VIROLOGY

179,

759-767

Molecular

(1990)

Characterization YOSHIHIRO

*St. Jude

of a New Hemagglutinin,

Subtype

KAWAOKA,*r’ SVETLANA YAMNIKOVA,t THOMAS DMITRI K. LVOV,t AND ROBERT G. WEBSTER*

Children’s Research Hospital, Tennessee 38 10 1; and

Department of Virology 1. lvanovsky Institute

t D.

Received

April

H14, of Influenza

A Virus

M. CHAMBERS,*

and Molecular Biology, 332 N. Lauderdale, P. 0. Box 3 18, Memphis, of Virology, Gamaleya Str. 16, 123098 Moscow, USSR

17, 1990; accepted

August

7, 1990

Two influenza A viruses whose hemagglutinin (HA) did not react with any of the reference antisera for the 13 recognized HA subtypes were isolated from mallard ducks in the USSR. Antigenic analysis by hemagglutination inhibition and double immunodiffusion tests showed that the HAS of these viruses are similar to each other but distinct from the HAS of other influenza A viruses. Nucleotide sequence analysis showed that these HA genes differ from each other by only 21 nucleotides. However, they differ from all other HA subtypes at the amino acid level by at least 31 o/o in HAL Thus, we propose that the HAS of these viruses [A/Mallard/Gurjev/263/82 (H14N5) and A/MallardlGurjev/244/82 (H14N6)] belong to a previously unrecognized subtype, and are designated H14. Unlike any other HAS of influenza viruses, the H 14 HAS contained lysine at the cleavage site between HA1 and HA2 instead of arginine. Experimental infection of domestic poultry and ferrets with A/Mallard/Gurjev/263/82 (H14N5) showed that the virus is avirulent for these animals. Based on comparative sequence analysis of different HA genes, it is suggested that differences of 30% or more at the amino acid level in HA1 constitute separate subtypes. Phylogenetic analysis of representatives of each HA subtype showed that H14 is one of the most recently diverged lineages while H8 and H12 branched off early during the evolution of the HA subtypes. These latter two subtypes (H8 and l-l 12) have been isolated very infrequently in recent years, suggesting that these old subtypes may be disappearing frOtT7 the influenza reservoirs in nature. 8 1990 Academic PWSS, IW.

INTRODUCTION

(Lazarowitz and Choppin, 1975; Klenk et al., 1975). Cleavability of the HA is known to be a key feature for virulence of avian influenza viruses (Bosch et a/., 1979). The HA and NA subtypes were historically determined based on the host from which the virus was isolated and the antigenicity in HA inhibition (HI) and NA inhibition (NI) tests (WHO Memorandum, 197 1). However, subsequent immunological and biochemical studies indicated a need to reconsider the subtype designation of the HA and NA of influenza A viruses (WHO Memorandum, 1980). The current HA and NA subtypes were established based on the antigenic relationships among influenza Aviruses in double immunodiffusion tests. This technique was determined to be preferable to HI and NI tests, because double immunodiffusion tests reveal antigenic relationships which may not be apparent by the conventional HI and NI tests (Schild ef al., 1980). During surveillance studies of influenza virus in wild birds in the USSR, influenza A viruses whose HA did not react with any of the reference antisera for the 13 currently recognized HA subtypes [A/Mallard/Gurjev/ 263182 (Mal/263) and A/Mallard/Gurjev/244/82 (Mal/ 244)] were isolated. In the present paper, we describe antigenic and genetic analyses of the HAS of these viruses. The analyses showed that the HAS of these viruses are antigenically and genetically similar to each

Influenza A viruses have been isolated from a variety of animals including humans, pigs, horses, mink, sea mammals, and birds (see Hinshaw, 1981a; Murphy and Webster, 1990, for review). These viruses are classified according to antigenic differences between their glycoproteins-the hemagglutinin (HA) and neuraminidase (NA). Thirteen HA and nine NA subtypes have previously been identified in influenza A viruses. All of the HA subtypes are found in wild waterfowl, whereas only the Hl , H2, H3, H4, H7, Hl 0, and H13 HAS are recognized in mammals. Therefore, it has been suggested that wild waterfowl are an original reservoir of influenza viruses in mammals (Laver and Webster, 1973). Recent evolutionary studies on the influenza A virus genes support this hypothesis (Kawaoka et a/., 198913; Okazaki et al., 1989; Gorman et al., 1990). The HA molecule is responsible for attachment of the virus to a cell surface receptor, sialyloligosaccharide, and for penetration of the virus into the cell during the initial stages of infection [see Wharton et al. (1989) for review]. It is synthesized as a single polypeptide and then cleaved into HA1 and HA2. Cleavage of the HA is necessary for the virus particles to be infectious ’ To whom

requests

for reprints

should

be addressed. 759

0042-6822190

$3.00

CopyrIght 0 1990 by Academic Press, Inc All rights of reproduction I” any form reserved

760

KAWAOKA

other but distinct from the HAS of any other known influenza A viruses. We, therefore, propose that the HAS of Ma11263 and MaI/ belong to a new HA subtype, H14. MATERIALS Viruses

AND

METHODS

and viral RNA

Mali263 and MaI/ viruses were isolated during an outbreak of Clostridium botulinus C infection in wild ducks. MaI/ was isolated from a cloaca1 swab of a sick mallard duck with enteritis and paralyzed legs. MaI/ was isolated from the second passage of a cloaca1 swab of an apparently healthy duck. NA antigen of these viruses was identified as N5 and N6 for MaI/ and Mal/244, respectively. Viruses were grown in 1 l-day-old embryonated chicken eggs and purified by differential sedimentation through 25-70% sucrose gradients in a Beckman SW28 rotor. Virion RNA was isolated by treatment of purified virus with proteinase K and sodium dodecyl sulfate, followed by extraction with phenol:chloroform (1 :l), as described previously (Bean et a/., 1980). Antigenic

of the HA gene

Full-length cDNA was prepared by reverse transcription of virion RNA according to previously described methods (Huddleston and Brownlee, 1982). A 12-base synthetic oligonucleotide primer complementary to the 3’terminus of the negative-strand RNA was phosphorylated with T4 polynucleotide kinase and then used to prime reverse transcription of the total virion RNA in the presence of [32P]dATP. Second-strand DNA synthesis was performed with a phosphorylated 13-base synthetic primer complementary to the 3’end of the cDNA and the Klenow fragment of fscherichia co/i DNA polymerase I. Full-length double-stranded copies of the HA gene were blunt-end ligated into the Pvull site of a plasmid, pATX, obtained from Dr. C. Naeve (St. Jude Children’s Research Hospital, Memphis, Tennessee). Nucleic

