Xenobiotica, 2010; 40(1): 1–8
RESEARCH ARTICLE
Effects of organic solvents on the time-dependent inhibition of CYP3A4 by diazepam Y. Nishiya1, K. Nakamura1, N. Okudaira1, K. Abe1, N. Kobayashi1, and O. Okazaki1 Xenobiotica Downloaded from informahealthcare.com by Sankyo Co. Ltd (Japan) Active on 03/21/11 For personal use only.
Drug Metabolism & Pharmacokinetics Research Laboratories, Daiichi Sankyo Co., Ltd, Tokyo, Japan
Abstract 1. The effects of organic solvents, acetonitrile, dimethyl sulfoxide (DMSO), and methanol, which are used to dissolve lipophilic test compounds and cytochrome P450 (P450) substrates, and carried into preincubation at 1% (v/v), on time-dependent inhibition of CYP3A4 by diazepam, were evaluated using human liver microsomes (HLM) and recombinant human P450 expressed microsomes (rCYPs). 2. The inactivation kinetics of CYP3A4 by diazepam dissolved in acetonitrile and methanol were almost equal with kinact/KI values, 0.095 and 0.15 min−1 mM−1 for HLM and 1.1 and 1.4 min−1 mM−1 for rCYP3A4, respectively. In contrast, the inactivation by diazepam dissolved in 1% DMSO significantly decreased and the kinetic parameter could not be calculated. 3. The formation rate of nordiazepam and temazepam metabolized from diazepam dissolved in DMSO were approximately half of those using substrate dissolved in acetonitrile and methanol in both HLM and rCYP3A4. Dixon plots revealed that the metabolism of diazepam in rCYP3A4 were inhibited by DMSO in a competitive or mixed-type manner with Ki (inhibition constant) values of 6 and 24 mM for nordiazepam and temazepam, respectively. 4. In conclusion, the time-dependent inhibition of CYP3A4 by diazepam was attenuated by DMSO, while acetonitrile and methanol had no effect. The metabolite formation profile under the conditions tested suggested that DMSO competitively inhibit the formation of the reactive metabolites of diazepam by CYP3A4. The effect of organic solvents should be taken into consideration when evaluating the in vitro time-dependent inhibition of new chemical entities. Keywords: CYP3A4; time-dependent inhibition; dimethyl sulfoxide (DMSO); diazepam
Introduction Cytochrome P450s (P450) are the principal enzymes in drug metabolism. In particular, CYP3A4 is a major P450 due to its relatively high abundance in the liver (Wildt et al. 1999) and its responsibility for the metabolism of a large number of drugs. Thus, many clinically important pharmacokinetic drug–drug interactions result from the impairment of metabolic clearance via CYP3A4. P450 inhibition is classified as reversible (competitive or noncompetitive) and irreversible (mechanism-based inhibition). In general, irreversible inhibition more frequently results in unfavourable drug–drug interactions compared with reversible inhibition, since the inhibitions are sustained even after the inhibitor has disappeared from the
body, until the inactivated P450s is replaced by newly synthesized protein (Kalgutkar et al. 2007; Grimm et al. 2009). Therefore, mechanism-based inhibition of CYP3A4 should be carefully evaluated in vitro in new drug-development process to avoid drug–drug interactions in vivo. Because of the lipophilic nature of recent new drug candidate, organic solvents are often required for solubilization of the test substances as well as P450 substrates to add adequate concentrations to an in vitro incubation. However, some organic solvents may have inhibitory or stimulatory effects on the activity of P450s both in human liver microsomes (HLM) and recombinant human P450 expressed microsomes (rCYPs) (Hichman et al. 1998; Busby et al. 1998). Recently, we observed that time-dependent inhibition of CYP3A4 by a
Address for Correspondence: Y. Nishiya, Department of Drug Metabolism & Pharmacokinetics Research Laboratories, Daiichi Sankyo Co., Ltd, 1-2-58 Hiromachi, Shinagawa-Ku, Tokyo 140-8710, Japan. Tel: 81-3-3492-3131. Fax: 81-3-5436-8567. E-mail:
[email protected] (Received 11 August 2009; revised 11 September 2009; accepted 14 September 2009) ISSN 0049-8254 print/ISSN 1366-5928 online © 2010 Informa UK Ltd DOI: 10.3109/00498250903337392
http://www.informahealthcare.com/xen
2 Y. Nishiya et al.
Xenobiotica Downloaded from informahealthcare.com by Sankyo Co. Ltd (Japan) Active on 03/21/11 For personal use only.
