Journal of General Virology (2012), 93, 2215–2226 DOI 10.1099/vir.0.

044032-0

Antigenic analysis of highly pathogenic avian

influenza virus H5N1 sublineages co-circulating in
Egypt
Yohei Watanabe,1 Madiha S. Ibrahim,2 Hany F. Ellakany,3
Norihito Kawashita,4,5 Tomo Daidoji,6,7 Tatsuya Takagi,4,5
Teruo Yasunaga,5 Takaaki Nakaya6,7,8 and Kazuyoshi Ikuta1
Correspondence 1
Department of Virology, Research Institute for Microbial Diseases, Osaka University,
Yohei Watanabe 3-1 Yamadaoka, Suita, Osaka 565-0871, Japan
[email protected] 2
Department of Microbiology and Immunology, Faculty of Veterinary Medicine,
Damanhour University, Egypt
3
Department of Poultry Diseases and Hygiene, Faculty of Veterinary Medicine,
Damanhour University, Egypt
4
Department of Environmental Pharmacometrics, Graduate School of Pharmaceutical Sciences,
Osaka University, 1-6 Yamadaoka, Suita, Osaka 565-0871, Japan
5
Genome Information Research Center, Research Institute for Microbial Disease, Osaka University,
3-1 Yamadaoka, Suita, Osaka 565-0871, Japan
6
Department of Infectious Diseases, Kyoto Prefectural School of Medicine,
465 Kawaramachi-hirokoji, Kamigyo-ku, Kyoto 602-8566, Japan
7
International Research Center for Infectious Diseases, Research Institute for Microbial Diseases,
Osaka University, 3-1 Yamadaoka, Suita, Osaka 565-0871, Japan
8
Department of Infection Metagenomics, Research Institute for Microbial Diseases,
Osaka University, 3-1 Yamadaoka, Suita, Osaka 565-0871, Japan

Highly pathogenic avian influenza virus H5N1 has spread across Eurasia and Africa, and
outbreaks are now endemic in several countries, including Indonesia, Vietnam and Egypt.
Continuous circulation of H5N1 virus in Egypt, from a single infected source, has led to significant
genetic diversification with phylogenetically separable sublineages, providing an opportunity to
study the impact of genetic evolution on viral phenotypic variation. In this study, we analysed the
phylogeny of H5 haemagglutinin (HA) genes in influenza viruses isolated in Egypt from 2006 to
2011 and investigated the effect of conserved amino acid mutations in the HA genes in each of
the sublineages on their antigenicity. The analysis showed that viruses in at least four sublineages
still persisted in poultry in Egypt as of 2011. Using reverse genetics to generate HA-reassortment
viruses with specific HA mutations, we found antigenic drift in the HA in two influenza virus
sublineages, compared with the other currently co-circulating influenza virus sublineages in Egypt.
Moreover, the two sublineages with significant antigenic drift were antigenically distinguishable.
Our findings suggested that phylogenetically divergent H5N1 viruses, which were not
antigenically cross-reactive, were co-circulating in Egypt, indicating that there was a problem in
Received 27 April 2012 using a single influenza virus strain as seed virus to produce influenza virus vaccine in Egypt and
Accepted 10 July 2012 providing data for designing more efficacious control strategies in H5N1-endemic areas.

INTRODUCTION
The GenBank/EMBL/DDBJ accession numbers for representative Since the emergence of highly pathogenic avian influenza
sequence data determined in this study are AB496981–AB497040, virus (HPAI) subtype H5N1 in China in 1996, the H5N1
AB498019–AB498038 and AB551129–AB551136. virus has evolved to form 10 phylogenetically distinct
Three supplementary figures are available with the online version of this clades (0–9) (WHO/OIE/FAO H5N1 Evolution Working
paper. Group, 2008), spread into South-East and East Asia, and
Downloaded from www.microbiologyresearch.org by
044032 G 2012 SGM Printed in Great Britain 2215
IP: 49.149.78.196
On: Fri, 16 Nov 2018 14:28:06
Y. Watanabe and others

caused an epidemic in poultry and occasional infections in a2,3-linked sialylglycan (Couceiro et al., 1993; Ito et al.,
mammals (WHO, 2012c). In April 2005, HPAI H5N1 1998). HA is also an antigenically variable protein in which
viruses caused large outbreaks in wild waterfowl at Qinghai a large number of point mutations accumulate, mainly in
Lake in China (Chen et al., 2005) and one genotype, clade HA1 epitope regions (Nelson & Holmes, 2007).
2.2, unexpectedly spread west to central and southern Asia,
Interestingly, H5N1 was introduced into Egypt from a
Europe and Africa, including Egypt (Salzberg et al., 2007;
single infected source (Eladl et al., 2011; Watanabe et al.,
Wang et al., 2008). Since the initial outbreaks in Egypt, a
2011b). In contrast, in other countries, such as China and
distinct third-order H5N1 clade, clade 2.2.1, has evolved
Nigeria, co-circulation of different H5 sublineages has
in Egypt and diverged further into phylogenetically sepa-
rate branches within clade 2.2.1 (Cattoli et al., 2009). By allowed antigenic shift due to genetic reassortment among
early 2009, none of these new sublineages had become the sublineages (Chen et al., 2006; Fusaro et al., 2010).
dominant and all of the sublineages continued to co- Continuous replication of H5N1 virus in Egypt during the
circulate in birds in Egypt (Abdel-Moneim et al., 2009; last 5 years has provided an opportunity to study the
Arafa et al., 2010). relationship between genetic evolution and selection of
influenza virus phenotypes, including antigenicity, receptor-
Clade 2.2 H5N1 virus was first isolated in poultry in Egypt binding specificity and pathogenicity. Although previous
in February 2006, possibly after its introduction from in- studies focused on genetic evolution of influenza viruses,
fected migrating ducks (Saad et al., 2007). Thereafter, HPAI there are relatively few reports analysing the effect of genetic
H5N1 spread swiftly nationwide among birds, including evolution on biological characters (Cattoli et al., 2011a).
chickens, ducks, turkeys, geese and quail, and was declared
endemic in Egypt in July 2008. Other countries with endemic Several commercial inactivated H5 vaccines, produced
HPAI H5N1 are Indonesia, China and Vietnam (OIE, 2011). using different H5 virus strains, were used during the H5N1
To control and attempt to eradicate H5N1 viruses, the epidemic in Egypt (Peyre et al., 2009). However, mass
Egyptian authorities used a blanket vaccination programme vaccination has failed to control the continual H5N HPAI
and attempted to heighten biosecurity and quarantine mea- outbreaks in Egypt (Hafez et al., 2010). The vaccination
sures in both commercial and household sectors (Peyre et al., campaign limited the first wave of 2006 outbreaks. However,
2009). Nevertheless, HPAI H5N1 is still endemic in Egypt antigenic variants were detected in several vaccinated farms
and continues to pose a severe threat to the poultry industry in 2007 and are now the dominant strains in vaccinated and
(Hafez et al., 2010), causing more than a US$ 1 billion non-vaccinated flocks in Egypt (Abdelwhab & Hafez, 2011a).
annual loss (Meleigy, 2007). In addition, among countries Several studies have suggested that immune pressure due
surveyed by the WHO, Egypt has the second-highest number to the vaccines resulted in major antigenic drift of the
of human H5N1 infections (WHO, 2012a). As of April 2012, H5N1 virus, generating phylogenetically distinct clade 2.2.1
166 HPAI H5N1 cases, with 59 fatalities, have been reported variants (denoted here as sublineage C) (Cattoli et al., 2011b;
in Egypt. In particular, the cumulative number since 2009 Eladl et al., 2011). There are conflicting data on whether
is notable: of 249 HPAI H5N1 cases worldwide, 123 cases commercially available vaccines provide protection against
(49 % of the total) were in Egypt. Most human infections these new antigenic drift variants. Some studies have
were linked to close contact with and/or slaughtering of reported inadequate protection (Abdelwhab et al., 2011b;
infected birds and no sustained human–human transmis- Grund et al., 2011; Peyre et al., 2009), while others have
sion has been documented to date in Egypt (WHO, 2012b, reported sufficient protection (Kim et al., 2010; Terregino
c). However, the long-term endemic status of HPAI H5N1 et al., 2010). These discrepancies were probably due to
in Egypt could increase the opportunity for emergence of selection of strains in those studies with phylogenetically
potential pandemic strains through intra- and interspecies discordant or unrepresentative sequences. In addition,
transmission. polyclonal antibodies in infected chicken and ferret sera,
used in conventional analyses, might complicate sensitive
Influenza virus haemagglutinin (HA) is a virion-surface determination of the effect of HA amino acid mutations on
glycoprotein and the primary target for neutralizing antigenicity due to cross-reactivity with other viral struc-
antibodies (Skehel & Wiley, 2000; Smith & Helenius, tural proteins, such as neuraminidase and nucleoprotein
2004). The protein is initially synthesized as precursor HA0 (Kaminski & Lee, 2011).
and cleaved to yield HA1, a variable external subunit, and
HA2, a conserved transmembrane subunit (Stevens et al., In this study, we performed phylogenetic analyses of HA
2006). Most of the HA1 molecule forms a globular head genes in HPAI H5N1 strains isolated in Egypt from 2006 to
containing the binding pocket for cell-surface sialylglycans 2011, identified the amino acid mutations that were
(or sialylgangliosides), the primary receptor for influenza conserved in each of the newly formed H5 sublineages,
viruses (Suzuki, 2005). HA affinity for sialylglycans is one and investigated the effect of these mutations on the
of the determinants of influenza A virus host range antigenicity and immunogenicity of the HAs with mono-
(Horimoto & Kawaoka, 2005; Suzuki, 2005). Human and clonal and polyclonal antibodies. The results of this study
avian influenza viruses differ in their recognition of host- should be useful for understanding antigenic drift among
cell receptors: human viruses mainly bind a2,6-linked H5N1 viruses in Egypt and for planning more efficacious
sialylglycan, while avian viruses have a high affinity for control strategies for endemic HPAI H5N1.
Downloaded from www.microbiologyresearch.org by
2216 Journal of General Virology 93
IP: 49.149.78.196
On: Fri, 16 Nov 2018 14:28:06
 Antigenic analysis of H5N1 influenza virus sublineages

