﻿Chinese and Global Distribution of H9 Subtype Avian
Influenza Viruses
Wenming Jiang.
, Shuo Liu.
, Guangyu Hou, Jinping Li, Qingye Zhuang, Suchun Wang, Peng Zhang,
Jiming Chen*
The Laboratory of Avian Disease Surveillance, China Animal Health and Epidemiology Center, Qingdao, China
Abstract
H9 subtype avian influenza viruses (AIVs) are of significance in poultry and public health, but epidemiological studies about
the viruses are scarce. In this study, phylogenetic relationships of the viruses were analyzed based on 1233 previously
reported sequences and 745 novel sequences of the viral hemagglutinin gene. The novel sequences were obtained through
large-scale surveys conducted in 2008-2011 in China. The results revealed distinct distributions of H9 subtype AIVs in
different hosts, sites and regions in China and in the world: (1) the dominant lineage of H9 subtype AIVs in China in recent
years is lineage h9.4.2.5 represented by A/chicken/Guangxi/55/2005; (2) the newly emerging lineage h9.4.2.6, represented
by A/chicken/Guangdong/FZH/2011, has also become prevalent in China; (3) lineages h9.3.3, h9.4.1 and h9.4.2, represented
by A/duck/Hokkaido/26/99, A/quail/Hong Kong/G1/97 and A/chicken/Hong Kong/G9/97, respectively, have become
globally dominant in recent years; (4) lineages h9.4.1 and h9.4.2 are likely of more risk to public health than others; (5)
different lineages have different transmission features and host tropisms. This study also provided novel experimental data
which indicated that the Leu-234 (H9 numbering) motif in the viral hemagglutinin gene is an important but not unique
determinant in receptor-binding preference. This report provides a detailed and updated panoramic view of the
epidemiological distributions of H9 subtype AIVs globally and in China, and sheds new insights for the prevention of
infection in poultry and preparedness for a potential pandemic caused by the viruses.
Citation: Jiang W, Liu S, Hou G, Li J, Zhuang Q, et al. (2012) Chinese and Global Distribution of H9 Subtype Avian Influenza Viruses. PLoS ONE 7(12): e52671.
doi:10.1371/journal.pone.0052671
Editor: Martin Beer, Friedrich-Loeffler-Institut, Germany
Received August 16, 2012; Accepted November 19, 2012; Published December 21, 2012
Copyright: ß 2012 Jiang et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work is supported by the Sci-tech Basic Work Project of Ministry of Science and Technology (SQ2012FY3260033). The funder had no role in study
design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: jmchen66@yahoo.cn
. These authors contributed equally to this work.
Introduction
Influenza viruses in the family Orthomyxoviridae are classified into
three types, A, B, and C. Type A influenza viruses are further
classified into different subtypes based on antigenic differences in
the viral surface glycoproteins, hemagglutinin (HA) and neur-
aminidase (NA). Currently, 16 HA subtypes (H12H16) and 9 NA
subtypes (N12N9) of influenza viruses have been isolated from
birds, and a novel HA subtype (H17) and a novel NA subtype
(N10) of influenza viruses have been identified in bats [1]. Among
the 16 HA subtypes of avian influenza viruses (AIVs), H9 subtype
(mainly H9N2 subtype) in domestic fowls and wild birds have been
reported in various regions including North America, Europe,
Asia, Africa and the Pacific [2­9].
Usually, H9 subtype AIVs cause mild clinical signs in birds;
however, the infection can be exacerbated by a secondary
bacterial infection [10­21]. In recent years, H9 subtype AIVs
circulating in China aroused worldwide concerns, because several
cases of human infections in China have been identified since the
end of the 1990s, and some of the viruses displayed human
influenza virus-like receptor specificity [11,22­29]. H9 subtype
AIVs viruses are thus considered one of the most likely candidates
to cause a new influenza pandemic in humans [22­27]. Unlike
human infection with the H5 subtype highly pathogenic avian
influenza (HPAI) virus that usually causes a severe infection, the
associated disease symptoms in all human cases of H9 subtype
AIV infection have been mild, with no evidence of human-to-
human transmission [22­27,30].
As an essential component of the global strategy for
pandemic preparedness, selection of candidate H5 and H9
influenza vaccine viruses has been coordinated by the World
Health Organization (WHO) in recent years, through epidemi-
ological surveys and phylogenetic analysis [25,26]. However, as
clearly stated in two relevant reports issued by the WHO in
2011, data characterizing recent H9 subtype AIVs circulating in
the world are limited, and the majority of the viruses that have
been sequenced belong to the G1 clade, represented by A/
quail/Hong Kong/G1/97, or the Y280/G9 clade, represented
by A/duck/Hong Kong/Y280/97 or A/chicken/Hong Kong/
G9/97 [25,26].
The goal of this study was to elucidate the host, spatial and
phylogenetic distributions of H9 subtype AIVs circulating in
China and in the world in recent years, through epidemiological
surveys and phylogenic analysis of viral HA gene sequences. This
viral gene is vital for viral pathogenicity, antigenicity and host
range, although some evidence indicates that these traits are
polygenic [31].
PLOS ONE | www.plosone.org 1 December 2012 | Volume 7 | Issue 12 | e52671
Results
Swab Sample Detection
A total of 31406 swab samples were collected from 299 live
bird markets, 26 slaughtering houses, 303 backyard flocks and
348 poultry farms through the eight surveys conducted in
200822011. Among them, 57 were positive for Newcastle
disease viruses (NDVs), and 1349 were positive for AIVs. Among
the 1349 AIV positive samples, 950 were positive for H9 subtype
AIVs and 76 were positive for H5 subtype AIVs. The prevalence
of H9 subtype AIVs in chicken samples, 3.50% (884/25233), was
significantly higher than the prevalence in samples from
waterfowl including ducks and geese, 1.08% (66/6127), with
P,0.05. None of the 46 pigeon samples were positive for AIVs.
