Inoculation of mice with influenza A virus.

C57BL/6 and DBA/2 mice were inoculated with influenza A viruses intranasally in 30 μl of sterile phosphate-buffered saline (PBS) after sedation with avertin (2,2,2-tribromoethanol; Sigma-Aldrich, MO). The 50% mouse lethal dose (MLD₅₀) was determined after mice were infected with 10-fold serial dilutions of the viruses from 10⁶ EID₅₀ to 10¹ EID₅₀. Morbidity and mortality were monitored for 21 days, and the MLD₅₀ values were calculated by the Reed-Munch method (20). Groups of five mice per inoculum size per isolate were tested with the exception of seasonal H1N1 (at 10⁶ EID₅₀, n = 6; at 10⁴ EID₅₀, n = 8), H1N1pdm (10⁴ EID₅₀, n = 9), H2N3 (10⁶ EID₅₀, n = 3; 10⁴ EID₅₀, n = 4), H2N4 (10⁴ EID₅₀, n = 4), H4N6 (10⁶ EID₅₀, n = 3), H5N9 (10⁶ EID₅₀, n = 6; 10⁴ EID₅₀, n = 4), H5N7 (10⁶ EID₅₀, n = 10), H7N3 (10⁶ EID₅₀, n = 9; 10⁴ EID₅₀, n = 4), H7N9 (10⁴ EID₅₀, n = 4), H9N2/Y280 (10⁶ EID₅₀, n = 10), H9N5 (10⁶ and 10⁴ EID₅₀, n = 4), H10N5 (10⁶ EID₅₀, n = 6; 10⁴ EID₅₀, n = 8), and H10N7 (10⁴ EID₅₀, n = 4) for DBA/2 mice and H5N7 (10⁶ EID₅₀, n = 8), H6N5 (10⁶ EID₅₀, n = 6), H7N3 (10⁶ EID₅₀, n = 10), H7N9 (10⁶ EID₅₀, n = 4), and H9N2 (10⁶ EID₅₀, n = 4) for C57BL/6.

Lung viral titers.

Lungs were collected on days 2 and 7 postinoculation with 10⁴ EID₅₀ of influenza A virus and stored at −80°C. They were homogenized in 1.0 ml of minimal essential medium, and homogenates were spun for 5 min at 1,000 × g to remove cellular debris. The supernatant was used to quantify the amount of infectious virus present in the lungs. Depending on the virus isolate, virus titers were determined in eggs or Madin-Darby canine kidney (MDCK) cells as described previously (1).

Hemagglutination inhibition and virus neutralization assays.

Influenza A virus-neutralizing activity in serum was quantified by hemagglutination inhibition and virus microneutralization (VN) assay. Sera were first treated with receptor-destroying enzyme (RDE) (RDE II Seiken; Denka Seiken UK Ltd., United Kingdom) for 18 h at 37°C, followed by a 30-min inactivation at 56°C. HI assays were done with 4 hemagglutination units of the virus and 0.5% turkey red blood cells (H1N1pdm) or 0.5% chicken red blood cells (avian virus isolates), as described previously (10). For a VN assay the serum was diluted 2-fold starting at a 1:10 dilution in PBS and incubated for 1 h at 37°C with 100 50% tissue culture infective doses (TCID₅₀) of A/California/4/09 virus. Next, 100 μl of the mixture of virus and serum was added to MDCK cells for 1 h at 37°C. Following the aspiration of the supernatant, cells were washed with PBS, and 200 μl of fresh minimal essential medium supplemented with 0.1% bovine serum albumin (A8412; Sigma-Aldrich), antibiotics (Invitrogen, NY), vitamins (Invitrogen), and 1 μg/ml tosylsulfonyl phenylalanyl chloromethyl ketone (TPCK)-trypsin (Worthington, NJ) was added. After 3 to 4 days at 37°C, the assay was developed by HA assay using turkey red blood cells. The average HI and VN titers were calculated following log₂ transformation of the highest serum dilution able to inhibit hemagglutination or virus replication, respectively.

Passive antibody transfer.

