1 INTRODUCTION

The H3 viruses possess high adaptability to infect various hosts, ranging from birds to mammals (Bean et al., 1992). Among multiple cocirculating avian influenza virus (AIV) subtypes in domestic poultry in China, H3 viruses are one of the most frequently identified subtypes and have undergone extensive reassortments (Deng et al., 2013; Luo et al., 2017). Of note, some H3 viruses shared a close genetic relationship with highly pathogenetic H5N8 viruses (Cui et al., 2016). Additionally, H9N2 has undergone a constant change of genetics and pathogenicity since 1994 onward (Bi et al., 2011). By contributing their internal genes, H9N2 viruses have facilitated the emergence of novel reassortants such as H5N6, H7N9, and H10N8 (Bi et al., 2016; Liu, Li, et al., 2015; Liu, Xie, et al., 2015; Pu et al., 2015). This frequent reassortment raises the concern about its potential to have continuous reassortment with other subtypes such as H3. In the interim, constant surveillance in wild bird population has been carried out. Although substantial phylogenetic diversity has been proposed (Zhang, Li, Zhu, Chang, & Xu, 2019; author's unpublished results), the extent of reassortment and the potential emergence of novel strains are likely to be underestimated due primarily to the limited surveillance in wild birds.

2 MATERIALS AND METHODS

To elucidate a complete landscape of AIV ecology in wild birds, we undertook a routine bird ring survey in Poyang Lake, Jiangxi Province, in 2014–2016. Details of surveillance and isolation of AIVs have been presented in our previous studies (Zhang et al., 2019). Briefly, tracheal and cloacal swab samples were collected from migratory birds and domestic ducks and chickens. Virus isolation using these specimens was conducted in 9- to 11-day-old specific pathogen-free embryonated chicken eggs. The viral RNAs were extracted from allantoic fluid of samples with hemagglutination activity using RNeasy Mini Kit (Qiagen), and reverse transcription was carried out using the SuperScript III Reverse Transcription-PCR (RT-PCR) Kit (Invitrogen). The subtype of these positive samples was determined using PCR of a marker gene (Tsukamoto et al., 2009; Lee, Chang, Shien, Cheng, & Shieh, 2001). All segments of these samples were amplified using a Phusion High-Fidelity PCR System (New England Biolabs) (Hoffmann, Stech, Guan, Webster, & Perez, 2001) and sequenced as individual amplicons using Applied Biosystems Automated 3730xl DNA Analyzer.

Here, we report the isolation of two novel H3N6 reassortants from swan goose in 2014. They were designated as A/swan goose/Jiangxi/H23/2014(H3N6) (Gs/JX/H23) and A/swan goose/Jiangxi/H24/2014(H3N6) (Gs/JX/H24).

To investigate the phylogenetic relationship of Jiangxi H3N6 viruses with other AIV subtypes, we selected representative sequences by using the clustering algorithm and BLAST search. For each of the eight genes, we initially retrieved all the sequences available from GenBank and GISAID repositories. To reduce the size of this dataset, we firstly removed identical sequences by keeping the sequence with the earliest data. With the remaining sequences, we constructed the maximum-likelihood tree and used the clustering algorithm to select representative sequences. Additionally, we used BLAST search to identify sequences with >90% identity. Sequences selected by clustering algorithm and BLAST search were integrated as a dataset of reference sequences (Tables A1–A8). Molecular phylogenetic analyses were conducted using the maximum-likelihood method based on the Kimura 2-parameter model (Kimura, 1980) with 1,000 bootstrap replicates in MEGA 6.0 software (Tamura, Stecher, Peterson, Filipski, & Kumar, 2013). Nucleotide substitution models were compared and selected based on BIC (Table A9). Phylogenetic relationship between H3N6 and other subtypes was investigated based on the topology of phylogenies, which was then verified by estimating the evolutionary divergence.

