Beruflich Dokumente
Kultur Dokumente
Isolation of highly pathogenic H5N6 avian influenza virus in Southern Vietnam with
Running title: Highly pathogenic H5N6 avian influenza virus in Southern Vietnam
Ryota Tsunekunia,b,*, Kasumi Sudoc,*, Phuong Thanh Nguyend, Bach Duc Luud, Thai Duy
Phuongd, Tran Minh Tand, Nguyen Tunge, Junki Minea,b, Momoko Nakayamaa,b, Taichiro
This article has been accepted for publication and undergone full peer review but has not been through
the copyediting, typesetting, pagination and proofreading process, which may lead to differences
between this version and the Version of Record. Please cite this article as doi: 10.1111/tbed.13294
Japan
Russia
taksaito@affrc.go.jp,tune@affrc.go.jp
Since 2013, H5N6 highly pathogenic avian influenza viruses (HPAIVs) have
been responsible for outbreaks in poultry and wild birds around Asia. H5N6 HPAIV is
also a public concern due to sporadic human infections being reported in China. In the
Southern Vietnam in July 2018 and genetically characterized it. Basic Local Alignment
Search Tool (BLAST) analysis revealed that the eight genomic segments of AI470 were
was isolated in October 2018 in Russia. Furthermore, AI470 also shared 99.4–99.9%
China in 2018. Phylogenetic analyses of the entire genome showed that AI470 was
directly derived from H5N6 HPAIVs that were in South China from 2015 to 2018 and
clustered with four H5N6 HPAIV strains of human origin in South China from 2017 to
2018. This indicated that AI470 was introduced into Vietnam from China. In addition,
H5N6 HPAIV viruses, except PB2 E627K, were shared by AI470. These findings are
Keywords: Highly pathogenic avian influenza virus, human infection, Vietnam, G1.1,
South China
(Gs/Gd) lineage have become diversified since their original emergence in 1996 (Xu,
Subbarao, Cox, & Guo, 1999). Gs/Gd HPAIVs had been exclusively of the H5N1 subtype
until an H5N5 HPAIV Gs/Gd strain appeared in China in 2008–2009 due to genetic
reassortment (Gu et al., 2011). Since then, the so-called H5Nx HPAIVs have expanded
worldwide and have been responsible for poultry outbreaks on the Eurasian, North
American, and African continents (World Organisation for Animal Health, 2018a). The
clade 2.3.4.4 has been a major branch of the H5Nx HPAIVs of the Gs/Gd lineage and
classified into four sub-lineages, groups A, B, C, and D (D. H. Lee et al., 2016; Lee,
Bertran, Kwon, & Swayne, 2017). The clade 2.3.4.4 group A H5 HPAIVs were first
identified in China in 2013 as H5N8 (Fan et al., 2014) and were subsequently responsible
for outbreaks in South Korea and Japan in 2014 (Kanehira et al., 2015; Lee et al., 2014).
This sub-lineage disseminated to Eastern Siberia, Eurasia, and North America through
bird migration (Global Consortium for H5N8 and Related Influenza Viruses, 2016; Lee et
al., 2015; M. S. Lee et al., 2016; Marchenko et al., 2015; Tanikawa et al., 2016). The clade
2.3.4.4 group B H5 HPAIVs were identified in China and Korea in 2013–2014 (Fan et al.,
Sharshov, et al., 2017; Pohlmann et al., 2017). While group A and group B HPAIVs have
expanded worldwide, group C H5N6 HPAIVs have been mainly isolated from 2013 to
2017 in Asian countries such as China, Japan, Korea, Laos, and Vietnam (Hiono et al.,
2017; Mok et al., 2015; Okamatsu et al., 2017; Takemae et al., 2017; Wong et al., 2015).
Group D H5N6 HPAIVs have been enzootic in China and Vietnam (Chu et al., 2016; Lee
In Vietnam, the first wave of massive outbreaks started on December 23, 2003,
by H5N1 HPAIV of the clade 1 Gs/Gd lineage (Nguyen, 2005) with most poultry
outbreaks during this period being caused by clade 1 viruses (Wan et al., 2008).
