Swine challenge studies were performed at South Dakota State University and were approved by the Institutional Animal Care and Use Committee (approval number 11-051A) and were performed under biosafety level 2+ conditions.

Swine testicle (ST) cells (ATCC) were grown in DMEM containing 5% fetal bovine serum at 37°C with 5% CO₂. A/swine/Minnesota/0745/2010 H1N2 (MN745) and A/swine/Minnesota/0432/2010 H1N1 (MN432) were isolated from routine diagnostic specimens submitted to Newport Laboratories (Worthington, Minnesota). Viral isolation was performed on ST cells from nasal swabs collected from pigs displaying influenza-like illness. Full genome sequencing was performed at St. Jude Children's Research Hospital. For simplicity and clarity, H1N2 for MN745 and pH1N1 for MN432 are used throughout the manuscript. Viral growth studies were performed on a monolayer of ST or A549 cells (ATCC) using an inoculum of 1.0 TCID₅₀/mL (tissue culture infectious dose 50) in triplicate. Samples were removed at 0, 24, 48, 72 and 96 hours and infectious viruses were titrated on the same cell line and the titers determined by the method of Spearman-Karber.

T25 flasks containing a monolayer of ST cells were infected with either pH1N1 or H1N2 at a multiplicity of infection (MOI) of 0.1 or co-infected with both pH1N1 and H1N2 with various combinations of MOI's ranging from 0.1∶0.1, 1.0∶0.1 or 0.1∶1.0. 1 ml of inoculum(s) was added to each flask and the virus was adsorbed for 1 hour and then removed and 20 mL of DMEM was added back. Flasks were incubated at 37°C with 5% CO2 for 3–4 days and then 0.1 mL of fluid was passed to a T25 flask containing a confluent monolayer of ST's. This procedure was repeated for a total of 10 passages. A sample was collected from each flask after each passage and analyzed for the presence of H1N2-M and pH1N1-M and NA genes by real time-RT-PCR (rt-RT-PCR). Cell culture harvests of passage x+10 were titrated on ST cells and 10 plaques from each co-infection flask and one plaque from each parent virus flask were expanded. Plaques were analyzed by rt-RT-PCR to determine M and NA lineages and identify reassortant viruses. Two reassortant viruses, designated R1 and R2, were identified that contain pH1N1-derived M and H1N2-derived NA and these two viruses were used in this study.

RNA was extracted with the MagMAX-96 viral RNA isolation kit per manufacturer's instructions (Life Technologies). rt-RT-PCR was performed using QIAGEN QuantiTect RT-PCR (QIAGEN) and primers and probes designed to detect and differentiate the North American lineage M gene and pH1N1-derived Eurasian M gene [17]. For specific detection of the pH1N1 NA gene, primers and probes were designed targeting sequences unique to Eurasian-lineage NA. The pH1N1-NA specific PCR used primers NA Forward: 5′-TCTCCCTATCCAAACACCATTG-3′ and NA Reverse: 5′-AGACAATCCACGCCCTAATG-3′and NA probe: 5′- AGACGATACTGGACCACAACTGCC-3′ using Cy5 as the fluorophore. pH1N1-NA detection was multiplexed with the differential M gene primers and probes. For rt-RT-PCR, negative samples were assigned a value of 37.1, which corresponds to detection limit of the method. Viruses were sent to St. Jude Children's Research Hospital's Hartwell Sequencing Center for determination of full-genome sequence. Sequences for the parental viruses MN432 (pH1N1) and MN745 (H1N2) were submitted to GenBank with accession numbers JQ023759–JQ023774.

Plaque purified pH1N1, H1N2 and reassortant viruses R1 and R2 were expanded by one additional passage on ST cells and cell culture fluids were titrated on ST cells. Thirty-six 3-week-old pigs were obtained from a commercial high-health herd and their SIV-negative status at day 0 was confirmed using a commercial ELISA (FlockChek Avian Influenza MultiS-Screen Antibody Test Kit; Idexx Laboratories Inc, Westbrook, Maine). Nasal swabs were also negative for SIV by rt-RT-PCR prior to study commencement. Pigs were divided into five groups of four pigs housed in separate rooms. On day 0, pigs were challenged intranasally with 1 mL of 1×10⁶ TCID₅₀/mL virus in each nare with either pH1N1, H1N2, R1 or R2 delivered by a syringe. Control pigs were mock challenged with 1 mL DMEM in each nare. On day one, four naïve pigs were added to each challenge group. Pigs were observed daily for clinical signs of influenza and their temperatures recorded. Nasal swabs were collected daily to assess viral shedding. Viral titers in nasal swabs were quantified by rt-RT-PCR. Pigs were euthanized by a lethal dose of pentobarbitol on day five post-challenge and lungs were harvested.

