In vitro inhibition of human liver drug metabolizing enzymes by second generation antihistamines

Chemico-Biological Interactions 123 (1999) 63 – 79

www.elsevier.com/locate/chembiont

In vitro inhibition of human liver drug metabolizing enzymes by second generation antihistamines Jean-Marie Nicolas *, Rhys Whomsley, Philippe Collart, Jose Roba Department of Product Safety and Metabolism, UCB S.A. Pharma Sector, Chemin du Foriest, B-1420 Braine l’Alleud, Belgium Received 1 December 1998; received in revised form 4 July 1999; accepted 17 July 1999

Abstract Cetirizine, terfenadine, loratadine, astemizole and mizolastine were compared for their ability to inhibit marker activities for CYP1A2, CYP2C9, CYP2C19, CYP2D6, CYP3A4 and for some glucuronidation isoenzymes in human liver microsomes. The most pronounced effects were observed with terfenadine, astemizole and loratadine which inhibited CYP3A4mediated testosterone 6b-hydroxylation (IC50 of 23, 21 and 32 mM, respectively) and CYP2D6-mediated dextromethorphan O-demethylation (IC50 of 18, 36 and 15 mM, respectively). In addition, loratadine markedly inhibited the CYP2C19 marker activity, (S)mephenytoin 4-hydroxylation (Ki of 0.17 mM). Furthermore, loratadine activated the CYP2C9-catalyzed tolbutamide hydroxylation (ca. 3-fold increase at 30 mM) and inhibited some glucuronidation enzymes. Mizolastine appeared to be a relatively weak and unspecific inhibitor of CYP2E1, CYP2C9, CYP2D6 and CYP3A4 (IC50s in the 100 mmolar range). Cetirizine demonstrated no effect on the investigated activities. A comparison of the inhibitory potencies of cetirizine, terfenadine, loratidine, astemizole and mizolastine with their corresponding plasma concentrations in humans suggests that these antihistamines are not likely to interfere with the metabolic clearance of coadministered drugs, with the exception of loratidine, which appears to inhibit CYP2C19 with sufficient potency to warrant additional investigation. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Antihistamines; Cytochrome P-450; Drug – drug interactions; Glucuronidation enzymes



A portion of this work was presented at the 6th European ISSX Meeting, 1997, Gotenburg * Corresponding author. E-mail address: [email protected] (J.-M. Nicolas)

0009-2797/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 0 0 9 - 2 7 9 7 ( 9 9 ) 0 0 1 3 1 - 3

64

J.-M. Nicolas et al. / Chemico-Biological Interactions 123 (1999) 63–79

1. Introduction In recent years, drug – drug interactions concerning second-generation antihistamines have been given special consideration since the discovery that concomitant therapy with terfenadine and azole antifungals or macrolide antibiotics may give rise to life-threatening arrhythmic side effects [1]. Terfenadine undergoes a near complete first-pass metabolism by cytochrome P-450 3A4 (CYP3A4) to form a carboxylic acid, fexofenadine, which probably accounts for most of the clinical antihistaminic effect [2,3]. Several studies have demonstrated that the coadministration of known CYP3A4 inhibitors (e.g. ketoconazole and erythromycin) produce an accumulation of unchanged terfenadine in the blood-stream which could in turn trigger cardiac side-effects [1,4,5]. Other second-generation antihistaminic drugs, such as astemizole and loratadine, are also extensively metabolized by CYP3A4 to generate active metabolites [6 – 8]. It is not surprising, therefore, that pharmacokinetic studies have indicated that CYP3A4 inhibitors are able to reduce the metabolic clearance of these drugs thereby elevating their plasma levels [9,10]. However, not all second generation antihistamines are prone to giving rise to drug – drug interactions with CYP3A4 inhibitors. For example, cetirizine can be coadministered with ketoconazole, erythromycin and cimetidine without any modification in its disposition [Data on UCB files, [11,12]]. This finding is explained by the low metabolic extraction of cetirizine (more than 60% of the dose excreted unchanged in urine within 24 h) [13]. In vitro models have been extensively used to characterize the metabolic pathways of terfenadine, loratadine and astemizole, and to predict the risks of pharmacokinetic interactions [3,6,7,14 – 17]. Most studies have focused on the ability of CYP inhibitors to impair the metabolism of these antihistamines. However, few studies comparing the inhibitory potency of second-generation antihistamines on liver drug metabolizing enzymes have been performed. In vitro drug – drug interactions studies have traditionally involved the screening of compounds for inhibitory potency against the major CYP isozymes known to be involved in the biotransformation of drugs in man. However, it is now being recognized that interactions involving conjugative enzymes may also play a role in the interactions of concomitantly administered substances and that screening for potential interactions should be extended to these enzymes, where appropriate. The present study compared the ability of cetirizine, terfenadine, astemizole, loratadine and mizolastine (Fig. 1) to inhibit some of the major drug metabolizing CYP and UDP-glucuronidation (UGT) isoenzymes present in human liver microsomes. 2. Materials and methods

2.1. Chemicals and reagents Cetirizine and loratadine were synthesized at UCB S.A., Pharma Sector (Braine l’Alleud, Belgium). Mizolastine was kindly provided by V. Rovei (Synthelabo,

J.-M. Nicolas et al. / Chemico-Biological Interactions 123 (1999) 63–79

65

Bagneux, France). Astemizole was obtained from Janssen Biotech N.V. (Beerse, Belgium). Terfenadine, tolbutamide, chlorpropamide, Brij 58, uridine 5%-diphosphoglucuronic acid (UDPGA), paracetamol, [14C]paracetamol, phenacetin and ethinyloestradiol were obtained from Sigma (St. Louis, MO). Glucose-6-phosphate (disodium salt), glucose-6-phosphate dehydrogenase (from yeast, grade I), NADPH (tetrasodium salt) and NADP (disodium salt) were purchased from Boehringer Mannheim GmbH Biochemica (Mannheim, Germany). Other chemicals were obtained from the following sources: dextromethorphan, dextrorphan and triton X-100 from ICN Biomedicals (Costa Mesa, CA); hydroxytolbutamide, (S)mephenytoin and 4-hydroxymephenytoin from Ultrafine Chemicals (Manchester,

Fig. 1. Chemical structures of second-generation H1 receptor antagonists.

