Biologia 65/2: 183—190, 2010Section Cellular and Molecular BiologyDOI: 10.2478/s11756-010-0003-0

Protein homology modeling and structure-function relationshipof 2009 swine influenza virus hemagglutinin (HA1):more human than swine

Mushtaq Hussain, Rafiq M. Khanani*, Nusrat Jabeen, Shaheen S. Shoaib & Talat Mirza

Department of Molecular Pathology, DDRRL, Dow University of Health Sciences, Karachi, Pakistan;e-mail: r.khanani@duhs.edu.pk

Abstract: The virulence and transmissibility of viruses are highly associated with their binding specificity to the host cellreceptor. In influenza, this initial event of viral pathogenesis is mediated by a glycoprotein known as hemagglutinin (HA).In the present study we constructed homology models of the chain A of hemagglutinin (HA1) of 2009 swine influenza strain.The modeled proteins were compared with atomic coordinates of 1918 (Spanish flu strain) and 1930 HA1 (swine influenzastrain). HA1 of recent swine influenza strain showed 84.83% and 93.14% homology with the same versions of 1918 and1930 strains, respectively. Discrepancies in multiple sequence alignment particularly at the ligand-binding residues notifiedits receptor specificity to α-2,6 sialic acids in 1918 and 2009 viral strains in contrast to α-2,3 sialic acids as found in 1930swine flu strain. This implicated the relatively closer relationship of 2009 strain with 1918 strain rather than swine originstrain of 1930. Similarly, the spatial orientations of receptor-binding residues, located in 190-helix, 130-loop and 220-loop,were found more aligned in 1918 and 2009 (RMSD 0.98 A) than in 1930 and 2009 (RMSD 1.06 A) strains HA1. Moresimilarities were established between both human origin influenza viruses (1918 and 2009 strains) by the receptor-bindingcavity architecture and the orientation of protease cleavage site (Arg327). Briefly, the present finding is expected to showmolecular discrepancies and congruencies among the recent and past pandemic influenza strains and may also potentiallyillustrate the drug targets to rein the infection at earlier stages.

Key words: swine flu; influenza virus; hemagglutinin; virus evolution; genetic reassortment; Mexican flu.

Abbreviations: FD, fusion domain; HA, hemagglutinin; HA1, chain A of hemagglutinin; HA2, chain B of hemagglutinin;RBD, receptor-binding domain; RMSD, root mean square deviation; VED, vestigial esterase domain.

Introduction

Influenza virus, a negative stranded RNA containingvirus of orthomyxoviridae family, is perhaps the bestparadigm of protean virus, whose genome is a subject ofcontinuous changes in terms of genetic shifts and drifts(Reid et al. 1999; Ha et al. 2002). Three major typesof influenza viruses (A, B and C) have been catego-rized on the antigenic nature of their internal proteinsnamely nucleoprotein and matrix protein. However, itis the influenza A virus, which is considered a ma-jor source of human morbidity and mortality (Crosby1989). On the basis of sequential variations and conse-quently antigenic heterogenicities found in two struc-tural proteins, i.e. hemagglutinin (HA) and neurami-nadase, the influenza A virus has been further catego-rized into 16 (H1–H16) and 9 (N1–N9) antigenic types,respectively (Ha et al. 2002; Stevens et al. 2006). How-ever, historically, the H1 strains of influenza A virushave proven themselves as most obnoxious, H1 influenzastrain caused 1918 pandemic, killed 20–40 million peo-

ple in a span of less than a year (Crosby 1989). Evenafter the resurrection of this fearsome 1918 flu virusand the undertaking of its complete genome sequenc-ing, the extensive mortality rate of this virus has re-mained still a mystery (Reid 1999; Stevens et al. 2004).As HA is the primary target of host antibodies (Bushet al. 1999), many studies have been focused to un-ravel the structure and function interrelation ship ofHA protein of 1918 virus in order to decipher the ex-ceptional lethality associated with it (Reid et al. 1999;Gamblin et al. 2004; Stevens et al. 2004). HA is a gly-coprotein, involved in the initial attachment of virus tohuman cells via sialic acids receptors of cell surface gly-colipids and/or glycoproteins (Skehel & Wiley 2000).Furthermore, HA is also responsible for internalizationand membrane fusion events during the infective cycleof the virus (Stevens et al. 2004). The mature HA isa homotrimer, each monomer is synthesized as a singleprecursor polypeptide (HA0), which then gets cleavedby host-coded protease into two chains: the chain A(HA1) and the chain B (HA2) (Wiley & Skehel 1987). It

