Influence of acylation sites of influenza B virus hemagglutinin on fusion pore formation and dilation

The cytoplasmic tail (CT) of hemagglutinin (HA) of influenza B virus (BHA) contains at positions 578 and 581 two highly conserved cysteine residues (Cys578 and Cys581) that are modified with palmitic acid (PA) through a thioester linkage. To investigate the role of PA in the fusion activity of BHA, site-specific mutagenesis was performed with influenza B virus B/Kanagawa/73 HA cDNA. All of the HA mutants were expressed on Cos cells by an expression vector. The membrane fusion ability of the HA mutants at a low pH was quantitatively examined with lipid (octadecyl rhodamine B chloride) and aqueous (calcein) dye transfer assays and with the syncytium formation assay. Two deacylation mutants lacking a CT or carrying serine residues substituting for Cys578 and Cys581 promoted full fusion. However, one of the single-acylation-site mutants, C6, in which Cys581 is replaced with serine, promoted hemifusion but not pore formation. In contrast, four other single-acylation-site mutants that have a sole cysteine residue in the CT at position 575, 577, 579, or 581 promoted full fusion. The impaired pore-forming ability of C6 was improved by amino acid substitution between residues 578 and 582 or by deletion of the carboxy-terminal leucine at position 582. Syncytium-forming ability, however, was not adequately restored by these mutations. These facts indicated that the acylation was not significant in membrane fusion by BHA but that pore formation and pore dilation were appreciably affected by the particular amino acid sequence of the CT and the existence of a single acylation site in CT residue 578.     

Regulation of receptor binding affinity of influenza virus hemagglutinin by its carbohydrate moiety.

The hemagglutinin (HA) of the fowl plague virus (FPV) strain of influenza A virus has two N-linked oligosaccharides attached to Asn123 and Asn149 in the vicinity of the receptor binding site. The effect of these carbohydrate side chains on the binding of HA to neuraminic acid-containing receptors has been analyzed. When the oligosaccharides were deleted by site-specific mutagenesis, HA expressed from a simian virus 40 vector showed enhanced hemadsorbing activity. Binding was so strong under these conditions that erythrocytes were no longer released by viral neuraminidase and that release was significantly reduced when neuraminidase from Vibrio cholerae was used. Similarly, when these oligosaccharides were removed selectively from purified viruses by N-glycosidase F, such virions were unable to elute from receptors, although they retained neuraminidase activity. Thus, release of FPV from cell receptors depends on the presence of the HA glycans at Asn123 and Asn149. On the other hand, receptor binding was abolished when these oligosaccharides were sialylated after expression in the absence of neuraminidase (M. Ohuchi, A. Feldmann, R. Ohuchi, and H.-D. Klenk, Virology 212:77-83, 1995). These observations indicate that the receptor affinity of FPV HA is controlled by oligosaccharides adjacent to the receptor binding site.     

Modification of the cytoplasmic domain of influenza virus hemagglutinin affects enlargement of the fusion pore

The fusion activity of chimeras of influenza virus hemagglutinin (HA) (from A/fpv/Rostock/34; subtype H7) with the transmembrane domain (TM) and/or cytoplasmic tail (CT) either from the nonviral, nonfusogenic T-cell surface protein CD4 or from the fusogenic Sendai virus F-protein was studied. Wild-type or chimeric HA was expressed in CV-1 cells by the transient T7-RNA-polymerase vaccinia virus expression system. Subsequently, the fusion activity of the expression products was monitored with red blood cells or ghosts as target cells. To assess the different steps of fusion, target cells were labeled with the fluorescent membrane label octadecyl rhodamine B-chloride (R18) (membrane fusion) and with the cytoplasmic fluorophores calcein (molecular weight [MW], 623; formation of small aqueous fusion pore) and tetramethylrhodamine-dextran (MW, 10,000; enlargement of fusion pore). All chimeric HA/F-proteins, as well as the chimera with the TM of CD4 and the CT of HA, were able to mediate the different steps of fusion very similarly to wild-type HA. Quite differently, chimeric proteins with the CT of CD4 were strongly impaired in mediating pore enlargement. However, membrane fusion and formation of small pores were similar to those of wild-type HA, indicating that the conformational change of the ectodomain and earlier fusion steps were not inhibited. Various properties of the CT which may affect pore enlargement are considered. We surmise that the hydrophobicity of the sequence adjacent to the transmembrane domain is important for pore dilation.     