The prevalence of H9 subtype AIVs in the samples from live
bird markets, 6.84% (655/9558), or in the samples from
slaughtering houses, 3.81% (47/1232), was significantly higher
than that in the samples from backyard flocks, 1.82% (172/
9464), or from poultry farms, 0.68% (76/11152), all with
P,0.05. On average, infection of H9 subtype AIVs was
identified in 35.12% (105/299) of the sampled live bird markets,
34.62% (9/26) of the sampled slaughtering houses, 17.49% (53/
303) of the sampled backyard flocks, and 11.21% (39/348) of the
sampled poultry farms. The prevalence of H9 subtype AIVs in
Northern China, 2.85% (350/12273), was similar to its
counterpart in Southern China, 3.14% (600/19133), with
P.0.05.
Panorama of the Global Phylogenetic Distribution of H9
Subtype AIVs
We analyzed the sequence phylogenetic relationships of this
report using the neighbor-joining method and the maximum
likelihood method (G+I), and found the results of the two methods
were of little difference. Therefore, only the results of the
neighbor-joining method were presented.
As of September 17, 2011, HA sequences (.900 bp) of 1233
H9 subtype AIVs with clear background were available in
GenBank, namely that the source, isolation place and isolation
time of the sequences were available and they have not been
suspected by any publications [14,16]. Of these, 55 were reported
by our laboratory. The remaining 1178 sequences were submitted
by other research groups.
Phylogenetic analysis of the 1178 sequences revealed four
primary lineages, h9.12h9.4, some secondary and tertiary lineages
(Figure S1). The temporal and spatial distribution of the lineages is
showed in Table S1and Figure S2.
Genetic distances in the HA1 subunit of the viral HA gene
among the lineages are showed in Table S2. In general, the genetic
distances were 20.00% to 24.39% between primary lineages,
11.93% to 17.27% between secondary, and 5.81% to 18.28%
between tertiary lineages.
At the primary lineage level, lineages h9.1 and h9.2 corre-
sponded to only two viruses isolated in 1966 in North America and
three viruses isolated in the 1990s in North America, respectively.
Lineage h9.3 corresponded to 140 viruses isolated in the Eastern
Hemisphere, and 33 viruses isolated in North America. Lineage
h9.4 corresponded to hundreds of viruses from the Eastern
Hemisphere. Though lineage 9.4 harbored many more sequences
than lineage h9.3, the latter was more widely distributed, with
viruses from Asia, Europe, Africa, the Pacific and North America,
while lineage h9.4 circulated exclusively in Asia. Lineage h9.3 also
had a longer circulation history (197622010) than lineage h9.4
(199422011).
At the secondary lineage level, most avian H9N2 viruses
isolated in recent years in China, and those isolated in the
Asian countries west to China including Pakistan, India, Iran,
and Israel, belonged to the secondary lineages h9.4.2 and
h9.4.1, respectively. Previously, these two secondary lineages
were designated as the Y280/G9-like and G1-like viruses,
respectively [25,26]. Lineage h9.4.2 comprised more sequences
than lineage h9.4.1, which further comprised more sequences
than lineage h9.3.3 represented by A/duck/Hokkaido/26/99.
However, lineage h9.3.3 was more widely distributed than
h9.4.2 and h9.4.1, with isolates from Asia, Europe, Africa, the
Pacific and North America. Lineages h9.4.1 circulated almost
exclusively in Asian countries and h9.4.2 in China. In addition,
lineages h9.3.3, h9.4.1 and h9.4.2 had a similar circulation
history, from approximately the 1990s to the present.
At the tertiary lineage level, lineage h9.3.3.2 was distributed
more widely than the other tertiary lineages with isolates from
Asia, Europe, Africa, the Pacific and North America.
Taken together, at the primary, secondary and tertiary lineage
levels, lineages h9.3, h9.3.3, h9.3.3.2 should be globally dominant
compared to their counterparts. In addition, lineages h9.4.1 and
h9.4.2, namely the G1-like and Y280/G9-like viruses, were also
assumed to be globally dominant because they were prevalent in
many countries in Asia and in many provinces of China (a country
covering a large land area), respectively.
Host Distribution and Transmission of Different H9
Subtype AIVs
Most (.90%) of the viruses in lineages h9.4.1, h9.4.2 and
h9.3.3.1 were from domestic terrestrial birds like chickens and
quails, while most (.90%) in lineages h9.3.1 and h9.3.3.2 were
from domestic or wild waterfowl (Figure S1). Furthermore, most
branches within lineages h9.4.1, h9.4.2 and h9.3.3.1, namely those
mainly from domestic terrestrial birds, covered viruses from one
country only, indicating that transboundary transmission was
infrequent within these lineages. In contrast, most branches in
lineages h9.3.1 and h9.3.3.2, namely those mainly from domestic
or wild waterfowl, covered viruses from many countries, indicating
that transboundary transmission was frequent within these lineages
(Figure S2).
Phylogenetic Distribution of H9 Subtype AIVs in China
Among the 1178 H9 subtype AIVs, 833 were from China
including Hong Kong SAR with a clear background. These 833
sequences were reported by other research groups. As shown in
Table S1 and Figure S2, among these 833 viruses, 767 from
dozens of provinces in China belonged to lineage h9.4.2, 48
from Hong Kong and a city in Guangdong Province, (Shantou)
belonged to lineage h9.4.1, and the remaining 18 belonged to
other lineages including h9.3.1, h9.3.3 or h9.3.2. Therefore,
lineage h9.4.2 was more dominant than lineage h9.4.1 in
China.
As shown in Table S1 and Figure S2, 88.42% (504/570) of the
viruses in lineages h9.4.2.12h9.4.2.4 were isolated before 2007,
while 75.65% (166/193) in lineage h9.4.2.5, represented by A/
chicken/Guangxi/55/2005, were isolated in dozens of provinces
in China in 200722011, while lineage h9.4.2.6, represented by A/
chicken/Guangdong/FZH/2011, comprised 26 viruses isolated in
the provinces of Fujian and Guangdong in China in 201022011.
Therefore, lineage h9.4.2.5 was more dominant in China in recent
years than other tertiary lineages.
