Infection, Genetics and Evolution 29 (2015) 138–145
Contents lists available at ScienceDirect
Infection, Genetics and Evolution journal homepage: www.elsevier.com/locate/meegid
Phylogenetic and pathogenic analyses of three H5N1 avian influenza viruses (clade 2.3.2.1) isolated from wild birds in Northeast China Zhaobin Fan a,b,c,1, Yanpeng Ci b,1, Liling Liu b, Yixin Ma d, Ying Jia b, Deli Wang b, Yuntao Guan b, Guobin Tian b, Jianzhang Ma a,⇑, Yanbing Li b,⇑, Hualan Chen b a
College of Wildlife Resources, Northeast Forestry University, Harbin 150040, China Harbin Veterinary Research Institute, Harbin 150001, China College of Animal Science and Veterinary Medicine, Liaoning Medical University, Jinzhou 121001, China d College of Information and Computer Engineering, Northeast Forestry University, Harbin 150040, China b c
a r t i c l e
i n f o
Article history: Received 2 August 2014 Received in revised form 20 November 2014 Accepted 22 November 2014 Available online 26 November 2014 Keywords: H5N1 Avian influenza virus Wild bird Phylogenetic analysis Northeast China
a b s t r a c t From April to September 2012, periodic surveillance of avian influenza H5N1 viruses from different wild bird species was conducted in Northeast China. Three highly pathogenic avian influenza (HPAI) H5N1 viruses were isolated from a yellow-browed warbler, common shoveler, and mallard. To trace the genetic lineage of the isolates, nucleotide sequences of all eight gene segments were determined and phylogenetically analyzed. The data indicated that three viruses belonged to the same antigenic virus group: clade 2.3.2.1. To investigate the pathogenicity of these three viruses in different hosts, chickens, ducks, and mice were inoculated. The results showed that chickens were susceptible to each of the three HPAI H5N1 viruses, resulting in 100% mortality within 2–6 days after infection, whereas the three isolates exhibited distinctly different virulence in ducks and mice. The results of this study demonstrated that HPAI H5N1 viruses of clade 2.3.2.1 are still circulating in wild birds through overlapping migratory flyways. Therefore, continuous monitoring of H5N1 in both domestic and wild birds is necessary to prevent a potentially wider outbreak. Ó 2014 Elsevier B.V. All rights reserved.
1. Introduction During the last 15 years of circulation in poultry, the H5N1 virus has undergone significant genetic diversification and antigenic drift and 10 distinct viral clades (0–9) with subclades have been reported (WHO, 2008). Highly pathogenic avian influenza virus (HPAIV) H5N1 clade 2.3.2.1 has spread geographically and evolved genetically in Asia since it was first isolated from a dead Chinese pond heron in Hong Kong in 2004. The most probable route of this transmission was suspected to be the movement of land-based poultry or local migratory birds (WHO, 2012). In 2009/10, clade 2.3.2.1 was repeatedly detected in China, Japan, Mongolia, and Russia (Sharshov et al., 2010; Li et al., 2011). In early 2010, this clade was confirmed in Nepal, which was considered the first reported introduction of this virus into South Asia. Meanwhile, the H5N1 virus was detected in poultry and wild birds in Romania and Bulgaria in March 2010, which ⇑ Corresponding authors. Tel.: +86 18724591358 (J. Ma), +86 13946015508 (Y. Li). E-mail addresses: [email protected] (J. Ma), [email protected] (Y. Li). 1 These authors contributed equally to this study. http://dx.doi.org/10.1016/j.meegid.2014.11.020 1567-1348/Ó 2014 Elsevier B.V. All rights reserved.
was also the first report of clade 2.3.2.1 in poultry in Europe (Reid et al., 2011). The possible dissemination of influenza A H5N1 throughout Eurasia through migratory wild birds has been previously discussed (Chen et al., 2005). The emergence and spread of HPAIV H5N1 clade 2.3.2.1 in Southeast Asian supports the contention that this clade is probably established in wild birds and land-based poultry and is spreading throughout the geographical range, just as clade 2.2 before it. Before 2009, clade 2.3.4 was the dominant clade in poultry in China (Li et al., 2011). During this period, clade 2.2 caused widespread outbreaks in wild birds in the wetlands of Gengahai Lake, Qinghai Province, China, and subsequently spread westward to the Middle East and south Asia, Europe, and Africa in 2006–2007 and became established in poultry populations in some Asian and African countries (Li et al., 2011). In the spring of 2009, clade 2.3.2.1 viruses were isolated from four sick wild swans at a park in Shanghai (Zhao et al., 2012a,b) following repeated detection H5N1 clade 2.3.2.1 viruses in wild birds in Hong Kong in 2006– 2008 and in waterfowl. Subsequently, from May 8 to June 15, 2009, a total of 273 wild birds died of HPAI H5N1 viruses of clade 2.3.2.1, when they moved to Gengahai Lake, Qinghai Province during their northward migration. The virus was the second
139
Z. Fan et al. / Infection, Genetics and Evolution 29 (2015) 138–145
lineage, in addition to clade 2.2, to emerge in wild birds in Qinghai (Li et al., 2011). Zhou et al. also reported that the wild pikas dwelling in Qinghai harbored the same clade of viruses (Zhou et al., 2009). When investigating the prevalence and evolution of the H5 subtype HPAIVs circulating in poultry in China during the period 2007–2009, Jiang et al. found that clade 2.3.2.1 viruses were prevalent in poultry in many provinces in China (Jiang et al., 2010). Northeast China is a vast geographic area with a broad diversity of ecosystems, including forests, plains, wetlands, meadows, rivers, and lakes, that serve as major migratory flyways (East Asia-Australasian Flyway) and provide important breeding and stopover sites for migrating wild birds. Therefore, we conducted epidemic surveillance of avian influenza H5N1 viruses among wild birds in Northeast China from April to September 2012 and isolated three HPAI H5N1 viruses from different species. We performed phylogenetic analyses based on sequence data and assessed the replication and pathogenic potential of three isolated viruses in chickens, ducks, and mice. The elucidation of the molecular and biological features of the H5N1 AIVs will help to reveal the potential evolutionary and transmission features of them to benefit disease control and pandemic preparedness. 2. Materials and methods 2.1. Ethics statement All animal studies were approved by the Institutional Animal Care and Use Committee (IACUC) of the Harbin Veterinary Research Institute (HVRI), Chinese Academy of Agricultural Sciences. All animal procedures were carried out in strict accordance with the recommendation in the Guide for the Care and Use of Laboratory Animals of the Ministry of Science and Technology of the People’s Republic of China. All experiments were performed in a biosafety level 3 laboratory at Harbin Veterinary Research Institute (Harbin, China).