RESULTS isolated in 2009–2011 clustered in two sublineages (A and B)

and most avian H5N1 viruses isolated in 2009–2011 clustered
Phylogenetic analysis of H5N1 HAs and in two different sublineages (C and D), as described pre-
identification of conserved mutations viously (Abdel-Moneim et al., 2009; Balish et al., 2010;
Watanabe et al., 2011b). This phylogenetic tree topology was
The phylogeny of H5N1 influenza viruses in Egypt was the same as that for a phylogenetic tree reconstructed from
investigated by analysing 62 HA sequences of representative 492 HA sequences from H5N1 viruses isolated in Egypt and
viruses isolated from birds and humans from 2006 to 2011. posted in GenBank (data not shown). Analysis of the 492 H5
Phylogenetic analysis showed that all of the HA genes HA sequences also identified amino acid mutations that were
clustered in clade 2.2.1 with an overall monophyletic conserved in each of the sublineages compared with ancestral
topology, indicating that these viruses diverged from a single Egyptian viruses; these conserved mutations are listed in Fig.
origin (Fig. 1). In addition, most human H5N1 viruses 1. Most HPAI H5N1 viruses isolated in Egypt at the time of

Fig. 1. Phylogenetic tree of HA genes of H5N1 viruses isolated in Egypt. This tree includes HA sequences of 51 H5N1
influenza A viruses isolated in Egypt, available in GenBank, and 11 HA sequences determined in our study. Strains whose HA
sequences were also analysed for antigenicity are marked with $. Amino acid mutations conserved in each of the sublineages
are shown using H5 numbering. Colours are used to highlight human and avian virus strains isolated in 2009–2011. Bar, 0.005
nucleotide substitutions per site.

Downloaded from www.microbiologyresearch.org by

http://vir.sgmjournals.org 2217
IP: 49.149.78.196
On: Fri, 16 Nov 2018 14:28:06
Y. Watanabe and others

this study were in sublineages B and D, which were recently different mAb reactivity patterns against the HAs of
classified as group 2.2.1/C and subclade 2.2.1.1, respectively, recombinant viruses from different sublineages (Fig. 2). All
in the WHO classification. In addition, viruses closely mAbs reacted similarly to cells infected by viruses with
related to ancestral viruses were isolated at low frequency. sublineage A and B HAs, except for mAb 4G6, which did not
Sublineages A and B were probably formed in early 2008, and react with sublineage B HAs due to the mutation at the
contained 96 of the 97 isolates from human infection cases in epitope (D43N) recognized by mAb 4G6 (Du et al., 2009).
Egypt since 2008. We previously found that HA mutations However, the mAbs did not react with cells infected with
Q192H and S129del/I151T (H5 numbering), which were viruses with sublineage C and D HAs. The reactivity of
conserved in sublineages A and B, respectively, increased the mAbs with the Egyptian HAs correlated with their
HA binding affinity for a2,6-linked sialylglycan, possibly neutralizing activity: parental, sublineage A and sublineage B
accounting for the increase in human H5N1 infections in viruses were neutralized by the mAbs with different efficacies,
Egypt (Watanabe et al., 2011b). However, sublineages C and while sublineage C and D viruses were not neutralized by the
D were probably formed in 2007, mainly from isolates from mAbs (data not shown). These results indicated that
vaccinated birds in commercial farms (Balish et al., 2010). ancestral, sublineage A and sublineage B viruses isolated in
These two sublineages shared several amino acid mutations Egypt shared epitopes in the HA globular head with
in the HA globular head and retained the classical binding antigenicity similar to the Asian H5 lineages, but sublineage
preference for a2,3-linked sialylglycan (Watanabe et al., C and D HAs did not have this antigenicity.
2011b). Previous studies suggested a vaccine-driven emer-
gence of sublineage C viruses (Abdel-Moneim et al., 2011; Prevalence of mutations characteristic of
Cattoli et al., 2011b). Sublineage D diverged from sublineage sublineages C and D
C in 2007, with additional mutations compared with the
ancestral viruses. Viruses in the most recent phylogenetic To investigate the effect of vaccination on antigenic drift of
branches were in sublineage D. Therefore, no single H5N1 sublineage C and D viruses, the prevalence of conserved
sublineage has become dominant, and phylogenetically amino acid mutations in these sublineages was compared
distinct viruses have persisted in Egypt. between HAs from H5N1 viruses isolated from 28 vaccinated
and 10 non-vaccinated geographically distant poultry flocks
in northern Egypt during 2007–2009, which was the putative
Antigenic analysis using HA recombinant viruses time when these sublineages arose (Arafa et al., 2010; Balish
A panel of five mAbs against the HA of A/crow/Kyoto/53/ et al., 2010). In H5N1 viruses isolated from vaccinated flocks,
2004 (H5N1), which had the antigenicity of other contem- mutations in HA characteristic of sublineage C were identified
porary Asian strains, was used for antigenic analysis of H5N1 in 30–80 % of viruses isolated in 2007, 54–77 % of viruses
viruses circulating in Egypt. We have previously mapped the isolated in 2008, and 80–100 % of viruses isolated in 2009,
epitopes recognized by these mAbs to conserved regions in although the number of virus strains from vaccinated flocks
the globular head of the HAs of East Asian H5 viruses and was small (Table 1). In H5N1 viruses isolated from vaccinated
shown that they have broad cross-neutralizing activity flocks, mutations in HA characteristic of sublineage D were
against Asian H5 lineage strains (Du et al., 2009). The broad identified in 10–30 % of viruses isolated in 2007, 8–23 % of
cross-reactivity of the mAbs with H5N1, H5N2 and H5N3 viruses isolated in 2008, and 0–80 % of viruses isolated in
HAs was confirmed in this study (Fig. S1, available in JGV 2009 (Table 1). In contrast, in H5N1 viruses isolated from
Online). The pattern of mAb reactivity with HA of A/duck/ non-vaccinated flocks, the prevalence of HA mutations
Egypt/D2Br21/2007, one of the ancestral H5 viruses in Egypt, characteristic of sublineage C was essentially zero in 2007
was comparable to that of the Asian H5 lineage, indicating an and 2008 and 100 % in 2009, and was essentially zero for HA
antigenic similarity between the Asian H5 lineage and mutations characteristic of sublineage D in 2007, 2008 and
ancestral Egyptian H5 viruses. In these studies, mAbs C43, 2009, although the number of virus strains from non-
which binds influenza nucleoprotein (Okuno et al., 1993), vaccinated flocks was small. For H5N1 viruses isolated in
and C179, which binds the HA stalk with cross-reactivity to Asia, the prevalence of HA mutations characteristic of
H1, H2, H5 and H6 viruses (Okuno et al., 1993; Smirnov sublineages C and D was low in 2007, 2008 and 2009, with
et al., 1999), were used as controls. The reactivity of these a few exceptions (at HA residues 120, 141, 154, 156 and 162).
mAbs with 11 HAs that contained sequences representative These results indicated that sublineage C and D viruses spread
of sublineage A, B, C and D viruses was analysed by immuno- preferentially among vaccinated flocks in northern Egypt,
implying vaccine-driven evolution of these viruses.
fluorescence assays. Both a human- and a bird-derived virus
HA were included as representatives of sublineages A and B.
To investigate specific reactivity between the mAbs and HAs, Antigenic analysis of recombinant viruses with
we generated recombinant H5N1 viruses, each containing specific mutations
one of the sublineage HA genes and the other genes from A/ To compare antigenic variation among sublineage HAs, the
duck/D1Br12/2007 (EG/D1), and infected Madin–Darby effect of strain-specific amino acids had to be excluded.
canine kidney (MDCK) cells with these viruses. Antigenic Therefore, mutations characteristic of each sublineage were
analysis of the infected cells with the panel of mAbs showed introduced in the HA gene of EG/D1, and recombinant
Downloaded from www.microbiologyresearch.org by
2218 Journal of General Virology 93
IP: 49.149.78.196
On: Fri, 16 Nov 2018 14:28:06
 Antigenic analysis of H5N1 influenza virus sublineages

