Background
Wild water birds usually can be classified as Anseriformes and Charadriiformes, which are represented by duck and gull respectively. They are the natural reservoirs for all avian influenza subtypes. Influenza A virus H13 subtype seems to be highly gull associated [
1] and it is rare to isolate H13 viruses from Anseriformes, such as duck and goose. Since its isolation firstly reported in 1977 [
2], the H13 subtype influenza has been divided into Eurasian and North American lineages according to the evolutional relationship [
3]. The intercontinental and interspecies reassortment of viruses had happened occasionally [
4,
5].
Wild aquatic birds serve as natural reservoirs harboring 16 Hemagglutinin (HA) and 9 Neuraminidase (NA) subtypes of influenza A virus. Influenza A viruses, except for highly pathogenic avian influenza H5 and H7 subtype viruses, usually cause mild or even asymptomatic infection among birds (known as low pathogenic avian influenza). The viruses in migratory wildfowl may spillover to birds or mammals, or reassort with other influenza A viruses, causing diseases with pandemic potential.
An outbreak caused by highly pathogenic avian influenza H5N1 resulted in more than 10,000 deaths of migratory birds in Qinghai Lake in year 2005 [
6,
7]. The virus was then spread to Mongolia, Russian, Europe, and Africa along the migratory flyways in the following years of 2005–09 [
8,
9]. This event is a typical example of the global transmission of avian influenza viruses. Since then, Qinghai Lake has become an animal and human public health concern due to its geographic location, which is major breeding site for migratory birds flying to Australia, India, Siberia and Southeast Asia via the Central Asian-Indian flyway and the East Asian-Australian flyway.
The data of the ecology and distribution of influenza A virus subtype, especially the low pathogenic avian influenza subtypes in the Qinghai Lake region is lacking. In view of this, the project-based avian influenza surveillance of relevant environmental samples in Qinghai Lake was implemented in year 2012 Samples of bird feces and lake surface water were collected in core areas of Qinghai Lake, such as Bird Island and Xiannvwan. A few of influenza A subtype viruses were isolated from collected samples, among which two H13N8 subtype influenza viruses were identified and their characteristics were studied.
Discussion
Influenza A virus surveillance among gulls was systematically conducted in Netherlands, Norway and Georgia [
15,
17,
18]. The annual epidemics in gulls caused by H13 and H16 subtype viruses often occurred. First year gull is more susceptive of H13 and H16 infection than gulls older than one year [
16]. Adult gulls had antibody reaction against H13 and H16 viruses and H16 antibodies were most common [
17]. To date, gulls are considered the natural reservoir of influenza A H13 and H16 subtype viruses and gulls also host other subtypes of influenza A virus with diversity. It is possible to cause potential reassortment within species. Gulls sharing same habitats with wild ducks and shorebirds will increase the risk of cross-species transmission or reassortment of viruses resulting in novel subtype viruses.
Usually, based on its geographic location, influenza H13 subtype viruses are separated into Eurasian and North American lineages. Substantial genetic reassortment of the two continents had happened through wild bird migration. The overlapping places of wild bird flyways became the hot spot for avian influenza ecological and epidemiological study. The surveillance took place in Georgia showed Georgian influenza A virus subtype distribution was different by wild bird flyways. In East Africa and West Asia, H7, 11, 13 and N6 were more concentrated [
18]. Most genes of our two H13N8 viruses are closely related to Georgian isolates that support the concept of virus transmission by wild bird through long-distance migration of east Asia and west Asia flyway. Qinghai lake is located in the crossing places of three wild bird migratory flyways: central Asia, east Asia-Australia, east Africa-west Asia. Hence its importance of avian influenza ecology, evolution should be paid more attention.
Although influenza H13 subtype virus was mostly associated with gulls, the finding of interspecies reassortment with genes from Anseriformes (such as mallard) viruses have also been reported [
19]. In our study, PB1 genes of the two H13N8 viruses were phylogenetically relevant to the viruses from Anseriformes, which is deemed an evidence of interspecies reassortment.
H13 subtype influenza virus can infect gulls to induce antibody reaction [
20]. The susceptibility to H13 virus is presented differently among avian species. Gulls are highly susceptible, ducks and turkeys are resistant to some strains, and chickens are refractory to infections of all strains [
21]. The tissue tropism and pathology of H13 natural infection of black-headed gulls showed that H13 virus has adapted to gulls with minimal pathogenicity, with non-clinical signs [
22]. In addition, gull-related H13 subtype influenza viruses also caused the infection and stranding of marine mammals such as whales [
23].
In our study, the two H13N8 showed dual receptor binding properties, which means that they have a capacity to attach to both human receptor and avian receptor. These viruses may infect human being under certain suitable conditions as binding a 2, 6 linked sialic acids (SA) is a pre- requirement for AIV transmission to humans. The molecular basic of receptor binding specificity is subtyped dependent. Different subtypes, or even different strains of same type, might have different molecular markers of receptor binding specificity. The substitution or mutation of HA protein relative to receptor binding preference was concentrated on sites 226,228,186,190 and 135–137. The soluble H13 HA glycoprotein of A/gull/Maryland/704/1977(H13N6) was purified, and its receptor binding specificity was characterized as binding exclusively of the avian a 2, 3 linked sialic acids receptor [
24], which was dissimilar with the two H13N8 viruses. This is probably due to the different amino acids composition of position 135–136 of HA. We should monitor the ecology of this type of virus in birds and potential reassortment with other subtypes of avian influenza.
According to the criteria of pathogenicity of influenza A virus adopted by OIE, the highly pathogenic influenza was determined by the intravenous pathogenicity index (IVPI) test on chickens. In terms of deduction of in-vivo tests, the determination of the cleavage site of HA by sequencing and trypsin dependence assay should be initiatively taken [
25]. We found that the cleavage site of HA of the two H13N8 viruses was mono-basic cleavage sites, which contained only one basic amino acid in the critical position PAISN
R↓GLF. And the virus growth presented clearly trypsin dependence due to not having multi-basic amino acids at the cleavage site of HA. The low pathogenicity of the two H13N8 viruses was confirmed by the virus inoculation on eggs. No egg death was shown.
Virus titration on different cell types can reflect virus growth and infection abilities. The two H13N8 viruses presented low TCID50 value on A549 and MDCK(10^3.25/100ul, 10^2.75/100ul)with limited growth characteristics. The reason of poor virus replication on A549 and MDCK is probably that these cell lines are of mammalian origin. These viruses can reach higher titer on embryonated eggs that showed avian tissues preference. This result will be ideal basis for avian influenza H13 subtype virus isolation and propagation.
To our knowledge, this is the first time that H13N8 subtype influenza virus isolated from Qinghai lake region is reported. Although there is no evidence showing that this low pathogenic avian influenza subtype virus is a risk to wild birds or human beings, it is helpful to understand the ecology and evolution of avian influenza virus in this region, and to provide foundation for the study of correlation between wild bird migration and virus transmission.