycosylation of this structural protein appeared to be insensitive to tunicamycin. Lactoperoxidase-catalyzed radioiodination was used to determine which of the virus structural glycoproteins are exposed on the virion surface.

Recent advances in the chemistry of parainfluenza-1 (Sendai) virus inhibitors

Purine and pyrimidine derivatives, antioxidants, fusion inhibitors, statins, prostaglandins, antibiotic nucleosides, inhibitors of Ca(2+) homeostasis, carbohydrate derivatives, antisense polynucleotides and chimeras, are described as inhibitors of parainfluenza-1 (Sendai) viral infections.

Functional reconstitution of the integral membrane proteins of influenza virus into phospholipid liposomes

The integral membrane proteins of influenza virus, a hemagglutinin and a neuraminidase, have been incorporated into liposomes composed of either phosphatidylcholine or a mixture of phosphatidylcholine and phosphatidylethanolamine (2:1 w/w) using detergent dialysis. The virus spike glycoproteins for reconstitution were selectively solubilized by using cetyltrimethylammonium bromide to leave a "core particle", which lacked a lipid bilayer but possessed quaternary structure as observed by electron microscopy. The viral spike proteins were combined with exogenous phospholipid in excess sodium cholate followed by exhaustive dialysis for 150 h. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis showed that only the viral glycoproteins were associated with all the complexes formed. The level of sodium cholate remaining after dialysis was shown to be reduced to less than 1 molecule per 80 protein molecules. Viral proteins reconstituted into dimyristoylphosphatidylcholine liposomes were shown to have retained hemagglutination, low-pH-dependent hemolysis, and neuraminidase activities and were associated with a lipid bilayer in two types of complexes with average lipid to protein mole ratios after sucrose density gradient purification of either 590:1 or 970:1. The bilayer vesicles formed were of similar sizes and were shown by negative-stain electron microscopy to be 150-300 nm in diameter with well-defined spikes on their surface. Reconstituted liposomes of dimyristoylphosphatidylcholine were found to be unstable with respect to their trapped volume and therefore were unsuitable for fusion studies, unlike complexes formed with phosphatidylcholine or a mixture of phosphatidylcholine/phosphatidylethanolamine derived from hen eggs.(ABSTRACT TRUNCATED AT 250 WORDS)

The Thy-1 antigen exhibits rapid lateral diffusion in the plasma membrane of rodent lymphoid cells and fibroblasts

Thy-1 is a plasma membrane protein, but its primary structure lacks the typical membrane-spanning sequence. Recent studies revealed that a glycophospholipid is covalently bound to the carboxyl terminus, suggesting that the protein is integrated into the plasma membrane by this lipid moiety. Lateral diffusion of Thy-1 was measured in mouse thymocytes, lymphoma cells, and fibroblasts by the fluorescence recovery after photobleaching technique. Thy-1 was labeled with rhodamine-conjugated anti-Thy-1 monoclonal antibodies. Diffusion coefficients of 2-4 X 10(-9) cm2/sec were obtained for the antigen-antibody complex in all the cell types. About 50% of the Thy-1 was mobile. The diffusion coefficient for the mobile fraction of Thy-1 is considerably larger than the diffusion coefficients of many other plasma membrane proteins. Rather, the diffusion coefficient of Thy-1 is similar to those of lipid analogs embedded in the same membrane, providing strong support for the suggested lipid anchoring of this antigen.

Effect of the vesicular stomatitis virus matrix protein on the lateral organization of lipid bilayers containing phosphatidylglycerol: use of fluorescent phospholipid analogues

In order to investigate the mode of interaction of peripheral membrane proteins with the lipid bilayer, the basic (pI approximately 9.1) matrix (M) protein of vesicular stomatitis virus was reconstituted with small unilamellar vesicles (SUV) containing phospholipids with acidic head groups. The lateral organization of lipids in such reconstituted membranes was probed by fluorescent phospholipid analogues labeled with pyrene fatty acids. The excimer/monomer (E/M) fluorescence intensity ratios of the intrinsic pyrene phospholipid probes were measured at various temperatures in M protein reconstituted SUV composed of 50 mol % each of dipalmitoylphosphatidylcholine (DPPC) and dipalmitoylphosphatidylglycerol (DPPG). The M protein showed relatively small effects on the E/M ratio either in the gel or in the liquid-crystalline phase. However, during the gel to liquid-crystalline phase transition, the M protein induced a large increase in the E/M ratio due to phase separation of lipids into a neutral DPPC-rich phase and DPPG domains presumably bound to M protein. Similar phase separation of bilayer lipids was also observed in the M protein reconstituted with mixed lipid vesicles containing one low-melting lipid component (1-palmitoyl-2-oleoylphosphatidylcholine or 1-palmitoyl-2-oleoylphosphatidylglycerol) or a low mole percent of cholesterol. The self-quenching of 4-nitro-2,1,3-benzoxadiazole (NBD) fluorescence, as a measure of lipid clustering in the bilayer, was also studied in M protein reconstituted DPPC-DPPG vesicles containing 5 mol % NBD-phosphatidylethanolamine (NBD-PE). The quenching of NBD-PE was enhanced at least 2-fold in M protein reconstituted vesicles at temperatures within or below the phase transition.(ABSTRACT TRUNCATED AT 250 WORDS)

