HA Glycoprotein Research Articles

Genetic content of Influenza H3N2 vaccine seeds

Influenza vaccine seeds produced in chicken eggs are selected through HA and NA surface glycoproteins antigenicity, as well as through high replicative ability. Here we characterize the genetic content of recently used thirteen H3N2 influenza vaccine seeds. Interestingly, sequence analysis of the vaccine seeds shows reassortment events leading to PR8:H3N2 segment constellations, ranging from the 6:2 to 2:6 constellations. This study shows that the H3N2 PB1 is the most frequent internal segment incorporated in the tested vaccines seeds.

Recombinant Soluble, Multimeric HA and NA Exhibit Distinctive Types of Protection against Pandemic Swine-Origin 2009 A(H1N1) Influenza Virus Infection in Ferrets

The emergence and subsequent swift and global spread of the swine-origin influenza virus A(H1N1) in 2009 once again emphasizes the strong need for effective vaccines that can be developed rapidly and applied safely. With this aim, we produced soluble, multimeric forms of the 2009 A(H1N1) HA (sHA(3)) and NA (sNA(4)) surface glycoproteins using a virus-free mammalian expression system and evaluated their efficacy as vaccines in ferrets. Immunization twice with 3.75-microg doses of these antigens elicited strong antibody responses, which were adjuvant dependent. Interestingly, coadministration of both antigens strongly enhanced the HA-specific but not the NA-specific responses. Distinct patterns of protection were observed upon challenge inoculation with the homologous H1N1 virus. Whereas vaccination with sHA(3) dramatically reduced virus replication (e.g., by lowering pulmonary titers by about 5 log(10) units), immunization with sNA(4) markedly decreased the clinical effects of infection, such as body weight loss and lung pathology. Clearly, optimal protection was achieved by the combination of the two antigens. Our observations demonstrate the great vaccine potential of multimeric HA and NA ectodomains, as these can be easily, rapidly, flexibly, and safely produced in high quantities. In particular, our study underscores the underrated importance of NA in influenza vaccination, which we found to profoundly and specifically contribute to protection by HA. Its inclusion in a vaccine is likely to reduce the HA dose required and to broaden the protective immunity.

Increasing Prevalence of Unique Mutation Patterns in H5N1 Avian Influenza Virus HA and NA Glycoproteins from Human Infections in Egypt

Highly pathogenic avian influenza H5N1 virus (HPAIV) continues to be a candidate of a further influenza virus pandemic. Egypt is the country worst affected by human cases of HPAIV H5N1 infection in 2009. Increased infection of preschool children and decreased mortality rates suggested subtle changes in the epidemiology of the infection. Among other factors, the evolution of several conspicuous viral genetic markers in the HA and NA genes of HPAIV H5N1 viruses of human cases from Egypt and their putative influence on biological virus characteristics described here may contribute to this situation.

Characterization of pseudotyped viruses coated with hemagglutinin of H5N1 avian influenza

To construct pseudovirus bearing H5N1 HA based on a lentivirus vector system. Then we study the biological feature of the pseudovirus. With the newly established viral particles, we performed the serological tests. H5N1 avian influenza virus that isolated from human case was cloned to construct pLP-HA, then pLP-HA co-transfected with lentivirus vector plasmids pLP1, pLP2 and pEmGFP into 293T cells. The supernatant was harvested 48h post-transfection. Concentrated by super centrifuge, the pseudotyped viruses were analyzed by infection test, HA test and micro-neutralization test. At the same time, optimized HA gene and a Vietnam H5N1 HA gene were used to construct pseudotyped virus for comparison. Pseudotyped virus particles can be observed with electronic microscope. Western-blot revealed that HA glycoprotein can be expressed in the virions. Our neutralization assay by using the pseudoviruses was comparable with the conventional microneutralization assay with wild-type viruses. A high degree of correlation was detected. Pseudotyped Viruses coated with HA of H5N1 High Pathogenic Avian Influenza were successfully constructed; it can be used to for the microneutralization assay. The HA gene from different sources affect the efficiency of the packaging of the pseudovirus. But the optimized HA gene can not obviously improve packaging efficiency of the pseudovirus.

