Genome-wide mutagenesis of influenza virus reveals unique plasticity of the hemagglutinin and NS1 proteins

Abstract

The molecular basis for the diversity across influenza strains is poorly understood. To gain insight into this question, we mutagenized the viral genome and sequenced recoverable viruses. Only two small regions in the genome were enriched for insertions, the hemagglutinin head and the immune-modulatory nonstructural protein 1. These proteins play a major role in host adaptation, and thus need to be able to evolve rapidly. We propose a model in which certain influenza A virus proteins (or protein domains) exist as highly plastic scaffolds, which will readily accept mutations yet retain their functionality. This model implies that the ability to rapidly acquire mutations is an inherent aspect of influenza HA and nonstructural protein 1 proteins; further, this may explain why rapid antigenic drift and a broad host range is observed with influenza A virus and not with some other RNA viruses.

Keywords: insertional mutagenesis; protein flexibility; viral evolution.

Fig. 1.

Construction of mutant libraries and rescue of viruses. (A) Influenza segments (blue) were cloned into a pGEM shuttling vector and mutagenized in vitro. The virus segment containing the transposon (red) was removed from the shuttling vector and ligated into the viral RNA expression plasmid pPolI. Finally, the body of the transposon was removed and the viral segment was ligated back together to leave the 15-nt insertion (red striped). (B) The mutant library for a given segment was transfected with the other seven WT segments into 293T cells. Rescued virus was used to infect MDCK cells. After 24 h, virus was collected and used to infect naïve MDCK cells. RNA was collected from the second set of MDCK cells and subjected to RT-PCR for the mutant segment. The input library was PCR-amplified from the expression plasmid. Both sets of amplified DNA were prepared for sequencing and submitted for Illumina next-generation sequencing.

Fig. 2.

Insertional mutagenesis of the IAV genome reveals unique plasticity of NS and HA segments. (A–H) Results of the Illumina sequencing analysis for all eight segments. For all panels, input libraries (Left) were sequenced from the DNA rescue plasmids to determine mutational coverage. Rescued virus was amplified via RT-PCR from MDCK cell culture (Right). The x axis corresponds to the nucleotide position along the viral segment. The y axis corresponds to raw number of reads for the input and fold increase above background for the output. For output samples, a cutoff of 10-fold enrichment is indicated with the dotted line. Green lines represent viruses that tolerate insertions at nucleotide positions greater than 10-fold above background. Red lines indicate viruses that are enriched less than 10-fold. Asterisks on A and B indicate values greater than the y-axis limit; actual values are denoted on the panel.

Fig. 3.

The HA head domain tolerates insertional mutations. (A) The HA library rescued under noncompetitive conditions in MDCK cells (red) and under competitive conditions in embryonated chicken eggs (blue). The nucleotide position of the HA segment is indicated along the x axis. The HA1 (head) and HA2 (stalk) domains are denoted above the graph. (B) Venn diagram showing the overlap of the two rescue conditions from A.

Fig. 4.

Locations of transposon insertions in the HA head. (A) Diagram of the HA segment showing the locations of transposons (red and green triangles) recovered from both the MDCK and egg rescues as well as HA mutants verified by plaque purification. All insertional mutants in the HA1 domain were also denoted on the primary sequence of the HA1 domain (black triangles). Escape mutation locations defining antigenic sites are indicated by the colored boxes. Insertion locations from A are shown as red or green spheres on the crystal structure (PDB ID code 1RU7) of the HA monomer (B) and trimer (C). B Inset is an enlargement of the HA head domain. (D) Individual viruses (denoted by green in A–C) were cloned and rescued. MDCK cells were infected at an MOI of 0.001, and samples were plaqued at the indicated times.