Role of P–Glycoprotein in Restricting in vivo Ocular Absorption of Erythromycin in Rabbit Cornea

Abstract

Abstract: : Purpose: The purpose of this work was to determine if P–glycoprotein restricts in vivo absorption of erythromycin (P–glycoprotein substrate) in rabbit cornea Method: New Zealand Albino (NZW) rabbits were used for all in vivo experiments. Anesthetized rabbits were given constant topical infusions of [¹⁴C] erythromycin with and without inhibitors. Testosterone, verapamil and quinidine were used as inhibitors. Dose dependant inhibition was performed with different concentrations of testosterone. MDCKII–MDR1 cells were used for in vitro studies. Also cultured primary rabbit corneal epithelial cells (rPCEC) and a corneal cell line (SIRC) were used to demonstrate the existence of P–gp. Results: All the inhibitors inhibited P–glycoprotein efflux pump. Maximum inhibition was observed with 500 µM of testosterone. AUC_{0–<font face="symbol">¥</font>} with 500 µM of testosterone was 3 times higher than AUC_{0–<font face="symbol">¥</font>} without inhibitor. Rate of elimination (k_el) for [¹⁴C] erythromycin and [¹⁴C] erythromycin+ 500 µM of testosterone was found to be similar (141 ± 23 min) suggesting that the path of elimination was the same. Rate of elimination of mannitol (a paracellular uptake marker) and diazepam (a transcellular marker) was found to be 76 mins and 120 mins respectively. C_max was increased 5 fold in the presence of inhibitor. Verapamil and quinidine also inhibited the efflux pump with moderate increase in AUC_{0–<font face="symbol">¥</font>} and C_max compared to control. Uptake of Rho–123 (a P–gp substrate) across both SIRC and rPCEC was time and temperature dependent. Rho–123 uptake in SIRC was 14.4 pmoles/mg protein and in the presence of 10 µM of cyclosporine A (potent P–gp inhibitor) was 70.8 pmoles/mg protein at 3 hrs. Western blot analysis indicated a 170 kDa band confirming the presence of P–gp. Human cornea was also checked for the presence of P–gp. RT–PCR data indicated one single band which was subcloned and sequenced to confirm the presence of P–gp. The protein sequence deduced from the fragment product indicated 89% homology with human MDR1. Conclusion: Functional and molecular characterization proves the existence of P–gp in human cornea, rabbit cornea and the cell lines. Furthermore, the efflux pump was active in vivo and it restricted erythromycin absorption into the aqueous humor which was inhibited by known P–gp inhibitors. This knowledge of existence of P–gp will help in better ocular drug delivery strategies.

Keywords: cornea: epithelium • anterior chamber • pharmacology