Persistence of pdm2009-H1N1 internal genes of swine influenza in pigs, Thailand

Swine influenza is one of the important zoonotic diseases of pigs. We conducted a longitudinal survey of swine influenza A viruses (S-IAV) circulating in a pig farm with history of endemic S-IAV infection from 2017 to 2018. The samples were collected from 436 pigs including nasal swab samples (n = 436) and blood samples (n = 436). Our result showed that 18.81% (82/436) were positive for influenza A virus and subsequently 57 S-IAV could be isolated. Then 24 out of 57 S-IAVs were selected for whole genome sequencing and could be subtyped as S-IAV-H1N1 (n = 18) and S-IAV-H3N2 (n = 6). Of 24 S-IAVs, we observed 3 genotypes of S-IAVs including rH1N1 (pdm + 1), rH1N1 (pdm + 2), and rH3N2 (pdm + 2). Since all genotypes of S-IAVs in this study contained internal genes from pdmH1N1-2009, it could be speculated that pdmH1N1-2009 was introduced in a pig farm and then multiple reassorted with endemic S-IAVs to generate diversify S-IAV genotypes. Our study supported and added the evidences that pdmH1N1-2009 and it reassortant have predominately persisted in pig population in Thailand. Thus, monitoring of S-IAVs in pigs, farm workers and veterinarians in pig farms is important and should be routinely conducted.


Scientific Reports
| (2020) 10:19847 | https://doi.org/10.1038/s41598-020-76771-2 www.nature.com/scientificreports/ could contribute the generation of a novel virus with high infectivity and more virulence in pigs or humans. Due to the novel S-IAV reassortant could be found in pig farms, routine genetic monitoring of S-IAVs is important. It has been reported that the transmission of human-origin viruses to pigs could contribute the adaptations or mutations for the fitness of the viruses. For example, adaptive mutations of pdmH1N12009 after infection in pigs, the NP gene of the virus develop D53E mutation correlated to less resistance to the antiviral factor in pigs. While some human-origin viruses contain H289Y substitution to reduce resistance to the antiviral factor and increase viral replication 11 . Some human-origin viruses maintain or loss glycosylation sites in HA protein due to less antibody selection pressure in pigs 12 . Thus genetic analysis of the S-IAVs is important to identify the potential adaptations or mutations that might relate to potential zoonotic of the viruses. The objective of this study was to conduct a longitudinal survey of S-IAVs circulating in a pig farm with history of endemic S-IAV infection from 2017 to 2018. Genetic diversity of S-IAVs and evidence of genetic reassortment of S-IAVs in a pig farm was investigated.

Results
Survey for swine influenza viruses in a pig farm. In Table 1).