Changes in H5N1 influenza virus hemagglutinin receptor binding domain affect systemic spread

The HA of influenza virus is a receptor-binding and fusion protein that is required to initiate infection. The HA receptor-binding domain determines the species of sialyl receptors recognized by influenza viruses. Here, we demonstrate that changes in the HA receptor-binding domain alter the ability of the H5N1 virus to spread systemically in mice. The A/Vietnam/1203/04 (VN1203) and A/Hong Kong/213/03 (HK213) viruses are consistently lethal to domestic chickens but differ in their pathogenicity to mammals. Insertion of the VN1203 HA and neuraminidase (NA) genes into recombinant HK213 virus expanded its tissue tropism and increased its lethality in mice; conversely, insertion of HK213 HA and NA genes into recombinant VN1203 virus decreased its systemic spread and lethality. The VN1203 and HK213 HAs differ by 10 aa, and HK213 HA has shown greater binding affinity for synthetic alpha2,6-linked sialyl receptor. Introduction of an S227N change and removal of N-linked glycosylation at residue 158 increased the alpha2,6-binding affinity of VN1203 HA. Recombinant VN1203 virus carrying the S227N change alone or with the residue-158 glycosylation site removed showed reduced lethality and systemic spread in mice but not in domestic chickens. Wild-type VN1203 virus exhibited the greatest efficiency in systemic spread after intramuscular inoculation and in infection of mouse bone marrow-derived dendritic cells and conventional pulmonary dendritic cells. These results show that VN1203 HA glycoprotein confers pathogenicity by facilitating systemic spread in mice; they also suggest that a minor change in receptor binding domain may modulate the virulence of H5N1 viruses.

A Novel Intranasal Virus-Like Particle (VLP) Vaccine Designed to Protect against the Pandemic 1918 Influenza A Virus (H1N1)

We have prepared a virus-like particle (VLP) vaccine bearing the surface glycoproteins HA and NA of the 1918 influenza A virus by infecting Sf9 cells with a quadruple recombinant baculovirus that expresses the four influenza proteins (HA, NA, M1, and M2) required for the assembly and budding of the VLPs. The presence of HA and M1 in the purified VLPs was confirmed by Western blot, and that of NA by a neuraminidase enzymatic assay. For in vivo studies, the 1918 VLP vaccine was formulated with or without an oligonucleotide containing two CpG motifs and administered in two doses 2 wk apart via the intranasal route. The antibody titers in mice immunized with VLP vaccines were higher than in mice vaccinated with an inactivated swine virus (H1N1) control, when CHO cells expressing 1918 HA were used as antigen. The opposite result was obtained when disrupted swine virus was the antigen for the ELISA test. Vaccine efficacy was evaluated by challenging immunized mice with the 1918 antigenically related influenza virus A/swine/Iowa/15/30 (H1N1) and measuring viral titers in the upper and lower respiratory tract. Mice immunized with VLP vaccine plus CpG demonstrated significantly lower viral titers in the nose and lungs than did the control on days 2 and 4 postchallenge and completely cleared the virus by day 6. Furthermore, they did not show symptoms of disease although there was a minor decrease in body weight. Mice vaccinated with VLP alone also demonstrated significantly lower viral titers in the nose and lungs than did the placebo group as well as the inactivated virus group on days 4 and 6 postchallenge. These results suggest that it is feasible to make a safe and immunogenic vaccine to protect against the extremely virulent 1918 virus, using a novel and safe cell-based technology.

Phenotypic properties resulting from directed gene segment reassortment between wild-type A/Sydney/5/97 influenza virus and the live attenuated vaccine strain

Widespread use of a live-attenuated influenza vaccine (LAIV) in the United States (licensed as FluMist) raises the possibility that vaccine viruses will contribute gene segments to the type A influenza virus gene pool. Progeny viruses possessing new genotypes might arise from genetic reassortment between circulating wild-type (wt) and vaccine strains, but it will be difficult to predict whether they will be viable or exhibit novel properties. To begin addressing these uncertainties, reverse-genetics was used to generate 34 reassortant viruses derived from wt influenza virus A/Sydney/5/97 and the corresponding live vaccine strain. The reassortants contained different combinations of vaccine and wt PB2, PB1, PA, NP, M, and NS gene segments whereas all strains encoded wt HA and NA glycoproteins. The phenotypes of the reassortant strains were compared to wt and vaccine viruses by evaluating temperature-sensitive ( ts) plaque formation and replication attenuation ( att) in ferrets following intranasal inoculation. The results demonstrated that the vaccine virus PB1, PB2, and NP gene segments were dominant when introduced into the wt A/Sydney/5/97 genetic background, producing recombinant viruses that expressed the ts and att phenotypes. A dominant attenuated phenotype also was evident when reassortant strains contained the vaccine M or PA gene segments, even though these polypeptides are not temperature-sensitive. Although the vaccine M and NS gene segments typically are not associated with temperature sensitivity, a number of reassortants containing these vaccine gene segments did exhibit a more restricted ts phenotype. Overall, no reassortant strains were more virulent than wt, and in fact, 33 of the 34 recombinant viruses replicated less efficiently in infected ferrets. These results suggest that genetic reassortment between wt and vaccine strains is unlikely to produce viruses having novel properties that differ substantially from either progenitor, and that the likely outcome of reassortment will be attenuated viruses.