Seeburg (1985) using alkali-denatured DNA templates. The sequence corresponding to the cleavage site between HA1 and HA2 was also obtained by direct RNA sequencing by the dideoxynucleotide chain termination method (Sanger et a/., 1977; Air, 1979) using the purified viral RNA as template, and reverse transcriptase. To determine the nucleotide sequence of the Ma11244 HA gene, direct RNA sequencing was performed. A cDNA clone of the Ma11244 HA gene was used to determine the sequence of the 5’ (viral RNA sense) noncoding region. The MaI/ sequence is complete except for the 3’ noncoding region. Oligonucleotide primers complementary to the HA gene segment were synthesized on an Applied Biosystems Model 280A DNA synthesizer by the solid-phase phosphoramidite method. The reaction products were resolved on 6% polyacrylamide-7 n/l urea thin gels containing a 1x to 5X TBE (90 mll/lTris-borate, pH 8.0, 1 mn/l EDTA) gradient. The sequences of the oligonucleotides used as primers will be provided upon request. Nucleotide sequences were analyzed by the maximum parsimony method to determine the minimum number of mutations needed to account for sequence differences (Fitch, 1971).

analysis

HA titration and HI tests were performed with RDEtreated antisera in microtiter plates (Palmer et a/., 1975). Monospecific antiserum to the HA of MaI/ virus was prepared as described (Webster eta/., 1974). Double immunodiffusion tests were done according to Schild et a/. (1980). Cloning

ET AL.

acid sequencing

The nucleotide sequence of the MaI/ HA gene was determined by sequencing both strands of four full-length clones according to the method of Chen and

Experimental

infection

Chickens and ducks were inoculated into the nares and orally with approximately lo6 EID,, of MaI/ virus. Two ferrets were also inoculated intranasally with a similar amount of the virus. Tracheal and cloaca1 swabs were obtained from infected birds 3 days after infection. Nasal washes were obtained from infected ferrets 3 and 6 days after infection. Virus replication was detected by inoculation of these samples into embryonated chicken eggs. RESULTS Antigenic

analysis

To establish the antigenic relationships among Mal/ 263, Mal/244, and other HA subtypes, they were analyzed in HI tests using the monospecific antisera to the HAS of reference viruses for each subtype shown in Table 1. MaI/ and MaI/ viruses did not react with any of the antisera to the reference subtypes of HA. We, therefore, prepared a monospecific antiserum to the isolated HA of MaI/ virus. To prepare the monospecific antiserum, the reassot-tant virus containing the MaI/ HA and the A/Bellamy/42 (Hl Nl) NA was made. This was done to facilitate isolation of the HA molecules, because the H14 HA was stable in 1Yo sodium dodecyl sulfate without reducing agents but the A/Bellamy/42 NA was not. The HI tests showed

H14

HEMAGGLUTININ

TABLE 1 INFLUENZA A VIRUSES USED FOR ANTIGENIC ANALYSIS AND FOR PREPARATION OF MONOSPECIFIC ANTISERA FOR EACH HA SUBTVPE HA subtype Hl

H2 H3

H4 H5 H6 H7 H8 H9 HlO Hll H12 H13

Virus

strain

A/Puerto Rico/8/34 (H 1 N 1) A/Fort Monmouth/1/47 (Hl Nl) A/Swine/Iowa/l 5/30 (H 1 N 1) A/Duck/Alberta/35/76 (Hl Nl) A/Japan/305/57 (H2N2) A/Gull/Maryland/l 9/77 (H2N9) A/Aicht/2/68 (H3N2) A/Equine/Miami/l/63 (H3N8) A/Duck/Ukraine/63 (H3N8) A/Duck/Czechoslovakia/56 (H4N6) WernlSouth Africa761 (H5N3) A/Turkey/Massachusetts/3740/65 (H6N2) A/Shear-water/Australia/l/73 (H6N5) A/Equrne/Prague/l/56 (H7N7) A/FPV/Dutch/27 (H7N7) AJTurkeylOntariolGl 18/68 (H8N4) A/Turkey/Wisconsin/l/66 (H9N2) A/Chicken/Germany/N/49 (H 1 ON7) A/Duck/England/56 (Hl 1 N6) AiDuck/Memphis/546/74 (Hl 1 N9) A/Duck/Alberta/60/76 (H12N6) A/Gull/Maryland/704/77 (H13N6)

that this antiserum reacts with only MaI/ and Mall 244 viruses and did not react with any other reference viruses of the H 1 through H 13 subtypes (data not shown). Although the HI tests are routinely used for HA subtyping, some virus variants whose HAS are slightly different do not react with the reference antiserum to their subtypes in the HI tests. Furthermore, current HA subtypes were established based on the results of the double immunodiffusion tests @child et al., 1980). We therefore examined the cross-reactivity of the monospecific antiserum to Mali263 virus in double immunodiffusion tests. Since the nucleotide sequence analysis (see below) revealed that the deduced amino acid sequence of the MaI/ HA was most closely related to that of H4 viruses and secondly to that of H3 viruses, the cross-reactivity of the antiserum with H3 and H4 viruses of both avian and mammalian origin was examined. Only MaI/ virus produced a precipitation line, whereas the other viruses did not (Fig. 1). Thus, the MaI/ and MaI/ HAS are antigenically distinct from the HAS of any other influenza A viruses. Sequencing analysis To determine the difference between the MaI/ and MaI/ HA genes and other reference strains at the nucleotide and amino acid levels, the nucleotide

761

sequences of these viruses were determined. The sequences have been entered in the GenBank database as accession numbers M35996 and M35997. The MaI/ HA gene consists of 1748 nucleotides and has the potential to encode an unprocessed precursor polypeptide of 568 amino acids. The amino-terminal signal peptide cleavage site has not been identified directly by amino acid sequencing. However, using an algorithm which predicts signalase cleavage sites with 80% accuracy (Von Heijne, 1986) we located the processing site at the carboxyl end of Ser at -1 (Fig. 2). The cleavage site between the HA1 and HA2 subunits of the mature HA protein is predicted to be at Lys-364. This would result in a HA1 polypeptide of 330 amino acids and a HA2 polypeptide of 221 amino acids. There are seven potential glycosylation sites (Asn-XSer/Thr, Struck et al., 1978) in HA1 (Asn-4, -21, -45, -165, -225, and -296) and one in HA2 (Asn-154). The MaI/ HA gene was similar but not identical to Mall 263 HA. A total of 21 nucleotide differences between the MaI/ and MaI/ HA genes was found where comparison was made (nucleotide residues 33-l 748) 2 of which resulted in amino acid differences; these are amino acid residue 13, His + Gln, and 67, Asn + Lys. To examine the phylogenetic relationship between the MaI/ and MaI/ and other HA genes, the deduced amino acid sequences of the HA genes of MaI/ and representative viruses of all other HA subtypes were first aligned (Fig. 2) and their nucleotide sequences were then aligned accordingly. Phylogenetic analysis was done by the maximum parsimony method (Fitch, 1971) using the region between nucleotides 63 and 343 (H3 numbering), because complete

FIG. 1. Double immunodiffusion test with monospecific antiserum to the A/Mallard/Gurjev/263/82 HA. lmmunodiffusion was done in 1% agarose (SeaKem HGT, FMC, Maine) with monospecific antrserum to the Ma17263 HA (n-H14) against purified viruses disrupted In 1% sodium dodecyl sulfate of MaI/ (H 14) A/Aichi/2/68 (H3N2) (H3 Human), A/Duck/Hokkaido/8/80 (H3N8) (H3 Duck), A/Duck/ Czeckoslovakia/56 (H4N6) (H4 Duck), A/Seal/Massachusetts/l 33/ 82 (H4N5) (H4 Seal), and saline.