Fig. 2. Antigenic variation in H5N1 viruses

isolated in Egypt. MDCK cells were infected
with recombinant viruses with different HAs at
an m.o.i. of 0.5 and the reactivity of the HAs
against a panel of mAbs against influenza A
viruses was determined by immunofluores-
cence assays. mAbs 3C11, 3H4, 3H12,
4C12 and 4G6 were cross-reactive to HA1
of the Asian H5 lineage, and mAbs C43 and
C179 were controls. The recombinant virus
designations are on the left and the phylogen-
etic sublineages are on the right. The mAb
designations are at the top, with their antigens
in parentheses.

viruses with the mutated HA gene and the unmodified attachment of N-glycans to the globular head of sublineage
other seven protein genes were generated. For sublineage HAs, recombinant HAs with specific mutations were
D, the 11 mutations found in both sublineages C and D prepared as described in Methods. Each protein carried
were also introduced. The antigenicity of these viruses was the mutations conserved in one of the sublineages. In
investigated by using them to infect MDCK cells and addition, variant forms of sublineage D HA were prepared in
analysing the reactivity patterns of the infected cells with which amino acid residues were introduced in the NGS of
the panel of mAbs described above. With one exception, sublineage D HA. Electrophoretic analysis of the proteins
HA mutations characteristic of sublineages A and B had showed different patterns of N-glycosylation among the
little effect on the reactivity patterns of the mAbs (Fig. 3). sublineage HAs (Fig. 4). The mobility shift for sublineage D
In contrast, no reactivity was seen in cells infected by HA1 with mutated NGS showed that the slower mobility of
viruses with HA mutations characteristic of sublineages C HA1 with mutations characteristic of sublineage D was due
and D. These results confirmed that sublineage C and D to glycosylation of residue 154N as a result of amino acid
HAs generally had different antigenicity from ancestral changes D154N and A156T in sublineage D HA1. Amino
Asian and Egyptian H5 viruses. acid changes P74S and N165H, which were conserved in
both sublineage C and D HAs, resulted in the generation and
loss of glycosylation sites, respectively, which explained the
Glycosylation of recombinant HAs with specific
lack of a mobility shift for sublineage C HA1. The N-
mutations
glycosylation patterns identified in sublineage HAs are
The HA1 proteins of Egyptian H5 lineage viruses contain summarized in Table 2. The effect of N-glycosylation
different combinations of three N-linked glycosylation sites changes in sublineage D HA on antigenicity was analysed
(NGS) at residues 72, 154 and 165. To determine the by immunofluorescence assays. The results suggested that
Downloaded from www.microbiologyresearch.org by
http://vir.sgmjournals.org 2219
IP: 49.149.78.196
On: Fri, 16 Nov 2018 14:28:06
Y. Watanabe and others

Table 1. Prevalence of HA mutations characteristic of H5 sublineages C and D in viruses isolated in Egypt and Asia

Sublineage/mutation in HA* Strains (%) with mutations isolated in:D

Egypt (n538) Asia (n5363)

From vaccinated flocks From non-vaccinated flocks

(n528) (n510)

2007 2008 2009 2007 2008 2009 2007 2008 2009

(n510) (n513) (n55) (n54) (n53) (n53) (n5198) (n5112) (n553)

C
P74S 50.0 61.5 100.0 0.0 0.0 100.0 0.5 18.8 1.9
D97N 50.0 61.5 100.0 0.0 0.0 100.0 1.5 0.9 1.9
H110R 40.0 53.8 100.0 0.0 0.0 100.0 0.0 0.9 0.0
S123P 40.0 61.5 100.0 0.0 0.0 100.0 15.2 5.4 15.1
R140G 30.0 61.5 80.0 0.0 0.0 100.0 0.0 0.9 0.0
S141P 80.0 61.5 100.0 0.0 0.0 100.0 40.4 32.1 20.8
F144Y 70.0 61.5 100.0 0.0 0.0 100.0 0.0 0.9 0.0
R162K 60.0 76.9 100.0 0.0 33.3 100.0 32.3 20.5 62.3
N165H 40.0 61.5 80.0 0.0 0.0 100.0 0.0 0.9 0.0
A184E 60.0 61.5 100.0 0.0 0.0 100.0 5.1 38.4 64.2
M226V 40.0 61.5 100.0 0.0 0.0 100.0 0.0 0.9 0.0
D
S120L 20.0 7.6 0.0 0.0 0.0 100.0 31.7 39.2 75.4
D154N 30.0 23.0 20.0 25.0 0.0 0.0 93.0 79.4 41.5
A156T 10.0 23.0 80.0 0.0 0.0 0.0 76.4 47.3 24.5
L190I 10.0 23.0 20.0 0.0 0.0 0.0 0.5 17.8 1.8
A238T 10.0 7.6 40.0 0.0 0.0 0.0 1.0 8.0 3.7

*Mutations are shown according to H5 numbering.

DPercentage of H5N1 viruses that have mutation(s) characteristic of sublineages C and D for each geographical region, type of flock and year of
virus isolation. Sequence information is from GenBank and from sequences analysed in this study.