66

J.-M. Nicolas et al. / Chemico-Biological Interactions 123 (1999) 63–79

England); [3H]ethinyloestradiol from Dupont de Nemours (Wilmington, DE); testosterone from Fluka Chemika (Buchs, Switzerland) and 6b-hydroxytestosterone from Steraloids Inc. (Wilton, NH). All other chemicals were of analytical grade.

2.2. Human li6er homogenates and microsomes Human liver specimens were obtained under strict ethical conditions from transplant donors with no known liver disease. Samples were obtained from the Department of Toxicology, Vrije Universiteit, Brussels, Belgium (Prof. V Rogiers) and the Department of Pediatric Hepatology, Universite´ Catholique Louvain, Brussels, Belgium (Prof. E Sokal). Microsomes were prepared by differential centrifugation of liver homogenates as described elsewhere [18]. Homogenates and 105 000×g microsomal pellets were stored at − 80°C until use. Protein concentration was determined using the method described by Smith et al. [19].

2.3. Inhibition assays The following probe substrate assays were used: phenacetin O-deethylation for CYP1A2 [20], tolbutamide 4-methylhydroxylation for CYP2C9 [21], (S)mephenytoin 4-hydroxylation for CYP2C19 [22], dextromethorphan O-demethylation for CYP2D6 [23], testosterone 6b-hydroxylation for CYP3A4 [24] and, ethinyloestradiol-, paracetamol- and P-nitrophenol glucuronidation for UGT enzymes [25 – 27]. Under the conditions used, the enzyme activities were linear with respect to time and protein concentration. For determination of inhibitory potency of compounds on each CYP, microsomal samples from a bank held at UCB were selected having a high level of marker enzyme activity. As a first screen, all the marker activities were measured separately in two human liver microsomal preparations in the presence and absence of the antihistamines at a final concentration of 100 mM. Antihistamines were dissolved in DMSO and sequentially diluted with aqueous buffer to achieve the desired concentration. The final DMSO concentration in the incubates was kept at 1% (v/v) with the exception of incubates containing cetirizine which was dissolved in aqueous buffer. Control incubates were performed under the same conditions using these vehicles. Selected inhibition assays were extended to the measurement of IC50 (test substance concentration producing half of the maximal inhibition) using six final inhibitor concentrations (0, 3, 10, 30, 100 and 300 mM). Finally, the effect of loratadine on (S)-mephenytoin 4-hydroxylation was submitted to a full kinetic study to determine Ki values and to characterize the nature of the inhibition. For this purpose, the activities were measured using five probe substrate concentrations and four inhibitor concentrations (ca. 0-, 0.5-, 1-, and 2.5 × IC50). Unless otherwise stated, IC50 and Ki determinations were performed on a single human liver microsomal preparation.

J.-M. Nicolas et al. / Chemico-Biological Interactions 123 (1999) 63–79

67

2.4. Phenacetin O-deethylation (CYP1A2) The incubation mixture (0.5 ml reaction vol.) contained 0.5 ml human liver microsomes (0.8 mg protein/ml), 100 mM phenacetin, 1 mM NADPH and 5 mM MgCl2 in 100 mM phosphate buffer (pH 7.4). After a 25 min incubation at 37°C, the reaction was stopped by the addition of 250 ml acetonitrile containing 30 mM 3-acetamidophenol (internal standard). The proteins were pelleted by a 10 min centrifugation at 3000×g, and the supernatant was used for reverse phase HPLC analysis. Phenacetin and 3-acetamidophenol were separated on a Lichrospher 100 RP-18, 5 mm (250 × 4 mm) column (Merck, Darmstadt, Germany) using a binary mixture of 20 mM phosphate buffer (solvent A) and acetonitrile (solvent B) at a flow rate of 1 ml/min. The mobile phase was initially composed of solvent A/solvent B (90:10, v/v) for 12 min, followed by a linear gradient to solvent A/solvent B (80:20, v/v) over 17 min. The chromatography was then conducted under isocratic conditions for 25 min. Detection was by UV absorbance at 254 nm.

2.5. Tolbutamide 4 -methylhydroxylation (CYP2C9) The incubation mixture (0.5 ml reaction vol.) contained human liver microsomes (1 mg protein/ml), 150 mM tolbutamide, 1 mM NADPH and 5 mM MgCl2 in 100 mM phosphate buffer (pH 7.4). After a 20 min incubation at 37°C, the reaction was stopped by adding 60 ml of 3 M HCl and 20 ml of the internal standard (100 mM chlorpropamide). The hydroxytolbutamide formed and the internal standard were extracted by shaking for 10 min with 6 ml diethyl ether. The aqueous phase was frozen in an ice-cold methanol bath and the organic phase separated and evaporated under a stream of nitrogen. The dry residue was reconstituted in 100 ml mobile phase and analyzed by reverse phase HPLC. Tolbutamide, hydroxytolbutamide and chlorpropamide were separated at a flow rate of 1 ml/min on a Lichrospher 100 RP 18, 5 mm (125 × 4 mm) column (Merck, Darmstadt, Germany) using a mobile phase consisting of ethanol:water (40:60, v/v) containing 0.05% (v/v) ortho phosphoric acid. Detection was by UV absorbance at 230 nm.

