DOI: https://doi.org/10.21203/rs.3.rs-2147685/v1
H11N9 viruses in wild birds might have provided the NA gene of human H7N9 virus in early 2013 in China, which evolved with highly pathogenic strains in 2017 and caused severe fatalities. To investigate the prevalence and evolution of the H11N9 influenza viruses, 16781 samples were collected and analyzed during 2016–2020. As a result, a novel strain of influenza A (H11N9) virus with several characteristics that increase virulence was isolated. Phylogenetic analyses showed that it was a sextuple-reassortant virus of H11N9, H3N8, H3N6, H7N9, H9N2, and H6N8 viruses present in China, similar to the H11N9 strains in Japan and Korea during the same period. This was the H11N9 strain isolated from China most recently, which add a record to viruses in wild birds. Therefore, comprehensive surveillance and enhanced biosecurity precautions are particularly important for the prediction and prevention of potential pandemics resulting from reassortant viruses with continuous evolution and expanding geographic distributions.
Avian influenza viruses are ribonucleic acid viruses of the family Orthomyxoviridae and possess 8 negative-sense RNA segments encoding 11 known proteins [30]. Of these, the two major surface antigens, hemagglutinin (HA) and neuraminidase (NA), form the basis of multiple serologically distinct virus subtypes. With 18 hemagglutinin (H1– H18) and 11 (N1–N11) neuraminidase subtypes, there is considerable antigenic differences among influenza viruses [15, 25]. Currently, 16 HA and 9 NA subtypes combinations exist in harmony with wild waterfowl, the major natural reservoir for all influenza A viruses, cause no overt disease, and emerge to infect domestic poultry and occasionally mammals [3, 15].
Influenza A viruses can be divided into two distinct groups of high or low pathogenicity on the basis of their pathogenicity in chickens. All highly pathogenic avian influenza viruses known to date that mutate from LPAIV have been restricted to subtypes H5 and H7 [1, 23]. Highly pathogenic avian influenza A virus infections caused by H5 and H7 subtypes in humans have been placed on the top priority list among other zoonotic AIVs and have raised concerns that a new influenza pandemic will occur in the future [11, 22]. H7N9 infections caused significant negative impacts on public health, the economy, and national and even global security that had resulted in 1,567 human cases with 615 deaths. Human H7N9 has almost disappeared in 2018 because the effective response including management of LPMs and the vaccination strategy [4, 26]. However, H7N9 AIVs isolated in 2019 were antigenically distinct from the vaccine strain, so that the H7N9 AIV has not been eradicated from poultry in China [34].
The NA gene of human influenza A(H7N9) virus might have originated from influenza A (H2N9, H4N9, H11N9) viruses that circulated in eastern China [17, 27, 28]. Most H11N9 strains usually were found in wild birds. Some studies have described that H11N9 isolated in China can replicate in mammalian cells in vitro [18], and even in mice in vivo without prior adaption [32, 33].Although no H11N9 virus was isolated from human till now, serologic evidence of human past infection with influenza A/H11N9 suggested a potent risk of direct transmission of AIV to humans [7, 16, 31].
Considering that H11N9 viruses might contributed to H7N9 ressortment and have the threat to public health, active surveillance on influenza was required urgently. Recently, H11N9 viruses were detected in China (Feb 2016) [33] and South Korea (2016–2018)[16, 31]. However, the prevalence of them afterward in China was unclear. Thus, this study focused on the surveillance of H11N9 viruses during 2016–2020 to analyze their evolution and epidemic risk. As a result, we isolated one strain of the H11N9 influenza virus in Shanghai in November 2016, and analyzed the genetic origin of it, indicating that it was a local inter-subtype reassortant present in China and might be transmitted to Japan and South Korea, which prompted us to conduct further influenza surveillance in wild birds in the future.
To respond to the H7N9 outbreak, a total of 16781 swab and fecal samples were collected from waterfowl in natural reserves in Shanghai and Jiangxi during 2016–2020. For virus isolation, 9–10-day-old specific-pathogen-free (SPF) embryonated chicken eggs were inoculated with the sample supernatants. Viral RNA was extracted from 200 ml of allantoic fluid and subjected to reverse transcriptase polymerase chain reaction (RT-PCR). First, reverse transcription was performed with the primer Uni12 and GoScript™ Reverse Transcriptase System (Promega). This was followed by PCR in which the reverse-transcription product(s) was amplified by the universal primer set MBTuni-12 and − 13 to amplify the short segments for hemagglutinin (HA), nucleoprotein (NP), neuraminidase (NA), matrix (M), and nonstructural gene (NS). Segment-specific primers were used to amplify the long segments for polymerase basic protein 2 (PB2), polymerase basic protein 1 (PB1), and polymerase acidic protein (PA) (Table S1) [10, 35]. Full-genome sequences of the new isolates were annotated using the Influenza Virus Sequence Annotation Tool of the Influenza Virus Database of the National Centre for Biotechnology Information (NCBI; https://www.ncbi.nlm.nih.gov/genomes/FLU/annotation), then deposited in GenBank. The sequence feature was annotated on Influenza Research Database (IRD;
https://www.fludb.org/brc/sequenceFeatureDetailsReport.spg?decorator=influenza&seqFeaturesNamesId=564986 .).