Subsequently, clade 2.3.4 viruses gradually replaced the clade 1 viruses in parts of
Northern and Central Vietnam between 2005 and 2010 with the descendants of the clade
1 viruses, so-called clade 1.1, being followed by clade 1.1.2 viruses, which have continued
to predominate in Southern Vietnam (Nguyen et al., 2012; Wan et al., 2008). Three
genetically distinct subgroups of clade 2.3.2.1 (2.3.2.1a, 2.3.2.1b, and 2.3.2.1c) replaced
clade 2.3.4 viruses in parts of Northern and Central Vietnam between 2010 and 2012
(Creanga et al., 2013). Clade 2.3.2.1c viruses have spread to parts of Southern Vietnam
Since the first reported case of human infection with H5N1 that occurred in
1997 in Hong Kong, more than 900 cases in 16 countries have been reported by 2015
with half of them being fatal (Lai et al., 2016). Subsequently, as of March 6, 2019, the
incidence of human cases of H5N1 infection has been decreasing with none being
human cases of H5N1 virus infections were recorded between 2003 and 2014 with 64 of
been no reports of human cases of H5N1 or H5N6 HPAIV infections since reports in
Vietnam in 2015. Recently, the H5N6 HPAIV clade 2.3.4.4 groups C and D have been
responsible for 23 human infections, leading to seven deaths in China since 2014
(https://apps.who.int/iris/bitstream/handle/10665/279855/AI-20190201.pdf?sequence=4
&isAllowed=y).
Southern Vietnam in July 2018. The entire genome of the isolate was sequenced for
This article is protected by copyright. All rights reserved.
detailed molecular analyses, including a comparison of the new isolate with recent
On July 25, 2018, six Muscovy ducks on a farm in Long An Province, Vietnam,
died suddenly after exhibiting clinical signs of HPAI. The farm included 1,202 Muscovy
ducks, which had not been vaccinated against HPAIV and had been scavenging in rice
fields during the days. A spleen tissue sample was collected from a dead Muscovy duck
on an affected farm, and the tissue homogenate was inoculated into 10-day-old
embryonated chicken eggs in a BSL-3 facility at the National Institute of Animal Health,
RNA was purified from the allantoic fluid with the RNeasy Micro kit (Qiagen,
Valencia, CA, USA). The sequence library was prepared using an NEBNext Ultra RNA
Library Prep Kit for Illumina (NEB, Valencia, CA, USA) and subjected to whole-genome
sequence analysis using a MiSeq system (Illumina, San Diego, CA, USA) with MiSeq
Reagent Kit v2 as described previously (Tanikawa et al., 2019). The sequence data were
analyzed using FluGAS software v0.9.0 (World Fusion, Tokyo, Japan) to create
EPI1363700-7).
Basic Local Alignment Search Tool (BLAST) search of Global Initiative on Sharing All
Influenza Data (GISAID). For phylogenetic analysis, all AIV sequences registered to
This article is protected by copyright. All rights reserved.
GISAID as of March 8, 2019 were downloaded and aligned using MAFFT (Katoh et al,
2013). Non-coding regions of each segment were trimmed and sequences that included
ambiguous and mixed bases were removed. Segment sets in the maximum likelihood
(ML) tree were screened as having 0.995 (H5) or 0.970 (PB2, PB1, PA, NP, N6, M and
NS) sequence similarity by CD-HIT and added to the sequences of the top 250 blast hits
for AI470 on GISAID. The number of analyzed sequences of each segment included 1399
for PB2, 1414 for PB1, 1342 for PA, 701 for H5, 1235 for NP, 348 for N6, 1008 for M, and
1104 for NS. Phylogenetic trees of the eight gene segments of HPAIVs were constructed
with MEGA7.0.26 software using the ML method. Bootstrap analysis was performed
with 200 replicates. For tanglegrams, all segment sets of strains belonging to clade
2.3.4.4 in the ML tree of H5 genes were downloaded from GISAID. The sequence sets
were added the sequences of each top 250 blast hits for AI470 and human-origin H5N6
HPAIVs on GISAID with the duplicate strains being removed. Tanglegrams were
generated based on the phylogenetic trees using the Dendroscope 3 program. A Bayesian
maximum clade credibility (MCC) tree of the hemagglutinin (HA) gene was constructed
from the sequences of the top 100 blast hits for AI470 on GISAID using BEAST 1.8.4
software. A world map was generated using SPREAD software to perform discrete
(https://www.google.co.jp/intl/ja/earth/).
