12 October 2017 in PLOS Pathogens
Influenza A virus (IAV) infection can be severe or even lethal in toddlers, the elderly and patients with certain medical conditions. Infection of apparently healthy individuals nonetheless accounts for many severe disease cases and deaths, suggesting that viruses with increased pathogenicity co-circulate with pandemic or epidemic viruses. Looking for potential virulence factors, we have identified a polymerase PA D529N mutation detected in a fatal IAV case, whose introduction into two different recombinant virus backbones, led to reduced defective viral genomes (DVGs) production. This mutation conferred low induction of antiviral response in infected cells and increased pathogenesis in mice. To analyze the association between low DVGs production and pathogenesis in humans, we performed a genomic analysis of viruses isolated from a cohort of previously healthy individuals who suffered highly severe IAV infection requiring admission to Intensive Care Unit and patients with fatal outcome who additionally showed underlying medical conditions. These viruses were compared with those isolated from a cohort of mild IAV patients. Viruses with fewer DVGs accumulation were observed in patients with highly severe/fatal outcome than in those with mild disease, suggesting that low DVGs abundance constitutes a new virulence pathogenic marker in humans. Influenza A viruses are the causative agents of annual epidemics, sporadic zoonotic outbreaks and occasionally pandemics. Worldwide, acute respiratory infections caused by influenza A viruses continue to be one of the main causes of acute illness and death. The appearance in 2009 of a new H1N1 pandemic influenza strain reinforced the search to identify viral pathogenicity determinants for evaluation of the consequences of virus epidemics and potential pandemics for human health. Here we identify a new general virulence determinant found in a cohort of severe/fatal influenza virus-infected patients, a reduced accumulation of viral defective genomes. These molecules are incomplete viral genome segments that activate the innate immune response. This data will contribute to the prediction of influenza disease severity, to improved guidance of patient treatment and will enable the development of risk-based prevention strategies and policies.