Domestic Pigs Are Susceptible to Infection with Influenza B Viruses

ABSTRACT

Influenza B virus (IBV) causes seasonal epidemics in humans. Although IBV has been isolated from seals, humans are considered the primary host and reservoir of this important pathogen. It is unclear whether other animal species can support the replication of IBV and serve as a reservoir. Swine are naturally infected with both influenza A and C viruses. To determine the susceptibility of pigs to IBV infection, we conducted a serological survey for U.S. Midwest domestic swine herds from 2010 to 2012. Results of this study showed that antibodies to IBVs were detected in 38.5% (20/52) of sampled farms, and 7.3% (41/560) of tested swine serum samples were positive for IBV antibodies. Furthermore, swine herds infected with porcine reproductive and respiratory syndrome virus (PRRSV) showed a higher prevalence of IBV antibodies in our 2014 survey. In addition, IBV was detected in 3 nasal swabs collected from PRRSV-seropositive pigs by real-time RT-PCR and sequencing. Finally, an experimental infection in pigs, via intranasal and intratracheal routes, was performed using one representative virus from each of the two genetically and antigenically distinct lineages of IBVs: B/Brisbane/60/2008 (Victoria lineage) and B/Yamagata/16/1988 (Yamagata lineage). Pigs developed influenza-like symptoms and lung lesions, and they seroconverted after virus inoculation. Pigs infected with B/Brisbane/60/2008 virus successfully transmitted the virus to sentinel animals. Taken together, our data demonstrate that pigs are susceptible to IBV infection; therefore, they warrant further surveillance and investigation of swine as a potential host for human IBV.

IMPORTANCE IBV is an important human pathogen, but its ability to infect other species, for example, pigs, is not well understood. We showed serological evidence that antibodies to two genetically and antigenically distinct lineages of IBVs were present among domestic pigs, especially in swine herds previously infected with PRRSV, an immunosuppressive virus. IBV was detected in 3 nasal swabs from PRRSV-seropositive pigs by real-time reverse transcription-PCR and sequencing. Moreover, both lineages of IBV were able to infect pigs under experimental conditions, with transmissibility of influenza B/Victoria lineage virus among pigs being observed. Our results demonstrate that pigs are susceptible to IBV infections, indicating that IBV is a swine pathogen, and swine may serve as a natural reservoir of IBVs. In addition, pigs may serve as a model to study the mechanisms of transmission and pathogenesis of IBVs.

INTRODUCTION

Influenza viruses are classified as genera A, B, and C, in accordance with the antigenic differences in their nucleoproteins (NP) and matrix 1 (M1) proteins (28). Influenza A (IAV) and B (IBV) viruses can result in severe upper respiratory disease in humans, while influenza C viruses (ICV) cause relatively mild disease (9, 23). Among influenza viruses, IAV and IBV are very similar in terms of genome structure and organization. IBV, along with influenza A(H3N2) and A(H1N1) viruses [including A(H1N1)pdm09 virus], cause seasonal influenza epidemics annually (9, 23). In the United States alone during 1976 to 2007, approximately 3,000 to 49,000 deaths each year have been attributed to these epidemics (42). Some reports indicate that in older children and healthy adults, influenza A(H3N2) virus is responsible for the most severe cases, followed by IBV, while influenza A(H1N1) virus infections tend to manifest as the mildest cases of illness (1, 5, 23, 25). In some seasons, however, IBV may be the predominate strain responsible for influenza activities. This was best exemplified by the 1979-1980 season, in which IBV was the predominant strain circulating in the United States; therefore, it was responsible for influenza outbreaks and excess pneumonia and influenza deaths nationwide (39). Furthermore, IBV has been reported to be associated with central nervous system complications, such as Reye's syndrome and encephalitis in children (1).

IBVs continue to circulate worldwide alongside IAVs. Actively circulating IBVs are divided into two genetically and antigenically distinct lineages, represented by B/Yamagata/16/1988 and B/Victoria/2/1987 viruses (14, 21, 22, 37, 38). Antigenic differences between the lineages can be readily revealed by the hemagglutination inhibition (HI) assay (35).

The ecologic and natural reservoirs of IAV have been well characterized (9). In contrast, little is known about animal reservoirs of IBVs. It was long assumed that the natural host and reservoir of IBVs were restricted exclusively to humans (9, 27, 28), although studies published sporadically from 1960 to 1980 indicated that IBVs were isolated from pheasants, horses, and dogs (8, 15, 33). In addition, some recent studies have revealed that seals in Europe, South America, and North America were infected by IBVs (3, 4, 26, 27). It is still not clear whether the virus is transmissible among seals or if human activity played a role in transmission to seals. Despite this concern, the previous studies provide compelling evidence that the seal is another mammalian species susceptible to IBV infection. Furthermore, antibodies specific to IBVs were identified from guinea pigs raised as livestock in Ecuador (16).

