NEW PHARMACOKINETIC IN-VITRO MODEL FOR STUDIES OF ANTIBIOTIC-ACTIVITYAGAINST INTRACELLULAR MICROORGANISMS

The capacity for intracellular growth is an important survival strateg y for a large group of common pathogens. Helicobacter pylori, the etio logical agent for gastritis and duodenal ulcer, has been shown by both in vivo and in vitro studies to have the capacity to invade epithelia l cells. In vitro models are used to study the effect of antibiotics o n microorganisms. Most investigations are performed in broth culture o r on agar plates, but kinetic models for bacteria in broth have been d escribed. We present a new, kinetic model adapted for intracellular pa thogens. A glass chamber, with a metal rack fitting Falcon cell cultur e inserts, was connected to a pump by rubber tubes. Different tube dia meters and pump speeds were evaluated, and the assay was designed to m imic the half-lives of the antibiotics in vivo, i.e., 11.5 h for azith romycin, 5 h for clarithromycin, and 1 h for amoxicillin. Monolayers o f HEp-2 cells were grown in the inserts for 2 days, after which H. pyl ori (clinical strain 88-23), was added to the system. Internalization was allowed for 12 h, and extracellular H. pylori cells were eradicate d with gentamicin. The inserts were moved to the glass chamber, contai ning medium with 12.5 mg of either amoxicillin or azithromycin per lit er or 2.4 mg of clarithromycin per liter. This represents 12.5, 50, an d 80 times the extracellular minimum bactericidal concentration value, respectively. Samples were taken at 0, 2, 4, 6, 8, and 24 h. The HEp- 2 cells were lysed, and intracellular bacteria were counted by plating . Inserts with infected cells grown in drug-free medium were included as controls for each time interval. A 3-log(10) reduction of H. pylori was achieved in the experiments with azithromycin, and a 4-log(10) re duction was achieved in the clarithromycin experiments, while no intra cellular effect was seen when amoxicillin was used. The antibiotic con centrations at the sampling intervals were 12.5, 3.1, 0.8, 0.2, 0.05, and 0 mg/liter for amoxicillin; 12.5, 11.5, 10, 9, 8, and 3 mg/liter f or azithromycin; and 2.4, 1.8, 1.4, 1, 0.8, and 0 mg/liter for clarith romycin. This new model for pharmacokinetic studies provides a useful tool, with applications for a broad range of microorganisms.