Structure and Function Analysis of an Antibody Recognizing All Influenza A Subtypes.

Influenza virus remains a threat because of its ability to evade vaccine-induced immune responses due to antigenic drift. Here, we describe the isolation, evolution, and structure of a broad-spectrum human monoclonal antibody (mAb), MEDI8852, effectively reacting with all influenza A hemagglutinin (HA) subtypes. MEDI8852 uses the heavy-chain VH6-1 gene and has higher potency and breadth when compared to other anti-stem antibodies. MEDI8852 is effective in mice and ferrets with a therapeutic window superior to that of oseltamivir. Crystallographic analysis of Fab alone or in complex with H5 or H7 HA proteins reveals that MEDI8852 binds through a coordinated movement of CDRs to a highly conserved epitope encompassing a hydrophobic groove in the fusion domain and a large portion of the fusion peptide, distinguishing it from other structurally characterized cross-reactive antibodies. The unprecedented breadth and potency of neutralization by MEDI8852 support its development as immunotherapy for influenza virus-infected humans.

Structural transitions in influenza haemagglutinin at membrane fusion pH.

Infection by enveloped viruses involves fusion of their lipid envelopes with cellular membranes to release the viral genome into cells. For HIV, Ebola, influenza and numerous other viruses, envelope glycoproteins bind the infecting virion to cell-surface receptors and mediate membrane fusion. In the case of influenza, the receptor-binding glycoprotein is the haemagglutinin (HA), and following receptor-mediated uptake of the bound virus by endocytosis1, it is the HA that mediates fusion of the virus envelope with the membrane of the endosome2. Each subunit of the trimeric HA consists of two disulfide-linked polypeptides, HA1 and HA2. The larger, virus-membrane-distal, HA1 mediates receptor binding; the smaller, membrane-proximal, HA2 anchors HA in the envelope and contains the fusion peptide, a region that is directly involved in membrane interaction3. The low pH of endosomes activates fusion by facilitating irreversible conformational changes in the glycoprotein. The structures of the initial HA at neutral pH and the final HA at fusion pH have been investigated by electron microscopy4,5 and X-ray crystallography6-8. Here, to further study the process of fusion, we incubate HA for different times at pH 5.0 and directly image structural changes using single-particle cryo-electron microscopy. We describe three distinct, previously undescribed forms of HA, most notably a 150 Å-long triple-helical coil of HA2, which may bridge between the viral and endosomal membranes. Comparison of these structures reveals concerted conformational rearrangements through which the HA mediates membrane fusion.

Receptor binding and priming of the spike protein of SARS-CoV-2 for membrane fusion.

Infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is initiated by virus binding to the ACE2 cell-surface receptors1-4, followed by fusion of the virus and cell membranes to release the virus genome into the cell. Both receptor binding and membrane fusion activities are mediated by the virus spike glycoprotein5-7. As with other class-I membrane-fusion proteins, the spike protein is post-translationally cleaved, in this case by furin, into the S1 and S2 components that remain associated after cleavage8-10. Fusion activation after receptor binding is proposed to involve the exposure of a second proteolytic site (S2'), cleavage of which is required for the release of the fusion peptide11,12. Here we analyse the binding of ACE2 to the furin-cleaved form of the SARS-CoV-2 spike protein using cryo-electron microscopy. We classify ten different molecular species, including the unbound, closed spike trimer, the fully open ACE2-bound trimer and dissociated monomeric S1 bound to ACE2. The ten structures describe ACE2-binding events that destabilize the spike trimer, progressively opening up, and out, the individual S1 components. The opening process reduces S1 contacts and unshields the trimeric S2 core, priming the protein for fusion activation and dissociation of ACE2-bound S1 monomers. The structures also reveal refolding of an S1 subdomain after ACE2 binding that disrupts interactions with S2, which involves Asp61413-15 and leads to the destabilization of the structure of S2 proximal to the secondary (S2') cleavage site.

Reversible structural changes in the influenza hemagglutinin precursor at membrane fusion pH.