Samples of lungs collected from infected pigs on day five post-challenge were fixed in 10% buffered formalin and submitted to the Iowa State University Veterinary Diagnostic Laboratory for histopathology and immunohistochemistry for SIV.

The Student's t-test was used to determine statistical significance of titers in vitro as well as for Ct values in nasal swabs and infectious viral titers in lung samples using a probability value of 0.05 to indicate significance using the JMP software program (SAS, Cary, NC)

Routine characterization of SIV isolated from diagnostic samples identified MN745 as a H1N2 virus with homology to the δ-subcluster and MN432 as H1N1 with homology to pH1N1 based upon HA, M and NA gene sequencing. MN432 and MN745 were selected randomly as representatives of swine pH1N1 and endemic TRIG SIV, respectively. The percentage nucleotide identity for each segment between these two viruses is 94.0% for PB2, 94.2% for PB1, 94.1% for PB1, 76.0% for HA, 94.7% for NP, 53.8% for NA, 87.6% for M, and 94.2% for NS, respectively. rt-RT-PCR analysis of ST cell culture harvests after each passage indicated that flasks coinfected with viruses at MOI ratios of 1.0∶1.0 (Flask A) or 1.0∶0.1 H1N2∶pH1N1 (Flask C) only contained the endemic M gene from H1N2 (MN745) (Figure 1A, 1C). Additionally, the pH1N1-M and NA genes from pH1N1 (MN432) could not be detected after passage 10. In contrast, the flask coinfected with viruses at a MOI ratio of 0.1∶1.0 H1N2∶pH1N1 (Flask B) contained a mixed population after 10 passages as evidenced by positive PCR values for both the endemic and pH1N1-M genes (Figure 1B). In conjunction with the absence of signal for pH1N1-NA after 10 passages, these data also suggested that at least one of the viral populations present was a reassortant virus.

Ten plaques derived from each flask after the tenth passage were analyzed by rt-RT-PCR for the presence of endemic M and pH1N1-M and NA genes and partial HA gene sequence. Consistent with PCR data from cell culture harvests, all plaques from the flasks inoculated with MOI ratios of 1.0∶1.0 or 1.0∶0.1 H1N2∶pH1N1 had genotypes consistent with H1N2 based on HA, NA and M gene analysis. Eight of ten plaques from the flask coinfected with viruses at a MOI ratio of 0.1∶1.0 H1N2∶pH1N1 had a H1N2 genotype. The remaining two plaques were identified as reassortant viruses because they contained a pH1N1-derived M gene, were negative by PCR for pH1N1-NA but positive for H1N2 NA, and had HA gene sequences identical to H1N2.

Full genome sequences were determined for H1N2 and pH1N1 from the initial cell passage and following plaque purification after 10 sequential cell culture passages. Reassortants R1 and R2 were also sequenced following plaque purification. Pairwise ClustalW alignment of nucleotide sequences allowed determination of parentage of each gene (Table 1). For R1, polymerase basic 1 (PB1), polymerase acidic (PA), nucleoprotein (NP), M and NS (non-structural) genes were all derived from the pH1N1. The other three genes PB2 (polymerase basic 2), HA and NA were derived from H1N2. For R2, only PB1, NP and M were derived from pH1N1, with the remaining genes originating from H1N2.

Full genome sequences identified synonymous mutations in seven and non-synonymous mutations in six of the eight genes following 10 passages in cell culture. Nucleotide and amino acid changes, as compared to the parental gene sequence in the inoculum, are shown in Table 2. The majority of mutations were identified only in a single virus with the exception of a R528Q mutation that was found in the HA protein of H1N2 and both reassortants R1 and R2. Additionally, interestingly, both reassortants also had a S224R mutation in the M1 protein within the M segment, which was absent in both parental viruses.

