Contribution of P-glycoprotein to efflux of ramosetron, a 5-HT₃ receptor antagonist, across the blood-brain barrier

In-situ rat and mouse brain perfusion data indicated that the brain distribution of ramosetron (R-ramosetron), a 5-hydroxytryptamine₃ (5-HT₃) receptor antagonist, was extremely low compared with that expected from its lipophilicity. We hypothesized the involvement of an efflux system(s) and investigated the contribution of P-glycoprotein to efflux transport of ramosetron across the blood-brain barrier by means of an in-vitro uptake study in cell lines that over-express P-glycoprotein. We examined the contribution of mdr1a, mdr1b and MDR1 P-glycoprotein by using LV500 cells, MBEC4 cells and LL-GA5-COL300 cells, which over-express mdr1a P-glycoprotein, mdr1b P-glycoprotein and MDR1 P-glycoprotein, respectively. The uptake of [¹⁴C]ramosetron by LV500 cells and LLC-GA5-COL300 cells was significantly lower than that by the respective parental cells. Next, we studied the effects of P-glycoprotein inhibitors, verapamil and ciclosporin, on uptake of [¹⁴C] ramosetron by these cell lines. The uptake of [¹⁴C]ramosetron by LV500 cells and LLC-GA5-COL300 cells was significantly increased in the presence of verapamil or ciclosporin, while verapamil did not affect the uptake of [¹⁴C]ramosetron by MBEC4 cells. These results indicate that the efflux of [¹⁴C]ramosetron is partly mediated by mdr1a P-glycoprotein, but not by mdr1b P-glycoprotein, and that there is a difference in substrate specificity between mdr1a P-glycoprotein and mdr1b P-glycoprotein. Further, [¹⁴C]ramosetron was confirmed to be effluxed by human MDR1 P-glycoprotein. We conclude that the limited distribution of ramosetron to the brain is due, at least in part, to efflux mediated by the P-glycoprotein at the blood-brain barrier.