The population dynamics of plasmid-mediated antibiotic resistance in salmonella typhimurium in chickens
A model of growth and plasmid transfer between strains of Escherichia coli and Salmonella typhimurium was developed with reference to the literature. This was the organising principle for the collection of a complete set of in vitro life history parameters of one S. typhimurium and one E. coli strain. In the course of estimating these parameters two results of note were obtained. Fits of the Lotka-Volterra competition model were obtained for data on S. typhimuiurm growing in competition with E. coli. The first noteworthy discovery was the failure of this model to account for several characteristics of growth of these strains under competition. The growth rates of plasmid-bearing and plasmid-free strains were obtained. The second main result came from examination of the results of the growth rate data, which revealed that the cost to S. typhimuiurm 576 of bearing the resistance plasmid was low (4%). The model was also used to simulate the effect of antibiotic dose on the density of the donor, recipient and transconjugant populations over time. These simulations predicted that there would be a convex relationship between antibiotic dose and transconjugant density (i.e. that the density would first rise, then fall, with increasing dose). Following from this result, laboratory experiments and in vivo experiments in chickens were directed towards obtaining information on the relationship between these two variables. This convex relationship was not demonstrated within a single experiment, although some experimental environments produced an increase in transconjugant density with dose, and others, a decrease. Few transconjugants were formed in vivo. In order to investigate the low cost of resistance and low rate of in vivo transconjugant production, cost of resistance and plasmid transfer rate of this plasmid in several strain combinations of E. coli and S. typhimuiurm was evaluated.