Asymmetric distribution of phosphatidylethanolamine fatty acyl chains in the membrane of vesicular stomatitis virus

The membrane of vesicular stomatitis virus (VSV) contains two distinct pools of phosphatidylethanolamine molecules which reside in the inner and outer phospholipid monolayers, respectively. 36% of the total membrane phosphatidylethanolamine is found in the outer monolayer while 64% is found in the inner. The two pools of VSV phosphatidylethanolamine can be distinguished operationally by the fact that only outer phosphatidylethanolamine is reactive in intact virions with the membrane-impermeable reagent trinitrobenzenesulfonate (TNBS). We have made use of this property to separate inner from outer VSV phosphatidylethanolamine and to determine the fatty acyl chain compositions of the two phosphatidylethanolamine pools separately. The results show that compared to outer phosphatidylethanolamine, inner phosphatidylethanolamine molecules contain a significantly higher proportion of unsaturated fatty acyl chains. Furthermore, whereas the proportion of unsaturated fatty acyl chains was found to be quite similar at the 1 and 2 glycerol carbon atoms in inner phosphatidylethanolamine, a marked dissimilarity was observed in outer phosphatidylethanolamine; outer phosphatidylethanolamine was enriched in saturated fatty acyl chains at the 1 position and in unsaturated fatty acyl chains at the 2 position. The differential fatty acyl chain composition of inner compared to outer phosphatidylethanolamine indicates that rapid, random transmembrane migration (flip-flop) of phosphatidylethanolamine does not occur in the VSV membrane. The nature of the fatty acyl chain asymmetry observed in VSV phosphatidylethanolamine does not support the view that the     

Asymmetric distribution of phosphatidylethanolamine fatty acyl chains in the membrane of vesicular stomatitis virus.

The membrane of vesicular stomatitis virus (VSV) contains two distinct pools of phosphatidylethanolamine molecules which reside in the inner and outer phospholipid monolayers, respectively. 36% of the total membrane phosphatidylethanolamine is found in the outer monolayer while 64% is found in the inner. The two pools of VSV phosphatidylethanolamine can be distinguished operationally by the fact that only outer phosphatidylethanolamine is reactive in intact virions with the membrane-impermeable reagent trinitrobenzenesulfonate (TNBS). We have made use of this property to separate inner from outer VSV phosphatidylethanolamine and to determine the fatty acyl chain compositions of the two phosphatidylethanolamine pools separately. The results show that compared to outer phosphatidylethanolamine, inner phosphatidylethanolamine molecules contain a significantly higher proportion of unsaturated fatty acyl chains. Furthermore, whereas the proportion of unsaturated fatty acyl chains was found to be quite similar at the 1 and 2 glycerol carbon atoms in inner phosphatidylethanolamine, a marked dissimilarity was observed in outer phosphatidylethanolamine; outer phosphatidylethanolamine was enriched in saturated fatty acyl chains at the 1 position and in unsaturated fatty acyl chains at the 2 position. The differential fatty acyl chain composition of inner compared to outer phosphatidylethanolamine indicates that rapid, random transmembrane migration (flip-flop) of phosphatidylethanolamine does not occur in the VSV membrane. The nature of the fatty acyl chain asymmetry observed in VSV phosphatidylethanolamine does not support the view that the identity of the fatty acyl chains can uniquely specify or determine which side of the membrane individual phosphatidylethanolamine molecules come to occupy. Although fatty acyl chain asymmetry and phosphatidylethanolamine asymmetry are correlated in VSV, no simple rules can be discerned which uniquely relate the two paramaters.     

Lipid analysis of cells and chromatophores of Rhodopseudomonas sphaeroides

The phospholipids and fatty acid analysis of four strains of Rhodopseudomonas sphaeroides and of chromatophores from two strains show some differences and also show the presence of an unusual polar neutral lipid which is ninhydrin positive and which on acid hydrolysis yields ornithine and an unidentified amino compound. This lipid is called aminolipid-X and has a fatty acid composition very different from the phospholipids. Phosphatidylcholine (PC) and phosphatidylethanolamine (PE) contain a very small amount of plasmalogen forms as determined by combined mild alkaline hydrolysis, acetic acid hydrolysis and phospholipase A₂ hydrolysis.The reaction of intact cells and chromatophores with trinitrobenzenesulfonate (TNBS), fluorodinitrobenzene (FDNB) and isethionylacetimidate (IA) show that 78% of the total PE in chromatophores is localized on the outer membrane surface. In intact cells about 15–35% of the total PE is localized on the outer surface of the plasma membrane.     

