Influenza is a zoonotic disease caused by influenza A virus, which belongs to the family Orthomyxoviridae. Genetic reassortment in these segmented viruses provides an efficient way of generating novel strains with variable epidemic potential and host ranges. Pigs are prone to infection with human and avian influenza viruses because they express specific receptors for both species. Therefore, pigs are a reservoir for genetic reassortment of influenza viruses that can lead to the development of novel viruses with different host ranges or transmission capabilities (1–4).
Swine influenza virus causes respiratory diseases in pigs and is disseminated worldwide. SIVs, which were first isolated from pigs in the USA in 1930, are believed to have been acquired from humans during the Spanish influenza pandemic of 1918–1919 (1). Three dominant subtypes of SIV, H1N1, H1N2 and H3N2, currently circulate in the pig population worldwide. Classical H1N1 subtype SIVs have spread throughout the major swine populations of the world (5). In North America, classical H1N1 subtype SIVs are still predominantly of the H1N1 subtype (6), whereas H1N1 SIVs found in Europe are entirely of avian origin (7). Human H3N2 viruses, known as “Hong Kong” strains, were frequently transmitted to pigs after the human “Hong Kong flu” pandemic of 1968 (8). In the late 1990s, novel triple reassortant H3N2 SIVs emerged in North America. Genetic analysis of these triple reassortant H3N2 viruses showed that they consisted of HA, NA, and PB1 gene segments of human H3N2 viruses; NS, NP, and M gene segments of classical H1N1 subtype SIVs; and PB2 and PA gene segments of North American avian H1N1 viruses. The six internal genes (PA, PB1, PB2, NP, M, and NS) of these triple reassortant viruses, termed TRIG cassette, played a significant role in the emergence of novel SIVs, because further reassortment between these triple reassortant viruses and classical H1N1 subtype SIVs led to emergence of reassortant H3N1 (9, 10), H1N2 (11) and H1N1 (12) viruses, all of which retained TRIG cassette, among pigs in North America.(13, 14)
A(H1N1)pdm09 virus was first identified in humans in April 2009 in Mexico and the USA. This pandemic strain is believed to have arisen from reassortment of two SIVs, because the NA and M gene segments originated from a Eurasian avian-like H1N1 SIV, whereas the remaining six segments were from a triple reassortant H1 SIV that was mainly circulating in the North American swine population (15, 16). Shortly after the first isolation of A(H1N1)pdm09 virus from humans, researchers found genetically similar pig isolates in the USA and Canada (17, 18). Since then, A(H1N1)pdm09 virus has been isolated from pigs worldwide (19–22).
In Japan, reassortant H1N2 viruses possessing the HA gene, which originates from classical H1N1 subtype SIVs, and the NA gene, from human-like H3N2 viruses, have been predominantly isolated from pigs (23–25). Classical H1N1 subtype SIVs probably appeared in the Japanese swine population around 1977, although there have been few reports of its isolation after the 1980s. Serological examination of pig sera from some regions in Japan has confirmed that “human-like” H3N2 viruses have entered Japanese pig populations from time to time (26). However, there are no published reports about the incidence of SIVs since the 2009 pandemic. Because circulation of SIVs is not only an economic burden on the pig industry, but could also have public health implications, information about the incidence and genetic characterization of SIVs is important.
In this study, we genetically characterized 11 SIVs. These viruses were isolated at various domestic institutions and sent to the National Institute of Animal Health in Japan for detailed characterization between 2009 and 2012. HA and NA subtypes of the 11 SIV isolates were determined to be H1N1, H3N2, and H1N2 by RT-PCR: their designations and origins are listed in Table 1.