We obtained the HA gene sequences of 800 H9 subtype
AIVs, through an unpublished pioneer survey in 2007 and the
eight surveys reported here. Among these 800 sequences, 302
Distribution of H9 Avian Influenza
PLOS ONE | www.plosone.org 2 December 2012 | Volume 7 | Issue 12 | e52671
were not fully identical to any others. In addition, among these
800 sequences, 55 were published previously [14], and the
remaining 745 ones were first reported herein. Their GenBank
accession numbers were JN8025292JN802662,
JN8039472JN804557, FJ4345612FJ434587,
FJ8077082FJ807719, and GQ4632122GQ463227. As shown
in Figure S3, 769 of the 800 viruses belonged to lineage
h9.4.2.5, and these were isolated in dozens of provinces during
200722011. Another 22 viruses belonged to lineage h9.4.2.6
and these were isolated in five provinces (two in South China,
one in Southeast China, one in East China and one in
Northwest China) in 2011. The remaining nine viruses belonged
to lineage h9.4.2.4, and were isolated in one province in 2008.
Therefore, our data were consistent with the suggestion of
Figure S1 using the sequences reported by other research
groups. Furthermore, our data indicated that lineage h9.4.2.6
had also been prevalent in China in 2011, as it was identified
from South, Southeast, East, and Northwest China.
We separated the phylogenetic analysis of the HA gene
sequences of H9 subtype AIVs reported by others and us, in
order to minimize potential analytical bias caused by so many HA
gene sequences reported by us.
Phylogenetic Distribution of H9 Subtype AIVs in
Countries Near to China
Among the 1178 viruses, all the 65 H9 subtype AIVs from
Korea belonged to h9.3, distinct from the dominant lineage h9.4
in China. Similarly, all the viruses from Vietnam, Pakistan and
India were distinct from the dominant lineages circulating in
China in the corresponding periods. Although H9 subtype AIVs in
lineage 9.4.2 have been widely circulating in China for years, only
a few viruses within the lineage were isolated in another country,
Japan. This indicated that the transboundary spreading of H9
subtype between China and its neighboring countries occurred
infrequently. This is consistent with the finding that transboundary
transmission was infrequent within the lineages mainly from
domestic terrestrial birds, because the viruses from China and its
neighboring countries belonged to lineages mainly from domestic
terrestrial birds.
In Asian countries west to China including Pakistan, India, Iran,
and Israel, h9.4.1 has been dominant in recent years. At the
tertiary lineage level, as partially shown in Figure S2 and Table S1,
lineage h9.4.1.5 has been dominant over others in these Asian
countries since 2007, based on the limited sequences available in
GenBank.
Antigenic Analysis Using the HI Assay
Antigenic cross-reactivity of randomly selected viruses was
investigated by the HI assay. As showed in Table 1, the results
suggested that the viruses in lineages h9.4.2.3 and h9.4.2.4 did not
react well to the chicken antisera against the viruses in lineages
h9.4.2.5 and h9.4.2.6, although viruses in lineages h9.4.2.5 and
h9.4.2.6 reacted well to the antisera against viruses in lineages
h9.4.2.3 and h9.4.2.4, indicating that lineages h9.4.2.5 and
h9.4.2.6 have become somehow antigenically distinct from
lineages h9.4.2.3 and h9.4.2.4.
Prevalence of the Leu-234 Motif in the Viral HA Gene
Leu-234 (corresponding to residue 226 in H3 numbering) at the
receptor-binding site (RBS) in the HA gene of H9 subtype AIVs
likely has human influenza virus-like receptor specificity [24,32­
35]. Among the 1178 H9 subtype AIVs (Figure S2), none of the
178 viruses in lineages h9.1, h9.2 and h9.3 carried the motif Leu-
234, while nearly 75% of the viruses in lineages h9.4.1 (165/222)
and h9.4.2 (582/778) carried the motif.
Analysis of Receptor-binding Preference Using the HA
Assay
All the five NDVs and the two human influenza viruses reacted
well with the untreated goose RBCs and a2,3-specific sialidase-
treated goose RBCs with no difference in HA titers. This indicated
that the viruses did not bind to a2,3-linked sialic acid on the
surface of the goose RBCs. All the ten H5 subtype avian influenza
viruses reacted only with untreated goose RBCs, indicating that
they bound only to a2,3-linked sialic acid on the surface of the
goose RBCs. The 76 H9 subtype AIVs (72 in lineage h9.4.2.5, 3 in
lineage h9.4.2.6, 1 in lineage h9.4.2.3) reacted well with untreated
goose RBCs, but their HA titers using treated goose RBCs
declined to 02100% of their counterparts using untreated goose
RBCs (Table S3). This indicated that these H9 subtype AIVs had
different affinities for binding to a2,6-linked sialic acid. Of the 76
H9 subtype viruses, 55 (54 carrying the Leu-234 motif) reacted
with untreated and treated goose RBCs, indicating that they could
bind to a2,6-linked sialic acid on the surface of the goose RBCs.
The remaining 21 (20 carrying the Leu-234 motif) viruses reacted
only with untreated goose RBCs and not treated goose RBCs,
indicating that they bound only to a2,3-linked sialic acid on the
surface of the goose RBCs.
Discussion
As mentioned in the introduction of this report, epidemiological
studies on distribution of H9 subtype AIVs in some countries have
been reported previously. However, most of them covered limited
sequences or limited regions, not at Chinese national level or
global level. This report presents a detailed and updated
panoramic view of the distribution of H9 subtype AIVs globally
and in China, partially based on 745 novel HA gene sequences of
H9 subtype AIVs obtained through large-scale surveys. This is also
the first report using surveillance data to indicate the wide
existence of the new lineage h9.4.2.6 in China and the viral
prevalences in different sites in China. In addition, this report
provides novel experimental data to show that the Leu-234 motif
in the viral HA gene is not a unique determinant in receptor-
binding preference for a2,6-linked sialic acid.
