www.elsevier.com/locate/yviro

Virology 340 (2

Evolution of H9N2 influenza viruses from domestic

poultry in Mainland China

Chengjun Li, Kangzhen Yu, Guobin Tian, Dandan Yu, Liling Liu, Bo Jing,

Jihui Ping, Hualan Chen*

Animal Influenza Laboratory of the Ministry of Agriculture and National Key Laboratory of Veterinary Biotechnology,

Harbin Veterinary Research Institute, Chinese Academy of Agricultural Sciences,

427 Maduan Street, Harbin 150001, People’s Republic of China

Received 2 April 2005; returned to author for revision 6 May 2005; accepted 13 June 2005

Available online 18 July 2005

Abstract

H9N2 viruses have circulated in domestic poultry in Mainland China since 1994, and an inactivated vaccine has been used in chickens to

control the disease since 1998. The present study analyzed 27 H9N2 avian influenza viruses that were isolated from chickens and ducks from

1996 to 2002. Infection studies indicated that most of the viruses replicate efficiently but none of them is lethal for SPF chickens. However,

these viruses exhibit different phenotypes of replication in a mouse model. Five viruses, including 4 early isolates and one 2000 isolate, are

not able to replicate in mice; 14 viruses replicate to moderate titers in mouse lungs and cause less than 5% weight loss, while other 8 viruses

could replicate to high titers in the lungs and 7 of them induce 10–20% weight loss of the mice on day 5 after inoculation. Most of the viruses

isolated after 1996 are antigenically different from the vaccine strain that is currently used in China. Three viruses isolated in central China in

1998 are resistant to adamantanes. Phylogenetic analysis revealed that all of the viruses originated from CK/BJ/1/94-like virus and formed

multiple genotypes through complicated reassortment with QA/HK/G1/97-, CK/HK/G9/97-, CK/SH/F/98-, and TY/WI/66-like viruses. This

study is a description of the previously uncharacterized H9N2 avian influenza viruses recently circulating in chickens and ducks in Mainland

China. Our findings suggest that urgent attention should be paid to the control of H9N2 influenza viruses in animals and to the human’s

influenza pandemic preparedness.

D 2005 Elsevier Inc. All rights reserved.

Keywords: Avian influenza viruses; H9N2; Evolutionary analysis

Introduction

Wild aquatic birds are the natural hosts of influenza A

viruses (Hinshaw et al., 1980; Sharp et al., 1993; Webster et

al., 1992). Currently, 16 HA and 9 NA subtype viruses have

been identified from avian species (Baker et al., 1987;

Fouchier et al., 2004), but only a few of them transmitted

and are widespread in the domestic poultry and caused mild

respiratory or severe systemic disease (Campitelli et al.,

2002; Guo et al., 2000; Hall, 2004; Webster et al., 2002). In

the last century, four human influenza pandemics occurred

and resulted in significant mortality and morbidity; genetic

0042-6822/$ - see front matter D 2005 Elsevier Inc. All rights reserved.

doi:10.1016/j.virol.2005.06.025

* Corresponding author. Fax: +86 451 82733132.

E-mail address: hlchen1@yahoo.com (H. Chen).

evidences indicate that these pandemic strains were partially

or entirely derived from avian influenza viruses (Kawaoka

et al., 1989; Taubenberger et al., 1997; Webster et al., 1992).

In 2003 and 2004, H5N1 highly pathogenic influenza

viruses spread across Southeast Asia countries and caused

disastrous avian influenza outbreaks, and were also trans-

mitted into humans, resulting in 32 of 42 infected cases fatal

(Hien et al., 2004). Therefore, the existing avian influenza

viruses in the poultry also bear big public health threats.

H9N2 subtype avian influenza viruses (AIV) have widely

circulated in the world since its first detection from turkeys in

Wisconsin in 1966 (Homme and Easterday, 1970). In North

America, H9N2 viruses were mainly found in shore birds and

wild ducks (Kawaoka et al., 1988; Shaw et al., 2002) and no

H9-associated disease has been reported in chickens (Perez et

005) 70 – 83

C. Li et al. / Virology 340 (2005) 70–83 71

al., 2003). In Asia, H9N2 viruses were isolated regularly

from ducks before the 1990s (Shortridge, 1992). However,

infections of H9 subtype AIV in chickens have been reported

in many Asian countries in chickens since later 1990s

(Alexander, 2000; Guo et al., 2000; Naeem et al., 1999).

H9N2 subtype viruses were isolated from pigs in 1998 (Peiris

et al., 1999a) and, subsequently, were isolated from humans

with influenza-like illness in Hong Kong and Mainland

China (Guo et al., 1999; Peiris et al., 1999b). Detailed

antigenic and genetic analysis of the H9N2 influenza viruses

from avian species in Hong Kong live bird markets revealed

the presence of three groups, represented by A/Quail/Hong

Kong/G1/97 (G1), A/Duck/Hong Kong/Y280/97(Y280),

and A/Duck/Hong Kong/Y439/97(Y439) (Guan et al.,

1999). The six internal genes of the G1 group viruses and

the PB2 and PB1 genes of A/chicken/Hong Kong/G9/97(G9)

are closely related to the corresponding gene segments of the

1997 Hong Kong H5N1 influenza viruses (Lin et al., 2000;

Shaw et al., 2002). The two human H9N2 isolates from Hong

Kong belong to the G1 group, while at least 1 human isolate

from China and 2 isolates from pigs in Hong Kong are G9-

like viruses (Guo et al., 1999; Lin et al., 2000).

In Mainland China, H9N2 AIV was firstly isolated from

chickens in Guangdong province in 1994 (Chen et al., 1994;

Li et al., 2003; Liu et al., 2003a, 2003d), and then the

viruses spread to several other southern provinces and

resulted in severe economic losses for poultry industries. An

inactivated vaccine derived from an early isolate, A/chicken/

Table 1

Antigenic analysis of H9N2 influenza viruses

Virusa HI titers with post-infection antiserab

CK/SD/6/96 CK/GD/5/97 CK/SJZ/2/9

CK/SD/6/96 160c 160 40

CK/SD/7/96 160 160 40

DK/NJ/2/97 40 160 40

CK/GD/5/97 20 160 40

CK/GD/6/97 10 80 40

CK/SJZ/2/98 <d 40 160

CK/BJ/8/98 40 320 160

CK/HN/5/98 20 320 160

CK/GX/9/99 10 80 80

CK/GX/10/99 < 80 80

CK/NX/4/99 20 320 160

CK/HLJ/35/00 < < 20

CK/GD/4/00 20 320 80

CK/HB/31/00 40 320 160

CK/SH/10/01 80 320 160

CK/JL/53/01 20 320 80

CK/GD/47/01 40 320 80

CK/HLJ/48/01 20 320 80

CK/GD/56/01 160 640 640

CK/HN/43/02 160 640 320

CK/GD/21/02 10 320 80

a Virus abbreviations—animals: chicken, CK; duck, DK. Place: Guangdong, GD

Guangxi, GX; Ningxia, NX; Heilongjiang, HLJ; Hebei, HB; Shanghai, SH; Jilin,b Antisera were 10-fold diluted.c Homologous titers are shown in bold.d <, Below the level of detection of 10.