|Virus||Subtype||Sampling date||Samples||Clinical signs||Farm (number of livestock)/institution||Segments sequenced in this study||Accession No.|
|A/sw/Osaka/1/2009||H1N1||10/2/2009||Nasal swabs from 10 fattening pigs (80–120 days old)*||Not observed||Domestic farm in Osaka prefecture (sows 100, fattening 900)||HA, NA and M||AB53144, AB531445, AB740971|
|A/sw/Yamagata/11/2010||H1N1||1/12/2010||Nasal swab from a sow (3 years old)||Fever and respiratory symptoms||Domestic farm in Yamagata prefecture (sows 280, boars 20, fattening 2100)||Full genome||AB740972-AB74979|
|A/sw/Yamagata/12/2010||H1N1||1/12/2010||Nasal swab from a sow (2 years old)||Fever and respiratory symptoms||Same as above||Full genome||AB740980-AB740987|
|A/sw/Yamagata/14/2010||H1N1||1/12/2010||Nasal swab from a sow (1 year old)||Fever and respiratory symptoms||Same as above||HA, NA, PB1, PA, NP, NS and M||AB740988-AB740994|
|A/sw/Yamagata/17/2010||H1N1||1/12/2010||Nasal swab from a fattening pig (age unknown)||Fever and respiratory symptoms||Same as above||HA||AB740995|
|A/sw/Yamagata/19/2010||H1N1||1/12/2010||Nasal swab from a fattening pig (5 months old)||Fever and respiratory symptoms||Same as above||Full genome||AB740996-AB741003|
|A/sw/Tochigi/2/2011||H1N2||2/21/2011||Nasal swab from a sow (1 year old)||Anorexia, respiratory symptoms||Domestic farm in Tochigi prefecture (sows 850, piglets 3500, fattening 6500)||Full genome||AB741004-AB741011|
|A/sw/Mie/R01/2012||H1N2||1/19/2012||Pulmonary emulsion from a fattening pig (30 days old)||Dysstasia (slaughtered)||Domestic farm in Mie prefecture (sows 75)||Full genome||AB741012-AB741019|
|A/sw/Yokohama/aq114/2011||H3N2||3/8/2011||Nasal swab from a pig imported from Canada for breeding (2–4months old)||Not observed||Animal Quarantine Service in Yokohama city, Kanagawa prefecture||Full genome||AB741020-AB741027|
|A/sw/Yokohama/aq138/2011||H3N2||3/8/2011||Nasal swab from a pig imported from Canada for breeding (2–4months old)||Not observed||Same as above||Full genome||AB741028-AB741035|
|A/sw/Narita/aq21/2011||H1N1||2/17/2011||Nasal swab from a pig imported from Denmark for experiments (< 1year old)||Anorexia, fever, and respiratory symptoms||Animal Quarantine Service in Narita city, Chiba prefecture||Full genome||AB741036-AB741043|
We determined the complete coding sequences of gene segments and carried out BLAST searches for each segment. We also performed phylogenetic analyses using the neighbor-joining method of the MEGA program (Fig. 1a, b, c, d, e, and Fig. 2). We found that all seven H1N1 viruses had originated from the A(H1N1)pdm09 virus, displaying high identities with human A(H1N1)pdm09 isolates (Table 2). Phylogenetic analysis revealed that the HA, NA and M (Fig. 1a, c, e) genes were clearly distinct from the classical H1N1 subtype SIVs. The HA and NA genes of one virus, A/sw/Narita/aq21/2011, which was isolated from a pig imported from Denmark, showed close homology to a Danish human viral isolate and is suspected to have infected host pigs before or during transit (Fig. 1a, c). Phylogenetic analysis of HA, NA and M genes showed that A/sw/Yamagata/17/2010 and A/sw/Yamagata/19/2010 were genetically distinguishable from A/sw/Yamagata/11/2010, A/sw/Yamagata/12/2010 and A/sw/Yamagata/14/2010 (Figs. 1a, c, e). We identified these viruses in specimens collected from pigs on the same farm at the same time. We isolated A/sw/Yamagata/17/2010 and A/sw/Yamagata/19/2010 from fattening pigs raised in a separated pigpen from that of breeding pigs, specimens yielding A/sw/Yamagata/11/2010, A/sw/Yamagata/12/2010 and A/sw/Yamagata/14/2010 were collected from the latter (Table 1). More detailed phylogenetic analysis of the M gene along with 1197 A(H1N1)pdm viruses (Fig. 2) and the estimated evolutionary rate for the M gene of pH1N1 viruses in previous studies (27, 28), that is 2.55 × 10−3 to 3.87 × 10−3 substitutions per site per year, supports this genetic distinguishability. A/sw/Yamagata/19/2010 was positioned in a distinguishable branch from A/sw/Yamagata/11/2010, A/sw/Yamagata/12/2010 and A/sw/Yamagata/14/2010 in the comprehensive tree and there were three nucleotide differences between A/sw/Yamagata/19/2010 and A/sw/Yamagata/11/2010, A/sw/Yamagata/12/2010 and A/sw/Yamagata/14/2010. Considering the estimated evolutionary rate given above, it can be calculated that almost one year would be needed for three substitutions to occur in the M gene of pH1N1. Therefore, it is more likely that the simultaneous detection of genetically distinguishable isolates from one farm resulted from multiple introductions of A(H1N1)pdm09 viruses into the farm rather than genetic drift during infection within the farm. However, we cannot completely rule out genetic drift.