2.6. (S) -Mephenytoin 4 -hydroxylation (CYP2C19) The incubation mixture (0.5 ml reaction volume) contained human liver microsomes (1.0 mg protein/ml), 50 mM S-mephenytoin, 1 mM NADPH and 15 mM MgCl2 in 50 mM Tris – HCl buffer (pH 7.4). After a 30 min incubation at 37°C, the reaction was stopped by the addition of 100 ml of a solution containing 2% (w/v) sodium azide and 10 mM phenobarbital (internal standard). Hydroxymephenytoin and phenobarbital were extracted by shaking for 10 min with 5 ml dichloromethane. The organic and aqueous phases were separated by a 5 min centrifugation at 1000× g. The organic layer was evaporated under a stream of nitrogen and the dry residue reconstituted in 160 ml of mobile phase and analyzed by reverse phase HPLC. (S)-Mephenytoin, hydroxymephenytoin and phenobarbital were separated on a Lichrospher 100 RP 18, 5 mm (125× 4 mm) column (Merck,

68

J.-M. Nicolas et al. / Chemico-Biological Interactions 123 (1999) 63–79

Darmstadt, Germany) at a flow rate of 1 ml/min in a mobile phase consisting of acetonitrile:water (30:70, v/v). Detection was by UV absorbance at 204 nm.

2.7. Dextromethorphan O-demethylation (CYP2D6) The incubation mixture (0.2 ml reaction volume) contained human liver microsomes (0.4 mg protein/ml), 50 mM dextromethorphan, 1 mM NADPH and 5 mM MgCl2 in 100 mM phosphate buffer (pH 7.4). After a 30 min incubation at 37°C, the reaction was stopped by adding 50 ml of 50% trichloroacetic acid. The proteins were pelleted by centrifugation at 3000×g for 10 min. The supernatant was analyzed by reverse phase HPLC using fluorescence detection (excitation wavelength of 270 nm, emission wavelength of 312 nm). Dextromethorphan and dextrorphan were separated on a Lichrospher 100 RP 18, 5 mm (125× 4 mm) column (Merck, Darmstadt, Germany) at a flow rate of 1 ml/min in a mobile phase consisting of 10 mM KH2PO4:methanol:acetonitrile:TEMED (69:5:22:1, v/ v/v/v).

2.8. Testosterone 6b-hydroxylation (CYP3A4) The incubation mixture (0.5 ml reaction volume) contained human liver microsomes (0.2 or 0.6 mg protein/ml), 250 mM testosterone, 1 mM NADPH, 1 mM EDTA, 3 mM MgCl2 in 100 mM phosphate buffer (pH 7.4). After a 10 min incubation at 37°C, the reaction was stopped by adding 50% (w/v) trichloroacetic acid solution (125 ml) containing 60 mM corticosterone (internal standard). The proteins were pelleted by a 10 min centrifugation at 1000× g. The supernatant was analyzed by reverse phase HPLC. Testosterone, 6b-hydroxytestosterone and corticosterone were separated on an Ultrasphere IP, 5 mm (250×4.6 mm) column (Beckman, Fullerton, CA) at a flow rate of 1 ml/min in a mobile phase of methanol:water:tetrahydrofuran (33:58:9, v/v/v). Detection was by UV absorbance at 254 nm.

2.9. Ethinylestradiol glucuronidation The microsomal sample was preincubated for 20 min at 4°C in 100 mM Tris – HCl buffer (pH 7.4) containing Triton X-100 (final detergent: protein ratio of 0.4:1 w/w). The assays were carried out in a 0.2 ml incubate containing preincubated microsomal sample (final protein concentration of 0.7 mg/ml), 192 mM [3H]ethinyloestradiol (5.2 mCi/mmol), 5 mM UDPGA and 5 mM MgCl2 in 100 mM Tris – HCl buffer (pH 7.4). After incubation at 37°C for 80 min, the reaction was stopped by the addition of 0.8 ml of 0.9 M trichloroacetic acid/0.6 M glycine. Unreacted ethinyloestradiol was removed by extraction with 5 ml of water-saturated ether, and a 200 ml aliquot of the aqueous phase was subjected to liquid scintillation spectrometry in 7 ml scintillation fluid.

J.-M. Nicolas et al. / Chemico-Biological Interactions 123 (1999) 63–79

69

2.10. Paracetamol glucuronidation The microsomal sample was preincubated for 30 min at 4°C in 100 mM Tris – HCl buffer (pH 7.4) containing Brij 58 (final detergent: protein ratio of 0.1:1 w/w). The assays were carried out in a 0.2 ml incubate containing preincubated microsomal sample (final protein concentration of 3 mg/ml), 500 mM [14C]paracetamol (200 nCi/mmol), 5 mM UDPGA and 5 mM MgCl2 in 100 mM Tris – HCl buffer (pH 7.4). After incubation at 37°C for 40 min, the reaction was stopped by the addition of 0.8 ml of 0.4 M trichloroacetic acid/0.6 M glycine. The proteins were pelleted by a 10 min centrifugation at 1500×g, and a 700 ml aliquot of the supernatant was extracted with 7 ml of ethyl acetate to remove unreacted paracetamol. Following centrifugation for 5 min at 1500×g, a 200 ml aliquot of the aqueous phase was subjected to liquid scintillation spectrometry in 7 ml scintillation fluid.

2.11. p-Nitrophenol glucuronidation The microsomal sample was preincubated for 30 min at 4°C in 125 mM Tris – HCl/125 mM maleate buffer (pH 7.4) containing Triton X-100 (final detergent: proteins ratio of 0.4:1 w/w). The assays were carried out in a 0.25 ml incubate containing preincubated microsomal sample (final protein concentration of 0.4 mg/ml), 500 mM p-nitrophenol, 4 mM UDPGA and 5 mM MgCl2 in 125 mM Tris – HCl/125 mM maleate buffer (pH 7.4). After a 20 min incubation at 37°C, the reaction was stopped by the addition of 0.5 ml of 0.5 M trichloroacetic acid/0.6 M glycine. The proteins were pelleted by a 10 min centrifugation at 1000× g, and an aliquot (0.4 ml) of the supernatant was mixed with 1.6 ml of 0.5 M NaOH. Unreacted P-nitrophenol was measured spectrophotometrically at 405 nm.