Shandong/6/96(CK/SD/6/96), has been used in layers and

breeders in chicken farms since 1998. However, virus

infection still occurred, sometimes even in the vaccinated

flocks. And most importantly, H9N2 virus infection in pig

farms has been confirmed in several provinces, including

those in Northern China (Li et al., 2004; Xu et al., 2004). In

this study, 27 representative H9N2 viruses isolated from

chickens and ducks from different provinces in Mainland

China from 1996 to 2002 were fully sequenced, and their

antigenic variation, replication, and pathogenecity in chick-

ens and mice were also investigated. Our results suggest that

the widely distributed H9N2 avian influenza viruses may

pose severe public health threat, and attention needs to be

paid for the pandemic preparedness.

Results

Chicken study

H9N2 viruses are usually associated with egg production

drops and, sometimes, 10–20% mortality in the field. To

understand the pathogenicity of H9N2 viruses in chickens,

6-week-old white leghorn SPF chickens were i.n. inoculated

with 0.1 ml 106.0 EID50 of virus and were observed for 21

days for clinical signs and deaths. None of the 21 viruses

listed in Table 1 induce clinical signs or deaths in chickens

during the observation period of 3 weeks. Virus shedding

8 CK/GX/10/99 CK/HLJ/35/00 CK/SH/10/01

160 10 320

160 20 320

160 10 160

80 40 80

80 20 80

20 10 20

320 20 160

320 20 320

640 20 80

640 20 80

320 20 320

20 160 <

160 20 160

320 40 160

320 20 640

160 40 160

160 20 640

320 20 160

640 160 1280

640 80 1280

80 20 80

; Shandong, SD; Nanjing, NJ; Beijing, BJ; Henan, HN; Shijiazhuang, SJZ;

JL.

C. Li et al. / Virology 340 (2005) 70–8372

was detected from the tracheas of inoculated chickens, with

the titers ranging from 1.2 to 6.5 log10EID50/ml on day 3

after infection. However, only a small proportion of

inoculated chickens shed detectable viruses in the cloacal

swabs, indicating that the H9N2 influenza viruses in

Mainland China are mainly spread via the respiratory route.

All of the chickens were seroconverted on day 21 after

inoculation. These results implied that the disease or death

in the poultry farms where these viruses were isolated may

not be caused by the H9N2 avian influenza viruses alone

and may be a result of co-infection with other pathogens

(Kishida et al., 2004).

Antigenic analysis

Post-infection chicken antisera to 6 selected viruses of

each year were used to analyze the antigenic diversity of 21

Table 2

Replication of H9N2 influenza viruses in micea

Viruses and their stock titers Amino acids of the RBS in

Namee TiterHA gene (H3 number)

(logEID50/ml) 183 190 226

CK/SD/6/96 9.2 N A Q

CK/SD/7/96 9.2 N A Q

DK/NJ/1/97 8.5 N T Q

DK/NJ/2/97 9.5 N T Q

CK/GD/5/97 8.8 N V Q

CK/GD/6/97 9.2 N V Q

CK/SZ/9/97 8.6 N V Q

CK/BJ/8/98 9.4 N V L

CK/SJZ/2/98 8.8 N V Q

CK/HN/5/98 9.6 N V L

CK/GX/9/99 10.0 N T Q

CK/GX/10/99 9.9 N T L

CK/NX/4/99 9.4 N V L

CK/HLJ/35/00 9.5 H E Q

CK/GD/4/00 9.2 N T L

CK/HB/31/00 8.5 N V L

CK/FJ/25/00 9.4 N V L

CK/GD/10/00 9.8 N T L

CK/HN/26/00 8.8 N V Q

CK/JS/1/00 9.5 N T L

CK/SH/10/01 9.0 N A Q

CK/GD/56/01 9.5 N A L

CK/JL/53/01 9.0 N V L

CK/GD/47/01 9.4 N A Q

CK/HLJ/48/01 10.0 N A L

CK/HN/43/02 10.4 N A L

CK/GD/21/02 8.7 N V Q

a 6-week old female BALB/c mice were used in all experiments and were obseb Mice were inoculated i.n. with 106.0 EID50 of each virus in a 50-Al volume. O

infectivity in 10-day-old embryonated chicken eggs. Virus titers were given in units

in the undiluted samples. The infection experiments were repeated one more time

mice. Six mice per group were used; three were killed on day 3 and day 5, respe

described above.c The phenotype groups of the H9N2 avian influenza viruses were divided on th

the text.d Genotypes were determined on the basis of the diversity of the nucleotide seqe Virus abbreviation: Fujian, FJ; Jiangsu, JS.f ND: not done.

H9N2 isolates (Table 1). CK/SD/6/96 was one of the early

isolates and was used as a vaccine strain; however, the HI

test showed that 16 of 21 viruses were antigenic heterolo-

gous with this virus (the HI titers were fourfold lower than

the homologous HI titers). The antisera to CK/SJZ/2/98

could cross-react well with all of the viruses isolated after

1998, except the CK/HLJ/35/00, but not with those viruses

isolated before 1998. The cross-reaction HI titers of the

antisera of CK/HLJ/35/00 with 18 viruses were fourfold

lower than the homologous titer, indicating that CK/HLJ/35/

00 was antigenically quite different with others. CK/SH/10/

01 and CK/GX/10/99 were antigenically heterologous;

however, antisera to CK/SH/10/01 and CK/GX/10/99

reacted equally well with 19 of the 21 viruses. CK/SJZ/2/

98 and CK/HLJ/35/00 could only react well with the

homologous antisera but poorly with other antisera. This

analysis indicated that the H9N2 viruses circulated in China

Virus isolation in mouseb Phenotypec Genotyped

lungs after inoculation group

Day3 Day5

< < I A

< < I A

< < I A

< < I A

6.1 T 0.3 NDf III A

6.6 T 0.1 ND III B

6.0 T 0.7 ND III B

4.5 T 0.3 ND II A

6.4 T 0.1 ND III A

5.2 T 0.6 ND II A

5.3 T 0.4 ND II D

5.7 T 0.5 ND II C

2.8 T 0.7 ND II A

< < I E

5.2 T 0.4 ND II A

6.8 T 0.4 ND III A

5.5 T 1.3 ND II A

7.0 T 0.4 ND III F

6.8 T 0.4 ND III G

4.3 T 0.5 ND II F

5.2 T 1.4 ND II H

< 1.2 T 2.0 II I

4.8 T 1.2 ND II F

5.8 T 1.5 ND II F

< 1.6 T 1.4 II H

< 2.0 T 1.9 II F

7.3 T 0.5 ND III F

rved for 14 days post-infection.

rgans were collected on day 3 p.i., and homogenates were titrated for virus

of log10EID50 per 1 ml T standard deviation (SD). <, virus was not detectedif the virus was not detected in the undiluted lung samples of the inoculated

ctively. The virus titers in the mouse lungs were determined by the method

e basis of their ability to replicate and cause disease in mice, as described in

uences of all the eight gene segments, as described in Fig. 2.