Acylation-mediated membrane anchoring of avian influenza virus hemagglutinin is essential for fusion pore formation and virus infectivity

Attachment of palmitic acid to cysteine residues is a common modification of viral glycoproteins. The influenza virus hemagglutinin (HA) has three conserved cysteine residues at its C terminus serving as acylation sites. To analyze the structural and functional roles of acylation, we have generated by reverse genetics a series of mutants (Ac1, Ac2, and Ac3) of fowl plague virus (FPV) containing HA in which the acylation sites at positions 551, 559, and 562, respectively, have been abolished. When virus growth in CV1 and MDCK cells was analyzed, similar amounts of virus particles were observed with the mutants and the wild type. Protein patterns and lipid compositions, characterized by high cholesterol and glycolipid contents, were also indistinguishable. However, compared to wild-type virus, Ac2 and Ac3 virions were 10 and almost 1,000 times less infectious, respectively. Fluorescence transfer experiments revealed that loss of acyl chains impeded formation of fusion pores, whereas hemifusion was not affected. When the affinity to detergent-insoluble glycolipid (DIG) domains was analyzed by Triton X-100 treatment of infected cells and virions, solubilization of Ac2 and Ac3 HAs was markedly facilitated. These observations show that acylation of the cytoplasmic tail, while not necessary for targeting to DIG domains, promotes the firm anchoring and retention of FPV HA in these domains. They also indicate that tight DIG association of FPV HA is essential for formation of fusion pores and thus probably for infectivity.     

Infectivity studies of influenza virus hemagglutinin receptor binding site mutants in mice

The replicative properties of influenza virus hemagglutinin (HA) mutants with altered receptor binding characteristics were analyzed following intranasal inoculation of mice. Among the mutants examined was a virus containing a Y98F substitution at a conserved position in the receptor binding site that leads to a 20-fold reduction in binding. This mutant can replicate as well as wild-type (WT) virus in MDCK cells and in embryonated chicken eggs but is highly attenuated in mice, exhibiting titers in lungs more than 1,000-fold lower than those of the WT. The capacity of the Y98F mutant to induce antibody responses and the structural locations of HA reversion mutations are examined.     

Analysis of influenza virus hemagglutinin receptor binding mutants with limited receptor recognition properties and conditional replication characteristics

To examine the range of selective processes that potentially operate when poorly binding influenza viruses adapt to replicate more efficiently in alternative environments, we passaged a virus containing an attenuating mutation in the hemagglutinin (HA) receptor binding site in mice and characterized the resulting mutants with respect to the structural locations of mutations selected, the replication phenotypes of the viruses, and their binding properties on glycan microarrays. The initial attenuated virus had a tyrosine-to-phenylalanine mutation at HA1 position 98 (Y98F), located in the receptor binding pocket, but viruses that were selected contained second-site pseudoreversion mutations in various structural locations that revealed a range of molecular mechanisms for modulating receptor binding that go beyond the scope that is generally mapped using receptor specificity mutants. A comparison of virus titers in the mouse respiratory tract versus MDCK cells in culture showed that the mutants displayed distinctive replication properties depending on the system, but all were less attenuated in mice than the Y98F virus. An analysis of receptor binding properties confirmed that the initial Y98F virus bound poorly to several different species of erythrocytes, while all mutants reacquired various degrees of hemagglutination activity. Interestingly, both the Y98F virus and pseudoreversion mutants were shown to bind very inefficiently to standard glycan microarrays containing an abundance of binding substrates for most influenza viruses that have been characterized to date, provided by the Consortium for Functional Glycomics. The viruses were also examined on a recently developed microarray containing glycans terminating in sialic acid derivatives, and limited binding to a potentially interesting subset of glycans was revealed. The results are discussed with respect to mechanisms for HA-mediated receptor binding, as well as regarding the species of molecules that may act as receptors for influenza virus on host cell surfaces.     