* Corresponding author

c©2010 Institute of Molecular Biology, Slovak Academy of Sciences

184 M. Hussain et al.

is HA1, which is mainly involved in the receptor bind-ing, whereas HA2 is primarily involved in membranefusion activities (Shangguan et al. 1998). To date struc-tural information of HA of all the three influenza virusesA, B and C have been reported, for instance HAs of hu-man H3 (Wilson et al. 1981), swine H9 (Ha et al. 2001,2002), avian H5 (Ha et al. 2001, 2002; Stevens et al.2006), H1 of 1918 influenza virus (Gamblin et al. 2004;Stevens et al. 2004), HA1 of influenza B (Tung et al.2004; Wang et al. 2008) and HA-esterase fusion pro-tein (Rosenthal et al. 1998). This repertoire of studiesstrongly indicates the importance of HA in the infec-tivity and pathogenesis rendered by the virus.Ironically, the genetic drift and shift are con-

tinuing, and has caused the emergence of a newstrain(s) of H1N1, a swine flu virus that is theetiological agent of the recent pandemic of swineflu. As of 27 November 2009 more than 200 coun-tries have been inflicted with the swine flu includ-ing more than 622,000 confirmed cases and morethan 7,800 deaths (World Health Organization 2009,http://www.who.int/csr/don/2009 11 27a/en/).In the present study, comparative sequential varia-

tions of HA1 of current influenza virus (swine flu virus)have been illustrated with its ancient counterparts, in-fluenza virus strains of 1918 (human influenza) and1930 (swine influenza). The findings have also been ex-tended to the tertiary structure level by constructingthe homology models of HA1 of 2009 influenza virusand comparing it with the same versions found in 1918and 1930 strains. It is expected that the present studywill explicate the receptor specificity, congruencies anddissimilarities in the architectural attributes of HA1 ofrecent influenza strain with its ancient versions. Thestudy could further be exploited to elucidate the poten-tial drug targets and development of vaccine against thecurrent propagating influenza strain. To the best of ourknowledge, this is the first direct report on the struc-tural analysis of HA1 protein of 2009 swine influenzavirus using the protein homology modeling tools.

Methods

Multiple sequence alignmentPrimary structure sequences of HA1 protein of 2009swine flu virus (A/California/04/2009/H1N1; accessionnumber FJ966082.1; protein id. ACP41105.1), 1918 in-fluenza virus (A/South Carolina/1/18; accession numberAF117241.1; protein id. AAD17229.1) and 1930 swine fluvirus (A/swine/Iowa/30; PDB code 1RUY) were retrievedfrom the NCBI (National Center for Biotechnology In-formation; http://www.ncbi.nlm.nih.gov/) server (Wheeleret al. 2005). Primary and tertiary structure homologuesof the above-mentioned proteins were detected using pro-gram FASTA (Pearson 1990) and BLAST (Altschul et al.1997). Multiple sequence alignment was generated by de-fault parameters of the Clustal X program (Thomson et al.1997). With some non-redundant manual modification, thealignment file was analyzed using GeneDoc (Nicolas et al.1997) and visualized by CLC Sequence Viewer 6.0.2 (http://www.clcbio.com/index.php?id=28).

Glycosylation siteGlycosylation sites were predicted using Center for Bio-logical Sequence server (http://www.cbs.dtu.dk/services/)(Hallin & Ussery 2004).

Homology modelingAs templates, the crystal structure coordinates of HA1 of1918 flu virus (PDB code 1RD8) (Stevens et al. 2004) andof 1930 swine flu virus (PDB code 1RUY) (Gamblin etal. 2004) were retrieved from Protein Data Bank (PDB)(Berman et al. 2000). The tertiary structure models of HA1of 2009 swine flu virus were constructed using Geno3D(Combet et al. 2002) and SWISS-MODEL (Schwede et al.2003) with the manual input of PDB code of both the tem-plates.

