Influenza B viruses in pigs, Taiwan

Background Influenza B viruses (IBVs) have never been isolated from natural‐infected pigs in clinical cases, although the susceptibility of domestic pigs to experimental IBV infections had been confirmed as well as IBV‐specific antibodies were detected from pigs under natural and experimental conditions. Objectives We aimed to assess and investigate the activities for infection and circulation of IBVs in pigs. Methods Annual active surveys for influenza have been implemented on swine populations in Taiwan since July 1998. Nasal swabs, trachea, lungs, and blood from pigs were tested using virological and serological assays for influenza. Gene sequences of influenza viral isolates were determined and characterized. Preliminary sero‐epidemiological data for influenza virus were investigated. Results Three strains of IBV were isolated and identified from natural‐infected pigs in 2014. Genetic characterization revealed the highest identities (>99%) of molecular sequence with the contemporary IBVs belonged to the B/Brisbane/60/2008 genetic clade of Victoria lineage in the phylogenetic trees for all 8 genes. IBV‐specific antibodies were detected in 31 (0.2%; 95%CI: 0.1%‐0.2%) of 15 983 swine serum samples from 29 (2.8%; 95%CI: 1.9%‐3.9%) of 1039 farm visits under annual active surveys from 2007 through 2017. Seropositive cases have been found sparsely in 1‐5 of test prefectures every year except 2015 and 2017 as well as scattered loosely over 26 townships/districts of 11 prefectures in Taiwan cumulatively in 11 years. Conclusions Influenza B viruse infections from humans to pigs remained sporadic and accidental currently in Taiwan but might have paved potential avenues for newly emerging zoonotic influenza in the future.

had been sporadically isolated from pheasants (in Hungary), 9 horses (in Japan), 10 dogs (in 3 urban communities of Taiwan, July 1971), 11 and seals (in the Netherlands, B/seal/Netherlands/1/1999) 12 as well. To our knowledge, IBVs have never been isolated from naturalinfected pigs in clinical cases. Experimental infection of 12 piglets with infectious IBV (B/Budapest/10/1965) on February 1, 1968, was first conducted in Hungary. 13 The susceptibility of the domestic pig to IBV was confirmed. A serological survey for Midwest domestic swine herds of the United States from 2010 to 2012 was conducted.
Meanwhile, an experimental challenge study was performed in pigs using two representative IBVs: B/Yamagata /16/1988 andB/ Brisbane/60/2008. 14 The susceptibility of domestic pigs to IBV infection in the US study supports and extends former observations made in the 1960s. 13 In this report, we presented gene sequence analyses and comparisons for IBVs isolated from natural infections in Taiwanese

| Sample studied
Official veterinarians at local animal disease control centers had in-  Hilden, Germany). Several sets of primer specific for detection and amplification of IAV genes and IBV genes (primer names and sequences provided in: Table S1) were used in reverse transcriptionpolymerase chain reaction (RT-PCR).

| Sequence analyses
The resultant RT-PCR products were purified using a QIAquick PCR purification kit (QIAGEN, Hilden, Germany) and sequenced with IBV gene-specific primers (primer names and sequences provided in: Table S1). Nucleotide sequences were determined with an ABI

| Serological analyses
Hemagglutination-inhibition (HI) assays 15 were used to detect antibodies against IBVs in swine serum samples. To eliminate non-specific inhibitors, swine sera were treated with 100 U/mL of receptor-destroying enzyme (Cholera filtrate, Cat. No. C8772; Sigma-Aldrich/Merck KGaA, Darmstadt, Germany) at 37°C overnight, heat-inactivated at 56°C for 30 minutes, and absorbed with 10% (v/v) rooster or male turkey erythrocytes at 4°C for 1 hour.
The reciprocal of the highest serum dilution that completely inhibits hemagglutination is considered HI titer. The virus neutralization (VN) assays 15 were performed to confirm the results of the HI assays. Untreated original aliquots of HI-positive swine sera were 10-fold-diluted with sterilized phosphate-buffered saline (pH 7.2 ± 0.2) and then heat-inactivated at 56°C for 30 minutes. A serial 1:2 dilution of inactivated serum samples was assayed using above-mentioned reference viruses (100 TCID 50 /50 μL) on MDCK cell line cultures.