Monoclonal antibodies to the M protein of vesicular stomatitis virus (Indiana serotype) and to a cDNA M gene expression product

Twenty-nine independent hybridomas producing monoclonal antibodies to the matrix (M) protein of vesicular stomatitis virus (Indiana serotype) were prepared by fusion of SP2/0 myeloma cells with spleen lymphocytes obtained from BALB/c mice which had been immunized with the purified M protein. The specific reactivity of each monoclonal antibody was determined by an enzyme-linked immunosorbent assay and a competitive binding assay. Most of the antibodies were of the immunoglobulin G2a and G2b isotypes, although some were immunoglobulin M. By measuring the competitive binding of 125I-antibody, we identified four antigenic determinants in the M protein of the virus; two of these determinants, however, exhibited a large degree of overlap. Western blot analysis revealed little or no cross-reactivity of the antibodies with other viral proteins or with the M protein of the New Jersey serotype. Prolonged trypsin proteolysis removed the first 43 amino acids from the amino-terminal region of the M protein, but it retained its reactivity with monoclonal antibodies to each epitope, except for diminished reactivity with one. To aid in future mapping of these epitopes, we inserted a cDNA clone of the mRNA encoding the M protein of vesicular stomatitis virus into an inducible lac expression vector; the M protein produced in the JM103 strain of Escherichia coli under induced conditions was found to be approximately the same size as native M protein and was recognized by the monoclonal antibodies. These monoclonal antibodies and the cDNA clone should be useful for studying the role of M protein in virus maturation and the regulation of viral transcription.

Cell-mediated cytotoxicity against targets bearing Sendai virus glycoproteins in the absence of viral infection

The glycoproteins HN and F and the lipids were solubilized from Sendai virus envelopes by using the nonionic detergent beta-D-octylglucoside. When beta-D-octylglucoside was removed by dialysis, the glycoproteins and lipids reassociated to form vesicles. These vesicles displayed hemagglutinating, neuraminidase, and hemolysin activities comparable to those expressed by the intact virus. The vesicles were used as carriers to transfer the glycoproteins to the surface of P815 cells. The recipient cells were tested for the acquisition of the glycoproteins by demonstration of surface neuraminidase, hemadsorption activity, and antigens. The modified cells were used as targets for natural cell-mediated lysis and were found to be sensitive.

Inhibition of sendai virus-induced hemolysis by long chain fatty acids

A number of fatty acids were found to inhibit Sendai virus-induced hemolysis. cis-Unsaturated fatty acids such as oleate, as well as the methyl-branched isostearate, completely inhibited viral hemolysis at concentrations as low as 5-10 micrograms/ml, whereas the saturated, normal acids such as palmitate and stearate were comparably inhibitory only at 2-5 times those concentrations. trans-Unsaturated acids, as well as several other amphiphilic compounds, were either not or only weakly inhibitory. In contrast to their disparate effects on viral hemolysis, cis- and trans-unsaturated acids lysed erythrocytes in the same concentration range, which is several times higher than that at which the former compounds inhibited viral hemolysis. The mechanism of inhibition of viral hemolysis by isostearate involves the inactivation of viral hemolytic activity per se, since isostearate neither inhibited viral hemagglutination nor rendered erythrocytes significantly less susceptible to hemolysis. Furthermore, the concentration dependence of hemolysis inhibition by isostearate was biphasic, increasing sharply at the critical micelle concentration from a linear relationship below that concentration. Finally, an inhibitory concentration of isostearate was well below that at which amphiphiles dissolved membranes and did not dissolve Sendai virus, as shown by sucrose gradient centrifugation and sodium dodecyl sulfate-polyacrylamide gel electrophoresis. It was concluded that low concentrations of fatty acids--particularly cis-unsaturated or fluid-phase types--could block the fusion, as opposed to agglutination, step of viral hemolysis by perturbing hydrophobic regions of the Sendai virus membrane.

