Interactions of human P-glycoprotein transport substrates and inhibitors at the drug binding domain: Functional and molecular docking analyses

Rhodamine 123 (R123) transport substrate sensitizes P-glycoprotein (P-gp) to inhibition by compound 2c (cis-cis) N,N-bis(cyclohexanolamine)aryl ester isomer in a concentration-dependent manner in human MDR1-gene transfected mouse T-lymphoma L5178 cells as shown previously. By contrast, epirubicin (EPI) concentration changes left unaltered 2c IC50 values of EPI efflux. To clarify this discrepancy, defined molecular docking (DMD) analyses of 12 N,N-bis(cyclohexanolamine)aryl esters, the highly flexible aryl ester analog 4, and several P-gp substrate/non-substrate inhibitors were performed on human P-gp drug- or nucleotide-binding domains (DBD or NBD). DMD measurements yielded lowest binding energy (LBE, kcal/mol) values (mean ± SD) ranging from -11.8 ± 0.54 (valspodar) to -3.98 ± 0.01 (4). Lys234, Ser952 and Tyr953 residues formed H-bonds with most of the compounds. Only 2c docked also at ATP binding site (LBE value of -6.9 ± 0.30 kcal/mol). Inhibition of P-gp-mediated R123 efflux by 12 N,N-bis(cyclohexanolamine)aryl esters and 4 significantly correlated with LBE values. DMD analysis of EPI, (3)H-1EPI, (3)H-2EPI, (14)C-1EPI, (14)C-2EPI, R123 and 2c before and after previous docking of each of them indicated that pre-docking of either 2c or EPI significantly reduced LBE of both EPI and R123, and that of both (3)H-2EPI and (14)C-2EPI, respectively. Since the clusters of DBD amino acid residues interacting with EPI were different, if EPI docked alone or after pre-docking of EPI or 2c, the existence of alternative secondary binding site for EPI on P-gp is credible. In conclusion, 2c may allocate the drug-binding pocket and reduce strong binding of EPI and R123 in agreement with P-gp inhibition experiments, where 2c reduced efflux of EPI and R123.