Background
Both animal and human infections with influenza A virus have been reported. Sometimes mammals, such as pigs, horses, and seals, and poultry can be infected under natural conditions. Wild aquatic birds are the major reservoir of avian influenza, harbouring 16 haemagglutinin (HA) and 9 neuraminidase (NA) subtypes of viruses [
1]. Although avian influenza viruses present limited replication ability in humans, direct human infection with avian influenza and pandemics caused by reassortment of human influenza have both occurred. To date, human infections have been reported with the H5, H6, H7, H9, and H10 influenza virus subtypes [
2,
3].
Exposure to live poultry markets (LPMs) is an important risk factor for highly pathogenic avian influenza infection. Closure of LPMs has been reported to be efficient in blocking avian influenza transmission [
4,
5]. In China, LPMs are major places enabling influenza dissemination and potential reassortment because of the high densities and the mixture of poultry, pet birds, and wild birds that are often present in LPMs. LPMs are also considered valuable places for influenza ecology research and research on the emergence and re-emergence of influenza type A virus [
6]. Therefore, avian influenza surveillance in LPMs provides not only clues for tracing infection sources but also evidence for risk assessment and decision making [
7‐
9].
Along with avian influenza control and prevention, regular LPM surveillance has been conducted in China annually since 2008. Specimens are sampled from relevant environments, and viruses are isolated and identified. The distribution and prevalence of influenza A virus subtypes in LPMs are then analysed. Multiple subtypes of avian influenza virus have been identified in the more than 10 years of surveillance. One avian influenza virus subtype, H1N8, which is uncommon among animals and animal-associated environments, was identified from poultry drinking water in a LPM in Fujian Province in 2014. To better understand its potential risk to human health, studies were conducted on the genetic and biological characteristics of the virus.
Methods
RNA extraction and real-time RT-PCR
A QIAamp Viral RNA Mini Kit (Qiagen, Hilden, Germany) was used to carry out RNA extraction. Real-time reverse transcription polymerase chain reaction (RT-PCR) assays for the influenza A were performed on each of the samples. The reactions were carried out using an AgPath-ID™ One-Step RT-PCR Kit (ThermoFisher, Waltham, USA) under the following conditions: 10 min at 45 °C; 10 min at 95 °C; and 40 cycles of 15 s at 95 °C and 45 s at 60 °C. The sequences of the primers and probe are as follows: forward primer, 5’GACCRATCCTGTCACCTCTGAC3’; reverse primer, 5’AGGGCATTYTGGACAAAKCGTCTA3’; and probe, 5’FAM-TGCAGTCCTCGCTCACTGGGCACG-BHQ1–3′.
Virus amplification
Viruses or influenza A-positive specimens were inoculated into the allantoic cavity of 9-day-old embryonated chicken eggs for 48 h at 37 °C for virus propagation. The allantoic fluid was harvested, and haemagglutination assays were performed using 1% turkey red blood cells to detect the titres of the influenza viruses.
Full-genome sequencing
PathAmp FluA Reagents (ThermoFisher, Waltham, USA) were used to amplify the full genome. The primer sequences were as follows: 5′-CTGGATACGCCAGCRAAAGCAGG-3′ (sense) and 5′-GACCTGATGCGGAGTAGAAACAAGG-3′ (antisense). Whole-genome sequencing was then performed on an Ion Torrent™ Personal Genome Machine™ platform (Thermo Fisher Scientific) with a read length of 200 base pairs following the instructions, and the data were analysed using CLC Genomics Workbench 7.5.1 software.
Phylogenetic analysis
Phylogenetic analysis was performed using MEGA7.0, and a phylogenetic tree was constructed using the maximum likelihood method. The bootstrap value was tested with 1000 replications for each gene segment. Homology analysis of nucleic acids was performed on the NCBI website with the BLAST platform.