(Jilin, Liaoning and Heilongjiang) were sent to the Harbin Veterinary Research Institute for AIV diagnosis. The detailed information of wild birds samples is shown in Table 1. 2.3. Virus isolation and identification Excrement or lung, kidney and brain (dead birds) were suspended in antibiotic-treated phosphate-buffered saline (PBS; pH 7.2) and inoculated into the allantoic cavities of 10-day-old embryonated specific pathogen-free (SPF) chicken eggs (Chen et al., 2004). Viral subtype detection was performed using a hemagglutination-hemagglutination inhibition (HA-HI) assay and reverse transcription polymerase chain reaction (RT-PCR) methods (Zhao et al., 2012a,b). All of the isolated viruses were purified three times in SPF chicken eggs and stored at 80 °C after propagation for future use (Li et al., 2005). 2.4. Sequence and phylogenetic analysis Viral RNA was extracted from allantoic fluid using TRIzol reagent (Invitrogen Carlsbad, CA, USA) and then reverse transcribed with 12-bp primers. RT-PCR was performed using a set of sequence-specific primers (primer sequences available on request) to obtain eight full-length gene fragments of influenza virus. The RT-PCR products were purified using the Watson PCR purification kit (Watson Biotechnologies Inc., Shanghai, China) and directly sequenced using the CEQ DTCS-Quick Start Kit on a CEQ 8800 DNA sequencer (Beckman Coulter, Inc., Brea, CA, USA). MEGA 5 software was used to perform multiple sequence alignment with the Clustal W algorithm and phylogenetic trees were generated by the neighbor-joining method and Kimura two-parameter model with 1000 bootstrap replications. Potential glycosylation sites were analyzed with NetNGlyc 1.0 online software (www.cbs.dtu.dk/services/NetNGlyc/). 2.5. Animal experiments
2.2. Samples collection The collection of wild birds samples is managed by National Terrestrial Wildlife Pathogen-origins and Epidemic Diseases Monitoring Master Stations of State Forestry Administration. There are many monitoring substations in Jilin, Liaoning and Heilongjiang provinces of China, which are responsible for collecting samples of wild birds and sending to research institutions for research. These substations collected samples by collecting excrement or looking for dead wild birds from nature reserve areas and forest farm, not capturing wild birds alive. Therefore, the sample collection did not do harm to wild birds. From April to September in 2012, a total of 798 frozen samples collected from National Terrestrial Wildlife Pathogen-origins and Epidemic Diseases Monitoring Stations in 3 provinces in Northeast China
The 50% egg infection dose (EID50) and 50% lethal dose in mice (MLD50) were calculated in the respective models using the method of Reed and Muench (Reed and Muench, 1938; Li et al., 2005). To examine the replication and transmission of the three isolated H5N1 viruses in chickens and ducks, three groups of 4week-old SPF chickens and 3-week-old SPF Sheldrake ducks (eight birds/group) were intranasally inoculated with 105.0 EID50 per 100 lL. After 24 h, three additional chickens and ducks were placed into the same isolation units to monitor contact infection. All birds were monitored daily every 6 h for clinical signs or death during the course of the 21-day experimental period. Viral shedding was monitored through oropharyngeal and cloacal swabs sampled from infected and contacted birds at 3, 5, 7, and 9 days
Table 1 Surveillance work on wild birds in 2012. Spot
Province
Sampling time
Habitat
No. of sample
Shuangtaihekou National Nature Reserve Xianghai Nation Nature Reserve
Liaoning
Apr.
Wetland
201
Jilin
May
Wetland
Xingkai Lake Nature Reserve Mao’er forest farm
Heilongjiang Heilongjiang
May Jul.
Zhalong Nature Reserve
Heilongjiang
Sep.
Total
No. of positive
Positive rate (%)
Subtype (n)
Host
Order
3
1.49
H3N8(3)
Mallard
Anseriformes
198
4
2.02
Mallard
Anseriformes
Lake Forest
116 178
1 1
0.86 0.56
H5N1(1) H3N8(3) H5N1(1) H5N1(1)
Anseriformes Passeriformes
Wetland
105
1
0.95
H3N8(1)
Common shoveler Yellow-browed Warbler Red-crowned crane
798
10
1.25
Griformes
140
Z. Fan et al. / Infection, Genetics and Evolution 29 (2015) 138–145
post-inoculation (dpi). Three birds (dead or illness including weakness, respiratory difficulty and neurological signs) were euthanized with intravenous sodium pentobarbital for doses of 100 mg/kg and tissues including brain, lung, kidney, spleen, bursa, thymus, trachea, cecal tonsil, heart, liver, and pancreas were collected aseptically on 3 dpi for virus titration. On 14 and 21 dpi, serum samples were obtained from each bird for serologic testing. To investigate the virulence of three H5N1 viruses in mice, six groups of 6-week-old female BALB/c mice (five mice/group) were lightly anesthetized with ketamine formulated to provide doses of 25 mg/kg ketamine to each mouse. The mice were separately inoculated intranasally with 101.0–106.0 EID50 of the H5N1 influenza virus in a volume of 50 lL (Zhao et al., 2012a,b). The control group (five mice) was mock-infected with PBS. Each group was monitored daily for weight loss and mortality for 14 days. In addition, another group of six mice was inoculated intranasally with 106.0 EID50 of the H5N1 influenza viruses per 50 lL. Three mice (lethargy, emaciation or dyspnea) were euthanized by peritoneal injection of sodium pentobarbital for doses of 200 mg/kg on 3 dpi and lung, kidney, spleen, turbinal bone, and brain tissues were collected for viral titration. An additional three mice were euthanized on 5 dpi and organs were harvested for histopathological evaluation.
