Plasma protein binding of fluoroquinolones affects antimicrobial activity

Abstract

OBJECTIVES: In contrast to most antimicrobial classes, there is a doubt about the impact of protein binding (PB) on the antimicrobial activity of fluoroquinolones. We set out to evaluate the suitability of previously used models for investigating the influence of PB on bacterial killing by fluoroquinolones. METHODS: PB of moxifloxacin and trovafloxacin was determined in Mueller-Hinton broth (MHB) containing different concentrations of human serum or albumin. Bacterial growth curves of Staphylococcus aureus and Pseudomonas aeruginosa were determined in pure serum, pure MHB and MHB containing different amounts of serum or albumin. Killing of both strains at concentrations equal to the MIC was investigated for moxifloxacin and trovafloxacin in MHB and also in medium that showed PB values identical to those of pure serum. RESULTS: Frequently used media for investigating the impact of PB, i.e. MHB containing 20% to 70% serum or 4% albumin, did not reach the level of PB achieved in pure serum or significantly hampered bacterial growth compared with pure MHB. PB in MHB containing 12% albumin was identical to that in pure serum but did not impair bacterial growth. Addition of 12% albumin significantly reduced bacterial killing by both fluoroquinolones compared with that found in pure MHB. CONCLUSIONS: For fluoroquinolones, standard media might be insufficient to investigate the impact of PB on bacterial killing. MHB containing 12% albumin seems to be a promising medium in this context. For moxifloxacin and trovafloxacin, PB leads to significant reduction of antimicrobial activity.