|Isolated virus||Segment||Virus with greatest homology||Lineage||Identity (%)|
|A/sw/Yokohama/aq114/2011(H3N2) and||PB2||A/Saskatchewan/5350/2009(H1N1)||Triple reassortant||98|
The HA and NA of two H1N2 viruses (A/sw/Tochigi/2/2011 and A/sw/Mie/R01/2012) showed high identity with Japanese classical swine H1N2 isolates by BLAST search (Table 2). Phylogenetic analysis confirmed that the HA and NA genes of these two isolates are of Japanese H1N2 SIVs, which have established themselves as a single clade since their first recognition by A/sw/Ehime/1/1980(H1N2) (Fig. 1a, d). However, all the internal genes, PB2, PB1, PA, NP, M and NS, showed high homology with those of the A(H1N1)pdm09 virus (Table 2). This was supported by the phylogenetic tree of the M gene, as a representative for internal genes (Fig. 1e). This indicates that these two H1N2 SIVs arose from reassortment between A(H1N1)pdm09 and Japanese classical swine H1N2 viruses. Isolation of A(H1N1)pdm viruses from pigs has been reported worldwide since 2009 (17–19); several patterns of reassorted A(H1N1)pdm influenza viruses have been identified in pigs in North America (29, 30), Europe (31), Australia (20), Asia (22, 32) and Africa (21). The isolates from Tochigi and Mie prefectures in this study are another example of reassorted A(H1N1)pdm viruses identified in pigs. Phylogenetic analysis of 1197 A(H1N1)pdm viruses demonstrated that the M genes of these two isolates were positioned in distinguished clades (Fig. 2) and there were six nucleotide differences between them. Considering the mean evolutionary rate of the M gene of pH1N1 as described above (27, 28), more than a year and a half would be needed for six substitutions to occur if they originated from a single introduction of pH1N1 virus to pigs. Since they were isolated only a year apart and there is approximately 500 km between the sites of isolation, it is likely that these isolates arose as a consequence of at least two independent reassortment events. It is worth noting that no closely related ancestral strain to A/sw/Tochigi/2/2011 and A/sw/Mie/R01/2012 has been isolated in Japan, although HA and NA genes of both strains belong to an H1N2 SIV cluster which has been maintained among Japanese pigs since the 1980s (24). An H1N2 virus was isolated in Tochigi prefecture in a previous study in 2008 (A/sw/Tochigi/1/2008) (25). Although the HA and NA genes of A/sw/Tochigi/2/2011 belong to the same cluster as A/sw/Tochigi/1/2008, these viruses are not closely genetically related to each other (Fig. 1a, d). This indicates that the genetic diversity of HA and NA genes of Japanese H1N2 SIVs is larger than one might expect.
Phylogenetic analysis revealed that the HA and NA genes of A/sw/Yokohama/aq114/2011 and A/sw/Yokohama/aq138/2011 are triple reassortant SIVs of H3N2 subtype. They belong to cluster-IV (33), their highest identity value is with North American isolates, A/turkey/BC/1529–3/2005(H3N2) and A/sw/Indiana/A01076191/2010(H3N2), and they were separated from H3N2 viruses of human and avian lineages (Fig. 1b, d). All the internal genes, PB2, PB1, PA, NP, M and NS, are genetically related to H1N1 triple reassortant viruses isolated from humans in Canada (Fig. 1e). We found by BLAST search that the PB2, PA and NS genes have high identity with A/Saskatchewan/5320/2009(H1N1), whereas for PB1 and M genes it is with A/sw/SK/11–16/2009(H1N1) (Table 2). These isolates were from a pig and a human who worked in a pig farm in Saskatchewan, Canada. They were originally thought to be reassortants of TRIG-bearing SIV and either classical swine or seasonal human influenza A viruses. In addition, the NS gene of A/sw/SK/11–16/2009 closely matches that of the A(H1N1)pdm09 virus (34). A/sw/Yokohama/aq114/2011 and A/sw/Yokohama/aq138/2011 isolates are suspected to be triple reassortant H3N2 viruses that have been maintained in pigs in Canada without further reassortment. In Japan, human-like H3N2 viruses have been isolated sporadically, but they appear to be distinct from the two H3N2 isolates in this study (Fig. 1b, d, e). This suggests the possibility that a novel SIV could invade via imported pigs.
The risk of a novel SIV invading from other countries was thought to be negligible in Japan because it is completely surrounded by sea and the frequency and quantity of live animal movement is less than that in other continental countries. Circulation of H1N2 virus in Japanese pig populations for approximately 30 years has led to the belief that the genetic diversity of SIV is low (24, 25). However, our results demonstrate that A(H1N1)pdm viruses and their reassortant strains are already present in domestic pigs in Japan. In addition, it is possible that previously unrecognized SIV genotypes in Japan are being imported from foreign countries. Since the 2009 pandemic, blanket tests have been performed on all imported pigs in Japanese animal quarantine services. The numbers of tested pigs per year were 664 and 1288 in 2010 and 2011, respectively. Although the positive rate was not high, the importance of exhaustive testing in quarantine is now recognized.
The epidemiology of Japanese SIVs remains unclear. Pigs play an important role as mixing vessels for animal and human influenza viruses, providing a place for reassortment and host adaptation to take place (4, 1). To understand how genetically diverse SIVs continuously evolve and circulate, and to predict outbreaks of swine influenza, systematic surveillance of domestic swine populations is urgently needed.