Efflux transport of a new quinolone antibacterial agent, HSR-903, across the blood-brain barrier.

The distribution of HSR-903, a new quinolone antibacterial agent, to the brain after i.v. administration to rats was low compared with that to other tissues. The blood-brain barrier permeability to HSR-903 determined by the brain perfusion method was low, and increased nonlinearly with increasing concentration of HSR-903 in the perfusate. When the brain-to-plasma concentration ratio (Kp, brain) was measured in mdr1a gene-knockout mice, the value was 8 times higher than that in normal mice. The uptake of [14C]HSR-903 by multidrug-resistant K562/ADM cells, which express P-glycoprotein (P-gp), was significantly lower than that by the drug-sensitive parent K562 cells. In addition, the uptake of [14C]HSR-903 by K562/ADM cells was significantly increased in the presence of cyclosporin A and ATP-depleting agents. These observations support the idea that P-gp participates in HSR-903 efflux from the brain. The steady-state uptake of HSR-903 by a monolayer of primary cultured bovine brain capillary endothelial cells was increased in the presence of several quinolone antibacterial agents or anionic compounds, such as 4,4'-diisothiocyanatostilbene-2,2'-disulfonic acid, and in bicarbonate ion-free medium, as well as by P-gp inhibitors (cyclosporin A and quinidine). These results suggested that the efflux of HSR-903 proceeds at least partly via an anion-sensitive efflux transport mechanism as well as via P-gp. In conclusion, the low brain distribution of the new quinolone antibacterial agent HSR-903 can be ascribed to multiple efflux mechanisms including P-gp and an unidentified anion-sensitive transporter operating in the brain capillary endothelial cells that constitute the blood-brain barrier.