Swine are naturally infected by both IAV and ICV and are hosts for many human and avian IAVs (28). Recent studies also pointed out that this agricultural animal species is a host of newly identified influenza D virus (IDV) (11, 12). Pigs are considered a mixing vessel for generation of reassortant IAVs. Some of the reassortant viruses were capable of triggering influenza pandemics, and the recent 2009 influenza pandemic is one example (19). Reports from eastern Europe in the 1960s showed that IBV-specific antibodies were detected in pigs under natural and experimental conditions (40, 41). Since then, there has been significant evolution of IBVs in humans and numerous changes in infectious disease ecology in pigs in North America. These changes may have affected the susceptibility of pigs to IBV infection. In this study, we conducted a swine serological survey and an experimental challenge study in pigs using two representative IBVs: B/Brisbane/60/2008 and B/Yamagata/16/1988.

MATERIALS AND METHODS

Cells and viruses.

Madin-Darby canine kidney (MDCK) cells were maintained in Dulbecco's modified Eagle medium (DMEM) containing 10% fetal bovine serum at 37°C with 5% CO₂. IBVs B/Florida/04/2006 (Yamagata lineage), B/Wisconsin/10/2010 (Yamagata lineage), B/Brisbane/60/2008 (Victoria lineage), and B/Georgia/07/2010 (Victoria lineage) were provided by the Centers for Disease Control and Prevention (CDC). IAV A/swine/Iowa/73 (H1N1) was obtained from the National Veterinary Services Laboratories of the United States Department of Agriculture. IBVs B/Brisbane/60/2008, B/Yamagata/16/1988, B/Wisconsin/10/2010, and B/Georgia/07/2010 were propagated in 9-day-old specific-pathogen-free chicken embryonated eggs. IBV B/Florida/04/2006 and IAV A/swine/Iowa/73 (H1N1) were amplified in MDCK cells using DMEM (Gibco, Grand Island, NY) supplemented with 2 μg/ml of tosyl phenylalanyl chloromethyl ketone (TPCK)-treated trypsin (Sigma-Aldrich, St. Louis, MO), 0.3% bovine serum albumin (BSA; Sigma-Aldrich, St. Louis, MO), and antibiotics.

Sera and antigens.

Five hundred sixty swine serum samples from growing pigs were randomly selected from the serum pool of the Animal Disease Research and Diagnostic Laboratory, South Dakota State University (SDSU), and Newport Laboratories; the serum pool originated from 52 swine farms throughout the Midwest in the United States (see Table S1 in the supplemental material). Information that serum samples derived from these pigs were seropositive or seronegative for PRRSV was not available. Seventy-nine samples were collected from 5 swine farms in 2010 (January to April) and 2011 (April), and 481 samples were collected from 47 swine farms from January to July 2012. Ferret antisera to IBVs B/Brisbane/60/2008, B/Florida/04/2006, and B/Georgia/07/2010, a more recent B/Victoria lineage reference virus, as well as IAVs A/California/07/2009 (pdm09H1N1) and A/Minnesota/11/2010 variant (H3N2v), were kindly provided by the Centers for Disease Control and Prevention (CDC). Swine A(H1N1)-positive sera and sera from gnotobiotic piglets (as a negative control) were kindly provided by Mojun Zhao, SDSU. Ether-treated IBV antigens from B/Wisconsin/1/2010 (Yamagata lineage) and B/Brisbane/60/2008 (Victoria lineage), as well as whole IAV antigens derived from A/California/07/2009 (pdm09H1N1) and A/Minnesota/11/2010 variant (H3N2v), were provided by the CDC. Ether-treated IBV antigen was selected for our HI antibody assay due to its increased ability in the detection of virus-specific antibody to IBVs (24, 30).

To determine whether immunosuppressive infection by PRRSV affected the susceptibility of pigs to IBV infection, 115 serum samples were collected from nursery and finisher pigs in 4 PRRSV-positive swine herds and in 4 PRRSV-negative swine herds located in South Dakota, Texas, Iowa, Minnesota, and Kansas from August to October 2014. These serum samples were tested and confirmed to be PRRSV seropositive or seronegative by the commercial IDEXX PRRS X3 enzyme-linked immunosorbent assay (ELISA) kit (a sample-to-positive [S/P] ratio of ≥0.4 was classified as positive for PRRSV antibodies) or by the Tetracore PRRSV multiplex real-time reverse transcription-PCR (RT-PCR) assay (a threshold cycle [C_T] value of ≤35 was classified as positive for PRRSV). Thirty nasal swab samples also were collected from one PRRSV-positive swine herd and one PRRSV-negative swine herd, herd D (see Table 4). These serum samples were tested by the HI assay to detect virus-specific antibody to IBVs, and nasal swabs were examined by real-time RT-PCR assays targeting the NS gene of IBV and the M gene of IAV (36). Samples positive for IBV were used for virus isolation in chicken embryonated eggs and MDCK cells. A portion of the IBV NS gene (approximately 100 to 250 nucleotides [nt]) was amplified and sequenced from these samples (primers are available upon request).

TABLE 4.