For phylogenetic analysis, the sequences of full genomes of the top 100 basic local alignment search tool (BLAST) hits for the new isolate were downloaded from NCBI (https://www.ncbi.nlm.nih.gov/genomes/FLU/Database/nph-select.cgi?go=database) and the EpiFlu database from the Global Initiative on Sharing All Influenza Data (GISAID; https://platform.epicov.org/epi3/cfrontend).
Multiple sequence alignments were produced using MUSCLE in MEGA7. Maximum clade credibility phylogenetic trees were generated for the full-genome sequences of the top 100 BLAST hits. ModelTest-NG was used to select the best-fit model for nucleotide substitutions. We used relaxed molecular clock models (uncorrelated exponential clock models) to estimate divergence times. Markov chain Monte Carlo (MCMC) chains were run for 100–1100 million iterations according to the number of sequences. The best-fit substitution and tree models are listed in Table S2. TRACER 1.6 was used to confirm appropriate burn-in and adequate effective sample sizes (ESS > 200) for the MCMC analyses [5]. All phylogenetic trees, visualized using Figtree, are presented in Supplementary Fig. 1.
A novel strain of the H11N9 subtype AIV was isolated from a spot-billed duck and named A/Anas poecilorhyncha/shanghai/SH2/2016 (SH2; GenBank accession numbers MW575006–MW575013). SH2 harbors a single basic residue at the cleavage site and was classified as an LPAI virus [2] (Table 1). Six residues in the HA protein (i.e., A138, E190, L194, G225, Q226, and G228; H3 numbering) were conserved, contributing to the strain’s avian receptor-binding characteristics [20]. Multiple virulence-increasing substitutions were present in SH2, including L89V, G309D, T339K, R477G, I495V, K627E, and A676T in PB2, all of which have been shown to increase virulence and replication in mammals [19]. SH2 contained residues R57, I62, S65, and V100 in PA, which suppress host-cell protein synthesis during infection, attenuating the antiviral response [6]. The putative zinc-finger motif CCHH in helix 9 of M1 in SH2 also plays a critical role in virulence in mice [12]. Remarkably, SH2 had S42 and avian-like NS1 C-terminal PDZ domain ligand (PL) residues of ESEV in the NS1 protein, which appeared to increase virulence in mice [9, 13, 14, 24]. Interestingly, two highly conserved NS1 residues in 99% human influenza A viruses that enhance virulence in mice, F at 103 and M at 106, are also present in SH2 [29]. These H11N9 characteristics appear to have considerable pathogenic potential for humans.
Gene | Sequence variation | Sequence feature | Reference |
---|---|---|---|
HA | PAIASR↓GLF | Low pathogenic cleavage sites | [2] |
HA (H3 num) | A138, E190, L194, G225, Q226, G228 | Conserved characteristics contribute to avian receptor-binding | [20] |
PB2 | K627E, N701D | Decreased virulence and replication efficiency | PubMed: 21849466; PubMed: 16140781 |
PB2 | L89V, G309D, T339K, R477G, I495V, K627E, A676T | Increased polymerase activity | [19] |
PB1 | Y436H | Decreased virulence in mice | PubMed: 17553873 |
PA | Q57R, V62I, L65S, A100V | Increased production of viral proteins | [6] |
NP | A184K | Increased replication and pathogenicity in chickens. | PubMed: 19475480 |
M1 | N30D, T215A | Increased virulence | PubMed:19117585 |
M1 | C148, C151, H159, H162 | CCHH increased virulence in mice | [12] |
M2 | S50C | Modest increasement in virulence in mice | PubMed: 19553312 |
NS1 | 80–84 delete | Attenuated in virus replication in vitro and in vivo (chicken and mice) | PubMed: 20854176 |
NS1 | P42S | Increased virulence in mice | [14] |
NS1 | N101D | Increased virulence in mice | PubMed: 10873787 |
NS1 | L103F, I106M | Increased virulence | [29] |
NS1 | E227, S228, E229, V230 | ESEV increased virulence in mice | [24]; [13] |
Next, nucleotide sequence similarity was conducted to investigate the relationships between SH2 with other influenza A viruses. We found that the HA, NA, NP, and M genes of SH2 were most closely related to viruses in Japan with 99.50–99.94% identities, while the other SH2 inner genes (PB2, PB1, PA, and NS) were most closely related to isolates from Korea and China (Table S3). Phylogenetic analysis showed similar results (Fig. 1; Fig. S1): the surface genes HA and NA were clustered together with those of the A/duck/Ibaraki/99/2016 H11N9 strain, with 99.50% and 99.7% nucleotide identities, respectively; the internal genes NP and M were clustered with those of A/duck/Fukuoka/401202/2016 (H4N6), with identities of 99.94% and 99.80%, respectively; and the internal genes PB2 and PA were most closely related to the strains isolated in Korea. However, the PB1 and NS genes were related to the newly isolated viruses from wild waterfowl in China. The spatial locations of all virus strains, including those related to the SH2 strain, are illustrated in Fig. 1.