Multiple cycle growth kinetics were performed on pH1N1 and H1N2, both prior to and following 10 sequential passages in ST cells, as well as for R1 and R2. In ST cells, R1 and R2 had maximal titers similar to parent H1N2 after 10 passages in cell culture and 2 log₁₀ TCID₅₀/mL higher than parent pH1N1 following ten cell culture passages (Figure 2). These differences were statistically significant (P<0.05, denoted by different subscripts). The maximal titer for both pH1N1 and H1N2 increased 1.4 and 0.8 log₁₀ TCID₅₀/mL, respectively following 10 cell culture passages. Statistical analysis of the differences of virus replication among different groups of infections here and elsewhere in the manuscript was performed by the Student's t-test coupled with the JMP software program. Statistically significant differences (P<0.05) for the level of virus replication at different time points among experimental groups were indicated by different subscripts (a, b, c, d, and e).

Additionally, multiple cycle growth curves were performed using A549 cells. pH1N1 and H1N2, prior to passaging, reached maximal titers of 7.9 and 9.0 log₁₀ TCID₅₀/mL, respectively, at 48 hours (Figure 3). Sequential passage in ST cells decreased pH1N1 and H1N2 growth rates and titers as seen by a slower growth phenotype for viruses derived from x+10 in A549 cells. Reassortant viruses R1 and R2 had similar growth kinetics and maximal titers as compared to the two parents pH1N1 and H1N2 following 10 cell passages.

SIV was detected in nasal swabs by rt-RT-PCR on day one post infection (p.i.) among all intranasally challenged pigs and was consistently detected until pigs were euthanized on day 5 p.i. (Figure 4). pH1N1 and R2 shedding was significantly higher (P<0.05) than H1N2 and R1 by 5 d.p.i. Similarly, virus was detected in naïve contact pigs beginning day one post exposure (day 2 post infection) continually until the pigs were euthanized (Figure 5). Interestingly, pH1N1 and R1 shed virus in contact exposure pigs at significantly higher levels than R2. Despite viral shedding, no clinical signs of influenza infection or elevated temperatures were observed (data not shown). All four viruses were capable of infecting and transmitting between pigs as determined by rt-RT-PCR. Lung samples collected on day 5 p.i. from directly challenged and day 5 p.e. contact pigs were all positive for virus by rt-RT-PCR with the exception of one pig directly infected with H1N2 and two contact pigs exposed to R2 (Figure 6). Titration of lung homogenates on ST cells only detected virus replication in two of the four pigs challenged with H1N2 (mean 0.5 log₁₀ TCID₅₀/mL) and three of the four pigs challenged with pH1N1 (mean = 1.8 log₁₀ TCID₅₀/mL, Figure 7). In contrast, all pigs challenged with R1 and R2 were negative for the presence of infectious viruses as measured in ST cells. Further analysis of lung homogenates from contact exposure pigs showed that these pigs were all positive for SIV with the exception of 1 pig exposed to R2. However, viral titers were significantly lower for reassortant viruses R1 and R2 (1.3 and 0.7 log₁₀ TCID₅₀/mL, respectively) than parental H1N2 and pH1N1 (3.6 and 3.4 log₁₀ TCID₅₀/mL, respectively). Histopathology of nasal challenge pigs found necrotizing bronchiolitis ranging from bronchiolar epithelial attenuation to necrosis. There were no discernible differences in microscopic lesions that could be attributed to different viruses. In addition, no obvious macroscopic lung lesions were observed in infected pigs. Immunohistochemistry (IHC) was positive for SIV in two of the four pigs intranasally challenged with H1N2 and three of the four pigs challenged with pH1N1, R1 and R2. Contact exposure pigs also contained lesions consistent with influenza infection, namely disorganization and sloughing of epithelial cells lining the airways, which were occasionally lightly cuffed with infiltrating lymphocytes. Approximately half of the pigs exposed to H1N2, R1 and pH1N1 were IHC positive for SIV. All pigs exposed to R2 were IHC negative for SIV. No lesions were detected in mock-challenged pigs and were IHC negative for SIV. Similarly, all nasal swabs from mock-challenged pigs were negative by rt-RT-PCR and no virus was detected in lung samples by rt-RT-PCR or titration in cells (data not shown).