The new isolate was determined to be an H5N6 subtype and was designated as
A/ Muscovy duck/Long An/AI470/2018 (H5N6; AI470). The amino acid sequence of the
HA protein cleavage site of AI470 was RERRRKRGLF, which indicated it was an HPAIV
according to the molecular criteria of the World Organisation for Animal Health (OIE)
Terrestrial Animal Health Code (World Organisation for Animal Health, 2018b). A
BLAST search revealed that the new isolate had the highest HA gene homology with
followed by A/Guangxi/32797/2018 (H5N6), which was isolated on October 25, 2018, from
specimens of a patient. The HA gene homology of AI470 compared to that of other H5N6
HPAIVs from human patients is shown in Table S1. The other seven genomic segments
constructed using the eight genomic segments. The phylogenetic trees revealed that the
HA segment of AI470 was derived from H5N6 HPAIV clade 2.3.4.4 group C in South
China in 2015–2018 (Figure 1). The other segments, except polymerase basic protein 2
(PB2), were shown by phylogenic analyses to have origins similar to that of HA (Figure 2
and Figures S1–S7). In contrast, the PB2 segment of AI470 was derived from H5N6
HPAIV strains in China in 2014 to 2017, which initially originated from an H6 avian
influenza virus that circulated among wild birds. Intriguingly, five HPAIVs of human
origin in South China from 2017 to 2018 clustered with AI470 in the HA tree, with four of
them also demonstrating results similar to those of the other seven segments. Analysis
based on the genetic nomenclature defined by Bi et al. demonstrated that AI470 and the
four HPAIVs of human origin in South China from 2017 to 2018 were characterized to be
of the G1.1 genotype (Bi et al., 2016). The transmission of four HPAIVs of human origin
to humans suggests that other H5N6 HPAIV strains in this cluster may possess the
among AI470 and the four human-origin H5N6 HPAIVs in the same cluster of a G1.1
genotype (Table 1). The substitutions of D101N and T160A in the HA protein (using H3
human-like receptor (Gao et al., 2018; Su, Yang, Zhang, Jia, & Tien, 2008), and these
substitutions are commonly found in clade 2.3.4.4. The receptor binding site (RBS) of HA
has conserved secondary elements (130-loop, 190-helix, and 220-loop), and substitutions
proximal to the RBS can affect receptor binding specificity (Tharakaraman et al., 2013;
site at amino acid 129 of AI470. This new glycosylation site adjoins the 130-loop of the
RBS of HA and may affect receptor specificity. In addition, this putative glycosylation
site is characteristic of most H5N6 viruses of human origin and may play a role in human
adaptation. Some amino acid substitutions that are considered to increase virulence in
mammalian hosts such as mice were shared by AI470, recent H5N6 HPAIVs from
patients, and closely related poultry isolates. Those substitutions include a deletion in
the neuraminidase (NA) stalk region (position 59–69), N30D and T215A substitutions in
the matrix protein 1 (M1), and a S42 substitution and amino acid deletion (position
Liu, Wu, & Liu, 2008; Matsuoka et al., 2009). AI470 also shared a G309D substitution in
PB2, which is known to enhance polymerase activity in mammalian cells (Hatta, Gao,
Halfmann, & Kawaoka, 2001; Subbarao, London, & Murphy, 1993; Yamaji et al., 2015).
AI470 did not possess PB2 substitutions E627K and D701N, but some of the human
Our genetic analyses revealed that the H5N6 HPAIV AI470 was closely related
to H5N6 HPAIV strains isolated from patients in China and that it was introduced into
Southern Vietnam (Figure 3). Vietnam has been at risk of HPAIV invasion because of its
2006). After crossing the border from China into Vietnam, HPAIVs have usually
disseminated from the North to the South through the poultry trade and other human
activities (Nguyen et al., 2017). It is surprising that there has been no report prior to our
study of an H5N6 HPAIV in Vietnam that is genetically related to AI470. This suggests
that the dissemination route of AI470 may have been different than that described
(H5N6), which was isolated from wild bird in Russia, it is conceivable that AI470 was
have not been determined, this hypothesis should be further scrutinized. Another
potential route of virus transmission may be the illegal transportation of live birds or
poultry products, which was observed in cases of HPAIV being isolated from eagles that
had been illegally smuggled into Belgium (Borm et al., 2005) and from raw poultry
illegally imported from China into Japan by airline passengers (Shibata et al., 2018).
Since the outbreak by AI470, eight outbreaks of H5N6 viruses in Vietnam have been
reported to OIE through 2018 (Table S2). Therefore, AI470-like viruses may have
already been circulating in Vietnam, and the genetic analyses of those viruses are
needed.