2.12. Analysis of data IC50 values were calculated by non-linear regression analysis assuming a sigmoid dose-effect relationship (Statistica software, version 5.0, Statsoft, Tulsa, OK). Reaction velocities at varying concentrations of probe substrate and inhibitor were analyzed by nonlinear regression using the conventional relationships for Michaelis – Menten kinetics and either competitive, noncompetitive, uncompetitive or linear mixed-type inhibition [28]. The model providing the best fit of parameters to the actual data was chosen based on the residual sum of squares, the distribution of residuals and visual inspection of the double reciprocal plots of the data.

3. Results Table 1 shows the effects of antihistamines on CYP and UGT probe activities when added at a final concentration of 100 mM. The activities of the marker enzyme

70

J.-M. Nicolas et al. / Chemico-Biological Interactions 123 (1999) 63–79

activities in the presence and absence of 1% DMSO are also shown. DMSO inhibited CYP2C9, CYP2C19 and CYP3A4 activities to some extent (up to 59%), however, the residual activity was sufficiently high for the measurement of inhibitory effects by the test compounds. IC50 and Ki data for inhibition of enzyme activities by the antihistamines are displayed in Table 2 and Fig. 2. At a concentration of 100 mM, mizolastine inhibited the marker activity for CYP1A2, phenacetin O-deethylation, by 45%. The other compounds evaluated did not inhibit this activity by more than 26%. CYP2C9 activity was monitored by measuring the rates of formation of 4-hydroxytolbutamide following incubation of tolbutamide with human liver microsomes. Terfenadine, astemizole and mizolastine had similar inhibitory effects on CYP2C9, with IC50 values of 66, 114 and 69 mM, respectively. At 100 mM, cetirizine had no effect, while loratadine appeared to increase this enzyme activity (Fig. 4). This latter finding was subsequently shown to be dose related, being maximal at 30 mM (ca 3-fold control value). Cetirizine and mizolastine (100 mM) had little effect on S-mephenytoin 4-hydroxylase activity, a marker for CYP2C19. The activity of this enzyme was, however, inhibited in the presence of the same concentration of terfenadine and astemizole by 47 and 46%, respectively. The IC50 values for the inhibition of CYP2C19 by these compounds were 138 and 96 mM, respectively. No enzymic activity was detectable in the presence of loratadine (100 mM) for which an IC50 value of B 1 mM was obtained. Detailed kinetic analysis demonstrated that this inhibition was competitive with a Ki of 0.17 mM (Fig. 3). With the exception of cetirizine, inhibitory effects on dextromethorphan Odemethylation were observed in the presence of all compounds at an inhibitor concentration of 100 mM. Astemizole, loratadine and terfenadine inhibited enzyme activity by more than 75% and the IC50 values for these compounds were 36, 15 and 18 mM, respectively. Mizolastine inhibited CYP2D6 by 43% at 100 mM with an IC50 value of 118 mM Fig. 4. At 100 mM all compounds except cetirizine inhibited CYP3A4, as determined by the measurement of testosterone 6b-hydroxylase activity, to some extent. The IC50 value for each compound was determined and found to be \ 300 mM for cetirizine and 23, 21, 32 and 119 mM, respectively, for terfenadine, astemizole, loratadine and mizolastine. No inhibitory effects greater than 20% were observed for ethinylestradiol glucuronidation in the presence of the test compounds at 100 mM. Paracetamol glucuronidation was inhibited to some extent by all compounds, except cetirizine, at an inhibitor concentration of 100 mM, loratadine being the most potent inhibitor (39% inhibition, IC50 123 mM). Similarly loratadine was the most potent inhibitor of p-nitrophenol glucuronidation, with an IC50 value of 73 mM. Some inhibitory effects (B 30%) were also observed in the presence of 100 mM terfenadine, astemizole and mizolastine. Cetirizine (100 mM) was non-inhibitory towards all UGT activities investigated Fig. 5.

J.-M. Nicolas et al. / Chemico-Biological Interactions 123 (1999) 63–79 71

Fig. 2. Effect of antihistamines on CYP marker activities. Effects of (A) cetirizine, (B) terfenadine, (C) astemizole, (D) loratadine and (E) mizolastine were determined on phenacetin O-deethylation ( ), tolbutamide hydroxylation ( ), (S)-mephenytoin hydroxylation ( ), dextromethorphan O-demethylation () and testosterone 6b-hydroxylation (). Activities are expressed as a percentage of control activity in the presence of the corresponding vehicle. The inhibition assays were conducted using a single human liver microsomal sample. The exception was inhibition of testosterone 6b-hydroxylation (all antihistamines) and (S)-mephenytoin hydroxylation (loratadine), which were determined on three microsomal samples with results reported as means.

J.-M. Nicolas et al. / Chemico-Biological Interactions 123 (1999) 63–79

72

Table 1 Effect of antihistamines (100 mM) on human liver microsomal CYP and UGT marker activitiesa (iso) Enzyme

CYP1A2

CYP2C9 CYP2C19 CYP2D6

CYP3A4

UGT1A1 UGT1A6 Nonspecific UGT

Probe assay

Phenacetin O-deethylationb Tolbutamide hydroxylationc (S)-mephenytoin hydroxylationd Dextromethorphan Odemethylatione Testosterone 6b-hydroxylationf Ethinylestradiol glucuronidationg Paracetamol glucuronidationh P-nitrophenol glucuronidationi