3. Results 3.1. Virus isolation and identification A total of 798 samples were collected from the wild birds in northeast China from April to September 2012. The habitats
included wetlands, lakes, and forests in Liaoning, Jilin, and Heilongjiang provinces as shown in Table 1. By adopting the method based on chick embryos, 10 HA-positive allantoic fluid samples were found (positive rate = 1.25%). Of the 10 HA-positive samples, three H5N1 subtype AIVs were isolated from a yellow-browed warbler, common shoveler and mallard, named A/Yellow-browed Warbler/Heilongjiang/18/2012(H5N1)(YBW/HLJ/18), A/Common shoveler/Heilongjiang/137/2012(H5N1)(CS/HLJ/137) and A/Mallard/Jilin/36/2012(H5N1)(MD/JL/36) separately. 3.2. Molecular characteristic analysis The HA genes of CS/HLJ/137, MD/JL/36 and YBW/HLJ/18 contain open reading frames of 1701, 1701, and 1704 nucleotides, coding for 567, 567, and 568 amino acids respectively, including a signal peptide (residues 1–16). Based on the deduced amino acid sequences, the HA cleavage sites of the three viruses are PQRERRRKR⁄GLF (CS/HLJ/137, MD/JL/36) and PQIERRRRKR⁄GLF (YBW/HLJ/18) (Table 2), which contain a series of basic amino acids that meet the characteristic of highly pathogenic AIV in chickens (Horimoto and Kawaoka, 1994). Both CS/HLJ/137 and MD/JL/36 had seven potential glycosylation sites in HA1 [positions 27 (NST), 39 (NVT), 181 (NNT), 209 (NST), and 302 (NSS)] and HA2 [positions 499 (NGT) and 558 (NGS)], while YBW/HLJ/18 had additional sites at positions 289 (NCS) and 302 (NTS). In addition, the three isolated viruses did not contain a glycosylation site at position 170, which was characteristic of HPAI H5N1 viruses in wild birds (Chen et al., 2006). The conserved amino acids residues within the receptor binding site of the HA protein (H3 numbering), including Y98, S136, W153, H183, and Y195, were each present in the three isolated viruses, implying that they retained typical avian
Table 2 Selected characteristic amino acids of H5N1 subtype AIVs isolated from wild birds. Key residues
Comments
The isolates YBW/HLJ/18
MD/JL/36
CS/HLJ/137
GCG/QH/1/09
HA(H3 numbering) Cleavage sites Ser138Ala(S-A) Thr160Ala(T-A) Gln226Leu(Q-L) Gly228Ser(G-S)
Characteristic of high-pathogenic in chickens Favour mammalian adaptation Increased affinity toward a2, 6-type receptor Increased affinity toward a2, 6-type receptor Increased affinity toward a2, 6-type receptor
PQIERRRRKR⁄GLF A A Q G
PQRERRRKR⁄GLF A A Q G
PQRERRRKR⁄GLF A A Q G
PQRERRRKR⁄GLF A A Q G
NA(N1 numbering) Stalk deletion(49–68) His275Tyr(H-Y)
Increased virulence Neuraminidase resistance
Deletion H
Deletion H
Deletion H
Deletion H
PB2 Leu89Val(L-V) Glu627Lys(E-K) Asp701Asn(D-N)
Enhanced polymerase activity Mammalian adaptation Mammalian adaptation
V E D
V E D
V E D
V E D
PB1 His99Tyr(H-Y) Lys207Arg(K-R) Ile368Val(I-V) Tyr436His(Y-H)
Enables droplet transmission in ferrets Enhanced virulence in duck Enables droplet transmission in ferrets Enhanced virulence in duck
H K I Y
H K V Y
H K I Y
H K I Y
PB1-F2 Full-length(87–90) Asn66Ser(N-S)
Full-length needed for virulence in mice Increased virulence in a mice model
Full-length N
Full-length S
Full-length N
Full-length N
PA Ala36Thr(A-T) Thr515Ala(T-A)
Increased replication in mice Enhanced virulence in duck
A T
A T
A T
A T
M1 Asn30Asp(N-D) Thr215Ala(T-A)
Increased virulence in a mice model Increased virulence in a mice model
D A
D A
D A
D A
M2 Ser31Asn(S-N)
Amantadine resistance
N
S
S
S
NS1 Deletion(80–84) Pro42Ser(P-S)
Signalling of host proteins Increased virulence in mice
Deletion S
Deletion S
Deletion S
Deletion S
141
Z. Fan et al. / Infection, Genetics and Evolution 29 (2015) 138–145
virus-like receptor specificity (Skehel and Wiley, 2000). In addition, the HA amino acids at positions 226 and 228 were Gln and Gly (Table 2), indicating that these viruses may retain the characteristic of preferential binding to the a2, 3-type receptor, which is predominant in avian species (Yamada et al., 2006). Compared to A/goose/Guangdong/1/96(H5N1), the three isolated H5N1 viruses all had a deletion of 20 residues (position 49–68) in the stalk region of the NA protein (Table 2), which resulted in the loss of four potential glycosylation sites at positions 50 (NQS), 58 (NNT), 63 (NQT), and 68 (NIS). Therefore, each of the three viruses had three potential glycosylation sites in the NA protein at positions 88 (NSS), 146 (NGT), and 235 (NGS). The H275Y mutation in the NA protein, which has been associated with oseltamivir resistance (Brookes et al., 2011), was not detected. The three isolated viruses had no mutations at positions 627 (E-K) and 701 (D-N) of the PB2 protein, which are important for adaptation of avian influenza viruses to mammals (Katz et al., 2000; Li et al., 2005). Remarkably, two mutations in PB1 and PB1-F2 (I368V and N66S, respectively), which are associated with
A
92
98
62 88
increased virulence in mice and ferrets, were found in the MD/JL/ 36 strain (To et al., 2013). The S31N mutation in the M2 protein, which causes amantadine resistance (Lee et al., 2008), was detected in YBW/HLJ/18. Moreover, the NS1 protein of each of the three isolated viruses had a deletion of five amino acids (80–84) and the PDZ binding motif was EPEV or ESEV, which denote avian-like specificity. Remarkably, P42S mutations were found in the three isolated viruses, which are all associated with interferon (IFN) resistance (Seo et al., 2002; Jiao et al., 2008). 3.3. Phylogenetic analysis To elucidate the genetic relationships of the three H5N1 viruses, the whole genome of each virus was sequenced and all eight gene fragments were compared to sequences of representative influenza viruses that were retrieved from the GenBank database (https:// www.ncbi.nlm.nih.gov/genbank/). In the HA phylogenetic tree (Fig. 1A), the three viruses were clustered into clade 2.3.2.1, and shared 95.3–99.3% homology with