Tertiary structure analysisThe constructed models of HA1 of 2009 swine flu virus wereviewed by Swiss PDB viewer (Guez et al. 1997) and Ac-celrys Discovery Studio visualizer 2.0 (http://accelrys.com/products/index.html#drugdiscovery). The structural andthermodynamic stability of all models were checked usingSwiss-PDB viewer, PROCHECK, Whatcheck (Laskowaski& Kato 1980), ANOELA (Melo & Feytman 1998) and Ver-ify3D (Elsenberg et al. 1997). Folds in the modeled proteinwere detected from 3D-PSSM algorithm (Kelley et al. 2000).

Result and discussion

Sequence analysisMultiple sequence alignment of 2009 swine flu virusHA1 showed 84.83% and 93.14% homology with thesame proteins found in 1918 and 1930 flu virus, respec-tively. The sequence similarities and/or differences werefound more or less homogenously distributed in all thefunctionally different regions of proteins namely the fu-sion domain (FD), receptor-binding domain (RBD) andvestigial esterase domain (VED). Notably, 58 sites werefound with substitutions across all the three sequenceswhere more precisely at least one sequence showedamino acid substitution (Fig. 1). However, in terms ofbiophysico-chemical characteristics of residues, around50% of these substitutions are iso-functional in nature;they may not thus render significant effect on the pro-tein structure and/or function. For instance, HA1 se-quence alignment of 1918 and 1930 influenza viruses ex-hibited 52 and 43 substitutions, respectively, with HA1of 2009 strain, out of which 24 and 22 were found iso-functional, respectively (Fig. 1). As HA1 is the primarytarget of host antibodies and to date at least 5 epitopeshave been described (Bush et al. 1999). The increas-ing differences among the HA1 protein of 1918 to 1930flu virus and subsequently to 2009 influenza strain maybe inferred in terms of ongoing genetic reassortmentin order to evade the selective pressure exerted by thehost immune arsenals (Reid et al. 1999). Other stud-ies also support this notion where it has been observedthat vaccine efficacy varies substantially from year toyear, depending on the antigenic distances between thecirculating influenza strain and strains used in vaccinepreparation (Deen & Pan 2009).

Structure-function relationship of HA1 185

Fig. 1. Multiple sequence alignment of HA1 of influenza virus strains. The conservation percentage has been deduced in the histogram.Different functional domains, i.e. the fusion domain, vestigial esterase domain and receptor-binding domain are underlined by blue,green and red lines, respectively. Glycosylation sites are indicated by orange rectangles.

On the basis of earlier crystal structure coordinatesof HA1 with ligand, 16 sites have been stipulated to me-diate the receptor binding of HA1 protein of influenzavirus (Reid et al. 1999; Kovacova et al. 2002; Gam-blin et al. 2004; Stevens et al. 2006). Interestingly, theHA1 of both 1918 and 2009 strains were found identi-cal in this connection, however, two substitutions wereobserved when compared with 1930 swine influenzavirus HA1. Briefly, Tyr98, Thr136, Trp153, His183,Asp190, Leu194, Tyr195, Gln196, Glu216, Pro221,Lys222, Gln226, Ala227 and Gly228 were found con-served in all the three proteins. However, variationswere clear at position 193 where both 1918 and 2009HA1 showed presence of serine in contrast to asparaginefound in 1930 strain. Similarly, again in both 1918 and2009 strains the aspartate was found present at the po-sition 225 instead of glycine, which was detected in 1930virus HA1. Such changes are substantially significant todefine the receptor specificity of the virus (this will be

discussed later). However, it is worth mentioning herethat the later substitutions (Asp/Gly 225) have alsobeen noted among the five resurrected strains of 1918influenza with the ratio of 3:2 (Reid et al. 1999; Gam-blin et al. 2004).Both mammalian and avian species have membra-

ne-linked sialic acids, which offer binding sites for HA1of influenza virus. However, subtle differences in thebinding configuration exist between these taxonomicgroups, as birds have α-2,3 linkage (Nobusawa et al.1991), while mammals predominantly have α-2,6 link-age with sialic acid (Rogers & D’Souza 1989; Cou-ceiro et al. 1993). Since both swine and human in-fluenza viruses have been largely proposed of avian ori-gin (Kanegae et al. 1994), cross species transfer essen-tially requires changes in the binding specificity. In thisconnection presence of the Asp/Glu and Asp/Gly atpositions 190 and 225, respectively, has proven vitalfor defining the receptor and consequently host speci-

