Archives of Virology

, 154:677

A comparison of the pathogenicity of avian and swine H5N1 influenza viruses in Indonesia

Authors

  • Ryo Takano
    • Division of Virology, Department of Microbiology and ImmunologyUniversity of Tokyo
  • Chairul A. Nidom
    • Faculty of Veterinary MedicineAirlangga University
    • Collaborating Research Center Emerging and Reemerging Infectious Diseases, Tropical Disease CentreAirlangga University
  • Maki Kiso
    • Division of Virology, Department of Microbiology and ImmunologyUniversity of Tokyo
  • Yukiko Muramoto
    • Division of Virology, Department of Microbiology and ImmunologyUniversity of Tokyo
  • Shinya Yamada
    • Division of Virology, Department of Microbiology and ImmunologyUniversity of Tokyo
  • Kyoko Shinya
    • International Centre for Medical Research and Treatment, Graduate School of MedicineKobe University
  • Yuko Sakai-Tagawa
    • Division of Virology, Department of Microbiology and ImmunologyUniversity of Tokyo
    • Division of Virology, Department of Microbiology and ImmunologyUniversity of Tokyo
    • International Centre for Medical Research and Treatment, Graduate School of MedicineKobe University
    • International Research Center for Infectious Diseases, Institute of Medical ScienceUniversity of Tokyo
    • Department of Pathological Sciences, School of Veterinary MedicineUniversity of Wisconsin-Madison
Brief Report

DOI: 10.1007/s00705-009-0353-5

Cite this article as:
Takano, R., Nidom, C.A., Kiso, M. et al. Arch Virol (2009) 154: 677. doi:10.1007/s00705-009-0353-5
  • 163 Views

Abstract

Highly pathogenic avian influenza H5N1 viruses are circulating in many countries. We recently discovered that these viruses have been transmitted to pigs on multiple occasions in Indonesia. To investigate whether avian H5N1 influenza viruses adapted to mammals through their introduction into pigs, we examined the growth of avian and swine isolates in cell culture and compared their pathogenicity in mice. We found that swine isolates were less virulent to mice than avian isolates, suggesting that the viruses became attenuated during their replication in pigs. Continuous surveillance of H5N1 viruses among pigs is clearly warranted.

Abbreviations

HA

Hemagglutinin

MDCK

Madin–Darby canine kidney

EID50

50% Egg infective dose

TCID50

50% Tissue culture infective dose

MLD50

50% Mouse lethal dose

Ck

Chicken

Dk

Duck

Fl

Feline

Qa

Quail

IDN

Indonesia

Sw

Swine

PB2

Polymerase basic protein 2

The highly pathogenic H5N1 avian influenza A virus was first isolated from geese in Guangdong province, China, in 1996 [1]. Since then, the viruses have reached across Asia, Europe, and Africa, giving rise to >380 human infections, with a high mortality rate of >60% [24]. In Indonesia, the H5N1 virus outbreaks were first confirmed in poultry in December 2003. The virus spread rapidly and was identified in almost all of the provinces of Indonesia by the end of 2006 [4, 5]. Human infection with an H5N1 virus was first confirmed in western Java in July 2005. As of 16 December 2008, Indonesia had more than 139 confirmed human cases with 113 deaths [4]. This is the highest case number and mortality rate (>80%) among all affected countries, increasing the risk of emergence of a pandemic virus.

Pigs have been considered an intermediate host for avian influenza viruses to adapt to humans [6]. Epithelial cells in pig trachea can be infected with avian and human influenza viruses and support the replication of both types of virus because they bear receptors for these viruses [7, 8]. These features raise two possibilities for the emergence of a human pandemic virus. One possibility is the emergence of a reassortant virus of avian and human viruses via the co-infection of both viruses in a pig [8, 9]. Alternatively, repeated replication in pigs may allow an avian virus to adapt to grow efficiently in pig epithelial cells and replicate efficiently in mammalian cells [8].

In the 20th century, we experienced three human pandemics caused by the introduction of avian influenza virus into the human population. Previous studies have revealed that the Spanish pandemic was caused by an avian influenza virus that adapted to humans, and the Asian and Hong Kong pandemics were caused by avian–human reassortant viruses [1012], although it was not proven that these strains were generated in pigs.