number of in-house compounds disappeared due to the effects of organic solvents. Since this disappearance of time-dependent inhibition caused false-negative results, the effect of organic solvents appears to be serious in the evaluation of time-dependent inhibition. Thus, we evaluated the effect of organic solvents, acetonitrile, dimethyl sulfoxide (DMSO) and methanol, on the time-dependent inhibition of CYP3A4 by diazepam, one of the inhibitors for CYP3A4 as well as the probe activity (Kenworthy et al. 2001; Andersson et al. 1994), using HLM and rCYP3A4. In addition, the inhibitory effect of DMSO on the metabolism of diazepam by CYP3A4 was determined.
Materials and methods Materials Diazepam, temazepam, nordiazepam (desmethyldiazepam), mibefradil, troleandomycin, reserpine, β-nicotinamide adenine dinucleotide phosphate (β-NADP) sodium salt, D-glucose 6-phosphate (G-6-P) disodium salt hydrate and glucose 6-phosphate dehydrogenase (G-6-PDH) from baker’s yeast were purchased from Sigma-Aldrich Co. (St. Louis, MO, USA). S-(+)-Mephenytoin, midazolam hydrochloride (±)-4′–hydroxymephenytoin and 1′–hydroxymidazolam were purchased from Ultrafine Chemicals (Manchester, UK). Three pools of human liver microsomes (HLM), prepared by combining the liver microsomal fractions from 50 donors (20 mg protein ml−1, Lot Nos 0610351, 0710403 and 0810433), were purchased from XenoTech, LLC. (Lenexa, KS, USA). Recombinant human P450 expressed microsomes (rCYPs), rCYP3A4 (Lot Nos 27001, 66311 and 82636), rCYP3A5 (Lot Nos 04156, 34084 and 85426), and rCYP2C19 (Lot Nos 10141, 33767 and 58971) were purchased from BD Gentest (Woburn, MA, USA). All other reagents and solvents were commercially available and of the highest purity. Effect of the organic solvent on time-dependent inhibition by diazepam The activities of midazolam 1′–hydroxylase and S-mephenytoin 4′–hydroxylase were determined as typical CYP3A4/5 and CYP2C19 activities, respectively (Yuan et al. 2002). The pre-incubation mixture contained 0.1 M potassium phosphate buffer (pH 7.4), HLM 0.2 mg-protein ml−1, 500 µM diazepam, 10 µM mibefradil or 10 µM troleandomycin dissolved in acetonitrile, and 1% DMSO or methanol in a total volume of 300 µl. For the rCYP’s experiment, the pre-incubation mixture contained 0.1 M potassium phosphate buffer (pH 7.4), 10 pmol P450 ml−1 of rCYP3A4, rCYP3A5 and rCYP2C19, and varying concentrations of diazepam in a total volume of 300 µl.