16 a Hl HZ H3 H4 H5 H7 HlO H13 H14

* * * * TICIGYHANNSTDTVDTIFEKNVAVTHSVNLLEDRHNGKLCKIKGIAPLQLCK ICIGYHANNSTEKVDTNLERNVlVTtL4KDILEKTHNGKLCKLNGIPPLELGD DFPGNDNSTAT~LGHHAVPNGTLVKTITNDCIEPTNATELVOSSSTGKICN-NP~ILDGID

*

MKAKLLVLLYAFVATD MKTIIALSYIF&lL:mAvR'

Hl HZ H3 H4 H5 H7 HlO H13 HI4

* * * * * * CNITGWLLCNPECDSLLPARSWSYIVBTPNSENGACYPGDFIDYEELREQLSSVSSLERFEIFPKESSW-PNHTFNDVTV CSIAGWLLGNPECDRLLSVPEWSYIMIU(ENPRDCLCPPGSR~SSVKHFEKVKILPK-DRW-TQHTTTOGSR CTLIDALLGDPHCDGF-QNETWDLFVERSKA-FSNCYPYDVPDYASLRLLVASSGTLEFINEGF---TL-TGVTQNGGSN CDIINAALCSPGCDR-LDGAEWDVFIBRPTA-VDTCPPFDVPDYQSLRSILANNGKFEFIAEEF---QW-STVKQffiDSG CSVAGLVLGNPMCDEFLNAPEWSYIVEKNNPINGLCYPGDSSTNLFEKIRIIPR-NSWANHDASSCVSS CGLLGTITCP~DQFL-EFSADLIIERREG-NDVCYPGKFVNEEALRQILGSGGIDKETMGF---~-SGIRTNoTTS CHPVGMLICTPVCDPHL-TGTWDTLIBRENA-IAHCYPGATINEEALRQKIMESGGISKMSTGF---TYGSSITSACTTK C~FEG~IGCNPACTSNFGIRE~~~LIED~~~~G~R~GE~LF~GIR~F~RTRLI~-~T~~--G~~L~T~ CHLINGALGSPGCDR-~DTTWDVFIERPTA-VDTCYPFDVPDYQS~SI~SSGSLEFIAEQF---TU-NGVKVWSSS

HZ HZ H3 H4 H5 H7 HlO H13 H14

* * * * * * * SCSHR-GKSSFYRNL~TKD--SYPKLTNSYVWM(GKEVZVLPSSSD~QSLYSNGNA~SVASS~~ ACAVS-GNPSPFRNMVVLTKEGS--DYPVAKGS~SGEQMLIIWmRPIPIDETgQRTL~NVGTYVSVGTSTLNKRST ACKRG-PDSGPFSRLNnuKSGS--T~LNVTM~DNFDKtYIVmnMPSTDQBQTSL~QASGRV~ST~S~TII TCKRG-NVNGFFRQLNWLTKSNG-DAYP~NLT~GDYARtYIUOVtMPSTDTBQTNLYKNSTKTSQTSW ACPHL-GRSSFFRJWUL IKKNN--VYF'TIKRTYN#TNVEDLLIlXGItlHPNDAAt3QAKLYQNLNAyvSV-TSTIXQRSI ACRRS-G-SSFIAEMEYLLSNTDNASFPQMTKSYKNTRRESALIVWOUP1SGSTTgQTKL~SGNLI~GSSKYHQSFV ACMRN-GGDSFYAELKULVSKTKGQNFPqTTNTYRllTDTAEHLIIUBItPlPSSTQffKNDLYGTQSLSISVESSTYQNNFV ACRDNTGTNSFYRNLVVFIKN--RYOVISKTYNNTTGRDVLVLWCrtMPVSVDBTKTLPVNS ACLRG-GRNSFFSRLNWLTKATNGNYGP-INVTK~GSWPSSDNRQTDLYKVATGRVTSDQISIV

Hl H2 H3 H4 H5 H7 HlO H13 H14

* * * * * * * PEIAARPKVKDQHGRHNWUT LLEPGDTIIFEAT~LIhPWYAFALSRGFESGIITSNASMHECNTKCqTPQGSINSN~ PEIATBPKVNG9GGRHEFSIL~~DTINFES~L~EYG~ISKRGSSGIMKTEGTLENCETKaQTPLGAfNTTLP PNIGSRPWVROQSSRISIYWTIVKPGDILVINSN~LIAORGYFKMRTGKSS-IMRSDAPIGTCISECITPNGSIPNDKP PNIGSBmJVRGqSGRISMIVEPGDLIVFNTI~NL~RG~~SQ~STI~TAVPIGSCTSK~TDRGSIST~ PKIATBPKVNOQSGRMEFRI~PNDTISFESTbNFI~EYAYKIVDSAIMRSELEYGNCDTK~TPLVAINSSMP PSPGTRPQINOQSGRIDFMnILDPNDTVTFSFNOAF~SFL-RGKSMGIQSDVQVDANCEGECYHSGGTITSRLP PWG~VNCQSGRIDFHZrrLVQPGDNITFSHL~SRVSKL-TGRDLGIQSEALIDNSCESKCFWRGGSINTKLP LETGVRPGYN~RSWMKIYUSLIHPGEMITFESNGGFLAPRYGYIIEEYGKGRIFQSRIRMSRCNTK~TSVGGINTNRT PNIGSBPRV~SGRISIYLVNPGDSIIFNSI(;NLI~RGH~ISKSTKSTV~SD~IGSCTSPCLTDKGSIQSD~