Fig. 1.

C. Li et al. / Virology 340 (2005) 70–83 73

Fig. 1 (continued).

C. Li et al. / Virology 340 (2005) 70–8374

Fig. 1 (continued).

C. Li et al. / Virology 340 (2005) 70–83 75

Fig. 1 (continued).

C. Li et al. / Virology 340 (2005) 70–8376

C. Li et al. / Virology 340 (2005) 70–83 77

are quite different antigenically. Our previous challenge

study indicated that chickens inoculated with the CK/SD/6/

96 inactivated vaccine were poorly protected from shedding

virus in the trachea after challenge with the heterologous

viruses of CK/GX/10/99 or CK/GD/21/02 (authors unpub-

lished data). These results suggested that to maintain

optimal protection by vaccination, the prevailing strains of

influenza virus need to be included in each year’s influenza

vaccine, requiring yearly reevaluation and frequent changes

to the vaccine formulation (Lee et al., 2004).

Pathogenicity of the H9N2 isolates in mice

Avian influenza viruses were regarded to be restricted to

replicate in avian species, but H9N2 avian influenza viruses

have been isolated from pigs and humans previously (Guo

et al., 1999; Peiris et al., 1999a, 1999b; Xu et al., 2004). We

use BALB/c mice as a model to evaluate the ability of

replication in the mammalian hosts of the H9N2 viruses

isolated from avian species in China. Groups of 6-week-old

female BALB/c mice were inoculated i.n. with 106.0 EID50

of the H9N2 viruses, organs from 3 mice were collected on

day 3 or 5 p.i. for virus titration, and 5 mice from each

group were observed for 2 weeks. On the basis of their

ability to replicate and cause disease in mice, these viruses

could be divided into 3 groups (Table 2). Group I consists of

5 viruses, including 4 early isolates, CK/SD/6/96, CK/SD/7/

96, DK/NJ/1/97, and DK/NJ/2/97, and one 2000 isolate,

CK/HLJ/35/00. These 5 viruses could not be recovered from

any of the organs of the inoculated mice on day 3 or day 5

after inoculation, and the inoculated mice stayed healthy and

kept gaining weight during the observation period. Group II

includes 14 viruses. Eleven viruses in this group could

replicate well in the lungs with titers of 2.8–5.8 log10EID50

of inoculated mice on day 3 after inoculation, while 3 other

viruses, CK/GD/56/01, CK/HLJ/48/01, and CK/HN/43/02,

replicate slowly and could only be detected in the mouse

lungs with titers of 1.2–2.0 log10EID50 on day 5 after

inoculation. The weight loss of the mice induced by this

group of viruses is less than 5%. Group III contains 8

viruses that replicate very well in the lungs on day 3 after

inoculation, and the titers reach 6.0–7.3 log10EID50. The

mice showed disease signs as ruffled fur and 10–20%

weight loss. The only exception in group III is the CK/GD/

10/00 virus, which could replicate in the mouse lungs with a

high titer of 7.0 log10EID50, but did not induce any weight

loss. None of the viruses was recovered from spleens and

Fig. 1. Phylogenetic trees for the HA (A), NA (B), PB2 (C), PB1 (D), PA (E),

generated with the PHYLIP program of the CLASTALX software package (versio

(916 bp) of HA; (B) 88–1271 (1184 bp ) of NA; (C) 1036–2289 (1254 bp) of PB

999 (933 bp) of NP; (G) 26–984 (959 bp) of M; and (H) 41–859 (819 bp) of NS.

phylogenetic trees of PB2, NP, NA, M, and NS are rooted to A/Equine/Praque/1/

London/1416/73 (H7N7). Horizontal distances are proportional to the minimum nu

for spacing branches and labels. Internal branching probabilities were determined b

at the key branches in the trees. The sequences obtained from GenBank are under

Hong Kong, HK; Korea, Kor; Hokkaido, Hok; Wisconsin, WI.

brains and no deaths were observed from the inoculated

mice.

Genetic and phylogenetic analysis

To understand the molecular evolution of the H9N2

viruses in Mainland China, the entire genomes of the 27

Chinese isolates and the North American strain, TY/WI/66,

were completely sequenced. The sequences were compared

with representative H9N2 viruses, including CK/BJ/1/94,

G1, G9, Y-280, CK/SH/F/98, and some duck viruses

isolated in Hokkaido and Shantou, reported previously by

other researchers (Guan et al., 1999; Li et al., 2003; Liu et

al., 2003b, 2003c, 2004; Lu et al., 2003).

All of the HA genes contain 1742 nucleotides, except the

DK/NJ/2/97, which has a 21-nucleotide deletion from 1267

to 1287 encoding 7 amino acids. Twenty-six HA genes

shared 94.3% to 97.7% nucleotide homology with the HA

gene of CK/BJ/1/94, whereas the HA gene of CK/HLJ/35/

00 shared a lower than 79.9% homology with the HA genes

of other 26 isolates but 99.3% homology with TY/WI/66,

the prototype virus of North American lineage. Twenty-five

viruses share the same amino acid sequence of PARSSR,Gat the cleavage site (arrow) between HA1 and HA2, while

the sequences are PAVSSR,G and PARLSR,G in the HA of

CK/HLJ/35/00 and CK/GD/56/01, respectively. The

sequences of the cleavage site of these viruses are of the

characters of the low pathogenic avian influenza virus

(Steinhauer, 1999; Taubenberger, 1998). Five potential

glycosylation sites (PGS) in HA1 (11, 123, 200, 280, and

287) and two PGS in HA2 (154 and 213) are conserved in

most of the viruses, while the amino acid changes of N to S

at position 200 of CK/HLJ/35/00, N to D at position 200 of

CK/SD/6/96, CK/SD/7/96, DK/NJ/1/97, and DK/NJ/2/97, S

to G at position 289 of CK/GD/4/00, and S to P at position

215 (HA2) of CK/BJ/8/98 led to the loss of the related PGS.

However, the amino acid S to N change at position 127 led

to the addition of one PGS in CK/GX/9/99, CK/GX/10/99,

and CK/HB/31/00 viruses. These viruses also have different

amino acid substitutions in the receptor binding sites of the

HA gene at positions 183, 190, and 226 (H3 numbers)

(Table 2).