Summary of the serological survey in PRRSV-positive and -negative domestic swine farms by HI assay

Farm and PRRSV status	No. of sera positivea for B/Yamagata lineageb	Positive (%)	No. of sera positive for B/Victoria lineagec	Positive (%)
Positive
A	4/10 (48)d	40	0/10	0
B	0/9	0	0/9	0
C	0/6	0	0/6	0
D	29/90 (51)	32.2	48/90 (69)	53.3
Negative
A	0/7	0	0/7	0
B	0/9	0	0/9	0
C	0/9	0	0/9	0
D	0/90	0	0/90	0

^aData are the number of positive samples out of the total number tested.

^bB/Yamagata/16/1988 was used.

^cB/Brisbane/60/2008 was used.

^dNumbers in parentheses indicate geometric mean HI titers for samples with a value of ≥40.

HI assay.

The HI assay was conducted by following the WHO standard procedure (43). Briefly, 100 μl serum was mixed with 300 μl of receptor-destroying enzyme (RDE) (Denka Seiken Co., Ltd., Japan) and incubated overnight in a 37°C water bath. The mixture was heated in a 56°C water bath for 30 min to inactivate the remaining activity of RDE. After cooling to room temperature, 100 μl physiological saline (0.85% NaCl) was added to serum so that the final dilution of serum stock was 1:5. To remove nonspecific agglutinins, 25 μl of 5% packed chicken red blood cells (RBCs) (Lampire Biological Laboratories, Pipersville, PA) was added to 500 μl RDE-treated and heat-inactivated sera. After an additional incubation at 4°C for 1 h, the mixture was centrifuged at 1,200 rpm for 10 min; the supernatant was collected, aliquoted, and frozen at −20°C or directly used for the following assays.

Reference antigens were titrated by 2-fold serial dilutions with 50 μl physiological saline and then incubated with 50 μl of 0.5% RBCs. Standardized antigens of 4 hemagglutination units per 25 μl were prepared based on the titration. Pig serum samples and reference sera were diluted serially 2-fold from 1:5 with 25 μl physiological saline and incubated with 25 μl standardized antigens at room temperature for 15 min, followed by the addition of 50 μl of 0.5% RBCs before incubating for an additional 30 min. Physiological saline and serum controls, as well as antigen back titration, also were performed for validation. A titer of 40 was used as a threshold, i.e., a sample with a titer of less than 40 was judged as negative, and those with a titer equal to or higher than 40 were viewed as positive.

MN assay.

The microneutralization (MN) assay was performed on MDCK cells as described in the WHO standard manual (43). Prior to neutralization, a 50% tissue culture infectious dose (TCID₅₀) of reference viruses was determined on MDCK cells, and then the viruses were diluted to 200 TCID₅₀/100 μl with DMEM containing 2 μg/ml of TPCK-trypsin, 0.3% BSA, and antibiotics. RDE-treated and heat-inactivated pig serum samples and reference ferret antisera were 2-fold serially diluted with DMEM containing 2 μg/ml of TPCK-trypsin, 0.3% BSA, and antibiotics and mixed with 100 μl TCID₅₀ of the reference viruses to react for 1 h at 37°C. The mixture was transferred into 96-well plates that contained 100 μl MDCK cell suspension (1.5 × 10⁴ cells/well). The plate was incubated overnight at 37°C with 5% CO₂. Negative and positive serum controls, virus controls, cell controls, and virus back titration also were performed to validate procedures and results. The cells were fixed with 100 μl of 80% cold acetone at room temperature for 10 min and used for ELISA. Briefly, cells were incubated with the primary antibody, anti-influenza B virus nucleoprotein antibody (Abcam, Cambridge, MA), and then with the secondary antibody, goat anti-mouse IgG-horseradish peroxidase (HRP) (Sigma-Aldrich, St. Louis, MO), before being washed. The substrate ortho-phenylenediamine (OPD) and stop solution then were added. The absorbance (optical density at 490 nm [OD₄₉₀]) was read and recorded for further analysis.

Infection and transmission studies of IBVs in pigs.

Four-week-old pigs that were negative for antibodies against IAVs, IBVs, and PRRSV were used in this study. The pig experiment was conducted in a biosafety level 2 (BSL-2) facility and approved by the Institutional Animal Care and Use Committee at Kansas State University. A total of 2 pig experiments were performed to investigate infection, pathogenicity, and transmission of IBVs.

For the first experiment, 18 4-week-old pigs were randomly allocated into 2 groups (12 pigs in the infected group and 6 pigs in the control group). Eight pigs (pigs 62 to 69) in the infected group were inoculated intratracheally (2 ml) and intranasally (1 ml) with B/Brisbane/60/2008 virus at a titer of 10⁶ TCID₅₀/ml that was derived from MDCK cells. Four contact pigs (pigs 70 to 73) were commingled with the challenged pigs at 2 days postinfection (dpi) to investigate viral transmission. Six control pigs were intratracheally (2 ml) and intranasally (1 ml) inoculated with virus-free minimum essential medium (MEM). Clinical signs and body temperatures were recorded daily from 0 to 14 dpi for inoculated pigs and from 0 to 7 dpi for control pigs. Three control and 3 infected pigs were necropsied at 5 dpi, and 3 control and 5 infected pigs were necropsied at 14 dpi. Four contact pigs were euthanized at 12 days postcontact (dpc).