62 A/mandarin duck/Korea/PSC24-24/2010(H5N1) 68 A/goshawk/Tochigi/64/2011(H5N1) 98 A/tundra swan/Fukushima/207/2011(H5N1) 92 A/whooper swan/Mongolia/21/2010(H5N1) 56 A/whooper swan/Mongolia/6/2009(H5N1) 92 A/Anas clypeata/Heilongjiang/137/2012(H5N1) A/great crested-grebe/Qinghai/1/2009(H5N1) 100 A/wild duck/Jilin/HF/2011(H5N1) A/duck/Hunan/S4150/2011(H5N1) 100 68 A/Anas platyrhynchos/Jilin/36/2012(H5N1) 98 A/barnswallow/Hong Kong/1161-SJC001_RG1/2010(H5N1) A/wild bird/Hong Kong/07035-1/2011(H5N1) 100 100 A/wild duck/Fujian/1/2011(H5N1) 100 A/Phylloscopus inornatus/Heilongjiang/18/2012(H5N1) 88 A/Common Magpie/Hong Kong/5052/2007 A/duck/Guangxi/89/2006(H5N1) A/chicken/Hunan/999/2005(H5N1) A/Goose/Guangdong/1/96(H5N1) 100 A/chicken/Shanxi/2/2006(H5N1) A/chicken/Liaoning/23/2005(H5N1) A/goose/Guiyang/3422/2005(H5N1) A/duck/Laos/3295/2006(H5N1) A/Anhui/1/2006(H5N1) 100
Clade 2.3.2.1
0.005
B
89
39
A/bar-headed goose/Mongolia/X53/2009(H5N1)
A/ruddy shelduck/Mongolia/X42/2009(H5N1)
74
A/grebe/Tyva/3/2009(H5N1) 87
44
A/common buzzard/Bulgaria/38WB/2010(H5N1)
Group 1
A/black-headed gull/Tyva/115/2009(H5N1) A/Anas clypeata/Heilongjiang/137/2012(H5N1)
84
72
A/great crested-grebe/Qinghai/1/2009(H5N1) 99
A/Anas platyrhynchos/Jiling/36/2012(H5N1) A/wild duck/Jilin/HF/2011(H5N1)
100 87
A/duck/Lao/469/2010(H5N1))
96
Group 2
A/chicken/Hubei/QE8/2009(H5N1)
A/environment/Hunan/5-25/2007(H5N1) A/house crow/Hong Kong/7677/2008(H5N1) 64
A/Phylloscopus inornatus/Heilongjiang/18/2012(H5N1) A/oriental magpie robin/Hong Kong/9298/2009(H5N1) A/feral pigeon/Hong Kong/3409/2009(H5N1)
67
A/barn swallow/Hong Kong/1161/2010(H5N1)
27
Group 3
A/muscovy duck/Vietnam/LBM57/2011 (H5N1)
58 61
A/duck/Vietnam/LBM133/2012(H5N1)
0.005
Fig. 1. Phylogenetic tree based on the open reading frame sequences of HA (A) and NA (B) gene segment of H5 and N1 subtype viruses. The three H5N1 viruses isolated from Northeast in China were highlighted with a close circle. Different lineage was highlighted in different color. The trees were constructed using the neighbour-joining method of MEGA5.0 with 1000 bootstrap trials performed to assign confidence to the grouping. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
142
Table 3 The highest nucleotide identity to whole genomes of three H5N1 influenza viruses. Gene segment
HA NA PB2 PB1 PA NP M NS
CS/HLJ/137
MD/JL/36
YBW/HLJ/18
Closest viruses
Nucleotide identity
Accession No.
Closest viruses
Nucleotide identity
Accession No.
Closest viruses
Nucleotide identity
Accession No.
WS/MG/6/09 (H5N1) GCG/QH/1/09 (H5N1) CB/BG/38/10 (H5N1) GCG/QH/1/09 (H5N1) GCG/QH/1/09 (H5N1) BG/Tyva/10/09 (H5N1) DK/Lao/463/10 (H5N1) GS/JS/k0403/10 (H5N1)
99.3% 99.1% 99.3% 99.2% 99.0% 99.4% 99.3% 99.6%
AB523767.1 CY063320.1 CY110851.1 GU182156.1 CY063317.2 GQ386152.1 CY098340.1 JQ638682.1
DK/HN/S4150/11(H5N1) WD/JL/HF/11(H5N1) CK/AH/HF/10(H9N2) CK/HB/1102/10(H5N2) CK/SD/BD/08(H9N2) DK/JS/26/04(H3N2) WD/JL/HF/11(H5N1) EV/CS/10/09(H5N1)
99.0% 99.1% 99.3% 99.2% 98.5% 99.2% 99.2% 99.4%
CY1466921 JX534575.1 JX312549.1 JQ041392.1 JF795077.1 KC261670.1 JX534576.1 JN558599.1
WD/FJ/1/11(H5N1) OMR/HK/9298/09(H5N1) WD/FJ/1/11(H5N1) MD/VT/LBM228/12(H5N1) WD/FJ/1/11(H5N1) MD/VT/LBM228/12(H5N1) BHG/HK/709/11(H5N1) GH/HK/1046/08(H5N1)
99.1% 98.7% 99.1% 98.0% 99.4% 99.5% 99.0% 98.6%
JX534589.1 AB557634.1 JX534594.1 AB786682.1 JX534588.1 AB786685.1 KC436117.1 CY036249.1
Table 4 Animal experiments of three H5N1subtype AIVs in chickens and ducks. Birds
Chickens
Viruses
YBW/HLJ/18 CS/HLJ/137 MD/JL/36
Ducks
YBW/HLJ/18 CS/HLJ/137 MD/JL/36
Infected method
3 dpi
5 dpi
7 dpi
9 d pi
Sera-conversion (positive/total)
OP
CL
OP
CL
OP
CL
OP
CL
14 dpi
21 dpi
Inoculated Contacted Inoculated Contacted Inoculated Contacted
2/2a(2.6 ± 0.3)b 1/3(1.25) N 0/3 5/5(2.3 ± 0.7) 2/3(1.6 ± 0.8)
2/2(2.4 ± 0.2) 2/3(
Contents lists available at ScienceDirect
Infection, Genetics and Evolution journal homepage: www.elsevier.com/locate/meegid
Phylogenetic and pathogenic analyses of three H5N1 avian influenza viruses (clade 2.3.2.1) isolated from wild birds in Northeast China Zhaobin Fan a,b,c,1, Yanpeng Ci b,1, Liling Liu b, Yixin Ma d, Ying Jia b, Deli Wang b, Yuntao Guan b, Guobin Tian b, Jianzhang Ma a,⇑, Yanbing Li b,⇑, Hualan Chen b a
College of Wildlife Resources, Northeast Forestry University, Harbin 150040, China Harbin Veterinary Research Institute, Harbin 150001, China College of Animal Science and Veterinary Medicine, Liaoning Medical University, Jinzhou 121001, China d College of Information and Computer Engineering, Northeast Forestry University, Harbin 150040, China b c
a r t i c l e
i n f o
Article history: Received 2 August 2014 Received in revised form 20 November 2014 Accepted 22 November 2014 Available online 26 November 2014 Keywords: H5N1 Avian influenza virus Wild bird Phylogenetic analysis Northeast China