Vesicular stomatitis virus infects and matures only through the basolateral surface of the polarized epithelial cell line,MDCK

We have used Madin-Darby canine kidney (MDCK) cells grown on nitrocellulose filters to study the polarity of virus infection and maturation. The cells form epithelia-like monolayers, which display high (>1000 Ω cm²) electrical resistance and a cuboidal morphology. Vesicular stomatitis virus (VSV) was found to infect the monolayer at least 100 times more efficiently when applied through the filter to the basolateral surface than when applied to the apical surface. The avian influenza, fowl plague virus (FPV), infected the monolayer through either the apical or basolateral surface. The polarity of virus budding was evaluated by harvesting virus from the two sides of the monolayer. More than 99% of released influenza hemagglutinin titre was found on the apical side of the filter, while more than 98% of budded VSV was found on the basal side. This polarity of budding was retained through 10 hr of viral infection, as was the polarity of surface expression of viral envelope proteins revealed by immunofluorescence. The strong preference of VSV for basolateral maturation is paralleled by an equally strong preference for infection through the basolateral membrane of this polar epithelial cell.     

Differential effect of monensin on enveloped viruses that form at distinct plasma membrane domains

We have observed a striking differential effect of the ionophore, monensin, on replication of influenza virus and vesicular stomatitis virus (VSV) in Madin-Darby canine kidney (MDCK) and baby hamster kidney (BHK21) cells. In MDCK cells, influenza virus is assembled at the apical surfaces, whereas VSV particles bud from the basolateral membranes; no such polarity of maturation is exhibited in BHK21 cells. A 10(-6) M concentration of monensin reduces VSV yields in MDCK cells by greater than 90% as compared with controls, whereas influenza virus yields are unaffected. In BHK21 cells, monensin also inhibits VSV production, but influenza virus is also sensitive to the ionophore. Immunofluorescent staining of fixed and unfixed MDCK monolayers indicates that VSV glycoproteins are synthesized in the presence of monensin, but their appearance on the plasma membrane is blocked. Electron micrographs of VSV-infected MDCK cells treated with monensin show VSV particles aggregated within dilated cytoplasmic vesicles. Monensin-treated influenza virus-infected MDCK cells also contain dilated cytoplasmic vesicles, but virus particles were not found in these structures, and numerous influenza virions were observed budding at the cell surface. These results indicate that influenza virus glycoprotein transport is not blocked by monensin treatment, whereas there is a block in transport of VSV G protein. Thus it appears that at least two distinct pathways of transport of glycoproteins to the plasma membrane exist in MDCK cells, and only one of them is blocked by monensin.     

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.     

Delayed appearance of pseudotypes between vesicular stomatitis virus influenza virus during mixed infection of MDCK cells.

In intact Madin-Darby canine kidney (MDCK) cell monolayers, vesicular stomatitis virus (VSV) matures only at basolateral membranes beneath tight junctions, whereas influenza virus buds from apical cell surfaces. Early in the growth cycle, the viral glycoproteins are restricted to the membrane domain from which each virus buds. We report here that phenotypic mixing and formation of VSV pseudotypes occurred when influenza virus-infected MDCK cells were superinfected with VSV. Up to 75% of the infectious VSV particles from such experiments were neutralized by antiserum specific for influenza virus, and a smaller proportion (up to 3%) were resistant to neutralization with antiserum specific for VSV. The latter particles, which were neutralized by antiserum to influenza A/WSN virus, are designated as VSV(WSN) pseudotypes. During mixed infections, both wild-type viruses were detected 1 to 2 h before either phenotypically mixed VSV or VSV(WSN) pseudotypes. Coincident with the appearance of cytopathic effects in the monolayer, the yield of pseudotypes rose dramatically. In contrast, in doubly infected BHK-21 cells, which do not show polarity in virus maturation sites and are not connected by tight junctions, VSV(WSN) pseudotypes were detected as soon as VSV titers rose to the minimum levels which allowed detection of pseudotypes, and the proportion observed remained relatively constant at later times. Examination of thin sections of doubly infected MDCK monolayers revealed that polarity in maturation sites was preserved for both viruses until approximately 12 h after inoculation with influenza virus, when disruption of junctional complexes was evident. Even at later periods, the majority of each virus type was associated with its normal membrane domain, suggesting that the sorting mechanisms responsible for directing the glycoproteins of VSV and influenza virus to separate surface domains continue to operate in doubly infected MDCK cells. The time course of VSV(WSN) pseudotype formation and changes in virus maturation sites are compatible with progressive mixing of viral glycoproteins at either intracellular or plasma membranes of doubly infected cells.     

