Pathway of vesicular stomatitis virus entry leading to infection

The entry of vesicular stomatitis virus into Madin-Darby canine kidney (MDCK) cells was examined both biochemically and morphologically. At low multiplicity and 0 °C, viruses bound to the cell surface but were not internalized. Binding was very dependent on pH. More than ten times more virus bound at pH 6.5 than at higher pH values. At the optimal pH, binding failed to reach equilibrium after more than two hours. The proportion of virus bound was irreproducible and low, relative to the binding of other enveloped viruses. Over 90% of the bound viruses were removed by proteases. When cells with pre-bound virus were warmed to 37 °C, a proportion of the bound virus became protease-resistant with a half-time of about 30 minutes. After a brief lag period, degraded viral material was released into the medium. The protease-resistant virus was capable of infecting the cells and probably did so by an intracellular route, since ammonium chloride blocked the infection and slightly reduced the degradation of viral protein.When the entry process was observed by electron microscopy, viruses were seen bound to the cell surface at 0 °C and, after warming at 37 °C, within coated pits, coated vesicles and larger, smooth-surfaced vesicles. No fusion of the virus with the plasma membrane was observed at pH 7.4.When pre-bound virus was incubated at a pH below 6 for 30 seconds at 37 °C, about 40 to 50% of the pre-bound virus became protease-resistant. On the basis of this result and previously published experiments (White et al., 1981), it was concluded that vesicular stomatitis virus fuses to the MDCK cell plasma membrane at low pH.These experiments suggest that vesicular stomatitis virus enters MDCK cells by endocytosis in coated pits and coated vesicles, and is transported to the lysosome where the low pH triggers a fusion reaction ultimately leading to the transfer of the genome into the cytoplasm. The entry pathway of vesicular stomatitis virus thus resembles that described earlier for both Semliki Forest virus and fowl plague virus.     

An efficient method for introducing defined lipids into the plasma membrane of mammalian cells

An efficient method has been devised to introduce lipid molecules into the plasma membrane of mammalian cells. This method has been applied to fuse lipid vesicles with the apical plasma membrane of Madin-Darby canine kidney cells. The cells were infected with fowl plague or influenza N virus. 4 h after infection, the hemagglutinin (HA) spike glycoprotein of the virus was present in the apical plasma membrane of the cells. Lipid vesicles containing egg phosphatidylcholine, cholesterol, and an HA receptor (ganglioside) were then bound to the cells at 0 degrees C. More than 85% of the vesicles were released by external neuraminidase at 0 degrees C or by simply warming the cells to 37 degrees C for 10 s, probably because of the action of the viral neuraminidase at the cell surface. However, when the cells were warmed to 37 degrees C in a pH 5.3 medium for 30 s, 50% of the bound vesicles could no longer be released by external neuraminidase. This only occurred when the HA protein had been cleaved into its HA1 and HA2 subunits. When we used influenza N virus, whose HA is not cleaved in Madin-Darby canine kidney cells, cleavage with external trypsin was required. The fact that the HA protein has fusogenic properties at low pH only in its cleaved form suggests that fusion of the vesicles with the plasma membrane had taken place. Further confirmation for fusion was obtained using an assay based on the decrease of energy transfer between two fluorescent phospholipids in a vesicle upon fusion of the vesicle with the plasma membrane (Struck, D. K., D. Hoekstra, and R. E. Pagano. 1981. Biochemistry, 20:4093-4099).     

Release of putative exocytic transport vesicles from perforated MDCK cells.

Mechanically perforated MDCK cells were used to study membrane transport between the trans-Golgi network and the apical and basolateral plasma membrane domains in vitro. Three membrane transport markers--an apical protein (fowl plague virus haemagglutinin), a basolateral protein (vesicular stomatitis virus G protein), and a lipid marker destined for both domains (C6-NBD-sphingomyelin)--were each accumulated in the trans-Golgi by a 20 degrees C block of transport and their behaviour monitored following cell perforation and incubation at 37 degrees C. In the presence of ATP and in the absence of calcium ions a considerable fraction of the transport markers were released from the perforated cells in sealed membrane vesicles. Control experiments showed that the vesicles were not generated by non-specific vesiculation of the Golgi complex or the plasma membrane. The vesicles had well defined sedimentation properties and the orientation expected of transport vesicles derived from the trans-Golgi network.     