186 M. Hussain et al.

Fig. 2. Tertiary structure of HA1. (a) Modeled tertiary structure of HA1 of swine influenza virus 2009. Four disulphide bridges areemphasized in red. (b) Stereo view of superimposition of the HA1 of 1918 (green), 1930 (cyan) and 2009 (red) influenza viruses. Otherstructural elements are annotated correspondingly.

ficity of the influenza virus (Reid et al. 1999; Stevenset al. 2006). Depending on the type of HAs, differentmechanisms have been adopted by the human virusesto triumphantly undertake the cross species transmis-sion. Among human viruses with H2 and H3 types HA,Gln226 is substituted with Leu226, while Gly228 sub-stitutes Ser228 (Connor et al. 1994). Conversely, HAs ofhuman H1 retains both Gln226 and Gly228 but possessAsp190 and Asp225 which substantially increases thebinding specificity to α-2,6 ligand (Rogers & D’Souza1989). In the present study it was observed that HA1of 2009 and 1918 influenza strains have Asp190 andAsp225 suggesting their complete specificity to α-2,6linkage which are only found in humans or swine. In-triguingly, in the old classical 1930 swine flu strain,Asp225 is substituted with Gly225. This substitutionmay hamper its specificity to α-2,6 type linkage for α-2,3 type linkage which are abundantly found in birdsenteric tracts (Ito et al. 1998) but are also observedin swine trachea (Rogers and D’Souza 1989). Consid-ering these observations, despite more discrepancies inthe sequence alignment, 2009 swine flu strain is closerto exceptionally virulent 1918 strain in terms of theirreceptor-binding specificity and possibly species trans-mission. As established, genetic reassortment among in-fluenza virus strains of different origins is the majorphenomenon involved in the emergence of new strains ofinfluenza virus, the present findings implicates that cur-rent swine flu virus is genetically well contributed withhuman associated influenza viruses rather than swineorigin viruses. This notion is further supported by theearlier findings, which have concluded on the note thatthe infected swine do not survive to the extent to exertimmune selective pressure on the virus rendering to theslower genetic drift in swine as compared to humans(Sugita et al. 1991).

Furthermore, it has recently been concluded by dif-ferent means that the recent influenza strain is possiblythe product of triple reassortment of two circulatingstrains namely swine influenza that is found prevalentin North America since 1998 and H1N1 strain that is ex-isting in Europe and Asia (Kingsford et al. 2009; Smithet al. 2009).

Glycosylation sitesSome glycosylation sites have been proposed necessaryfor the biological activity of the HA1 (Schulze 1997).With some positional differences, as compared to 5predicted glycosylation sites in both 1918 and 1930influenza strains, 6 such sites were predicted in 2009strain. However, in all proteins only 4 were juried tocross over the threshold level (Fig. 1). Moreover, theadditional glycosylation site of HA1 of 2009 strain wasfound located towards the C-terminal cytoplasmic tail,which is unlikely to be glycosylated because of the re-ducing environment (Stevens et al. 2006). It is sug-gested that increase in the glycosylation sites are theimportant post translation modification, which the HAshave to undergo as a part of adaptation and appar-ently involved in the masking of antigens (Reid et al.1999; Wang et al. 2007). Retaining the same numberof glycosylation sites may also be inferred, as the ad-ditional glycosylation site has no necessity for efficientviral replication in the host at least in humans (Reid etal. 1999).

Overall structureHolistically, the protein structure of HA1 monomercould be segregated into two basic components: (i) theextracellular globular domain, which is mainly involvedin the substrate binding and contributes toward anti-genicity of the protein; and (ii) the stem domain which

Structure-function relationship of HA1 187

Fig. 3. Receptor binding domain of HA1. The residues are shown that define the receptor specificity and binding of HA1 withhuman/swine membrane-bound sialic acid in 1918 (a), 1930 (b) and 2009 (c) influenza virus.

