Resistance of pig pathogens and commensals to antimicrobial drugs: mechanisms and avoidance

This project will provide an ideal training for a veterinary graduate in in vitro molecular bacteriology, mathematical modelling and in vivo experimental skills. The PK/PD modelling techniques used will provide ideal training for a veterinary scientist to enter either academia or the pharmaceutical industry, where these skills are desperately needed in research which predicts drug dosing regimens which are both optimally efficacious and least likely to induce resistance in clinical use of antimicrobial drug products. Hypothesis The emergence and spread of resistance to antimicrobial drugs in porcine bacterial pathogens can be minimized by establishing dosage schedules that optimize bactericidal effects and by determining the mechanisms through which resistance arises. Aims Two key pathogens (A. pleuropneumonia, P. multocida) and the g.i.t. commensal potential pathogen (E. coli) and 4 drugs will be studied. The project will: (a) establish by PK-PD modeling methods, concentrations of each antimicrobial drug required to achieve defined levels of bacteriostatic and bactericidal effects and eradication of bacteria; (b) identify resistance and define the molecular mechanisms involved; and (c) design dosage schedules based on the derived PK-PD parameters and bacterial drug resistance findings. Investigational Plan In years 1 and 2, 6 porcine strains of each of the organisms, A. pleuropneumoniae, P. multocida and E. coli will be obtained from VLA centres. For each strain, in two matrices [Mueller Hinton Broth (MHB) and pig serum], MIC, MBC and 24-hour time-kill curves will be established in vitro. Serum will be used in addition to the conventional artificial medium (MHB), as it is a more therapeutically relevant matrix than MHB. Data will be obtained for 4 drugs, florfenicol, marbofloxacin, oxytetracycline and tiamulin, each representing a drug class with differing pharmacodynamic properties, using several multiples ranging from 0.5 to 8.0 x MIC. The time-kill data will be subjected to PK-PD modeling programmes to elucidate AUC/MIC values required for bacteriostatic, bactericidal and eradication responses. From these data dosages will be determined to provide bactericidal and eradication effects. Resistance of the selected bacteria to the antimicrobial drugs will be induced in vitro with 2-10 passages . In order to associate drug resistance with doses, a range of drug concentrations will be used for induction. The development of resistance will be identified by MIC tests and determination of the resistant genes. When the resistance is established, the in vitro experiments will be carried out using the drugs at MIC and MBC concentrations and the bacteria carrying or not carrying resistant genes. Bacteria will be isolated and their DNAs or total RNAs extracted for determination of resistant genes using PCR and qPCR technology. The proposed resistant genes are FlorA for florfenicol, a group of tet and otr (such as A-E, G and Y) for oxytetracycline; gyrA, parC and acrAB for marbofloxacin and cfr methyltransferase for tiamulin. In years 3 and 4 investigations will be conducted in vivo in a porcine model of enzootic pleuropneumonia currently in use in our laboratory. Based on the findings relating to PK/PD modelling and resistance pattern monitored in years 1 and 2, two drugs will be selected. Each will be administered in the pleuropneumonia model. Appropriate clinical, bacteriological, pathological and pharmacokinetic measurements will be made. These will include: (a) establishing plasma concentration-time profiles; and (b) harvesting organisms from the lung at post mortem and from faeces (E. coli) prior to induction of disease and at post mortem to identify any reduced sensitivity to the drugs and the molecular mechanism of the resistance, for comparison with in vitro studies. Data will be used to calculate optimal dosing regimens for clinical use.