Membrane recognition by vesicular stomatitis virus involves enthalpy-driven protein-lipid interactions

Vesicular stomatitis virus (VSV) infection depends on the fusion of viral and cellular membranes, which is mediated by virus spike glycoprotein G at the acidic environment of the endosomal compartment. VSV G protein does not contain a hydrophobic amino acid sequence similar to the fusion peptides found among other viral glycoproteins, suggesting that membrane recognition occurs through an alternative mechanism. Here we studied the interaction between VSV G protein and liposomes of different phospholipid composition by force spectroscopy, isothermal titration calorimetry (ITC), and fluorescence spectroscopy. Force spectroscopy experiments revealed the requirement for negatively charged phospholipids for VSV binding to membranes, suggesting that this interaction is electrostatic in nature. In addition, ITC experiments showed that VSV binding to liposomes is an enthalpically driven process. Fluorescence data also showed the lack of VSV interaction with the vesicles as well as inhibition of VSV-induced membrane fusion at high ionic strength. Intrinsic fluorescence measurements showed that the extent of G protein conformational changes depends on the presence of phosphatidylserine (PS) on the target membrane. Although the increase in PS content did not change the binding profile, the rate of the fusion reaction was remarkably increased when the PS content was increased from 25 to 75%. On the basis of these data, we suggest that G protein binding to the target membrane essentially depends on electrostatic interactions, probably between positive charges on the protein surface and negatively charged phospholipids in the cellular membrane. In addition, the fusion is exothermic, indicating no entropic constraints to this process.     

Comparative studies on cytopathic effects induced by vesicular stomatitis virus in two cell types.

Infection of mouse L cells with vesicular stomatitis virus (VSV) leads to an extensive cell fusion, while porcine kidney stable (PS) cells infected with VSV show only cell rounding. Therefore, comparative morphological studies on the infection of the two cell lines were carried out using a transmission or scanning electron microscope and an immunofluorescence microscope. PS cells infected with VSV contrasted to L cells infected with the same virus in the following two points; (1) the principal site of VSV maturation was the intracytoplasmic vacuolar membrane in PS cells and the plasma membrane in L cells. However, it was found that viral glycoprotein was present on the cell surface of infected PS cells; (2) the morphological changes at the cell surface of infected PS cells occurred much earlier and were severer than those at the cell surface of infected L cells. From these observations, we discuss the possibility that the surfaceembrane of PS cells is too sensitive to the VSV-induced cell damage to cause cell fusion.     

Structural changes in BHK cell plasma membrane caused by the binding of vesicular stomatitis virus

Spin label electron spin resonance techniques using a nitroxide derivative of stearic acid were used to detect changes in plasma membrane structure caused by the binding of vesicular stomatitis virus (VSV) to cell plasma membranes of intact BHK-21 cells. The results indicate that binding of VSV to cell surface receptors causes an increase in the observed rigidity of the plasma membrane lipid bilayer. This change in membrane structure, which appears to be caused by the cross-linking of receptors in the plane of the plasma membrane, could be prevented by treating the cells with colchicine before addition of virus and could be reversed by treating the cells with colchicine after addition of virus. Cells treated with a monovalent, water-soluble derivative of VSV G-protein (Gs) did not show an increase in plasma membrane bilayer rigidity. However, addition of anti-VSV G-protein immunoglobulin G to cells pretreated with G8 caused an increase in plasma membrane bilayer rigidity. This increased rigidity could also be reversed by the addition of colchicine. Fluorescence microscopy was used to determine the distribution of fluorescein-labeled VSV particles on the cell surface after addition of virus. Approximately 30 min after addition of virus, discrete areas on the cell surface showed fluorescent staining, which coalesced to apical regions of the cell after approximately 40 min.     

