There is increasing interest in phage therapy as an alternative to antibiotics for treating bacterial infections, especially using phages that select for evolutionary trade-offs between increased phage resistance and decreased fitness traits such as virulence in target bacteria. A vast repertoire of virulence factors allows the opportunistic bacterial pathogen,
Shigella flexneri
, to invade human gut epithelial cells, replicate intracellularly, and evade host immunity through intercellular spread. It is previously shown that OmpA is necessary for intercellular spread of
S. flexneri
. We hypothesized that a phage which uses OmpA as a receptor to infect
S. flexneri
, should select for phage-resistant mutants with attenuated intercellular spread. Here we show that phage A1-1, requires OmpA as a receptor and selects for reduced virulence in
S. flexneri
. We characterized five phage-resistant mutants by measuring phenotypic changes in various traits: cell-membrane permeability, total lipopolysaccharide (LPS), sensitivity to antibiotics, and susceptibility to other phages. Results separated the mutants into two groups: R1 and R2 phenotypically resembled
ompA
knockouts, whereas R3, R4 and R5 were similar to LPS-deficient strains. Whole genome sequencing confirmed that R1 and R2 had mutations in
ompA
, while R3, R4 and R5 showed mutations in LPS inner-core biosynthesis genes
gmhA
and
gmhC
. Bacterial plaque assays confirmed that all phage-resistant mutants were incapable of intercellular spread. We concluded that selection for
S. flexneri
resistance to phage A1-1 generally reduced virulence (i.e. intercellular spread), but this trade-off could be mediated either by mutations in
ompA
or in LPS-core genes that likely altered OmpA conformation.
Author Summary
Shigella flexneri
is a facultative intracellular pathogen of humans, and a leading cause of bacillary dysentery. With few effective treatments and rising antibiotic resistance in these bacteria, there is increasing interest in alternatives to classical infection management of
S. flexneri
infections. Phage therapy poses an attractive alternative, particularly if a therapeutic phage can be found that results in an evolutionary trade-off between phage resistance and bacterial virulence. Here, we isolate a novel lytic phage from water collected in Cuatro Cienegas, Mexico that uses the OmpA porin of S. flexneri as a receptor. We use phenotypic assays and genome sequencing to show that phage A1-1 selects for phage-resistant mutants that can be grouped into two categories: OmpA-deficient mutants and LPS-deficient mutants. Despite these underlying mechanistic differences, we confirmed that naturally-occurring phage A1-1 selected for evolved phage resistance that coincided with impaired intercellular spread of
S. flexneri
in a eukaryotic infection model.
Experimental evolution and bacterial resistance: (co)evolutionary costs and trade-offs as opportunities in phage therapy research