Fig. 3. Effect of HA conserved mutations in different sublineages on antigenicity. MDCK cells were infected with recombinant
viruses with HAs containing the mutations conserved in one of the sublineages and the other genes from virus EG/D1, and the
reactivity of the HAs against a panel of mAbs was determined by immunofluorescence assays as described in the legend to Fig.
2. The mutations introduced into each EG/D1 HA are on the left and the HA sublineages are on the right. Mutations specific to
sublineage D are underlined. The mAb designations are at the top, with their antigens in parentheses.

Downloaded from www.microbiologyresearch.org by

2220 Journal of General Virology 93
IP: 49.149.78.196
On: Fri, 16 Nov 2018 14:28:06
 Antigenic analysis of H5N1 influenza virus sublineages

Table 2. Glycosylation patterns on HAs investigated in this

study
Glycosylation at the indicated residue in the globular head of HAs was
determined by the mobility shift of HAs with mutated NGS as shown
in Fig. 4. Residues are shown according to H5 numbering. +,
Glycosylated; 2, non-glycosylated.

Sublineage N-Glycosylation at residue:

72 154 165

Parental 2 2 +
A 2 2 +
B 2 2 +
Fig. 4. Glycosylation patterns of HA1 proteins. Flag-tagged HA1 C + 2 2
proteins with mutations characteristic one of the sublineages were D + + 2
produced in 293T cells as described in Methods. Variant forms of DS74P 2 + 2
sublineage D HA1 with mutated NGS were also produced. DN154D,T156A + 2 2
Electrophoretic patterns of recombinant HA1 proteins were DH165N + + +
visualized by Western blotting with an anti-Flag antibody. DS74P,N154D,T156A 2 2 2
Sublineage designations are at the top: Pa, parental HA1; A,
sublineage A HA1; B, sublineage B HA1; C, sublineage C HA1; D,
sublineage D HA1. –, Mock sample from empty-plasmid-trans- sublineage C or D HAs are not cross-reactive with sublineage
fected cells. A and B HAs.

the mutations at NGS and the other sites (non-NGS) in DISCUSSION

sublineage D HA impacted antigenic variation synergisti- In this study, we elucidated the antigenic drift among the
cally (Fig. S2). H5N1 viruses currently co-circulating in Egypt. Our analyses
indicated that the HAs in sublineage C and D viruses were
antigenically different from those in ancestral Asian and
Variations in immunogenicity of recombinant Egyptian H5 viruses and had undergone significant antigenic
viruses with specific mutations drift since their divergence from the sublineage A and B
To evaluate the immunogenicity of sublineage HAs, recom- phylogenetic branches. To our knowledge, this is the first
binant HA proteins with specific mutations were prepared report that sublineage D HAs were generally not cross-
and purified as described in Methods. Electrophoretic analysis immunogenic with other sublineage HAs, including even
of the purified proteins showed the specificity and purity of sublineage C HAs, although sublineages C and D form a
the HA1 preparations (Fig. S3). We examined whether the single phylogenetic branch and have a number of amino acid
amino acid changes in each sublineage affected induction of mutations in common.
serum antibody responses to HA1. Groups of mice were Viruses in the sublineage C phylogenetic branch were first
vaccinated intra-peritoneally with purified HA1s, and serum isolated in 2007. Our survey of vaccination records found
HAI and neutralizing antibody titres against viruses with that sublineage C viruses preferentially circulated in vacci-
homologous and heterologous HAs were determined post- nated poultry flocks during this time. These data supported
immunization. Sera from mice immunized with sublineage A previous suggestions that emergence of sublineage C may
and B HA1s had similar haemagglutination inhibition (HAI) have resulted from suboptimal vaccinations in Egypt (Abdel-
titres against homologous virus (453 and 570) and against Moneim et al., 2011; Cattoli et al., 2011b), although there is
each other (403 and 570), and much lower HAI titres against no direct evidence of this. Indeed, seven of the 11 mutations
viruses with sublineage C and D HA1s (7–36) (Table 3). Sera conserved in sublineage C HAs were in residues correspond-
from mice immunized with sublineage C and D HA1s had ing to the A and B antigenic sites in H3 HAs, which contain
significantly different HAI titres against homologous virus epitopes with high neutralizing efficiency (Kaverin et al.,
(508 and 127), lower titres against each other (18 and 50), and 2007; Wiley et al., 1981) (Fig. 5). H5N1 variants with some of
much lower titres against viruses with sublineage A and B the mutations characteristic of sublineage C HAs have been
HA1s (18–57) (Table 3). There was good correlation between isolated in Egypt since April 2007, and viruses with all of the
the patterns of HAI and neutralizing antibody titres for mutations characteristic of sublineage C HAs have been
sublineages A–D (Table 3). These results indicated that there isolated since December 2007. The first date, April 2007, was
was significant antigenic drift in sublineage C and D HAs after about 13 months after the launch of mass vaccination in
divergence of these phylogenetic branches from sublineage A industrial poultry sectors in Egypt, and about 5 months
and B HAs, and that mouse polyclonal antibodies induced by before the reported emergence of sublineage C (Arafa et al.,
Downloaded from www.microbiologyresearch.org by
http://vir.sgmjournals.org 2221
IP: 49.149.78.196
On: Fri, 16 Nov 2018 14:28:06
Y. Watanabe and others

Values are geometric mean antibody titres (GMT) in serum from six mice, obtained 2 weeks after the second administration of virus with HA mutations characteristic of the indicated viral

(423215 504)
2010; Balish et al., 2010). Similar findings were reported in

(15021364)
(9021144)
(382341)
(752682)
China, where possible vaccine-escape variants emerged
1 year after implementation of vaccination in poultry

D
(Smith et al., 2006).

2560
113
226
320
(4806–43 634) 453
Sublineage D HAs were more phylogenetically and anti-
genically distant from ancestral, sublineage A and B HAs

(150–1364)
than were sublineage C HAs. Interestingly, sublineage D,

(90–1144)

(90–1144)
(38–341)
C which diverged rapidly from sublineage C in 2007, was
antigenically distinct from sublineage C with little polyclonal
Table 3. Serum antibody response in mice immunized with H5N1 viruses isolated in Egypt with HA mutations characteristic of each sublineage

*Each HAI titre against the challenge virus is the reciprocal of the highest serum dilution that inhibited haemagglutination by 8 virus haemagglutination units.
113
320
453

320
14 482
antibody cross-reactivity. These data were in agreement with
NT GMTD (95 % confidence interval)

a previous antigenic analysis using sublineage D strain A/

DEach neutralizing titre (NT) against the challenge virus is the reciprocal of the highest serum dilution that reduced infection of 100 virus f.f.u. by .50 %.
Egypt/3300-NAMRU3/2008 (Balish et al., 2010). Sublineage
7 241 (876259 843) 10 240 (1691262 017)
(13224) 5120 (564246 482) 10 240 (4522231 811) 20 480 (5731273 188)
(15267) 5120 (1433218 297) 10 240 (8012130 773) 20 480 (5731273 188)
640 (10623876)
D viruses also circulated preferentially in vaccinated poultry
113 (382341)
flocks, implying vaccination-driven phylogenetic emer-
B

gence. The sublineage D viruses had five conserved amino

acid changes, which were in residues corresponding to H3
antigenic sites A and B, in addition to 11 mutations also
found in sublineage C viruses (Fig. 5). These amino acid
changes included D154N and A156T mutations, which have
453 (15021364)
113 (382341)

resulted in 154N glycosylation. Post-translational N-glyco-

sylation of the HA protein can influence its antigenicity and
A

receptor-binding affinity (Skehel et al., 1984; Wang et al.,

2010). As expected, HA1 with sublineage D characteristic
mutations, including 154N and 156T, migrated more slowly
electrophoretically than other sublineage HA1s with residues
7241 (876259 843)