Hl H2 H3 H4 H5 H7 HlO H13 H14

FQNIHP:TIGECPK&TKLRMVT&NIPSIQY----R FHNVHPLTIGECPKYVKSEKLVIATGLRN'VpQIES----R ~NVNKITYGACPKYVKQNTLKLATGMRNVPEKRT----R FQNISRISIGDCPKYVKQGSI.KIATGMRNIPEk4T----R FHNVHPLTIGECPKYVKSDKLVIATGMRBvpQKtX----R FQNINSRAVGKCPRYVKQESLLIATGMKtivpEPSKKREKR FQNLSPRTVCQCPKYVNQRSLLL4TCMIUWp EWQG---R FQNIDKNAlBDCPKYIKSGQLKL4TGLRNvpAISN----R FQNVSRIAIGNCPKYVKQGSLNIATGHRNIPGKQA----K *

*

*

63

*

137

*

214

*

293

329

*

*

*

*

*

b HI H2 H3 H4 H5 H7 HlO H13 HI4

80

*

*

*

*

*

*

*

*

NF@ESVR#G%?!DYP 160 H5

NKNLE&FID

Hl ~2 ~3 H4 H5 ~7 H10 H13 H14

IqrSE@KLNREK~DGVKLES-MGVYQIUIYSTVASSLVLLVSLGAIS~S-N@SIQCRICI KBEE~BK~E~KG~KLSS-MGVYQI~.AIYATVAGSLSL~I~GISFWM~-~S~CRI~ &&ALNN&-QLKG~RLKS-GYKDWI~~WISFAISCFLLC-~LLGFIHWA~-~IRCNIC-~ I%RD&4INNttFg~QGVKLTQ-GyKDIIPWISFSISCFLLV-ALLL4FILWACQ-K@NIRCQI~ WSE&SKLKPI(EXDGVKWES-MGTYOIhSIYSTVASSLAlAIMTAGLSF-N&LCCRI~ K$&~~IQIDP~LSS-GYKDVI~~FSFGASC~LL-AIAVGLVFI-&I@N~&CTI~X Q'fRE&XLNRLNXNPVKLSS-GYKDIILWFSFGESCFVLL-AVVMGLVFF-CIKX@MRCTI~ ~~AEE~KL~~P~E~DGIK~XSEDN~~KAL~I~~CIA~~~L~GLILSFIM-~CRFNVC~ IYRD&4INN8IKINPVTLTM-GYKDIILWISFSMSCFVFV-ALILGFVLWACQ-N@NIRCQI~1:

*

*

*

*

*

* 221

Hl HZ ~3 H4 H5 H7 HlO H13 H14

FIG. 2. Comparison of the amino acid sequence of the reference influenza A HA subtypes. Amino acids that have been conserved are shaded. The amino acid numbering for the H3 subtype is shown on the right side; separate numbering is used for HA1 (Fig. 2a) and HA2 (Fig. 2b). The dashes are included to give maximum sequence homology. The amino-terminal amino acid of HA1 (actual or predicted) is indicated by an arrow. H 1: A/WSN/33 (Hit1 et al.. 198 1); H2: A/Japan/305/57 (Gething et al., 1980); H3: A/Memphis/l 02172 (Sleigh eta/., 1979); H4: A/Seal/Massachusetts/l 33/82 (Donis et al., 1989); H5: A/Chicken/Pennsylvania/l/83 (Kawaoka et al., 1984); H7: A/FPV/Rostock/34 (Porter et a/., 1979); H 10: A/chicken/Germany/N/49 (Feldmann et a/.. 1988); HI 3: A/Gull/Maryland/704/77 (Chambers et a/., 1989).

H14 TABLE NUCLEOTIDE

HEMAGGLUTININ

2

AND AMINO ACID HOMOLOGIES BETWEEN THE A/MALLARD/ Gu~~~v/263/82 AND OTHER HA SUBTYPES’ ?h homology Nucleotide sequence

Hl H2 H3 H4 H5 H6 H7 H8 H9 HlO Hll H12 H13

Amino acid sequence

HA1

HA2

HA1

HA2

43.5 43.8 54.7 60.6 43.7 32.8 43.5 36.3 35.1 45.2 31.7 31.3 42.7

54.7 56.8 72.1 78.6 56.2 b

36.1 35.5 55.8 68.5 37.0 26.6 36.7 27.7 26.6 39.7 26.6 26.6 34.8

50.0 47.7 82.3 92.3 49.1

a The nucleotide [resrdues 62 acid [residues -5 (HAI) to 221 were compared when complete wise, nucleotrde residues 62 to 90 were compared. lnsertrons calculatron. * No data.

65.9

64.2

56.1

68.2

67.3

52.3

to 1750 (H3 numbering)] and amino (HA2) (H3 numbering)] sequences sequences were available. Other292 and amino acid residues -5 to or deletions were not included for

763

The phylogenetic analysis also shows that H8, H9, and H12 subtypes probably diverged early during the evolution of the HA subtypes of influenza A viruses (Fig. 3). Other HA subtypes including H 14 fall into two groups; the first group consists of H6, H2, H5, Hll, and H13 and the second group consists of Hl , H7, H 10, H3, H4, and H 14. This analysis indicates that the H 14 is one of the most recently diverged subtypes. Since the HA of each subtype contains a unique sequence near the 5’ end (mRNA sense) sequence (Air, 1981) and nucleotide sequence variation within the subtypes is limited in this region (e.g., Donis et al., 1989) we have compared this region of the Ma11263 HA gene with others (Fig. 4). Although the number of nucleotides at the 5’ noncoding region of Ma11263 is the same as that of the H6 HA gene, the following region encoding the signal peptide of MaI/ differs substantially from that of H6 and is more similar to that of H4. Thus, the 5’ sequence of viral RNA encoding the MaV263 HA is unique to this virus. The amino acid sequences at the cleavage site between HA1 and HA2 are known to be important for virulence of avian influenza A viruses (Bosch et al., 1979). All of the virulent avian viruses contain a series

B/Lee Hfl H12

sequences of some of the HA subtypes (H6, H8, H9, H 1 1, H 12) are not yet available and the length of the noncoding region at the 3’ end differs among the HA subtypes (Air, 1981). The degrees of homologies of nucleotide and amino acid sequences between Mal/ 263 and other HA subtypes were also determined (Table 2). These analyses showed that the MaV263 HA gene was most closely related to the H4 HA gene (Fig. 3 and Table 2). Among the H4 HA genes sequenced so far, A/Seal/Massachusetts/l 33/82 (H4N5) showed the highest degree of homology with MaV263 (data not shown). The H2 and H5 HA genes were known to be most closely related to each other among all of the HA subtypes completely sequenced so far (Air, 1981, Kawaoka et al., 1984). The degree of nucleotide (60.6%) and deduced amino acid (68.5%) sequence homologies in HA1 between the MaV263 and AISeaVMassachusetts/l33/82 HA genes were similar to those found between the H2 and H5 HA genes (62.9 and 65.2% for nucleotide and amino acid sequences, respectively). The degree of homology was higher in HA2 than in HAl, as in other HA subtypes. The near identity (92.3% homology) in HA2 between H4 and H 14 subtypes argues for very recent divergence of these subtypes,