Effect of membrane protein on lipid bilayer structure: a spin-label electron spin resonance study of vesicular stomatitis virus.

Spin-label electron spin resonance (ESR) methods have been used to study the structure of the envelope of vesicular stomatitis virus (VSV). The data indicate that the lipid is organized in a bilayer structure. Proteolytic digestion of the glycoproteins which are the spike-like projections on the outer surface of the virus particle increases the fluidity of the lipid bilayer. Since the lipid composition of the virion reflects the composition of the host plasma membrane and the protein composition is determined by the viral genome, VSV was grown in both MDBK and BHK21-F cells to determine the effect of a change in lipid composition on the structure of the lipid bilayer of VSV. The lipid bilayer of the virion was found to be more rigid when derived from MDBK cells than from BHK21-F cells. Studies comparing spin-labeled intact cells and cell membrane fractions suggest that upon labeling the whole cell the spin label probes the plasma membrane. Comparison of spin-labeled VSV particles and their host cells indicates that the lipid bilayer of the plasma membrane is considerably more fluid than that of the virion. These results are discussed in terms of the effect of membrane-associated protein on the structure of the lipid bilayer.     

Fluorescence photobleaching recovery measurements reveal differences in envelopment of sindbis and vesicular stomatitis viruses

Fluorescence photobleaching recovery (FPR) measurements of virus glycoproteins on the surfaces of cells infected with vesicular stomatitis virus (VSV) and Sindbis virus showed that the VSV glycoprotein (G) remained mobile throughout the infectious cycle, whereas Sindbis virus glycoproteins (E1, E2) were partially mobile early after infection and immobile at later times when greater amounts of these proteins were on the cell surface. A highly mobile fraction of Sindbis virus glycoproteins was detected throughout the replication cycle of a temperature-sensitive mutant unable to form virus particles. Thus immobilization of E1 and E2 was the result of increasing surface glycoprotein concentrations and virus budding. Together with other data, which included the detection of E1 and E2 in particles as soon as these proteins were transported to the cell surface, the FPR results suggest that Sindbis virus assembly initiates on intracellular vesicles, where glycoproteins aggregate and bind nucleocapsids. In contrast, our FPR data on VSV support a model previously suggested by others, in which a small fraction of cell-surface G is immobilized into budding sites formed by interactions with virus matrix and nucleoproteins. FPR measurements also provide direct evidence for strong interactions between E1 and E2, as well as between E1 and PE2, the precursor form of E2.     

Differential Targeting of Vesicular Stomatitis Virus G Protein and Influenza Virus Hemagglutinin Appears During Myogenesis of L6 Muscle Cells

Exocytic organelles undergo profound reorganization during myoblast differentiation and fusion. Here, we analyzed whether glycoprotein processing and targeting changed during this process by using vesicular stomatitis virus (VSV) G protein and influenza virus hemagglutinin (HA) as models. After the induction of differentiation, the maturation and transport of the VSV G protein changed dramatically. Thus, only half of the G protein was processed and traveled through the Golgi, whereas the other half remained unprocessed. Experiments with the VSV tsO45 mutant indicated that the unprocessed form folded and trimerized normally and then exited the ER. It did not, however, travel through the Golgi since brefeldin A recalled it back to the ER. Influenza virus HA glycoprotein, on the contrary, acquired resistance to endoglycosidase H and insolubility in Triton X-100, indicating passage through the Golgi. Biochemical and morphological assays indicated that the HA appeared at the myotube surface. A major fraction of the Golgi-processed VSV G protein, however, did not appear at the myotube surface, but was found in intracellular vesicles that partially colocalized with the regulatable glucose transporter. Taken together, the results suggest that, during early myogenic differentiation, the VSV G protein was rerouted into developing, muscle-specific membrane compartments. Influenza virus HA, on the contrary, was targeted to the myotube surface.     