226 (2721870)

154D and 156A, due to glycosylation of residue 154N in

80 (222286)
Parental

sublineage D HA1s. No other putative glycosylation sites

were generated by the sublineage D conserved mutations.
These data confirmed that the D154N and A156T amino
acid changes resulted in acquisition of a glycosylation site
(702230)

and that the 154N residue in sublineage D HAs was indeed

(10231)
(6219)

glycosylated. These amino acid changes may produce

D

substantial differences in immunogenicity between subline-

11
18
32
508 (24021077) 18
127

age C and D HAs, especially in HAI antibody levels in the

sera of immunized mice.
(2772587) 453 (3042674) 570 (24421335) 50 (192136)
57 (262121)

50 (132195)
28 (13261)

Influenza virus receptor affinity can be altered during virus

HAI GMT* (95 % confidence interval)

evolution by selection of immune-escape variants (Hensley

et al., 2009). Unexpectedly, sublineage A and B viruses,
which acquired increased affinity for a2,6-sialylglycan in
combination with residual affinity for a2,3-sialylglycan,
(2082621) 254 (1402460) 254 (1742370)

(2772587) 403 (2232730) 570 (4242767)

36 (21262)
11 (5223)

retained the antigenicity of ancestral Asian and Egyptian

B

viruses. However, mAb 4G6 did not react with sublineage

sublineage. Homologous titres are shown in bold.

B HAs, indicating a minor antigenic drift in these HAs.

Viruses in sublineages A and B were first isolated in early
2008, several months after viruses in sublineages C and D
18 (9236)
7 (4213)

were first isolated. The factors involved in the emergence of

A

sublineages A and B remain uncertain. Previous studies

reported that sublineage B virus hosts were restricted to
aquatic poultry, mainly ducks (Arafa et al., 2010; Cattoli
et al., 2011b), implying that specific mutations were
(10253)
Parental

(5211)

selected and conserved among these avian virus strains.

We also reported genome diversification in H5N1 viruses
359
403
403
22
7

that infect wild ducks (Watanabe et al., 2011a) and host-

specific genetic evolution in these viruses in Egypt
Sublineage

Parental

(Ibrahim et al., 2011). It is possible that antigenic changes

may gradually accumulate during influenza virus circula-
D
C
A
B

tion in waterfowl, unrelated to whether there is vaccination

Downloaded from www.microbiologyresearch.org by
2222 Journal of General Virology 93
IP: 49.149.78.196
On: Fri, 16 Nov 2018 14:28:06
 Antigenic analysis of H5N1 influenza virus sublineages

Fig. 5. Locations of amino acid mutations

conserved in sublineage C and D HAs. (a)
Three-dimensional structure of the H5 HA
trimer, showing the locations of H3 HA
antigenic sites A–E. (b) Locations of amino
acid mutations conserved in sublineage C and
D HAs. The mutated residues, which were
conserved in sublineage C and D HAs, are
shown in red. Orange circles indicate N-
glycosylation sites. Residue D43, which is
recognized by mAb 4G6, is shown in blue. The
crystal structure of the HA of A/Vietnam/1194/
2004 (H5N1) (PDB ID 2IBX) was used as a
template.

in the field. Aquatic poultry probably plays a substantial continue to evolve without the selective pressure due to
role in H5N1 evolution in Egypt, along with terrestrial vaccination. However, since the ecology of influenza viruses
birds. Therefore, the ecology of sublineage A and B viruses and that of many avian species in Egypt are closely related,
in aquatic birds needs to be studied further. evolution will continue to produce genetically divergent viruses
in Egypt and, perhaps, other countries where H5N1 infection is
The standard method for control of an HPAI outbreak is
testing and culling of all poultry in a farm (Suarez, 2010). endemic (Watanabe et al., 2012). Continuous circulation
However, it has been suggested by the OIE that, when among hosts may allow H5N1 virus to acquire amino acid
outbreaks spread to a broad area and become uncontrollable, changes enabling more bird–human transmission and even-
ring vaccination would be an additional method to reduce tually human–human transmission. Large-scale surveillance of
nascent virus production and, thereby, suppress further avian influenza viruses in endemic areas should be expanded
virus infection (OIE, 2003). Vaccination of poultry is now to better understand evolution of the virus and enable more
considered a preventive or auxiliary control approach in efficacious control strategies in these regions.
several H5N1-endemic countries, including Egypt (Peiris
et al., 2007; Swayne & Kapczynski, 2008). The problem is that
antigenically divergent groups of viruses, which are not METHODS
cross-reactive, are co-circulating in Egypt. A similar situation
Generation of recombinant viruses. Recombinant H5N1 viruses
was also found in Indonesia even though major antigenic were generated using a plasmid-based reverse-genetics system as
variation had not been detected there as of 2008 (Wibawa described previously (Fodor et al., 1999; Watanabe et al., 2011b). Each
et al., 2011). Only several of the recently isolated H5N1 strains virus generated by reverse genetics (denoted here as rEG/X) carried
circulating in Indonesia had lost reactivity to mAb 4G6 the HA gene of the virus being studied, with the other genes coming
(Mieko Kosaka, personal communication). This makes selec- from A/duck/D1Br12/2007 (EG/D1), one of the ancestral influenza
tion of one influenza virus strain as the vaccine seed virus virus strains isolated in Egypt. The 11 recombinant viruses, each
especially problematic. Furthermore, random rearing of many containing an HA from an Egyptian virus, were EG/D1, A/chicken/
Egypt/RIMD/1-5/2008 (EG/1), A/chicken/Egypt/RIMD4-3/2008 (EG/4),
bird species and their hybrid breeds with uncontrolled
A/chicken/Egypt/RIMD5-3/2008 (EG/5), A/chicken/Egypt/RIMD11-1/
confinement is common in rural areas (Suarez, 2010), leaving 2008 (EG/11), A/chicken/Egypt/RIMD12-3/2008 (EG/12), A/chicken/
ducks and geese free to fly away. Therefore, circulation of Egypt/RIMD28-1/2009 (EG/28), A/chicken/Egypt/RIMD29-3/2008 (EG/
viruses in each sublineage in Egypt was not restricted in terms 29), A/goose/Egypt/0929-NLQP/2009 (EG/0929), A/Egypt/N04822/2009
of geography or host species, complicating efforts to use a (EG/4822) and A/Egypt/N02039/2009 (EG/2039). The HA genes of EG/
vaccine produced against antigens from a single virus strain. 0929, EG/4822 and EG/2039 were synthesized using the sequences
In the future, when vaccination is implemented as part of a registered in GenBank (http://www.ncbi.nlm.nih.gov/nucleotide) and
comprehensive control strategy for endemic HPAI in Egypt, site-directed mutagenesis PCR. Mutant HA genes were generated by
PCR-based site-directed mutagenesis of the HA genes of EG/D1, EG/11,
the most promising method may be development and use of
EG/12 and EG/29. The HA genes of the virus stocks were sequenced to
multivalent or universal vaccines. Broad-spectrum efficacy detect the possible emergence of revertants or unwanted mutations
would need to be revised periodically (Swayne, 2009), although during amplification as described previously (Watanabe et al., 2007).
vaccination in backyard settings was suspended provision-
ally in July 2009 in Egypt until a new vaccine strategy could Cells. MDCK and 293T cells were purchased from the RIKEN
be adopted (FAO, 2011). BioResource Center Cell Bank (http://www.brc.riken.jp/lab/cell/
english/). Chicken embryo fibroblast (CEF) cells were prepared from
It is not known whether sublineage B and D viruses, which 11-day-old embryonated eggs. These cell lines were maintained as
are now the dominant strains co-circulating in Egypt, described previously (Watanabe et al., 2009).