Infectious Cell Entry Mechanism of Influenza Virus

Interaction between influenza virus WSN strain and MDCK cells was studied by using spin-labeled phospholipids and electron microscopy. Envelope fusion was negligibly small at neutral pH but greatly activated in acidic media in a narrow pH range around 5.0. The half-time was less than 1 min at 37°C at pH 5.0. Virus binding was almost independent of the pH. Endocytosis occurred with a half-time of about 7 min at 37°C at neutral pH, and about 50% of the initially bound virus was internalized after 1 h. Electron micrographs showed binding of virus particles in coated pits in the microvillous surface of plasma membrane and endocytosis into coated vesicles. Chloroquine inhibited virus replication. The inhibition occurred when the drug was added not later than 10 min after inoculation. Chloroquine caused an increase in the lysosomal pH 4.9 to 6.1. The drug did not affect virus binding, endocytosis, or envelope fusion at pH 5.0. Electron micrographs showed many virus particles remaining trapped inside vacuoles even after 30 min at 37°C in the presence of drug, in contrast to only a few particles after 10 min in vacuoles and secondary lysosomes in its absence. Virus replication in an artificial condition, i.e., brief exposure of the inoculum to acidic medium followed by incubation in neutral pH in the presence of chloroquine, was also observed. These results are discussed to provide a strong support for the infection mechanism of influenza virus proposed previously: virus uptake by endocytosis, fusion of the endocytosed vesicles with lysosome, and fusion of the virus envelope with the surrounding vesicle membrane in the secondary lysosome because of the low pH. This allows the viral genome to enter the target cell cytoplasm.     

Attenuation of pathogenicity of fowl plague virus by recombination with other influenza A viruses nonpathogenic for fowl: nonexculsive dependence of pathogenicity on hemagglutinin and neuraminidase of the virus.

A number of antigenic hybrids of influenza A viruses were produced possessing either the hemagglutinin or the neuraminidase of fowl plague virus and the corresponding antigen derived from another influenza A virus. Other recombinants were obtained carrying both surface antigens of fowl plague virus but differing from the parent in certain biological properties. None of the recombinants isolated were pathogenic for adult chickens. Most recombinants obtained after crosses between reciprocal recombinants carrying both fowl plague virus surface antigens were also apathogenic for chickens. Using the same parent recombinants for double infection some of the progeny "back-recombinants" were pathogenic, whereas others were not. From these results it is concluded that the surface components do not by themselves determine the pathogenicity of influenza A viruses.     

Infectious entry of murine retroviruses into mouse cells: evidence of a postadsorption step inhibited by acidic pH.

The entry into cells by many enveloped RNA viruses is accomplished by endocytosis and subsequent penetration of the endosomal membrane by an acidic pH-dependent fusion event. In the current study, we examined early events in the infectious entry of mouse retroviruses, using as a framework the observation that infection of a mouse tail skin cell line by the ecotropic virus Friend murine leukemia virus was inhibited at mildly acidic pH (pH 6). This inhibition operated on a postadsorption step, since binding of virus was unaffected at this pH. The rate of penetration of preadsorbed virus, which displayed first-order kinetics, was markedly affected by changes in the pH of the medium. The half-time for disappearance of infectious cell surface virus at 37 degrees C was approximately 10 min at pH 7.6. At pH 6.0, however, greater than 98% of the adsorbed infectivity remained at the cell surface after 45 min. This cell surface virus, though not infecting the cell at pH 6.0, retained its capacity to enter and infect the cell when the pH of the medium was raised. Acidic pH had little effect on the rate of fluid uptake by the cells, as measured by internalization of [3H]sucrose, indicating that global inhibition of endocytosis had not occurred. In contrast, cell fusion induced by Friend murine leukemia virus was optimal at pH 7.6 but markedly inhibited at a pH of less than 6.4. This inhibitory effect of acidic pH on membrane fusion is unique among the enveloped viruses which have been studied and would preclude entry of Friend murine leukemia virus from within acidified endocytic vesicles. Entry of other members of the ecotropic, mink cell focus-forming, and xenotropic host range groups displayed similar pH sensitivity. However, one xenotropic virus was relatively resistant to the effect of acidic pH, suggesting that differences might exist in the requirements for entry of different retroviruses.     

