Many P-glycoprotein (P-gp) inhibitors studied in vitro and in vivo are also
known or suspected to be substrates and/or inhibitors of cytochrome P-450
3A (CYP3A). Such overlap raises the question of whether CYP3A inhibition is
an intrinsic characteristic of P-gp inhibitors, a matter of concern in the
development and rational use of such agents. Thus, the purpose of the pres
ent study was to determine whether the ability to inhibit P-gp and CYP3A is
, in fact, linked and whether specific P-gp inhibitors with limited ability
to inhibit CYP3A can be identified. Therefore, the potency of a series of
14 P-gp inhibitors was assessed by measuring their inhibition of the transe
pithelial flux across Caco-2 cells of digoxin, a prototypical P-gp substrat
e CYP3A inhibition was determined from the impairment of nifedipine oxidati
on by human liver microsomes. Determination of the apparent K-i values for
CYP3A inhibition and the IC(50)s for P-gp and CYP3A inhibition allowed comp
arison of the relative inhibitory potency of the compounds on the two prote
ins' function. The IC(50)s for P-gp inhibition ranged from 0.04 to 3.8 mu M
. All compounds inhibited CYP3A with apparent K-i values of between 0.3 and
76 mu M and IC(50)s between 1.5 and 50 mu M. However, no correlation was f
ound between the extent of P-gp inhibition and CYP3A inhibition, and the ra
tio of the IC50 for CYP3A inhibition to the IC50 fur P-gp inhibition varied
from 1.1 to 125. These results demonstrate that, although many P-gp inhibi
tors are potent inhibitors of CYP3A, a varying degree of selectivity is pre
sent. The development and use of P-gp inhibitors with minimal or absent CYP
3A inhibitory effects should decrease the impact of drug interactions on th
e therapeutic use of such compounds.