Drug penetration model of vinblastine-treated Caco-2 cultures

Abstract

The penetrability of new chemical entities (NCE) is routinely screened in preclinical drug research. Although Caco-2 is a well-established model for human absorption, the identification of P-glycoprotein (P-gp) substrates and therefore the predictive accuracy of this model is not always satisfactory. Vinblastine has been reported to affect P-gp expression in Caco-2 cells. Therefore, this study was intended to assess the effect of sustained vinblastine treatment on the expression of P-gp, using RT-PCR and Western blot techniques. The P-gp functionality was monitored in transport assay, and metabolic enzyme activities were studied using probe substrates. Completion of culture medium with vinblastine (10nM during both the growing and the differentiation period) increased the P-gp mRNA and the expression at protein level. These changes were associated with the sensitive and steady identification of P-gp substrates in the bidirectional transport assay. While the vinblastine-treated Caco-2 (VB-Caco-2) based model reliably identified the P-gp substrates, the native Caco-2 model failed to recognize 7 out of the 11 reference substrates. The penetrability of passively permeating compounds correlated strongly (r2=0.9830) in the two models as expected. Omitting vinblastine from established VB-Caco-2 cultures did not affect either the protein level or the functionality of P-gp. Vinblastine did not alter the CYP mediated activities of the cells either. The higher sensitivity of VB-Caco-2 culture is also supported by the test results of NCEs, where 37% of NCEs were found to be P-gp substrate in VB-Caco-2 verified by verapamil, but only 9% by native Caco-2.