The subunits of the influenza hemagglutinin (HA) trimer are synthesized as single-chain precursors (HA0s) that are proteolytically cleaved into the disulfide-linked polypeptides HA1 and HA2. Cleavage is required for activation of membrane fusion at low pH, which occurs at the beginning of infection following transfer of cell-surface-bound viruses into endosomes. Activation results in extensive changes in the conformation of cleaved HA. To establish the overall contribution of cleavage to the mechanism of HA-mediated membrane fusion, we used cryogenic electron microscopy (cryo-EM) to directly image HA0 at neutral and low pH. We found extensive pH-induced structural changes, some of which were similar to those described for intermediates in the refolding of cleaved HA at low pH. They involve a partial extension of the long central coiled coil formed by melting of the preexisting secondary structure, threading it between the membrane-distal domains, and subsequent refolding as extended helices. The fusion peptide, covalently linked at its N terminus, adopts an amphipathic helical conformation over part of its length and is repositioned and packed against a complementary surface groove of conserved residues. Furthermore, and in contrast to cleaved HA, the changes in HA0 structure at low pH are reversible on reincubation at neutral pH. We discuss the implications of covalently restricted HA0 refolding for the cleaved HA conformational changes that mediate membrane fusion and for the action of antiviral drug candidates and cross-reactive anti-HA antibodies that can block influenza infectivity.

The effect of the D614G substitution on the structure of the spike glycoprotein of SARS-CoV-2.

The majority of currently circulating severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) viruses have mutant spike glycoproteins that contain the D614G substitution. Several studies have suggested that spikes with this substitution are associated with higher virus infectivity. We use cryo-electron microscopy to compare G614 and D614 spikes and show that the G614 mutant spike adopts a range of more open conformations that may facilitate binding to the SARS-CoV-2 receptor, ACE2, and the subsequent structural rearrangements required for viral membrane fusion.

Influenza hemagglutinin membrane anchor.

Viruses with membranes fuse them with cellular membranes, to transfer their genomes into cells at the beginning of infection. For Influenza virus, the membrane glycoprotein involved in fusion is the hemagglutinin (HA), the 3D structure of which is known from X-ray crystallographic studies. The soluble ectodomain fragments used in these studies lacked the "membrane anchor" portion of the molecule. Since this region has a role in membrane fusion, we have determined its structure by analyzing the intact, full-length molecule in a detergent micelle, using cryo-EM. We have also compared the structures of full-length HA-detergent micelles with full-length HA-Fab complex detergent micelles, to describe an infectivity-neutralizing monoclonal Fab that binds near the ectodomain membrane anchor junction. We determine a high-resolution HA structure which compares favorably in detail with the structure of the ectodomain seen by X-ray crystallography; we detect, clearly, all five carbohydrate side chains of HA; and we find that the ectodomain is joined to the membrane anchor by flexible, eight-residue-long, linkers. The linkers extend into the detergent micelle to join a central triple-helical structure that is a major component of the membrane anchor.

Receptor binding by H10 influenza viruses.

H10N8 follows H7N9 and H5N1 as the latest in a line of avian influenza viruses that cause serious disease in humans and have become a threat to public health. Since December 2013, three human cases of H10N8 infection have been reported, two of whom are known to have died. To gather evidence relating to the epidemic potential of H10 we have determined the structure of the haemagglutinin of a previously isolated avian H10 virus and we present here results relating especially to its receptor-binding properties, as these are likely to be major determinants of virus transmissibility. Our results show, first, that the H10 virus possesses high avidity for human receptors and second, from the crystal structure of the complex formed by avian H10 haemagglutinin with human receptor, it is clear that the conformation of the bound receptor has characteristics of both the 1918 H1N1 pandemic virus and the human H7 viruses isolated from patients in 2013 (ref. 3). We conclude that avian H10N8 virus has sufficient avidity for human receptors to account for its infection of humans but that its preference for avian receptors should make avian-receptor-rich human airway mucins an effective block to widespread infection. In terms of surveillance, particular attention will be paid to the detection of mutations in the receptor-binding site of the H10 haemagglutinin that decrease its avidity for avian receptor, and could enable it to be more readily transmitted between humans.

Receptor binding by a ferret-transmissible H5 avian influenza virus.