Interactions of Semliki Forest virus spike glycoprotein rosettes and vesicles with cultured cells

Semliki Forest virus (SFV)-derived spike glycoprotein rosettes (soluble octameric complexes), virosomes (lipid vesicles with viral spike glycoproteins), and liposomes (protein-free lipid vesicles) have been used to investigate the interaction of subviral particles with BHK-21 cells. Cell surface binding, internalization, degradation, and low pH- dependent membrane fusion were quantitatively determined. Electron microscopy was used to visualize the interactions. Virosomes and rosettes, but not liposomes, bound to cells. Binding occurred preferentially to microvilli and was inhibited by added SFV; it increased with decreasing pH but was, in all cases, less efficient than intact virus. At 37 degrees C the cell surface-bound rosettes and virosomes were internalized via coated pits and coated vesicles. After a lag period of 45 min the protein components of the internalized ligands were degraded and appeared, as acid-soluble activity, in the medium. The uptake of rosettes and virosomes was found to be similar to the adsorptive endocytosis of SFV except that their average residence times on the cell surface were longer. The rosettes and the liposomes did not show low pH-induced membrane fusion activity. The virosomes, however, irrespective of the lipid compositions used, displayed hemolytic activity at mildly acidic pH and were able to fuse with the plasma membrane of cells with an efficiency of 0.25 that observed with intact viruses. Cell-cell fusion activity was not observed with any of the subviral components. The results indicated that subviral components possess some of the entry properties of the intact virus.     

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.     

Fusion and infection of influenza and Sendai viruses as modulated by dextran sulfate: a comparative study

We have directly compared the effect of two types of dextran sulfate with distinct molecular weights (500 kDa and 5 kDa) on the fusion activity and infectivity of both Sendai and influenza viruses, two lipid-enveloped viruses that differ in their routes of entry into target cells. To correlate membrane merging and infectivity MDCK cells were used as targets for the viruses in both approaches. In either case pronounced inhibition of virus–cell interactions by dextran sulfate was only observed at low pH, even though Sendai virus fuses maximally at pH 7.4. Although membrane merging could not be fully abolished, the inhibitory effect was always greater when the higher molecular weight dextran sulfate was used. The presence of this residual fusion activity, that could not be reduced even with high concentrations of agent, suggests that a limited number of binding sites for dextran sulfate may exist on the viral envelopes. The compounds also inhibited fusion of bound virions, and all results could be reproduced using erythrocyte ghosts as target membranes in the fusion assay, instead of MDCK cells. In agreement with these observations only the infectivity of influenza virus (which requires a low pH-dependent step to enter target cells) was affected by dextran sulfate, again the higher molecular weight compound showing a more pronounced inhibitory effect.     

Influenza virus hemagglutinin with multibasic cleavage site is activated by furin, a subtilisin-like endoprotease.

Many viruses have membrane glycoproteins that are activated at cleavage sites containing multiple arginine and lysine residues by cellular proteases so far not identified. The proteases responsible for cleavage of the hemagglutinin of fowl plague virus, a prototype of these glycoproteins, has now been isolated from Madin-Darby bovine kidney cells. The enzyme has a mol. wt of 85,000, a pH optimum ranging from 6.5 to 7.5, is calcium dependent and recognizes the consensus sequence R-X-K/R-R at the cleavage site of the hemagglutinin. Using a specific antiserum it has been identified as furin, a subtilisin-like eukaryotic protease. The fowl plague virus hemagglutinin was also cleaved after coexpression with human furin from cDNA by vaccinia virus vectors. Peptidyl chloroalkylketones containing the R-X-K/R-R motif specifically bind to the catalytic site of furin and are therefore potent inhibitors of hemagglutinin cleavage and fusion activity.     

Conformational change of the coronavirus peplomer glycoprotein at pH 8.0 and 37 degrees C correlates with virus aggregation and virus-induced cell fusion.