a b s t r a c t From April to September 2012, periodic surveillance of avian influenza H5N1 viruses from different wild bird species was conducted in Northeast China. Three highly pathogenic avian influenza (HPAI) H5N1 viruses were isolated from a yellow-browed warbler, common shoveler, and mallard. To trace the genetic lineage of the isolates, nucleotide sequences of all eight gene segments were determined and phylogenetically analyzed. The data indicated that three viruses belonged to the same antigenic virus group: clade 2.3.2.1. To investigate the pathogenicity of these three viruses in different hosts, chickens, ducks, and mice were inoculated. The results showed that chickens were susceptible to each of the three HPAI H5N1 viruses, resulting in 100% mortality within 2–6 days after infection, whereas the three isolates exhibited distinctly different virulence in ducks and mice. The results of this study demonstrated that HPAI H5N1 viruses of clade 2.3.2.1 are still circulating in wild birds through overlapping migratory flyways. Therefore, continuous monitoring of H5N1 in both domestic and wild birds is necessary to prevent a potentially wider outbreak. Ó 2014 Elsevier B.V. All rights reserved.
1. Introduction During the last 15 years of circulation in poultry, the H5N1 virus has undergone significant genetic diversification and antigenic drift and 10 distinct viral clades (0–9) with subclades have been reported (WHO, 2008). Highly pathogenic avian influenza virus (HPAIV) H5N1 clade 2.3.2.1 has spread geographically and evolved genetically in Asia since it was first isolated from a dead Chinese pond heron in Hong Kong in 2004. The most probable route of this transmission was suspected to be the movement of land-based poultry or local migratory birds (WHO, 2012). In 2009/10, clade 2.3.2.1 was repeatedly detected in China, Japan, Mongolia, and Russia (Sharshov et al., 2010; Li et al., 2011). In early 2010, this clade was confirmed in Nepal, which was considered the first reported introduction of this virus into South Asia. Meanwhile, the H5N1 virus was detected in poultry and wild birds in Romania and Bulgaria in March 2010, which ⇑ Corresponding authors. Tel.: +86 18724591358 (J. Ma), +86 13946015508 (Y. Li). E-mail addresses: [email protected] (J. Ma), [email protected] (Y. Li). 1 These authors contributed equally to this study. http://dx.doi.org/10.1016/j.meegid.2014.11.020 1567-1348/Ó 2014 Elsevier B.V. All rights reserved.
was also the first report of clade 2.3.2.1 in poultry in Europe (Reid et al., 2011). The possible dissemination of influenza A H5N1 throughout Eurasia through migratory wild birds has been previously discussed (Chen et al., 2005). The emergence and spread of HPAIV H5N1 clade 2.3.2.1 in Southeast Asian supports the contention that this clade is probably established in wild birds and land-based poultry and is spreading throughout the geographical range, just as clade 2.2 before it. Before 2009, clade 2.3.4 was the dominant clade in poultry in China (Li et al., 2011). During this period, clade 2.2 caused widespread outbreaks in wild birds in the wetlands of Gengahai Lake, Qinghai Province, China, and subsequently spread westward to the Middle East and south Asia, Europe, and Africa in 2006–2007 and became established in poultry populations in some Asian and African countries (Li et al., 2011). In the spring of 2009, clade 2.3.2.1 viruses were isolated from four sick wild swans at a park in Shanghai (Zhao et al., 2012a,b) following repeated detection H5N1 clade 2.3.2.1 viruses in wild birds in Hong Kong in 2006– 2008 and in waterfowl. Subsequently, from May 8 to June 15, 2009, a total of 273 wild birds died of HPAI H5N1 viruses of clade 2.3.2.1, when they moved to Gengahai Lake, Qinghai Province during their northward migration. The virus was the second
139
Z. Fan et al. / Infection, Genetics and Evolution 29 (2015) 138–145
lineage, in addition to clade 2.2, to emerge in wild birds in Qinghai (Li et al., 2011). Zhou et al. also reported that the wild pikas dwelling in Qinghai harbored the same clade of viruses (Zhou et al., 2009). When investigating the prevalence and evolution of the H5 subtype HPAIVs circulating in poultry in China during the period 2007–2009, Jiang et al. found that clade 2.3.2.1 viruses were prevalent in poultry in many provinces in China (Jiang et al., 2010). Northeast China is a vast geographic area with a broad diversity of ecosystems, including forests, plains, wetlands, meadows, rivers, and lakes, that serve as major migratory flyways (East Asia-Australasian Flyway) and provide important breeding and stopover sites for migrating wild birds. Therefore, we conducted epidemic surveillance of avian influenza H5N1 viruses among wild birds in Northeast China from April to September 2012 and isolated three HPAI H5N1 viruses from different species. We performed phylogenetic analyses based on sequence data and assessed the replication and pathogenic potential of three isolated viruses in chickens, ducks, and mice. The elucidation of the molecular and biological features of the H5N1 AIVs will help to reveal the potential evolutionary and transmission features of them to benefit disease control and pandemic preparedness. 2. Materials and methods 2.1. Ethics statement All animal studies were approved by the Institutional Animal Care and Use Committee (IACUC) of the Harbin Veterinary Research Institute (HVRI), Chinese Academy of Agricultural Sciences. All animal procedures were carried out in strict accordance with the recommendation in the Guide for the Care and Use of Laboratory Animals of the Ministry of Science and Technology of the People’s Republic of China. All experiments were performed in a biosafety level 3 laboratory at Harbin Veterinary Research Institute (Harbin, China).