For the second experiment, nine 4-week-old pigs were randomly allocated into 2 groups (7 pigs in the infected group and 2 pigs in the control group). Five pigs in the infected group were inoculated with B/Yamagata/16/1988 virus derived from chicken embryo eggs, and two control pigs were inoculated with virus-free MEM using the same route and dose as that described above for the first experiment. Two contact pigs were commingled with infected pigs at 2 dpi. Clinical signs and body temperatures were recorded daily as described for the first experiment. Two control and 2 infected pigs were necropsied at 5 dpi, and 3 infected pigs and 2 contact animals were necropsied at 14 dpi and 12 dpc, respectively.

For both experiments, blood samples were collected from each pig at 0 dpi (or dpc) and at the day of necropsy. Nasal swab samples were collected from infected pigs at 0, 2, 4, 5, 7, and 9 dpi and from contact pigs at 0, 2, 3, 5, and 7 dpc. Nasal swab samples were suspended in 2 ml MEM and stored at −80°C. During necropsy the lungs were removed in toto from pigs. The percentage of gross lesions on each lung lobe was scored by a single experienced veterinarian (31). Bronchoalveolar lavage fluid (BALF) was obtained from each pig lung by flushing with 50 ml MEM. The right cardiac lung lobe was collected and fixed in 10% buffered formalin and then stained with hematoxylin and eosin for histopathologic examination. Lung sections were examined by a veterinary pathologist in a blinded fashion and given a score of 0 to 3 to reflect the severity of bronchial epithelial injury as described previously (17). Virus titers of BALF and nasal swabs were determined in a 96-well plate on MDCK cells, as previously described (31), using a temperature of 33°C. The detection limit of virus titration is approximately 1 log₁₀ TCID₅₀.

RESULTS

No antigenic cross-reactivity between IAV and IBV.

We first investigated the potential cross-reactivity between various IAVs and IBVs by the HI assay. Ferret reference antisera to IAVs and IBVs and a swine positive serum to H1N1 A/swine/Iowa/1973 (swH1N1), as well as a swine negative-control serum, were tested. Two IBV ether-treated antigens, B/Wisconsin/1/2010 (Yamagata lineage) and B/Brisbane/60/2008 (Victoria lineage), and three influenza A antigens (two H1 and one H3 subtype) were used as depicted in Table 1. Although results showed no detectable cross-reactivity between antigens and sera of IAVs and IBVs, a low level of cross-reactivity was observed between the two IBVs tested with ferret antisera (Table 1). These results confirmed the specificity of our HI assay in the detection of anti-IBV antibodies, demonstrating that it could be used for screening of pig serum samples for antibodies to IBVs.

TABLE 1.

Cross-reactivity between influenza A and B viruses tested by HI assay

Antigen	HI titer toa:
	SwH1N1	pdmH1N1	H3N2v	B/Yamagata	B/Victoria	Negative control
A/swine/IA/73 (H1N1)	160	ND	ND	<10	<10	<10
A/California/07/2009 (pdmH1N1)	ND	1,280	<10	<10	<10	<10
A/Minnesota/11/2010 (H3N2v)	ND	<10	640	<10	<10	<10
B/Yamagata lineage (ether treated)
B/Wisconsin/1/2010	<10	<10	<10	1,280	40	<10
B/Brisbane/60/2008	<10	<10	<10	40	5,120	<10

^aND, not determined.

Detection of IBV-specific antibodies in swine serum samples.

A total of 560 serum samples collected from 52 pig farms throughout the Midwest were tested in the HI assays; ether-treated IBVs were used as antigens in the assay to increase sensitivity for the detection of influenza B antibodies. Sera with HI antibody titers equal to or greater than 40 were scored as positive. The HI assay showed that serum specimens from 20 out of 52 (38.5%) swine farms were positive for antibodies to IBVs (see Table S1 in the supplemental material), and a total of 41 serum samples (41/560; 7.3%) collected from the 20 positive swine farms tested positive (HI titers for ≥40) for anti-IBV antibodies. Among the 41 positive samples, 2 samples (2/560, 0.36%) from 2 farms (2/52, 3.9%) tested positive for antibodies to both B/Wisconsin/10/2010 and B/Brisbane/60/2008 viruses. In addition, 20 serum samples (25/560; 4.5%) from 12 farms (12/52; 23.1%) tested positive for anti-B/Wisconsin/10/2010 antibodies, and 23 samples (23/560; 4.1%) from 11 farms (11/52; 21.2%) tested positive for anti-B/Brisbane/60/2008 antigen (Table 2; also see Table S1).

TABLE 2.