Cell-surface-receptor binding by influenza viruses is a key determinant of their transmissibility, both from avian and animal species to humans as well as from human to human. Highly pathogenic avian H5N1 viruses that are a threat to public health have been observed to acquire affinity for human receptors, and transmissible-mutant-selection experiments have identified a virus that is transmissible in ferrets, the generally accepted experimental model for influenza in humans. Here, our quantitative biophysical measurements of the receptor-binding properties of haemagglutinin (HA) from the transmissible mutant indicate a small increase in affinity for human receptor and a marked decrease in affinity for avian receptor. From analysis of virus and HA binding data we have derived an algorithm that predicts virus avidity from the affinity of individual HA-receptor interactions. It reveals that the transmissible-mutant virus has a 200-fold preference for binding human over avian receptors. The crystal structure of the transmissible-mutant HA in complex with receptor analogues shows that it has acquired the ability to bind human receptor in the same folded-back conformation as seen for HA from the 1918, 1957 (ref. 4), 1968 (ref. 5) and 2009 (ref. 6) pandemic viruses. This binding mode is substantially different from that by which non-transmissible wild-type H5 virus HA binds human receptor. The structure of the complex also explains how the change in preference from avian to human receptors arises from the Gln226Leu substitution, which facilitates binding to human receptor but restricts binding to avian receptor. Both features probably contribute to the acquisition of transmissibility by this mutant virus.

Receptor binding by an H7N9 influenza virus from humans.

Of the 132 people known to have been infected with H7N9 influenza viruses in China, 37 died, and many were severely ill. Infection seems to have involved contact with infected poultry. We have examined the receptor-binding properties of this H7N9 virus and compared them with those of an avian H7N3 virus. We find that the human H7 virus has significantly higher affinity for α-2,6-linked sialic acid analogues ('human receptor') than avian H7 while retaining the strong binding to α-2,3-linked sialic acid analogues ('avian receptor') characteristic of avian viruses. The human H7 virus does not, therefore, have the preference for human versus avian receptors characteristic of pandemic viruses. X-ray crystallography of the receptor-binding protein, haemagglutinin (HA), in complex with receptor analogues indicates that both human and avian receptors adopt different conformations when bound to human H7 HA than they do when bound to avian H7 HA. Human receptor bound to human H7 HA exits the binding site in a different direction to that seen in complexes formed by HAs from pandemic viruses and from an aerosol-transmissible H5 mutant. The human-receptor-binding properties of human H7 probably arise from the introduction of two bulky hydrophobic residues by the substitutions Gln226Leu and Gly186Val. The former is shared with the 1957 H2 and 1968 H3 pandemic viruses and with the aerosol-transmissible H5 mutant. We conclude that the human H7 virus has acquired some of the receptor-binding characteristics that are typical of pandemic viruses, but its retained preference for avian receptor may restrict its further evolution towards a virus that could transmit efficiently between humans, perhaps by binding to avian-receptor-rich mucins in the human respiratory tract rather than to cellular receptors.

Structures of complexes formed by H5 influenza hemagglutinin with a potent broadly neutralizing human monoclonal antibody.

H5N1 avian influenza viruses remain a threat to public health mainly because they can cause severe infections in humans. These viruses are widespread in birds, and they vary in antigenicity forming three major clades and numerous antigenic variants. The most important features of the human monoclonal antibody FLD194 studied here are its broad specificity for all major clades of H5 influenza HAs, its high affinity, and its ability to block virus infection, in vitro and in vivo. As a consequence, this antibody may be suitable for anti-H5 therapy and as a component of stockpiles, together with other antiviral agents, for health authorities to use if an appropriate vaccine was not available. Our mutation and structural analyses indicate that the antibody recognizes a relatively conserved site near the membrane distal tip of HA, near to, but distinct from, the receptor-binding site. Our analyses also suggest that the mechanism of infectivity neutralization involves prevention of receptor recognition as a result of steric hindrance by the Fc part of the antibody. Structural analyses by EM indicate that three Fab fragments are bound to each HA trimer. The structure revealed by X-ray crystallography is of an HA monomer bound by one Fab. The monomer has some similarities to HA in the fusion pH conformation, and the monomer's formation, which results from the presence of isopropanol in the crystallization solvent, contributes to considerations of the process of change in conformation required for membrane fusion.

Recent evolution of equine influenza and the origin of canine influenza.