Diazepam was dissolved in acetonitrile, DMSO and methanol, and added to the pre-incubation mixture at a final concentration of 1%(v/v), ranging from 1 to 100 µM. Control samples were prepared by the addition of solvent alone. To the pre-incubation mixture previously maintained at 37°C, 30 µl of an NADPH-generating system containing 2.5 mM β-NADP, 25 mM G-6-P, 0.5 units ml−1 G-6-PDH and 10 mM MgCl2 was added and the mixture was incubated for 30 min. Subsequently, a 30-µl aliquot of the mixture was collected and added to an assay mixture (300 µl), consisting of 0.1 M potassium phosphate buffer (pH 7.4), 40 µM midazolam for CYP3A4/5 or 80 µM S-mephenytoin for CYP2C19, and the NADPH-generating system. After the assay mixture was incubated for 5 min at 37°C, a 200-µl aliquot of the assay mixture was collected and added to a mixture of 100 µl of methanol and 100 µl of acetonitrile, which contains 0.1 μM reserpine as an internal standard, to terminate the reaction. Determination of the inactivation kinetic parameters of diazepam for CYP3A4 The pre-incubation mixture contained 0.1 M potassium phosphate buffer (pH 7.4), 0.2 mg-protein ml−1 of HLM, or 10 pmol P450 ml−1 of rCYP3A4, and varying concentrations of diazepam in a total volume of 300 µl. Diazepam was dissolved in acetonitrile, DMSO and methanol and added to the pre-incubation mixture at a final concentration of 1% (v/v), ranging from 5 to 200 µM. Control samples were prepared by the addition of solvent alone. To the pre-incubation mixture previously maintained at 37°C, 30 µl of the NADPH-generating system was added and the mixture was incubated at zero, 15, 30, and 45 min for HLM and at zero, 10, 20, 30, and 40 min for CYP3A4, respectively. At each time point, a 30-µl aliquot of the mixture was collected and added to an assay mixture for midazolam 1′–hydroxylase activity (300 µl), consisting of 0.1 M potassium phosphate buffer (pH 7.4), 40 µM midazolam, and the NADPH-generating system. The incubation and termination of the reaction were performed in the same manner as described above. Determination of the metabolism kinetic parameters of diazepam The incubation mixture contained 0.1 M potassium phosphate buffer (pH 7.4), 0.2 mg-protein ml−1 of HLM, or 10 pmol P450 ml−1 of rCYP3A4, and varying concentrations of diazepam in a total volume of 300 µl. Diazepam was dissolved in acetonitrile, DMSO and methanol and added to the pre-incubation mixture at a final concentration of 1% (v/v), ranging from 5 to 200 µM. After the incubation mixture was incubated for 10 min at 37°C, a 200-µl aliquot of the assay mixture was collected and added to a mixture of 100 µl of methanol and 100 µl of acetonitrile,
Effects of organic solvents on the time-dependent inhibition of CYP3A4 3 which contains 0.1 μM reserpine as an internal standard, to terminate the reaction. Determination of the inhibition constant (Ki) of DMSO on the metabolism of diazepam
Xenobiotica Downloaded from informahealthcare.com by Sankyo Co. Ltd (Japan) Active on 03/21/11 For personal use only.
The incubation mixture contained 0.1 M potassium phosphate buffer (pH 7.4), 10 pmol P450 ml−1 of rCYP3A4, 12.5–100 µM diazepam dissolved in acetonitrile, and varying concentration of DMSO in a total volume of 300 µl. The incubation and termination of the reaction were performed in the same manner as described above. LC-MS/MS analysis The samples were centrifuged at 1870g for 15 min and the supernatant fractions were directly injected into the HPLC system (Alliance 2795; Waters Corporation, Milford, MA, USA) equipped with a Capcell Pak C18 MGII column (5 µm, 2.0 × 100 mm, Shizeido Co., Ltd, Tokyo, Japan). Detection and quantitation of 1′–hydroxymidazolam, 4′–hydroxymephenytoin, nordiazepam and temazepam were performed utilizing a mass spectrometer (Quattro Micro API, Waters Corporation). Elution was performed using a mixture of solvent A consisting of formic acid, 0.1 M ammonium acetate, purified water and methanol (2:5:90:5, v/v) and solvent B consisting of formic acid, 0.1 M ammonium acetate and methanol (2:5:95, v/v) as a mobile phase. The proportions of solvent B in the mobile phase increased from 40% to 90% linearly from zero to 5 min for 1′–hydroxymidazolam, kept at 20% for 1 min, increased from 20% to 80% linearly from 1 to 4 min for 4′–hydroxymephenytoin, and increased from 50% to 100% linearly from zero to 2 min for nordiazepam and temazepam. The peak areas of the m/z 342 → 324 product ion of 1′–hydroxymidazolam, 235 → 150 product ion of 4′–hydroxymephenytoin, 271 → 140 product ion of