The global views of H9 subtype AIVs reported here reflected
only a part of the reality, mainly because relevant data are scarce
for many countries and regions. Lineages of this report were
classified mainly according the genetic distances and the topology
of relevant phylogenetic trees. It should be noted that the
bootstrap values of some lineages including h9.4.1.3, h9.4.2.3
and h9.4.2.4 in Figure S1 were not high. This may result from that
some intermediate strains were covered in the analysis, as they, in
nature, could not be assigned to any lineages with confidence.
Similar phenomena have been observed in our previous phyloge-
netic analysis of other influenza A and influenza B viruses [36,37].
Furthermore, also due to the existence of some intermediate, some
viruses in a tertiary lineage were of genetic distances longer than
the genetic distances of some pairs between the secondary lineages.
Similar phenomena have been observed in phylogenetic analysis of
subtype H5 avian influenza A viruses, e.g. some viruses in clade
2.3.2 were of genetic distances longer than some pairs between
clade 2.3.2 and clade 2.3.4 [38].
This report suggests that some lineages of H9 subtype AIVs are
more prevalent in terrestrial birds including chickens and quails
than in waterfowl, e.g. .90% of the viruses in lineages h9.4.1,
h9.4.2 and h9.3.3.1 in Figure S1 were from domestic terrestrial
Distribution of H9 Avian Influenza
PLOS ONE | www.plosone.org 3 December 2012 | Volume 7 | Issue 12 | e52671
birds. These lineages may have adapted to terrestrial birds rather
than to waterfowl. Furthermore, the distinct host tropisms of the
viruses have likely led to different frequencies of transboundary
transmission, as domestic and wild waterfowl frequently share
open water bodies which can facilitate the transmission of the
viruses. In addition, wild waterfowl can also transmit viruses into
another country via migration.
The surveys revealed that the prevalence of H9 subtype AIVs
was similar in Southern and Northern China. This is different
from the distribution of H5 subtype HPAI viruses [39]. This might
be because the H9 subtype lineages dominant in China have
become more adapted to terrestrial birds. Thus, the ecological
features of Southern China with prevalent open water bodies and
waterfowl, which facilitate the transmission of H5 subtype HPAI
viruses, do not facilitate the transmission of H9 subtype viruses.
Live poultry markets have a high prevalence of infection with
H5 subtype HPAI viruses in some countries [39,40]. We also
found that the prevalence of H9 subtype of AIVs in live poultry
markets was significantly higher than in backyard flocks and
poultry farms, suggesting that live poultry markets are important
for transmission of multiple subtypes of AIVs.
The surveys indicated that, on average, infection of H9 subtype
AIVs was identified in 35.12% of the sampled live bird markets,
17.49% of the sampled backyard flocks, and 11.21% (39/348) of
the sampled poultry farms. These data suggested that the viruses
are widely circulating in poultry in China. Recent reports also
indicated that the viruses were also widely circulating in many
other countries or regions [225,7212,15,17,20,21]. For example,
in Jordan, a survey suggested that more than 50% of broiler or
layer flocks were carrying H9 subtype AIVs [18]. In South Korea,
a survey suggested that 8.09% of fecal samples of ducks and
chickens collected from live poultry markets were H9 subtype
AIVs positive [19].
The avian influenza reports issued by the WHO in 2010 and
2011 all suggested that lineages h9.4.1 and h9.4.2 have been
globally dominant in recent years [25,26]. This report provides
evidence through phylogenetic analysis that another lineage,
h9.3.3.2, is also globally dominant.
Lineage h9.4.2, or the Y280/G9-like clade, has been dominant
in China in recent years; however, we found that the lineage
recently evolved into two main lineages, h9.4.2.5 and h9.4.2.6.
Table S2 indicates that lineages h9.4.2.5 and h9.4.2.6 have
become genetically distinct from their precedent lineages h9.4.2.3
and h9.4.2.4 which have largely disappeared in China. Therefore,
in principle, it should be better to select the vaccine strains used for
prevention of the viral infection in poultry from lineages h9.4.2.5
and h9.4.2.6. However, most of the current H9 subtype AIVs used
for the production of inactivated vaccines in China, such as A/
chicken/Guangdong/SS/94, A/chicken/Shandong/6/96, A/
chicken/Shanghai/F/98, A/chicken/Guangxi/10/99, belong to
lineages h9.4.2.3 or h9.4.2.4. Interestingly, as showed in Table 1,
although viruses from the earlier lineages h9.4.3.3 and h9.4.3.4
(YB06, BZ02 and GX10) did not react well with antisera raised
against viruses from the currently prevalent lineages h9.4.3.5 and
h9.4.3.6, viruses from lineages h9.4.3.5 and h9.4.3.6 still reacted
well with antisera raised against the earlier strains. Therefore,
whether antibodies raised by vaccination can protect a host from
infection of currently circulating strains should be evaluated in the
future by neutralization tests and/or animal infection experiments,
rather than the HI assays only.
It has been reported that the Leu-234 motif in the HA gene of
H5 and H9 subtype influenza viruses is an important determinant
in receptor-binding preference for a2,6-linked sialic acid
[24,32235]. This was supported by our analysis of receptor-
binding preference using a2,3-specific sialidase-treated goose
RBCs, because most (54/76) of the viruses carrying the Leu-234
motif bound to a2,6-linked sialic acid on the surface of a2,3-
specific sialidase-treated goose RBCs. According to the distribu-
tion of the Leu-234 motif in the viral HA gene, this report
indicates that lineages h9.4.1 and h9.4.2 are of greater risk to
public health than others. Therefore, in principle, the vaccine
strains for pandemic preparedness should be selected from lineages
h9.4.1 and h9.4.2. Favorably, they should be from the tertiary
lineages h9.4.2.5 and h9.4.1.5, which are likely to be dominant in
China and the Asian countries west to China, respectively.
Although lineage h9.3.3.2 is also globally dominant, it seemed to
be less of a threat to public health, because no viruses in this
lineage carrying the Leu-234 motif. In addition, all the 11 H9
subtype AIVs identified from humans so far belonged to lineage
h9.4.1 or lineage h9.4.2 (Figure S2), except one (A/Korea/KNBP-
0028/2000) from lineage h9.3.3.1.
