The ferret model of human influenza: understanding the interplay between the immune response and human influenza viruses in vivo

Influenza viruses cause seasonal, epidemic, and pandemic respiratory infections in humans, resulting in significant morbidity and mortality, with huge economic costs. The innate abilities of influenza viruses to escape from pre-existing immunity induced by infection or vaccination (antigenic drift), the emergence of novel strains of influenza virus (antigenic shift), and the development of resistance to antiviral treatments, ensures that influenza viruses continue to evolve and pose great challenges to prevention and control of infection. The ferret model is considered the ‘gold standard’ for human influenza. Ferrets can be directly infected with human influenza viruses without the need for adaptation; they also experience similar pathogenesis and clinical symptoms to humans. In this thesis, the ferret model was used to investigate the influence of the immune response on influenza virus infection. Two distinct concepts were studied: firstly, how the immune response to vaccination can drive antigenic drift of currently circulating influenza viruses and, secondly, whether a prior influenza virus infection may influence susceptibility to subsequent infection by the same, or a different, influenza virus. An in vivo ferret model of antigenic drift was established that enabled passaging of human influenza viruses in naïve or vaccinated ferrets multiple times. The potential for antigenic drift of the A(H1N1)pdm09 virus was assessed in this model. Following multiple passages in vivo, a mutation in an antigenic site of the HA protein (N156K) emerged, enabling the A(H1N1)pdm09 virus to escape immune pressure. The N156K antigenic mutant could not be isolated using routine culturing techniques without significant adaptation and had an altered receptor binding specificity compared to wildtype virus. The mutant virus was fit in vivo. The N156K mutation has been detected in human samples at a low frequency, yet as culturing methods may bias against the detection of the N156K mutation, the true frequency may be underestimated. As vaccine selection is dependent on the ability to isolate viruses that are representative of those circulating in the human population, without significant adaptation during the culturing process, these findings have significant public health implications. The theory of ‘non-specific immunity’ hypothesizes that after a primary influenza virus infection, the host has a reduced susceptibility to infection with an unrelated influenza virus. This challenges classical immunological theory as influenza virus types, such as A and B, elicit minimal cross-reactive immune responses. In this thesis, studies demonstrated that prior infection of ferrets with A(H1N1)pdm09 virus reduced their susceptibility to infection with influenza B virus, 1-7 days later. However, primary infection of influenza B virus had no effect on the susceptibility of ferrets to subsequent challenge with A(H1N1)pdm09 virus. Strikingly, these ferrets experienced dual influenza virus infections, with both the influenza B and A(H1N1)pdm09 viruses displaying similar infection kinetics to those seen in ferrets infected with only one of these viruses. The fact that ‘non-specific immunity’ was observed only following A(H1N1)pdm09 virus infection, and not following influenza B virus infection, suggests that this phenomenon may be virus specific. ‘Temporary susceptibility’ describes incidences of re-infection of individual patients, or multiple individuals in a population, with the same virus after a primary infection. This may be due to (i) heterogeneity in each individual’s immune response, (ii) temporary susceptibility whilst immunological memory is developing, and/or (iii) waning immunity. The potential for re-infection with the same virus was assessed in the ferret model in intervals of 1-8 weeks after primary virus infection. At no time intervals were ferrets susceptible to re-infection with either the A(H1N1)pdm09 virus or the influenza B virus following a successful primary infection. Although the phenomenon of re-infection was not observed in the ferret model in this work, further investigation by in vivo modeling of this phenomenon is warranted. In summary, this thesis highlights the complex interplay between the immune response and influenza virus infection. By using the ferret model, the effect of the immune response on driving antigenic changes in influenza viruses was demonstrated. Furthermore, the hypothesis of non-specific immunity to unrelated influenza viruses was shown, for the first time. Data obtained through these studies are now being applied to mathematical modeling to improve our understanding of influenza in the population.