Conformational intermediates and fusion activity of influenza virus hemagglutinin

Three strains of influenza virus (H1, H2, and H3) exhibited similar characteristics in the ability of their hemagglutinin (HA) to induce membrane fusion, but the HAs differed in their susceptibility to inactivation. The extent of inactivation depended on the pH of preincubation and was lowest for A/Japan (H2 subtype), in agreement with previous studies (A. Puri, F. Booy, R. W. Doms, J. M. White, and R. Blumenthal, J. Virol. 64:3824-3832, 1990). While significant inactivation of X31 (H3 subtype) was observed at 37 degrees C at pH values corresponding to the maximum of fusion (about pH 5.0), no inactivation was seen at preincubation pH values 0.2 to 0.4 pH units higher. Surprisingly, low-pH preincubation under those conditions enhanced the fusion rates and extents of A/Japan as well as those of X31. For A/PR 8/34 (H1 subtype), neither a shift of the pH (to >5.0) nor a decrease of the temperature to 20 degrees C was sufficient to prevent inactivation. We provide evidence that the activated HA is a conformational intermediate distinct from the native structure and from the final structure associated with the conformational change of HA, which is implicated by the high-resolution structure of the soluble trimeric fragment TBHA2 (P. A. Bullough, F. M. Hughson, J. J. Skehel, and D. C. Wiley, Nature 371:37-43, 1994).     

ph-Dependent hydrophobicity profile of hemagglutinin of influenza virus and its possible relevance in virus fusion

The hydropathy profile of hemagglutinin (HA) subunits HA1 and HA2 of influenza virus X31 and A/PR 8/34 is analyzed at different pH. At neutral pH (7.4) pronounced hydrophobic sequences of HA correspond to the N-terminus and the transmembrane spanning sequence of HA2. At pH 5.0 where influenza virus is known to fuse with biological membranes several hydrophobic sequences in the ectodomain exist which are comparable in both the hydrophobicity and length of the N-terminus of HA2. It is suggested that these hydrophobic stretches are important for the fusion complex, in addition to the N-terminal site of HA2.Abbreviations HA hemagglutinin - NHA2 N-terminus of HA2     

Lysolipids do not inhibit influenza virus fusion by interaction with hemagglutinin

The interaction of a spin-labeled lysophosphatidylcholine analog with intact and bromelain-treated influenza viruses as well as with the bromelain-solubilized hemagglutinin ectodomain has been studied. The inhibition of fusion of influenza viruses with erythrocytes by the lysophosphatidylcholine analog was similar to that observed for non-labeled lysophosphatidylcholine. Only a weak interaction of the lysophosphatidylcholine analog with the hemagglutinin ectodomain was observed even upon triggering the conformational change of the ectodomain at a low pH. A significant interaction of spin-labeled lysophosphatidylcholine with the hemagglutinin ectodomain of intact viruses was observed neither at neutral nor at low pH, whereas a strong interaction of the lipid analog with the viral lipid bilayer was evident. We suggest that the high number of lipid binding sites of the virus bilayer and their affinity compete efficiently with binding sites of the hemagglutinin ectodomain. We conclude that the inhibition of influenza virus fusion by lysolipids is not mediated by binding to the hemagglutinin ectodomain, preventing its interaction with the target membrane. The results unambiguously argue for an inhibition mechanism based on the action of lysolipid inserted into the lipid bilayer.     

Oligosaccharide composition of an influenza virus hemagglutinin with host-determined binding properties