Human sera were collected as part of a clinical trial conducted during the influenza seasons of 2007 to 2008 and 2008 to 2009 in the Greater Vancouver Area of British Columbia, Canada, or in the vicinity of the Greater Hartford Area of Connecticut. All participants received the standard dose of the licensed trivalent split-virus influenza vaccine containing A/Solomon Islands/3/2006-like (H1N1), A/Wisconsin/67/2005-like (H3N2), and B/Malaysia/2506/2004-like viruses in 2007 to 2008 or A/Brisbane/59/2007 (H1N1)-like, A/Brisbane/10/2007 (H3N2)-like, and B/Florida/4/2006-like viruses in 2008 to 2009. Sera were collected before vaccination and 4 weeks after vaccination. Postvaccination sera from individuals aged 65 years and older with detectable HI and VN titers toward H1N1pdm (A/California/4/2009) were pooled and heat inactivated for 30 min at 56°C. To study the effect of neutralizing antibodies, we used age-matched pooled human sera without detectable HI and VN titers to H1N1pdm (A/California/4/2009), seasonal H1N1 (A/Brisbane/59/2007), and H7N3 (A/shorebird/Delaware/22/2006) viruses. Ferret polyclonal sera obtained from ferrets 14 days after inoculation with the H1N1pdm virus (HI titer of 2,560; VN titer of 320) or PBS was used as a positive or negative control, respectively. A total of 400 μl of pooled human sera, diluted 1:1 in PBS, was injected intraperitoneally into 10 mice 24 h prior to inoculation with a lethal dose of virus. The positive and negative controls were also injected into 10 mice each for the H1N1pdm experiment, while five PBS control mice were included in the H7N3 and seasonal H1N1 follow-up experiment.

Cytokine analysis.

Lungs were collected on days 2 and 7 postinoculation with 10⁴ EID₅₀ of influenza A virus, and concentrations of CCL2, CCL5, interleukin-6 (IL-6), tumor necrosis factor-alpha (TNF-α), and interferon-gamma (IFN-γ) were determined as described previously (1). Enzyme-linked immunosorbent assays (ELISAs) were performed according to the manufacturer's instructions (Quantikine kits; R&D Systems, Minneapolis, MN). At least four animals infected with a particular strain of influenza A virus were tested for one cytokine at a given time point.

Nonadapted avian influenza A viruses can replicate to high titers in C57BL/6 and DBA/2 mice.

We hypothesize that the large difference in the degrees of pathogenicity of avian influenza A viruses between DBA/2 and C57BL/6 mice may be due to increased replication efficiency. To test this, we measured day 2 and day 7 postinoculation pulmonary viral loads in mice infected with 10⁴ EID₅₀ of H7N3 or H10N5 virus (Fig. ​(Fig.1).1). These avian virus isolates were selected for their exceptionally large difference in pathology scores between DBA/2 and C57BL/6 mice. High-dose inoculation (10⁶ EID₅₀) with H7N3 or H10N5 virus induced 7% and 4% maximum weight loss in C57BL/6 mice, while DBA/2 mice succumbed to infection with 10² or 10⁴ EID_50, respectively. On day 2 postinoculation, lung viral titers of DBA/2 and C57BL/6 mice were similar for both H7N3 (10^6.25 versus 10^6.75 EID₅₀/ml, respectively) and H10N5 virus (10^5.3 versus 10^6.0 EID₅₀/ml, respectively) (Fig. ​(Fig.1).1). Interestingly, titers in H7N3-infected lungs were approximately 1 log₁₀ higher than those in H10N5-infected lungs. At day 7 postinoculation, H7N3-infected lungs of DBA/2 mice had higher virus loads than those of C57BL/6 mice (P < 0.05) (Fig. ​(Fig.1).1). Lungs of H10N5-infected DBA/2 mice also contained more virus than those of C57BL/6 mice; however, this difference was not significant (P = 0.07).

Open in a separate window

FIG. 1.