Summary of the serological survey from 20 virus-positive domestic swine farms by HI assay

Yr of sampling and farm	No. of sera positivea for B/Yamagata lineageb	Positive (%)	No. of sera positive for B/Victoria lineagec	Positive (%)
2010-2011
11-A	0/27	0	5/27 (557)	18.5
11-B	0/10	0	2/10 (905)	20.0
11-C	0/7	0	1/7 (640)	16.7
11-D	0/30	0	1/30 (1,280)	3.3
2012
12-A	0/12	0	1/12 (40)	8.3
12-B	4/12 (67)d	33.3	1/12 (40)	8.3
12-E	0/16	0	5/16 (243)	31.3
12-G	1/11 (40)	9.1	0/11	0
12-L	1/13 (40)	7.7	0/13	0
12-T	2/12 (40)	16.7	0/12	0
12-V	1/12 (80)	8.3	0/12	0
12-W	2/10 (40)	20.0	0/10	0
S12-A	2/10 (40)	20.0	1/10 (160)	10.0
S12-B	5/10 (61)	50.0	0/10	0
S12-D	1/10 (160)	10.0	0/10	0
S12-E	1/10 (40)	10.0	0/10	0
S12-G	0/12	0	4/10 (48)	40.0
S12-J	0/5	0	1/5 (320)	20.0
S12-K	4/12 (40)	33.3	0/12	0
S12-S	1/8 (1,280)	12.5	1/8 (160)	12.5

^aData are positive sample numbers out of the total number tested.

^bYamagata lineage B/Wisconsin/1/2010 was used.

^cVictoria lineage B/Brisbane/60/2008 was used.

^dNumbers in parentheses indicate geometric mean HI titers for samples with a value of ≥40.

Our data also identified that in several positive farms, high proportions of animal sera tested positive for anti-IBV antibodies (Table 2). For example, 50% of pig sera tested from farm S12-B were positive for antibodies to B/Wisconsin/10/2010, while 31.3% of pig sera tested from another farm (12-E) were positive for antibodies to B/Brisbane/60/2008 (Table 2). Samples collected in the 2010-2011 influenza season were positive only to the B/Victoria lineage test antigen (11.4%; 9/79), whereas in 2012 antibodies to both influenza B lineages (0.4%; 2/481) were found even concurrently in animals at individual farms (Table 2). Also in 2012, the percentage of samples positive for the B/Yamagata lineage test antigen (5.2%; 25/481) was higher than the positive percentage for the B/Victoria lineage test antigen (2.9%; 14/481) (Table 2; also see Table S1 in the supplemental material). It has been noticed that the overall HI antibody titers to the B/Yamagata lineage antigen were much lower than those to the B/Victoria lineage antigen. For instance, there were 17 samples with an HI titer of 160 to 1,280 for the B/Victoria lineage antigens, while only two samples had a titer of 160 for antigens to the B/Yamagata lineage (Table 2; also see Table S1).

To confirm the results of the HI assays, the MN assays were performed independently by two laboratories (CDC and SDSU) for 20 serum samples, including 8 positive and 12 negative samples pretested by HI assay; these serum samples were collected from 5 farms during the 2010-2011 influenza season. Two B/Victoria lineage viruses (B/Brisbane/60/2008 and B/Georgia/07/2010) and one B/Yamagata lineage virus (B/Florida/04/2006) and their corresponding sera were used as controls for analysis. As summarized in Table 3, MN assay results from the two laboratories were highly comparable and further confirmed the findings from the HI assays. Some low degree of cross-reactivity between two lineages of IBVs was observed in the MN assay, similar to that observed in the HI assay; however, a lineage-specific response was clearly demonstrated in both HI and MN assays. In addition to the lineage specificity, the MN assay also was highly sensitive, since the neutralization titers of positive samples were higher than those obtained with the HI assay (Tables 2 and 3). In summary, results of the MN assays confirmed the results of the HI assay. Data obtained from the MN assays further demonstrated that IBV-specific antibodies were detected in swine at U.S. pig farms.

TABLE 3.

Comparison of neutralization and HI antibody titers for a subset of serum samples

Serum sample	Neutralization titer				HI titer
	B/Florida/04/06 (B/Yamagata)	B/Brisbane/60/08 (B/Victoria)	B/Georgia/07/10 (B/Victoria)	B/Georgia/07/10 (B/Victoria)	B/Florida/04/06 (B/Yamagata)	B/Brisbane/60/08 (B/Victoria)
Gnotobiotic piglet serum	≤5	≤5	≤5	≤5	≤5	≤5
S1 (11-A)	≤5	≥1,280	640	320	≤5	320
S2 (11-A)	≤5	≤5	≤5	≤5	≤5	≤5
S3 (11-A)	≤5	≤5	≤5	≤5	≤5	≤5
S4 (11-A)	≤5	≤5	≤5	≤5	≤5	≤5
S5 (11-A)	≤5	≥1,280	≥1,280	320	≤5	640
S6 (11-A)	≤5	≥1,280	≥1,280	320	≤5	640
S7 (11-A)	≤5	≤5	≤5	≤5	≤5	≤10
S8 (11-A)	≤5	≥1,280	≥1,280	320	≤5	640
S9 (11-A)	≤5	≤5	≤5	≤5	≤5	≤5
S10 (11-A)	≤5	≤5	≤5	≤5	≤5	≤5
S11(11-A)	≤5	≤5	≤5	≤5	≤5	≤5
S12 (11-A)	≤5	≤5	≤5	≤5	≤5	≤5
S13 (11-B)	≤5	≤5	≤5	≤5	≤5	≤5
S14 (11-B)	≤5	≥1,280	640	320	≤5	1,280
S15 (11-B)	≤5	≥1,280	≥1,280	320	≤5	640
S16 (11-C)	≤5	≤5	≤5	≤5	≤5	≤5
S17 (11-C)	≤5	≤5	≤5	≤5	≤5	≤5
S18 (11-C)	≤5	≤5	≤5	≤5	≤5	≤5
S19 (11-C)	≤5	≥1,280	≥1,280	320	≤5	640
S20 (11-D)	≤5	≥1,280	≥1,280	320	≤5	1,280

Prevalence of IBV antibodies in PRRSV-positive and -negative herds.