Diversity of swine influenza viruses in a pig farm.
Twenty four out of 57 swine influenza viruses were selected for whole genome sequencing. The 24 S-IAVs were selected based on time of sample collection, subtypes of the viruses and virus titer (low Ct-value). To identify subtype of S-IAVs, nucleotide sequences of HA and NA genes were compared with nucleotide sequences in the NCBI database by using BLAST program. Our result showed that 24 S-IAVs could be identified as S-IAV-H1N1 (n = 18) and S-IAV-H3N2 (n = 6) ( Table 1). Phylogenetic tree of H1 was constructed by comparing S-IAV-H1N1 (n = 18) to 125 references viruses. Phylogenetic tree of H1 gene showed that all S-IAV-H1N1s in this study were clustered with alpha sublineage (1A.1.2) of the classical swine lineage (Fig. 1A). It is noted that alpha sublineage (1A.1.2) is a common lineage of endemic Thai-S-IAVs. Phylogenetic tree of N1 was constructed by comparing 18 S-IAV-H1N1 to 70 references viruses. N1 of S-IAV-H1N1s in this study was clustered with either Eurasian avian-like swine lineage (n = 6) or pdm09 lineage (n = 12) (Fig. 1B). The Eurasian avian-like swine lineage is a common lineage of N1 of endemic S-IAVs in Thailand. Phylogenetic tree of H3 was constructed by comparing S-IAV-H3N2 (n = 6) to 142 references viruses. Phylogenetic analysis showed that H3 of S-IAV-H3N2 in this study was clustered with North America lineage, sublineage human-like swine (both Ha and Hb), which is previously identified as common lineages of endemic Thai-S-IAV-H3N2 ( Fig. 2A). Phylogenetic tree of N2 was constructed by comparing S-IAV-H3N2 (n = 6) to 82 references viruses. The reference of NA2 gene included 2 major lineages; North American lineage and Eurasian swine lineage. The North American lineage contains 2 subclusters: N2-1998 and N2-2002 lineages from human seasonal influenza. Phylogenetic analysis showed that N2 genes of S-IAV-H3N2s were clustered into North American lineage with sublineage human-like swine of endemic Thai S-IAV-H3N2 (Fig. 2B). Phylogenetic trees of the internal genes of the S-IAV-H1N1 and S-IAV-H3N2 were constructed by comparing S-IAV-H1N1 (n = 18) and S-IAV-H3N2 (n = 6) to references viruses. The reference viruses included Classical swine lineage, Eurasian swine lineage, North America triple reassortant lineage, human seasonal lineage and pdm09 lineage. Phylogenetic analysis of each internal genes showed that internal genes of S-IAV-H1N1s (n = 18) and S-IAV-H3N2s (n = 6) were clustered with pdm09 lineage indicating S-IAVs in this study acquired internal genes (backbone) from pdmH1N1-2009 virus (Supplement Figures 1-6).
Gene constellation of the S-IAVs was identified by designated by using the combination of eight lineages or clusters of the virus. Of 24 S-IAVs, we observed 3 genotypes of S-IAVs including (1) rH1N1 (pdm + 1) contained H1 from classical swine lineage (CS) of endemic Thai-S-IAVs and the other 7 genes (NA and internal genes) from pdm09 lineage. (2) rH1N1 (pdm + 2) contained H1 from classical swine lineage (CS) of endemic Thai-S-IAVs, while N1 from Eurasian avian-like swine lineage (EA) of endemic Thai-S-IAVs and the 6 internal genes from pdm09 lineage, and (3) rH3N2 contained HA3 and NA2 genes from human-like swine lineage of endemic Thai-S-IAVs and internal genes from pdm09 lineage (Fig. 3A,B). Our result suggested that internal genes of pdm09 lineage persist and become predominant lineage in endemic Thai-S-IAVs (Figs. 1A and 2A, Supplement Table 2). It should be concerned that persistence of gene especially pdmH1N1-2009 in pigs for long period in the pig farms could contribute the generation of a novel virus with high infectivity and transmissibility in pigs or humans.

