Together, these findings uncover the co-evolutionary dynamics governing DelVG behavior within influenza populations and lay the groundwork for engineering DVGs with optimized interference properties for antiviral applications.

Using influenza A virus (PR8) as a model system, we showed that a single point mutation within a bona fide DelVG can modulate its interfering activity and reveal antagonistic pleiotropy, reflecting distinct fitness optima between DelVG and WT genomes.

Utilizing various assays, we demonstrated that DI250:A94G replicates and interferes with the WT virus more efficiently, while exhibiting no detectable impact on packaging (Fig 4 and 5).

Interestingly, the same A94G change that benefited the DelVG was fatal when introduced into the WT virus (Fig 4)

This led us to start analyzing the mutation profiles of both the wild-type (WT) and the selected DelVGs. We found that DI250, and not the WT, acquired a point mutation (A94G) around passage 15, which continued to accumulate until its selection at later passages (Fig 4).

We rebuilt that dominant DelVG (called DI250) and compared it to others that had been outcompeted earlier. Surprisingly, in controlled competition experiments, DI250 didn’t win again (Fig 3).

We observed distinct evolutionary patterns across the viral segments, with strong evidence of DelVG selection in the PB2 segment, where one DelVG progressively dominated the population at later passages (Fig 2).

We serially passaged influenza virus 72 times and used deep sequencing to track how deletion-containing viral genomes (DelVGs) evolved over time. Hundreds of DelVGs emerged across all gene segments but as passages went on, their diversity narrowed (Fig 1).