Available evidence suggests that swine H1N1 influenza viruses are sporadically transmitted to humans [13, 14]. Moreover, swine workers have been shown to have increased levels of antibodies to swine influenza viruses, suggesting that humans are susceptible to swine-adapted influenza viruses [15, 16]. H5N1 infections in pigs have been reported rarely or gone unnoticed altogether because pigs infected with H5N1 influenza viruses show no influenza-like signs or symptoms [17, 18]. There have been no reports on whether adaptation to pigs makes avian viruses adapt to replicate in other mammalian hosts.

We previously investigated the prevalence of H5N1 influenza viruses among pigs in Indonesia during 2005–2007 and discovered that pigs were asymptomatically infected with H5N1 viruses every year [Nidom et al., unpublished]. While Indonesian H5N1 viruses were exclusively classified into three clades, i.e., clade 2.1.1–2.1.3, based on phylogenetic analysis of their HA genes [19], our phylogenetic analysis of the HA gene revealed that our Indonesian swine H5N1 influenza isolates were classified into three sub-lineages, i.e., clade 2.1.1, 2.1.3, and an undetermined clade (Fig. 1). In addition, phylogenetic analyses of the remaining seven viral genes illustrated that the phylogenetic relationships established for the HA gene were maintained, and swine isolates were most closely related to isolates from chickens kept nearby, indicating transmission of avian H5N1 viruses to pigs via infected chickens on at least three independent occasions [Nidom et al., unpublished].
https://static-content.springer.com/image/art%3A10.1007%2Fs00705-009-0353-5/MediaObjects/705_2009_353_Fig1_HTML.gif
Fig. 1

Phylogenetic relationship between the HA genes of H5N1 influenza A viruses from pigs and birds. HA residues 281–1,672 of Indonesian viruses were analyzed by neighbor-joining methods using Clustal W based on evolutionary validation. The tree was rooted to A/goose/Guangdong/1/96. Numbers below or above the branch nodes indicate neighbor-joining bootstrap values. Viruses isolated from pigs and their closest relatives isolated from domestic poultry are indicated by boldface. The sequence data used in this analysis have been submitted for publication, and more detailed analyses were performed. Here, we show an abbreviated phylogenetic tree to indicate the genetic relationship between the test viruses

Here, to determine the pathogenicity of H5N1 pig isolates for mammalian hosts, we compared the virus growth properties in mammalian cells and pathogenicity to mice of the swine isolates and the most closely related H5N1 chicken isolates.

We analyzed two chicken and three swine isolates, i.e., A/chicken/East Java/UT6044/07, A/swine/East Java/UT6005/06, and A/swine/East Java/UT6010/07 of clade 2.1.3 and A/chicken/East Java/UT6016/06 and A/swine/East Java/UT6003/06 of the undetermined clade (A/Indonesia/6/05-like). Although A/chicken/Indonesia/R60/05 was most closely related to our swine isolate in clade 2.1.1, we were unable to obtain this virus and therefore could only analyze these five strains.

All of the H5N1 viruses were isolated in 10-day-old embryonated chicken eggs. All virus stocks were prepared from the allantoic fluid of infected eggs and stored at −80°C until use. The infectivity of stock viruses was determined in 10-day-old embryonated chicken eggs and Madin–Darby canine kidney (MDCK) cells according to standard procedures. The 50% egg infectious dose (EID50) and the 50% tissue culture infectious dose (TCID50) were calculated prior to mouse experiments by the Reed–Muench method [20]. To determine the 50% lethal dose for mice (MLD50), female 5-week-old BALB/c mice (Japan SLC, Hamamatsu, Japan) were intranasally inoculated with 50 μl of tenfold serial dilutions containing doses ranging from 101 to 107 EID50 and monitored for survival over a 14-day period.

Growth properties in eggs and MDCK cells demonstrated that all viruses, whether from chickens or pigs, grew well in both eggs and MDCK cells, indicating that these swine H5N1 viruses grow as well as chicken viruses (Table 1). Among the avian influenza viruses isolated from humans, several have adapted to grow better in mammalian cells than avian isolates [2123]. However, among the swine viruses isolated in our study, none adapted to grow substantially better in mammalian cells than chicken isolates.
Table 1