Semliki Forest virus: a probe for membrane traffic in the animal cell

The traffic among the cellular compartments is thought to be mediated by membrane vesicles, which bud from one compartment and fuse with the next. Despite the continuous exchange of membrane components among them, the organelles maintain their characteristic protein and lipid compositions such that the traffic remains selective, thus, avoiding intermixing of components. This membrane traffic recycles components from the cell surface to the interior of the cell and back to the cell surface again. The membrane traffic between the ER and the cell surface involves a major sorting problem. Little is known of how the animal cell has solved this problem in molecular terms. One experimental tool in this direction is provided by some enveloped animal viruses, which mature at the cell surface of infected cells. Such viruses include influenza virus, Semliki Forest virus (SFV), Sindbis virus, and vesicular stomatitis virus (VSV). They are extremely simple in makeup and hence are very well characterized. The purpose of this article is to illustrate the use of the enveloped viruses as tools in the study of membrane traffic in the animal cell. This is done in the context of the life cycle of the virus in the host cell. The article will be concerned mainly with Semliki Forest virus (SFV), which is the virus that has been worked upon in the chapter. SFV belongs to the alphaviruses, a genus of the togavirus family.

Membrane fusion proteins of enveloped animal viruses

In a living cell membrane-bound compartments are continuously either separated or united through fusion reactions, and literally thousands of such reactions take place every minute. The formation of membrane vesicles from pre-existing membranes, and their fusion with specific acceptor membranes, constitute a prerequisite for the transport of most impermeant molecules and macromolecules into the cell by endocytosis, out of the cell by exocytosis, and between the cellular organelles (Palade, 1975; Silverstein, 1978; Evered & Collins, 1982). Less frequent, but equally crucial, are fusion events in fertilization, cell division, polykaryon formation, enucleation, etc. (for reviews see Poste & Nicholson, 1978). Although a great deal is known about the properties and consequences of individual forms of membrane fusion in cellular systems, and about fusion in artificial lipid membranes, the molecular basis for the reactions remain largely unclear.

Lipid and protein contributions to the membrane surface potential of vesicular stomatitis virus probed by a fluorescent pH indicator, 4-heptadecyl-7-hydroxycoumarin

The surface potential of membranes of vesicular stomatitis virus and liposomes was determined by shift of ionization over a wide pH range of the membrane-inserted fluorophore, 4-heptadecyl-7-hydroxycoumarin. Incorporation into sonicated vesicles of negatively charged phosphatidylserine markedly increased the surface potential of uncharged phosphatidylcholine, but no significant effect on surface potential was produced by polar but uncharged glucocerebroside incorporated in phosphatidylcholine vesicles. The membrane of vesicular stomatitis virus was found to have a moderately high surface potential. Contributing to this viral membrane surface potential were glycoprotein spikes and phospholipid headgroups as determined by lowered charge after treatment of intact virions with thermolysin to remove glycoprotein or phospholipase C to remove phospholipid headgroups. The role of viral glycoprotein was confirmed by demonstrating increased surface charge of vesicles reconstituted with both viral glycoprotein and lipids compared with vesicles reconstituted with viral lipids alone. An unexpected finding was the large contribution to surface potential of cholesterol present in viral membrane. Increasing cholesterol concentration in virions by interaction with cholesterol-complexed serum lipoproteins resulted in a marked decrease in surface potential, whereas 75% depletion of virion cholesterol by interaction with sphingomyelin-complexed serum lipoproteins resulted in a significant increase in virion membrane surface potential. Although removal of glycoprotein spikes or depletion of cholesterol causes reduction in infectivity of vesicular stomatitis virus, no direct correlation could be found between alteration in surface charge and infectivity.