(Jilin, Liaoning and Heilongjiang) were sent to the Harbin Veterinary Research Institute for AIV diagnosis. The detailed information of wild birds samples is shown in Table 1. 2.3. Virus isolation and identification Excrement or lung, kidney and brain (dead birds) were suspended in antibiotic-treated phosphate-buffered saline (PBS; pH 7.2) and inoculated into the allantoic cavities of 10-day-old embryonated specific pathogen-free (SPF) chicken eggs (Chen et al., 2004). Viral subtype detection was performed using a hemagglutination-hemagglutination inhibition (HA-HI) assay and reverse transcription polymerase chain reaction (RT-PCR) methods (Zhao et al., 2012a,b). All of the isolated viruses were purified three times in SPF chicken eggs and stored at 80 °C after propagation for future use (Li et al., 2005). 2.4. Sequence and phylogenetic analysis Viral RNA was extracted from allantoic fluid using TRIzol reagent (Invitrogen Carlsbad, CA, USA) and then reverse transcribed with 12-bp primers. RT-PCR was performed using a set of sequence-specific primers (primer sequences available on request) to obtain eight full-length gene fragments of influenza virus. The RT-PCR products were purified using the Watson PCR purification kit (Watson Biotechnologies Inc., Shanghai, China) and directly sequenced using the CEQ DTCS-Quick Start Kit on a CEQ 8800 DNA sequencer (Beckman Coulter, Inc., Brea, CA, USA). MEGA 5 software was used to perform multiple sequence alignment with the Clustal W algorithm and phylogenetic trees were generated by the neighbor-joining method and Kimura two-parameter model with 1000 bootstrap replications. Potential glycosylation sites were analyzed with NetNGlyc 1.0 online software (www.cbs.dtu.dk/services/NetNGlyc/). 2.5. Animal experiments
2.2. Samples collection The collection of wild birds samples is managed by National Terrestrial Wildlife Pathogen-origins and Epidemic Diseases Monitoring Master Stations of State Forestry Administration. There are many monitoring substations in Jilin, Liaoning and Heilongjiang provinces of China, which are responsible for collecting samples of wild birds and sending to research institutions for research. These substations collected samples by collecting excrement or looking for dead wild birds from nature reserve areas and forest farm, not capturing wild birds alive. Therefore, the sample collection did not do harm to wild birds. From April to September in 2012, a total of 798 frozen samples collected from National Terrestrial Wildlife Pathogen-origins and Epidemic Diseases Monitoring Stations in 3 provinces in Northeast China
The 50% egg infection dose (EID50) and 50% lethal dose in mice (MLD50) were calculated in the respective models using the method of Reed and Muench (Reed and Muench, 1938; Li et al., 2005). To examine the replication and transmission of the three isolated H5N1 viruses in chickens and ducks, three groups of 4week-old SPF chickens and 3-week-old SPF Sheldrake ducks (eight birds/group) were intranasally inoculated with 105.0 EID50 per 100 lL. After 24 h, three additional chickens and ducks were placed into the same isolation units to monitor contact infection. All birds were monitored daily every 6 h for clinical signs or death during the course of the 21-day experimental period. Viral shedding was monitored through oropharyngeal and cloacal swabs sampled from infected and contacted birds at 3, 5, 7, and 9 days
Table 1 Surveillance work on wild birds in 2012. Spot
Province
Sampling time
Habitat
No. of sample
Shuangtaihekou National Nature Reserve Xianghai Nation Nature Reserve
Liaoning
Apr.
Wetland
201
Jilin
May
Wetland
Xingkai Lake Nature Reserve Mao’er forest farm
Heilongjiang Heilongjiang
May Jul.
Zhalong Nature Reserve
Heilongjiang
Sep.
Total
No. of positive
Positive rate (%)
Subtype (n)
Host
Order
3
1.49
H3N8(3)
Mallard
Anseriformes
198
4
2.02
Mallard
Anseriformes
Lake Forest
116 178
1 1
0.86 0.56
H5N1(1) H3N8(3) H5N1(1) H5N1(1)
Anseriformes Passeriformes
Wetland
105
1
0.95
H3N8(1)
Common shoveler Yellow-browed Warbler Red-crowned crane
798
10
1.25
Griformes
140
Z. Fan et al. / Infection, Genetics and Evolution 29 (2015) 138–145
post-inoculation (dpi). Three birds (dead or illness including weakness, respiratory difficulty and neurological signs) were euthanized with intravenous sodium pentobarbital for doses of 100 mg/kg and tissues including brain, lung, kidney, spleen, bursa, thymus, trachea, cecal tonsil, heart, liver, and pancreas were collected aseptically on 3 dpi for virus titration. On 14 and 21 dpi, serum samples were obtained from each bird for serologic testing. To investigate the virulence of three H5N1 viruses in mice, six groups of 6-week-old female BALB/c mice (five mice/group) were lightly anesthetized with ketamine formulated to provide doses of 25 mg/kg ketamine to each mouse. The mice were separately inoculated intranasally with 101.0–106.0 EID50 of the H5N1 influenza virus in a volume of 50 lL (Zhao et al., 2012a,b). The control group (five mice) was mock-infected with PBS. Each group was monitored daily for weight loss and mortality for 14 days. In addition, another group of six mice was inoculated intranasally with 106.0 EID50 of the H5N1 influenza viruses per 50 lL. Three mice (lethargy, emaciation or dyspnea) were euthanized by peritoneal injection of sodium pentobarbital for doses of 200 mg/kg on 3 dpi and lung, kidney, spleen, turbinal bone, and brain tissues were collected for viral titration. An additional three mice were euthanized on 5 dpi and organs were harvested for histopathological evaluation.