nordiazepam and 301 → 255 product ion of temazepam were measured against the peak areas of the m/z 609 → 195 product ion of the internal standard. Data analysis All incubations were performed in duplicate. The mean value of the midazolam 1′–hydroxylation and S-mephenytoin 4′–hydroxylation activity expressed as a percentage against the control activity was used to estimate the kinetic parameters of inactivation. The natural logarithm of the residual activities (LN percentage residual activity) was plotted against the pre-incubation time to calculate the observed inactivation rate constants (kobs) (Silverman 1988). The hyperbolic relationship between kobs and the concentrations of the test compounds was fitted by equation (1) using WinNonlin Professional to estimate the kinetic parameters of inactivation (Mayhew et al. 2000):
kobs
kinact [ I ] KI [ I ]
(1)
where kinact is the maximum inactivation rate constant; KI is the concentration of the test substance that produces a half-maximal rate of inactivation; and [I] is the concentration of the test substance. To estimate the kinetic parameters, maximum velocity of the metabolic reaction (Vmax), and Michaelis constant (Km), the mean values of nordiazepam and temazepam formation activities and the diazepam concentrations were fitted with a Michaelis–Menten equation using WinNonlin Professional. The inhibition constant (Ki) value of DMSO was evaluated by the Dixon plot. The kinact, KI, kinact/KI, Km, Vmax, and Vmax/Km values are expressed as the mean and standard deviation (SD). The differences between acetonitrile and the other solvents, DMSO and methanol, were statistically analysed by a Dunnett’s test with a significance level of 5%. The analyses were performed using the software EXSAS (Arm Co., Ltd, Osaka, Japan).
Results Effect of organic solvents on time-dependent inhibition by diazepam Figure 1 shows the effects of DMSO and methanol on time-dependent inhibition by diazepam, mibefradil and troleandomycin for midazolam 1′–hydroxylation activities in human liver microsomes. The time-dependent inhibition by diazepam decreased by adding DMSO, but was not affected by methanol. Among three mechanism-based inhibitors tested, this decrease was observed only for diazepam. The effect of DMSO on the time- and concentration-dependent inhibition by dizepam was observed in rCYP3A4, not in rCYP3A5 and rCYP2C19, which are involved in the metabolism of diazepam (Figure 2). Effect of organic solvents on the CYP3A4 inactivation kinetic parameters for diazepam The inactivation kinetics of CYP3A4 by diazepam are shown in Figure 3 and Table 1. The kinact/KI values for diazepam added as a solution in acetonitrile and methanol were almost equal. By contrast, inactivation by diazepam in DMSO was weaker and the kinetic parameter could not be calculated. Effect of organic solvents on metabolism kinetic parameters of diazepam Formations of the metabolites of diazepam, temazepam and nordiazepam, in rCYP3A4 decreased when the
4 Y. Nishiya et al. substrate was dissolved in DMSO as compared with acetonitrile and methanol with Vmax/Km values of 0.048, 0.20 and 0.20 µl min−1 mg protein−1 for nordiazepam and of 0.47, 0.90 and 0.98 µl min−1 mg protein−1 for temazepam,
Determination of the inhibition constant (Ki) of DMSO for the metabolism of diazepam by CYP3A4
Residual activity (% of control)
100.0 ***
80.0
DMSO inhibited the metabolism of diazepam by CYP3A4 in a competitive or mixed-type manner based on the results of the Dixon plot (Figure 5). The Ki values of DMSO for the formation of nordiazepam and temazepam were determined to be 6 and 24 mM, respectively.
60.0 40.0 20.0
Discussion
0.0 Diazepam
Mibefradil
Troleandomycin
Figure 1. Effect of organic solvents on the time-dependent inhibition of CYP3A4 by diazepam, mibefradil and troleandomycin in human liver microsomes. Three lots of pooled human liver microsomes were incubated for 30 min in the presence of an NADPH-generating system, 500 µM diazepam, 10 µM mibefradil or 10 µM troleandomycin dissolved in acetonitrile, water (the black bar), 1% DMSO (the grey bar) or 1% methanol (the white bar), and then the residual CYP3A4 enzyme activity as midazolam 1′–hydroxylation was assayed. Control was the addition of solvent alone instead of the inhibitors. For details on the determination of the activity, see the Materials and methods section. Each error bar represents the standard deviation (SD) of the mean of three separate determinations with human liver microsomes. Significant differences from water were shown (***p
