The role of polymorphic CYP2B6 in cyclophosphamide (CPA) bioactivation was investigated in human liver microsomes. A total of 67 human liver specimens were first genotyped with respect to the CYP2B6*5 and CYP2B6*6 variant alleles. CYP2B6 apoprotein levels in 55 liver microsomal preparations were assessed by immunoblotting. 4-Hydroxy-CPA and hydroxy-bupropion were quantified by using HPLC and LC-MS, respectively. 7-Ethoxy-4-trifluoromethyl coumarin O-deethylase activity was measured fluorometrically. The frequencies of CYP2B6*5 and CYP2B6*6 mutant alleles were 9.0 and 16.4%, respectively. CYP2B6 protein expression was detected in 80% of the samples, with a large variation (0.003-2.234, arbitrary units). There was a high correlation between CYP2B6 apoprotein content and CPA 4-hydroxylation (n ¼ 55, r ¼ 0.81, Po0.0001). When based on the CYP2B6 apoprotein levels, the *6 carriers had significantly higher CPA 4-hydroxylation (Po0.05). CPA 4-hydroxylation also correlated significantly with other CYP2B6-specific reactions (n ¼ 20, Po0.0001). V max and K m for CPA 4-hydroxylation in recombinant CYP2B6 enzyme were 338 nmol/min/nmol enzyme and 1.4 mM, respectively. CYP2B6 showed much higher in vitro intrinsic clearance than previously observed in recombinant CYP2C19 and CYP2C9 variants in yeast expression system. Our results demonstrate that the polymorphic CYP2B6 is a major enzyme in the bioactivation of CPA. Moreover, we identified a strong impact of CYP2B6*6 on CPA 4-hydroxylation.

The CYP2C9*3 allele was shown to be associated with decreased formation of E-3174 from losartan. The significant differences between genotypes in plasma and urine losartan/E-3174 ratios and the good correlation between the plasma and urine ratios suggest that the losartan/E-3174 ratio in 0- to 8-hour urine specimens may serve as a phenotyping assay for CYP2C9 activity. Further studies in larger populations will be required to establish this.