Exchange of gene segments through reassortment is a major feature of influenza A virus evolution and frequently contributes to the emergence of novel epidemic, pandemic and zoonotic strains. It has long been evident that viral diversification through reassortment is constrained by genetic incompatibility between divergent parental viruses. In contrast, the role of virus-extrinsic factors in determining the likelihood of reassortment has remained unclear. To evaluate the impact of such factors in the absence of confounding effects of segment mismatch, we previously reported an approach in which reassortment between wild-type (wt) and genetically tagged variant (var) viruses of the same strain is measured. Here, using wt/var systems in the A/Netherlands/602/2009 (pH1N1) and A/Panama/2007/99 (H3N2) strain backgrounds, we have tested whether inoculation of parental viruses into distinct sites within the respiratory tract limits their reassortment. Using a ferret model, matched parental viruses were either co-inoculated intranasally, or one virus was instilled intranasally while the second was instilled intratracheally. Dual intranasal inoculation resulted in robust reassortment for wt/var viruses of both strain backgrounds. In contrast, when infections were initiated simultaneously at distinct sites, strong compartmentalization of viral replication was observed and minimal reassortment was detected. The observed lack of viral spread between upper and lower respiratory tissues may be attributable to localized exclusion of super-infection within the host, mediated by innate immune responses. Our findings indicate that dual infections in nature are more likely to result in reassortment if viruses are seeded into similar anatomical locations and have matched tissue tropisms.ImportanceGenetic exchange between influenza A viruses (IAVs) through reassortment can facilitate the emergence of antigenically drifted seasonal strains and plays a prominent role in the development of pandemics. Typical human influenza is concentrated in the upper respiratory tract; however, lower respiratory tract (LRT) infection is an important feature of severe cases, which are more common in the very young, the elderly, and individuals with underlying conditions. In addition to host factors, viral characteristics and mode of transmission can also increase the likelihood of LRT infection: certain zoonotic IAVs are thought to favor the LRT and transmission via small droplets allows direct seeding into lower respiratory tissues. Toward gauging the likelihood of reassortment in co-infected hosts, we assessed the extent to which initiation of infection at distinct respiratory sites impacts reassortment frequency. Our results reveal that spatially distinct inoculation results in anatomical compartmentalization of infection, which in turn strongly limits reassortment.