% of control activity remaining Cetirizine

Terfenadine

Astemizole

Loratadine

Mizolastine

99

74

75

90

55

98

56

54

184

51

98

53

54

B6

86

100

B20

B20

21

57

98

10

14

43

55

107

94

80

100

99

101

80

89

61

65

98

76

79

53

85

a Activity remaining is given as % of vehicle control. Incubations were conducted in triplicate for each of two microsomal samples and the inhibition of the reaction calculated for each microsomal preparation. The degree of inhibition was similar using either microsomal preparation and the data presented is the mean activity remaining for the two samples. b Mean control enzyme activity in the absence and presence of DMSO was 905 and 789 pmol/min per mg protein, respectively. c Mean control enzyme activity in the absence and presence of DMSO was 271 and 172 pmol/min per mg protein, respectively. d Mean control enzyme activity in the absence and presence of DMSO was 109 and 45 pmol/min per mg protein, respectively. e Mean control enzyme activity in the absence and presence of DMSO was 238 and 221 pmol/min per mg protein, respectively. f Mean control enzyme activity in the absence and presence of DMSO was 12056 and 6801 pmol/min per mg protein, respectively. g Mean control enzyme activity in the absence and presence of DMSO was 81 and 87 pmol/min per mg protein, respectively. h Mean control enzyme activity in the absence and presence of DMSO was 207 and 219 pmol/min per mg protein, respectively. i Mean control enzyme activity in the absence and presence of DMSO was 27 and 23 pmol/min per mg protein, respectively.

4. Discussion The current study represents the first detailed in vitro investigation on the ability of second-generation antihistamines to inhibit human liver cytochrome P450 and

J.-M. Nicolas et al. / Chemico-Biological Interactions 123 (1999) 63–79

73

UDP glucuronyltransferase isoenzymes. Initially, the compounds were all screened for inhibitory activity at 100 mM and further assays were then performed in order to provide IC50 or Ki values for selected activities and compounds. The data obtained provided evidence for marked differences in the inhibitory spectrum of the antihistamines tested. No marked inhibition of CYP1A2 or CYP2C9 activity was observed under the conditions used for the present study during the preliminary screen. Mizolastine was the most effective inhibitor of CYP1A2 activity. Mizolastine (IC50 = 69 mM) and terfenadine (IC50 =66 mM) were the most potent inhibitors of CYP2C9 activity. Interestingly, loratadine appeared to increase CYP2C9 activity in a dose-dependent manner. This may have been due to an allosteric mechanism as already reported for 7,8-benzoflavone and CYP2E1 and CYP3A4 activities [29,30]. It is unlikely that the effect was due to inhibition of secondary metabolism of hydroxytolbutamide as the enzymes involved, alcohol and aldehyde dehdrogenase, are both cytosolic. Some inhibition of CYP2C19 was observed in the presence of both terfenadine and astemizole, however, loratadine was shown to be a very potent inhibitor of this enzyme with a Ki value of 0.17 mM. In comparison, omeprazole, one of the most effective in vitro and in vivo inhibitors of CYP2C19, exhibits a Ki value of 1 mM [31]. Therefore loratadine is amongst the most potent inhibitors known for this enzyme. CYP2C19 plays a major role in the metabolism of omeprazole, however studies using correlation analysis have demonstrated that this isozyme is not involved in the metabolism of loratadine [6]. The chemical structure of loratadine combines some of the characteristics of known substrates for CYP2C19 such as amitryptyline, imipramine, proguanil and warfarin. Molecular modelling studies [32] of the human CYP2C subfamily have attempted to rationalise binding affinities of CYP2C19 substrates and inhibitors through substrate specificity and site–diTable 2 IC50 (mM) of antihistamines for inhibition of human microsoma CYP and UGT marker activitiesa (iso)Enzyme

Cetirizine

Terfenadine

Astemizole

Loratadine

Mizolastine

CYP1A2 CYP2C9 CYP2C19 CYP2D6 CYP3A4 Ethinylestradiol glucuronidation Paracetamol glucuronidation p-nitrophenol glucuronidation

\100 \100 \100 \100 \300 \100

\100 66 138 18 23 96 \100

\100 114 96 36 21 95 \100

\100 increase B1 (0.17*) 15 32 910 \100

\100 69 \100 118 119 913 \100

\100

\100

\100

123

\100

\100

\100

\100

73

\100

a

Assays were performed on either one or three human liver microsomal samples. In this latter case, results are expressed as mean 9 standard deviation. * Ki value in mM.

74

J.-M. Nicolas et al. / Chemico-Biological Interactions 123 (1999) 63–79

Fig. 3. Eadie–Hofstee plot of the inhibitory effect of loratadine on (S)-mephenytoin hydroxylation. Loratadine concentrations used were 0 ( ), 0.25 ( ), 0.5 () and 1.25 () mM. Data were fitted by non-linear regression analysis using a competitive inhibition model. These inhibition assays were conducted using a single human liver microsomal sample.

rected mutagenesis experiments. If this model is taken into consideration, one may speculate that the high affinity of loratadine might be explained through the cooperativity of a hydrogen bond interaction of a His 99 residue with the ketone moiety and the p stacking of the chlorophenyl group with Ph75, placing the chloro substituent in hydrophobic contact with Leu332. It appears that such interactions may allow loratadine to bind strongly to the active site of CYP2C19 without itself being metabolized by the enzyme. In view of the lack of a selective inhibitor for CYP2C19, loratadine may be an useful agent for the study of in vitro drug interactions, although its selectivity with regard to inhibition of CYP2D6 and CYP3A4, the two isozymes known to be involved in its metabolism, would need to be more clearly defined. The discovery that loratadine is a potent inhibitor of this enzyme should also give further insight into the structure activity relationships for substrates and inhibitors of CYP2C19. The IC50 value (18 mM) for inhibition of dextromethorphan hydroxylation by terfenadine was in the range found in previous studies [33] for CYP2D6 substrates, where terfenadine inhibited bufuralol 1-hydroxylase activity with a Ki of 3.6 mM and was also found to be a substrate for CYP2D6, having a Km of 13 mM. In the present study, loratadine inhibited dextromethorphan hydroxylase to a similar extent and this drug is also known to be metabolized by CYP2D6 [6]. The inhibitory effect of astemizole on dextromethorphan hydroxylation (approx. 50% of that observed with terfenadine) was more unexpected, as there are no reports in the

