Martin F. Fromm

The ATP‐binding cassette transporter P‐glycoprotein is now recognized as an important determinant for disposition of multiple drugs. The use of P‐glycoprotein‐expressing cell lines, the generation of P‐glycoprotein knockout mice as well as studies in animals and humans contributed to a better understanding on the role of active transport processes for drug disposition. P‐glycoprotein is located in tissues with excretory function such as intestine, liver and kidney. Moreover, due to its expression in important blood–tissue barriers (blood–brain and blood–testis barriers), in lymphocytes and in placenta it limits tissue penetration of its substrates. Induction and inhibition of P‐glycoprotein have now been identified as important underlying mechanisms of drug interactions in humans. Using selected examples, this review summarizes currently available data on the impact of P‐glycoprotein for bioavailability of drugs, drug interactions and drug effects.

Aims  The C3435T polymorphism in the human MDR1 gene is associated with lower intestinal P‐glycoprotein expression, reduced protein function in peripheral blood cells and higher plasma concentrations of the P‐glycoprotein substrate digoxin. Using fexofenadine, a known P‐glycoprotein substrate, the hypothesis was tested whether this polymorphism also affects the disposition of other drugs in humans.Methods  Ten Caucasian subjects homozygous for the wild‐type allele at position 3435 (CC) and 10 individuals homozygous for T at position 3435 participated in this study. A single oral dose of 180 mg fexofenadine HCl was administered. Plasma and urine concentrations of fexofenadine were measured up to 72 h using a sensitive LC/MS method. In addition, P‐glycoprotein function was assessed using efflux of the P‐glycoprotein substrate rhodamine 123 from CD56⁺ cells. Results  Fexofenadine plasma concentrations varied considerably among the study population. However, fexofenadine disposition was not significantly different between the CC and TT groups (e.g. AUC(0,∞) CC vs TT: 3567.1±1535.5 vs 3910.1±1894.8 ng ml⁻¹ h, NS; 95% CI on the difference −1364.9, 2050.9). In contrast, P‐glycoprotein function was significantly decreased in CD56⁺ cells of the TT compared with the CC group (rhodamine fluorescence CC vs TT: 45.6±7.2% vs 61.1±12.3%, P Conclusions  In spite of MDR1 genotype‐dependent differences in P‐glycoprotein function in peripheral blood cells, there was no association of the C3435T polymorphism with the disposition of the P‐glycoprotein substrate fexofenadine in this German Caucasian study population. These data indicate that other mechanisms including uptake transporter function are likely to play a role in fexofenadine disposition.

Objectives: Intestinal metabolism and transport are now recognized as protective barriers against orally ingested xenobiotics, including drugs. However, in vitro studies of the expression and function of intestinal proteins are hampered by the limited availability of human intestinal tissues. Because enterocytes are constantly shed in large numbers into the gut lumen, this study investigated whether these cells could be collected with a multilumen perfusion catheter and whether they are functionally active.Methods: In healthy volunteers, a 20‐cm isolated jejunal segment was generated with the perfusion catheter by inflating 2 balloons with air. Shed cells were characterized by fluorescence‐activated cell sorting analysis for leukocyte‐specific CD45 and enterocyte‐specific villin, as well as for apoptosis. Homogenates of the cells were used for reverse transcriptase polymerase chain reaction and Western blotting. Cytochrome P450 enzyme activity was determined with the calcium channel blocker verapamil as a substrate.Results: On average, 4.83 mg protein and 56.23 million cells were collected from a 20‐cm segment during 2 hours. A total of 84.2% of the cells were positive for enterocyte‐specific villin, and only 1.6% of the collected cells were positive for CD45. The majority of cells (65.3%) were not in early or late apoptosis or necrosis. In all volunteers, drug‐metabolizing enzymes (such as members of the cytochrome P450 family) could be detected as both messenger ribonucleic acid and proteins. Consistent with expression data, formation of verapamil metabolites catalyzed by CYP3A4 and CYP2C was shown.Conclusions: The majority of shed human enterocytes collected with a multilumen perfusion catheter were still functionally active and not apoptotic. Harvesting of spontaneously shed enterocytes provides a new tool for studies on expression and function of intestinal proteins.Clinical Pharmacology & Therapeutics (2002) 71, 131–140; doi: 10.1067/mcp.2002.121370

BACKGROUND: Therapy with racemic compounds produces effects that can be attributed to both (S)- and (R)-enantiomers. Here we have tested the hypothesis that an enantiomer-enantiomer interaction would modulate the effects of treatment with a racemate, the antiarrhythmic propafenone. Previous studies have shown that while the enantiomers of propafenone exert similar sodium channel-blocking (QRS widening) effects, it is the (S)-enantiomer that produces beta-blockade; moreover, we have demonstrated recently that (R)-propafenone inhibits the metabolism of (S)-propafenone in vitro. METHODS AND RESULTS: This single-blind, randomized study compared the effects of (R/S)-, (S)-, (R)-propafenone (150 mg q 6 hours for 4 days) and placebo on QRS duration (delta QRS) and on maximum exercise heart rate (delta HRmax), an index of beta-blockade. The clearance of (S)-propafenone was significantly lower (-55 +/- 24%, P < .001) during treatment with (R/S)-propafenone than with the (S)-enantiomer alone, and delta HRmax was significantly altered during (R/S)-propafenone (-8.8 +/- 6.6 beats per minute; P < .01) and during (S)-propafenone (-4.3 +/- 4.8 beats per minute; P < .01) but not during (R)-propafenone (-1.8 +/- 6.4 beats per minute) or placebo (0.3 +/- 7.1 beats per minute). In contrast, (R/S)-, (S)-, and (R)-propafenone all prolonged QRS compared with placebo. CONCLUSIONS: These data indicate that (R)-propafenone impairs the disposition of (S)-propafenone in humans. As a result, the beta-blocking effects of 150 mg of racemic propafenone (75 mg of the [S]-enantiomer) were more pronounced than those of 150 mg of (S)-propafenone alone. Thus, the effects of racemic drug therapy are not necessarily those predicted by summation of the effects of the individual enantiomers.