We have obtained biochemical and electron microscopic evidence of conformational changes at pH 8.0 and 37 degrees C in the coronavirus spike glycoprotein E2 (S). The importance of these changes is reflected in the loss of virus infectivity, the aggregation of virions, and increased virus-induced cell fusion at the same pH. Coronavirus (MHV-A59) infectivity is exquisitely sensitive to pH. The virus was quite stable at pH 6.0 and 37 degrees C (half-life, approximately 24 h) but was rapidly and irreversibly inactivated by brief treatment at pH 8.0 and 37 degrees C (half-life, approximately 30 min). Virions treated at pH 8.0 and 37 degrees C formed clumps and large aggregates. With virions treated at pH 8.0 and 37 degrees C, the amino-terminal peptide E2N (or S1) was released from virions and the remaining peptide, E2C (S2), was aggregated. Viral spikes isolated from detergent-treated virions also aggregated at pH 8.0 and 37 degrees C. Loss of virus infectivity and E2 (S) aggregation at pH 8.0 and 37 degrees C were markedly enhanced in the presence of dithiothreitol. On the basis of the effects of dithiothreitol on the reactions of the peplomer, we propose that release of E2N (S1) and aggregation of E2C (S2) may be triggered by rearrangement of intramolecular disulfide bonds. The aggregation of virions and the isolated E2 (S) glycoprotein at pH 8.0 and 37 degrees C or following treatment with guanidine and urea at pH 6.0 and 37 degrees C indicate that an irreversible conformational change has been induced in the peplomer glycoprotein by these conditions. It is interesting that coronavirus-induced cell fusion also occurred under mildly alkaline conditions and at 37 degrees C. Some enveloped viruses, including influenza viruses and alphaviruses, show conformational changes of spike glycoproteins at a low pH, which correlates with fusion and penetration of those viruses in acidified endocytic vesicles. For coronavirus MHV-A59, comparable conformational change of the spike glycoprotein E2 (S) and cell fusion occurred at a mildly alkaline condition, suggesting that coronavirus infection-penetration, like that of paramyxoviruses and lentiviruses, may occur at the plasma membrane, rather than within endocytic vesicles.     

Rescue of vector-expressed fowl plague virus hemagglutinin in biologically active form by acidotropic agents and coexpressed M2 protein.

The hemagglutinin of the Rostock strain of fowl plague virus was expressed in CV-1 cells by a simian virus 40 vector, and its stability in the exocytotic transport process was examined by a fusion assay. A 50-fold increase in the fusion activity of the hemagglutinin was observed when expression occurred in the presence of ammonium chloride, Tris-HCl, or high doses of amantadine. When chloroquine, another acidotropic agent, was used, the hemagglutinin exposed at the cell surface had to be activated by trypsin, because intracellular cleavage was inhibited by this compound. Hemagglutinin mutants resistant to intracellular cleavage did not require acidotropic agents for full expression of fusion activity, when treated with trypsin after arrival at the cell surface. These results indicate that fowl plague virus hemagglutinin expressed by a simian virus 40 vector is denatured in the acidic milieu of the exocytotic pathway and that cleavage is a major factor responsible for the pH instability. Coexpression with the M2 protein also markedly enhanced the fusion activity of the hemagglutinin, and this effect was inhibited by low doses of amantadine. These results support the concept that M2, known to have ion channel function, protects the hemagglutinin from denaturation by raising the pH in the exocytotic transport system. The data also stress the importance of acidotropic agents or coexpressed M2 for the structural and functional integrity of vector-expressed hemagglutinin.     

Two distinct low-pH steps promote entry of vaccinia virus

Entry of vaccinia virus into cells occurs by an endosomal route as well as through the plasma membrane. Evidence for an endosomal pathway was based on findings that treatment at a pH of <6 of mature virions attached to the plasma membrane enhances entry, whereas inhibitors of endosomal acidification reduce entry. Inactivation of infectivity by low-pH treatment of virions prior to membrane attachment is characteristic of many viruses that use the endosomal route. Nevertheless, we show here that the exposure of unattached vaccinia virus virions to low pH at 37 degrees C did not alter their infectivity. Instead, such treatment stably activated virions as indicated by their accelerated entry upon subsequent addition to cells, as measured by reporter gene expression. Moreover, the rate of entry was not further enhanced by a second low-pH treatment following adsorption to the plasma membrane. However, the entry of virions activated prior to adsorption remained sensitive to inhibitors of endosomal acidification, whereas virions treated with low pH after adsorption were resistant. Activation of virions by low pH was closely mimicked by proteinase digestion, suggesting that the two treatments operate through a related mechanism. Although proteinase cleavage of the virion surface proteins D8 and A27 correlated with activation, mutant viruses constructed by individually deleting these genes did not exhibit an activated phenotype. We propose a two-step model of vaccinia virus entry through endosomes, in which activating or unmasking the fusion complex by low pH or by proteinase is rate limiting but does not eliminate a second low-pH step mediating membrane fusion.     