The ion channels coded by viruses.

Pathogenicity and virulence are multifactorial traits, depending on interaction of viruses with susceptible cells and organisms. The ion channels coded by viruses, viroporins, represent only one factor taking part in the cascade of interactions between virus and cell, leading to the entry of virus, replication and to profound changes in membrane permeability. The M2 protein from influenza A virus forms proton-selective, pH-regulated channel involved in regulating vesicular pH, a function important for the correct maturation of HA glycoprotein. The NB glycoprotein of influenza B viruses is an integral membrane protein with an ion channel activity. The CM2 protein of influenza C virus is an integral membrane glycoprotein structurally analogous to influenza A virus M2 and influenza B virus NB proteins. The picornavirus 3A protein is involved in cell lysis and shows homology with other lytic proteins. Vpu is an oligomeric integral membrane protein encoded by HIV-1, which forms ion channels. The togavirus 6K protein shows structural similarities with other viroporins.

The development of live attenuated cold-adapted influenza virus vaccine for humans.

A procedure to attenuate live influenza virus of type A and type B was developed using adaptation of the virus to grow at 25 degrees C (cold adaptation; ca). Through a series of stepwise passages, two stable mutants were obtained and designated as 'Master' strains, one for type A influenza virus (A/Ann Arbor/6/60-H2N2) and one for type B influenza virus (B/Ann Arbor/1/66). These mutants were used in genetic reassortment using either the classical method or more recently described reverse genetics to update the relevant surface antigens of the circulating strains of influenza virus. The derivation is based on the concept of 6/2 where 6 signifies the six internal genes of the master strain and 2 refers to the two genes coding for the two surface glycoproteins HA and NA of the circulating influenza virus. The advantages of this vaccine were demonstrated to be (1) proper level of attenuation, (2) non-transmissibility, (3) genetic stability, (4) presence of the ca and ts markers and (5) immunogenicity involving both local and the cell-mediated immune responses. The clinical trials in infants, children, adults and elderly have provided the necessary data for eventual licensing of this vaccine. The ease of administration (intranasal) safety and high efficacy make this vaccine suitable to prevent influenza virus infection in all age groups.

Genomic study of hemagglutinins of swine influenza (H1N1) viruses associated with acute and chronic respiratory diseases in pigs.

The nucleotide sequences of HA1 domain of hemagglutinin of clinical H1N1 influenza viruses, isolated during recent outbreaks of respiratory problems in pig farms of Quebec, was determined. The viruses A/Sw/Quebec/3291/90 (SwQc3291) and A/Sw/Quebec/1747/90 (SwQc1747), associated with chronic respiratory disease, showed close similarity for their deduced aa sequences. When compared with the published data of A/Sw/Quebec/5393/91 (SwQc91), the variations observed included Cb and Ca antigenic sites in SwQc3291 and Sb and Ca sites in SwQc1747 isolates. These variants were antigenically related to SwQc91 virus associated with chronic respiratory disease, but differed from the more classical A/Sw/Quebec/192/81 (SwQc81) strain. In contrast, A/Sw/Quebec/1192/86 (SwQc1192) isolate, associated with acute respiratory influenza, showed maximum number of differences including Ca, Cb, Sa and Sb antigenic sites. The latter, as well as the SwQc81 strain, were antigenically distinct from SwQc91 virus on the basis of its cross-reactivity to MAbs directed against the HA glycoprotein. Estimation of genetic distances and phylogenic tree analysis showed that SwQc1747 and SwQc3291 were closely related, but these viruses along with SwQc1192 were considerably divergent from SwQc91 virus.