H4 I 100 Nucleotide

I changes

H14

FIG. 3. Phylogenetic tree for the Influenza A HA genes. The nucleotide sequences corresponding to resrdues 63-343 (H3 numbering) were analyzed wrth PAUP (D. Swofford), which uses a maxrmum parsimony algorrthm. The tree was rooted to the B/Lee/40 HA gene (Krystal et al., 1982). The number of vanable characters represented is 189, the total tree length IS 1344 nucleotide changes, and the consistency index (proportron of changes due to forward mutations) is 0.448. Horizontal drstance IS proportronal to the minrmum number of nucleotrde distances to join nodes and HA gene sequences. Vertcal lines are of arbrtrary length for spacing branches and labels. The viruses used for the analysis are the same as In Fig. 2 for subtypes Hl, H2, H4, H7, HlO, H13, and H14 and additronally Include the following: H3, A/Archr/2/68 (Verhoeyen et a/., 1980); H6, A/Shearwater/Australra/72 (Arr, 1981); H8, PJTurkeylOregon/71 (Air, 1981); H9, A/Turkey/Wrsconsrn/l/66 (Air, 1981); Hl 1, A/Duck/Memphis/ 546176 (Arr, 1981), H 12, AIDuck/Alberta/60/76 (Air, 1981).

764

KAWAOKA

Hl

AGCAAMGCAGGGGAAAATAAAAACAACCAAA

Ii2 ii3

AGWLAAAGWLGGGGTTATACCATAGACAACAUIAAAAGCAGGa;ATAATTCTATTAATC

H4

ATG AAG GCA MET Lys Ala A ATG MET

AAA Lys RX Ala

CTA Leu ATC Ile

ATG AAG ACC ATC ATT GCT TTG AGC TAC ATT TTC TGT CTG 'XT MET Lys Thr Ile Ile Ala Leu Ser Tyr Ile Phe Cys Leu Ala -GSGGMACA

85 H6 H7

ATG CTA TCA ATC ACG ATT CTG TTT CTG MET Leu Ser Ile Thr Ile Leu Phe Leu AGCAAAAGCAGGGGTATAATTCATCAAA ATG GAA AGA MET Glu Ar6 AGWAAGUdMA ATG ATT GCA MET Ile Ala AGWVLAAGCAGGGGTTACAAA

ATG AAC ACT CAA MET Asn Thr Gln AGCAAAAGCACGGGTCACA ATG GAG AM MET Glu Lys

H6 H9

-GGGAACMCATAGCCAATCAAG

H10

SGGGTCACA ARXUAGC

Hll

-QZGGTCACA

813

AGCAAMGWLGGGGMATATTAACAATCAGMACMACAAG AGCAAAAGCAGGGGAAA

ATC Ile TTC Phe

CTG Leu ATT Ile

GTC Vsl TAT Tyr

CTG Leu CTC Leu

TTA Leu ATT Ile

TAT Tyr CTC Leu

GCA Ala CTG Leu

TTT Phe TTC Phe

GTA GCT ACA Val Ala Thr ACA GCA GTG Thr Ala Val

CTC G4X +CM Leu Gly Gin

GAT Asp MA ArS

CTC Leu ACA Thr ATC Ile

GAC CTT CCA GGA MT Asp Leu Pro Gly Am ATA CXA GAA GGC TCC TCT 1 CAG MT Ile Ala Glu Gly Ser Ser Gin Am GTG ATT GCC CTT GCA ATA ATC AGC Val Ile Ala Leu Ala Ile Ile Ser ATT GTA GTA GCG ATA CTG GCA ACA Ile Val Val Ala Ile Leu Ala Thr

TAC Tyr GTT Val EC Ala

ACA Thr GTC Val GGA Gly

GGG Gly AM Lys AGG Ar6

CTG Leu ATC Ile

GTT Val GCA Ala

CCC Pro ACA Thr

ACA Thr MT Am

MT Am tXA Ala

TTC Phe ATA Ile

GCC Ala GCA Ala

CTT Leu ATG MST

GTG Val CTC Leu

GCA Ala TTG Leu

GTC Val GCG Ala

ATC Ile A@2 Ser

MT Asn

ATG TAC AAA GTA GTA GTA ATA ATT GCG CTC CTT GGA GCA GTG MA a;T MET Tyr Lys Val Val Val Ile Ile Ala Leu Leu Gly Ala Val Lys Gly ATG MET ATG GAA MST Gly

AAG Lys AAA Lys

AM Lys TTC Ftm

GTA Val ATC Ile

CTG Leu ATT Ile

CTT Leu TTG Leu

TTT Phe AGT Ser

GCA Ala ACT Thr

ATG GCT CTC MT GTC ATT GCA ACT TTG ACA MEZT Ala Leu Asn Val Ile Ala Thr Leu Thr ATG ATT GCA CTC ATA TTG GTT GCA CTG GCT CTG AGC CAC ACT GCT TAT TCT 4CAG MET Ile Ala Leu Ile Leu Val Ala Leu Ala Leu Ser His Thr Ala Tyr Ser Gin

GCA Ala GTC Val

ATC Ile TTG Leu

ATC ATC TGT Ile Ile Cya GCA IXA AGC Ala Ala Ser

ATT Ile TTT Phe

CGA Ar& GCA Ala

CTT ATA AGT GTA TGT GTA CAT Leu Ile Ser Val Cys Val His

Experimental

infection

Since the MaV263 and MaV244 viruses were isolated from sick birds, the possibility exists that they may have contributed to the disease signs that were

ACA Thr CAG Gln

ATA Ile ATA Ile

MT CCT Am Pro GGT1 GAC Gly Asp TCT 4GAC Ser Asp KA 1 GAC Ala Asp TAC 1GAT Tyr Asp GCT I GAT Ala Asp CTT 1GAC Leo Asp GCA 1 GAC &la Asp TAT+ GAC 5r Asp GCA 4GAC Ala Asp