| Isolation and identification of influenza B virus from pigs
The nasal swabs of fattening pigs collected from 3 swine farms in 2014 had been confirmed to harbor IBVs. Case history for swine herds where 3 sIBVs isolated is provided in Table S2. Culture su-

| Amino acid sequence analysis of swine influenza B virus
Comparisons with B/Taiwan/113/2014 (the best-matching virus for HA and NA genes) revealed that all sIBVs have N197D and

| Serological data of influenza B virus in swine populations
Based on a conservative HI cut-off titer of 1:40 for seropositivity, 14,19 antibodies against sIBVs were detected in 31 (0.2%; 95% CI:  (Table 1). Notably, the first one of sIBVs was isolated from swine farm in the Wandan Township of Pingtung County (data provided in: Table S2), but none of positive swine serum samples was found from specimens collected at the same time and any subsequent visits thereafter. As for the other 2 swine farms which IBVs were isolated, none of IBV-specific antibodies had ever detected from serum samples collected at the same farms and any other farms in the same township or district.

| D ISCUSS I ON
Even IBV is thought to be primarily a human pathogen, several serological, molecular, and experimental infection evidences have Case history for swine herds where 3 sIBVs were isolated manifested that asymptomatic influenza infections occurred independently in pigs on swine farms, which are small scale as well as have poor animal health management, biosecurity, and hygiene practices, in different parts of Taiwan (data provided in: Table S2).
Our long-term experiences indicated that most of IAVs were isolated from symptomless or mild sick pigs under annual active survey. Therefore, it would not be surprised that IBVs could be isolated from subclinical pigs in this study. Besides, a matter worthy of note is that none of personnel in those 3 commercial swine farms had received seasonal influenza vaccine. In Taiwan In Taiwanese human populations, influenza B predominated over influenza A in several influenza seasons and some of the interseason periods as well as two lineages changed rotationally or cocirculated in different periods. 20 TA B L E 1 Seropositive a cases on swine farms for swine strains of influenza B virus at prefecture level in Taiwan high weekly percentage of IBVs appeared in 6 periods: 2008/31-42, 2010/08-38, 2011/04-15, 2011/29-34, 2011/39-2012/13, and 2014/03-18 (y/wk Phylogenetic analyses revealed also that our sIBVs might be originated from B/ Taiwan In NA protein, drug-resistant amino acids have been reported, such as the oseltamivir-resistant mutations E105K, G109E, E119A,   D197E, D198N, I221V, I222T, H274Y, R292K, and N294S; zanamivirresistant mutations G109E, E119A, R152K, D197E, and I222T; and peramivir-resistant mutations E105K, D197E, and H273Y(H274Y). 16 None of them was found on all three sIBVs. It suggests that our sIBVs are sensitive to NA inhibitors. However, 3 of major signature amino acid substitutions (S295R, N340D, and E358K), which shared by V1A-2 subclade viruses of the 2012-14 Malaysian IBVs, based on the recent WHO genetic groupings 27 were observed from all sIBVs (data provided in: and 2017. Meanwhile, no more than 6 positive farms found in positive prefectures. In positive townships/districts, no more than 2 farms are seropositive. In all positive farms, most of them have only one positive swine serum and no more than 2 in the remaining farms. In the period of study, therefore, positive cases for isolation of IBVs and detection of IBV-specific antibodies were not confined to localized area and found infrequently. The trend for temporal increment and spatial expansion of seropositive cases year by year was not observed. The findings mean that transmission of IBVs remained probably sporadic and accidental at the human-swine interface as none of available evidences that swine workers were positive for IBV. Public health threats, however, might be considered because swine populations have served as a reservoir of IAV genes that have circulated in humans. 32 Human IBVs would be kept in swine populations and provided as genetic sources for reassortment of novel influenza viral variants. Even only sporadic cases of IBV were found during our study period in Taiwanese swine populations, reassortment events of swine IBV with other kind of influenza viruses (IAV and any unpredictable types) in pigs might be happened to create newly emerging zoonotic influenza in the future.
In conclusion, public health concerns might arise from the finding that IBVs could be transmitted naturally from humans to pigs, even cross-species infections are sporadic and accidental currently in Taiwan. It implies a cycle of human-swine-human transmission could be or might have been initiated as some of IAVs did in human and swine populations. 33 Therefore, IBV infections in Taiwanese pigs should deserve much more attention, better surveillance, and further detailed investigations.

ACK N OWLED G EM ENTS
We like to greatly thank Meng-Jung Lee and Yi-Chaun Chang, research assistants of Agricultural Technology Research Institute, for their help in processing swine specimens as well as conducting virological and serological experiments. This study was supported by several project grants from Bureau of Animal and Plant Inspection and Quarantine, Council of Agriculture, Taiwan in the period of