Interdependence of hemagglutinin glycosylation and neuraminidase as regulators of influenza virus growth: a study by reverse genetics

The hemagglutinin (HA) of fowl plague virus A/FPV/Rostock/34 (H7N1) carries two N-linked oligosaccharides attached to Asn123 and Asn149 in close vicinity to the receptor-binding pocket. In previous studies in which HA mutants lacking either one (mutants G1 and G2) or both (mutant G1,2) glycosylation sites had been expressed from a simian virus 40 vector, we showed that these glycans regulate receptor binding affinity (M. Ohuchi, R. Ohuchi, A. Feldmann, and H. D. Klenk, J. Virol. 71:8377-8384, 1997). We have now investigated the effect of these mutations on virus growth using recombinant viruses generated by an RNA polymerase I-based reverse genetics system. Two reassortants of influenza virus strain A/WSN/33 were used as helper viruses to obtain two series of HA mutant viruses differing only in the neuraminidase (NA). Studies using N1 NA viruses revealed that loss of the oligosaccharide from Asn149 (mutant G2) or loss of both oligosaccharides (mutant G1,2) has a pronounced effect on virus growth in MDCK cells. Growth of virus lacking both oligosaccharides from infected cells was retarded, and virus yields in the medium were decreased about 20-fold. Likewise, there was a reduction in plaque size that was distinct with G1,2 and less pronounced with G2. These effects could be attributed to a highly impaired release of mutant progeny viruses from host cells. In contrast, with recombinant viruses containing N2 NA, these restrictions were much less apparent. N1 recombinants showed lower neuraminidase activity than N2 recombinants, indicating that N2 NA is able to partly overrule the high-affinity binding of mutant HA to the receptor. These results demonstrate that N-glycans flanking the receptor-binding site of the HA molecule are potent regulators of influenza virus growth, with the glycan at Asn149 being dominant and that at Asn123 being less effective. In addition, we show here that HA and NA activities need to be highly balanced in order to allow productive influenza virus infection.     

Autocatalytic activation of influenza hemagglutinin

Enveloped viruses contain surface proteins that mediate fusion between the viral and target cell membranes following an activating stimulus. Acidic pH induces the influenza virus fusion protein hemagglutinin (HA) via irreversible refolding of a trimeric conformational state leading to exposure of hydrophobic fusion peptides on each trimer subunit. Herein, we show that cells expressing fowl plague virus HA demonstrate discrete switching behavior with respect to the HA conformational change. Partially activated states do not exist at the scale of the cell, activation of HA leads to aggregation of cell surface trimers, and newly synthesized HA refold spontaneously in the presence of previously activated HA. These observations imply a feedback mechanism involving self-catalyzed refolding of HA and thus suggest a mechanism similar to the autocatalytic refolding and aggregation of prions.     

Interplay between lipids and viral glycoproteins during hemolysis and fusion by influenza virus

Since mixtures of lipids alone are known to elicit membrane fusion without participation of fusion proteins, the role of viral lipids in the so-called virus-induced hemolysis and cell fusion has been investigated, using as a model the fowl plague virus (influenza A/FPV/Rostock/H7N1). The experiments were planned in a way that allowed quantitative modification of viral lipids without changing envelope glycoproteins. Under the conditions employed, cholesterol oxidase of Nocardia erythropolis and phospholipase C of Bacillus cereus were shown to completely modify their substrates in the virus without altering virus-associated hemagglutinating and neuraminidase activities. It was found with such enzyme treatment that virus-induced hemolysis and cell fusion are greatly influenced by cholesterol and phospholipids of the envelope. It became clear, that hemolysis and fusion are differently dependent on the nature of lipid components even though mediated by the same viral glycoproteins.     