We then conduct phylogenetic analyses to identify genomic sources of the SH2 virus. The results showed that all eight genes originated from southern China (Fig. 2; Fig. S1). The HA and NA genes originated from an A/duck/Jiangxi/22620/2012 (H11N9)-like gene pool. The PB2 and PB1 genes originated from the A/goose/Wuxi/7276/2016 (H3N8)-like gene pool, and the PA, NP, M, and NS genes originated from A/EN/Sichuan/03404/2015 (H3N6)‐like, A/environment/Hunan/SD009/2015 (H7N9)‐like, A/duck/Wuhan/WHYF05/2014 (H9N2)‐like, and A/wild bird/Jiangxi/P419/2016 (H6N8)‐like gene pool, respectively. Meanwhile, we found that the H11N9 viruses isolated during 2016–2017 in Japan and Korea were similar with SH2 and shared the same origin. Overall, SH2 is a sextuple‐reassortant virus of H11N9, H3N8, H3N6, H7N9, H9N2, and H6N8 viruses in China. This was the most recent report of the H11N9 virus in China, indicating that it circulated in this country until late 2016, and may have been transmitted to Japan and Korea.
In this study, a reassortant, SH2, was isolated in Shanghai in 2016 and was found to possess several phenotypic characteristics that increase virulence. The substitution L89V in PB2, carried by the SH2 strain, is located in the region involved in interactions with PB1 (residues 51–259) and the region binding to the heat shock protein 90 (HSP90) (residues 1–515) [21]. G309D, T339K, R477G, and I495V are also located in the HSP90-binding region. These substitutions increase polymerase activity by enhancing the interaction of polymerase subunits or between polymerase and host factors to increase virulence in mammals [8, 19]. In addition, most avian viruses, including SH2, have the PA protein residues R57, I62, and S65, which are located in a flexible loop (51–74) and play a major role in turning off host protein synthesis [6]. Surprisingly, the cleavage and polyadenylation specificity factor (CPSF30) binding site in the SH2 NS1 protein contains residues F103 and M106, which are common in human isolates, that stabilize CPSF30 binding and inhibit the production of beta interferon (IFN-β) mRNA [29]. Despite the lower significance of H11N9 LPAIVs for public health, antibody against influenza A/H11N9 has been reported in waterfowl hunters [7]. Thus, some reassortment viruses may possess new characteristics that have implications for public health, requiring more extensive surveillance of avian reservoirs.
Next, we analyzed the origin of the SH2 H11N9 virus. The HA and NA genes originated from A/duck/Jiangxi/22620/2012 (H11N9). Though A/wild bird/Anhui Shengjin Lake/S119/2014 (H11N9), A/wild bird/Anhui Caizi Lake/L306/2014 (H11N9), and A/waterfowl/Korea/S353/2016 [16] had similar genes, their phylogenetic relationships did not clearly indicate whether SH2 had obtained genes from them (Fig. 2; Fig. S1D; Fig. S1F). Apart from the NS gene, the other seven genes were similar to those of viruses in Japan and Korea in 2016. We assumed that viruses isolated in China in 2012–2015 had been transmitted to Japan and Korea by wild birds, and circulated locally in China at the same time. The H11N9 viruses in China, Japan, and Korea in 2016 may have been generated from independent reassortment events involving different NS genes.
In summary, we isolated a novel H11N9 strain in late 2016 that possesses new characteristics of increasing virulence. Although our phylogenetic analysis was based on strains with limited genetic diversity, SH2 is most likely a reassortant virus that originated from domestic birds in China. So, continued influenza surveillance in wild birds is extremely essential to prevent a potent pandemic caused by virus ressortment.
Conflict of Interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as potential conflicts of interest.
Author Contributions
H.H. and B.W. conceived the study. B.W. conducted the experiments, analyzed experimental data, drafted the manuscript, and designed the figures. Q.S. and Z.Y. assisted with virus isolation and sequencing. Q.S. and S.H. helped conduct the genetic analyses and contributed to revising the manuscript.
Acknowledgments
This study was supported by the Strategic Priority Research Program of the Chinese Academy of Sciences (XDA19050204), National Forestry and Grassland Administration, China and National Key R&D Program of China.