Downloaded from www.microbiologyresearch.org by

http://vir.sgmjournals.org 2223
IP: 49.149.78.196
On: Fri, 16 Nov 2018 14:28:06
Y. Watanabe and others

Virus propagation. The recombinant viruses generated in this study use of laboratory animals in the Research Institute for Microbial
were propagated by single passage in the allantoic cavity of 11-day-old Diseases, Osaka University.
embryonated chicken eggs. The allantoic fluids were harvested 3 days
post-infection and stored at 280 uC. Virus titres were assayed as f.f.u. Serological analysis. Antibody levels against homologous and
by focus-forming assays (Di Lonardo et al., 2002) on MDCK cells. All heterologous viruses in post-immunization mouse sera were deter-
experiments with live H5N1 viruses were performed in Biosafety Level mined by HAI and microneutralization assays as follows. Sera were
3+ (BSL 3+) conditions at Osaka University, as approved for work treated overnight with receptor-destroying enzyme (Denka-Seiken) at
with these viruses by the Ministry of Agriculture, Forestry and 37 uC for 18 h and heat-inactivated at 56 uC for 45 min. HAI assays
Fisheries, Japan. were performed as described previously (Okuno et al., 1993), with
serial twofold dilutions of mouse serum starting from a 1 : 10 dilution.
Genetic analysis. For phylogenetic analysis of HA genes, we used The HAI antibody titre was defined as the reciprocal of the highest
HA sequences of 51 representative H5N1 influenza A viruses isolated serum dilution that inhibited haemagglutination. For HAI calcula-
in Egypt from 2006 to 2011 and obtained from GenBank, and 11 HAs tions, it was useful to assign HAI titres ,10 HAI a value of 5 HAI.
representative of the 21 HAs sequenced in our studies. Phylogenetic For neutralization assays, serial twofold dilutions of serum starting
analysis of the 62 HA nucleotide sequences was done by the from a 1 : 10 dilution were incubated with an equal volume of virus
neighbour-joining method using MEGA4 software (Tamura et al., containing 100 f.f.u. in a 96-well U-bottom plate for 60 min at 37 uC.
2007), with the nucleotide sequences covering most of the HA genes. The virus–serum mixture was transferred to an MDCK cell monolayer
Estimates of the variability of the reconstructed phylogenetic trees and incubated at 37 uC for 8 h. The neutralization antibody titre was
were calculated for 1000 bootstrap replicates. For reconstruction of a defined as the reciprocal of the highest serum dilution that
detailed phylogenetic tree, 398 HA sequences from avian viruses and neutralized .50 % of the viruses in a focus-forming assay.
98 HA sequences from human viruses isolated in Egypt from 2006 to
2011 were obtained from a GenBank search and analysed. For
comparison, 366 published HA sequences of H5N1 influenza A ACKNOWLEDGEMENTS
viruses recently isolated in Asia were obtained from GenBank. The
prevalence of mutations characteristic of H5N1 sublineages C and D We thank R. Kubota-Koketsu and M. Yasugi for valuable advice
was calculated for 17 HA sequences in previous studies and 21 HA and discussions and A. Yamashita for computational assistance and
sequences in our studies, all isolated from different poultry sectors in resources. This work was supported by a Grant in Aid for Scientific
northern Egypt (detailed data are available on request). These Research from the Ministry of Education, Culture, Sports, Science
sequences were aligned by the MAFFT program (Katoh et al., 2002) and Technology, Japan (grant numbers 23791134 and 23406017) and
and the HA1 regions were compared with each other. the Japan Society for the Promotion of Science (JSPS) through the
‘Bilateral Programs’.
Antigenic analysis. The antigenic specificity of HAs in Egypt was
assessed by immunofluorescence assays with a panel of five mouse
mAbs against A/crow/Kyoto/53/2004 (H5N1), which we produced REFERENCES
and characterized previously (Du et al., 2009). MDCK cells were
infected with recombinant viruses at an m.o.i. of 0.5 and analysed for Abdel-Moneim, A. S., Shany, S. A., Fereidouni, S. R., Eid, B. T.,
reactivity with each of the mAbs by immunofluorescence using an el-Kady, M. F., Starick, E., Harder, T. & Keil, G. M. (2009). Sequence
FITC-conjugated secondary antibody. diversity of the haemagglutinin open reading frame of recent highly
pathogenic avian influenza H5N1 isolates from Egypt. Arch Virol 154,
Purification of recombinant HA1s. Each HA1 expression plasmid 1559–1562.
was produced by fusing an HA1 gene with the C terminus of a Flag
Abdel-Moneim, A. S., Afifi, M. A. & El-Kady, M. F. (2011). Genetic drift
tag and inserting it into a pcXN2 vector (Watanabe et al., 2009).
evolution under vaccination pressure among H5N1 Egyptian isolates.
Insertion of the correct sequences for wild-type and mutant HA1s was
Virol J 8, 283.
confirmed by DNA sequencing and Western blot analysis of the
expressed proteins. Recombinant HA1s were produced by transfecting Abdelwhab, E. M. & Hafez, H. M. (2011a). An overview of the
293T cells with plasmid DNAs using TransIT-LT1 (Mirus Bio), epidemic of highly pathogenic H5N1 avian influenza virus in Egypt:
according to the manufacturer’s instructions. At 5 days post- epidemiology and control challenges. Epidemiol Infect 139, 647–657.
transfection, culture supernatants containing released HA1s were Abdelwhab, E. M., Grund, C., Aly, M. M., Beer, M., Harder, T. C. &
collected and centrifuged twice at 8400 g for 20 min. The super- Hafez, H. M. (2011b). Multiple dose vaccination with heterologous
natants were incubated with pre-equilibrated anti-Flag M2 agarose H5N2 vaccine: immune response and protection against variant clade
(Sigma-Aldrich) for 5 h at 4 uC with gentle rotation. The beads were 2.2.1 highly pathogenic avian influenza H5N1 in broiler breeder
then collected by centrifugation at 4700 g for 40 s and washed four chickens. Vaccine 29, 6219–6225.
times with PBS containing 0.1 % Tween 20 (PBST). The proteins Arafa, A., Suarez, D. L., Hassan, M. K. & Aly, M. M. (2010).
immunoprecipitated by anti-Flag agarose were eluted with 36 Flag Phylogenetic analysis of hemagglutinin and neuraminidase genes of
peptide (Sigma-Aldrich) in PBST and dialysed against PBS for 40 h at highly pathogenic avian influenza H5N1 Egyptian strains isolated
4 uC. Western blotting was performed as described previously from 2006 to 2008 indicates heterogeneity with multiple distinct
(Watanabe et al., 2009). sublineages. Avian Dis 54 (1 Suppl. ), 345–349.
Immunization of mice. Groups of six mice (Japan SLC Inc.) were Balish, A. L., Davis, C. T., Saad, M. D., El-Sayed, N., Esmat, H.,
immunized with two intraperitoneal injections of 100 mg purified Tjaden, J. A., Earhart, K. C., Ahmed, L. E., Abd El-Halem, M. & other
HA1 protein with Freund’s complete adjuvant, with 2 weeks between authors (2010). Antigenic and genetic diversity of highly pathogenic
injections. Two weeks after the second administration, sera were avian influenza A (H5N1) viruses isolated in Egypt. Avian Dis 54
collected and specific antibody responses were examined by HAI and (1 Suppl. ), 329–334.
microneutralization assays as described below. Control mice were Cattoli, G., Monne, I., Fusaro, A., Joannis, T. M., Lombin, L. H., Aly,
immunized with PBS with Freund’s adjuvant. All animal studies were M. M., Arafa, A. S., Sturm-Ramirez, K. M., Couacy-Hymann, E. &
conducted under the applicable laws and guidelines for the care and other authors (2009). Highly pathogenic avian influenza virus