We have previously reported that the binding properties of the hemagglutinin (HA) of the WSN-F strain of influenza A are affected by the cells in which the virus is grown (Crecelius, D. M., Deom, C. M., and Schulze, I.T. (1984) Virology 139, 164-177); at 37 degrees C chick embryo fibroblast-grown F virus has a greater affinity for host cells than does the same virus grown in Madin-Darby bovine kidney (MDBK) cells. In an attempt to explain this host-determined property, we have characterized the carbohydrate put onto the viral HA by these two cells. Experiments using tunicamycin indicate that the HA made by MDBK cells contains about 4000 daltons of carbohydrate in excess of that on the HA from chick embryo fibroblast. Serial lectin affinity chromatography of the asparagine-linked oligosaccharides on the HA subunits, HA1 and HA2, detected a number of host-dependent differences in the complex oligosaccharides. Both HA1 and HA2 from MDBK cells contained more highly branched (i.e. tri- and tetraantennary) complex oligosaccharides than did the subunits from chick embryo fibroblasts. In addition, the HA subunits from the two sources differed in the amount of galactose-containing "bisected" complex oligosaccharides and in the presence of certain fucosylated triantennary oligosaccharides. Profiles of the asparagine-linked oligosaccharides from the host cells did not show these differences, indicating that the HA subunit profiles were not necessarily representative of the structures found on the cellular glycoproteins. The data support the conclusion that bulky oligosaccharides on the MDBK-HA subunits of WSN-F reduce the affinity of the virus for cellular receptors.     

Dilation of the influenza hemagglutinin fusion pore revealed by the kinetics of individual cell-cell fusion events

We have monitored kinetics of fusion between cell pairs consisting of a single influenza hemaglutinin (HA)-expressing cell and a single erythrocyte (RBC) that had been labeled with both a fluorescent lipid (Dil) in the membrane and a fluorescent solute (calcein) in the aqueous space. Initial fusion pore opening between the RBC and HA-expressing cell produced a change in RBC membrane potential (delta psi) that was monitored by a decrease in Dil fluorescence. This event was followed by two distinct stages of fusion pore dilation: the flux of fluorescent lipid (phi L) and the flux of a large aqueous fluorescent dye (phi s). We have analyzed the kinetics of events that occur as a result of transitions between a fusion pore (FP) and a solute permissive fusion pore (FPs). Our data are consistent with a fusion pore comprising six HA trimers.     

A single amino acid substitution in 1918 influenza virus hemagglutinin changes receptor binding specificity

The receptor binding specificity of influenza viruses may be important for host restriction of human and avian viruses. Here, we show that the hemagglutinin (HA) of the virus that caused the 1918 influenza pandemic has strain-specific differences in its receptor binding specificity. The A/South Carolina/1/18 HA preferentially binds the alpha2,6 sialic acid (human) cellular receptor, whereas the A/New York/1/18 HA, which differs by only one amino acid, binds both the alpha2,6 and the alpha2,3 sialic acid (avian) cellular receptors. Compared to the conserved consensus sequence in the receptor binding site of avian HAs, only a single amino acid at position 190 was changed in the A/New York/1/18 HA. Mutation of this single amino acid back to the avian consensus resulted in a preference for the avian receptor.     

Cholesterol promotes hemifusion and pore widening in membrane fusion induced by influenza hemagglutinin

Cholesterol-specific interactions that affect membrane fusion were tested for using insect cells; cells that have naturally low cholesterol levels (< 4 mol %). Sf9 cells were engineered (HAS cells) to express the hemagglutinin (HA) of the influenza virus X-31 strain. Enrichment of HAS cells with cholesterol reduced the delay between triggering and lipid dye transfer between HAS cells and human red blood cells (RBC), indicating that cholesterol facilitates membrane lipid mixing prior to fusion pore opening. Increased cholesterol also increased aqueous content transfer between HAS cells and RBC over a broad range of HA expression levels, suggesting that cholesterol also favors fusion pore expansion. This interpretation was tested using both trans-cell dye diffusion and fusion pore conductivity measurements in cholesterol-enriched cells. The results of this study support the hypothesis that host cell cholesterol acts at two stages in membrane fusion: (1) early, prior to fusion pore opening, and (2) late, during fusion pore expansion.     