Effects of protease inhibitors on the cleavage of influenza HA glycoprotein mimic by human respiratory tract protease: M. Hosoya 1, S. Mori 2, H. Suzuki 1, and S. Shigeta 2 Department of Pediatrics 1 and Microbiology 2, Fukushima Medical College, Fukushima 960-12, Japan

Assessment of the extent of bioavailability (EBA) of salmon calcitonin (sCT) from hypocalcemic effects after intranasal administration was presented in rats. An integrated pharmacokineticpharmacodynamic (PK-PD) model with the endogenous Ca regulation system was applied. The influence of camostat mesilate, a protease inhibitor, on absorption of sCT was also estimated. Camostat, coadministered intravascularly, delayed the elimination of sCT. Although the hypocalcemic effect of sCT after i.v. administration was accelerated when camostat was coadministered intravenously, the enhanced effect could not be expressed only by pharmacokinetic change of sCT, and then the pharmacological data in the presence of camostat were analyzed to obtain optimal PD parameters. For the absorption of sCT after i.n. administration, a saturable absorptive process and a zero-order kinetic clearance from the nasal cavity were introduced to the model. The regression curves fitted the observed data, and camostat caused both an increase in maximum absorption rate and a decrease in the clearance parameter compared with the control. According to this modified PK-PD relationship, plasma sCT concentrations following i.n. administration of sCT with camostat were predicted well using its pharmacological effects. The EBA of sCT calculated from the simulated concentrations increased more than 4-folds compared with the control study. These results indicate the potential for prediction of plasma sCT concentration from the hypocalcemic effect.

Abortive replication of influenza A viruses in HeLa 229 cells

Several experimental data support the idea that certain mammalian cells are unable to replicate influenza viruses type A, although these viruses can efficiently penetrate the cells. This cannot be attributed to a lack of specific receptors on the cell surface, but depends upon the failure of specific step(s) to occur during viral growth. Here we report a study of abortiveness of human and avian type A influenza viruses in HeLa 229 cells. Viral polypeptide synthesis was monitored by [ 35S]methionine pulse labelling at several time points after infection, showing that normal amounts of virus-induced components were synthesized. Cellular fractionation of HeLa 229 cells infected by influenza viruses showed that the distribution of viral proteins into nuclear and cytoplasmic compartments was comparable to that seen in the permissive host, chick embryo fibroblasts. Viral HA glycoprotein, produced during the infectious cycle, was entirely found in the cytoplasm of infected HeLa 229 cells. The polypeptide was able to agglutinate red blood cells but did not show positive haemadsorption even at late times of infection. Therefore it seems that during the maturation of viral particles there is a failure of the haemagglutinin to perform a correct insertion into the plasma membrane of infected HeLa 229 cells.

Sequence of the fusion protein gene of human parainfluenza type 2 virus and its 3′ intergenic region: Lack of small hydrophobic (SH) gene

cDNA clones representing the fusion (F) gene of human parainfluenza virus type 2 (PIV-2) were isolated from cDNA libraries constructed from virus-specific mRNA and genomic RNA, and the complete nucleotide sequence of the F gene was determined. The F gene is 1854 nucleotides long and encodes one long open reading frame of 551 amino acids. The cleavage site for activation of the precursor F o protein is ThrArgGlnLysArg. The F gene of PIV-2 is most closely related to those of simian virus 5 (SV5) and mumps virus (MuV). Interestingly, although the HN glycoprotein of PIV-2 shows no relatedness to the HA glycoprotein of measles virus (MV), a distinct homology is found in the F proteins of PIV-2 and MV. As concerns F proteins, paramyxoviruses can be divided intt two subgroups; that is, PIV-2, SV5, and MuV belong to one group, and HPIV-1, SV, and PIV-3 belong to the other group. Newcastle disease virus (NDV) and MV are intermediate. Coding regions for small hydrophobic (SH) proteins have been found between the HN and F genes of SV5 and MuV, which are the viruses most closely related to PIV-2. However, such a gene could not be detected in two different strains of PIV-2.