Announcement

Recently we described a saturable, high-affinity binding site for vesicular stomatitis virus (VSV) on the surface of Vero cells that appears to mediate viral infectivity. To isolate this binding site, we have extracted Vero cells with the detergent, octyl-β-d-glucopyranoside. The dialyzed detergent extract specifically inhibits the saturable, high-affinity binding of ³⁵S-methionine-labeled VSV to Vero cells. The inhibitory activity is resistant to protease, neuraminidase and heating to 100°C. It is soluble in chloroform-methanol and inactivated by phospholipase C, suggesting that it is a phospholipid. Of various puriifed lipids tested, only phosphatidylserine was capable of totally inhibiting the high-affinity binding of VSV. The half-maximal inhibitory concentration for phosphatidylserine was 1 μM. Phosphatidylserine also inhibited VSV plaque formation by 80%–90%; Herpes simplex virus plaque formation was unaffected. Centrifugation and electron microscopy studies have shown that phosphatidylserine-containing liposomes bind to VSV. The finding that phosphatidylserine directly binds to VSV and inhibits VSV attachment and infectivity suggests that plasma membrane phosphatidylserine could function as a binding site or portion of a binding site for VSV.     

Cerulenin blocks fatty acid acylation of glycoproteins and inhibits vesicular stomatitis and Sindbis virus particle formation

Cerulenin, an antibiotic that inhibits de novo fatty acid and cholesterol biosynthesis, effectively inhibited the formation and release of virus particles from chicken embryo fibroblasts infected with Sindbis or vesicular stomatitis virus (VSV). When added for 1 h at 3 h postinfection, the antibiotic blocked VSV particle production by 80 to 90% and inhibited incorporation of [3H]palmitic acid into the VSV glycoprotein by an equivalent amount. The effect of this antibiotic on virus protein and RNA biosynthesis was significantly less than that on fatty acid acylation. Nonacylated virus glycoproteins accumulated inside and on the surface of cerulenin-treated cells. These data indicate that fatty acid acylation is not essential for intracellular transport of these membrane proteins, but it may have an important role in the interaction of glycoproteins with membranes during virus assembly and budding.     

Role of heterologous and homologous glycoproteins in phenotypic mixing between Sendai virus and vesicular stomatitis virus.

Phenotypic mixing between Sendai virus and vesicular stomatitis virus (VSV) or the mutant VSV ts045 was studied. Conditions were optimized for double infection, as shown by immunofluorescence microscopy. Virions from double-infected cells were separated by sequential velocity and isopycnic gradient centrifugations. Two types of particles with mixed protein compositions were found. One type was VSV particles with Sendai virus spikes, i.e., phenotypically mixed particles. A second type was Sendai virus-VSV associations, which in plaque assays also behaved as phenotypically mixed particles. The ratio of VSV G protein to Sendai virus glycoproteins on the cell surface was varied, using the VSV mutant ts045 in double infections. Thus, different amounts of the VSV G protein were allowed to reach the cell surface at 32, 38, and 39 degrees C in Sendai virus-infected cells. However, a fixed number of Sendai virus spikes was always found in the ts045 virions. This represented 12 to 16% of the number of G proteins present in normal VSV. Furthermore, the yield of ts045 virions was radically reduced during double infection when the temperature was raised to block G-protein transport to the cell surface, suggesting that the Sendai virus glycoproteins were not able to compensate for G protein in budding. These results emphasize the role of the G protein in VSV assembly.     

Viral glycoproteins destined for apical or basolateral plasma membrane domains traverse the same Golgi apparatus during their intracellular transport in doubly infected Madin-Darby canine kidney cells