3. Results 3.1. Virus isolation and identification A total of 798 samples were collected from the wild birds in northeast China from April to September 2012. The habitats
included wetlands, lakes, and forests in Liaoning, Jilin, and Heilongjiang provinces as shown in Table 1. By adopting the method based on chick embryos, 10 HA-positive allantoic fluid samples were found (positive rate = 1.25%). Of the 10 HA-positive samples, three H5N1 subtype AIVs were isolated from a yellow-browed warbler, common shoveler and mallard, named A/Yellow-browed Warbler/Heilongjiang/18/2012(H5N1)(YBW/HLJ/18), A/Common shoveler/Heilongjiang/137/2012(H5N1)(CS/HLJ/137) and A/Mallard/Jilin/36/2012(H5N1)(MD/JL/36) separately. 3.2. Molecular characteristic analysis The HA genes of CS/HLJ/137, MD/JL/36 and YBW/HLJ/18 contain open reading frames of 1701, 1701, and 1704 nucleotides, coding for 567, 567, and 568 amino acids respectively, including a signal peptide (residues 1–16). Based on the deduced amino acid sequences, the HA cleavage sites of the three viruses are PQRERRRKR⁄GLF (CS/HLJ/137, MD/JL/36) and PQIERRRRKR⁄GLF (YBW/HLJ/18) (Table 2), which contain a series of basic amino acids that meet the characteristic of highly pathogenic AIV in chickens (Horimoto and Kawaoka, 1994). Both CS/HLJ/137 and MD/JL/36 had seven potential glycosylation sites in HA1 [positions 27 (NST), 39 (NVT), 181 (NNT), 209 (NST), and 302 (NSS)] and HA2 [positions 499 (NGT) and 558 (NGS)], while YBW/HLJ/18 had additional sites at positions 289 (NCS) and 302 (NTS). In addition, the three isolated viruses did not contain a glycosylation site at position 170, which was characteristic of HPAI H5N1 viruses in wild birds (Chen et al., 2006). The conserved amino acids residues within the receptor binding site of the HA protein (H3 numbering), including Y98, S136, W153, H183, and Y195, were each present in the three isolated viruses, implying that they retained typical avian
Table 2 Selected characteristic amino acids of H5N1 subtype AIVs isolated from wild birds. Key residues
Comments
The isolates YBW/HLJ/18
MD/JL/36
CS/HLJ/137
GCG/QH/1/09
HA(H3 numbering) Cleavage sites Ser138Ala(S-A) Thr160Ala(T-A) Gln226Leu(Q-L) Gly228Ser(G-S)
Characteristic of high-pathogenic in chickens Favour mammalian adaptation Increased affinity toward a2, 6-type receptor Increased affinity toward a2, 6-type receptor Increased affinity toward a2, 6-type receptor
PQIERRRRKR⁄GLF A A Q G
PQRERRRKR⁄GLF A A Q G
PQRERRRKR⁄GLF A A Q G
PQRERRRKR⁄GLF A A Q G
NA(N1 numbering) Stalk deletion(49–68) His275Tyr(H-Y)
Increased virulence Neuraminidase resistance
Deletion H
Deletion H
Deletion H
Deletion H
PB2 Leu89Val(L-V) Glu627Lys(E-K) Asp701Asn(D-N)
Enhanced polymerase activity Mammalian adaptation Mammalian adaptation
V E D
V E D
V E D
V E D
PB1 His99Tyr(H-Y) Lys207Arg(K-R) Ile368Val(I-V) Tyr436His(Y-H)
Enables droplet transmission in ferrets Enhanced virulence in duck Enables droplet transmission in ferrets Enhanced virulence in duck
H K I Y
H K V Y
H K I Y
H K I Y
PB1-F2 Full-length(87–90) Asn66Ser(N-S)
Full-length needed for virulence in mice Increased virulence in a mice model
Full-length N
Full-length S
Full-length N
Full-length N
PA Ala36Thr(A-T) Thr515Ala(T-A)
Increased replication in mice Enhanced virulence in duck
A T
A T
A T
A T
M1 Asn30Asp(N-D) Thr215Ala(T-A)
Increased virulence in a mice model Increased virulence in a mice model
D A
D A
D A
D A
M2 Ser31Asn(S-N)
Amantadine resistance
N
S
S
S
NS1 Deletion(80–84) Pro42Ser(P-S)
Signalling of host proteins Increased virulence in mice
Deletion S
Deletion S
Deletion S
Deletion S
141
Z. Fan et al. / Infection, Genetics and Evolution 29 (2015) 138–145
virus-like receptor specificity (Skehel and Wiley, 2000). In addition, the HA amino acids at positions 226 and 228 were Gln and Gly (Table 2), indicating that these viruses may retain the characteristic of preferential binding to the a2, 3-type receptor, which is predominant in avian species (Yamada et al., 2006). Compared to A/goose/Guangdong/1/96(H5N1), the three isolated H5N1 viruses all had a deletion of 20 residues (position 49–68) in the stalk region of the NA protein (Table 2), which resulted in the loss of four potential glycosylation sites at positions 50 (NQS), 58 (NNT), 63 (NQT), and 68 (NIS). Therefore, each of the three viruses had three potential glycosylation sites in the NA protein at positions 88 (NSS), 146 (NGT), and 235 (NGS). The H275Y mutation in the NA protein, which has been associated with oseltamivir resistance (Brookes et al., 2011), was not detected. The three isolated viruses had no mutations at positions 627 (E-K) and 701 (D-N) of the PB2 protein, which are important for adaptation of avian influenza viruses to mammals (Katz et al., 2000; Li et al., 2005). Remarkably, two mutations in PB1 and PB1-F2 (I368V and N66S, respectively), which are associated with