GTG Val CAA Gin AAG Lys

Ila ATC Ile ATC Ile

ATA TGC

Cys TGC Cys TGC Cys

AAA Lya AGG ArS

ATT Ile ATA Ile

TGT Cys TGC Cys

MA ATC TGC Lys Ile Qs A&A ATC TGC ArS Ile Cys GAA Glu AAA Lys

ATC Ile ATT Ile

TABLE

is from the 5’end by an arrow. The

3

REPLICATION OF ~MALMRD/GuRJEv/~~~/~~ DOMESTIC POULTRY AND FERRETS Virus

replication

Chickens

VIRUS IN

in Ferrets

Ducks Cloaca

8/10

3/10

T'X Cys TGC Cys

AGA ATA TGC ArS Ile Cys

caused primarily by C. botulinus. Furthermore, their HAS contain Lys instead of Arg at the cleavage site, which is novel and of unknown effect on virus virulence. We have, therefore, examined the pathogenicity of MaV263 virus in avian and mammalian species (i.e., chickens, ducks, and ferrets). None of animals infected with MaV263 virus showed disease signs. In chickens, the virus replicated well in the trachea after oral inoculation (Table 3). By intravenous inoculation, virus was recovered from all of the chickens either from the trachea or cloaca (data not shown). Replication of Mal/ 263 virus in ducks is similar to that of other duck viruses (Webster et a/., 1978); i.e., the virus was recovered only from cloaca1 swabs but not from tracheal swabs. Like other avian influenza A viruses, MaV263

Trachea

TGT Cys TGC Cys

ATC ACA MT GGG ACA ACA GGA MC CCC ATT ATA TGC Ile Thr Am Gly Thr Thr Gly Am Pro Ila Ile Cys

FIG. 4. The nucleotide and amino acid sequences of the 5’(mRNA sense) region of different HA subtypes. The region shown to the first cysteine conserved among all HA subtypes. The amino-terminal amino acid of HA1 (actual or predicted) is indicated viruses shown are the same as those in Fig. 3.

of basic amino acids ending with Arg at this region, whereas the avirulent avian viruses and the viruses from other species contain a single arginine (Webster and Rott, 1987). Comparison of the MaV263 and Mal/ 244 HAS with others at the cleavage site showed that these viruses contain a single Lys instead of Arg. Thus, the MaV263 and MaV244 viruses are the only influenza viruses examined so far that do not contain Arg at the cleavage site between HA1 and HA2. Many amino acid residues are conserved among the HA of different subtypes (Kawaoka et a/., 1984; Feldmann et al., 1988). To extend the analysis of these conserved residues, such residues were highlighted in Fig. 2. Most of the residues which are conserved among the HA molecules of the different subtypes were also conserved in the MaV263 and MaV244 HAS; these include most of the Cys residues, some of Gly and Pro residues, the hydrophobic residues at the amino terminus of the HA2, and the amino acid residues that are in contact with a sialic acid residue of the receptor moiety (Weis et a/., 1988) [Tyr-98, Gly-134, Try-l 53, His-l 83, Gln-190, Leu-194, and Tyr-195 (H3 numbering)]. This indicates that strict structural and presumably functional constraints exist in all the HA molecules examined above.

GCA 4GAC Ala Asp GGG4GAC Gly Asp

GAC AAC AGC ACA GCA ACG CTG TCC Asp Am Ser Thr Ala Thr Leu Cys

ATG GAA ACA AAA GCA ATA ATT GCT GCA CTG CTA ATG GTA ACA GCA QX MET Glu Thr Lys Ala Ila Ile Ala Ala Leu Leu MET Val 'hr Ala Ala

AGEGAAATATCTAGAAATCAAA

812

H14

ET AL.

Trachea o/2

Cloaca

Day 3

Day 6

212

212

o/2

Nore. Tracheal and cloaca1 swabs were obtained from birds 3 days after infection. Nasal washes were obtained fected ferrets 3 and 6 days after infection. The figures give ber of animals from which virus was recovered over the inoculated.

infected from inthe numnumber

H14

HEMAGGLUTININ

virus also showed the ability to experimentally replicate in mammals. Two infected ferrets both shed the virus up to 3 days after infection. These findings indicate that the replicating ability of MaV263 virus is similar to other influenza viruses isolated from ducks (Hinshaw e2 a/., 198 1 b). DISCUSSION In the present study, we have shown that the HAS of MaV263 and Mali244 viruses are antigenically and genetically distinct from those of any other known HA subtypes. The degrees of homology at the nucleotide and amino acid levels between these and the most homologous (H4) HAS were equivalent to those between the HAS of H2 and H5 (Air, 1981; Kawaoka eta/., 1984). Thus, we propose that the MaV263 and MaI/ 244 viruses be the prototype viruses for a new subtype, H14. The current HA and NA subtypes were established based only on antigenic analysis (WHO Memorandum, 1980). We suggest that the time has come to also include nucleotide and amino acid sequence information in determining these subtypes. The recent accumulation of sequence information on the HA genes of the different subtypes allows us to determine the molecular basis of antigenic and genetic divergence. Since HA2 is relatively more conserved among subtypes compared to HAI, and since all the antigenic sites are located in the HA1 (Wiley et al., 198 l), it is reasonable to compare only HA1 sequences. Homology of the amino acid sequence of the HA1 portion within the subtype is at least 80% (Donis et a/., 1989; Kawaoka et al., 1989a), whereas that between the subtypes is maximum 68.5%, when all pair-wise comparisons were made (data not shown). Therefore, viruses that differ by more than 30% at the amino acid level may be considered for a new subtype. Additionally, some influenza A viruses (e.g., H 1 viruses) cross-react with monospecific antiserum to more than one subtype in the HI test (Kawaoka and Webster, unpublished observation), making subtype differentiation based only on serological reactions not completely reliable. Further study is required to understand such cross reactivity at the molecular level. Although both Mali263 and Ma11244 viruses were isolated at the same time in the same region, these viruses differ in their NA antigen. Additionally, two other viruses that reacted with antiserum to MaV263 were isolated from birds in 1983 in the same region (Gurjev, USSR). This indicates multiple viruses of this subtype circulate in wild birds. Determination of the extent of the gene pool of this subtype in the USSR and the United States is in progress.