Sendai virus-mediated lysis of liposomes requires cholesterol

Vesicles were constituted with glycophorin, the Sendai virus receptor of human erythrocytes, and loaded with calcein, a polar derivative of fluorescein, at self-quenching concentrations. On exposure to Sendai virus and mild hypo-osmotic stress, vesicles of the appropriate composition released a significant portion of their internal contents, as indicated by an increase in calcein fluorescence. Susceptible liposomes were not induced to leak by heat-inactivated virus or by trypsin-treated virus. The response of the vesicles to virus attachment is thus analogous to virus-induced hemolysis and presumably involves fusion of the vesicle and virus membranes. In addition to glycophorin and phosphatidylcholine, cholesterol was absolutely required for the lytic response to the virus. The need for cholesterol was not attributable to inactivation of the virus by liposomes without cholesterol. The presence of gangliosides increased the encapsulated volume of the liposomes, but gangliosides did not effectively substitute for glycophorin. Thin-layer chromatography of lipid extracted from incubated virus and liposomes containing a small amount of a fluorescent phosphatidylcholine indicated that phosphatidylcholine in the vesicle is not chemically altered by functional interaction with the virus.

Interactions of chlorinated hydrocarbon insecticides with membranes

Chlorinated hydrocarbon insecticides quench the fluorescence of N-alkyl derivatives of carbazole. We used phospholipids with covalently attached carbazole as probes for the interactions of chlorinated hydrocarbon insecticides with lipid bilayers, the object being to understand better the toxicities of chlorinated hydro-carbons. Fluorescence quenching measurements revealed the lipid-water partition coefficients of the chlorinated hydrocarbons, their diffusion coefficients in the membranes, and the binding capacities of the membranes for the chlorinated hydrocarbons. Active insecticides were compared with inactive analogues to test whether activities correlated with chlorinated hydrocarbon-membrane interactions. Thus DDT and methoxychlor were compared with inactive DDE, and insecticidal gamma-lindane was compared with three less active stereoisomers. The partition coefficients, diffusion coefficients and membrane saturation capacities did not correlate with insecticidal potency. The partition coefficients of these chlorinated hydrocarbons were larger in bilayers containing unsaturated fatty acyl chains as compared to bilayers containing saturated fatty acyl chains. Interestingly, neural membranes are known to contain a large percentage of unsaturated lipids. Our results indicate that the activities of chlorinated hydrocarbons are not a result of specific interactions of these compounds with the lipids of membranes. However, the neurotoxicity of chlorinated hydrocarbons may be amplified by selective partitioning in the unsaturated neural membranes.

Viruses budding from either the apical or the basolateral plasma membrane domain of MDCK cells have unique phospholipid compositions

Influenza virus and vesicular stomatitis virus (VSV) obtain their lipid envelope by budding through the plasma membrane of infected cells. When monolayers of Madin-Darby canine kidney (MDCK) cells, a polarized epithelial cell line, are infected with fowl plague virus (FPV), an avian influenza virus, or with VSV, new FPV buds through the apical plasma membrane whereas VSV progeny is formed by budding through the basolateral plasma membrane. FPV and VSV were isolated from MDCK host cells prelabeled with [32P]orthophosphate and their phospholipid compositions were compared. Infection was carried out at 31 degrees C to delay cytopathic effects of the virus infection, which lead to depolarization of the cell surface. 32P-labeled FPV was isolated from the culture medium, whereas 32P-labeled VSV was released from below the cell monolayer by scraping the cells from the culture dish 8 h after infection. At this time little VSV was found in the culture medium, indicating that the cells were still polarized. The phospholipid composition of the two viruses was distinctly different. FPV was enriched in phosphatidylethanolamine and phosphatidylserine and VSV in phosphatidylcholine, sphingomyelin, and phosphatidylinositol. When MDCK cells were trypsinized after infection and replated, non-infected control cells attached to reform a confluent monolayer within 4 h, whereas infected cells remained in suspension. FPV and VSV could be isolated from the cells in suspension and under these conditions the phospholipid composition of the two viruses was very similar. We conclude that the two viruses obtain their lipids from the plasma membrane in the same way and that the different phospholipid compositions of the viruses from polarized cells reflect differences in the phospholipid composition of the two plasma membrane domains.

Transcription of vesicular stomatitis virus activated by pardaxin, a fish toxin that permeabilizes the virion membrane

The toxic protein, Pardaxin, of the Red Sea flatfish Pardachirus marmoratus readily induced transcription of vesicular stomatitis virus by making the virion membrane permeable to nucleoside triphosphates in the absence of nonionic detergents. Virion transcription was activated over a wide range of Pardaxin concentrations, but at optimal concentrations, the rate of transcription exceeded that induced by Triton X-100. The inhibitory effect of M protein was manifested for both Pardaxin-induced and Triton-induced transcription at high concentrations of vesicular stomatitis virions; however, unlike the Triton-induced reaction, the inhibitory effect of M protein was not reversed by polyglutamic acid added to the Pardaxin-induced transcription reaction. We propose that activation of virion transcription by Pardaxin resembles more closely intracellular transcription initiated by virion penetration than does detergent-activated transcription of vesicular stomatitis virus.

