Bacterial plasmid transfer on surfaces : theoretical and experimental modelling approaches
The aim of this work was to develop and test a mathematical model for DNA transfer by conjugation which considered spatial separation of cells on surfaces. Model parameters include the initial donor and recipient numbers, maximum specific population growth rate, microcolony radial extension rate, maximum cell yield, maximum incubation time, standard deviation of the mean intercellular distance (σ), cellular area and area available for colonization. Sensitivity analysis indicated that growth areas and σ had the greatest effects on conjugation. The model was extended to consider the presence of a strain that did not conjugate and its effect on plasmid transfer. Filter matings with Pseudomonas fluorescens MON787 RP4 as the plasmid donor and P. fluorescens MON787 R lux as recipient were used to test the model. Model predictions were generally accurate but transconjugants were consistently underestimated. This was attributed to intercellular distances not following a strict Gaussian distribution. Nevertheless, predicted and experimental data were qualitatively similar, which increased confidence in the validity of the mechanisms proposed. Conjugation occurred over a wide range of cell densities, donor to recipient ratios, nutrient levels and incubation temperatures. Starved cells retained the capacity to conjugate, but plasmid transfer frequency was higher in the presence of nutrient. Above a minimum level, conjugation ability was not enhanced by nutrients. Temperature affected conjugation ability, the optimum being 20°C - 30°C. The presence of a nonconjugative strain decreased conjugation, by leading to earlier nutrient exhaustion and growth arrest which limited meetings between donor and recipient microcolonies. The model and the experimental system demonstrate the importance of spatial effects on conjugation. Description and prediction of gene transfer in natural environments will require models of greater complexity, and more sophisticated experimental testing, but this study provides a basis for theoretical descriptions of gene transfer in heterogeneous natural environments, such as soil and biofilms on solid surfaces in aquatic environments.