Pharmacokinetic drug-drug interactions often occur at the level of P-glycoprotein (Pgp). To study possible interactions caused by the newer antidepressants we investigated citalopram, fluoxetine, fluvoxamine, paroxetine, reboxetine, sertraline, and venlafaxine and their major metabolites desmethylcitalopram, norfluoxetine, paroxetine-metabolite (paroxetine-M), desmethylsertraline, N-desmethylvenlafaxine, and O-desmethylvenlafaxine for their ability to inhibit Pgp. Pgp inhibition was studied by a fluorometric assay using calcein-acetoxymethylester as Pgp substrate and two different cell systems: L-MDR1 cells (model for human Pgp) and primary porcine brain capillary endothelial cells (pBCECs, model for the blood-brain barrier). Both cell systems proved to be suitable for the evaluation of Pgp inhibitory potency of drugs. All antidepressants tested except O-desmethylvenlafaxine showed Pgp inhibitory activity with sertraline, desmethylsertraline, and paroxetine being the most potent, comparable with the well known Pgp inhibitor quinidine. In L-MDR1 cells fluoxetine, norfluoxetine, fluvoxamine, reboxetine, and paroxetine-M revealed intermediate Pgp inhibition and citalopram, desmethylcitalopram, venlafaxine, and N-desmethylvenlafaxine were only weak inhibitors. The ranking order was similar in pBCECs. The fact that some of the compounds tested exert Pgp inhibitor effects at similar concentrations as quinidine suggests that pharmacokinetic drug-drug interactions between the newer antidepressants and Pgp substrates should now be thoroughly studied in vivo.