Twenty-six of 27 HA genes of the viruses from this study

belong to the CK/BJ/1/94 sub-lineage and the gradual evo-

lution of these HA genes are observed over the last several

years. The HA gene of CK/HLJ/35/00 is quite different from

others and goes to the North American lineage (Fig. 1A). No

NP (F), M (G), and NS (H) genes of the H9N2 influenza A viruses were

n 1.81) by using the neighbor-joining algorithm. (A) Nucleotides 162–1077

2; (D) 40–1473 (1434 bp) of PB1; (E) 754–2151 (1398 bp) of PA; (F) 67–

The HA phylogenetic tree is rooted to A/Duck/Alberta/60/76 (H12N5). The

56 (H7N7). The phylogenetic trees of PB1 and PA are rooted to A/Equine/

mber of nucleotide differences required to join nodes. The vertical lines are

y bootstrap analysis using 1000 replications and are indicated as percentages

lined. Turkey, TY; Liaoning, LN; Shantou, ST; Tianjin, TJ; California, CA;

Table 3

Effect of 1-adamantylamine on the replication of H9N2 influenza viruses

Viruses Virus yield in the

supernatant with

1-adamantylaminea

Residue at

31 of M2

Phenotype

Absence Presence

CK/SD/6/96 128 0 S Sensitive

CK/GD/6/96 256 0 S Sensitive

CK/SJZ/2/98 64 64 N Resistant

CK/HN/5/98 256 512 N Resistant

CK/BJ/8/98 128 128 N Resistant

CK/GX/10/99 128 0 S Sensitive

CK/HN/43/02 128 0 S Sensitive

a The HA titer of the supernatant from MDCK cells was determined 48 h

after infection with 100 TCID50 infectious dose of the indicated virus

strains in the absence or presence of 2 Ag/ml of 1-adamantylamine in the

cultures.

C. Li et al. / Virology 340 (2005) 70–8378

G1- or Y439-like HA genes are found in this study. The 8 HA

genes of Shantou duck viruses cluster to the clades of the CK/

BJ/1/94 sub-lineage, while the HA genes from the duck

viruses isolated in Hokkaido are closely related with the

reported Y439 virus.

The NA gene of all the H9N2 viruses in this study

belongs to the Eurasian lineage and forms two sister sub-

lineages (Fig. 1B), with CK/BJ/1/94 at the base of one sub-

lineage and with G9 at the base of another one. Analysis of

the deduced amino acid sequence reveals that G9 sub-

lineage viruses have full-length NA gene, while our 16 NA

genes derived from the CK/BJ/1/94 virus have deletions, 15

of them have a 9-nt deletion at positions 206–214 in the

stalk region as reported by Liu et al. (2003b), and the NA of

CK/FJ/25/00 has a deletion from nucleotide 190 to 228,

resulting in a loss of 13 amino acids, whereas CK/BJ/1/94

itself does not. Despite the deletion, the homology of the

viruses between the two sub-lineages is less than 95%,

suggesting that they evolved independently in nature. The

NA genes of the viruses isolated from ducks in Shantou

cluster to the G9 or CK/BJ/1/94 sub-lineages, while the

genes of the viruses from ducks in Hokkaido form a distinct

sub-lineage.

The topology of the phylogenetic trees of PB2, PB1 and

PA are very similar (Figs. 1C, D, E). All of the polymerase

genes of the viruses in the present study belong to the

Eurasian lineage. They form three sub-lineages represented

by the CK/BJ/1/94, G1, and CK/SH/F/98 viruses, respec-

tively. The genes share 94–100% homology within the sub-

lineages, but 84–92% homology between the sub-lineages.

The PB2 gene of the duck viruses isolated from Shantou

cluster to all of the 3 sub-lineages, while some PB1 and PA

genes of these viruses form a separate sub-lineage in the

phylogenetic trees. The NP gene is different from the

polymerase (Fig. 1F); twenty-three of the NP genes obtained

from this study share very high homology and form a large

sub-lineage rooted to CK/BJ/1/94, and NP genes of CK/

HLJ/48/01, CK/SH/10/01, and CK/GD/56/01 share a 97–

99% homology with CK/SH/F/98 and form a distinct sub-

lineage. The NP gene of CK/HLJ/35/00 shares less than

90% homology with other Chinese isolates analyzed in this

study but a 99.6% homology with the NP gene of TY/WI/66

and belongs to the North American lineage.

The M genes of the 27 viruses analyzed in the present

study share 96–100% homology; phylogenetically, they all

belong to the CK/BJ/1/94 sub-lineage (Fig. 1G). Detail

analysis indicated that 3 viruses, CK/BJ/8/98, CK/HN/5/98,

and CK/SJZ/2/98, have an S to N substitution at residue 31

in the M2, which is related to the resistance phenotype of

adamantanes of influenza A viruses (Takeda et al., 2002),

and the resistance to adamantanes of these three viruses was

experimentally confirmed, as shown in Table 3. The NS

genes of 25 viruses in this study are highly related and

belonged to the CK/BJ/1/94 sub-lineage, while the NS

genes of CK/GX/9/99 and CK/GX/10/99 are identical and

belong to the G1 sub-lineage (Fig. 1H).

The genes in the same sub-lineage in Fig. 1 share over

94% homology and are considered as one genotypic group.

On the basis of the genomic diversity, the viruses in the

present study are divided into 9 genotypes, whose evolution

and relationship are shown in Fig. 2. Twelve of the 27

analyzed viruses are classified as genotype A virus and they

are genetically close to the reported CK/BJ/1/94 virus, and

the two genotype B viruses and the two genotype H viruses

are similar to the previously reported G9 or CK/SH/F/98

viruses, respectively (Guan et al., 1999; Lu et al., 2003); the

other 6 genotypes are newly discovered in the present study.

G1-like polymerase genes and NS genes are found in 6

viruses of genotype B, C, D, E, and G. The CK/HLJ/35/00

is likely a reassortant virus that derived the genes from CK/

BJ/1/94-, G1-, G9-, and the North American TY/WI/66-like

viruses. Most of the duck H9N2 viruses isolated in Shantou

described by Li et al. (2003) have different polymerase and

NP genes from the viruses in the present study and form 8

more different genotypes (Li et al., 2003). These results

suggest that complicated reassortments of avian influenza

viruses occurred in the poultry in China. It is worthy to note

that there are no strong links between the genotypes and the

phenotypes of replication in mice of these H9N2 viruses.

Detailed analysis and comparison reveal that 34 amino acids

of the entire genome correlate with the replication pheno-

types in mice (Table 4), although the residue or residues

contributed to the difference of this phenotype remain to be

pinpointed.

Discussion

H9N2 avian influenza viruses always cause disease

outbreaks in the field. Guo et al. reported that the CK/BJ/1/

94 virus was lethal for chickens in field and also under the

experimental condition (Guo et al., 2000). Kida et al. reported

that co-infection with Staphylococcus or Haemophilus

increases the virulence of H9N2 viruses in chickens (Kishida

et al., 2004). Our study indicated that none of the analyzed

Fig 2. The genotypic evolution of the H9N2 viruses isolated from chickens and ducks in Mainland China from 1996 through 2002. The eight gene segments in

each schematic virus particle are (top to bottom) the PB2, PB1, PA, HA, NP, NA, M, and NS genes. Genes of the same sub-lineage are shown in the same color.

The capital letters indicate the genotypes. Table 2 lists the viruses of each genotype.

C. Li et al. / Virology 340 (2005) 70–83 79

viruses isolated from 1996 to 2002 induce disease signs or

deaths in SPF chickens during the observation period,

implying that co-infections with other pathogens might be

the cause for these H9N2 viruses to be lethal in the field.