comprises distal cytoplasmic end. The stem domain isconstituted by residues found both in the N-terminusand the distal end of the C-terminus. The overall struc-ture of the modeled HA1 mainly found to contain β-pleated sheets particularly at stem domain. However,intervening 4 α-helices were found in the HA1 of 2009strain. Comparable structural elements were also ob-served in both structures of HA1 of 1918 and 1930influenza strains (Gamblin et al. 2004; Stevens et al.2004). The selected model of 2009 HA1 bears the rootmean square deviation (RMSD) of 1.23 A and 1.29A with the HA1 of 1918 (human influenza virus) and1930 (swine influenza virus), respectively. This implic-itly suggests a closer structural resemblance, at least interms of Cα backbone, between both human-associatedinfluenza viruses (1918 and 2009) as compared to swinestrain of 1930. Four disulfide bridges were observed inHA1 of both 1918 and 1930 influenza virus strains. HA1of 2009 also possessed equivalent number of disulfidebridges with some subtle positional and spatial differ-ences (Fig. 2). The selected tertiary structure model ofHA1 of 2009 influenza strain was deduced to containthe free energy of –14099.80 kcal/mol. Moreover, Ra-machandran plot assessment also referred that 99.6%residues of modeled HA1 were in the acceptable rangeof φ and ψ angles. Ramachandran plot values are oftenconsidered as good indicator of quality of tertiary pro-tein structure (Wilson et al. 1998). Conclusively, con-sidering the size of proteins, the free energy and Ra-machandran plot values strongly imply to the structuraland thermodynamic fidelity of the modeled protein.Additionally, as reflected by the RMSD, more holisticstructural resemblance between HA1 of 1918 and 2009viruses supports the inference drawn using multiple se-quence alignment.

Receptor (ligand) binding sitesThe receptor-binding sites and/or residues of HA1 havebeen reportedly located at the membrane distal tip.Structurally, in HA1 of all three strains (1918, 1930and 2009), the residues responsible for binding with theligand were found more or less confined to three con-served secondary structural elements namely 190 he-

lix (residues 190–198), the 130 loop (residues 135–138)and 220 loop (residues 221–228) (Gamblin et al. 2004;Stevens et al. 2006). In addition to the receptor-bindingresidues found in the mentioned regions, some residueslike Tyr98, Trp153, His183 and Glu216, which were notthe part of these conserved secondary structural ele-ments but considered to be involved in binding withsialic acid molecules, located on the host cell membrane(Gamblin et al. 2004; Stevens et al. 2006). However, it isimportant to note here that spatially all these residueswere situated in a way that they allowed the devel-opment of a central region and/or cavity to surroundthe ligand (Fig. 3). Consistent to the whole structure,spatial orientation of receptor-binding residues of HA1of 2009 and 1918 influenza virus (human) resembledmore to each other than 1930 (swine) HA1. This factwas conspicuously indicated by the superimposition ofthe residues of all three proteins over each other. In-deed, the RMSD deviation between 1918 and 2009 in-fluenza strains was found as 0.98 A compared to 1.29A found between 1930 and 2009 influenza virus strains.Moreover, the cavity as developed by the RBD was alsofound more similar in terms of topology and peripheralelectrostatic potential in HA1 of 1918 and 2009 strains(Fig. 4). Considering the orientations of the residues itcould be inferred that the Asp190 may not be involvedin the binding with the sialic acid, however, its con-servation among all H1 influenza strain (Raymond etal. 1986) implicates its functional role. In the light ofprevious findings (Gamblin et al. 2004) it is likely thatAsp190 and Ser193 of HA1 of 2009 strain may bindwith the amino and hydroxyl termini of sialic acid, re-spectively. Similarly, Lys222 and Asp225 are expectedto form the hydrogen bonds with the hydroxyl of sialicacid molecules. Shortly, the values again suggest thecloser relationship between human origin 1918 and 2009influenza strains and the present findings are in linewith the observations taken from the multiple sequencealignment and overall structural comparison of HA1 of1918, 1930 and 2009 viruses.

Protease cleavage siteAs discussed above the HA0 (precursor of HA) is

188 M. Hussain et al.

Fig. 4. Substrate binding sites of HA1. Electrostatic surface potential and cavity topology of the receptor-binding domain of HA1 of1918 (a), 1930 (b) and 2009 (c) influenza virus. Note the similarities between (a) and (c). For details, please see the text.

