Cellular accumulation of fluoroquinolones is not predictive of their intracellular activity: studies with gemifloxacin, moxifloxacin and ciprofloxacin in a pharmacokinetic/pharmacodynamic model of uninfected and infected macrophages.

Fluoroquinolones enter eukaryotic cells but the correlation between cellular accumulation and activity remains poorly established. Gemifloxacin is known to accumulate to a larger extent than most other fluoroquinolones in tissues. Using murine J774 macrophages and human THP-1 monocytes, we show that gemifloxacin accumulates more than ciprofloxacin and even moxifloxacin. Whilst showing indistinguishable kinetics of accumulation in J774 macrophages, gemifloxacin was released at an approximately two-fold slower rate than ciprofloxacin and its release was only partial. Gemifloxacin was also a weaker substrate than ciprofloxacin for the efflux transporter Mrp4 active in J774 macrophages. In cells infected with Listeria monocytogenes or Staphylococcus aureus (typical cytoplasmic and phagolysosomal organisms, respectively), gemifloxacin was equipotent to moxifloxacin and ciprofloxacin in concentration-dependent experiments if data are normalised based on the minimum inhibitory concentration (MIC) in broth. Thus, larger cellular concentrations of gemifloxacin than of moxifloxacin or ciprofloxacin were needed to obtain a similar target effect. Fractionation studies showed a similar subcellular distribution for all three fluoroquinolones, with approximately two-thirds of the cell-associated drug recovered in the soluble fraction (cytosol). These data suggest that cellular accumulation of fluoroquinolones is largely a self-defeating process as far as activity is concerned, with the intracellular drug made inactive in proportion to its accumulation level. Whilst these observations do not decrease the intrinsic value of fluoroquinolones for the treatment of intracellular infections, they indicate that ranking fluoroquinolones based on cell accumulation data without measuring the corresponding intracellular activity may lead to incorrect conclusions regarding their real potential.