In situ 125I-labelling of endosome proteins with lactoperoxidase conjugates.

Asialoorosomucoid was conjugated to lactoperoxidase and bound specifically to the asialoglycoprotein receptor on the human cell line Hep G2 at 4 degrees C. The bound conjugates incorporated 125I into cell surface proteins in the presence of H2O2. When Hep G2 cells were allowed to endocytose the prebound conjugates by warming to 37 degrees C for 10 min or were incubated for 1 h at 23 degrees C in the presence of conjugate, addition of 125I and H2O2 at 4 degrees C now resulted in labelling of endocytic vesicle proteins. The cell surface labelling pattern and the endosome labelling pattern were compared and found to be distinct. A major component labelled by the endocytosed asialoorosomucoid conjugate is shown to be the transferrin receptor. This protein and a component of 230 000 daltons are enriched in the endosome relative to the cell surface. The endocytosed lactoperoxidase conjugate was also visualised at the morphological level. Characteristic endosome tubules and vesicles contained electron-dense peroxidase reaction product as did cell surface coated pits. Selective capture of some cell surface proteins but not others by coated pits presumably gives rise to the distinct polypeptide composition of the endosome.     

Membrane fusion activity of influenza virus.

A simple assay is described to monitor fusion between fowl plague virus (FPV, an avian influenza A virus) and liposomes which allows the simultaneous quantitation of both lytic and non-lytic fusion events. As in fusion between viruses and the plasma membrane and in FPV-induced cell-cell fusion, the reaction only occurs at pH 5.5 or below, and it is fast, highly efficient, and essentially non-lytic when fresh virus and liposomes are used. The fusion occurs over a broad temperature range, and has no requirement for divalent cations. The fusion factor of influenza virus is a hemagglutinin (HA) spike which protrudes from the virus membrane and which is also responsible for virus binding to the host cell. The finding that fusion occurs as efficiently with liposomes containing or lacking virus receptor structures, further emphasizes the remarkable division of labor in the HA molecule: the receptor-binding sites are located in the globular HA1 domains and the fusion activation peptide is found at the N-terminal of HA2 in the stem region of the protein. The mechanism of fusion is discussed in terms of the three-dimensional structure of the HA and the conformational change which the protein undergoes at the fusion pH optimum.     

N-Acetylneuraminyllactosylceramide, GM3-NeuAc, a new influenza A virus receptor which mediates the adsorption-fusion process of viral infection. Binding specificity of influenza virus A/Aichi/2/68 (H3N2) to membrane-associated GM3 with different molecular species of sialic acid

Agglutinates of native chicken erythrocytes caused by influenza virus A/Aichi/2/68 (H3N2) at 4 degrees C were potently fused and lysed at low pH (optimum pH 5.3) at 37 degrees C. Exogenous gangliosides GM3 (Sia alpha 2-3Gal beta 1-4Glc beta 1-ceramide) and GM2 (GalNAc beta 1-4(Sia alpha 2-3)-Gal beta 1-4Glc beta 1-ceramide) were integrated into the membranes of chicken asialoerythrocytes within 5-min incubation at 37 degrees C. We found that the incorporation of ganglioside GM3 containing N-acetylneuraminic acid into asialoerythrocytes restored the biological responsiveness to the virus as established by agglutination at 4 degrees C and fusion and hemolysis at 37 degrees C at pH 5.3. Biological responsiveness of GM3-NeuAc-erythrocytes to the virus was considerably higher than that of GM3-NeuGc-erythrocytes under the same experimental conditions. Treatment of the GM3-NeuAc-erythrocytes with neuraminidase again resulted in the complete abolishment of the response to the virus. Erythrocytes containing GM2-NeuAc showed no detectable biological responses toward the virus. The above results indicate that the hemagglutinin of influenza virus A/Aichi/2/68 (H3N2) recognizes the sialyloligosaccharide chain of ganglioside GM3 as its receptor which mediates the adsorption and fusion process on the virus entry into the host cells and has more preferential specificity for binding to N-acetylneuraminic acid-containing GM3 than that to N-glycolyl type in the target cell membranes.