Antigenic characterization of H3N2 influenza A ... - Journal of Virology

2 downloads 61 Views 536KB Size Report
May 1, 2013 - All sera were aliquoted and stored at -70˚C until use. 118 ..... and Dr. John Harkness for proofreading this manuscript. ..... Jin, H., H. Zhou, H. Liu, W. Chan, L. Adhikary, K. Mahmood, M. S. Lee, and G. 509. Kemble. 2005.
JVI Accepts, published online ahead of print on 1 May 2013 J. Virol. doi:10.1128/JVI.00804-13 Copyright © 2013, American Society for Microbiology. All Rights Reserved.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

Antigenic characterization of H3N2 influenza A viruses from Ohio agricultural fairs Running title: Antigenic profile of swine-origin H3N2 IAVs Zhixin Feng1,2$, Janet Gomez1$, Andrew S. Bowman3, Jianqiang Ye1, Li-Ping Long1, Sarah W. Nelson3, Jialiang Yang1, Brigitte Martin1, Kun Jia1, Jacqueline M. Nolting3, Fred Cunningham4, Carol Cardona5, Jianqiang Zhang6, Kyoung-Jin Yoon6, Richard D. Slemons3, and Xiu-Feng Wan1* 1

Department of Basic Sciences, College of Veterinary Medicine, Mississippi State University, Mississippi State, MS 39762, the United States; 2Institute of Veterinary Medicine, Jiangsu Academy of Agricultural Sciences, Nanjing 210014, People's Republic of China; 3Department of Veterinary Preventive Medicine, College of Veterinary Medicine, The Ohio State University, Columbus, OH 43210, the United States; 4USDA/APHIS/WS, National Wildlife Research Center, 5 Mississippi Field Station, Mississippi State, MS 39762, the United States; College of Veterinary Medicine, University of Minnesota, St. Paul, MN 55108, the United States; 6Department of Veterinary Diagnostic and Production Animal Medicine, College of Veterinary Medicine, Iowa State University, Ames, IA 50011, the United States $

Equal contribution to this work.

*Correspondence: [email protected] or [email protected].

27 28 29

Abstract

30

viruses (IAVs) and swine exposure at agricultural fairs has raised concerns about the human

31

health risk posed by IAV-infected swine. Understanding the antigenic profiles of IAVs

32

circulating in pigs at agricultural fairs is critical to developing effective prevention and control

33

strategies. Here 68 H3N2 IAV isolates recovered from pigs at Ohio fairs (2009 to 2011) were

34

antigenically characterized. These isolates were compared with other H3 IAVs recovered from

35

commercial swine, wild birds, and canines, along with human seasonal and variant H3N2 IAVs.

36

Antigenic cartography demonstrated that H3N2 IAV isolates from Ohio fairs could be divided

37

into two antigenic groups: 1) the 2009 fair isolates and 2) the 2010 and 2011 fair isolates. These

38

same two antigenic clusters have also been observed in commercial swine populations in recent

39

years. Human H3N2v isolates from 2010 and 2011 are antigenically clustered with swine-origin

40

IAVs from the same time period. The isolates recovered from pigs at fairs did not cross react

41

with ferret antisera produced against the human seasonal H3N2 IAVs circulating during the past

42

decade, raising the question of the degree of immunity the human population has to swine-origin

43

H3N2 IAVs. Our results demonstrate that H3N2 IAVs infecting pigs at fairs and H3N2v isolates

44

were antigenically similar to the IAVs circulating in commercial swine, demonstrating that

45

exhibition swine can function as a bridge between commercial swine and the human population.

46 47

The demonstrated link between the emergence of H3N2 variant (H3N2v) influenza A

48 49 50

Introduction

51

bi-directional interspecies transmission of influenza A viruses (IAVs). Numerous cases of variant

52

influenza A have been linked to swine exposure at agricultural fairs across the United States and

53

the first documented transmissions of the 2009 pandemic (pdm09) H1N1 IAV from humans to

54

pigs also occurred at agricultural fairs (1-5). Since July 2011, more than 325 confirmed human

55

cases of H3N2 variant (H3N2v) influenza A have occurred in 14 states (6). Many of these cases

56

reported direct or indirect contact to exhibited swine with limited subsequent human-to-human

57

transmission of H3N2v IAVs (7, 8).