Interestingly, this report also indicates that the Leu-234 motif is
not a unique determinant in receptor-binding preference for a2,6-
linked sialic acid, because a few (20/76) H9 subtype viruses
carrying the Leu-234 motif did not react with a2,3-specific
sialidase-treated goose RBCs, and one virus not carrying the Leu-
234 motif reacted well with a2,3-specific sialidase-treated goose
RBCs (Table S3). Previous reports also indicated that some other
Table 1. Homologous and heterogeneous hemagglutination inhibition titers of some H9 lineagesa
.
Antisera Antigen
NX184 (h9.4.2.6) GB26 (h9.4.2.6) AK4 (h9.4.2.5) AE15 (h9.4.2.5) YB06 (h9.4.2.4) BZ02 (h9.4.2.3) GX10 (h9.4.2.3)
NX184 (h9.4.2.6) 64 16 32 16 4 4 4
GB26 (h9.4.2.6) 64 64 64 32 2 4 4
AK4 (h9.4.2.5) 64 64 128 128 16 8 32
AE15 (h9.4.2.5) 32 32 128 128 8 8 16
YB06 (h9.4.2.4) 128 64 128 64 128 32 64
BZ02 (h9.4.2.3) 64 64 64 16 32 64 64
GX10 (h9.4.2.3) 32 32 64 64 16 16 64
a
Homologous titers are in boldface; lineage designations are in parentheses; virus designations are abbreviated as follows: NX184 = A/chicken/Ningxia/NX184/2011,
GB26 = A/chicken/Guizhou/GB26/2011, AK4 = A/chicken/Anhui/AK4/2011, AK15 = A/chicken/Anhui/AK15/2011, YB06 = A/chicken/Shandong/YB06/2006, BZ02 = A/
chicken/Shandong/BZ02/2002, GX10 = A/chicken/Guangxi/10/1999.
doi:10.1371/journal.pone.0052671.t001
Distribution of H9 Avian Influenza
PLOS ONE | www.plosone.org 4 December 2012 | Volume 7 | Issue 12 | e52671
residues, like those at sites 191, 197, 198 and 238, may also affect
cell tropism [34,35].
Taken together, this report presents an updated, detailed and
panoramic view of the recent epidemiological distribution of H9
subtype AIVs in China and around the world, and sheds new
insights for prevention of the infection in poultry and preparedness
for a potential pandemic caused by the viruses.
Materials and Methods
Animal Welfare
This study was conducted according to the animal welfare
guidelines of the World Organization for Animal Health [41], and
approved by the Animal Welfare Committee of China Animal
Health and Epidemiology Center.
The Hemagglutination (HA) and Hemagglutination
Inhibition (HI) Assays
The HA and HI assays were performed using 96-well U-bottom
microtiter plates as described previously [42]. For the HA assay,
each virus stock to be tested was diluted serially in the plates using
phosphate-buffered saline (PBS, pH 7.2). Thereafter, 0.5% (v/v)
chicken or goose RBCs were added to each well, and the plates
were incubated at 4uC for 45260 min. The HA titer was the
inverse of the last dilution of the virus stock showing complete
agglutination of the RBCs. For the HI assay, each antiserum to be
tested was serially diluted in the plates using PBS, and then four
hemaglutination unit (HAU) of working solution of standard viral
antigen was added to each well of the plates, except those served as
serum and RBC controls. The plates were incubated at 37uC for
30 min. Thereafter, 0.5% (v/v) chicken or goose RBCs were
added into all wells, and the plates were incubated at 4uC for
45260 min. The HI titer was the inverse of the last dilution of sera
showing complete inhibition of the RBC agglutination.
Sampling and Virus Isolation
During 200822011, eight epidemiological surveys of AIVs were
conducted twice a year. Each survey targeted 8212 provinces
(Figure 1), from which dozens of poultry farms, backyard flocks,
slaughtering houses and live bird markets were selected at random.
For each site, usually 30 birds were selected randomly for swab
sample collection, and a total of 31406 swab samples were
collected. Swab samples were collected by taking smears from the
trachea and cloacae of the domestic fowls and placed in a transport
medium. The samples were clarified by centrifugation at 10000 g
for 5 min, and the supernatants were inoculated in 10-day-old
specific-pathogen-free (SPF) chicken embryonated eggs via the
allantoic sac route. The eggs were further incubated for four days,
and checked twice each day during the incubation period. The
dead ones were picked out and stored in a refrigerator. After the
incubation period, the allantoic fluids of the live embryos were
collected and tested by using the HA assay. All the hemaggluti-
nation-positive samples and the allantoic fluid of the dead eggs
were investigated further by RT-PCR, as described below.
Reverse Transcription-polymerase Chain Reaction (RT-
PCR)
RNA was extracted using an RNeasy Mini Kit (Qiagen, Hilden,
Germany), and amplified with the PrimeScript One-step RT-PCR
Kit (TaKaRa, Dalian, China). Most amplifications used the
primers 59-CTCAGGGAGCAAAAGCAGGGG-39 (upper) and
59-GTATTAGTAGAAACAAGGGTG TTTT-39 (down) which
cover the entire length of the HA gene [43]. Some used the
primers 59-CCAAAGAATTGCTCCACACAGA-39 (upper) and
59-GCACAAGAGATGAGGCGACA GT-39 (down) which flank
a 1500 bp segment of the HA gene of H9 subtype AIVs. RT-PCR
was performed in a 50-mL reaction using the One-Step RT-PCR
kit (TaKaRa, Dalian, China), with incubation at 50uC for 30 min
and denaturation at 94uC for 2 min, followed by 30 cycles at 94uC
for 30 s, 55uC for 30 s and 72uC for 2 min. RT-PCR products
were purified with an agarose gel DNA extraction kit (TaKaRa,
Dalian, China) and ligated into the pGEM-T Easy vector
(Promega, Shanghai, China). Positive clones were sequenced
using a Perkin-Elmer model 377 XL DNA sequencer in both
directions using the T7 and SP6 promoter primers. Subtypes of
these sequences were determined using the BLAST tool in the
Influenza Virus Resource [44]. Some RT-PCR positives targeting
the 1500 bp segment of the H9 subtype AIV gene were not
sequenced because they were from samples collected at the same
time from the same site.
