Objectives: Aggregatibacter actinomycetemcomitans (Aa), a pathogen strongly associated with aggressive periodontitis, resists phagocytic killing by PMNs. If PMNs could be loaded with an antibiotic, phagocytic killing of Aa could potentially be enhanced. Clarithromycin is a macrolide that is active against Aa and is known to penetrate cells. We hypothesize that PMNs possess active transporters that allow them to accumulate clarithromycin and other macrolides. In this study, differentiated HL-60 cells were used as an in-vitro model of PMNs to examine the characteristics of granulocytic clarithromycin transport. Methods: HL-60 cells were cultured in RPMI media and differentiated into granulocytes by treatment with 1.3% DMSO for 7 days. [3H]-Clarithromycin transport by aliquots of suspended granulocytes was assayed at 37°C by measuring changes in cell-associated radioactivity. Intracellular volume was determined with [3H]-water. Lineweaver-Burk analysis was used to estimate the Michaelis constant (Km) and maximum velocity (Vmax) of clarithromycin transport. Results: Clarithromycin transport by granulocytes exhibited Michaelis-Menten kinetics and was temperature-sensitive. The observed Km and Vmax values were 249.8 ug/ml and 473.5 ng/minute/10^6 cells, respectively. Transport activity exhibited a linear relationship with temperature between 0 and 37°C and reached saturation within 15 minutes. At steady-state in 37°C medium containing 2 ug/ml clarithromycin, the intracellular concentration was 30 ug/ml. When extracellular clarithromycin was decreased to 0.2 ug/ml, clarithromycin efflux from loaded cells was observed. The rate of efflux was slow, and intracellular clarithromycin content was reduced by 50% over a 60-minute period. Conclusions: Granulocytes possess an active transporter for clarithromycin. Intracellular accumulation of this agent could potentially augment phagocytic killing of Aa. Moreover, it could facilitate targeted delivery of clarithromycin by PMNs as they leave the vasculature and migrate to infection sites. This work was supported by NIDCR grants DE12601 and DE14320. |