Virus titer in lungs of C57BL/6 (•) and DBA/2 () mice at 2 and 7 days postinoculation with 10⁴ EID₅₀ of A/shorebird/Delaware/22/2006 (H7N3), A/blue winged-teal/Alberta/271/2007 (H10N5), and A/California/4/2009 (H1N1pdm). *, P < 0.01. Infectious titer values on the y axis represent EID₅₀/ml for H7N3 and H10N5 viruses and TCID₅₀/ml for H1N1pdm.

To assess the ability of other non- or low-pathogenic avian influenza A virus isolates to infect mice, we measured the serological response against the challenge virus in convalescent-phase serum as a surrogate marker for viral replication. In C57BL/6 mice, an HI titer was detected postinoculation with five of the eight virus isolates studied, suggesting that the majority of the isolates replicate in the respiratory tract of mice (Table ​(Table2).2). Inoculation of three viruses, an H4N1, H6N5, and an H5N7 virus, did not cause seroconversion in C57BL/6 mice. Because DBA/2 mice are generally more susceptible to influenza virus infection and because fewer isolates were non- or low-pathogenic isolates, we tested convalescent-phase serum following inoculation with only five virus isolates. Of the five convalescent-phase sera tested, three contained a detectable HI titer whereas two (an H4N6 and an H9N6 virus) did not (Table ​(Table2).2). These data suggest that most influenza virus isolates are capable of replicating in the respiratory tract of mice, but the outcome after infection depends entirely on the mouse strain, virus strain, or a combination of both.

Pandemic H1N1 2009 virus A/California/4/2009 is highly pathogenic in DBA/2 mice.

Based on the results presented above, we next looked at the replication of the H1N1pdm viruses in DBA/2 and C57BL/6 mice. Inoculation of mice with 10⁶ to 10² EID₅₀ of A/California/4/09, a representative H1N1pdm virus, resulted in 100% mortality after 8 to 12 days, and inoculation with 10¹ EID₅₀ caused significant weight loss in 100% and mortality in 50% of DBA/2 mice (Table ​(Table1).1). In contrast, C57BL/6 mice lost a significant amount of weight (14%; P < 0.05) by day 7 when inoculated with 10⁶ EID₅₀, and 60% of the mice died. Inoculation with 10⁵ to 10⁴ EID₅₀ did not cause death of C57BL/6 mice. Therefore, the MLD₅₀ for A/California/4/2009 was 10⁵-fold lower in DBA/2 than in C57BL/6 mice, a finding consistent with results with other virus strains.

Increased pathogenicity is often associated with higher viral loads and increased levels of proinflammatory cytokines such as CCL2, IL-6, and TNF-α. At day 2 postinoculation with 10⁴ EID₅₀ of H1N1pdm, lung viral titers of DBA/2 mice and C57BL/6 mice were similar (Fig. ​(Fig.1).1). There were also no significant differences in the levels of CCL2, CCL5, and IL-6 between the strains (Table ​(Table3),3), but levels of TNF-α were significantly higher in DBA/2 than C57BL/6 mice (P < 0.01). At day 7 postinfection, lung homogenates of DBA/2 mice had higher viral loads than C57BL/6 mice (P < 0.01) (Fig. ​(Fig.1)1) as well as significantly higher concentrations of CCL2 and IL-6 (P < 0.01) (Table ​(Table33).

We thank David Walker for isolating and propagating influenza A virus isolates. We also acknowledge Nancy Cox, Sasha Klimov, and Ruben Donis of the Centers of Disease Control, Ron Fouchier of Erasmus Medical Center, Malik Peiris and Guan Yi of the University of Hong Kong, and the World Health Organization Global Influenza Surveillance Network as a source of influenza A viruses. Finally, we thank M. Ducatez and S. Schultz-Cherry for critically reviewing the paper.

This project was funded in part by grants from the National Institute of Allergy and Infectious Diseases, National Institutes of Health, Department of Health and Human Services, under contract HHSN266200700005C, by Department of Defense award W81XWH-09-1-0391, by the Centers of Infectious Diseases Control at St. Jude Children's Research Hospital, and by the American Lebanese Syrian Associated Charities.