Distribution Analysis
Distributions of the samples detected positive for H9 subtype
AIVs through the RT-PCR and sequence analysis in hosts, sites
and regions were analyzed according to their background
information. Differences in prevalence were examined using
a chi-square test, and P,0.05 was considered statistically
significant. The sampled provinces in Northern and Southern
China were shown in Figure 1. The Eastern Hemisphere covers
Asia, Europe, Africa and the Pacific, while the Western Hemi-
sphere means North and South America.
Phylogenetic Analysis
All the HA sequences ($900 bp) of H9 subtype AIVs that were
available in GenBank were downloaded through the Influenza
Virus Resource [44]. They were analyzed along with the viral HA
sequences obtained through the surveys reported here using the
software MEGA 5.05 (http://www.megasoftware.net/) [45].
Sequences were aligned using the software MUSCLE [46].
Genetic distances and phylogenetic relationships were calculated
by the neighbor-joining method, based on the sequences of the
HA1 subunit of the viral HA gene [47]. Nucleotide substitutions
were set under the Kimura 2-parameter model, and substitution
rates among sites were set in gamma distribution [47]. The gaps
were handled by pairwise deletion. Bootstrap values were
calculated from 1000 replicates.
Numbering of Amino Acid Residues
Amino acid residues, unless noted otherwise, were numbered
according to the HA sequence of A/chicken/Anhui/AK13/2008
(H9) with GenBank accession number FJ434582.
Classification of Lineages and Sublineages
Lineages and sublineages were classified mainly according to
genetic distances and topology of the phylogenetic trees, as
described in two previous reports [14,36]. Primary lineage
designations began with the subtype name plus a point and
a number, e.g., h9.3 represents the third lineage of H9 subtype
AIVs. Secondary lineage designations began with their primary
lineage designations plus a point and a number, e.g., h9.3.2
represents the second sublineage of lineage h9.3. The numbers
within the lineage designations, if possible, were in order of
Western Hemisphere followed by Eastern Hemisphere in geogra-
phy, and in order of past to present in isolation time. Classification
of the sublineages within lineage h9.3 was modified here,
compared with a previous report [36], namely that lineage
Distribution of H9 Avian Influenza
PLOS ONE | www.plosone.org 5 December 2012 | Volume 7 | Issue 12 | e52671
h9.3.2 in the previous report was designated h9.3.1.3 in this
report, and h9.3.2 represented an unclassified lineage in the
previous report.
Antigenic Analysis Using the HI Assay
Antigenic cross-reactivity was investigated using the aforemen-
tioned HI assay. Polyclonal antibodies against seven randomly
selected strains of H9 subtype AIVs (two in lineage h9.4.2.6, two in
lineage h9.4.2.5, two in lineage h9.4.2.4, one in lineage h9.4.2.3)
were generated using 21 day-old SPF chickens. Four of the seven
strains, A/chicken/Ningxia/NX184/2011, A/chicken/Guizhou/
GB26/2011, A/chicken/Anhui/AK4/2011 and A/chicken/An-
hui/AK15/2011, were obtained through the surveys presented
here. The remaining three, A/chicken/Shandong/YB06/2006,
A/chicken/Shandong/BZ02/2002 and A/chicken/Guangxi/10/
1999, were obtained from a biological company (Yebio, Qingdao),
and the GenBank accession numbers of their HA gene sequences
were AJ123452 AJ12348. Each of the SPF chickens was
immunized with 0.5 mL virus-infected allantoic fluid inactivated
using 0.2% (v/v, final) formaldehyde through intramuscular
injection for only one time, and antisera were collected on the
21st day after the injection.
Analysis of Receptor-binding Preference Using the HA
Assay
Receptor-binding preference of 76 H9 subtype AIVs, ten H5
subtype AIVs, five NDVs, and two human influenza viruses was
examined using the HA assay with untreated goose RBCs and
a2,3-specific sialidase-treated goose RBCs, according to a previous
report [48]. All the 93 avian influenza or NDVs were isolated from
the samples collected through this study, except the aforemen-
tioned two, A/chicken/Shandong/YB06/2006 and A/chicken/
Shandong/BZ02/2002. The human influenza viruses, A/Qing-
dao/1038/2007(H3N2) and A/Qingdao/783/2007(H1N1), were
obtained from Qingdao Center for Disease Control. For sialidase
treatment, a 10% (v/v) suspension of goose RBCs was prepared in
Figure 1. Provinces targeted for the eight surveys in 200822011 marked with ``m'' for those in Northern China or ``.'' for those in
Southern China.
doi:10.1371/journal.pone.0052671.g001
Distribution of H9 Avian Influenza
PLOS ONE | www.plosone.org 6 December 2012 | Volume 7 | Issue 12 | e52671
PBS and a 100 ml aliquot treated with 1.25 U of recombinant
a2,3-sialidase (Takara, Japan) for 1 h at 37uC. Treated erythro-
cytes were washed twice and adjusted to a final working
concentration (0.5%, v/v) with PBS. With sialidase treatment,
a2,3-linked sialic acid was completely eliminated from the surface
of goose RBCs without damaging a2,6-linked sialic acid.
Supporting Information
Figure S1 Small version of the phylogenetic tree of 1178
H9 subtype AIVs based on HA1 subunit of the viral
hemagglutinin gene sequences. Lineage designations are to
the right of relevant branches, and bootstrap values are at relevant
nodes. The representative virus of each lineage is marked with an
asterisk. This figure is too large, and it actual size can be viewed
using Windows-Picture-&-Fax Viewer via pressing the keys ``Ctrl''
and ``A'' simultaneously.