Fig. 5. Protease cleavage sites of HA1. The protease cleavage residue Arg 325/327 of HA1 of 1918 (a), 1930 (b) and 2009 (c) influenzavirus is emphasized. Note the similarities in spatial orientations between (a) and (c).

cleaved into two chains HA1 and HA2 by host-codedprotease at the earlier phase of pathogenesis (Wileyet al. 1987). The hydrolysis is stipulated to occurat Arg327/329 however, being spatially influenced byAsn20 and Asn34 (Stevens et al. 2004). Similar to wholestructure and RBD, the spatial orientation of thoseresidues was found more similar in HA1 of 2009 and1918 strains as compared to 2009 and 1930 strains(Fig. 5). This provides another value to the earlier in-ference regarding close relatedness between human as-sociated influenza viruses (1918 and 2009) than swinerelated virus (1930).

Fold recognitionAs anticipated most of the folds present in the HA1 ofswine influenza virus were found similar to folds presentamong other HA1 structures of influenza viruses includ-ing 1918 and 1930 strains. However, quite interestinglywith relatively less sequential identity, some folds re-sembled the glutamate receptor (21%), Epstein-Barrvirus receptor (19%) and cupredoxins (13%). This re-semblance with the unrelated proteins may be inferredin terms of continuous genetic drift and shift in the HA1gene (Reid et al. 1999) and presence of VED (Rosenthalet al. 1998).

ConclusionModeled structure of HA1 of 2009 swine flu virus issubstantially conserved with the crystal structure co-ordinates of HA1 of 1918 and 1930 strains. Multi-ple sequence alignment has shown more or less ho-mogenous conservation both holistically and amongthe parts of FD, RBD and VED in all the com-pared proteins. Importantly, the overall tertiary struc-ture, protease cleavage sites and the RBD of 2009virus HA1 is more similar to 1918 strain in termsof primary and tertiary structure. Briefly, as com-pared to HA1 of 1930 swine influenza virus, 2009influenza virus HA1 has shown greater resemblancewith 1918 strain. This strongly suggests a closer re-lationship of recent circulating flu strain with its 1918(human) counterpart rather than with the swine in-fluenza virus, as generally being translated, unfortu-nately, by the designated misnomer swine flu. How-ever, we understand that the evolution of viruses,particularly members of orthomyxoviradae must betaken holistically instead of a single protein and/orits part. Further studies in this connection withneuraminadase are underway and will be reportedshortly.

Structure-function relationship of HA1 189

References

Altschul S.F., Madden T.L., Schaffer A.A., Zhang J. Zhang Z.,Miller W. & Lipman D.J. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search pro-grams. Nucleic Acids Res. 25: 3389–3402.

Berman H.M., Westbrook J., Feng Z., Gililand G., Bhat T.N.,Weissig H., Shindylaov I.N. & Bourne P.E. 2000. The ProteinData Bank. Nucleic Acid Res. 28: 1235–1242.

Bush R.M., Fitch W.M., Bender C.A. & Cox N.J. 1999. Positiveselection on the H3 hemagglutinin gene of human influenzavirus A. Mol. Biol. Evol. 16: 1457–1465.

Couceiro J.N., Paulson J.C. & Baum L.G. 1993. Influenza virusstrains selectively recognize sialyloligosaccharides on humanrespiratory epithelium; the role of the host cell in selection ofhemagglutinin receptor specificity. Virus Res. 29: 155–165.

Combet C., Jambon M., Deleage G. & Geourjon C. 2002.Geno3D: automatic comparative molecular modeling of pro-tein. 18: 213–214.

Connor R.J., Kawaoka Y., Webster R.G. & Paulson J.C. 1994.Receptor specificity in humans, avian and equine H2 and H3influenza virus isolates. Virology 205: 17–23.

Crosby A.W. 1989. America’s Forgotten Pandemic: The Influenzaof 1918. Cambridge University Press, Cambridge, 337 pp.

Deen M.W. & Pan K. 2009. The epitope regions of H1–subtypeinfluenza A, with application to vaccine efficacy. Protein Eng.Des. Sel. 22: 543–546.

Elsenberg D., Luthy R. & Bowie J.U. 1997. VERIFY3D: assess-ment of protein models with three-dimensional profiles. Meth-ods Enzymol. 277: 396–404.