J.-M. Nicolas et al. / Chemico-Biological Interactions 123 (1999) 63–79

75

Fig. 4. Effect of loratadine on tolbutamide hydroxylation. Activities are expressed as a percentage of control activity in the presence of the corresponding vehicle. This assay was conducted using a single human liver microsomal sample.

literature of this compound being a substrate for CYP2D6. Previous studies have shown that a number of H1 antagonists interact with CYP2D6 [33,34] and that the substrate structure-activity relationships of the H1 receptor and CYP2D6 show some similarities in that both proteins bind molecules via an aspartic acid residue [33,35,36]. Active site template models of substrates for CYP2D6 have the same

Fig. 5. Effect of loratadine on glucuronidation activity towards () p-nitrophenol and ( ) paracetamol. Activities are expressed as a percentage of control activity in the presence of the corresponding vehicle. These inhibition assays were conducted using a single human liver microsomal sample.

76

J.-M. Nicolas et al. / Chemico-Biological Interactions 123 (1999) 63–79

basic feature of a basic nitrogen atom at 5–7 A from the site of metabolism [37]. The aspartic acid residue is believed to provide the carboxylate residue which binds the basic nitrogen of the substrates. Astemizole, being a lipophilic arylalkylamine, contains some of the basic features considered necessary to interact with CYP2D6. However, it is unknown whether the effects observed in the present study result from an interaction of astemizole as an inhibitor or as a substrate. The metabolism of terfenadine by CYP3A4 is well documented and therefore it is not surprising that this compound inhibited testosterone 6ß-hydroxylation. Astemizole and loratadine, which are also metabolized by CYP3A4, inhibited testosterone 6b-hydroxylation to a similar extent. The effect of the test compounds on in vitro glucuronidation activities was performed using ethinylestradiol, paracetamol and p-nitrophenol as substrates. Ethinylestradiol is a substrate for UGT1A1 [38], while at the concentration used in the study, paracetamol is preferentially glucuronidated by UGT1A6 [39]. The glucuronidation of 4-nitrophenol is catalyzed by several UGTs including UGT1A3 and UGT1A6 [40,41]. Inhibition of glucuronidation enzymes has been reported for drugs such as cyproheptadine and ketotifen [42] which are structurally related to some second-generation antihistamines. Some compounds metabolized by UGT enzymes, e.g. Zileuton, have been shown to alter the pharmacokinetics of UGT substrates (eg theophylline) in vivo, and to cause pharmacologically relevant interactions [43]. In the present study, the test compounds had little effect on ethinylestradiol activity suggesting a low potential for interaction with substrates for UGT1A1, however loratadine inhibited both paracetamol and p-nitrophenol glucuronidation to some extent. At therapeutic concentrations of these compounds the potential for inhibition of metabolism of coadministered compounds may be limited. At therapeutic doses, maximum plasma levels of cetirizine, terfenadine, astemizole, loratadine and mizolastine are in the range of 556, 5 3, B2, 12 and 638 nM, respectively [44,45]. The reason for the low plasma levels of terfenadine, astemizole and loratadine is their extensive distribution in addition to an extensive first-pass metabolism after oral administration [46–48]. As a result, the hepatic levels of these compounds may exceed the plasma concentrations. Therefore, for loratadine, the human hepatic blood concentration following a therapeutic dose may be in the range of the CYP2C19 inhibitory concentration for this drug. Whether loratadine might have clinically important effects on drugs metabolized by CYP2C19 (e.g. S-mephenytoin, omeprazole, propranolol, imipramine [49]) remains to be clarified, however, there are no reports of such interactions in the literature. Up to the highest tested concentrations of 100–300 mM, cetirizine did not inhibit the investigated CYP and UGT marker activities. The absence of any inhibition demonstrated that cetirizine is not a high affinity substrate for these isoenzymes, consistent with its low in vivo metabolic clearance [13].

J.-M. Nicolas et al. / Chemico-Biological Interactions 123 (1999) 63–79

77

In conclusion, the inhibitory potencies of cetirizine, terfenadine, loratidine, astemizole and mizolastine with their corresponding plasma concentrations in humans suggests that these antihistamines are not likely to interfere with the metabolic clearance of coadministered drugs, with the exception of loratidine, which appears to inhibit CYP2C19 with sufficient potency to warrant additional investigation.

Acknowledgements We gratefully acknowledge C. Derwa, F. Meriaux and P. Piette for their excellent technical assistance.