An oil-emulsion, formalin-inactivated vaccine derived

from an early isolate, CK/SD/6/96, has been used for

chickens to control H9N2 virus infection, since 1998, in

China. The antigenic analysis of 21 representative viruses

performed in this study indicated a large extent of antigenic

diversity among the H9N2 viruses isolated over the years,

and the antisera induced by the vaccine strain could not react

efficiently with most of the viruses isolated after 1996,

although it reacted well to the CK/SH/10/01, CK/GD/56/01,

and CK/HN/43/02 viruses. The previous studies also

demonstrated that the currently used CK/SD/6/96 vaccine

could not completely prevent the replication and shedding of

antigenically different viruses (authors’ unpublished data),

which is similar to the observation in the H5N2 avian

influenza virus in Mexico, where the commercial vaccine

was not able to prevent virus shedding when chickens were

challenged with antigenically different isolates (Lee et al.,

2004). Human’s H1N1 and H3N2 viruses easily subject to

antigenic drift since the vaccines could not completely

prevent the virus replication (Kilbourne et al., 1990, 2002).

About half a billion doses of the inactivated H9N2 vaccines

have been used in chickens every year in China since 1998,

and the antigenic drift of the H9N2 viruses that emerged in

China may also be due to the incomplete protection induced

by the vaccination. The optimal strategy for control of

pandemic influenza is early intervention with a vaccine,

produced, ideally, from the actual pandemic strain or at least

from an antigenically well-matched strain (Lu et al., 2001).

Therefore, it is necessary to evaluate how well the prevailing

isolates match the present vaccine and update the vaccine

formulation accordingly (Lee et al., 2004).

Most of the avian influenza viruses are restricted in the

avian species. Our previous studies indicated that the H5N1

viruses that circulated in ducks in southern China have

gradually acquired the ability to replicate and kill mice (Chen

et al., 2004). Previous studies reported that G9 and G1

viruses and H9N2 viruses isolated in Hong Kong in 2003

replicate systemically and are lethal for mice (Choi et al.,

2004; Guo et al., 2000), while Lu et al. reported that the G9-

and G1-like H9N2 viruses only replicate in the respiratory

system of mice and are not lethal for mice (Lu et al., 2001).

The present study demonstrated that the H9N2 viruses

isolated in Mainland China exhibited diverse replication

phenotype in mice. The 4 early isolates and CK/HLJ/35/00

could not replicate in mice; all others are able to replicate in

the mouse lungs efficiently without any prior adaptation, but

none of the viruses are lethal, although some viruses could

induce a 10–20% weight loss of the inoculated mice.

Table 4

Amino acid residues of the H9N2 viruses that correlate with mouse

replication phenotype

Genes Position of

amino acids

Viruses that

Cannot replicate in micea Replicate in miceb

PB2 105 I T

156 V A

176 M I

285 N H

319 V I

448 D/E N

472 D E/G

PB1 157 I T/A

PA 27 Y D

129 M I

181 D E

388 N G/S

465 V I

672 F L

723 H Y

HA 4 T V/I/L

12 L V/A

15 T V/A

38 A T

218 D N

256 D N

393 M I

537 V I

NP 239 V M

398 K Q

NA 72 I T

370 T S/L

392 K I/T

427 V I

M1 195 A S

M2 14 E G

NS1 25 R Q

225 A T

NS2 14 T M/I/K

a Including the CK/SD/6/96, CK/SD/7/96, DK/NJ/1/97, and DK/NJ/2/97

viruses.b Including 22 viruses that can replicate in mice. CK/HLJ/35/00 contains

an HA gene of North American lineage and was not included in the

comparison.

C. Li et al. / Virology 340 (2005) 70–8380

Most of the viruses contain G1-like genes are able to

replicate to high titers in the mouse lungs without prior

adaptation; however, the CK/HLJ/35/00 virus cannot

replicate in mice, although it contains 3 polymerase genes

of G1-like virus. The CK/HLJ/35/00 virus contains the HA

and NP genes of the North American H9N2 viruses;

however, there is no evidence to indicate that the North

American lineage viruses have been stably established in

Mainland China. How the CK/HLJ/35/00 virus emerged has

not been determined.

Many factors are related to the host range of influenza A

viruses. Subbarao et al. reported that the amino acid at

position 627 of PB2 plays an important role in the host

range of the influenza A viruses (Subbarao et al., 1993). We

recently demonstrated that a single amino acid aspartic acid

to asparagine substitution at 701 of PB2 enables a duck

H5N1 virus to cross the host-barrier to infect and replicate in

mice (Li et al., in press). The receptor specificity of the HA

is important in determining host range, and changes in

preference for SAa2, 3Gal to SAa2, and 6Gal moieties have

been observed to accompany the establishment of avian

viruses (or their HAs) in human and porcine hosts

(Matrosovich et al., 2000). All of the viruses in this study

have conserved glutamine at position of 627 and aspartic

acid at position of 701 in PB2, suggesting that these amino

acids may not be the determinant for the host range

deference of these H9N2 viruses. A total of 34 amino acids

of the entire genome are observed to correlate with the

replication phenotypes of these viruses in mouse, and the

use of plasmid-based reverse genetics will enable an

evaluation of the contribution of the specific amino acid

residues to the host range phenotype of these H9N2 viruses.

Phylogenetic analysis indicated that complicated geno-

types of H9N2 viruses existed in the avian species in China,

and the viruses in the same clades appeared to reflect

temporal parameter rather than geographical parameter,

implying that the viruses were not restricted in the areas

where they appeared, but were spread very quickly to the

neighbor provinces or the provinces far away. Most of the

viruses are genetically closely related to the reported CK/BJ/

1/94 virus, although some of them contain a different NA

gene. A previous report described that the G1-like H9N2

viruses were mainly maintained in the quail and no

reassortant between the G1 and Y280 sub-lineages was

detected despite their co-circulation in the Hong Kong

poultry markets (Guan et al., 2000). G1-like virus was also

not identified in this study, but viruses that contain one or

more internal genes from G1-like viruses were found

primarily in Guangxi and Guangdong provinces, and it

seems that the reassortments happened independently,

implying that the viruses that contain the internal genes of

G1-like viruses may still circulate in those areas. The three

polymerase genes and NP gene of CK/SH/10/01, CK/GD/

56/01, and CK/HLJ/48/01 are closely related with the CK/

SH/F/98 virus and form a distinct sub-lineage in the present

study. These genes are also found in Shantou duck viruses

by Li et al. and in the H5N1 viruses that were isolated from

healthy ducks in Shanghai, Guangxi, and Fujian, including

the DK/SH/35/02, DK/SH/38/01, DK/FJ/01/02, DK/GX/50/

01, and DK/GX/53/02 viruses (Chen et al., 2004). Liu et al.

reported that 4 genotypes of H9N2 viruses were identified

from a live bird market in southern China; one of the

genotype is similar to our genotype A, while viruses in other

3 genotypes contain different NP and polymerase genes

from H3N2, H3N6, or H4N6 viruses (Liu et al., 2003a,

2003d) and those genotypes are different from the ones

detected in the present study. We cannot compare the

genotypes of the viruses described by Li et al. and Liu et al.,

since only partial sequences are available from their reports,

and these sequences are not overlapped; therefore, it is

difficult to confirm the relationship of those viruses. Based

on available information, at least 17 genotypes of H9N2

viruses have been detected in poultry in Mainland China

C. Li et al. / Virology 340 (2005) 70–83 81

and the complicated reassortment occurred not only within

H9N2 avian influenza viruses, but also among the H9N2,

H5N1, H3N2 and H4N6 viruses.