Discussion
In this study, we conducted a longitudinal survey in a pig farm from January 2017 to November 2018. The previous study on S-IAVs in this pig farm in 2015 showed that this pig farm had been infected with S-IAVs and estimated prevalence of S-IAVs was 6.66% 5 . Comparing to previous study, the S-IAV prevalence in this study was 18.81% (from January 2017 to November 2018), which was higher than previous report in 2015. The possible explanation is that this study was more focusing on target sample collection in piglets and weaning pigs (4-10 week-old), thus higher prevalence of S-IAVs was observed.
In this study, two subtypes of S-IAVs (S-IAV-H1N1 and S-IAV-H3N2) were identified. Comparing to the previous study in Thailand during 2000-2014, three subtypes of S-IAVs were identified, but S-IAV-H1N2 could not be identified in this study. Notably, S-IAV-H1N2 has lower prevalence than S-IAV-H1N1 and S-IAV-H3N2 5,19 . Phylogenetic analysis of H1 gene showed that S-IAVs in this study as well as endemic Thai-S-IAVs belonged to Table 1. Description of Thai S-IAV-H1N1 and S-IAV-H3N2 characterized in this study. a rH1N1 (pdm + 1): reassorted S-IAV-H1N1 genotype contained HA gene from classical swine lineage (CS) and other genes from pandemicH1N1-2009 lineage (pdm09). rH1N1 (pdm + 2): reassorted S-IAV-H1N1 genotype contained HA gene from classical swine lineage (CS), NA gene from Eurasian avian-like lineage (EA) and internal genes from pandemicH1N1-2009 lineage (pdm09). rH3N2 (pdm + 2): reassorted S-IAV-H3N2 genotype contained HA and NA genes from human-like swine and internal genes from pandemicH1N1-2009 lineage (pdm09). www.nature.com/scientificreports/  www.nature.com/scientificreports/ In this study, there were 3 reassortant genotypes of S-IAVs; rH1N1 (pdm + 1), rH1N1 (pdm + 2) and rH3N2 (pdm + 2). In Thailand, at least 7 genotypes have been reported including eH1N1 (6 + 2), eH1N1 (7 + 1), rH1N1 (pdm + 1), rH1N1 (pdm + 2), rH1N2 (pdm + 2), rH3N2 (pdm + 2) and pdmH1N1 (Supplement Table 2) 5,9,20 . The genotype rH1N1 (pdm + 2) and rH3N2 (pdm + 2), which identified in this study, have been previously reported in  www.nature.com/scientificreports/ Based on genetic analysis, the transmission of human-origin viruses to pigs could contribute the adaptations or mutations for the fitness of the viruses. In this study, HA cleavage site, receptor binding sites and antigenic sites of S-IAVs in this study resemble to human viruses, thus the S-IAVs could possibly infect and/or replicate in mammal host including humans. For example, outbreaks of swine origin-H3N2v from pigs to humans and humans to pigs have been documented 29 . In this study, antibodies against H1N1, H3N2 and pdmH1N1 were observed in this pig farm. It is noted that pig had antibody against pdmH1N1, even though the pdmH1N1 viruses www.nature.com/scientificreports/ could not be identified in the pig farm. It has been reported that no cross-reaction between antibodies against H1-S-IAV and H1 pandemic viruses 30 . Thus, the HI titer result confirmed that pigs in this farm were exposed to both endemic S-IAV-H1 and S-IAV-H3 and pdmH1N1. It should be noted that the HI antibody titer against pdmH1N1 could be either from pandemic HA already circulating in this swine farm or from pandemic HA that has been recently introduced from human. Based on our findings, we can speculate that reverse zoonotic infection from workers to pigs as well as zoonotic infection from pigs to workers in the pig farm could be occurred. Thus, monitoring of influenza virus in pigs and workers is important and should be routinely conducted. The management in pig farm is important for example personal hygiene, personal protective equipment (PPE), seasonal influenza vaccination should be practiced and used in the pig farms for prevention and control of influenza transmission in pigs and humans.