Growth and pathogenicity comparison between avian and swine H5N1 viruses

Virus

HA classificationa

Virus growthb

MLD50 (log10EID50)

log10EID50/ml

log10TCID50/ml

A/chicken/East Java/UT6016/06

A/Indonesia/6/05-like

8.5 ± 0.5

8.8 ± 0.3

3.8

A/swine/East Java/UT6003/06

9.6 ± 0.4

9.1 ± 0.3

>6.1

A/chicken/East Java/UT6044/07

2.1.3

8.4 ± 0.3

8.6 ± 0.1

2.2

A/swine/East Java/UT6005/06

9.4 ± 0.1

9.2 ± 0.4

>6.1

A/swine East Java/UT6010/07

9.5 ± 0.5

9.0 ± 0.4

3.5

aSee the phylogenetic tree of Fig. 1

bAll data are the mean ± standard deviations from three independent experiments

A comparison of pathogenicity in mice demonstrated that all of the swine H5N1 viruses were less virulent or attenuated in mice than the chicken viruses (Table 1); two swine strains (A/swine/East Java/UT6005/06 and A/swine/East Java/UT6003/06) were strongly attenuated in mice.

A genetic comparison between the chicken and swine strains identified several amino acid candidates potentially responsible for the attenuation in mice (Table 2). As for A/Indonesia/6/05-like viruses, there were 12 amino acid differences between the chicken and swine viruses. We were unable to determine which mutations were strongly correlated to low virulence in mice because these mutations are frequently found among avian and human H5N1 viruses. As for clade 2.1.3 viruses, A/swine/East Java/UT6005/06 had three amino acids that are thought to be responsible for attenuation in mice. Two amino acids in the PB2 protein, glutamine 13 and glycine 150, where mutations were found in Sw/UT6005, were highly conserved among the influenza A viruses, suggesting that these two mutations may be responsible for attenuation.
Table 2

Amino acid differences between the proteins of swine and chicken strains

HA classification

A/Indonesia/6/05-like

Virus

Ck

Sw

UT6016

UT6003

MLD50 (log10EID50)a

3.8

>6.1

Protein position

 PB2

  526

R

K

 PA

  231

T

A

  618

A

T

  684

G

E

 HA1b

  36

A

T

  133

A

S

 NP

  50

S

N

  318

S

P

 NA

  266

S

G

 NS1

  

  27

M

L

  74

G

D

  180

F

L

HA classification

Clade 2.1.3

Virus

Ck

Sw

Sw

UT6044

UT6005

UT6010

MLD50 (log10EID50)

2.2

>6.1

3.5

Protein position

 PB2

  13c

Q

H

Q

  150d

G

V

G

  184

T

A

A

  317

L

M

M

  365

I

M

M

  678

G

D

D

PB1-F2

  4

G

E

E

  47

S

N

N

 PA

  367

R

K

K

  397

K

E

E

  547

D

N

N

 HA1

  156

A

T

T

 NA

   

  17

I

V

V

  41

G

R

G

  197

K

R

R

 M2

   

  23

S

N

N

  29

A

V

V

  68

V

I

I

 NS1

  124

K

I

I

  212

N

S

S

 NS2

  60

I

V

V

  92

N

N

T

aThe same data as shown in Table 1

bH5 numbering

cGln at 13 position of the PB2 protein is widely conserved among H1N1, H3N2, and H5N1 viruses

dGly-to-Val substitution at 150 position of the PB2 protein is a novel mutation among H1N1, H3N2 and H5N1 viruses

Since our swine strains were isolated from pigs with no apparent influenza-like symptoms, the decrease of pathogenicity in mice suggests that the H5N1 viruses may have lost their pathogenicity in mammals during replication in pigs. Given that for the H5N1 viruses to cause a pandemic, they would likely become attenuated in humans, becoming attenuated in mammals may be a prelude to the generation of a pandemic strain.

The fact that highly pathogenic H5N1 avian influenza viruses were transmitted to pigs on multiple occasions in Indonesia has afforded the opportunity for avian H5N1 viruses to adapt to growth in pigs asymptomatically. Pigs also serve as a source for the emergence of avian–human hybrid reassortant viruses [9, 24], increasing the risk of a pandemic strain surfacing. Given the gravity of this situation, it is clear from this study that the continuous surveillance and management of H5N1 viruses in pigs is a necessity.

Acknowledgments

We thank Susan Watson for editing the manuscript. This work was supported by a grant-in-aid for Specially Promoted Research and by a contract research fund for the Program of Funding Research Centers for Emerging and Reemerging Infectious Diseases from the Ministries of Education, Culture, Sports, Science, and Technology and by grants-in-aid of Health, Labor, and Welfare of Japan, by ERATO (Japan Science and Technology Agency), and by the National Institute of Allergy and Infectious Diseases Public Health Service research grants.

Copyright information

© Springer-Verlag 2009