In 2004 an hemagglutinin 3 neuraminidase 8 (H3N8) equine influenza virus was transmitted from horses to dogs in Florida and subsequently spread throughout the United States and to Europe. To understand the molecular basis of changes in the antigenicity of H3 hemagglutinins (HAs) that have occurred during virus evolution in horses, and to investigate the role of HA in the equine to canine cross-species transfer, we used X-ray crystallography to determine the structures of the HAs from two antigenically distinct equine viruses and from a canine virus. Structurally all three are very similar with the majority of amino acid sequence differences between the two equine HAs located on the virus membrane-distal molecular surface. HAs of canine viruses are distinct in containing a Trp-222 → Leu substitution in the receptor binding site that influences specificity for receptor analogs. In the fusion subdomain of canine and recent equine virus HAs a unique difference is observed by comparison with all other HAs examined to date. Analyses of site-specific mutant HAs indicate that a single amino acid substitution, Thr-30 → Ser, influences interactions between N-terminal and C-terminal regions of the subdomain that are important in the structural changes required for membrane fusion activity. Both structural modifications may have facilitated the transmission of H3N8 influenza from horses to dogs.

Recovery of an antiviral antibody response following attrition caused by unrelated infection.

The homeostatic mechanisms that regulate the maintenance of immunological memory to the multiple pathogen encounters over time are unknown. We found that a single malaria episode caused significant dysregulation of pre-established Influenza A virus-specific long-lived plasma cells (LLPCs) resulting in the loss of Influenza A virus-specific Abs and increased susceptibility to Influenza A virus re-infection. This loss of LLPCs involved an FcγRIIB-dependent mechanism, leading to their apoptosis. However, given enough time following malaria, the LLPC pool and humoral immunity to Influenza A virus were eventually restored. Supporting a role for continuous conversion of Influenza A virus-specific B into LLPCs in the restoration of Influenza A virus immunity, B cell depletion experiments also demonstrated a similar requirement for the long-term maintenance of serum Influenza A virus-specific Abs in an intact LLPC compartment. These findings show that, in addition to their established role in the anamnestic response to reinfection, the B cell pool continues to be a major contributor to the maintenance of long-term humoral immunity following primary Influenza A virus infection, and to the recovery from attrition following heterologous infection. These data have implications for understanding the longevity of protective efficacy of vaccinations in countries where continuous infections are endemic.

Evolution of the receptor binding properties of the influenza A(H3N2) hemagglutinin.

The hemagglutinin (HA) of influenza A(H3N2) virus responsible for the 1968 influenza pandemic derived from an avian virus. On introduction into humans, its receptor binding properties had changed from a preference for avian receptors (α2,3-linked sialic acid) to a preference for human receptors (α2,6-linked sialic acid). By 2001, the avidity of human H3 viruses for avian receptors had declined, and since then the affinity for human receptors has also decreased significantly. These changes in receptor binding, which correlate with increased difficulties in virus propagation in vitro and in antigenic analysis, have been assessed by virus hemagglutination of erythrocytes from different species and quantified by measuring virus binding to receptor analogs using surface biolayer interferometry. Crystal structures of HA-receptor analog complexes formed with HAs from viruses isolated in 2004 and 2005 reveal significant differences in the conformation of the 220-loop of HA1, relative to the 1968 structure, resulting in altered interactions between the HA and the receptor analog that explain the changes in receptor affinity. Site-specific mutagenesis shows the HA1 Asp-225→Asn substitution to be the key determinant of the decreased receptor binding in viruses circulating since 2005. Our results indicate that the evolution of human influenza A(H3N2) viruses since 1968 has produced a virus with a low propensity to bind human receptor analogs, and this loss of avidity correlates with the marked reduction in A(H3N2) virus disease impact in the last 10 y.

A novel I221L substitution in neuraminidase confers high-level resistance to oseltamivir in influenza B viruses.

UNLABELLED: Influenza B viruses with a novel I221L substitution in neuraminidase (NA) conferring high-level resistance to oseltamivir were isolated from an immunocompromised patient after prolonged oseltamivir treatment. METHODS: Enzymatic characterization of the NAs (Km, Ki) and the in vitro fitness of viruses carrying wild-type or mutated (I221L) NA genes were evaluated. Proportions of wild-type and mutated NA genes were directly quantified in the patient samples. Structural characterizations by X-ray crystallography of a wild-type and I221L variant NA were performed. RESULTS: The Km and Ki revealed that the I221L variant NA had approximately 84 and 51 times lower affinity for oseltamivir carboxylate and zanamivir, respectively, compared with wild-type NA. Viruses with a wild-type or I221L variant NA had similar growth kinetics in Madin-Darby canine kidney (MDCK) cells, and 5 passages in MDCK cells revealed no reversion of the I221L substitution. The crystal structure of the I221L NA and oseltamivir complex showed that the leucine side chain protrudes into the hydrophobic pocket of the active site that accommodates the pentyloxy substituent of oseltamivir. CONCLUSIONS: Enzyme kinetic and NA structural analyses provide an explanation for the high level of resistance to oseltamivir while retaining good fitness of viruses carrying I221L variant NA.