The swine-human interface at agricultural fairs has proven to be an important site for the

58

Influenza A viruses can cause sporadic, epidemic, and pandemic respiratory disease in

59

humans and from mild to severe respiratory diseases in horses, pigs, domestic poultry, and sea

60

mammals (9-12). Because the tracheal epithelial cells of swine contain both avian-like alpha 2,3-

61

linked sialic acid receptors and human-like alpha 2,6-linked sialic acid receptors (13, 14), pigs

62

can be infected by avian- and mammalian-origin IAVs which may facilitate reassortment events

63

and result in the subsequent generation of novel IAV capable of infecting humans (13, 14).

64

Currently, multiple antigenically and genetically divergent IAVs co-circulate in swine

65

populations and antigenic recycling of IAVs can occur in modern swine management systems

66

where there is a continuous supply of naive pigs (15). Antigenic characterization of the IAV

67

(H1N1)pdm09 demonstrated that the H1 had been circulating among swine for decades prior to

68

the onset of the 2009 pandemic (16) confirming the potential for the emergence of pandemic

69

IAV from pigs.

70 71

Influenza A virus was first isolated from swine in 1930 (17). This H1N1 strain, likely derived from the 1918 H1N1 pandemic IAV (18), was the only major IAV circulating among US

72

swine for more than 60 years and is now known as “classical” swine influenza virus. A triple

73

reassortant H3N2 (trH3N2) IAV composed of three genes (HA, NA, and PB1) derived from

74

human H3N2 seasonal influenza viruses (huH3N2), two genes (PB2 and PA) from avian-origin

75

IAVs, and three genes (NP, MP, and NS) from “classical” swine influenza virus (19, 20)

76

emerged in swine populations in the mid-1990s and has since become well-established in North

77

America. Phylogenetic analyses of HA genes from H3N2 IAVs isolated from North American

78

pigs after 1998 demonstrates that these trH3N2 IAV can be classified into at least four genetic

79

groups (clusters I, II, III, and IV) (21) with one lineage (cluster IV) predominating since 2005.

80

Neutralization assays with swine antisera demonstrated that these four genetic clusters are also

81

antigenically distinct (22). Several reassortment events between trH3N2 IAVs, human-origin

82

IAVs, and classical swine influenza virus have occurred since 1998 resulting in multiple gene

83

constellations with various HA and NA genes in combination with the Triple Reassortant

84

Internal Gene (TRIG) cassette (which includes PB2, PB1, PA, NP, NS, and M segments) (23,

85

24). Human cases of infection with swine-origin IAVs reported by the Centers for Disease

86

Control and Prevention (CDC) after 1998 were all characterized as containing the TRIG cassette

87

and were linked antigenically or genetically, to IAVs concurrently circulating in pigs. Most of

88

these cases were isolated infections in which patients reported indirect or direct exposure to pigs

89

(25, 26).

90

The degree of immunity the human population has to swine-origin H3N2 IAVs is open to

91

question (6). Because agricultural fairs bring people into close contact with swine, defining the

92

antigenic profiles of the IAVs circulating in exhibited pigs at fairs is critical to developing

93

effective strategies to prevent and/or control human infections of swine-origin IAVs.

94

Hemagglutination inhibition (HI) tests were used to compare the antigenic profiles of IAV

95

isolates recovered from pigs at Ohio agricultural fairs during 2009-2011 with H3 IAV isolates of

96

human, swine, canine and avian origins. The antigenic profiles from HI tests were validated by

97

microneutralization (MN) assays. The genomic sequences of the HA and NA segments of

98

selected swine-origin H3N2 IAV isolates were also analyzed to investigate the genomic changes

99

contributing to antigenic profiles.