Since the PRRSV infection has been considered to result in host immunosuppression (10), we investigated and compared IBV antibody prevalence in PRRSV-positive herds to that observed in PRRSV-negative herds. Analysis of 115 serum samples from 4 PRRSV-negative swine herds by the HI assay revealed they all were seronegative for two IBV lineages. In contrast, we found the presence of IBV antibodies in serum samples among 2 of 4 PRRSV-positive herds (Table 4). Of two herds, one was seropositive only for the Yamagata lineage virus, with a 40% (4/10) prevalence rate, while the other herd had a total of 54 serum samples (54/90) that tested positive against Yamagata and/or Victoria lineages, with a total prevalence rate of 60% (Table 4). Specifically, among the 54 positive samples, 23 samples (23/90; 25.6%) tested positive for antibodies to both lineages; in addition, 48 serum samples (48/90; 53.3%) tested positive for the Victoria lineage, and 29 samples (29/90; 32.2%) tested positive for the Yamagata lineage (Table 4).

We next sought to determine the presence of IBV using a real-time RT-PCR assay targeting its NS segment in nasal swabs derived from PRRSV/IBV antibody-positive and -negative pigs. An IAV real-time RT-PCR assay targeting the M segment also was employed to screen these nasal swab samples. For this purpose, 30 nasal swabs were collected from PRRSV-positive and -negative pigs, respectively, for the real-time RT-PCR assays. The results showed that all 30 nasal swab samples from PRRSV/IBV antibody-negative pigs were negative for IBVs but were positive for IAVs (C_T of <35). Twenty-one out of 30 nasal swabs collected from PRRSV/IBV antibody-positive pigs were positive for IAV; however, 3 of these 21 samples also were positive for IBV (C_T of approximately 35). Attempts to isolate IBV from these real-time RT-PCR-positive nasal swabs were not productive. Nevertheless, the partial NS sequence (307 nt, 308 nt, and 489 nt) was obtained from these 3 samples in which 2 samples were seropositive for the Victoria lineage virus and 1 sample was seropositive for the Yamagata lineage virus. Further analysis revealed that 3 partial NS sequences showed 99% to 100% identity to those of B/New York/1019/1999 (GenBank accession no. CY174077) and B/Yamanashi/166/1998 (GenBank accession no. CY019535) viruses, respectively, at the nucleotide levels.

Both B/Brisbane/60/2008 (Victoria lineage) and B/Yamagata/16/1988 viruses were able to infect and replicate in pigs.

To investigate whether IBVs can infect and transmit among pigs, two groups of pigs were intratracheally and intranasally inoculated with either B/Brisbane/60/2008 or B/Yamagata/16/1988 virus at a titer of 10⁶ TCID₅₀/ml, and 2 to 4 sentinel animals were cohoused with the respective infected pigs (Tables 5 and 6). At 2 dpi, 7 out of 8 pigs (87.5%; pig 66 was the exception) infected with the B/Brisbane/60/2008 virus developed a fever (temperature of over 40°C) lasting 2 to 3 days, while 3 (pigs 1, 3, and 4) out of 5 pigs (60%) infected with the B/Yamagata/16/1988 virus displayed a fever which lasted for 1 day. None of the control pigs exhibited fever or any other clinical signs upon examination. Virus was detected in 2 out of 3 bronchoalveolar lavage fluid (BALF) samples from B/Brisbane/60/2008 virus-infected animals at 5 dpi, with a titer of 1.67 to 2.50 log₁₀ TCID₅₀/ml (Tables 5 and 6). No virus was detected in BALF samples collected from the 2 pigs infected with B/Yamagata/16/1988 virus or the samples collected from all control animals at 5 dpi (Tables 5 and 6). The detection of viruses in nasal swab samples revealed that specimens from 6 (except for pigs 62 and 69) out of 8 pigs infected with B/Brisbane/60/2008 virus were positive at 2 and 4 dpi (Fig. 1). The number of pigs shedding viruses significantly decreased at 5 dpi, with only 1 pig (68) having detectable virus load in nasal swabs. The virus titers among these samples varied from 2.2 to 4.5 log₁₀ TCID₅₀/ml (Fig. 1). In contrast, no virus was detected in nasal swab samples from all 5 pigs infected with B/Yamagata/16/1988 virus at 2 and 4 dpi. No IBV was detected among control animals. At 5 dpi, pigs were euthanized and necropsy was performed: 2 of 3 pigs (66%) inoculated with the B/Brisbane/60/2008 virus and 2 of 2 pigs (100%) infected with B/Yamagata/16/1988 virus exhibited minor typical influenza-like lung lesions (Tables 5 and 6), which were scored between 0.5 and 1.0 (score range, 0 to 3), and no macroscopic lung lesions were found in any of the control animals. No significant difference in lung lesions was observed between the two IBV infection groups. The lungs of all infected pigs showed variable degrees of microscopic lung lesions, which tended to range from minimal to mild peribronchiolitis and alveolitis (Fig. 2). Multifocal areas of alveolitis were identified and were characterized by the presence of degenerate and intact neutrophils in the alveolar lumen, as well as thickening of alveolar septa by inflammatory cells, in both IBV-infected pig lungs (Fig. 2). No bronchiolar epithelial necrosis was observed in any of the IBV-infected pigs, which is in contrast to lesions seen in the lungs of pigs infected with swine IAVs (17, 18). Seroconversion, as measured in the HI assay, was not observed in infected animals that were necropsied at 5 dpi but had occurred at 14 dpi in pigs infected with either B/Brisbane/60/2008 or B/Yamagata/16/1988 virus (Tables 5 and 6). When combined, the results of these assays indicate that IBVs have the ability to infect and replicate in pigs. This is in agreement with our serological survey data, which demonstrate the presence of IBV-specific antibodies in pigs.

