CYP3A4*16 and CYP3A4*18 alleles found in East Asians exhibit differential catalytic activities for seven CYP3A4 substrate drugs.

CYP3A4, the major form of cytochrome P450 (P450) expressed in the adult human liver, is involved in the metabolism of approximately 50% of commonly prescribed drugs. Several genetic polymorphisms in CYP3A4 are known to affect its catalytic activity and to contribute in part to interindividual differences in the pharmacokinetics and pharmacodynamics of CYP3A4 substrate drugs. In this study, catalytic activities of the two alleles found in East Asians, CYP3A4*16 (T185S) and CYP3A4*18 (L293P), were assessed using the following seven substrates: midazolam, carbamazepine, atorvastatin, paclitaxel, docetaxel, irinotecan, and terfenadine. The holoprotein levels of CYP3A4.16 and CYP3A4.18 were significantly higher and lower, respectively, than that of CYP3A4.1 when expressed in Sf21 insect cell microsomes together with human NADPH-P450 reductase. CYP3A4.16 exhibited intrinsic clearances (V(max)/K(m)) that were lowered considerably (by 84-60%) for metabolism of midazolam, carbamazepine, atorvastatin, paclitaxel, and irinotecan compared with CYP3A4.1 due to increased K(m) with or without decreased V(max) values, whereas no apparent decrease in intrinsic clearance was observed for docetaxel. On the other hand, K(m) values for CYP3A4.18 were comparable to those for CYP3A4.1 for all substrates except terfenadine; but V(max) values were lower for midazolam, paclitaxel, docetaxel, and irinotecan, resulting in partially reduced intrinsic clearance values (by 34-52%). These results demonstrated that the impacts of both alleles on CYP3A4 catalytic activities depend on the substrates used. Thus, to evaluate the influences of both alleles on the pharmacokinetics of CYP3A4-metabolized drugs and their drug-drug interactions, substrate drug-dependent characteristics should be considered for each drug.

Substrate-dependent functional alterations of seven CYP2C9 variants found in Japanese subjects.

CYP2C9 is a polymorphic enzyme that metabolizes a number of clinically important drugs. In this study, catalytic activities of seven alleles found in Japanese individuals, CYP2C9*3 (I359L), *13 (L90P), *26 (T130R), *28 (Q214L), *30 (A477T), *33 (R132Q), and *34 (R335Q), were assessed using three substrates (diclofenac, losartan, and glimepiride). When expressed in a baculovirus-insect cell system, the holo and total (apo and holo) CYP2C9 protein expression levels were similar among the wild type (CYP2C9.1) and six variants except for CYP2C9.13. A large part of CYP2C9.13 was present in the apo form P420. Compared with CYP2C9.1, all variants except for CYP2C9.34 exhibited substrate-dependent changes in K(m), V(max), and intrinsic clearance (V(max)/K(m)). For diclofenac 4'-hydroxylation, the intrinsic clearance was decreased markedly (by >80%) in CYP2C9.13, CYP2C9.30, and CYP2C9.33 and variably (63-76%) in CYP2C9.3, CYP2C9.26, and CYP2C9.28 due to increased K(m) and/or decreased V(max) values. For losartan oxidation, CYP2C9.13 and CYP2C9.28 showed 2.5- and 1.8-fold higher K(m) values, respectively, and all variants except for CYP2C9.34 showed >77% lower V(max) and intrinsic clearance values. For glimepiride hydroxylation, the K(m) of CYP2C9.13 was increased 7-fold, and the V(max) values of all variants significantly decreased, resulting in reductions in the intrinsic clearance by >80% in CYP2C9.3, CYP2C9.13, CYP2C9.26, and CYP2C9.33 and by 56 to 75% in CYP2C9.28 and CYP2C9.30. These findings suggest the necessity for careful administration of losartan and glimepiride to patients bearing these six alleles.

Effect of nonspecific binding to microsomes and metabolic elimination of buprenorphine on the inhibition of cytochrome P4502D6.

Recently, the pharmaceutical industry has employed the high-throughput method for the evaluation of cytochrome P450 (CYP) inhibition, using a combination of the heterologously expressed enzyme and a fluoregenic substrate. When buprenorphine (BN), a potent mixed agonist-antagonist analgesic, was evaluated by this method, it exhibited potent inhibition of CYP2D6 with an IC50 value of 0.25 microM in recombinant CYP2D6-expressing insect cell microsomes (rCYP2D6 microsomes). In contrast, the IC50 value was 22.7 microM in human liver microsomes (HLM) using a classical method. Although the substrate concentrations in each study were set to near the Km values, there was a large discrepancy in IC50 values. When we investigated the effect of nonspecific binding to microsomes on the inhibitory potency, with a view to clarifying this discrepancy, the unbound fraction in microsomes (fu,mic) was 0.06-0.21 and 0.99 in HLM and rCYP2D6 microsomes, respectively. The corrected IC50 value (1.74 microM) using free BN concentrations was much smaller than the uncorrected value. On the other hand, it was observed that the concentration of BN in HLM decreased rapidly due to metabolism by CYP3A4 while that in rCYP2D6 microsomes decreased only slightly. We then investigated the effect of incubation time on the inhibitory potency, since the rapid elimination of BN in HLM could have been a cause of the discrepancy. The IC50 value for BN was observed to decrease slightly from 22.7 to 17.1 microM, following the shortening of the incubation time from 10 to 2 min in HLM. We conclude that nonspecific binding to microsomes of the inhibitor could affect the inhibitory potency against CYP2D6. If this factor is considered, a more precise estimate of the risk of adverse drug interaction could be achieved.