Abstract
Some macrolide antibiotics cause clinical drug interactions, resulting in altered metabolism of concomitantly administered drugs, via formation of an inactive cytochrome P-450 complex. In the present study, the formation of a cytochrome P-450 type I binding spectrum and a metabolic intermediate complex by troleandomycin and dirithromycin was assessed in liver microsomes obtained from untreated rats and phenobarbital- or dexamethasone-pretreated rats. Troleandomycin produced a type I binding spectrum and metabolic intermediate complex in microsomes from dexamethasone- and phenobarbital-pretreated rats. Dirithromycin did not produce a detectable type I binding spectrum but formed a small cytochrome P-450 metabolic intermediate complex (6% of that formed by troleandomycin) in microsomes from dexamethasone-pretreated rats only. The formation of a cytochrome P-450 type I binding spectrum and a metabolic intermediate complex by troleandomycin, erythromycin, dirithromycin, and erythromycylamine was also assessed in microsomes prepared from human livers. Troleandomycin and erythromycin formed a type I binding spectrum and a metabolic intermediate complex which were larger in microsomes from subjects on barbiturate therapy than in microsomes from subjects with no recent barbiturate exposure. Erythromycylamine did not form a detectable type I binding spectrum with any of the human microsomal samples, but a small metabolic intermediate complex was formed with microsomes from a subject on phenobarbital, phenytoin, and propranolol therapy. Dirithromycin did not form a detectable type I binding spectrum or a metabolic intermediate complex in any human liver sample. Preclinical quantitation of the human metabolic intermediate complex may be helpful in predicting the possibility of clinical drug interactions of new drug candidates.