100 101

Materials and methods

102

Viruses and Sera. From 2009 to 2011, a systematically designed surveillance effort was

103

performed across 53 agricultural fairs in Ohio, USA (27). From the 1,073 pigs sampled, 155

104

were positive for IAV, with 128 H3N2 viruses isolated. Sixty-eight representative (from the 128

105

total) H3N2 IAV (TABLE 1) were selected for the present study. The HA, NA, and M genes of

106

these isolates were characterized as described previously (27). Additionally, four H3N2v IAV

107

isolates (TABLE 1) were kindly provided by CDC, Atlanta, GA. Also included in the study, and

108

listed in TABLE 2 and 3, were twelve H3N2 IAV isolates from commercial pigs, two H3N2 IAV

109

isolates from wild birds, two H3 IAV isolates from dogs (28, 29), and 12 human seasonal H3N2

110

isolates. After being propagated in Madin-Darby canine kidney (MDCK) cells (ATCC,

111

Manassas, VA, USA), all IAVs were aliquoted and stored at -70˚C until use.

112

Ferret antisera were generated in 6 to 8-week-old ferrets, which had baseline HI titers less

113

than 1:10 against A/Brisbane/10/2007(H3N2), A/Brisbane/59/2007(H1N1), and

114

A/California/4/2009(H1N1). Each ferret was inoculated intranasally with 106 TCID50 for each

115

IAV isolate (TABLE 1). Sera were collected from the ferrets at three weeks post inoculation if

116

the HI titer in a ferret was at least 1:160 at 2 weeks post inoculation. Otherwise, a second dose of

117

106 TCID50 of the corresponding virus was given and serum was collected five weeks post first

118

inoculation. All sera were aliquoted and stored at -70˚C until use.

119

Eight swine H3N2 IAV isolates representing six fair events were used to generate ferret

120

antisera. The virus designation is summarized in TABLE 4. In addition, ferret antisera were

121

produced against 12 human seasonal H3N2 IAV isolates, two H3N2 IAV isolates from wild

122

birds, and two canine H3 IAV isolates for comparison. All these isolates are contemporary H3

123

IAVs (TABLE 1, 2, and 3).

124 125

Genomic sequencing. Viral RNA was extracted using the RNeasy Mini Kit according to the

126

manufacturer’s instructions (QIAGEN Inc., Valencia, CA, USA). Reverse transcription and PCR

127

were performed using influenza-specific primers (30). PCR products were separated using

128

agorose gel electrophoresis and purified using the QIAquik Gel Extraction Kit (QIAGEN Inc,

129

Valencia, CA, USA). Sequencing was performed by the Life Sciences Core Laboratories Center

130

at Cornell University using the Applied Biosystems Automated 3730 DNA Analyzer, with Big

131

Dye Terminator chemistry and AmpliTaq-FS DNA Polymerase. The HA genes of nine

132

agricultural fair isolates are fully sequenced, and the HA genes of 12 commercial swine IAV

133

isolates were partially sequenced and at least covered the entire HA1. The GenBank accession

134

numbers of these sequences are xxxxxx-xxxxxx (pending).

135 136

Hemagglutination inhibition assay. Before conducting HI tests, ferret antisera were treated

137

with receptor-destroying enzyme (RDE) (Denka Seiken Co., Tokyo, Japan) at 37˚C for 18 hours,

138

and then heat inactivated at 55˚C for 30 minutes. The sera were then diluted with phosphate-

139

buffered saline (for a final dilution of 1:10), and tested by HI assay with 0.5% turkey red blood

140

cells according to the WHO manual on animal influenza diagnosis and surveillance

141

(http://www.who.int/vaccine_research/diseases/influenza/WHO_manual_on_animal-

142

diagnosis_and_surveillance_2002_5.pdf).

143 144

Microneutralization assay. The sera were tested by performing a microneutralization (MN)

145

assay in MDCK cells. Neutralizing titers were expressed as the reciprocal of the serum dilution

146

that inhibited 50% of viral growth of 100 tissue culture infectious dose (TCID50) of virus. The

147

MN titers were determined by HA assay using 0.5% turkey red blood cells (31).