H1N1 2009 pandemic influenza virus: resistance of the I223R neuraminidase mutant explained by kinetic and structural analysis.

Two classes of antiviral drugs, neuraminidase inhibitors and adamantanes, are approved for prophylaxis and therapy against influenza virus infections. A major concern is that antiviral resistant viruses emerge and spread in the human population. The 2009 pandemic H1N1 virus is already resistant to adamantanes. Recently, a novel neuraminidase inhibitor resistance mutation I223R was identified in the neuraminidase of this subtype. To understand the resistance mechanism of this mutation, the enzymatic properties of the I223R mutant, together with the most frequently observed resistance mutation, H275Y, and the double mutant I223R/H275Y were compared. Relative to wild type, K(M) values for MUNANA increased only 2-fold for the single I223R mutant and up to 8-fold for the double mutant. Oseltamivir inhibition constants (K(I)) increased 48-fold in the single I223R mutant and 7500-fold in the double mutant. In both cases the change was largely accounted for by an increased dissociation rate constant for oseltamivir, but the inhibition constants for zanamivir were less increased. We have used X-ray crystallography to better understand the effect of mutation I223R on drug binding. We find that there is shrinkage of a hydrophobic pocket in the active site as a result of the I223R change. Furthermore, R223 interacts with S247 which changes the rotamer it adopts and, consequently, binding of the pentoxyl substituent of oseltamivir is not as favorable as in the wild type. However, the polar glycerol substituent present in zanamivir, which mimics the natural substrate, is accommodated in the I223R mutant structure in a similar way to wild type, thus explaining the kinetic data. Our structural data also show that, in contrast to a recently reported structure, the active site of 2009 pandemic neuraminidase can adopt an open conformation.

Crystal structures of oseltamivir-resistant influenza virus neuraminidase mutants.

The potential impact of pandemic influenza makes effective measures to limit the spread and morbidity of virus infection a public health priority. Antiviral drugs are seen as essential requirements for control of initial influenza outbreaks caused by a new virus, and in pre-pandemic plans there is a heavy reliance on drug stockpiles. The principal target for these drugs is a virus surface glycoprotein, neuraminidase, which facilitates the release of nascent virus and thus the spread of infection. Oseltamivir (Tamiflu) and zanamivir (Relenza) are two currently used neuraminidase inhibitors that were developed using knowledge of the enzyme structure. It has been proposed that the closer such inhibitors resemble the natural substrate, the less likely they are to select drug-resistant mutant viruses that retain viability. However, there have been reports of drug-resistant mutant selection in vitro and from infected humans. We report here the enzymatic properties and crystal structures of neuraminidase mutants from H5N1-infected patients that explain the molecular basis of resistance. Our results show that these mutants are resistant to oseltamivir but still strongly inhibited by zanamivir owing to an altered hydrophobic pocket in the active site of the enzyme required for oseltamivir binding. Together with recent reports of the viability and pathogenesis of H5N1 (ref. 7) and H1N1 (ref. 8) viruses with neuraminidases carrying these mutations, our results indicate that it would be prudent for pandemic stockpiles of oseltamivir to be augmented by additional antiviral drugs, including zanamivir.

A human antibody recognizing a conserved epitope of H5 hemagglutinin broadly neutralizes highly pathogenic avian influenza H5N1 viruses.

Influenza A virus infection is a persistent threat to public health worldwide due to its ability to evade immune surveillance through rapid genetic drift and shift. Current vaccines against influenza A virus provide immunity to viral isolates that are similar to vaccine strains. High-affinity neutralizing antibodies against conserved epitopes could provide immunity to diverse influenza virus strains and protection against future pandemic viruses. In this study, by using a highly sensitive H5N1 pseudotype-based neutralization assay to screen human monoclonal antibodies produced by memory B cells from an H5N1-infected individual and molecular cloning techniques, we developed three fully human monoclonal antibodies. Among them, antibody 65C6 exhibited potent neutralization activity against all H5 clades and subclades except for subclade 7.2 and prophylactic and therapeutic efficacy against highly pathogenic avian influenza H5N1 viruses in mice. Studies on hemagglutinin (HA)-antibody complexes by electron microscopy and epitope mapping indicate that antibody 65C6 binds to a conformational epitope comprising amino acid residues at positions 118, 121, 161, 164, and 167 (according to mature H5 numbering) on the tip of the membrane-distal globular domain of HA. Thus, we conclude that antibody 65C6 recognizes a neutralization epitope in the globular head of HA that is conserved among almost all divergent H5N1 influenza stains.