In addition to vaccines, two classes of antiviral agents

have been used to treat influenza—the M2 ion channel

inhibitors and the neuraminidase inhibitors (Kiso et al.,

2004). In comparison to neuraminidase inhibitors, adaman-

tane has a low cost and is frequently used for the influenza

treatment in humans and also for H9N2 avian influenza

treatment in some areas in China. The S to N substitution at

residue 31 in M2 of the influenza virus may occur

spontaneously or as a result of the treatment of adaman-

tanes. The three viruses bearing the S to N substitution at

residue 31 in M2, which confers resistance to adamantanes,

were all isolated in 1998 from central China, supporting the

idea that this mutation may be a result of the application of

adamantanes for avian influenza treatment in some regions.

In summary, the present study systemically analyzed 27

previously uncharacterized H9N2 representative viruses

isolated from poultry in China and revealed that these viruses

posed multiple genotypes and big diversity of the biological

properties. The low pathogenic properties to poultry make

these viruses easy to be widespread and transmitted to a new

host; their gradually acquired ability to infect and replicate in

mammalian and the fact that some genotype viruses are

widespread and established in the pig population in China

rank H9N2 viruses on the top of the list to cause human

influenza pandemics in the future. The antigenic diversity and

the resistance to the adamantanes of these viruses will pose

great difficulties for future human influenza control caused

by H9N2 viruses. Therefore, urgent attention should be paid

to the control of H9N2 influenza viruses in animals and to

human’s influenza pandemic preparedness.

Materials and methods

Viruses and facilities

The H9N2 viruses used in this study were isolated from

the lung samples of sick chickens or ducks sent to the

National Avian Influenza Reference Laboratory for disease

diagnosis. In brief, the lungs were collected under sterile

conditions and stored under �70 -C before inoculation into

embryonated eggs. The lung samples were ground, and a

10% suspension with cold phosphate-buffered saline (PBS),

containing 1000 U/ml of penicillin and 1000 Ag/ml of

streptomycin, was made. Solid debris was pelleted by

centrifugation at 4000�g for 5 min, and the homogenates

were inoculated into the allantoic cavity of 10-day-old

embryonated chicken eggs and incubated for 48–72 h at

37 -C. The isolated viruses were conducted with a three-

round limited dilution purification in 10-day-old embryo-

nated, specific-pathogenic-free (SPF) chicken eggs and were

identified by hemagglutination inhibition (HI) and neura-

minidase inhibition (NI) tests with a panel of antisera

provided by the Office International Des Epizooties (OIE)

Reference Laboratory, Veterinary Laboratory Agency, Sur-

rey, United Kingdom. The virus stocks were grown in the

allantoic cavity of 10-day-old embryonated SPF chicken

eggs at 37 -C for 48 h and aliquoted and stored at �70 -Cfor use in all the experiments described herein. The 50% egg

infection dose (EID50) of each virus stock was determined in

10-day-old embryonated eggs and calculated by the method

of Reed and Muench as described previously (Hoffmann

et al., 2000). All animal experiments were conducted in

HEPA-filtered isolators.

Chicken study

6-week-old white leghorn SPF chickens were intra-

nasally inoculated with 0.1 ml 106.0 EID50 of virus and

were observed for 21 days for clinical signs and deaths.

Tracheal and cloacal swabs were collected on days 3 and 5

post-inoculation (p.i.) for virus titration in eggs as described

(Chen et al., 2003). Sera were collected on day 21 p.i. for

seroconversion confirmation and antigenic analysis.

Antigenic analysis

Antisera to 6 selected H9N2 viruses generated in 6-week-

old white leghorn SPF chickens were included for the

antigenic analysis of 21 H9N2 avian influenza viruses by

hemagglutination inhibition (HI) assays with 0.5% chicken

erythrocytes.

Mouse study

Groups of eight 6-week-old female BALB/c mice (Beijing

Experimental Animal Center, Beijing) were lightly anesthe-

tized with CO2 and inoculated i.n. with 106.0 EID50 of H9N2

influenza virus in a volume of 50 Al. Control mice were

inoculated with PBS. On day 3, three mice were killed and

their lungs, spleens, and brains were collected under sterile

conditions. The organs were ground and homogenized in 1ml

of cold phosphate-buffered saline (PBS), solid debris was

pelleted by centrifugation at 4000�g for 5 min, and the

homogenates were used to determine for virus titration in 10-

day-old embryonated chicken eggs. Virus titers are given in

units of log10EID50 per 1 ml T standard deviation (SD). The

remaining five mice were monitored daily for weight loss and

mortality. For those viruses that could not be recovered from

any organs of inoculated mice on day 3 p.i., a second

experiment was conducted by inoculating groups of 6 mice

i.n. with 106.0 EID50 in a volume of 50 Al. Three mice were

killed on day 3 and day 5, respectively, and the organs were

collected for virus titration.

Genetic and phylogenic analysis

Viral RNA was extracted from allantoic fluid by using

the RNeasy Mini Kit (Qiagen, Valencia, CA) and was

C. Li et al. / Virology 340 (2005) 70–8382

reverse-transcribed using a 12-nucleotide primer, 5VAGCAQAAAGCAGG 3V, with the M-MLV reverse transcriptase

(Life Technologies, Invitrogen) for 1 h at 37 -C. PCR

amplification was performed by using fragment-specific

primers (primer sequences available on request). The PCR

products were purified with the QIAquick PCR purification

kit (Qiagen) and sequenced by using the CEQ DTCS-Quick

Start Kit on a CEQ 8000 DNA sequencer (Beckman

Coulter). Sequence data were compiled with the SEQMAN

program (DNASTAR, Madison, WI), and the phylogenetic

analysis was carried out with the PHYLIP program of the

CLASTALX software package (version 1.81) to implement

a neighbor-joining algorithm.

Sensitivity to adamantanes

Biological susceptibility to 1-adamantylamine was deter-

mined by infecting monolayers of MDCK cells with a 100

TCID50 infectious dose, in OPTI-MEM (Invitrogen Corp.,

Carlsbad, CA) containing 2 Ag/ml trypsin-TPCK (Sigma,

Fluka), in the absence or presence of 2 Ag/ml of 1-

adamantylamine (Sigma, Fluka). Relative virus replication

was determined by measuring the HA titer of the super-

natant after 48-h incubation at 37 -C with 5% CO2.

Nucleotide sequence accession numbers

The nucleotide sequences obtained in this study are

available from GenBank under the access numbers

DQ064354–DQ064569 and DQ067437–DQ067444.

Acknowledgments

This study was supported by Chinese National Key Basic

Research Program (973 Program) Grants G199901190, by

the Chinese National Natural Science Foundation

30440008, and by the Chinese Science and Technology

Development Program (863) 2004AA3071. We thank Dr.