A
92
98
62 88
increased virulence in mice and ferrets, were found in the MD/JL/ 36 strain (To et al., 2013). The S31N mutation in the M2 protein, which causes amantadine resistance (Lee et al., 2008), was detected in YBW/HLJ/18. Moreover, the NS1 protein of each of the three isolated viruses had a deletion of five amino acids (80–84) and the PDZ binding motif was EPEV or ESEV, which denote avian-like specificity. Remarkably, P42S mutations were found in the three isolated viruses, which are all associated with interferon (IFN) resistance (Seo et al., 2002; Jiao et al., 2008). 3.3. Phylogenetic analysis To elucidate the genetic relationships of the three H5N1 viruses, the whole genome of each virus was sequenced and all eight gene fragments were compared to sequences of representative influenza viruses that were retrieved from the GenBank database (https:// www.ncbi.nlm.nih.gov/genbank/). In the HA phylogenetic tree (Fig. 1A), the three viruses were clustered into clade 2.3.2.1, and shared 95.3–99.3% homology with
62 A/mandarin duck/Korea/PSC24-24/2010(H5N1) 68 A/goshawk/Tochigi/64/2011(H5N1) 98 A/tundra swan/Fukushima/207/2011(H5N1) 92 A/whooper swan/Mongolia/21/2010(H5N1) 56 A/whooper swan/Mongolia/6/2009(H5N1) 92 A/Anas clypeata/Heilongjiang/137/2012(H5N1) A/great crested-grebe/Qinghai/1/2009(H5N1) 100 A/wild duck/Jilin/HF/2011(H5N1) A/duck/Hunan/S4150/2011(H5N1) 100 68 A/Anas platyrhynchos/Jilin/36/2012(H5N1) 98 A/barnswallow/Hong Kong/1161-SJC001_RG1/2010(H5N1) A/wild bird/Hong Kong/07035-1/2011(H5N1) 100 100 A/wild duck/Fujian/1/2011(H5N1) 100 A/Phylloscopus inornatus/Heilongjiang/18/2012(H5N1) 88 A/Common Magpie/Hong Kong/5052/2007 A/duck/Guangxi/89/2006(H5N1) A/chicken/Hunan/999/2005(H5N1) A/Goose/Guangdong/1/96(H5N1) 100 A/chicken/Shanxi/2/2006(H5N1) A/chicken/Liaoning/23/2005(H5N1) A/goose/Guiyang/3422/2005(H5N1) A/duck/Laos/3295/2006(H5N1) A/Anhui/1/2006(H5N1) 100
Clade 2.3.2.1
0.005
B
89
39
A/bar-headed goose/Mongolia/X53/2009(H5N1)
A/ruddy shelduck/Mongolia/X42/2009(H5N1)
74
A/grebe/Tyva/3/2009(H5N1) 87
44
A/common buzzard/Bulgaria/38WB/2010(H5N1)
Group 1
A/black-headed gull/Tyva/115/2009(H5N1) A/Anas clypeata/Heilongjiang/137/2012(H5N1)
84
72
A/great crested-grebe/Qinghai/1/2009(H5N1) 99
A/Anas platyrhynchos/Jiling/36/2012(H5N1) A/wild duck/Jilin/HF/2011(H5N1)
100 87
A/duck/Lao/469/2010(H5N1))
96
Group 2
A/chicken/Hubei/QE8/2009(H5N1)
A/environment/Hunan/5-25/2007(H5N1) A/house crow/Hong Kong/7677/2008(H5N1) 64
A/Phylloscopus inornatus/Heilongjiang/18/2012(H5N1) A/oriental magpie robin/Hong Kong/9298/2009(H5N1) A/feral pigeon/Hong Kong/3409/2009(H5N1)
67
A/barn swallow/Hong Kong/1161/2010(H5N1)
27
Group 3
A/muscovy duck/Vietnam/LBM57/2011 (H5N1)
58 61
A/duck/Vietnam/LBM133/2012(H5N1)
0.005
Fig. 1. Phylogenetic tree based on the open reading frame sequences of HA (A) and NA (B) gene segment of H5 and N1 subtype viruses. The three H5N1 viruses isolated from Northeast in China were highlighted with a close circle. Different lineage was highlighted in different color. The trees were constructed using the neighbour-joining method of MEGA5.0 with 1000 bootstrap trials performed to assign confidence to the grouping. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
142
Table 3 The highest nucleotide identity to whole genomes of three H5N1 influenza viruses. Gene segment
HA NA PB2 PB1 PA NP M NS
CS/HLJ/137
MD/JL/36
YBW/HLJ/18
Closest viruses
Nucleotide identity
Accession No.
Closest viruses
Nucleotide identity
Accession No.
Closest viruses
Nucleotide identity
Accession No.
WS/MG/6/09 (H5N1) GCG/QH/1/09 (H5N1) CB/BG/38/10 (H5N1) GCG/QH/1/09 (H5N1) GCG/QH/1/09 (H5N1) BG/Tyva/10/09 (H5N1) DK/Lao/463/10 (H5N1) GS/JS/k0403/10 (H5N1)
99.3% 99.1% 99.3% 99.2% 99.0% 99.4% 99.3% 99.6%
AB523767.1 CY063320.1 CY110851.1 GU182156.1 CY063317.2 GQ386152.1 CY098340.1 JQ638682.1
DK/HN/S4150/11(H5N1) WD/JL/HF/11(H5N1) CK/AH/HF/10(H9N2) CK/HB/1102/10(H5N2) CK/SD/BD/08(H9N2) DK/JS/26/04(H3N2) WD/JL/HF/11(H5N1) EV/CS/10/09(H5N1)
99.0% 99.1% 99.3% 99.2% 98.5% 99.2% 99.2% 99.4%
CY1466921 JX534575.1 JX312549.1 JQ041392.1 JF795077.1 KC261670.1 JX534576.1 JN558599.1
WD/FJ/1/11(H5N1) OMR/HK/9298/09(H5N1) WD/FJ/1/11(H5N1) MD/VT/LBM228/12(H5N1) WD/FJ/1/11(H5N1) MD/VT/LBM228/12(H5N1) BHG/HK/709/11(H5N1) GH/HK/1046/08(H5N1)
99.1% 98.7% 99.1% 98.0% 99.4% 99.5% 99.0% 98.6%
JX534589.1 AB557634.1 JX534594.1 AB786682.1 JX534588.1 AB786685.1 KC436117.1 CY036249.1
Table 4 Animal experiments of three H5N1subtype AIVs in chickens and ducks. Birds
Chickens
Viruses
YBW/HLJ/18 CS/HLJ/137 MD/JL/36
Ducks
YBW/HLJ/18 CS/HLJ/137 MD/JL/36
Infected method
3 dpi
5 dpi
7 dpi
9 d pi
Sera-conversion (positive/total)
OP
CL
OP
CL
OP
CL
OP
CL
14 dpi
21 dpi
Inoculated Contacted Inoculated Contacted Inoculated Contacted
2/2a(2.6 ± 0.3)b 1/3(1.25) N 0/3 5/5(2.3 ± 0.7) 2/3(1.6 ± 0.8)
2/2(2.4 ± 0.2) 2/3(