148 149

Antigenic cartography and antigenic cluster identification. Antigenic cartography was

150

generated based on HI data using AntigenMap (32, 33) by setting 1:10 as the low reactor. Data

151

normalization was performed as previously reported (34-36). Influenza A viruses were classified

152

into antigenic clusters using k-mean clustering (37). Student’s t-test was performed to assess the

153

significance of antigenic distances.

154 155

Molecular characterization and phylogenic analysis. The multiple sequence alignments were

156

conducted by using the MUSCLE software package (38). The phylogenetic analyses were

157

performed using maximum likelihood by GARLI version (39), and bootstrap resampling

158

analyses were conducted using PAUP* 4.0 Beta (40) with a neighborhood joining method, as

159

previously described (41).

160

161 162

Results

163

Genetic characterization of H3N2 influenza A viruses. Phylogenetic analysis of HA genes

164

showed the HA genes of four human H3N2v isolates and all H3N2 IAVs isolates from both

165

agricultural fairs and commercial swine farms belonged to cluster IV (FIG. 1A) (22). A high

166

degree of genetic diversity was observed among the viruses. The amino acid sequence identity

167

between the HA amino acid sequence of A/swine/Ohio/10SW130/2010(H3N2) and that of

168

A/swine/Ohio/11SW111/2011(H3N2) was 99.3%; the HA protein sequence identity between

169

A/swine/Ohio/09SW64/2009(H3N2) and A/swine/Ohio/10SW130/2010(H3N2) and that

170

between A/swine/Ohio/09SW64/2009(H3N2) and A/swine/Ohio/11SW111/2011(H3N2) was

171

95.8% and 95.1%, respectively and the human H3N2v isolate A/Iowa/07/2011(H3N2) was

172

99.5% identical to A/swine/Ohio/11SW111/2011(H3N2) (TABLE 5).

173

The phylogenetic trees of influenza NA genes showed that the NA genes of these

174

agricultural fair isolates were phylogenetically close to the NA gene of

175

A/swine/Ontario/33853/2005 and to those NA genes of other IAVs in the same cluster, as similar

176

to the HA genes (FIG. 1B).

177 178

Antigenic diversity of H3N2 IAVs from pigs at agricultural fairs. The antisera generated

179

against selected IAV isolates from pigs at Ohio fairs during 2010 and 2011 cross-reacted with

180

the isolates from 2010 and 2011 fairs, but were less reactive to IAV recovered from pigs at fairs

181

during 2009. Similarly, the antisera generated against IAV isolates from the 2009 fair season

182

reacted poorly with 2010 and 2011 swine isolates (TABLE 1). The IAV isolates from pigs at

183

agricultural fairs could be divided into two antigenic groups: 1) the 2009 fair isolates (antigenic

184

cluster H3N2-alpha), and 2) the 2010 and 2011 fair isolates (antigenic cluster H3N2-beta).

185

The k-mean clustering also grouped 84 isolates (68 exhibition swine isolates, 4 human

186

H3N2v isolates, and 12 commercial swine isolates) into one of the two identified antigenic

187

clusters (35 in cluster alpha and 49 in cluster beta) as summarized in TABLE 1. The average

188

distance between the viruses within cluster alpha and cluster beta was 1.45 and 1.07 units,

189

respectively. The minimum distance between antigenic clusters was 2.32 units (e.g.,

190

A/Pennsylvania/14/2010 and A/swine/Iowa/11333/2006) and the maximum distance between

191

clusters was 6.53 units (e.g., A/swine/North Carolina/23592/2009 and

192

A/swine/Ohio/11SW214/2011). The average distance between two clusters from their centers

193

was 4.18 units. Each unit corresponds to a 2-fold change in HI assay. Statistically, the antigenic

194

distances between clusters were significantly different from those within each antigenic cluster

195

(p