Downloaded from www.microbiologyresearch.org by

2224 Journal of General Virology 93
IP: 49.149.78.196
On: Fri, 16 Nov 2018 14:28:06
 Antigenic analysis of H5N1 influenza virus sublineages

subtype H5N1 in Africa: a comprehensive phylogenetic analysis and Ibrahim, M. S., Watanabe, Y., Ellakany, H. F., Yamagishi, A.,
molecular characterization of isolates. PLoS ONE 4, e4842. Sapsutthipas, S., Toyoda, T., Abd El-Hamied, H. S. & Ikuta, K.
Cattoli, G., Milani, A., Temperton, N., Zecchin, B., Buratin, A., Molesti, E., (2011). Host-specific genetic variation of highly pathogenic avian
Aly, M. M., Arafa, A. & Capua, I. (2011a). Antigenic drift in H5N1 avian influenza viruses (H5N1). Virus Genes 42, 363–368.
influenza virus in poultry is driven by mutations in major antigenic sites Ito, T., Couceiro, J. N., Kelm, S., Baum, L. G., Krauss, S., Castrucci,
of the hemagglutinin molecule analogous to those for human influenza M. R., Donatelli, I., Kida, H., Paulson, J. C. & other authors (1998).
virus. J Virol 85, 8718–8724. Molecular basis for the generation in pigs of influenza A viruses with
Cattoli, G., Fusaro, A., Monne, I., Coven, F., Joannis, T., El-Hamid,
pandemic potential. J Virol 72, 7367–7373.
H. S., Hussein, A. A., Cornelius, C., Amarin, N. M. & other authors Kaminski, D. A. & Lee, F. E. (2011). Antibodies against conserved
(2011b). Evidence for differing evolutionary dynamics of A/H5N1 antigens provide opportunities for reform in influenza vaccine design.
viruses among countries applying or not applying avian influenza Front Immunol 2, 76.
vaccination in poultry. Vaccine 29, 9368–9375. MAFFT: a novel
Katoh, K., Misawa, K., Kuma, K. & Miyata, T. (2002).
Chen, H., Smith, G. J., Zhang, S. Y., Qin, K., Wang, J., Li, K. S., method for rapid multiple sequence alignment based on fast Fourier
Webster, R. G., Peiris, J. S. & Guan, Y. (2005). Avian flu: H5N1 virus transform. Nucleic Acids Res 30, 3059–3066.
outbreak in migratory waterfowl. Nature 436, 191–192. Kaverin, N. V., Rudneva, I. A., Govorkova, E. A., Timofeeva, T. A.,
Chen, H., Smith, G. J., Li, K. S., Wang, J., Fan, X. H., Rayner, J. M., Shilov, A. A., Kochergin-Nikitsky, K. S., Krylov, P. S. & Webster, R. G.
Vijaykrishna, D., Zhang, J. X., Zhang, L. J. & other authors (2006). (2007). Epitope mapping of the hemagglutinin molecule of a highly
Establishment of multiple sublineages of H5N1 influenza virus in pathogenic H5N1 influenza virus by using monoclonal antibodies.
Asia: implications for pandemic control. Proc Natl Acad Sci U S A J Virol 81, 12911–12917.
103, 2845–2850. Kim, J. K., Kayali, G., Walker, D., Forrest, H. L., Ellebedy, A. H., Griffin,
Couceiro, J. N., Paulson, J. C. & Baum, L. G. (1993). Influenza virus Y. S., Rubrum, A., Bahgat, M. M., Kutkat, M. A. & other authors
strains selectively recognize sialyloligosaccharides on human respir- (2010). Puzzling inefficiency of H5N1 influenza vaccines in Egyptian
atory epithelium; the role of the host cell in selection of hemag- poultry. Proc Natl Acad Sci U S A 107, 11044–11049.
glutinin receptor specificity. Virus Res 29, 155–165. Meleigy, M. (2007). Egypt battles with avian influenza. Lancet 370,
Di Lonardo, A., Buttinelli, G., Amato, C., Novello, F., Ridolfi, B. & 553–554.
Fiore, L. (2002). Rapid methods for identification of poliovirus Nelson, M. I. & Holmes, E. C. (2007). The evolution of epidemic
isolates and determination of polio neutralizing antibody titers in influenza. Nat Rev Genet 8, 196–205.
human sera. J Virol Methods 101, 189–196. OIE (2003). The use of vaccination as an option for the control of
Du, A., Daidoji, T., Koma, T., Ibrahim, M. S., Nakamura, S., de Silva, avian influenza. Final report of the 71st general session of the OIE
U. C., Ueda, M., Yang, C. S., Yasunaga, T. & other authors (2009). international committee, Paris, 18–23 May 2003. http://www.oie.int/
Detection of circulating Asian H5N1 viruses by a newly established fileadmin/Home/eng/About_us/docs/pdf/A_RFinal_2003_wp.pdf
monoclonal antibody. Biochem Biophys Res Commun 378, 197–202. OIE (2011). Update on highly pathogenic avian influenza in animals
Eladl, A. E., El-Azm, K. I., Ismail, A. E., Ali, A., Saif, Y. M. & Lee, C. W. (type H5 and H7). Accessed 4 December 2011. http://www.oie.int/
(2011). Genetic characterization of highly pathogenic H5N1 avian animal-health-in-the-world/update-on-avian-influenza/2011/
influenza viruses isolated from poultry farms in Egypt. Virus Genes Okuno, Y., Isegawa, Y., Sasao, F. & Ueda, S. (1993). A common
43, 272–280. neutralizing epitope conserved between the hemagglutinins of
FAO (2011). H5N1 HPAI global overview. Accessed 4 December influenza A virus H1 and H2 strains. J Virol 67, 2552–2558.
2011. http://www.fao.org/avianflu/en/overview.htm Peiris, J. S., de Jong, M. D. & Guan, Y. (2007). Avian influenza
Fodor, E., Devenish, L., Engelhardt, O. G., Palese, P., Brownlee, G. G. virus (H5N1): a threat to human health. Clin Microbiol Rev 20, 243–
& Garcı́a-Sastre, A. (1999). Rescue of influenza A virus from 267.
recombinant DNA. J Virol 73, 9679–9682. Peyre, M., Samaha, H., Makonnen, Y. J., Saad, A., Abd-Elnabi, A.,
Fusaro, A., Nelson, M. I., Joannis, T., Bertolotti, L., Monne, I., Salviato, A., Galal, S., Ettel, T., Dauphin, G., Lubroth, J. & other authors (2009).
Olaleye, O., Shittu, I., Sulaiman, L. & other authors (2010). Evolutionary Avian influenza vaccination in Egypt: limitations of the current
dynamics of multiple sublineages of H5N1 influenza viruses in Nigeria strategy. J Mol Genet Med 3, 198–204.
from 2006 to 2008. J Virol 84, 3239–3247. Saad, M. D., Ahmed, L. S., Gamal-Eldein, M. A., Fouda, M. K., Khalil, F.,
Grund, C., Abdelwhab, S. M., Arafa, A. S., Ziller, M., Hassan, M. K., Yingst, S. L., Parker, M. A. & Montevillel, M. R. (2007). Possible avian
Aly, M. M., Hafez, H. M., Harder, T. C. & Beer, M. (2011). Highly influenza (H5N1) from migratory bird, Egypt. Emerg Infect Dis 13,
pathogenic avian influenza virus H5N1 from Egypt escapes vaccine- 1120–1121.
induced immunity but confers clinical protection against a hetero- Salzberg, S. L., Kingsford, C., Cattoli, G., Spiro, D. J., Janies, D. A.,
logous clade 2.2.1 Egyptian isolate. Vaccine 29, 5567–5573. Aly, M. M., Brown, I. H., Couacy-Hymann, E., De Mia, G. M. & other
Hafez, M. H., Arafa, A., Abdelwhab, E. M., Selim, A., Khoulosy, S. G., authors (2007). Genome analysis linking recent European and
Hassan, M. K. & Aly, M. M. (2010). Avian influenza H5N1 virus African influenza (H5N1) viruses. Emerg Infect Dis 13, 713–718.
infections in vaccinated commercial and backyard poultry in Egypt. Skehel, J. J. & Wiley, D. C. (2000). Receptor binding and membrane
Poult Sci 89, 1609–1613. fusion in virus entry: the influenza hemagglutinin. Annu Rev Biochem
Hensley, S. E., Das, S. R., Bailey, A. L., Schmidt, L. M., Hickman, 69, 531–569.
H. D., Jayaraman, A., Viswanathan, K., Raman, R., Sasisekharan, R. & Skehel, J. J., Stevens, D. J., Daniels, R. S., Douglas, A. R., Knossow, M.,
other authors (2009). Hemagglutinin receptor binding avidity drives Wilson, I. A. & Wiley, D. C. (1984). A carbohydrate side chain on
influenza A virus antigenic drift. Science 326, 734–736. hemagglutinins of Hong Kong influenza viruses inhibits recognition by
Horimoto, T. & Kawaoka, Y. (2005). Influenza: lessons from past a monoclonal antibody. Proc Natl Acad Sci U S A 81, 1779–1783.
pandemics, warnings from current incidents. Nat Rev Microbiol 3, Smirnov, Y. A., Lipatov, A. S., Gitelman, A. K., Okuno, Y., Van Beek, R.,
591–600. Osterhaus, A. D. & Claas, E. C. (1999). An epitope shared by the

