Background: Clarithromycin increases both hepatic and intestinal availability of the selective cytochrome P450 (CYP) 3A probe midazolam. This study was designed to identify determinants of variability in the extent of intestinal wall CYP3A inhibition by clarithromycin, such as CYP3A5 genotype, and the mechanism of inhibition.
Methods: Ten healthy volunteers received 500 mg oral clarithromycin twice a day for 7 days. Before and after administration of clarithromycin, small-bowel mucosal biopsy specimens were obtained endoscopically. Intestinal CYP3A activity was determined from the rate of 1'-hydroxymidazolam and 4-hydroxymidazolam formation by incubation of small-bowel homogenate with midazolam (25 micromol/L) and NADPH for 5 minutes. Intestinal CYP3A4 and CYP3A5 messenger ribonucleic acid was quantified by real-time reverse transcriptase-polymerase chain reaction. Intestinal CYP3A4 and CYP3A5 protein concentrations were determined by immunoblotting. Serum and homogenate concentrations of midazolam, clarithromycin, and metabolites were determined by liquid chromatography-mass spectrometry. CYP3A5 genotype was determined by real-time polymerase chain reaction.
Results: The formation of 1'-hydroxymidazolam (1.36 +/- 0.46 pmol . min(-1) . mg(-1) at baseline versus 0.35 +/- 0.16 pmol . min(-1) . mg(-1) after administration) and 4-hydroxymidazolam (0.39 +/- 0.12 pmol . min(-1) . mg(-1) at baseline versus 0.12 +/- 0.05 pmol . min(-1) . mg(-1) after administration) was significantly (P < .001) reduced after clarithromycin administration. Clarithromycin administration did not result in a significant change in intestinal CYP3A4 and CYP3A5 messenger ribonucleic acid and protein expression. All subjects had detectable serum clarithromycin concentrations after 7 days of clarithromycin (3.71 +/- 2.43 micromol/L). The mean concentration of clarithromycin in the intestinal biopsy homogenate was 1.2 +/- 0.7 nmol/L (range, 0.42-2.39 nmol/L). Compared with CYP3A5 nonexpressers, subjects with at least 1 CYP3A5*1 allele (CYP3A5 expressers) had greater inhibition of intestinal CYP3A activity after treatment with clarithromycin. There was a strong linear relationship between the decrease in intestinal CYP3A activity and baseline catalytic activity (R(2) = 0.9).
Conclusion: Baseline intestinal activity of CYP3A4 was a key determinant of variability of the inhibitory effect of clarithromycin among individuals. CYP3A5*1 alleles were associated with greater baseline intestinal CYP3A activity and, therefore, greater extent of inhibition. The primary in vivo mechanism was not rapidly reversible competitive or irreversible inhibition but was likely formation of metabolic intermediate complexes.