The structure of H5N1 avian influenza neuraminidase suggests new opportunities for drug design.

The worldwide spread of H5N1 avian influenza has raised concerns that this virus might acquire the ability to pass readily among humans and cause a pandemic. Two anti-influenza drugs currently being used to treat infected patients are oseltamivir (Tamiflu) and zanamivir (Relenza), both of which target the neuraminidase enzyme of the virus. Reports of the emergence of drug resistance make the development of new anti-influenza molecules a priority. Neuraminidases from influenza type A viruses form two genetically distinct groups: group-1 contains the N1 neuraminidase of the H5N1 avian virus and group-2 contains the N2 and N9 enzymes used for the structure-based design of current drugs. Here we show by X-ray crystallography that these two groups are structurally distinct. Group-1 neuraminidases contain a cavity adjacent to their active sites that closes on ligand binding. Our analysis suggests that it may be possible to exploit the size and location of the group-1 cavity to develop new anti-influenza drugs.

Haemagglutinin mutations responsible for the binding of H5N1 influenza A viruses to human-type receptors.

H5N1 influenza A viruses have spread to numerous countries in Asia, Europe and Africa, infecting not only large numbers of poultry, but also an increasing number of humans, often with lethal effects. Human and avian influenza A viruses differ in their recognition of host cell receptors: the former preferentially recognize receptors with saccharides terminating in sialic acid-alpha2,6-galactose (SAalpha2,6Gal), whereas the latter prefer those ending in SAalpha2,3Gal (refs 3-6). A conversion from SAalpha2,3Gal to SAalpha2,6Gal recognition is thought to be one of the changes that must occur before avian influenza viruses can replicate efficiently in humans and acquire the potential to cause a pandemic. By identifying mutations in the receptor-binding haemagglutinin (HA) molecule that would enable avian H5N1 viruses to recognize human-type host cell receptors, it may be possible to predict (and thus to increase preparedness for) the emergence of pandemic viruses. Here we show that some H5N1 viruses isolated from humans can bind to both human and avian receptors, in contrast to those isolated from chickens and ducks, which recognize the avian receptors exclusively. Mutations at positions 182 and 192 independently convert the HAs of H5N1 viruses known to recognize the avian receptor to ones that recognize the human receptor. Analysis of the crystal structure of the HA from an H5N1 virus used in our genetic experiments shows that the locations of these amino acids in the HA molecule are compatible with an effect on receptor binding. The amino acid changes that we identify might serve as molecular markers for assessing the pandemic potential of H5N1 field isolates.

Structures of receptor complexes formed by hemagglutinins from the Asian Influenza pandemic of 1957.

The viruses that caused the three influenza pandemics of the twentieth century in 1918, 1957, and 1968 had distinct hemagglutinin receptor binding glycoproteins that had evolved the capacity to recognize human cell receptors. We have determined the structure of the H2 hemagglutinin from the second pandemic, the "Asian Influenza" of 1957. We compare it with the 1918 "Spanish Influenza" hemagglutinin, H1, and the 1968 "Hong Kong Influenza" hemagglutinin, H3, and show that despite its close overall structural similarity to H1, and its more distant relationship to H3, the H2 receptor binding site is closely related to that of H3 hemagglutinin. By analyzing hemagglutinins of potential H2 avian precursors of the pandemic virus, we show that the human receptor can be bound by avian hemagglutinins that lack the human-specific mutations of H2 and H3 pandemic viruses, Gln-226Leu, and Gly-228Ser. We show how Gln-226 in the avian H2 receptor binding site, together with Asn-186, form hydrogen bond networks through bound water molecules to mediate binding to human receptor. We show that the human receptor adopts a very similar conformation in both human and avian hemagglutinin-receptor complexes. We also show that Leu-226 in the receptor binding site of human virus hemagglutinins creates a hydrophobic environment near the Sia-1-Gal-2 glycosidic linkage that favors binding of the human receptor and is unfavorable for avian receptor binding. We consider the significance for the development of pandemics, of the existence of avian viruses that can bind to both avian and human receptors.