Gamblin S.J., Haire, L.F., Russell R.J., Stevens D.J., Xiao B.,Ha Y., Vassisht, N., Steinhauer D.A., Daniels R.S., Elliot A.,Wiley D.C. & Skehel J.J. 2004. The structure and receptorbinding properties of the 1918 influenza hemagglutinin. Sci-ence 303: 1838–1842.

Guex N. & Peitsch M.C. 1997. SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein model-ing. Electrophoresis 18: 2714–2723.

Ha Y., Stevens D.J. Skehel J.J. & Wiley D.C. 2001. X-ray struc-tures of H5 avian and H9 swine virus hemagglutinins boundto avian and human receptor analogs. Proc. Natl. Acad. Sci.USA 98: 11181–11186.

Ha Y., Stevens D.J., Skehel J.J. & Wiley D.C. 2002. H5 avian andH9 swine influenza virus hemagglutinin structures: possibleorigin of influenza subtypes. EMBO J. 21: 865–875.

Hallin P.F. & Ussery D.W. 2004. CBS Genome Atlas Database: adynamic storage for bioinformatics results and sequence data.Bioinformatics 20: 3682–3686.

Ito T., Nelson J., Couceiro S.S., Kelm S., Baum G., Krauss S.,Castrucci M.R., Donatelli I., Kida H., Paulson J.C., WebsterR.G. & Kawaoka Y. 1998. Molecular basis for the generationin pigs of influenza A viruses with pandemic potential. J.Virol. 72: 7367–7373.

Kanegae Y., Sugita S., Shortridge K.F., Yoshioka Y. & NeromeK. 1994. Origin and evolutionary pathways of the H1 hemag-glutinin gene of avian, swine and human influenza viruses:Cocirculation of two distinct lineages of swine virus. Arch.Virol. 134: 17–28.

Kelley L.A., MacCallum R.M. & Sternberg M.J.E. 2000. En-hanced genome annotation using structural profiles in theprogram 3D-PSSM. J. Mol. Biol. 299: 499–520.

Kingsford C., Nagarajan N. & Salzberg S.L. 2009. 2009 Swine ori-gin influenza A (H1N1) resembles previous influenza isolates.PloS ONE 4: e6402.

Kovacova A., Ruttkay-Nedecky G., Haverlik I.K. & Janecek S.2002. Sequence similarities and evolutionary relationships ofinfluenza virus A hemagglutinin. Virus Genes 24: 57–63.

Laskowski M.J. & Kato L. 1980. PROCHECK. Annu. Rev.Biochem. 49: 593–626.

Melo F. & Feytmans E. 1998. Assessing protein structures with anon-local atomic interaction energy. J. Mol. Biol. 277: 1141–1152.

Nicholas K.B., Nicholas H.B., Jr. & Deerfield D.W. II. 1997.GeneDoc: analysis and visualization of genetic variation, EM-Bnet News 4 (2): 1–4.

Nobusawa E., Aoyama T., Kato H., Suzuki Y., Tateno Y. &Nakajima K. 1991. Comparison of complete amino acid se-quences and receptor-binding properties among 13 serotypesof hemagglutinins of influenza A viruses. Virology 182: 475–485.

Pearson W.R. 1990. Rapid and sensitive sequence comparisonwith FASTP and FASTA. Methods Enzymol. 183: 63–98.

Raymond F.L., Caton A.J., Cox N.J., Kendal A.P. & BrownleeG.G. 1986. The antigenicity and evolution of influenza H1hemagglutinin from 1950–1957 and 1977–1983: two pathwaysfrom one gene. Virology 148: 275–287.

Reid A.H., Fanning T.G., Hultin J.V. & Taubenbeger J.K. 1999.Origin and evolution of the 1918 ‘Spanish’ influenza virushemagglutinin gene. Proc. Natl. Acad. Sci. USA 96: 1651–1656.

Rogers G.N. & D’Souza B.L. 1989. Receptor binding propertiesof human and animal H1 influenza virus isolates. Virology173: 317–322.

Rosenthal P.B., Zhang X., Formanowki F., Fitz W., Wong C.H.,Meier-Ewert H. Skehel J.J. & Wiley D.C. 1998. Structure ofthe hemagglutinin-esterase fusion glycoprotein of influenza Cvirus. Nature 396: 92–96.