(TIF)
Figure S2 Phylogenetic distribution of 1178 H9 subtype
AIVs in GenBank based the sequences of the HA1
subunit of the viral hemagglutinin gene. Lineage designa-
tions are to the right of the relevant branches, and bootstrap values
are at relevant nodes. This figure is too large, and it actual size can
be viewed using Windows-Picture-&-Fax Viewer via pressing the
keys ``Ctrl'' and ``A'' simultaneously.
(TIF)
Figure S3 Phylogenetic distribution of 800 H9 subtype
AIVs based the sequences of the HA1 subunit of the viral
hemagglutinin gene. The representative virus of each lineage is
marked with an asterisk. Lineage designations are to the right of
relevant branches, and bootstrap values are at relevant nodes. This
figure is too large, and it actual size can be viewed using Windows-
Picture-&-Fax Viewer via pressing the keys ``Ctrl'' and ``A''
simultaneously.
(TIF)
Table S1 The lineage, temporal and spatial distribution of 1178
H9 subtype AIVs reported to GenBank.
(DOCX)
Table S2 Genetic distances between different levels of H9
lineages.
(DOCX)
Table S3 Hemagglutination titers of 76 H9 viruses against goose
RBCs untreated or treated with a2,3-specific sialidase.
(DOCX)
Acknowledgments
We thank Dr. Jiwang Chen in University of Illinois at Chicago for his
precious comments and suggestions.
Author Contributions
Conceived and designed the experiments: JC WJ SL. Performed the
experiments: WJ SL GH JL QZ SW PZ JC. Analyzed the data: JC SL WJ.
Contributed reagents/materials/analysis tools: JC. Wrote the paper: JC
WJ SL.
References
1. Tong S, Li Y, Rivailler P, Conrardy C, Castillo DA, et al. (2012) A distinct
lineage of influenza A virus from bats. Proc Natl Acad Sci U S A 109:
426924274.
2. Baumer A, Feldmann J, Renzullo S, Mu
¨ller M, Thu
¨r B, et al. (2010)
Epidemiology of avian influenza virus in wild birds in Switzerland between 2006
and 2009. Avian Dis 54: 875­884.
3. Hansbro PM, Warner S, Tracey JP, Arzey KE, Selleck P, et al. (2010)
Surveillance and analysis of avian influenza viruses, Australia. Emerg Infect Dis
16: 1896­1904.
4. Kang HM, Jeong OM, Kim MC, Kwon JS, Paek MR, et al. (2010) Surveillance
of avian influenza virus in wild bird fecal samples from South Korea, 2003­
2008. J Wildl Dis 46: 878­888.
5. Lee DH, Lee HJ, Lee YN, Park JK, Lim TH, et al. (2011) Evidence of
intercontinental transfer of North American lineage avian influenza virus into
Korea. Infect Genet Evo 11: 232­236.
6. Liu YH, Han CH, Wang XQ, Lin J, Ma M, et al. (2009) Influenza A virus
receptors in the respiratory and intestinal tracts of pigeons. Avian Pathol 38:
263­266.
7. Munster VJ, Baas C, Lexmond P, Waldenstro
¨m J, Wallensten A, et al. (2007)
Spatial, temporal, and species variation in prevalence of influenza A viruses in
wild migratory birds. PLoS Pathog 3: e61.
8. Ramey AM, Pearce JM, Ely CR, Guy LM, Irons DB, et al. (2010) Transmission
and re-assortment of avian influenza viruses at the Asian-North American
interface. Virology 406: 352­359.
9. Senne DA (2007) Avian influenza in North and South America, 2002­2005.
Avian Dis 51 (s1): 167­173.
10. Abolnik C, Gerdes GH, Sinclair M, Ganzevoort BW, Kitching JP, et al. (2010)
Phylogenetic analysis of influenza A viruses (H6N8, H1N8, H4N2, H9N2,
H10N7) isolated from wild birds, ducks, and ostriches in South Africa from 2007
to 2009. Avian Dis 54(sl): 313­322.
11. Chu YC, Cheung CL, Hung Leung CY, Man Poon LL, Chen H, et al. (2011)
Continuing evolution of H9N2 influenza viruses endemic in poultry in southern
China. Influenza Other Respi Viruses 5(s1): 68­71.
12. Fusaro A, Monne I, Salviato A, Valastro V, Schivo A, et al. (2011)
Phylogeography and evolutionary history of reassortant H9N2 viruses with
potential human health implications. J Virol 85: 8413­8421.
13. Ge FF, Zhou JP, Liu J, Wang J, Zhang WY, et al. (2009) Genetic evolution of H9
subtype influenza viruses from live poultry markets in Shanghai, China. J Clin
Microbiol 47: 3294­3300.
14. Ji K, Jiang WM, Liu S, Chen JM, Chen J, et al. (2010) Characterization of the
hemagglutinin gene of subtype H9 avian influenza viruses isolated in 2007­2009
in China. J Virol Methods 163: 186­189.
15. Krauss S, Obert CA, Franks J, Walker D, Jones K, et al. (2007) Influenza in
migratory birds and evidence of limited intercontinental virus exchange. PLoS
Pathog 3: e167.
16. Li CJ, Yu KZ, Tian GB, Yu DD, Liu LL, et al. (2005) Evolution of H9N2
influenza viruses from domestic poultry in Mainland China. Virology 340: 70­
83.
17. Moon HJ, Song MS, Cruz DJ, Park KJ, Pascua PN, et al. (2010) Active
reassortment of H9 influenza viruses between wild birds and live-poultry markets
in Korea. Arch Virol 155: 229­241.
18. Nagarajan S, Rajukumar K, Tosh C, Ramaswamy V, Purohit K, et al. (2009)
Isolation and pathotyping of H9N2 avian influenza viruses in Indian poultry.
Vet Microbiol 133: 154­163.
19. Negovetich NJ, Feeroz NM, Jones-Engel L, Walker D, Alam SM, et al. (2011)
Live bird markets of Bangladesh: H9N2 viruses and the near absence of highly
pathogenic H5N1 influenza. PLoS One 6: e19311.
20. Roussan DA, Khawaldeh GY, Al Rifai RH, Totanji WS, Shaheen IA (2009)
Avian influenza virus H9 subtype in poultry flocks in Jordan. Prev Vet Med 88:
77­81.
