Analysis of Growth Dynamics in a Three-Species Chemostat by Disturbing Coexistence with Antibiotic Pulse

From clinical data is known that treatment with antibiotics against Staphylococcus aureus, a bacterium that occurs in the first years in the life of CF-patients, leads to an earlier infection with the more pathogenic Pseudomonas aeruginosa [1]. The reasons for this effect are unknown. In our previous studies, a chemostat culture with three bacterial species relevant for cystic fibrosis (CF): P. aeruginosa, S. aureus and Burkholderia cepacia, was set up as a model system for microbial communities occurring in the lung of CF patients. In chemostat experiments with stationary conditions the coexistence of at least two of the species was observed. A quantitative T-RFLP method was adapted to monitor the specific cell number in mixed cultures of P. aeruginosa, S. aureus and B. cepacia [2]. The experimental work was supported by mathematical modelling, indicating a possible cause of coexistence on the basis of a food chain [3]. Based on this work and motivated by antibiotic therapies of CF patients, the dynamics of the mixed chemostat culture by setting antibiotic pulses will be investigated in the study presented here. The results from the model system will be compared to antibiotic pulse experiments in pure culture. This should lead to a better understanding of the behaviour of microbial communities on antibiotic treatment in contrast to pure cultures which are routinely used in antibiotic screening. It is expected that useful hints for the effect observed in antibiotic therapies of CF related infections can be derived from the results as well. First results of experiments are presented including establishment of quantification methods. Ceftazidim, the applied antibiotic, will be determined in culture broth by reversed phase HPLC. The susceptibility of ceftazidim to the different bacterial species will be tested, using a minimum inhibition concentration (MIC) assay. The adapted quantitative T-RFLP analyses will be further optimized in a way that intact but dead cells are separated from intact but viable cells and therefore do not appear in the calculation of the cell number. [1] Ratjen F., Döring G.: Lancet (2003), 361: 681-689 [2] Schmidt J.K., et al.: Biotechnology and Bioengineering, accepted Aug. 2006 [3] Heßeler J., et al.: Journal of Mathematical Biology (2006): 53: 556-58