Fusion-mediated microinjection of liposome-enclosed DNA into cultured cells with the aid of influenza virus glycoproteins

Influenza viruses were able to mediate fusion of DNA-loaded liposomes with living cultured cells such as monkey COS-7 cells. This was inferred from the appearance of CAT activity in recipient cells incubated with the combination of influenza viruses and liposomes loaded with the plasmid pSV2CAT. Influenza virions were found to be as efficient as intact Sendai virions in mediating microinjection of foreign DNA into living cells. Also, reconstituted envelopes bearing either influenza glycoproteins or the combination of Sendai and influenza glycoproteins were highly efficient in promoting fusion of loaded liposomes with recipient cells. Introduction of DNA into cultured cells required the presence of an active influenza fusion protein; namely, an active HA glycoprotein. Very little or no CAT activity was observed in cells incubated with loaded liposomes and unfusogenic influenza viruses. The virus-induced fusion event probably occurs within intracellular organelles such as endosomes following receptor-mediated endocytosis of virusliposome complexes. This is due to the fact that the viral fusion glycoprotein is activated only at acidic pH values such as those which characterize the intraendosomal environment. Results of the present work demonstrate for the first time microinjection of foreign DNA via fusion with membranes of intracellular organelles. The potential of the present system to serve as a biological carrier for in vivo use is discussed.

Increased viral pathogenicity after insertion of a 28S ribosomal RNA sequence into the haemagglutinin gene of an influenza virus

The haemagglutinin glycoprotein HA of influenza viruses is responsible for the attachment of the virus to neuraminic acid-containing receptors at the cell surface and subsequent penetration by triggering fusion of the viral envelope with cellular membranes. To express full activity of the newly synthesized precursor, HA has to be modified by post-translational proteolytic cleavage into the polypeptides HA1 and HA2 by cellular enzymes. If proteases suitable for cleavage are not present in the host cell, the resulting virus particles are non-infectious. During adaptation of the apathogenic influenza virus A/turkey/Oregon/71 to chicken embryo cells, which are not permissive for HA cleavage, we obtained an infectious virus variant with increased pathogenicity. Sequence analysis revealed that during adaptation 54 nucleotides were inserted into the HA gene; their sequence corresponds to a region of the 28S ribosomal RNA. This insertion is probably responsible for increased cleavability of HA, as well as for infectivity and pathogenicity of the adapted virus.

Human Monoclonal Antibodies to Influenza Virus: IgG Subclass and Light Chain Distribution

Three adults and three children were immunized with inactivated or live attenuated influenza vaccines and 98 IgG monoclonal antibodies derived from EBV immortalization of their blood lymphocytes were studied. All antibodies reacted specifically with influenza A H3N2 or H1N1 whole virus and 73 of 74 tested reacted with the purified HA glycoproteins. The majority (76%) of 77 monoclonal antibodies adequately tested by ELISA or solid phase RIA contained lambda light chains. ELISA analysis of the IgG subclass of the six persons' postimmunization anti-influenza serum activity and of monoclonal antibodies derived from them showed a similar predominance of IgG1.

Radiation inactivation analysis of influenza virus reveals different target sizes for fusion, leakage, and neuraminidase activities.

The size of the functional units responsible for several activities carried out by the influenza virus envelope glycoproteins was determined by radiation inactivation analysis. Neuraminidase activity, which resides in the glycoprotein NA, was inactivated exponentially with an increasing radiation dose, yielding a target size of 94 +/- 5 kilodaltons (kDa), in reasonable agreement with that of the disulfide-bonded dimer (120 kDa). All the other activities studied are properties of the HA glycoprotein and were normalized to the known molecular weight of the neuraminidase dimer. Virus-induced fusion activity was measured by two phospholipid dilution assays: relief of energy transfer between N-(7-nitro-2,1,3-benzoxadiazol-4-yl)dipalmitoyl-L-alpha- phosphatidylethanolamine (N-NBD-PE) and N-(lissamine rhodamine B sulfonyl)-dioleoyl-L-alpha-phosphatidylethanolamine (N-Rh-PE) in target liposomes and relief of self-quenching of N-Rh-PE in target liposomes. Radiation inactivation of fusion activity proceeded exponentially with radiation dose, yielding normalized target sizes of 68 +/- 6 kDa by assay i and 70 +/- 4 kDa by assay ii. These values are close to the molecular weight of a single disulfide-bonded (HA1 + HA2) unit (75 kDa), the "monomer" of the HA trimer. A single monomer is thus inactivated by each radiation event, and each monomer (or some part of it) constitutes a minimal functional unit capable of mediating fusion. Virus-induced leakage of calcein from target liposomes and virus-induced leakage of hemoglobin from erythrocytes (hemolysis) both showed more complex inactivation behavior: a pronounced shoulder was present in both inactivation curves, followed by a steep drop in activity at higher radiation levels.(ABSTRACT TRUNCATED AT 250 WORDS)

Sorting and endocytosis of viral glycoproteins in transfected polarized epithelial cells.