765

The phylogenetic relationships determined from analysis of the nucleotide sequences of the different HA subtypes are correlated with ecological features of the viruses. Two HA subtypes (H8 and H 12) that apparently branched off early (Fig. 3) are currently rare in wild waterfowl (Hinshaw eta/, 1985; Kawaoka eta/, 1988). It is conceivable that their rarity is due to some selective disadvantage compared to other HA subtypes. The remaining HA subtypes fell into two lineages (Fig. 3). Although the reason for the presence of two lineages is not certain, it may be associated with the existence of two major influenza virus reservoirs in wild waterfowl, i.e., ducks and shorebirds (Hinshaw et al., 1985; Kawaoka et al., 1988). HA subtypes in each lineage that are the youngest in evolutionary terms are currently predominant subtypes in either the shorebird reservoir (H 11 and H 13) or the duck reservoir (H3). Thus, viruses of the two lineages may originally be reservoir-specific but spill over to the other reservoir at some time after their emergence. The MaV263 and MaI/ are the only known influenza viruses that contain Lys at the cleavage site of the HA instead of Arg. Although these viruses were isolated from sick birds, the virus was not the primary disease agent. Experimental infection of Peking ducks and chickens with MaV263 virus indicated that this virus is not highly virulent for domestic poultry. It has been reported that a change from Arg to Lys at the cleavage site of Pr85env protein reduces virulence of murine leukemia virus (Freed and Risser, 1987). However, it is uncertain that the presence of Lys instead of Arg has any effect on virulence of the Ma11263 and MaV244 viruses. Since the HI4 and H4 HAS, which have Arg at this position, have evolved from a common ancestor, it is likely that there may be other H 14 viruses with Arg instead of Lys. The HA has been shown to be one of the determinant genes for host range of influenza A viruses (Rogers et a/., 1983). H 14 is one of several subtypes of influenza A viruses which have so far been detected only in avian species. However, there have been several examples of introduction of influenza A viruses from avian species into mammals. The HA subtypes previously introduced from avian species which have produced major influenza outbreaks in mammals include H 1 (Scholtissek et a/., 1983), H2 (Scholtissek et al., 1978), H3 (Scholtissek et a/., 1978; Kida et a/., 1988) H4 (Hinshaw et al., 1984), H7 (Webster et a/., 1981) and HI 0 (Klingeborn et al., 1985; Feldmann et a/., 1988). We note that with the exception of H2, these subtypes are all of the same lineage as HI4 (Fig. 3). Every subtype of this lineage has caused at least one such outbreak, except, so far, H14. Since MaV263 virus replicated in ferrets, it also has the potential to

766

KAWAOKA

transmit to mammals. It is, however, not certain whether involvement of this particular lineage of subtypes is simply fortuitous, or whether these HA subtypes contain unique structural features not found in other HA subtypes, which makes them peculiarly suited for interspecies transmission in nature. ACKNOWLEDGMENTS We thank Albert Bean, John M. Freeman, Lili Wang, and Scott Krauss for excellent technical assistance; Clayton Naeve and the St. Jude Children’s Research Hospital Molecular Resource Center for preparation of oligonucleotides; Patricia Eddy and the St. Jude Children’s Research Hospital Molecular Biology Computer Facility for computer support; and David Swofford for PAUP software. This work was supported by Public Health Service Contract Al52586 and Public Health Service Research Grants Al-08831, Al29680, and Al-29599 from the National Institute of Allergy and Infectious Diseases; Cancer Center Support (CORE) Grant CA-21 765; and American Lebanese Syrian Associated Charities.

REFERENCES AIR, G. M. (1979). Nucleotide sequence coding for the “signal peptide” and N terminus of the hemagglutinin from an Asian (H2N2) strain of influenza virus. virology 97, 468-472. AIR, G. M. (1981). Sequence relationships among the hemagglutinin genes of 12 subtypes of influenza A virus. Proc. Nat/. Acad. Sci. USA 78,7639-7643. BEAN, W. J.. SRIRAM, G., and WEBSTER, R. G. (1980). Electrophoretic analysis of iodine-labeled influenza virus RNA segments. Anal. Biothem. 102, 228-232. BOSCH, F. X., ORLICH, M., KLENK, H. D., and Roar, R. (1979). The structure of the hemagglutinin, a determinant forthe pathogenicity of influenza viruses. virology 95, 197-207. CHAMBERS, T. M., YAMNIKOVA, S., KAWAOKA, Y., Lvov, D. K., and WEBSTER, R. G. (1989). Antigenic and molecular characterization of subtype H 13 hemagglutinin of influenza virus. Virology 172, 180188. CHEN, E. Y., and SEEEJURG, P. H. (1985). Supercoil sequencing: A fast and simple method for sequencing plasmid DNA. DNA 4, 165170. DONIS, R. O., BEAN, W. J., KAWAOKA, Y., and WEBSTER, R. G. (1989). Distinct lineages of influenza virus H4 hemagglutinin genes in different regions of the world. Virology 169, 408-417. FELDMANN, H., KRETZSCHMAR, E., KLINGEBORN, B., Roar, R., KLENK, H. D., and GARTEN. W. (1988). The structure of serotype H 10 hemagglutinin of influenza A virus: Comparison of an apathogenic avian and a mammalian strain pathogenic for mink. Virology 165, 428-437. FITCH, W. M. (1971). Toward defining the course of evolution: Minimum change for a specific tree topology. Syst. Zoo/. 20,406-416. FREED, E. O., and RISSER, R. (1987). The role of envelope glycoprotein processing in murine leukemia virus infection. J. Viral. 61, 28522856. GETHING. M. I., BYE, I., SKEHEL, J., and WATERFIELD, M. (1980). Cloning and DNA sequence of double-stranded copies of hemagglutinin genes from H2 and H3 strains elucidates antigenic shift and drift in human influenza virus. Nature (London) 287, 301-306. GORMAN, 0. T., BEAN, W. J., UWAOKA, Y., and WEBSTER, R. G. (1990). Evolution of the nucleoprotein gene of influenza A virus. 1. Viral. 64, 1487-l 497. HINSHAW, V. S., BEAN, W. J., WEBSTER, R. G., et a/. (1984). Are seals