Materials and methods
A longitudinal survey of swine influenza viruses in a pig farm. In this study, a longitudinal sample collection was conducted in a pig farm from 2017 to 2018. A pig farm was selected based on a history of S-IAV subtypes H1N1, H1N2 and H3N2 circulated in a pig farm 5 . The pig farm is located in central province, where considered as high-density pig production area in Thailand. The pig farm is a large-scale pig farm contains approximately 1600 sow and produces 2000 piglets per month with 43 building as farm office, 32 pig housing and 10 worker housing. The farm has open-housing system with moderate biosecurity, which birds and domestic animals can access to pig housing areas. The samples were collected from 436 pigs, including nasal swab samples (n = 436) and blood samples (n = 436) from piglets and nursery pigs (4-10-week-old) with clinical signs of S-IAV infection such as coughing, sneezing and nasal discharge. Samples collection was carried out every 4 months in 2017 and every month in 2018 (Supplement Table 1). The nasal swabs were placed in viral transport media (MEM with 7% BSA, 100 U/ml penicillin, 100 mg/ml streptomycin and 1 mg/ml trypsin). The blood samples were withdraw from jugular vein and placed in 5 ml tubes. The samples were kept on ice and transported to the laboratory within 24 h. The Chulalongkorn University, Animal Care and Uses Protocol committee approved the animal study (CU-VET IACUC #1831103). All animal study procedures were performed in accordance with CU-VET IACUC guidelines and regulations.   (Table 1). For phylogenetic analysis, the nucleotide sequences of each gene of S-IAVs from this study were compared with reference S-IAVs. The reference nucleotide sequences of S-IAVs were obtained from Influenza Research Database (https ://www.fludb .org). The reference S-IAVs were selected to represent clades/clusters, geographic locations and times of isolation of the viruses. To construct phylogenetic trees, the nucleotide sequences of each gene of the viruses (both references viruses and viruses form this study) were aligned with Muscle program v3.6 34 in MEGA v7.0 software 35 . Phylogenetic analysis of HA and NA gene of S-IAV was performed by using a BEAST 2.0 program applying a Bayesian Markov Chain Monte Carlo (BMCMC) algorithm. The best-fit substitution model was implemented by bModelTest (Bayesian model test package for BEAST 2). A strict clock model with coalescent constant population was used as model parameters. The Bayesian MCMC chain lengths were 50,000,000 generations, with sampling every 10,000 generations. The tree iteration was discharged with 10% of the chains as a burn-in pattern by using a tree annotator. The resulting MCC tree was drawn with FigTree software (v1.4.2) (Molecular evolution, phylogenetic and epidemiology, Edinburgh, Scotland, UK). Posterior probability and times of most recent common ancestor (TMRCA) among S-IAVs are provided on branches of trees. Phylogenetic analysis of each internal gene of S-IAV was performed by using MEGA v.7.0 (Tempe, AZ, USA) with neighbor-joining method with Kimura 2-parameter with 1000 bootstrap replicates and Beast program with Bayesian Markov chain Monte Carlo (BMCMC) with 50,000,000 generations and an average standard deviation of split frequencies < 0.05. Substitution rates among sites were set in uniform rate and gabs in the sequences were treated in pairwise deletion. To assign genotype of the S-IAVs, lineages or clusters of each gene of the virus were assigned based on the comparison to reference viruses. After lineages or clusters of gene are assigned, the combination of eight lineages or clusters was assigned as genotype or genetic constellation of S-IAVs. The genotypes of the S-IAVs can help to identify reassortment and genetic diversity of the viruses.

Detection and isolation of swine influenza virus.
Serological test for antibodies against swine influenza virus. Hemagglutination inhibitor test (HI test) was used for detecting antibodies against S-IAVs. In this study, HI test was performed to detect antibodies against 3 reference antigens including endemic S-IAV-H1N1 (A/swine/Thailand/CU-CB1/06), pandemic H1N1-2009 (A/swine/Thailand/CU-RA29/2009) and endemic S-IAV-H3N2 (A/swine/Thailand/CU-CB8.4/2007). In detail, the serum sample was incubated at 56 °C for 30 min to remove heat-labile non-specific factor. Then 100 µl of serum sample was treated to remove non-specific inhibitor. For HI test of S-IAV-H3N2 virus, receptor destroying enzyme (RDE) was used to treat serum (50 µl of serum: 150 µl of RDE) and incubated at 37 °C for 24 h and inactivated at 56 °C for 1 h. The RDE-treated serum was mixed with 100 µl of 50% chicken erythrocyte and incubated at room temperature for 1 h. After centrifuging at 2000 rpm for 10 min, the supernatant was free from non-specific inhibitor serum. For HI test of S-IAV-H1N1 and pdmH1N1-2009, 20% of kaolin was used to treat serum (100 µl of serum: 400 µl of 20% kaolin) and incubated at 25 °C for 30 min. For sedimentation of kaolin, the kaolin-treated serum was centrifuged at 2000 rpm for 10 min. The serum was added with 100 µl of 50%