References [1] R.L. Woosley, Y. Chen, J.P. Freiman, R.A. Gillis, Mechanism of the cardiotoxic actions of terfenadine, J. Am. Med. Assoc. 269 (1993) 1532 – 1536. [2] D.A. Garteiz, R.H. Hook, B.J. Walker, R.A. Okerholm, Pharmacokinetics and biotransformation studies of terfenadine in man, Arzneim.-Frosch. Drug Res. 32 (1982) 1185 – 1190. [3] C.H. Yun, R.A. Okerholm, F.P. Guengerich, Oxidation of the antihistaminic drug terfenadine in human liver microsomes, role of cytochrome P-450 3A(4) in N-dealkylation and C-hydroxylation, Drug Metab. Dispos. 21 (1993) 403– 409. [4] P.K. Honig, D.C. Wortham, K. Zamani, D.P. Conner, J.C. Mullin, L.R. Cantilena, Terfenadineketoconazole interaction. Pharmacokinetic and electrocardiographic consequences, J. Am. Med. Assoc. 269 (1993) 1513–1518. [5] P.K. Honig, D.C. Wortham, K. Zamani, L.R. Cantilena, Comparison of the effect of the macrolide antibiotics erythromycin, clarithromycin and azithromycin on terfenadine steady-state pharmacokinetics and electrocardiographis parameters, Drug Invest. 7 (1994) 148 – 156. [6] N. Yumibe, K. Huie, K.J. Chen, R.P. Clement, M.N. Cayen, Identification of human liver cytochrome P450 enzymes that metabolize the nonsedating antihistamine loratadine, Biochem. Pharmacol. 51 (1996) 165–172. [7] K. Lavrijsen, J.V. Houdt, W. Meuldermans, M. Janssens, J. Heykants, The interaction of ketoconazole, itraconazole and erythromycin with the in vitro metabolism of antihistamines in human liver microsomes, Eur. J. Allergy-Clin. Immunol. 48 (1993) 34. [8] J.P. Desager, Y. Hormans, Pharmacokinetic-pharmacodynamic relationships of H1-antihitamines, Clin. Pharmacokinet. 28 (1995) 419– 432. [9] M.D. Brannan, M.B. Affrime, E. Radwanski, M.N. Cayen, C. Banfield, Effects of various cytochrome P450 inhibitors on the metabolism of loratadine, Clin. Pharmacol. Ther. 57 (1995) 193. [10] K. Bachmann, T.J. Sullivan, J.H. Reese, L. Jauregui, K. Miller, M. Scott, A study of the interaction between dirithromycin and astemizole in healthy adults, Am. J. Ther. 4 (1997) 73 – 79. [11] M. Sale, R.L. Woosley, K. Thakker, K. Phillips, F. Caridi, M. Chung, Effects of cetirizine and erythromycin, alone and in combination on QT interval and pharmacokinetics in healthy subjects, Ann. Allergy Asthma Immunol. 76 (1996) 93. [12] K.J. Simons, G.L. Suusman, F.E.R. Simons, The effect of the H2-antagonist cimetidine (C) on the pharmacokinetics (PK) and pharmacodynamics (PD) of the H1-antagonist (H1-RA) cetirizine (CET) in patients with chronic urticaria, Pharm. Res. 10 (1993) Suppl. 355. [13] S.G. Wood, B.A. John, L.F. Chasseaud, J. Yen, M. Chung, The metabolism and pharmacokinetics of 14C-cetirizine in humans, Ann. Allergy 59 (1987) 31 – 34.

78

J.-M. Nicolas et al. / Chemico-Biological Interactions 123 (1999) 63–79

[14] M. Jurima-Romet, K. Crawford, T. Cyr, T. Inaba, Terfenadine metabolism in human liver. In vitro inhibition by macrolide antibiotics and azole antifungals, Drug Metab. Dispos. 22 (1994) 849 – 857. [15] A.D. Rodrigues, D.J. Mulford, B.W. Surber, M.J. Kukula, J.L. Ferrero, S.B. Thomas, M.S. Shet, R.W. Estabrook, In vitro metabolism of terfenadine by a purified recombinant fusion protein containing cytochrome P-4503A4 and NADPH-P450 reductase, Drug Metab. Dispos. 23 (1995) 765–775. [16] L.L. von Moltke, D.J. Greenblatt, S.X. Duan, J.S. Harmatz, C.E. Wright, R.I. Shader, Inhibition of terfenadine matabolism in vitro by azole antifungal agents and by selective serotonin reuptake inhibitor antidepressants: relation to pharmacokinetic interactions in vivo, J. Clin. Psychopharmacol. 16 (1996) 104–112. [17] A.P. Li, M. Jurima-Romet, Applications of primary human hepatocytes in the evaluation of pharmacokinetic drug–drug interactions: evaluation of model drugs terfenadine and rifampicin, Cell Biol. Toxicol. 13 (1997) 365–374. [18] P. Kremers, P. Beaune, T. Cresteil, J. De Graeve, S. Columelli, J.P. Leroux, J. Gielen, Cytochrome P-450 monooxygenase activities in human and rat liver microsomes, Eur. J. Biochem. 118 (1981) 599–606. [19] P.K. Smith, R.I. Krohn, G.T. Hermanson, A.K. Mallia, F.H. Gartner, M.D. Provenzano, Fujimoto, N.M. Goeke, B.J. Olson, D.C. Klenk, Measurement of protein using bicinchoninic acid, Anal. Biochem. 150 (1985) 76–85. [20] M. Bourrie´, V. Meunier, Y. Berger, G. Fabre, Cytochrome P450 isoform inhibitors as a tool for the investigation of metabolic reactions catalysed by human liver microsomes, J. Pharmacol. Exp. Ther. 277 (1996) 321–332. [21] J.W. Ho, D.E. Moody, Determination of tolbutamide hydroxylation in rat liver microsomes by high-performance liquid chromatography: effect of psychoactive drugs on in vitro activity, Life Sci. 52 (1992) 21–28. [22] S.A. Wrighton, J.C. Stevens, G.W. Becker, M. Vandenbranden, Isolation and characterization of human liver cytochrome P450 2C19: correlation between 2C19 and S-mephenytoin 4%-hydroxylation, Arch. Biochem. Biophys. 306 (1993) 240 – 245. [23] D. Wu, S.V. Otton, B.A. Sproule, U. Busto, T. Inaba, W. Kalow, E.M. Sellers, Inhibition of human cytochrome P450 2D6(CYP2D6) by methadone, Br. J. Clin. Pharmacol. 35 (1993) 30 – 34. [24] M.P. Arlotto, J.M. Trant, R.W. Estabrook, Measurement of steroid hydroxylation reactions by high-performance liquid chromatography as indicator of P450 identity and function, Methods Enzymol. 206 (1991) 454–462. [25] G.M. Pacifici, D.J. Back, M. Orme, Sulphation and glucuronidation of paracetamol in human liver: assay conditions, Biochem. Pharmacol. 37 (1988) 4405 – 4407. [26] G.M. Pacifici, D.J. Back, Sulphation and glucuronidation of ethinyloestradiol in human liver in vitro, J. Steroid Biochem. 31 (1988) 345 – 349. [27] B. Burchell, P. Weatherill, 4-Nitrophenol UDPglucuronyltransferase (rat liver), Methods Enzymol. 77 (1981) 169–177. [28] I.H. Segel, Enzyme Kinetics, Behavior and Analysis of Rapid Equilibrium and Steady-State Enzyme Systems, Wiley–Interscience Publication, 1993. [29] D.J. Newton, R.W. Wang, A.Y.H. Lu, Evaluation of specificities in the in vitro metabolism of therapeutic agents by human liver microsomes, Drug Metab. Dispos. 23 (1995) 154 – 158. [30] Y.F. Ueng, T. Kuwabara, Y.J. Chun, F.P. Guengerich, Cooperativity in oxidations catalysed by cytochrome P450 3A4, Biochemistry 36 (1997) 370 – 381. [31] J.W. Ko, N. Sukhova, T. Tacker, P. Chen, D. Flockhart, Evaluation of omeprazole and lansoprazole as inhibitors of cytochrome P450 isoforms, Drug Metab. Dispos. 25 (1997) 853 – 862. [32] D.F.V. Lewis, M. Dickins, R.J. Weaver, P.J. Eddershaw, P.S. Goldfarb, M.H. Tarbit, Molecular modelling of human CYP2C subfamily enzymes CYP2C9 and CYP2C19: rationalization of substrate specificity and site-directed mutagenesis experiments in the CYP2C subfamily, Xenobiotica 28 (1998) 235–268. [33] B.C. Jones, R. Hyland, M. Ackland, C.A. Tyman, D.A. Smith, Interaction of terfenadine and its primary metabolites with cytochrome P450 2D6, Drug Metab. Dispos. 26 (1998) 875 – 882.