A470 and A470-like strains of influenza may be threats to public health since
the genetic features of A470 resemble those of human isolates. HA proteins of the viruses
belonging to clade 2.3.4.4 have the ability to bind to both avian-like (α-2,3) and
This article is protected by copyright. All rights reserved.
mammalian-like (α-2,6) receptors (Gao et al., 2018). However, in addition to this
characteristic, other mutations in the virus genome are necessary for avian HPAIVs to
important change that is required for efficient replication in humans (Moncorge, Mura, &
Barclay, 2010). It has been reported that when an H5N1 HPAIV with E627 in PB2 is
inoculated into mice, the E627K substitution becomes predominant by 6 days post
infection (dpi) (Min et al., 2013). Therefore, while AI470-like strains in poultry may not
possess the E627K substitution in PB2, it may be possible for them to acquire this
Vietnam. For instance, the H5N6 HPAIV A/Muscovy duck/Quang Ninh/4c111/2013 of the
G2 genotype was isolated in April 2013 in Vietnam and has the same origin as
A/Sichuan/26221/2014 (H5N6; Sichuan), which was isolated in April 2014 from a patient
in South China that had an H5N6 HPAIV infection (Figure 2 and Figure S8) (Pan et al.,
through January 2016 (Nguyen et al., 2017). H5N6 HPAIV A/Muscovy duck/Quang
acids 80–84 in NS1. These changes are related to the adaptation of the Sichuan strain for
replication in mammalian cells (Table S3). A D701N substitution in PB2 was unique to
the Sichuan strain, while deletions of HA amino acid E130 and amino acids 59–69 in the
NA stalk region were not detected in either strain (Long et al., 2008; Matsuoka et al.,
2009). The putative glycosylation site at HA amino acid 129 caused by the deletion of HA
E130 was found in 19 of 27 human H5N6 HPAIVs evaluated (Table S3) and in 4 of 19
strains that belonged to the G1.1 genotype and have infected humans for at least one
year. In contrast, 8 of 27 virus strains without the glycosylation site at amino acid 129
were scattered among six different genotypes and have been known to only sporadically
infect humans. These results suggested that the additional glycosylation site at amino
acid 129 of HA may be related to the adaptation of HPAIVs for replication in humans;
H5N6 HPAIVs may possess a greater potential for human infection than Sichuan-like
H5N6 HPAIVs. In fact, human infection by Sichuan H5N6 HPAIVs has not been
reported since 2014, and human infection by Sichuan-like H5N6 HPAIVs has never been
reported.
to the H5N6 HPAIVs that infect humans in China, suggesting that AI470 may be a risk
for human infection, at least based on genetic analyses. Genome analyses of HPAIVs
may be useful for evaluating the potential of human infection by HPAIVs based on
known genetic signatures. However, genetic analyses alone may be insufficient for the
evaluation since such signatures may act in combination with other unknown changes
that may be involved in the potential for human infection. Therefore, further studies,
such as receptor binding assays and infection and transmissibility studies in ferrets, are
needed to properly evaluate the potential of AI470 and AI470-like virus for human
Acknowledgments
sequences that were used in the present study to GISAID. The current research was
supported by the Japan Initiative for Global Research Network on Infectious Diseases
Japan and the Japan Agency for Medical Research and Development (AMED; grant
number 18fm0108008) and by a joint grant from a pilot program for international
collaborative research between the Agriculture, Forestry and Fisheries Research Council
(AFFRC), the Ministry of Agriculture, Forestry and Fisheries (MA) of Japan and the
Ethics
References
evolution and prevalence of H5N6 avian influenza viruses in China. Cell Host &
Chu, D. H., Okamatsu, M., Matsuno, K., Hiono, T., Ogasawara, K., Nguyen, L. T., . . .
influenza viruses circulating in live bird markets with intervention in the center
doi:10.1016/j.vetmic.2016.07.016
Creanga, A., Thi Nguyen, D., Gerloff, N., Thi Do, H., Balish, A., Dang Nguyen, H., . . .
doi:10.1016/j.virol.2013.06.005
Fan, S., Deng, G., Song, J., Tian, G., Suo, Y., Jiang, Y., . . . Chen, H. (2009). Two amino
doi:10.1016/j.virol.2008.11.044
pathogenic H5N8 avian influenza virus isolated from a wild duck in China.