ET AL. frequently infected with avian influenza viruses? 1. Viral. 51, 863865. HINSHAW, V. S., WEBSTER, R. G., BEAN, W. I., and SIRIAM, G. (1981 a). The ecology of influenza viruses in ducks and analysis of the influenza viruses with monoclonal antibodies. Comp. Immunol. Microbiol. Infect. Dis. 3, 155-l 64. HINSHAW, V. S., WEBSTER, R. G., EASTERDAY, B. C., and BEAN, W. J. (1981 b). Replication of avian influenza A viruses in mammals. Infect. Immun. 34, 354-361. HINSHAW, V. S., WOOD, J. M., WEBSTER, R. G., DEIEIEL, R., and TURNER, B. (1985). Circulation of influenza viruses and paramyxoviruses in waterfowl: Comparison of different migratory flyways in North America. Bull. WHO 63, 7 1 l-7 19. HITI, A. L., DAVIS, A. R., and NAYAK, D. P. (1981). Complete sequence shows that the hemagglutinin of the HO and H2 subtypes of human influenza virus are closely related. virology 111, 1 13-l 24. HUDDLESTON, J. A., and BROWNLEE, G. G. (1982). The sequence of the nucleoprotein gene of human influenza Avirus strain, A/NT/60/68. Nucleic Acids Res. 10, 1029-l 037. KAWAOKA, Y., BEAN, W. J., and WEBSTER, R. G. (1989a). Evolution of the hemagglutinin of equine H3 influenza viruses. Virology 169, 283-292. KAWAOKA, Y., CHAMBERS, T. M., SIADEN, W. L., and WEBSTER, R. G. (1988). Is the gene pool of influenza viruses in shorebirds and gulls different from that in wild ducks? Virology 163, 247-250. KAWAOKA, Y., KRAUSS, S., and WEBSTER, R. G. (1989b). Avian-to-human transmission of the PBl gene of influenza A viruses in the 1957 and 1968 pandemics. J. Viral. 63,4603-4608. KAWAOKA, Y., NAEVE, C. W., and WEBSTER, R. G. (1984). Is virulence of H5N2 influenza viruses in chickens associated with loss of carbohydrate from the hemagglutinin? Virology 139, 303-316. KIDA, H., SHORTRIDGE, K. F., and WEBSTER, R. G. (1988). Origin of the hemagglutinin gene of H3N2 influenza viruses from pigs in China. Virology 162, 160-l 66. KLENK, H. D., Roar, R., ORLICH, M., and BLODORN, J. (1975). Activation of influenza A viruses by trypsin treatment. Virology 68, 426439. KLINGEBORN, B., ENGLUND, L., Roar, R., JUNTA, N., and ROCKBORN, G. (1985). An avian influenza A virus killing a mammalian speciesthe Mink. Arch. Viral. 86, 347-351. KRYSTAL, M.. ELLIOT, R. M., BENZ, E. W., YOUNG, J. F., and PALESE, P. (1982). Evolution of influenza A and B viruses. Conservation of structural features in the hemagglutinin genes. Proc. Nat/. Acad. Sci. USA 79,4800-4804. LAVER, W. G., and WEBSTER, R. G. (1973). Studies on the origin of pandemic influenza. Ill. Evidence implicating duckand equine influenza viruses as possible progenitors of the Hong Kong strain of human influenza. Virology 51, 383-391. LQAROWITZ, S. G., and CHOPPIN, P. W. (1975). Enhancement of the infectivity of influenza A and B viruses by proteolytic cleavage of the hemagglutinin polypeptide. Virology 68, 440-454. MURPHY, B. R., and WEBSTER, R. G. (1990). Orthomyxoviruses. In “Virology” (B. N. Fields and D. M. Knipe, Eds.), 2nd ed., pp. 10911152. Raven Press, New York. OKAZAKI, K., KAWAOKA. Y., and WEBSTER, R. G. (1989). Evolutionary pathways of the PA genes of influenza A viruses. Virology 172, 601-608. PALMER, D. F., COLEMAN, M. T., DOWDLE, W. R., and SCHILD, G. C. (1975). Advanced laboratory techniques for influenza diagnosis. Immunol. Ser. 6, 51-52. PORTER, A. G., BARBER, C., CAREY, N. H., HALLEWELL, R. A., THRELFALL, G., and EMTAGE, J. S. (1979). Complete nucleotide sequence of an influenza virus hemagglutinin gene from cloned DNA. Nature (London) 282, 471-477.

H14

HEMAGGLUTININ

ROGERS, G. N., PAULSON, J. C., DANIEL& R. S., SKEHEL, J. J., WILSON, I. A., and WILEY, 0. C. (1983). Single amino acid substitution in influenza haemagglutinin change receptor binding specificity. Nature (London) 304, 76-78. SANGER, F. S., NICKLEN, S., and COULSON, A. R. (1977). DNA sequencing with chain-terminating inhibitors. Virology 74, 54635467. SCHILD, G. C., NEWMAN, R. W., WEBSTER, R. G., MAJOR, D., and HINSHAW. V. S. (1980). Antigenic analysis of influenza A virus surface antigens: Considerations for the nomenclature of influenza virus. Arch. Virol. 63, 17 1 - 184. SCHOLTISSEK, C., BURGER, H., BACHMANN, P. A., and HANNOUN, C. (1983). Genetic relatedness of hemagglutinin of the Hl subtype of influenza A vtruses isolated from swine and birds. Virology 129, 521-523. SCHOLTISSEK, C., ROHDE, W., VON HOYNINGEN, V., and ROIT, R. (1978). On the origin of the human influenza virus subtype H2N2 and H3N2. Virology 87, 13-20. SLEIGH, M. J., BOTH, G. W., and BROWNLEE, G. G. (1979). The influenza virus haemagglutinin gene: Cloning and characterisation of a double-stranded DNA copy. Nucleic Acids R&s. 7, 879-893. STRUCK, D. K., LENNARZ, W. J., and BREW, K. (1978). Primary structural requirements for the enzymatic formation of the N-glycosidic bond rn glycoproteln. Studies with alpha-lactal-bumin. /. Biol. Chem. 253, 5786-5794. VERHOEYEN, M., FANG, R., MIN Jou, W., et a/. (1980). Antigenic drift between the haemagglutinin of the Hong Kong influenza strains A/Alchi/2/68 and A/Victoria/3/75. Nature (London) 286, 771-776.

767

VON HEIJNE, G. (1986). A new method for predicting signal sequence cleavage sites. Nucleic Acids Res. 14, 4683-4690. WEBSTER. R. G., HINSHAW, V. S., BEAN. W. J., VAN WYKE, K. L., GERACI, J. R., ST. AUEIIN, D. J., and PETURSON, G. (1981). Characterization of an influenza A virus from seals. Virology 113, 7 12-724. WEBSTER, R. G., ISACHENKO, V. A., and CARTER, M. (1974). A new avian influenza virus from feral birds in the USSR: Recombination in nature? Bull. WHO 51, 325-332. WEBSTER, R. G., and ROTT, R. (1987). Influenza virus A pathogenicity: The pivotal role of hemagglutinin. Ce// 50, 665-666. WEBSTER, R. G., YAKHNO, M. A., HINSHAW, V. S., BEAN, W. J., and MURTI, K. G. (1978). Intestinal influenza: Replication and characterization of influenza viruses in ducks. Virology 84, 268-278. WEIS, W., BROWN, J. H., CUSACK, S., PAULSON, J. C., SKEHEL, J. J., and WILEY, D. C. (1988). Structure of the influenza virus haemagglutinin complexed with its receptor, sialic acid. Nature (London) 333, 426-431. WHARTON, S. A.. WEIS, W., SKEHEL. J. J., and WILEY, D. C. (1989). Structure, function, and antigenicity of the hemagglutinin of influenza virus. In “The Influenza Viruses” (R. M. Krug, Ed.), pp. 153173. Plenum, New York. WHO Memorandum (1971). A revised system of nomenclature for influenza viruses. Bull. WHO 45, 1 19-l 24. WHO Memorandum (1980). A revision of the system of nomenclature viruses. Bull. WHO 58, 585-59 1. WILEY, D. C., WILSON, I. A., and SKEHEL, J. J. (1981). Structural identification of the antibody-binding sites of Hong Kong influenza haemagglutinln and their involvement in antigenic variation. Nature (London) 289, 373-378.