Pandemic Risk Assessment for Swine Influenza A Virus in Comparative In Vitro and In Vivo Models

Abstract

Swine influenza A viruses pose a public health concern as novel and circulating strains occasionally spill over into human hosts, with the potential to cause disease. Crucial to preempting these events is the use of a threat assessment framework for human populations. However, established guidelines do not specify which animal models or in vitro substrates should be used. We completed an assessment of a contemporary swine influenza isolate, A/swine/GA/A27480/2019 (H1N2), using animal models and human cell substrates. Infection studies in vivo revealed high replicative ability and a pathogenic phenotype in the swine host, with replication corresponding to a complementary study performed in swine primary respiratory epithelial cells. However, replication was limited in human primary cell substrates. This contrasted with our findings in the Calu-3 cell line, which demonstrated a replication profile on par with the 2009 pandemic H1N1 virus. These data suggest that the selection of models is important for meaningful risk assessment.

Keywords: animal; epithelial cells; ferrets; humans; influenza A virus; mice; models; pandemics; risk assessment; swine; whole genome sequencing.

Figure 1

BEAST phylogenies for swine influenza H1Nx isolates collected from 2014 to 2019. (A,B) Phylogenetic reconstruction for hemagglutinin (HA) segment of swine H1Nx isolates. Nodes with a posterior support of greater than 95% are annotated with a 95% Bayesian Credible Interval in blue. Taxa for H1N2 variant isolates are colored in red.

Figure 2

BEAST phylogenies for swine influenza A HxN2 isolates collected from 2014 to 2019. (A,B) Phylogenetic reconstruction for neuraminidase (NA) segment of swine HxN2 isolates. Nodes with a posterior support of greater than 95% are annotated with a 95% Bayesian Credible Interval in blue.

Figure 3

Weight change and lung virus replication following infection of DBA/2 or BALB/c mice with A/sw/GA/27480/19 (H1N2). Five- to eight-week-old BALB/c and DBA/2 mice (n = 25 mice/group) were inoculated intranasally with either 1 × 10⁵ pfu of virus in a 30 µL volume (n = 20) or with PBS (n = 5). Weight loss was tracked for 13 days post-infection (dpi) (A); dashed lines represent mock-infected control groups. At 2 and 4 dpi, a subset of five mice from each infected group were euthanized, and their lungs were collected. Viral titers, described as pfu/mL of lung homogenate, were determined by plaque assay (B). Significance values of ≤0.005 and ≤0.0005 are denoted by ** and ***, respectively. Error bars indicate mean ± SD.

Figure 4

Replication and transmission of A/sw/GA/27480/19 (H1N2) in a swine model. Six-week-old pigs were inoculated intranasally with 2 × 10⁶ pfu of GA/19 in a 2 mL volume (n = 3). At 2, 4, and 6 dpi, BAL samples were collected (black data points). At 3 dpi, naïve contact animals (n = 3) were co-housed with infected animals, and nasal swabs were collected daily (red data points). Viral titers were determined by plaque assay. Error bars indicate mean ± SD.

Figure 5

Replication and transmission of A/sw/GA/27480/19 (H1N2) in ferrets. Twelve-week-old ferrets were inoculated intranasally with 1 × 10⁶ pfu of GA/19 in a 1 mL volume (n = 6). At 1, 3, 5, 7, 9, and 11 dpi, nasal wash samples were collected, and titers were evaluated by plaque assay (black data points). At 2 dpi, naïve contact animals (n = 4) were co-housed with infected animals (1:1), and nasal washes were taken and evaluated for virus titer by plaque assay (red data points). Error bars indicate mean ± SD.

Figure 6

Replication of H1 influenza viruses in vitro. Primary porcine nasal epithelial (PNE), primary human bronchial epithelial (NHBE), and Calu-3 cells (all at ALI) were infected apically with either GA/19 (A) or A/CA/07/09 (B) at an MOI of 0.01. At 12, 24, 48, 72, and 96 h post-infection, the apical surface of cultures was washed, the fluid collected, and then titered for virus by plaque assay. Dashed lines denote duplicate experiments. Error bars indicate mean ± SD.