J.-M. Nicolas et al. / Chemico-Biological Interactions 123 (1999) 63–79

79

[34] B.A. Hamelin, A. Bouayad, A. Gravel, J. Turgeon, In vitro characterization of cytochrome P450 2D6 inhibition by classic histamine H1 receptor antagonists, Drug Metab Dispos. 26 (1998) 536–539. [35] A.M. ter Laak, J. Venhorst, G.M. Donne-Op-den-Kelder, H. Timmerman, The histamine H1-receptor antagonist binding site. A stereoselective pharmacophoric model based upon (semi-)rigid H1-antagonists and including a known interaction site on the receptor, J. Med. Chem. 38(1995), 3351–60. [36] S.W. Ellis, G.P. Hayhurst, G. Smith, T. Lightfoot, M.M. Wong, A.P. Simula, M.J. Ackland, M.J. Sternberg, M.S. Lennard, G.T. Tucker, C.R. Woolf, Evidence that aspartic acid 301 is a critical substrate-contact residue in the active site of cytochrome P450 2D6, J. Biol. Chem. 270 (1995) 29055–29058. [37] D.S. Smith, B.C. Jones, Speculations on the structure-activity relationship (SSAR) of cytochrome P450 enzymes, Biochem. Pharmacol. 44 (1992) 2089 – 2098. [38] T Ebner, R.P. Remmel, B. Burcell, Human bilirubin UDP-glucuronosyltransferase catalyzes the glucuronidation of ethinylestradiol, Mol. Pharmacol. 43 (1993) 649 – 654. [39] B. Burchell, C.H. Brierly, D. Rance, Specificity of human UDP-glucuronyltransferases and xenobiotic glucuronidation, Life Sciences 57 (1995) 1819 – 1831. [40] M.D. Green, C.D. King, B. Mojarrabi, P.I. Mackenzie, T.R. Tephly, Glucuronidation of amines and other xenobiotics catalyzed by expressed human UDP-glucuronyltransferase 1A3, Drug Metab. Dispos. 26 (1998) 507–512. [41] M. Ciotti, A. Marrone, C. Potter, I.S. Owens, Genetic polymorphism in the human UGT1A6 (planar phenol) UDP-glucuronosyltransferase: pharmacological implications, Pharmacogenetics 7 (1997) 485–495. [42] S. Sharp, L.Y. Mak, D.J. Smith, M.W.H. Coughtrie, Inhibition of human and rabbit liver steroid and xenobiotic UDP-glucuronosyltransferases by tertiary amine drugs – implications for adverse drug reactions, Xenobiotica 22 (1992) 13 – 25. [43] G.R. Granneman, A.R. Braeckman, C.S. Locke, J.H. Cavanaugh, L.M. Dube, W.M. Awni, Effect of zileuton on theophylline pharmacokinetics, Clin. Pharmacokinet. 2 Suppl. 29 (1995) 77 – 83. [44] F. Estelle, R. Simons, K.J. Simons, Pharmacokinetic optimisation of histamine H1-receptor antagonist therapy, Clin. Pharmacokinet. 21 (1991) 372 – 393. [45] A. Prakash, H.M. Lamb, Mizolastine. A review of its use in allergic rhinitis and chronic idiopathic urticaria, Biodrugs 10 (1998) 41–63. [46] D.M. Richards, R.N. Brogden, R.C. Heel, T.M. Speight, G.S. Avery, Astemizole. A review of its pharmacodynamic properties and therapeutic efficacy, Drugs 28 (1984) 38 – 61. [47] S.P. Clissold, E.M. Sorkin, K.L. Goa, Loratadine. A preliminary review of its pharmacodynamic properties and therapeutic efficay, Drugs 37 (1989) 42 – 57. [48] D. McTavish, K.L. Goa, M. Ferrill, Terfenadine. An updated review of its pharmacological properties and therapeutic efficay, Drugs 39 (1990) 552 – 574. [49] D.A. Flockhart, Drug interactions and the cytochrome P450 system. The role of cytochrome P450 2C19, Pharmacokinetics 29 (1995) 45 – 52.

.