Interactions of misfolded influenza virus hemagglutinin with binding protein (BiP)

We have characterized the association between the binding protein, BiP (also known as GRP 78), and misfolded forms of the influenza virus hemagglutinin precursor, HA0. BiP is a heat-shock-related protein that binds to unassembled immunoglobulin heavy chain and to a variety of misfolded proteins in the lumen of the ER. A small fraction (5-10%) of newly synthesized HA0 in CV-1 cells was found to be misfolded and retained in the ER. When glycosylation was blocked with tunicamycin, all of the HA0 produced was similarly misfolded. The misfolded HA0 was retained as relatively small (9-25-S) complexes associated with BiP. In these complexes the top domains of HA0 were correctly folded judging by their reactivity with monoclonal antibodies, but the polypeptides were cross-linked via anomalous interchain disulfides. The association with BiP was non-covalent and easily broken by warming to 37 degrees C or by adding ATP to the lysate. Pulse-chase experiments showed that HA0's self-association into complexes occurred immediately after synthesis and was followed rapidly by BiP association. The misfolded, BiP-associated HA0 was not transported to the plasma membrane but persisted as complexes in the ER for a long period of time before degradation (t1/2 = 6 h). The results suggested that BiP may be part of a quality control system in the ER and that one of its functions is to detect and retain misfolded proteins.     

Fatty acids on the A/USSR/77 influenza virus hemagglutinin facilitate the transition from hemifusion to fusion pore formation

Influenza virus hemagglutinin (HA) has three highly conserved acylation sites close to the carboxyl terminus of the HA2 subunit, one in the transmembrane domain and two in the cytoplasmic domain. Each site is modified by palmitic acid through a thioester linkage to cysteine. To elucidate the biological significance of HA acylation, the acylation sites of HA of influenza virus strain A/USSR/77 (H1N1) were changed by site-directed mutagenesis, and the membrane fusion activity of mutant HAs lacking the acylation site(s) was examined quantitatively using transfer assays of lipid (R18) and aqueous (calcein) dyes. Lipid mixing, so-called hemifusion, activity was not affected by deacylation, whereas transfer of aqueous dye, so-called fusion pore formation, was dramatically restricted. When the fusion reaction was induced by a lower pH than the optimal one, calcein transfer with the mutant HAs was improved, but simultaneously a considerable calcein leakage into the medium was observed. From these results, we conclude that the palmitic acids on the H1 subtype HA facilitate the transition from hemifusion to fusion pore formation.     

Structural characterization of an early fusion intermediate of influenza virus hemagglutinin

The hemagglutinin (HA) envelope protein of influenza virus mediates viral entry through membrane fusion in the acidic environment of the endosome. Crystal structures of HA in pre- and postfusion states have laid the foundation for proposals for a general fusion mechanism for viral envelope proteins. The large-scale conformational rearrangement of HA at low pH is triggered by a loop-to-helix transition of an interhelical loop (B loop) within the fusion domain and is often referred to as the "spring-loaded" mechanism. Although the receptor-binding HA1 subunit is believed to act as a "clamp" to keep the B loop in its metastable prefusion state at neutral pH, the "pH sensors" that are responsible for the clamp release and the ensuing structural transitions have remained elusive. Here we identify a mutation in the HA2 fusion domain from the influenza virus H2 subtype that stabilizes the HA trimer in a prefusion-like state at and below fusogenic pH. Crystal structures of this putative early intermediate state reveal reorganization of ionic interactions at the HA1-HA2 interface at acidic pH and deformation of the HA1 membrane-distal domain. Along with neutralization of glutamate residues on the B loop, these changes cause a rotation of the B loop and solvent exposure of conserved phenylalanines, which are key residues at the trimer interface of the postfusion structure. Thus, our study reveals the possible initial structural event that leads to release of the B loop from its prefusion conformation, which is aided by unexpected structural changes within the membrane-distal HA1 domain at low pH.     

Membrane fusion activity of reconstituted vesicles of influenza virus hemagglutinin glycoproteins

Reconstituted vesicles of hemagglutinin glycoproteins into egg yolk phosphatidylcholine/spin-labeled phosphatidylcholine/cholesterol (molar ratio 1.6:0.4:1) were prepared by dialysis. Preparations at appropriate protein-to-lipid ratios (1:44 and 1:105 mol/mol) contained vesicles with a diameter of 100-300 nm and a high density of spikes on the surface. These vesicles showed low pH-induced membrane fusion activity. At pH 5.2 and 37 degrees C, fusion with erythrocyte membranes took place very rapidly within 1-2 min and reached a plateau at 63-66% fusion. The fusion was negligibly small at neutral pH and was induced to occur at pH values lower than 6.0. The reconstituted vesicles caused hemolysis and fusion of human erythrocyte cells in the same pH range as that of the fusion with erythrocyte membranes. The low pH-induced fusion activity of the reconstituted vesicles is essentially the same as that of the parent virus. These vesicles can be used to deliver some reagents or drugs into target cell cytoplasm via fusion at lysosomes.     