Previous studies (Rindler, M. J., I. E., Ivanov, H. Plesken, and D. D. Sabatini, 1985, J. Cell Biol., 100: 136-151; Rindler, M. J., I. E. Ivanov, H. Plesken, E. J. Rodriguez-Boulan, and D. D. Sabatini, 1984, J. Cell Biol., 98: 1304-1319) have demonstrated that in polarized Madin-Darby canine kidney cells infected with vesicular stomatitis virus (VSV) or influenza virus the viral envelope glycoproteins G and HA are segregated to the basolateral and apical plasma membrane domains, respectively, where budding of the corresponding viruses takes place. Furthermore, it has been shown that this segregation of the glycoproteins reflects the polarized delivery of the newly synthesized polypeptides to each surface domain. In transfection experiments using eukaryotic expression plasmids that contain cDNAs encoding the viral glycoproteins, it is now shown that even in the absence of other viral components, both proteins are effectively segregated to the appropriate cell surface domain. In transfected cells, the HA glycoprotein was almost exclusively localized in the apical cell surface, whereas the G protein, although preferentially localized in the basolateral domains, was also present in lower amounts, in the apical surfaces of many cells. Using transfected and infected cells, it was demonstrated that, after reaching the cell surface, the G protein, but not the HA protein, undergoes interiorization by endocytosis. Thus, in the presence of chloroquine, a drug that blocks return of interiorized plasma membrane proteins to the cell surface, the G protein was quantitatively trapped in endosome- or lysosome-like vesicles. The sequestration of G was a rapid process that was completed in many cells by 1-2 h after chloroquine treatment. The fact that in transfected cells the surface content of G protein was not noticeably reduced during a 5-h incubation with cycloheximide, a protein synthesis inhibitor that did not prevent the effect of chloroquine, implies that normally, G protein molecules are not only interiorized but are also recycled to the cell surface.

Sialic acid is incorporated into influenza hemagglutinin glycoproteins in the absence of viral neuraminidase

We have analyzed the pronase-derived glycopeptides of the hemagglutinin glycoproteins expressed from SV40 vectors carrying cloned cDNA copies of the HA gene and of HA isolated from influenza virions (A/Jap/305/57). The glycopeptides derived from the HA glycoprotein obtained from cloned genes were heterogeneous, ranging in size from 3800 to 2800 dallons. Upon treatment with neuraminidase, sialic acid was released from the glycopeptides and their size was reduced to 2900–2400 daltons. However, under the same conditions, no sialic acid was detected in the virion HA. The presence of sialic acid was confirmed by monosaccharide analysis of the HA glycoprotein derived from products of cloned genes. These results support the idea that during replication of influenza virus, the viral neuraminidase cleaves sialic acid from the HA glycoprotein in infected cells.

Biogenesis of epithelial cell plasma membranes.

Polarized monolayers of cultured epithelial cells, such as the kidney-derived MDCK cell line, when infected with enveloped viruses, provide a convenient model system for study of the intracellular routes followed by newly synthesized glycoproteins to reach specific domains of the plasma membrane. The polarized nature of the monolayers is reflected in the asymmetric assembly of enveloped viruses, some of which, such as influenza and simian virus 5 (SV5), bud from the apical surfaces of the cells, while others, such as vesicular stomatitis virus (VSV), emerge from the basolateral surfaces. MDCK cells can sustain double infection with viruses of different budding polarity, and within such cells the envelope glycoproteins of the two viruses are synthesized simultaneously and assembled into virions at different sites. Immunoelectron microscopic observations of doubly infected cells show that glycoproteins of influenza and VSV traverse the same Golgi apparatus. This indicates that critical sorting steps must take place during or after passage of the glycoproteins through the organelle. Following passage through the Golgi, the HA glycoprotein accumulates almost exclusively at the apical surface, where the influenza virions assemble. Significant amounts of the G protein, however, are detected on both plasma membranes in singly and doubly infected cells, although VSV virion assembly is limited to basolateral domains. These observations indicate that the site of VSV budding is not exclusively determined by the presence of G polypeptides on a given cell-surface domain. It is possible that other cellular or viral components are responsible for the selection of the appropriate budding domain or that the G protein found on the apical surface must be transferred to the basolateral domain before it becomes competent for assembly.