3 RESULTS AND DISCUSSION

A BLAST search in GenBank database showed that most of the gene segments of Jiangxi H3N6 viruses shared a high level of nucleotide identity with a local H5N6 strain, that is, A/chicken/Jiangxi/NCDZT1126/2014, with the identity over 93% (Table 1). Additionally, H3N6 viruses contain the closest HA gene to that of A/common shelduck/Mongolia/2076/2011(H3N8) with >98% identity. Of note, PB1 gene of Gs/JX/H23 and Gs/JX/H24 was closely related to an H9N2 strain, that is, A/chicken/Liaoning/1116/2012, with 98.46% and 98.63% identity, respectively.

Phylogenetic analyses showed that all segments of Jiangxi H3N6 viruses belonged to the Eurasian lineage (Figures A1-A8). The phylogeny of surface genes revealed that Jiangxi H3N6 viruses are likely originated from H3Ny (N3, N6, N8) and H5N6 viruses. Most of their internal genes (except for PB1) are derived from recent H5N6 viruses isolated from both wild and domestic birds. Notably, PB1 gene is of the H9N2-derived origin from domestic birds (Table 2). Also, the various origins possessed by novel Jiangxi H3N6 viruses identified in wild birds in Poyang Lake suggest the complex reassortment of AIV subtypes in China. It is acknowledged that cocirculation of different subtypes serves as the primary mechanism for the generation of novel variants through extensive gene reassortment. Among these variants, H5N6 and H9N2 were identified as two dominant AIV subtypes with dynamic reassortments among birds in China (Bi et al., 2016; Liu, Shi, & Gao, 2014). Our bird survey revealed the high prevalence of H5N6 viruses and mixed samples containing H5 and N6 genes (author's unpublished results), indicating the possibility that H5N6 may have greater compatibility with other subtypes within wild bird reservoir. As such, they are more likely to contribute their genes to generate novel strains such as the H3N6 viruses isolated in this study (Figure 1). Likewise, previous studies demonstrated that novel reassortant H3N6 viruses isolated from domestic ducks contain some gene segments from the dominant H5N6 viruses (Bi et al., 2016; Li et al., 2016; Liu, Li, et al., 2015; Liu, Xie, et al., 2015).

The genetic analysis strongly argues the crucial role that long-distance migration birds have played in viral introduction and circulation at Poyang Lake, a migratory corridor situated within East Asian Flyway. Some wild bird species could carry AIVs over several hundred kilometers of their movement and shed viruses at stopover sites. Given the free-range manner and limited biosafety measures of poultry raising in Poyang Lake, sharing common habitats such as the same waterbody and outdoor areas might have created opportunities for poultry to acquire viruses shed by migratory birds. Through frequent exposure and continuous reassortment, viruses isolated from wild and domestic birds share a high genetic relatedness which suggests a two-way viral transmission at the wild–domestic interface.

4 CONCLUSIONS

In summary, two H3N6 reassortants were isolated from wild birds in Jiangxi Province, presumably originated from Eurasian H3Ny, H5N6, and H9N2 viruses. The findings suggested the complexity of reassortment of multiple AIV subtypes. Therefore, active surveillance in wild birds could improve the early warning system for an avian influenza outbreak.

CONFLICT OF INTEREST

None declared.

AUTHOR CONTRIBUTIONS

Ruiyun Li: Conceptualization (lead); formal analysis (lead); methodology (lead). Tao Zhang: Conceptualization (equal); formal analysis (equal); investigation (equal). Jian Xu: Investigation (equal). Jianyu Chang: Investigation (equal). Bing Xu: Funding acquisition (lead).

ETHICAL STATEMENT

All animal work was approved by the Beijing Association for Science and Technology (approval SYXK [Beijing] 2007-0023). The laboratory animal research was performed in the microbiology laboratory of China Agricultural University, following the Beijing Laboratory Animal Welfare and Ethics guidelines of the Beijing Administration Committee of Laboratory Animals and China Agricultural University Institutional Animal Care and Use Committee guidelines (ID: SKLAB-B-2010-003).