Pharmacodynamics of antibiotics with respect to bacterial killing of and release of lipoteichoic acid by Streptococcus pneumoniae.

OBJECTIVES There are marked differences in the amount of immunoreactive components such as lipoteichoic acid (LTA) released from Gram-positive bacteria following exposure to different antibiotics. Little is known about the kinetics and amount of release of such components in relation to bacterial killing. METHODS Bacterial killing and LTA release from Streptococcus pneumoniae type 3 during exposure to ceftriaxone, meropenem, rifampicin, rifabutin, quinupristin/dalfopristin, and trovafloxacin in tryptic soy broth were quantified microbiologically and by ELISA, respectively. We applied a mathematical model to characterize quantitatively the amount of lipoteichoic acid released and the statistical moments of this release. RESULTS The model approach revealed that (i) the lag time to release of LTA was very similar for individually killed bacterial cells (approximately 120 min), whatever the killing mechanism effected by the antibiotic, and (ii) the amount of LTA released per killed bacterial cell, a value that we regard as an indicator of the relation between antibacterial efficacy and possible adverse immunostimulatory effects due to release of cell wall components, differs markedly between antibiotics, even at antibiotic concentrations inducing equal killing. Rifamycins were most effective in killing S. pneumoniae while causing the least LTA release per killed bacterial cell; the amount released was about one-half that by quinupristin-dalfopristin and trovafloxacin, and one-quarter that by ceftriaxone and meropenem. CONCLUSIONS In the evaluation of antibacterial drugs, the present model provides useful information on the whole process of bacterial killing and release of immunoreactive components from the bacterial cell wall.