Madin-Darby canine kidney (MDCK) cells can sustain double infection with pairs of viruses of opposite budding polarity (simian virus 5 [SV5] and vesicular stomatitis virus [VSV] or influenza and VSV), and we observed that in such cells the envelope glycoproteins of the two viruses are synthesized simultaneously and assembled into virions at their characteristic sites. Influenza and SV5 budded exclusively from the apical plasma membrane of the cells, while VSV emerged only from the basolateral surfaces. Immunoelectron microscopic examination of doubly infected MDCK cells showed that the influenza hemagglutinin (HA) and the VSV G glycoproteins traverse the same Golgi apparatus and even the same Golgi cisternae. This indicates that the pathways of the two proteins towards the plasma membrane do not diverge before passage through the Golgi apparatus and therefore that critical sorting steps must take place during or after passage of the glycoproteins through this organelle. After its passage through the Golgi, the HA accumulated primarily at the apical membrane, where influenza virion assembly occurred. A small fraction of HA did, however, appear on the lateral surface and was incorporated into the envelope of budding VSV virions. Although predominantly found on the basolateral surface, significant amounts of G protein were observed on the apical plasma membrane well before disruption of the tight junctions was detectable. Nevertheless, assembly of VSV virions was restricted to the basolateral domain and in doubly infected cells the G protein was only infrequently incorporated into the envelope of budding influenza virions. These observations indicate that the site of VSV budding is not determined exclusively by the presence of G polypeptides. Therefore, it is likely that, at least for VSV, other cellular or viral components are responsible for the selection of the appropriate budding domain.     

Phosphatidylserine is not the cell surface receptor for vesicular stomatitis virus

The envelope protein from vesicular stomatitis virus (VSV) has become an important tool for gene transfer and gene therapy. It is widely used mainly because of its ability to mediate virus entry into all cell types tested to date. Consistent with the broad tropism of the virus, the receptor for VSV is thought to be a ubiquitous membrane lipid, phosphatidylserine (PS). However, the evidence for this hypothesis is indirect and incomplete. Here, we have examined the potential interaction of VSV and PS at the plasma membrane in more detail. Measurements of cell surface levels of PS show a wide range across cell types from different organisms. We demonstrate that there is no correlation between the cell surface PS levels and VSV infection or binding. We also demonstrate that an excess of annexin V, which binds specifically and tightly to PS, does not inhibit infection or binding by VSV. While the addition of PS to cells does allow increased virus entry, we show that this effect is not specific to the VSV envelope. We conclude that PS is not the cell surface receptor for VSV, although it may be involved in a postbinding step of virus entry.     

Saturable binding sites for vesicular stomatitis virus on the surface of Vero cells.

The binding of vesicular stomatitis virus (VSV) to Vero monkey cells was studied by using virus metabolically labeled with [35S]methionine. Under conditions where viral uptake did not occur (4 degrees C), apparent binding equilibrium was achieved within 12 h at a level representing 12% of the input virus. Two distinct forms of virus-cell interaction were found. At low concentrations of VSV, corresponding to multiplicities used for tissue culture studies, saturable binding was the major form of interaction. Saturation was complete at approximately 4,000 VSV virions per cell. At higher virus concentrations, nonsaturable binding prevailed. Trypsin treatment of Vero cells did not decrease the binding of VSV to the saturable binding sites. Internalization of VSV at 37 degrees C also displayed a saturable component which was directly comparable to that observed for binding. VSV binding to high-affinity, saturable sites on the plasma membrane may represent a receptor-mediated route of viral uptake.     

Prediction and identification of a permissive epitope insertion site in the vesicular stomatitis virus glycoprotein

We developed a rational approach to identify a site in the vesicular stomatitis virus (VSV) glycoprotein (G) that is exposed on the protein surface and tolerant of foreign epitope insertion. The foreign epitope inserted was the six-amino-acid sequence ELDKWA, a sequence in a neutralizing epitope from human immunodeficiency virus type 1. This sequence was inserted into six sites within the VSV G protein (Indiana serotype). Four sites were selected based on hydrophilicity and high sequence variability identified by sequence comparison with other vesiculovirus G proteins. The site showing the highest variability was fully tolerant of the foreign peptide insertion. G protein containing the insertion at this site folded correctly, was transported normally to the cell surface, had normal membrane fusion activity, and could reconstitute fully infectious VSV. The virus was neutralized by the human 2F5 monoclonal antibody that binds the ELDKWA epitope. Additional studies showed that this site in G protein tolerated insertion of at least 16 amino acids while retaining full infectivity. The three other insertions in somewhat less variable sequences interfered with VSV G folding and transport to the cell surface. Two additional insertions were made in a conserved sequence adjacent to a glycosylation site and near the transmembrane domain. The former blocked G-protein transport, while the latter allowed transport to the cell surface but blocked membrane fusion activity of G protein. Identification of an insertion-tolerant site in VSV G could be important in future vaccine and targeting studies, and the general principle might also be useful in other systems.