Gao, R., Gu, M., Liu, K., Li, Q., Li, J., Shi, L., . . . Liu, X. (2018). T160A mutation-induced
Global Consortium for H5N8 and Related Influenza Viruses. (2016). Role of migratory
wild birds in the global spread of avian influenza H5N8. Science, 354, 213–217.
Gu, M., Liu, W., Cao, Y., Peng, D., Wang, X., Wan, H., . . . Liu, X. (2011). Novel
doi:10.3201/eid/1706.101406
Hatta, M., Gao, P., Halfmann, P., & Kawaoka, Y. (2001). Molecular basis for high
isolated from wild and captive birds in the winter season of 2016-2017 in
doi:10.1111/1348-0421.12506
Jiao, P., Tian, G., Li, Y., Deng, G., Jiang, Y., Liu, C., . . . Chen, H. (2008). A
doi:10.1128/JVI.01698-07
Kanehira, K., Uchida, Y., Takemae, N., Hikono, H., Tsunekuni, R., & Saito, T. (2015).
Katoh, K., Standley, D.M. (2013). MAFFT multiple sequence alignment software version
30 (4):772-780. doi:10.1093/molbev/mst010
Lee, D. H., Bahl, J., Torchetti, M. K., Killian, M. L., Ip, H. S., DeLiberto, T. J., & Swayne,
1283–1285. doi:10.3201/eid2207.160048
Lee, D. H., Bertran, K., Kwon, J. H., & Swayne, D. E. (2017). Evolution, global spread,
Lee, D. H., Sharshov, K., Swayne, D. E., Kurskaya, O., Sobolev, I., Kabilov, M., . . .
virus in wild aquatic birds, Russia, 2016. Emerging and Infectious Diseases, 23,
359–360. doi:10.3201/eid2302.161252
Lee, M. S., Chen, L. H., Chen, Y. P., Liu, Y. P., Li, W. C., Lin, Y. L., & Lee, F. (2016).
Highly pathogenic avian influenza viruses H5N2, H5N3, and H5N8 in Taiwan in
Lee, Y., Kang, H., Lee, E., Song, B., Jeong, J., Kwon, Y., . . . Lee, H. (2014). Novel
Long, J. X., Peng, D. X., Liu, Y. L., Wu, Y. T., & Liu, X. F. (2008). Virulence of H5N1
doi:10.1007/s00705-015-2570-4
of the hemagglutinin influence the virulence of influenza H5N1 viruses for mice.
Min, J. Y., Santos, C., Fitch, A., Twaddle, A., Toyoda, Y., DePasse, J. V., . . . Subbarao, K.
(2013). Mammalian adaptation in the PB2 gene of avian H5N1 influenza virus.
Mok, C. K., Da Guan, W., Liu, X. Q., Lamers, M. M., Li, X. B., Wang, M., . . . Yang, Z. F.
doi:10.3201/eid2112.150809
Moncorge, O., Mura, M., & Barclay, W. S. (2010). Evidence for avian and human host cell
factors that affect the activity of influenza virus polymerase. Journal of Virology,
Nguyen, D. T., Jang, Y., Nguyen, T. D., Jones, J., Shepard, S. S., Yang, H., . . . Davis, C.
doi:10.1128/jvi.01708-16
Nguyen, T., Rivailler, P., Davis, C. T., Hoa, D. T., Balish, A., Dang, N. H., . . . Donis, R. O.
doi:10.1016/j.virol.2012.06.021
Press.
Okamatsu, M., Ozawa, M., Soda, K., Takakuwa, H., Haga, A., Hiono, T., . . . Kida, H.
doi:10.3201/eid2304.161957
doi:10.1016/j.jinf.2015.06.009
Pohlmann, A., Starick, E., Harder, T., Hoper, D., Globig, A., Staubach, C., . . . Beer, M.
(2017). Outbreaks among wild birds and domestic poultry caused by reassorted
Qi, X., Cui, L., Yu, H., Ge, Y., & Tang, F. (2014). Whole-genome sequence of a reassortant
H5N6 avian influenza virus isolated from a live poultry market in China, 2013.
Shibata, A., Hiono, T., Fukuhara, H., Sumiyoshi, R., Ohkawara, A., Matsuno, K., . . .
doi:10.1111/tbed.12726
Su, Y., Yang, H. Y., Zhang, B. J., Jia, H. L., & Tien, P. (2008). Analysis of a point
doi:10.1007/s00705-008-0255-y
Subbarao, E., London, W., & Murphy, B. (1993). A single amino acid in the PB2 gene of
1761–1764.