Downloaded from www.microbiologyresearch.org by

http://vir.sgmjournals.org 2225
IP: 49.149.78.196
On: Fri, 16 Nov 2018 14:28:06
Y. Watanabe and others

hemagglutinins of H1, H2, H5, and H6 subtypes of influenza A virus. Watanabe, Y., Ibrahim, M. S., Hagiwara, K., Okamoto, M., Kamitani, W.,
Acta Virol 43, 237–244. Yanai, H., Ohtaki, N., Hayashi, Y., Taniyama, H. & other authors
Smith, A. E. & Helenius, A. (2004). How viruses enter animal cells. (2007). Characterization of a Borna disease virus field isolate which
Science 304, 237–242. shows efficient viral propagation and transmissibility. Microbes Infect 9,
417–427.
Smith, G. J., Fan, X. H., Wang, J., Li, K. S., Qin, K., Zhang, J. X.,
Watanabe, Y., Ohtaki, N., Hayashi, Y., Ikuta, K. & Tomonaga, K.
Vijaykrishna, D., Cheung, C. L., Huang, K. & other authors (2006).
(2009). Autogenous translational regulation of the Borna disease virus
Emergence and predominance of an H5N1 influenza variant in China.
negative control factor X from polycistronic mRNA using host RNA
Proc Natl Acad Sci U S A 103, 16936–16941.
helicases. PLoS Pathog 5, e1000654.
Stevens, J., Blixt, O., Tumpey, T. M., Taubenberger, J. K., Paulson,
Watanabe, Y., Ibrahim, M. S., Ellakany, H. F., Abd El-Hamid, H. S. &
J. C. & Wilson, I. A. (2006). Structure and receptor specificity of the
Ikuta, K. (2011a). Genetic diversification of H5N1 highly pathogenic
hemagglutinin from an H5N1 influenza virus. Science 312, 404–410.
avian influenza A virus during replication in wild ducks. J Gen Virol
Suarez, D. L. (2010). Avian influenza: our current understanding. 92, 2105–2110.
Anim Health Res Rev 11, 19–33.
Watanabe, Y., Ibrahim, M. S., Ellakany, H. F., Kawashita, N., Mizuike, R.,
Suzuki, Y. (2005). Sialobiology of influenza: molecular mechanism of Hiramatsu, H., Sriwilaijaroen, N., Takagi, T., Suzuki, Y. & Ikuta, K.
host range variation of influenza viruses. Biol Pharm Bull 28, 399–408. (2011b). Acquisition of human-type receptor binding specificity by new
Swayne, D. E. (2009). Avian influenza vaccines and therapies for H5N1 influenza virus sublineages during their emergence in birds in
poultry. Comp Immunol Microbiol Infect Dis 32, 351–363. Egypt. PLoS Pathog 7, e1002068.
Swayne, D. E. & Kapczynski, D. (2008). Strategies and challenges for Watanabe, Y., Ibrahim, M. S., Suzuki, Y. & Ikuta, K. (2012). The changing
eliciting immunity against avian influenza virus in birds. Immunol nature of avian influenza A virus (H5N1). Trends Microbiol 20, 11–20.
Rev 225, 314–331. WHO (2012a). Cumulative number of confirmed human cases of
Tamura, K., Dudley, J., Nei, M. & Kumar, S. (2007).MEGA4: molecular
avian influenza A(H5N1) reported to WHO. Accessed 2 April 2012.
evolutionary genetics analysis (MEGA) software version 4.0. Mol Biol http://www.who.int/influenza/human_animal_interface/H5N1_cumula
Evol 24, 1596–1599. tive_table_archives/en/index.html
Terregino, C., Toffan, A., Cilloni, F., Monne, I., Bertoli, E., WHO (2012b). Global alert and response: disease outbreak news.
Castellanos, L., Amarin, N., Mancin, M. & Capua, I. (2010).
Accessed 2 April 2012. http://www.who.int/csr/don/en/
Evaluation of the protection induced by avian influenza vaccines WHO (2012c). H5N1 avian influenza: timeline of major events.
containing a 1994 Mexican H5N2 LPAI seed strain against a 2008 Accessed 14 March 2012. http://www.who.int/influenza/human_animal_
Egyptian H5N1 HPAI virus belonging to clade 2.2.1 by means of interface/avian_influenza/H5N1_avian_influenza_update140312.pdf
serological and in vivo tests. Avian Pathol 39, 215–222. WHO/OIE/FAO H5N1 Evolution Working Group (2008). Toward a
Wang, G., Zhan, D., Li, L., Lei, F., Liu, B., Liu, D., Xiao, H., Feng, Y., Li, J. unified nomenclature system for highly pathogenic avian influenza
& other authors (2008). H5N1 avian influenza re-emergence of Lake virus (H5N1). Emerg Infect Dis 14, e1.
Qinghai: phylogenetic and antigenic analyses of the newly isolated Wibawa, H., Henning, J., Wong, F., Selleck, P., Junaidi, A., Bingham, J.,
viruses and roles of migratory birds in virus circulation. J Gen Virol Daniels, P. & Meers, J. (2011). A molecular and antigenic survey of
89, 697–702. H5N1 highly pathogenic avian influenza virus isolates from smallholder
Wang, W., Lu, B., Zhou, H., Suguitan, A. L., Jr, Cheng, X., Subbarao, K., duck farms in Central Java, Indonesia during 2007–2008. Virol J 8, 425.
Kemble, G. & Jin, H. (2010). Glycosylation at 158N of the hemag- Wiley, D. C., Wilson, I. A. & Skehel, J. J. (1981). Structural
glutinin protein and receptor binding specificity synergistically affect identification of the antibody-binding sites of Hong Kong influenza
the antigenicity and immunogenicity of a live attenuated H5N1 A/ haemagglutinin and their involvement in antigenic variation. Nature
Vietnam/1203/2004 vaccine virus in ferrets. J Virol 84, 6570–6577. 289, 373–378.

Downloaded from www.microbiologyresearch.org by

2226 Journal of General Virology 93
IP: 49.149.78.196
On: Fri, 16 Nov 2018 14:28:06