TABLE 5.

Lung lesions, virus lung replication (BALF virus titer), and seroconversion (HI titers) of infected and contact pigs for the B/Brisbane/60/2008 (Victoria lineage) infection group

Parameter	Value for:
	Infected pigs at:								Contact pigs (12 dpc)
	5 dpi			14 dpi
	62	63	64	65	66	67	68	69	70	71	72	73
Lung lesion	No	Yes	Yes	No	No	Yes	No	No	No	No	No	No
BALF virus titer	<1	2.50	1.67	NDa	ND	ND	ND	ND	ND	ND	ND	ND
HI titer	<10	<10	<10	40	80	160	40	40	40	<10	<10	<10

^aND, not determined.

TABLE 6.

Lung lesions, virus lung replication (BALF virus titer), and seroconversion (HI titers) of infected and contact pigs for the B/Yamagata/16/1988 (Yamagata lineage) infection group

Parameter	Value for:
	Infected pigs at:					Contact pigs (12 dpc)
	5 dpi		14 dpi
	1	2	3	4	5	6	7
Lung lesion	Yes	Yes	No	No	No	No	No
BALF virus titer	<1	<1	NDa	ND	ND	ND	ND
HI titer	<10	<10	80	40	80	<10	<10

^aND, not determined.

FIG 1.

Transmission of B/Brisbane/60/2008 virus from inoculated pigs to contact animals. Eight pigs were inoculated intratracheally and intranasally with influenza B/Brisbane/60/2008 virus at a dose of 3 × 10⁶ TCID₅₀/ml. Four contact pigs were comingled with the challenged pigs at 2 days postinfection (dpi) to investigate viral transmission. Six out of 8 inoculated pigs shed virus by nasal cavity, and 2 inoculated pigs did not shed virus. Nasal shedding was detectable in 1 out of 4 contact pigs. Solid lines represent inoculated pigs, while dotted lines are used to represent contact animals.

FIG 2.

Microscopic lung sections from pigs infected with IBVs at 5 days postinfection (dpi). (A) Control pigs. The bronchioles are lined by normal cuboidal epithelium (arrow), and the alveoli are clear (star). (B and C) Infected pigs. Alveolitis is observed in pigs infected with either influenza B/Brisbane/60/2008 (B) or B/Yamagata/16/1988 (C) viruses. Multifocal areas of alveolitis are seen and are characterized by the presence of degenerate and intact neutrophils within the alveolar lumen (star), and there is a thickening of alveolar septa caused by inflammation. No bronchiolar epithelial necrosis is observed in any of the infected pigs. Bars, 50 μm.

B/Brisbane/60/2008 (Victoria lineage) displayed limited transmission in pigs.

Naive sentinel pigs were commingled with infected pigs to investigate IBV transmissibility through direct contact. In the contact groups, animals infected with either B/Brisbane/60/2008 or B/Yamagata/16/1988 virus did not show obvious respiratory clinical signs during the study. In the B/Brisbane/60/2008-infected group, 3 (pigs 70, 71, and 73) out of 4 pigs (75%) developed fever at 4 dpc and 1 (pig 70) of the 3 pigs displayed fever lasting for 3 days. Virus was detected only in nasal swabs collected from this pig (70) at 5 and 7 dpc with titers of 4.5 and 3.2 log₁₀ TCID₅₀/ml, respectively (Fig. 1); no virus was detected in nasal swabs collected from the other 3 contact pigs on any of the days tested. Pig 70 seroconverted as detected at 12 dpc with an HI titer of 40, and the remaining 3 contact pigs did not show seroconversion (Tables 5 and 6) by 12 dpc. These results indicated that the B/Brisbane/60/2008 virus was able to transmit from infected pigs to contact animals. In the B/Yamagata/16/1988-infected group, the 2 contact pigs did not develop fever, and no virus was detected in nasal swabs collected on any day from either sentinel. The HI titers of both contact animals were lower than 10, indicating that seroconversion did not occur in either animal. These data suggest that the B/Yamagata/16/1988 virus was not transmissible among experimental pigs, despite its ability to infect pigs.

