The host phylogeny determines viral infectivity and replication across Staphylococcus host species

Author summaryVirus host shifts, where a virus jumps into a new host from another species, are a major source of emerging infectious diseases. Virus host shifts have been shown to be more likely between animals and plant species that are closely related. However, there is much to learn about how relatedness predicts susceptibility in bacterial hosts. Here, we used a panel of 64 Staphylococcus bacteria and a virus that infects them to investigate how the relationship between host species influences their susceptibility to infection. We find high variation in susceptibility to infection across the host panel and that most of that variation can be explained by the relationship between hosts. This effect is seen consistently using three methods for assessing susceptibility and can be seen both within and between species. Additionally, we find that the relationship between hosts allows us to predict the susceptibility of an unknown host with some accuracy, although there are instances where susceptibility cannot be predicted. Overall, this suggests that closely related hosts will show similar susceptibility to infection and that the relationship between bacterial hosts can be used to predict the susceptibility of an unknown host with limited success. Virus host shifts, where a virus transmits to and infects a novel host species, are a major source of emerging infectious disease. Genetic similarity between eukaryotic host species has been shown to be an important determinant of the outcome of virus host shifts, but it is unclear if this is the case for prokaryotes where anti-virus defences can be transmitted by horizontal gene transfer and evolve rapidly. Here, we measure the susceptibility of 64 strains of Staphylococcaceae bacteria (48 strains of Staphylococcus aureus and 16 non-S. aureus species spanning 2 genera) to the bacteriophage ISP, which is currently under investigation for use in phage therapy. Using three methods-plaque assays, optical density (OD) assays, and quantitative (q)PCR-we find that the host phylogeny explains a large proportion of the variation in susceptibility to ISP across the host panel. These patterns were consistent in models of only S. aureus strains and models with a single representative from each Staphylococcaceae species, suggesting that these phylogenetic effects are conserved both within and among host species. We find positive correlations between susceptibility assessed using OD and qPCR and variable correlations between plaque assays and either OD or qPCR, suggesting that plaque assays alone may be inadequate to assess host range. Furthermore, we demonstrate that the phylogenetic relationships between bacterial hosts can generally be used to predict the susceptibility of bacterial strains to phage infection when the susceptibility of closely related hosts is known, although this approach produced large prediction errors in multiple strains where phylogeny was uninformative. Together, our results demonstrate the ability of bacterial host evolutionary relatedness to explain differences in susceptibility to phage infection, with implications for the development of ISP both as a phage therapy treatment and as an experimental system for the study of virus host shifts.