21. Xu KM, Smith GJ, Bahl J, Duan L, Tai H, et al. (2005) The genesis and
evolution of H9N2 influenza viruses in poultry from southern China, 2000 to
2005. J Virol 81: 10389­10401.
22. Butt KM, Smith GJ, Chen H, Zhang LJ, Leung YH, et al. (2005) Human
infection with an avian H9N2 influenza A virus in Hong Kong in 2003. J Clin
Microbiol 43: 5760­5767.
23. Guo YJ, Li JW, Cheng I (1999) Discovery of men infected by avian influenza A
(H9N2) virus. Chin J Exp Clin Virol 13: 105­108.
24. Matrosovich MN, Krauss S, Webster RG (2001) H9N2 influenza A viruses from
poultry in Asia have human virus-like receptor specificity. Virology 281: 156­
162.
25. (2011) Recommended composition of influenza vaccines for use in the 2012
southern hemisphere influenza season. Wkly Epidemiol Rec 86: 457­468.
26. (2011) Antigenic and genetic characteristics of influenza A(H5N1) and influenza
A(H9N2) viruses for development of candidate vaccines viruses for pandemic
preparedness - February 2011. Wkly Epidemiol Rec 86: 93­100.
27. Saito T, Lim W, Suzuki T, Suzuki Y, Kida H, et al. (2001) Characterization of
a human H9N2 influenza virus isolated in Hong Kong. Vaccine 20: 125­133.
28. Sun YP, Qin K, Wang JJ, Pu J, Tang QD, et al. (2011) High genetic
compatibility and increased pathogenicity of reassortants derived from avian
H9N2 and pandemic H1N1/2009 influenza viruses. Proc NatI Acad Sci U S A
108: 4164­4169.
29. Xu CT, Fan WX, Wei R, Zhao HK (2004) Isolation and identification of swine
influenza recombinant A/Swine/Shandong/1/2003(H9N2) virus. Microbes
Infect 6: 919­925.
Distribution of H9 Avian Influenza
PLOS ONE | www.plosone.org 7 December 2012 | Volume 7 | Issue 12 | e52671
30. Chen JM, Chen JW, Dai JJ, Sun YX (2007) A survey of human cases of H5N1
avian influenza reported by the WHO before June 2006 for infection control.
Am J Infect Control 35: 351­353.
31. Wang Q, Tao YJ (2010) Influenza: molecular virology. Norwich: Caister
Academic Press. 69­108 p.
32. Wan H, Perez DR (2007) Amino acid 226 in the hemagglutinin of H9N2
influenza viruses determines cell tropism and replication in human airway
epithelial cells. J Virol 81: 5181­5191.
33. Wan H, Sorrell EM, Song H, Hossain MJ, Ramirez-Nieto G, et al. (2008)
Replication and transmission of H9N2 influenza viruses in ferrets: evaluation of
pandemic potential. PLoS One 3(8): e2923.
34. Ha Y, Stevens DJ, Skehel JJ, Wiley DC (2002) H5 avian and H9 swine influenza
virus haemagglutinin structures: possible origin of influenza subtypes. EMBO J
21: 865­875.
35. Sorrell EM, Wan H, Araya Y, Song H, Perez DR (2009) Minimal molecular
constraints for respiratory droplet transmission of an avian-human H9N2
influenza A virus. Proc Natl Acad Sci U S A 106: 7565­7570.
36. Liu S, Ji K, Chen J, Tai D, Jiang WM, et al. (2009) Panorama phylogenetic
diversity and distribution of type A influenza virus. PLoS One 4: e5022.
37. Chen JM, Guo YJ, Wu KY, Guo JF, Wang M, et al. (2007) Exploration of the
emergence of the Victoria lineage of influenza B virus. Arch Virol 152: 415­422.
38. WHO/OIE/FAO H5N1 Evolution Working Group (2012) Continued evolu-
tion of highly pathogenic avian influenza A (H5N1): updated nomenclature.
Influenza Other Respi Viruses 6(1): 1­5.
39. Martin V, Pfeiffer DU, Zhou X, Xiao X, Prosser DJ, et al. (2011) Spatial
distribution and risk factors of highly pathogenic avian influenza (HPAI) H5N1
in China. PLoS Pathog 7: e1001308.
40. Wan XF, Dong L, Lan Y, Long LP, Xu C, et al. (2011) Indications that live
poultry markets are a major source of human H5N1 influenza virus infection in
China. J Virol 85: 13432­13438.
41. World Organization for Animal Health (2011) Terrestrial animal health code
2011, ed 1, World Organization for Animal Health, Paris.
42. Guo Y, Cheng X (1997) Influenza viruses and relevant experimental techniques.
Beijing: Three Gorges Press. 65­78 p.
43. Hoffmann E, Stech J, Guan Y, Webster RG, Perez DR (2001) Universal primer
set for the full-length amplification of all influenza A viruses. Arch Virol 146:
2275­2289.
44. Bao Y, Bolotov P, Dernovoy D, Kiryutin B, Zaslavsky L, et al. (2008) The
Influenza Virus Resource at the National Center for Biotechnology Information.
J Virol 82: 596­601.
45. Edgar RC (2004) MUSCLE: multiple sequence alignment with high accuracy
and high throughput. Nucleic Acids Res 32: 1792­1797.
46. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, et al. (2011) MEGA5:
molecular evolutionary genetics analysis using maximum likelihood, evolution-
ary distance, and maximum parsimony methods. Mol Biol Evol 28: 2731­2739.
47. Li WH (1997) Molecular evolution. Sunderland: Sinauer Associates Inc. 149­
175 p.
48. Suptawiwat O, Kongchanagul A, Chan-It W, Thitithanyanont A, Wiriyarat W,
et al. (2008) A simple screening assay for receptor switching of avian influenza
viruses. J Clin Virol 42: 186­189.
Distribution of H9 Avian Influenza
PLOS ONE | www.plosone.org 8 December 2012 | Volume 7 | Issue 12 | e52671
