WISH-barcoding of Salmonella Typhimurium ATCC14028s strains for population dynamics studies in vivo

This is the authors' abstract. We don't add key points for this paper.

Background Barcoding of isogenic strains is a powerful approach to assess pathogen population dynamics during infection. Here, we adapted WISH-barcoding to Salmonella Typhimurium ATCC14028s to evaluate its suitability for pooled infection experiments in streptomycin-pretreated mouse models. Results WISH-barcoded wild-type pools showed pronounced population instability, characterized by reproducible strain loss and segregation into high- and low-fitness subpopulations. Whole-genome sequencing identified recurrent mutations in the methyltransferase gene rsmG and loss of the P3 plasmid carrying streptomycin resistance in low-fitness strains; neither was observed in Δ invG or Δ ssaV pools. We propose that rsmG mutations were enriched during strain construction carried out under streptomycin selection, following loss of the P3 plasmid. Control experiments demonstrated that rsmG mutations and P3 loss are counter-selected in vivo and attenuate gut-luminal colonization in streptomycin-pretreated mice. Conclusion While population dynamics experiments with ATCC14028s are feasible in principle, wild-type strains are prone to acquiring fitness-altering mutations during in vitro construction when using the P3 plasmid and streptomycin, highlighting the need for careful pool validation prior to use and for minimizing antibiotic exposure during strain construction when the strain of interest has no natural resistance, to reduce the risk of selecting mutations.