Parallel evolution between genomic segments of seasonal human influenza viruses reveals RNA-RNA relationships

Abstract

The influenza A virus (IAV) genome consists of eight negative-sense viral RNA (vRNA) segments that are selectively assembled into progeny virus particles through RNA-RNA interactions. To identify relationships between vRNA segments, we examined parallel evolution between vRNA segments of seasonal human IAV, finding that evolutionary relationships between vRNA segments differ between subtypes and antigenically-shifted strains. Intersegmental relationships were distinct between H3N2 and H1N1 viruses, but largely conserved over time in H3N2 viruses. However, parallel evolution of vRNA segments diverged between H1N1 strains isolated before and after the 2009 pandemic. Surprisingly, intersegmental relationships were not driven solely by protein sequence, which is potentially indicative of RNA-RNA driven coevolution. Colocalization of highly coevolved vRNA segments was enriched over other pairs at the nuclear periphery during a productive viral infection. This study illustrates how phylogenetics can be applied to interrogate putative RNA interactions underlying selective assembly of IAV.

Host origin is a determinant of parallel evolution between influenza virus gene segments

“…One plausible explanation for the apparent lack of convergence between gene trees in H9 viruses is that parallel evolution between gene segments is lineage specific. We previously reported a similar observation in H1N1 viruses isolated before and after the 2009 pandemic (21). Two geographically distinct H9N2 lineages have emerged from North America and Eurasia (8).…”

“…We used tree distance as a proxy for convergent evolution between genes, with two gene segments considered converging if the tree distance between them was low. The extent of convergence between pairs of genes varied widely, from robust convergence with tree distances approaching zero to no detectable parallel evolution (21). Parallel evolution between pairs of gene segments differed between viral subtypes but was relatively consistent over time in H3N2 viruses.…”

“…In this approach, representative sequences are selected from genomic trees (i.e., species trees) by a sequence similarity threshold (see Methods for additional details). This approach significantly improves gene tree reconstruction by pruning highly similar sequences that don’t resolve well (21). In this study, this approach offered the additional advantage of capturing similar degrees of diversity despite differing sequence availability in human and avian influenza sequences.…”

“…Following a modified version of our previously established workflow (21), gene trees were constructed for each of the eight gene segments of each set of viruses ( Figures S1-S2 and data not shown ). Evolutionary convergence between genes was determined by the clustering information distance (CID) between each pair of gene trees, where CID is inversely correlated to tree similarity and evolutionary convergence.…”

“…Similarly, packaging signals constrain evolution of the gene segments themselves (18)(19)(20). We recently demonstrated that such gene-driven constraints contribute to parallel evolution between gene segments in seasonal human influenza A viruses (21). However, whether evolutionary constraints imposed by interactions between viral genes or proteins extend to the evolution and emergence of avian influenza viruses remains unknown.…”