Denis Alexander (Office International Des E’pizooties

Reference Laboratory) for providing the avian influenza

virus reference strains and antisera. The manuscript has

been reviewed by Dr. Paul Dowling from ProofItRight.

References

Alexander, D.J., 2000. A review of avian influenza in different bird species.

Vet. Microbiol. 74, 3–13.

Baker, A.T., Varghese, J.N., Laver, W.G., Air, G.M., Colman, P.M., 1987.

Three-dimensional structure of neuraminidase of subtypes N9 from an

avian influenza virus. Proteins 2, 111–117.

Campitelli, L., Fabiani, C., Puzelli, S., Fioretti, A., Foni, E., Marco, A.D.,

Krauss, S., Webster, R.G., Donatelli, I., 2002. H3N2 influenza viruses

from domestic chickens in Italy: an increasing role for chickens in the

ecology of influenza. J. Gen. Virol. 83, 413–420.

Chen, B.L., Zhang, Z.J., Chen, W.B., 1994. Isolation and preliminary

serological characterization of type A influenza viruses from chickens.

Chin. J. Vet. Med. (Chinese) 22, 3–5.

Chen, H., Matsuoka, Y., Swayne, D., Chen, Q., Cox, N.J., Murphy, B.R.,

Subbarao, K., 2003. Generation and characterization of a cold-adapted

influenza A H9N2 reassortant as a live pandemic influenza virus

vaccine candidate. Vaccine 21, 4430–4436.

Chen, H., Deng, G., Li, Z., Tian, G., Li, Y., Jiao, P., Zhang, L., Liu,

Z., Webster, R.G., Yu, K., 2004. The evolution of H5N1 influenza

viruses in ducks in southern China. Proc. Natl. Acad. Sci. U.S.A.

101, 10452–10457.

Choi, Y.K., Ozaki, H., Webby, R.J., Webster, R.G., Peiris, J.S., Poon,

L., Butt, C., Leung, Y.H.C., Guan, Y., 2004. Continuing evolution

of H9N2 influenza viruses in southeastern China. J. Virol. 78,

8609–8614.

Fouchier, R.A.M., Munster, V., Wallensten, A., Bestebroer, T.M.,

Herfst, S., Smith, D., Rimmelzwaan, G.F., Olsen, B., Osterhaus,

A.D., 2004. Characterization of a novel influenza A virus

hemagglutinin subtype (H16) obtained from black-headed gulls. J.

Virol. 79, 2814–2822.

Guan, Y., Shortridge, K.F., Krauss, S., Webster, R.G., 1999. Molecular

characterization of H9N2 influenza viruses; were they the donors of the

‘‘internal’’ genes of H5N1 viruses in Hong Kong? Proc. Natl. Acad. Sci.

U.S.A. 96, 9363–9367.

Guan, Y., Shortridge, K.F., Krauss, S., Chin, P.S., Dyrting, K.C., Ellis,

T.M., Webster, R.G., Peiris, M., 2000. H9N2 influenza viruses

possessing H5N1-like internal genomes continue to circulate in poultry

in southeastern China. J. Virol. 74, 9372–9380.

Guo, Y.J., Li, J.W., Cheng, I., 1999. Discovery of humans infected

by avian influenza A (H9N2) virus. Chin. J. Exp. Clin. Virol. 15,

105–108.

Guo, Y.J., Krauss, S., Senne, D.A., Mo, I.P., Lo, K.S., Xiong, X.P.,

Norwood, M., Shortridge, K.F., Webster, R.G., Guan, Y., 2000.

Characterization of the pathogenicity of members of the newly

established H9N2 influenza virus lineages in Asia. Virology 267,

279–288.

Hall, C., 2004. Impact of avian influenza on U.S. poultry trade relations—

2002: H5 or H7 low pathogenic avian influenza. Ann. N. Y. Acad. Sci.

1026, 47–53.

Hien, T.T., Jong, M.D., Farrar, J., 2004. Avian influenza—A challenge to

global heath care structures. N. Engl. J. Med. 351, 2363–2365.

Hinshaw, V.G., Webster, R.G., Turner, B., 1980. The perpetuation of

orthomyxoviruses and paramyxoviruses in Canadian waterfowl. Can. J.

Microbiol. 26, 622–629.

Hoffmann, E., Stech, J., Leneva, I., Krauss, S., Scholtissek, C., Chin,

P.S., Peiris, M., Shortridge, K.F., Webster, R.G., 2000. Character-

ization of the influenza A virus gene pool in avian species in

southern China: was H6N1 a derivative or a precursor of H5N1?

J. Virol. 74, 6309–6315.

Homme, P.J., Easterday, B.C., 1970. Avian influenza virus infections: I.

Characteristics of influenza A-turkey-Wisconsin-1966 virus. Avian Dis.

14, 66–74.

Kawaoka, Y., Chambers, T.M., Sladen, W.L., Webster, R.G., 1988. Is the

gene pool of influenza viruses in shorebirds and gulls different from that

in wild ducks? Virology 163, 247–250.

Kawaoka, Y., Krauss, S., Webster, R.G., 1989. Avian-to-human trans-

mission of the PB1 gene of influenza A viruses in the 1957 and 1968

pandemics. J. Virol. 63, 4603–4608.

Kilbourne, E.D., Johansson, B.E., Grajower, B., 1990. Independent and

disparate evolution in nature of influenza A virus hemagglutinin and

neuraminidase glycoproteins. Proc. Natl. Acad. Sci. U.S.A. 87,

786–790.

Kilbourne, E.D., Smith, C., Brett, I., Pokorny, B.A., Johansson, B.,

Cox, N., 2002. The total influenza vaccine failure of 1947 revisited:

major intrasubtypic antigenic change can explain failure of vaccine

in a post-World-War II epidemic. Proc. Natl. Acad. Sci. U.S.A. 99,

10748–10752.

Kishida, N., Sakoda, Y., Eto, M., Sunaga, Y., Kida, H., 2004. Co-

C. Li et al. / Virology 340 (2005) 70–83 83

infection of Staphylococcus aureus or Haemophilus paragallinarum

exacerbates H9N2 influenza A virus infection in chickens. Arch

Virol. 149, 2095–2104.

Kiso, M., Mitamura, K., Tagawa, Y.S., Shiraishi, K., Kawakami, C.,

Kimura, K., Hayden, F.G., Sugaya, N., Kawaoka, Y., 2004. Resistant

influenza A viruses in children treated with oseltamivir: descriptive

study. Lancet 364, 759–765.

Lee, C.W., Senne, D.A., Suarez, D.L., 2004. Effect of vaccine use in the

evolution of Mexican lineage H5N2 avian influenza virus. J. Virol. 78,

8372–8381.

Li, K.S., Xu, K.M., Peiris, J.S.M., Poon, L.L.M., Yu, K.Z., Yuen, K.Y.,

Shortridge, K.F., Webster, R.G., Guan, Y., 2003. Characterization of

H9 subtype influenza viruses from the ducks of southern China: a

candidate for the next influenza pandemic in humans? J. Virol. 77,

6988–6994.

