Immune Pressure on Polymorphous Influenza B Populations Results in Diverse Hemagglutinin Escape Mutants and Lineage Switching

Abstract

Mutations arise in the genomes of progeny viruses during infection. Mutations that occur in epitopes targeted by host antibodies allow the progeny virus to escape the host adaptive, B-cell mediated antibody immune response. Major epitopes have been identified in influenza B virus (IBV) hemagglutinin (HA) protein. However, IBV strains maintain a seasonal presence in the human population and changes in IBV genomes in response to immune pressure are not well characterized. There are two lineages of IBV that have circulated in the human population since the 1980s, B-Victoria and B-Yamagata. It is hypothesized that early exposure to one influenza subtype leads to immunodominance. Subsequent seasonal vaccination or exposure to new subtypes may modify subsequent immune responses, which, in turn, results in selection of escape mutations in the viral genome. Here we show that while some mutations do occur in known epitopes suggesting antibody escape, many mutations occur in other parts of the HA protein. Analysis of mutations outside of the known epitopes revealed that these mutations occurred at the same amino acid position in viruses from each of the two IBV lineages. Interestingly, where the amino acid sequence differed between viruses from each lineage, reciprocal amino acid changes were observed. That is, the virus from the Yamagata lineage become more like the Victoria lineage virus and vice versa. Our results suggest that some IBV HA sequences are constrained to specific amino acid codons when viruses are cultured in the presence of antibodies. Some changes to the known antigenic regions may also be restricted in a lineage-dependent manner. Questions remain regarding the mechanisms underlying these results. The presence of amino acid residues that are constrained within the HA may provide a new target for universal vaccines for IBV.

Keywords: hemagglutinin; influenza B; lineage; next generation sequencing; quasispecies.

Figure 1

HA consensus (>50%) sequences of wild-type and escape viruses. Viruses are labeled as in Table 1. NGS sequence data for each escape virus was translated and aligned with the sequence from the input virus. The wild-type virus sequence is underlined an in bold font. Numbering is based on the Yamagata lineage B/Phuket/3073/2013 (PK/WT) sequence. Amino acid sequences common to the PK/WT sequence that differ from the B/Colorado/06/2017 (CO/WT) sequence are highlighted in blue, those common to CO/WT are highlighted in green; mutations resulting in an amino acid not present at levels greater than 50% in the wild-type viruses are shown in pink; and deletions gained through virus passage are highlighted in black. Amino acids associated with known antigenic regions are indicated by yellow bars.

Figure 2

Hemagglutinin inhibition titer for select escape viruses. HAI titers for Victoria lineage viruses (A) and Yamagata lineage viruses (B) are shown. Viruses are labeled in the legend as for Table 1 and Table 2. The data for the CO/WT and PK/WT viruses (from Table 1 and Table 2) are shown as the green and blue bars in panels A and B, respectively. Graph bars are colored to indicate the predominance of B/CO-like (green) or B/PK-like (blue) amino acids at positions highlighted in Figure 1. Red stripes indicate changes to the 120- and 150-loops; and red dots indicate the G141R change. Graph bars with a black outline indicate that the virus was generated using the sera indicated in that column.

Figure 3

Multistep growth curves for escape viruses. MDCK cells were infected with a low MOI (0.001) of virus and the number of plaque forming units (PFU) present in the supernatant determined at various timepoints. The PK/WT and CO/WT growth curves are shown in each panel using open squares and circles respectively. The growth curves for the escape viruses are displayed as filled symbols. Growth curves for viruses generated using sera from the 2003 vaccine recipients are shown in panels (A–C) and growth curves for viruses generated using sera from the 2011 vaccine recipients are shown in panels (D–F).