Receptor binding analysis
A solid-phase binding assay was used as described previously [
10,
11]. Synthetic sialylglycopolymers, including 3′-SLN, 6′-SLN, 3′-SL, and 6′-SL (referring to Neu5Acα2, 3Galβ1-4GlcNAc, Neu5Acα2,6Galβ1-4GlcNAc, Neu5Acα2,3Galβ1-4Glc, and Neu5Acα2,6Galβ1-4Glc, respectively), were coated on the plates. For testing, 32 haemagglutination units (HAUs) of each virus were added to each well. ELISA was then performed. The primary antibody was a universal monoclonal antibody against group I HAs, and a goat anti-human IgG-HRP antibody was chosen as the secondary antibody. A tetramethylbenzidine (TMB) substrate solution (BD Biosciences) was used to develop the results. The optical density was read at 450 nm.
Virus pathogenicity in vitro can be reflected by plaque formation with or without trypsin. The plaque formation protocol was performed as follows: 96-well MDCK plates with 3 × 10
4 cells/well were cultured at 37 °C overnight. Serial dilutions of virus were inoculated into MDCK cells. Two hours after virus absorption, overlay medium (2× DMEM and Avicel) with or without trypsin (final concentration of 2 μg/ml) was prepared and added. The cell plates were fixed with 4% paraformaldehyde 1 day later. ELISA was performed with a mouse monoclonal antibody against influenza type A (CDC-WHO kit used at 1:2000 in ELISA buffer) as the primary antibody and a goat anti-mouse IgG (H + L) HRP conjugate (Bio-Rad 172–1011, used at 1:1000 in ELISA Buffer) as the secondary antibody. Finally, plaque formation was visualized by adding True Blue™ peroxidase substrate (KPL 50–78-02) and 0.03% H
2O
2 (1:1000 dilution of a 30% solution) [
12].
Intravenous pathogenicity index test in chicken
Five-week-old specific pathogen-free (SPF) leghorn chickens were purchased from the company Beijing Vital River Laboratory Animal Technology. According to the OIE/WHO guidelines [
13], the chickens were divided into two groups. One group with 10 chickens was infected intravenously with 100 μl of diluted test virus (fresh allantoic fluid diluted 1:10 with sterile isotonic saline). The other group, with 2 chickens, was infected intravenously with 100 μl of isotonic saline. The chickens were observed for 10 days.
Virus replication kinetics in DF1, A549 and PK15 cells
The virus replication capacity in mammalian cells was evaluated. Human type II alveolar epithelial cell line (A549), chicken embryo fibroblasts cell line DF1 (DF1), and porcine kidney cell line (PK15) were inoculated into 96-well plates at a density of 3 × 104 cells/well at 37 °C overnight, and the test viruses were added with a multiplicity of infection of 0.01. The plates were then incubated with DMEM containing 0.5% BSA and 2 μg/ml TPCK-treated trypsin at 37 °C. The supernatants were collected at 12, 24, 36, 48, 60 and 72 h post infection. End-point titration of the supernatants was performed with MDCK cells.
Statistical analysis
The statistical significance of the differences was determined using T tests, and P < 0.05 was considered to indicate a significant difference.
Discussion
H1 subtype influenza viruses are known for their wide range of hosts (from birds to mammalian species such as pigs and humans) and rapid evolution as a result of antigenic drift and antigenic shift, which can cause seasonal influenza epidemics and even influenza pandemics [
1,
19,
20]. The phylogenetic analysis showed that H1 subtype influenza viruses have distinct lineages that can be separated based on avian, human and swine origins. Cross-species transmission events of H1 subtype influenza virus have happened occasionally [
21‐
23]. Our study showed that the H1N8 avian influenza virus A/Environment/Fujian/85144/2014(H1N8) is derived from avian-origin viruses and is evolutionarily distant from mammalian viruses. Although avian-origin influenza viruses are distant from mammalian ones, they still make genomic contributions to the evolution of mammalian viruses [
24]. We should be vigilant about the potential emergence of pandemics due to avian influenza H1 subtype viruses.