Li, H., Yu, K., Yang, H., Xin, X., Chen, J., Zhao, P., Bi, Y., Chen, H., 2004.

Isolation and characterization of H5N1 and H9N2 influenza viruses

from pigs in China. Chin. J. Prev. Vet. Med. (Chinese) 26, 1–6.

Li, Z., Chen H., Jiao P., Deng G., Tian G., Li Y., Hoffmann E., Webster

R.G., Matsuoka Y., Yu, K., in press. Molecular basis associated with

replication of duck H5N1 influenza viruses in a mammalian mouse

model. J. Virol.

Lin, Y.P., Shaw, M., Gregory, V., Cameron, K., Lim, W., Klimov, A.,

Subbarao, K., Guan, Y.L., Krauss, S., Shortridge, K., Webster, R., Cox,

N., Hay, A., 2000. Avian-to-human transmission of H9N2 subtype

influenza A viruses: relationship between H9N2 and H5N1 human

isolates. Proc. Natl. Acad. Sci. U.S.A. 97, 9654–9658.

Liu, H.Q., Liu, X.F., Cheng, J., Peng, D.X., Jia, L.J., Huang, Y., 2003a.

Phylogenetic analysis of the hemagglutinin genes of twenty-six avian

influenza viruses of subtype H9N2 isolated from chickens in China

during 1996–2001. Avian Dis. 47, 116–127.

Liu, J.H., Okazaki, K., Shi, W.M., Wu, Q.M., Mweene, A.S., Kida, H.,

2003b. Phylogenetic analysis of neuraminidase gene of H9N2 influenza

viruses prevalent in chickens in China during 1995–2002. Virus Genes

27, 197–202.

Liu, J.H., Okazaki, K., Shi, W.M., Kida, H., 2003c. Phylogenetic analysis

of hemagglutinin and neuraminidase genes of H9N2 viruses isolated

from migratory ducks. Virus Genes 27, 291–296.

Liu, M., He, S.Q., Walker, D., Zhou, N.N., Perez, D.R., Mo, B., Li, F.,

Huang, X.T., Webster, R.G., Webby, R.J., 2003d. The influenza

gene pool in a poultry market in south central China. Virology 305,

267–275.

Liu, J.H., Okazaki, K., Mweene, A., Shi, W.M., Wu, Q.M., Su, J.L., Zhang,

G.Z., Bai, G.R., Kida, H., 2004. Genetic conservation of hemaggluinin

gene of H9 influenza virus in chicken population in Mainland China.

Virus Genes 29, 329–334.

Lu, X.H., Renshaw, M., Tumpey, T.M., Kelly, G.D., Primmer, J.H., Katz,

J.M., 2001. Immunity to influenza A H9N2 viruses induced by infection

and vaccination. J. Virol. 75, 4896–4901.

Lu, J.H., Liu, X.F., Shao, W.X., Zhang, P.H., Wei, D.P., 2003. Genetic

characterization of the entire genome of an H9N2 Avian influenza

virus A/Chicken/Shanghai/F/98. Acta Microbiol. Sin. (Chinese) 43,

434–441.

Matrosovich, M., Tuzikov, A., Bovin, N., Gmbaryan, A., Klimov, A.,

Castrucci, M.R., Donatelli, I., Kawaoka, Y., 2000. Early altercations of

the receptor-binding properties of H1, H2, and H3 avian influenza virus

hemagglutinins after their introduction into mammals. J. Virol. 74,

8502–8512.

Naeem, K., Ullah, A., Manvell, R.J., Alexander, D.J., 1999. Avian influenza

A subtype H9N2 in poultry in Pakistan. Vet. Rec. 145, 560.

Peiris, M., Yam, W.C., Chan, K.H., Ghose, P., Shortridge, K.F., 1999a.

Influenza A H9N2: aspects of laboratory diagnosis. J. Clin. Microbiol.

37, 3426–3427.

Peiris, M., Yuen, K.Y., Leung, C.W., Chan, K.H., Ip, P.L., Lai, R.W., Orr,

W.K., Shortridge, K.F., 1999b. Human infection with influenza H9N2.

Lancet 354, 916–917.

Perez, D.R., Lim, W., Seiler, J.P., Guan, Y., Peiris, M., Shortridge,

K.F., Webster, R.G., 2003. Role of quail in the interspecies

transmission of H9 influenza A viruses: molecular changes on HA

that correspond to adaptation from ducks to chickens. J. Virol. 77,

3148–3156.

Sharp, G.B., Kawaoka, Y., Wright, S.M., Turner, B., Hinshaw, V., Webster,

R.G., 1993. Wild ducks are the reservoir for only a limited number of

influenza A subtypes. Epidemiol. Infect. 110, 161–176.

Shaw, M., Cooper, L., Xu, X., Thompson, W., Krauss, S., Guan, Y., Zhou,

N., Klimov, A., Cox, N., Webster, R., Lim, W., Shortridge, K.,

Subbarao, K., 2002. Molecular changes associated with the trans-

mission of avian influenza a H5N1 and H9N2 viruses to humans.

J. Med. Virol. 66, 107–114.

Shortridge, K.F., 1992. Pandemic influenza: a zoonosis? Semin. Respir.

Infect. 7, 11–25.

Steinhauer, D.A., 1999. Role of hemagglutinin cleavage for the pathoge-

nicity of influenza virus. Virology 258, 1–20.

Subbarao, E.K., London, W., Murphy, B.R., 1993. A single amino acid in

the PB2 gene of influenza A virus is a determinant of host range.

J. Virol. 67, 1761–1764.

Takeda, M., Pekosz, A., Shuck, K., Pinto, L.H., Lamb, R.A., 2002.

Influenza A virus M2 ion channel activity is essential for efficient

replication in tissue culture. J. Virol. 76, 1391–1399.

Taubenberger, J.K., 1998. Influenza virus hemagglutinin cleavage into

HA1, HA2: no laughing matter. Proc. Natl. Acad. Sci. U.S.A. 95,

9713–9715.

Taubenberger, J.K., Reid, A.H., Krafft, A.E., Bijwaard, K.E., Fanning, T.G.,

1997. Initial genetic characterization of the 1918 ‘‘Spanish’’ influenza

virus. Science 275, 1793–1796.

Webster, R.G., Bean, W.J., Gorman, O.T., Chambers, T.M., Kawaoka, Y.,

1992. Evolution and ecology of influenza A viruses. Microbiol. Rev. 56,

152–179.

Webster, R.G., Guan, Y., Peiris, M., Walker, D., Krauss, S., Zhou,

N.N., Govorkova, E.A., Ellis, T.M., Dyrting, K.C., Sit, T., Rerez,

D.R., Shortridge, K.F., 2002. Characterization of H5N1 influenza

viruses that continue to circulate in Geese in Southeastern China.

J. Virol. 76, 118–126.

Xu, C.T., Fan, W.X., Rong, W., Zhao, H.K., 2004. Isolation and

identification of swine influenza recombinant A/Swine/Shan-

dong/1/2003(H9N2) virus. Microbes Infect. 6, 919–925.