A complex of influenza hemagglutinin with a neutralizing antibody that binds outside the virus receptor binding site.

The structure of a complex of influenza hemagglutinin (HA) with a neutralizing antibody shows that the antibody binds to HA at a distance from the virus receptor binding site. Comparison of the properties of this antibody and its Fab with those of an antibody that recognizes an epitope overlapping the receptor binding site leads to two main conclusions. First, inhibition of receptor binding is an important component of neutralization. Second, the efficiency of neutralization by the antibodies ranks in the same order as their avidities for HA, and their large size makes these antibodies highly efficient at neutralization, regardless of the location of their epitope in relation to the virus receptor binding site. These observations provide rationales for the range of antibody specificities that are detected in immune sera and for the distribution of sequence changes on the membrane-distal surface of influenza HAs that occur during 'antigenic drift.'     

Variant influenza virus hemagglutinin that induces fusion at elevated pH.

The hemagglutinin (HA) glycoprotein of influenza virus performs two critical roles during infection: it binds virus to cell surface sialic acids, and under mildly acidic conditions it induces fusion of the virion with intracellular membranes, liberating the genome into the cytoplasm. The pH dependence of fusion varies for different influenza virus strains. Here we report the isolation and characterization of a naturally occurring variant of the X31 strain that fuses at a pH 0.2 units higher than the parent strain does and that is less sensitive to the effects of ammonium chloride, a compound known to elevate endosomal pH. The bromelain-solubilized ectodomain of the variant HA displayed a corresponding shift in the pH at which it changed conformation and bound to liposomes. Cloning and sequencing of the variant HA gene revealed amino acid substitutions at three positions in the polypeptide. Two substitutions were in antigenic determinants in the globular region of HA1, and the third occurred in HA2 near the base of the molecule. By using chimeric HA molecules expressed in CV-1 cells from simian virus 40-based vectors, we demonstrated that the change in HA2 was solely responsible for the altered fusion phenotype. This substitution, asparagine for aspartic acid at position 132, disrupted a highly conserved interchain salt bridge between adjacent HA2 subunits. The apparent role of this residue in stabilizing the HA trimer is consistent with the idea that the trimer dissociates at low pH. Furthermore, the results demonstrate that influenza virus populations contain fusion variants, raising the possibility that such variants may play a role in the evolution of the virus.     

Length requirements for membrane fusion of influenza virus hemagglutinin peptide linkers to transmembrane or fusion peptide domains

During membrane fusion, the influenza A virus hemagglutinin (HA) adopts an extended helical structure that contains the viral transmembrane and fusion peptide domains at the same end of the molecule. The peptide segments that link the end of this rod-like structure to the membrane-associating domains are approximately 10 amino acids in each case, and their structure at the pH of fusion is currently unknown. Here, we examine mutant HAs and influenza viruses containing such HAs to determine whether these peptide linkers are subject to specific length requirements for the proper folding of native HA and for membrane fusion function. Using pairwise deletions and insertions, we show that the region flanking the fusion peptide appears to be important for the folding of the native HA structure but that mutant proteins with small insertions can be expressed on the cell surface and are functional for membrane fusion. HA mutants with deletions of up to 10 residues and insertions of as many as 12 amino acids were generated for the peptide linker to the viral transmembrane domain, and all folded properly and were expressed on the cell surface. For these mutants, it was possible to designate length restrictions for efficient membrane fusion, as functional activity was observed only for mutants containing linkers with insertions or deletions of eight residues or less. The linker peptide mutants are discussed with respect to requirements for the folding of native HAs and length restrictions for membrane fusion activity.