DISCUSSION

Unlike IAVs that can infect a wide range of species, IBV infections are almost exclusively restricted to humans, despite the isolation of IBVs and presence of IBV-specific antibodies in seals (3, 4, 26, 27). Recent publications have documented that antibodies to IBV are present in guinea pigs (16), and its efficient replication and transmission in this species (29) raised an interesting question of whether other animal species can support the replication of IBV. In this study, we present serological, molecular, and experimental infection evidence that pigs are susceptible to infection with the two genetically and antigenically distinct lineages of IBVs. Infected pigs developed influenza-like lesions in the respiratory tract similar to those demonstrated in humans and guinea pigs. Therefore, IBV may be an important swine pathogen for the swine industry, although it is not routinely included for testing of swine samples by veterinary diagnostic laboratories in the United States and other countries.

Our experiments generated a number of interesting observations. Significantly, the serological evidence of IBV infection in pigs reported in this study corresponded to the circulation of IBV in humans. For example, pig serum samples collected in the 2010-2011 influenza season harbored only antibodies to B/Victoria lineage viruses. In humans, the predominant IBVs at that time also were from the B/Victoria lineage (6). Furthermore, antibodies against both B/Victoria and B/Yamagata lineages were detected in swine serum samples collected in 2012 and 2014 in which both lineages of IBVs were circulating in humans (7, 32). Whether reverse zoonotic transmission of IBV could exist between humans and swine remains unknown and needs to be investigated in future research. Importantly, vaccination of swine workers may be an efficient means to block possible IBV reverse zoonotic and zoonotic transmission to protect animal and human health. Whether swine can serve as a natural host for IBVs that can cause epidemics in humans needs to be investigated in future research.

Although both B/Victoria and B/Yamagata lineage-specific antibodies were detected in some pig samples, the serological survey data showed that the HI antibody titers against the B/Victoria lineage viruses were much higher than those against the B/Yamagata lineage virus. Interestingly, in our animal infection experiment, B/Victoria lineage virus replicated and transmitted more efficiently than B/Yamagata lineage virus in pigs. This is also in agreement with recent findings in the ferret model that the Victoria lineage virus replicates more efficiently and is more pathogenic than the Yamagata lineage virus (13). It is unclear if more efficient viral replication in vivo correlates with higher antibody titers. Further study is needed to understand the nature of IBV infection in pigs.

Despite the overall relatively low prevalence of IBV antibodies in pigs (7.3%) in this study, serum samples collected from a large proportion of pig farms across the Midwest region tested positive for influenza B antibodies, and extremely high percentages of pigs with high antibody titers were found in certain farms. This finding indicates that pigs are susceptible to IBVs that can be an important pathogen for U.S. herds. Compared to other swine pathogens, such as PRRSV, IAV, and porcine circus virus 2, there is very limited knowledge regarding pig infection with IBV and whether it has affected the swine industry economically. Based on our results, improved surveillance for IBVs among pigs may be useful to protect animal and public health.

The susceptibility of domestic pigs to IBV infection described in this study supports and extends former observations made in the 1960s (40, 41). At that time, PRRSV was not recognized and may not have circulated in pigs (2). A significant difference in infectious disease ecology is the widespread occurrence of PRRSV in current pig populations. PRRSV is known to cause persistent and immunosuppressive infection in pigs (20, 34). Our serological surveillance data revealed a higher antibody prevalence of IBVs (up to 50%) in PRRSV-positive swine herds, suggesting that the introduction of this pathogen since the late 1980s has affected the susceptibility of pigs to IBV infection. However, coinfection of pigs with PRRSV and IBV needs to be performed to confirm this hypothesis. Although the seropositivity of IBVs in PRRSV-positive herds is very high and the partial sequences of the IBV NS gene were detected in 3 nasal swab samples collected from PRRSV-positive herds that also were positive for PRRSV and IAV, no IBV was successfully isolated from these samples, most likely due to the following reasons. First, the failure of virus isolation could be due to very low viral titers present in the nasal swab samples. Second, there could be a short period of time from when pigs were infected with IBV to when the virus was cleared despite the detection of viral RNA by RT-PCR. Third, the isolation of IBV could be complicated by the presence of multiple pathogens, such as IAV, PRRSV, and other bacterial agents present in pigs. Finally, IBV grows inefficiently compared to IAV.

In summary, our data further confirmed that pigs are another animal species that supports successful replication and transmission of IBVs. Susceptibility of pigs to IBVs indicates that the viruses are a swine pathogen and shall allow for further investigation and better surveillance. In addition, pig may be used as an animal model to study IBV replication and test for anti-IBV therapeutics and vaccines.