Research on avian influenza virus ecology in LPMs in China has demonstrated that H5, H9, and H7 are not the only subtypes of avian influenza viruses; subtypes such as H3, H4, H6, H1, and H2 are also present [
6,
25]. A few studies have indicated that H1N2, H1N1, and H1N3 are relatively easy to detect, although both H1N4 and H1N9 had been isolated as well [
26,
27]. To our knowledge, the H1N8 subtype has rarely been detected [
28,
29]. Our study may be the first systematic study of the biological and genetic characteristics of the H1N8 avian influenza virus, which was isolated from live poultry market-related environments in China.
Receptor binding preference plays an important role in influenza replication and transmission. The ability of avian influenza virus to bind to the human receptor is the basis for efficient human-to-human transmission. The relationship between human receptor binding specificity and HA gene molecular markers has been extensively revealed in H5, H7 and H9 subtype viruses [
22,
30]. Based on the solid-phase binding assay results, the studied H1N8 virus still has an avian receptor binding preference.
Plaque formation assays are some of the most quantitative and useful biological methods for virus research and can be used for quantification of infectivity and identification of individual virus particles with specific biological features. We intended to determine the pathogenicity of the H1N8 avian influenza virus in vitro using this assay. If a virus is highly pathogenic, a plaque can be formed without trypsin present. This property is based on the presence of multiple basic amino acids at the cleavage site of the HA protein [
31]. We found that the H1N8 virus needed trypsin to form plaques, which indicated its low-pathogenic avian influenza characteristics. This phenotype is supported by the sequence of the HA cleavage site, PSVQSR/GLF, which has only one basic amino acid.
Virus growth curves in different cell lines can be used to describe virus replication and proliferation ability. In our study, the H1N8 virus grew better in A549 and PK15 cells than in DF1 cells, which indicated the possibility of a growth preference for mammalian cells. A sharp increase in virus replication was detected 12 h post infection in the A549 and DF1 cells. However, the virus growth in PK15 cells increased to a peak at 36 h post infection. A/Environment/Fujian/85144/2014 displayed avian receptor binding characteristics, indicating that the virus retains avian influenza binding capacity. Despite its great similarity with other avian influenza viruses and its low pathogenicity to poultry, this H1N8 subtype avian influenza virus presented efficient replication in mammalian cells, indicating its potential risk for poultry and even mammals. Further exploration will be beneficial to enhance understanding of the mechanism of H1N8 infection in birds and mammals.
The PB2, PA, and NS genes of the H1N8 virus studied showed the highest similarity with a highly pathogenic avian influenza (HPAI) H7N9 virus, A/Environment/Fujian/S10058/2017(H7N9), which was reported in 2018 [
32]. Although the H1N8 NP gene was most similar to that of the H1N2 virus, it shared high similarity and clustered with those of the H7N9 viruses detected in 2017 and 2018. These HPAI H7N9 viruses were detected in poultry in China [
32] and did not present the same gene constellations as HPAI H7N9 viruses possessing internal genes from H9N2 viruses that crossed species barriers and caused sporadic human infections [
11]. Our study showed that the same genetic pool (the PB2, PA, NP, and NS genes) that circulated in 2014 was still present in 2017, indicating that A/Environment/Fujian/85144/2014(H1N8) acted as a gene carrier and was involved in the evolution of H7N9 viruses in poultry. Active surveillance of avian influenza in LPMs is a major pillar supporting avian influenza control and response.
Conclusions
The H1N8 influenza virus in this study contained a gene constellation of avian origin and might have been involved in the evolution of H7N9 viruses in poultry. In addition, an avian receptor binding preference was present, and low pathogenicity to poultry was confirmed both in vitro and in vivo. This H1N8 virus can also grow well in mammalian cells. Although this subtype of virus is not frequently detected, it still contributes to the diversity of avian influenza ecology and provides some insights for virus surveillance. The results of this study also emphasize the forecasting function of LPMs regarding the trends in avian influenza activity in China.