Role of matrix protein in assembling the membrane of vesicular stomatitis virus: reconstitution of matrix protein with negatively charged phospholipid vesicles

The matrix (M) protein of vesicular stomatitis virus (VSV) was reconstituted into phospholipid vesicles by detergent dialysis. Reconstitution of the positively charged M protein occurred only in the presence of negatively charged phospholipids such as phosphatidylserine, phosphatidic acid, or phosphatidylinositol. Preformed vesicles containing negatively charged phospholipids also bound free M protein. Derivatization of the positively charged lysines in M protein with acetic anhydride or succinic anhydride prevented M protein reconstitution but did not affect the biological property of M protein to inhibit in vitro VSV transcription. An additional indication of the electrostatic nature of the M protein binding to the vesicles was that M protein could not be reconstituted in the presence of 0.5 M NaCl. Nonelectrostatic forces also appear to be involved in the association of the M protein with vesicles, since previously reconstituted M protein remained associated with the vesicles upon subsequent exposure to 0.5 M NaCl.

Localization of membrane-associated proteins in vesicular stomatitis virus by use of hydrophobic membrane probes and cross-linking reagents

The location of membrane-associated proteins of vesicular stomatitis virus was investigated by using two monofunctional and three bifunctional probes that differ in the degree to which they partition into membranes and in their specific group reactivity. Two hydrophobic aryl azide probes, [(125)I]5-iodonaphthyl-1-azide and [(3)H]pyrenesulfonylazide, readily partitioned into virion membrane and, when activated to nitrenes by UV irradiation, formed stable covalent adducts to membrane constituents. Both of these monofunctional probes labeled the glyco-protein G and matrix M proteins, but [(125)I]5-iodonaphthyl-1-azide also labeled the nucleocapsid N protein and an unidentified low-molecular-weight component. Protein labeling of intact virions was unaffected by the presence of cytochrome c or glutathione, but disruption of membrane by sodium dodecyl sulfate greatly enhanced the labeling of all viral proteins except G. Labeling of G protein was essentially restricted to the membrane-embedded, thermolysin-resistant tail fragment. Three bifunctional reagents, tartryl diazide, dimethylsuberimidate, and 4,4'-dithiobisphenylazide, were tested for their capacity to cross-link proteins to membrane phospholipids of virions grown in the presence of [(3)H]palmitate. Only G and M proteins of intact virions were labeled with (3)H-phospholipid by these cross-linkers; the reactions were not affected by cytochrome c but were abolished by disruption of virus with sodium dodecyl sulfate. Dimethylsuberimidate, which reacts with free amino groups, cross-linked (3)H-phospholipid to both G and M protein. In contrast, the hydrophilic tartryl diazide cross-linked phospholipid primarily to the M protein, whereas the hydrophobic 4,4'-dithiobisphenylazide cross-linked phospholipid primarily to the intrinsic G protein. These data support the hypothesis that the G protein traverses the virion membrane and that the M protein is membrane associated but does not penetrate very deeply, if at all.

Thermotropic behavior of dipalmitoylphosphatidylcholine vesicles reconstituted with the glycoprotein of vesicular stomatitis virus

The vesicular stomatitis virus glycoprotein reconstituted into dipalmitoylphosphatidylcholine (DPPC) vesicles exerts a profound effect upon the DPPC gel to liquid-crystalline phase transition. The glycoprotein was reconstituted into DPPC vesicles by octyl glucoside dialysis. The gel to liquid-crystalline phase transition of these vesicles was monitored by differential scanning calorimetry. Vesicles formed in the absence of glycoprotein (600--2100-A diameter) underwent the phase transition at 41.0 degrees C and had an associated enthalpy change of 8.0 +/- 1.6 kcal/mol. Increasing the mole ratio of glycoprotein to DPPC in the vesicles to 0.15 mol % reduced both the transition temperature and the transition enthalpy change. The enthalpy change as a function of the mole percent glycoprotein could be fit to a straight line by a least-squares procedure. Extrapolation of the results to the glycoprotein concentration where the enthalpy change was zero indicated one glycoprotein molecule bound 270 +/- 150 molecules of DPPC.