Takemae, N., Tsunekuni, R., Sharshov, K., Tanikawa, T., Uchida, Y., Ito, H., . . . Saito, T.
viruses affected Japan during the winter of 2016-2017. Virology, 512, 8–20.
doi:10.1016/j.virol.2017.08.035
wild bird fecal specimen and a chicken in Japan in 2014. Microbiology and
Tanikawa, T., Uchida, Y., Takemae, N., Tsunekuni, R., Mine, J., Liu, M.T., . . . Saito, T.
doi:10.1007/s00705-018-4102-5
Tharakaraman, K., Raman, R., Viswanathan, K., Stebbins, N. W., Jayaraman, A.,
1475–1485. doi:10.1016/j.cell.2013.05.035
Van Borm, S., Thomas, I., Hanquet, G., Lambrecht, B., Boschmans, M., Dupont, G., . . .
van den Berg, T. (2005). Highly pathogenic H5N1 influenza virus in smuggled
doi:10.1371/journal.pone.0003462
Wong, F. Y., Phommachanh, P., Kalpravidh, W., Chanthavisouk, C., Gilbert, J.,
doi:10.3201/eid2103.141488
World Organisation for Animal Health. (2018a). OIE Situation Report for avian
http://www.oie.int/en/animal-health-in-the-world/update-on-avian-influenza/2018
World Organisation for Animal Health. (2018b). Influenza A Cleavage Sites, version 31
January 2018;
http://www.offlu.net/fileadmin/home/en/resource-centre/pdf/Influenza_A_Cleavag
e_Sites.pdf
hemagglutinin gene to those of H5N1 viruses from the 1997 outbreaks in Hong
Yamaji, R., Yamada, S., Le, M. Q., Li, C., Chen, H., Qurnianingsih, E., . . . Kawaoka, Y.
H5N1 influenza viruses in human lung epithelial cells. Journal of Virology, 89,
3947–3956. doi:10.1128/JVI.03328-14
Yang, Z., Wei, C., Kong, W., Wu, L., Xu, L., Smith, D., & Nabel, G. (2007). Immunization
Host Strain
D101N E130 130 Loop T160A 190 Helix 220 Loop 59-69 Deletion H275K G309D E627K D701N N30D T215A S31N P42S 80-84 Deletion 92E
Avian A/Muscovy duck/Long An/AI470/2018 N del‡ SSGVSA A EQADLYKN SQVNGQRG Yes H D E D D A S S Yes E
A/chicken/Guangdong/G1012/2018 N del SSGVSA A EQADLYKN SQVNGQRG N.A§ N.A N.A N.A N.A N.A N.A N.A N.A N.A N.A
A/duck/Guangdong/G1378/2018 N del SSGVSA A EQTDLYKN SQVNGQRG N.A N.A N.A N.A N.A N.A N.A N.A N.A N.A N.A
†
H3 numbering was used.
‡
Deletion at this position creates a putative N-glycosylation site at N129
§
Sequences were not available.
pathogenic avian influenza viruses (HPAIVs). The trees were constructed with
analysis was performed with 200 replicates. The yellow square of the left tree is
enlarged in the right tree. AI470 is indicated in red, human isolates are indicated in
blue, and human isolates of the G1.1 genotype are indicated by the blue circles.
avian influenza viruses (HPAIVs). Phylogenetic trees were constructed using the
maximum likelihood (ML) method and then used to generate the tanglegrams using
the Dendroscope 3 program. The groups of H5N6 isolates in this study were
classified according to the tanglegrams. AI470, human H5N6 isolates, and HPAIVs
isolated in Vietnam that are adjacent to each other in the trees are connected by a
line. Colors for the different taxa and lines are as follows: AI470 and its related
H5N6 human isolates, red; H5N6 HPAIVs isolated in Vietnam, green; G1 genotype,
light purple; G1.1 genotype, orange; G1.1b genotype, yellow-green; G1.2 genotype,
follows: H5 HPAIVs, black; H6N6 avian influenza viruses (AIVs), blue; H9N2 or
Vietnam. The connection of areas was based on the Bayesian maximum clade
credibility (MCC) tree of the hemagglutinin (HA) gene constructed using BEAST
1.8.4 software. The world map was generated using SPREAD software to perform
discrete phylogeographical analyses, and the results were visualized using Google Earth
(https://www.google.co.jp/intl/ja/earth/).