Schulze I.T. 1997. Effects of glycosylation on the properties andfunctions of influenza virus hemagglutinin. J. Infect. Dis. 176:24–28.

Schwede T., Kopp J., Guex N. & Peitsch M.C. 2003. SWISS-MODEL: an automated protein homology-modeling server.Nucleic Acids Res. 31: 3381–3385.

Shangguan T., Siegel D.P., Lear J.D., Axelsen P.H., Alford D. &Bentz J. 1998. Morphological changes and fusogenic activityof influenza virus hemagglutinin. Biophys J. 74: 54–62.

Skehel J. & Wiley D.C. 2000. Receptor binding and membranefusion in virus entry: the influenza hemagglutinin. Annu. Rev.Biochem. 69: 531–569.

Smith G.H., Vijaykrishna D., Bahl J., Lycett S.J., Worobey M.,Pybus O.G., Ma S.K., Cheung C.L., Raghwani J., Bhatt S.,Peiris J.S., Guan Y. & Rambaut A. 2009. Origins and evo-lutionary genomics of the 2009 swine-origin H1N1 influeza Aepidemic. Nature 459: 1122–1125.

Stevens J., Blixt O. Tumpey T.M., Taubenberger J.K., PaulsonJ.C. & Wilson I.A. 2006. Structure and receptor specificityof the hemagglutinin form an H5N1 influenza virus. Science312: 404–410.

Stevens J., Corper A.L., Bsier C.F., Taubenberger J.K., PaleseP. & Wilson I.A. 2004. Structure of the uncleaved human H1hemagglutinin form the extinct 1918 influenza virus. Science312: 404–410.

Sugita S., Yoshioka Y., Itamura S., Kanegae Y., Oguchi K., Go-jobori T., Nerome K. & Oya A. 1991. Molecular evolution ofhemagglutinin genes of H1N1 swine and human influenza Aviruses. J. Mol. Evol. 32: 16–23.

Thompson J.D., Gibson T.J., Plewniak F., Jeanmougin F. & Hig-gins D.G. 1997. The Clustal X windows interface: flexiblestrategies for multiple sequence alignment aided by qualityanalysis tools. Nucleic Acids Res. 25: 4876–4882.

Tung C., Goodman J.L., Lu H. & Macken C.A. 2004. Homologymodel of the structure of influenza B virus HA1. J. Gen. Virol.85: 3249–3259.

Wang Q., Cheng F., Lu M., Tian X. & Ma J. 2008. Crystal struc-ture of unliganded influenza B virus hemagglutinin. J. Virol.82: 3011–3020.

Wang Q., Tian X., Chen X. & Ma J. 2007. Structural basis forreceptor specificity of influenza B virus hemagglutinin. Proc.Natl. Acad. Sci. USA. 104: 16874–16879.

Wheeler D.L., Barrett T., Benson D.A., Bryant S.H., Canese K.,Church D.M., DiCuccio M., Edgar R., Federhen S., HelmbergW., Kenton D.L., Khovayko O., Lipman D.J., Madden T.L.,Maglott D.R., Ostell J., Pontius J.U., Pruitt K.D., SchulerG.D., Schriml L.M., Sequeira E., Sherry S.T., Sirotkin K.,

190 M. Hussain et al.

Starchenko G., Suzek T.O., Tatusov R., Tatusova T.A., Wag-ner L. & Yaschenko E. 2005. Database resources of the Na-tional Center for Biotechnology Information. Nucleic AcidsRes. 33 (Database Issue): D39–D45.

Wiley D.C. & Skehel J.J. 1987. The structure and function ofthe hemagglutinin membrane glycoprotein of influenza virus.Annu. Rev. Biochem. 56: 365–394.

Wilson I.A., Skehel J.J. & Wiley D.C. 1981. Structure of thehemagglutinin membrane glycoprotein of influenza virus at3 A resolution. Nature 289: 366–373.

Wilson K.S., Butterworth S. & Dauter Z. 1998. Who checks thecheckers? Four validation tools applied to eight atomic reso-lution structures. J. Mol. Biol. 276: 417–436.

Received September 2, 2009Accepted December 7, 2009