Cross-Reactive Neutralizing Antibodies Directed against Pandemic H1N1 2009 Virus Are Protective in a Highly Sensitive DBA/2 Mouse Influenza Model

Our ability to rapidly respond to an emerging influenza pandemic is hampered somewhat by the lack of a susceptible small-animal model. To develop a more sensitive model, we pathotyped 18 low-pathogenic non-mouse-adapted influenza A viruses of human and avian origin in DBA/2 and C57BL/6 mice. The majority of the isolates (13/18) induced severe morbidity and mortality in DBA/2 mice upon intranasal challenge with 1 million infectious doses. Also, at a 100-fold-lower dose, more than 50% of the viruses induced severe weight loss, and mice succumbed to the infection. In contrast, only two virus strains were pathogenic for C57BL/6 mice upon high-dose inoculation. Therefore, DBA/2 mice are a suitable model to validate influenza A virus vaccines and antiviral therapies without the need for extensive viral adaptation. Correspondingly, we used the DBA/2 model to assess the level of protection afforded by preexisting pandemic H1N1 2009 virus (H1N1pdm) cross-reactive human antibodies detected by a hemagglutination inhibition assay. Passive transfer of these antibodies prior to infection protected mice from H1N1pdm-induced pathogenicity, demonstrating the effectiveness of these cross-reactive neutralizing antibodies in vivo.Respiratory tract infections are the third leading cause of mortality in the world (27). Influenza, a disease of the airways caused by influenza viruses, is responsible for approximately half a million deaths and 3 to 5 million hospitalizations per year (28). In addition to the annual disease burden, influenza A virus is more notoriously known for its ability to cause pandemics. Three pandemics have been reported in the twentieth century: the first that occurred in 1918 (Spanish influenza) killed 20 to 50 million individuals (15); the other two in 1957 and 1968, although less lethal, killed millions due to the lack of preexisting immunity. In April 2009, two cases of febrile illness were confirmed to be caused by swine-origin influenza A virus (H1N1) (4, 8). Continuous spread within North America and other parts of the world has signaled the first influenza pandemic of this century.To study the pathogenicity of influenza A viruses, including the current pandemic A (H1N1) 2009 virus (H1N1pdm), in mammalian hosts and to determine the effectiveness of pharmaceutical interventions, it is essential to have a sensitive animal model. Although influenza has some important differences in mice and humans, a murine model is the only animal model thus far described that allows for relatively high group numbers and any relatively high throughput. Unfortunately, only a few strains of influenza A virus—almost exclusively belonging to the highly pathogenic avian influenza virus isolates of the H5 and H7 subtypes—are pathogenic in most commonly used mouse strains without adaptation through serial passaging. The hemagglutinin (HA) proteins of these H5 and H7 viruses contain a basic amino acid cleavage site, allowing them to spread systemically (12, 19, 26). Most other subtypes of influenza virus, including H1N1 and H3N2, either do not infect or cause very mild disease in mice. The requirement for adaptation of a pandemic virus to commonly used mouse strains can lead to a delay in the gathering of important data to help guide public health control strategies. As such, the lack of a sensitive small-animal model to study the infection dynamics of various subtypes of avian influenza viruses severely hampers the rapid and effective response required during a pandemic or prepandemic situation.This study was designed to demonstrate the utility of DBA/2 mice, previously reported to be susceptible to highly pathogenic influenza viruses (1), to study infections caused by several influenza A virus subtypes isolated from birds or humans without the need for prior adaptation. To assess the utility of the model to respond to emerging strains, we used DBA/2 mice to examine the functional activity of sera from individuals previously shown to have preexisting cross-reactive H1N1pdm antibodies. It is hypothesized that these individuals may be partially protected from infection because of the presence of cross-reactive neutralizing antibodies produced after infection with a different but related H1N1 virus. This hypothesis is supported by in vitro microneutralization and hemagglutination inhibition (HI) assays (2, 10); however, it is not yet known whether these antibodies are also functional in vivo.